RepoMirrors
/
ceph
mirror of https://github.com/ceph/ceph
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
ceph/doc/changelog/v12.2.2.txt

9541 lines
332 KiB
Plaintext
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

commit cf0baeeeeba3b47f9427c6c97e2144b094b7e5ba
Author: Jenkins Build Slave User <ceph-release-team@redhat.com>
Date: Thu Nov 30 14:59:27 2017 +0000
12.2.2
commit 83684b91a3c6b31419114b83fc22106146885fb6
Merge: 6d9f2161b2 b5d02ac0fd
Author: Kefu Chai <tchaikov@gmail.com>
Date: Thu Nov 30 22:26:46 2017 +0800
Merge pull request #19240 from tchaikov/wip-22266-luminous
luminous: tools/ceph_monstore_tool: rebuild initial mgrmap also
Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
commit 6d9f2161b269b19b0301348dd9b17660420624d5
Merge: 0b4c3b8f1e b02075a963
Author: Gregory Farnum <greg@gregs42.com>
Date: Wed Nov 29 16:25:14 2017 -0500
Merge pull request #19230 from tchaikov/wip-22247-luminous
luminous: mon/PGMap: Fix %USED calculation
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
commit 0b4c3b8f1ed72c3a0053ed372e1c5c9b0d48de59
Merge: a57edd4f1a eb99a7ee36
Author: Andrew Schoen <andrew.schoen@gmail.com>
Date: Wed Nov 29 10:33:34 2017 -0600
Merge pull request #19239 from ceph/luminous-bz1518264
luminous: ceph-volume correctly fallback to bluestore when no objectstore is specified
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
commit b5d02ac0fd60c26417ad33dc6efcf80bff503fdd
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Nov 29 20:21:33 2017 +0800
tools/ceph_monstore_tool: rebuild initial mgrmap also
Fixes: http://tracker.ceph.com/issues/22266
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit f63d1da4afa3463730ecbc0be29df6375b79fa8f)
commit eb99a7ee36d541939f589a674d9890b14e017dbe
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 28 09:22:19 2017 -0500
ceph-volume lvm.activate correct detection and fallback of objectstore flags
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f0ac2dfdbc0a05cff0a5edbedca91274885b7870)
commit 10dbbb0a6e09c6181e81b04765cd368963753bb8
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 28 09:21:34 2017 -0500
ceph-volume tests.devices.lvm verify objectstore flags in activate
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 8d22856d6c95ac3b45918765b382cae507a8d1a4)
commit a57edd4f1a47f08c0f93d7a3ff57c376b72c9a8a
Merge: 1b0a5660f7 ee02b5c037
Author: Sage Weil <sage@newdream.net>
Date: Wed Nov 29 08:43:07 2017 -0600
Merge pull request #19217 from liewegas/wip-p2p-app-warning-luminous
upgrade/jewel-x/parallel: debug fuse
commit b02075a963c24c6e1b24d1b076275dcce5350a03
Author: Xiaoxi Chen <xiaoxchen@ebay.com>
Date: Sun Nov 26 22:51:58 2017 -0700
mon/PGMap: Fix %USED calculation bug.
Previous code forgot to multiple raw_used_ratio to calculate
used byte.
Fixes: http://tracker.ceph.com/issues/22247
Signed-off-by: Xiaoxi Chen <xiaoxchen@ebay.com>
(cherry picked from commit d10c6c26f9ef7372e2c95da79d23b07ce5f4e0e5)
commit ee02b5c037c67c708335745898328e990ecacc90
Author: Sage Weil <sage@redhat.com>
Date: Tue Nov 28 16:48:36 2017 -0600
qa/suites/upgrade/jewel-x/point-to-point: whitelist more stuff
Signed-off-by: Sage Weil <sage@redhat.com>
commit d2a284c782fed3bec4ed38c898351826bca0a8bc
Author: Sage Weil <sage@redhat.com>
Date: Tue Nov 28 15:02:18 2017 -0600
qa/suites/upgrade/jewel-x/parallel: only mount ceph-fuse once
Otherwise we do 4 mounts for no good reason.
Signed-off-by: Sage Weil <sage@redhat.com>
commit b7c6d9799cbfba3cff7bee5fd1de309a8299d911
Author: Sage Weil <sage@redhat.com>
Date: Tue Nov 28 15:00:05 2017 -0600
upgrade/jewel-x/parallel: debug fuse
Signed-off-by: Sage Weil <sage@redhat.com>
commit 1b0a5660f793714e8cb470c35e084a141135cd8c
Merge: 4ed747dccf e50af8bb04
Author: Sage Weil <sage@newdream.net>
Date: Tue Nov 28 14:48:08 2017 -0600
Merge pull request #19216 from liewegas/wip-p2p-app-warning-luminous
qa/suites/upgrade/jewel-x/point-to-point: fix suppression of pool app warning
commit e50af8bb04bd37bd0e28ddc7935e29bcdc106b05
Author: Sage Weil <sage@redhat.com>
Date: Tue Nov 28 14:46:02 2017 -0600
qa/suites/upgrade/jewel-x/point-to-point: fix suppression of pool app warning
It's generated on the mgr.
Signed-off-by: Sage Weil <sage@redhat.com>
commit 4ed747dccfaae15a9a807d3f3db2fc44b8b21fda
Merge: acb02717f6 6e1e33a88d
Author: Kefu Chai <tchaikov@gmail.com>
Date: Tue Nov 28 23:47:17 2017 +0800
Merge pull request #19205 from tchaikov/wip-22136-luminous
luminous: qa/ceph-disk: enlarge the simulated SCSI disk
Reviewed-by: Alfredo Deza <adeza@redhat.com>
commit acb02717f6e96f96d4128bbebd946238d3c79291
Merge: 55f5d7d1f1 bb03f6b20b
Author: Nathan Cutler <presnypreklad@gmail.com>
Date: Tue Nov 28 14:03:01 2017 +0100
Merge pull request #19152 from smithfarm/wip-22235-luminous
luminous: ceph-disk flake8 test fails on very old, and very new, versions of flake8
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 6e1e33a88d038d5b6a469d7fb4e52182b888010f
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Nov 28 14:42:31 2017 +0800
qa/ceph-disk: enlarge the simulated SCSI disk
100MB will be allocated for journal, and the remaining 100MB is for data
device. taking the inode into consideration, there will be approximately
87988 kB available for the activated OSD. and it will complain with a
"nearfull" state.
Fixes: http://tracker.ceph.com/issues/22136
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit b3c159e9fb8c31d0acd75f0702080f18959f672e)
commit 55f5d7d1f19696af981007c63a7503b2e5d0a5a8
Merge: 3bd0007d5f ad2393b8b8
Author: Nathan Cutler <presnypreklad@gmail.com>
Date: Tue Nov 28 03:23:09 2017 +0100
Merge pull request #19173 from smithfarm/wip-22251-luminous
luminous: build/ops: macros expanding in spec file comment
Reviewed-by: David Disseldorp <ddiss@suse.com>
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
commit 3bd0007d5f1c991a80ff80fc2b7bb983ceee971e
Merge: 7ce2013dfd 5633a58824
Author: Yuri Weinstein <yuri.weinstein@gmail.com>
Date: Mon Nov 27 10:23:57 2017 -0800
Merge pull request #19179 from ceph/wip-yuriw-BP-19175-luminous
qa/suites/rados/rest: move rest_test from qa/suites/rest/
Reviewed-by: Yuri Weinstein <yweins@redhat.com>
commit 5633a58824f6248b79af74780411746cb7518813
Author: Sage Weil <sage@redhat.com>
Date: Mon Nov 27 10:11:51 2017 -0600
qa/suites/rados/rest: move rest_test from qa/suites/rest/
...and add more health whitelists.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ddb274669ca16b95c452b256e58a2830af10c73d)
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit 7ce2013dfda6f06ac47878fd4635ac0862bfbbbd
Merge: ecec65906a 61e4de133d
Author: Yuri Weinstein <yuri.weinstein@gmail.com>
Date: Mon Nov 27 09:20:18 2017 -0800
Merge pull request #19177 from liewegas/wip-jewel-x-p2p
qa/suites/upgrade/jewel-x/point-to-point: skip ec tests when mons may be old
Reviewed-by: Yuri Weinstein <yweins@redhat.com>
commit 61e4de133d9fa1eddcd858152b7828515e0b280f
Author: Sage Weil <sage@redhat.com>
Date: Mon Nov 27 10:28:16 2017 -0600
qa/suites/upgrade/jewel-x/point-to-point: skip ec tests when mons may be old
Early point release mons don't handle legacy ruleset-* ec profiles, new
ones do. Skip the ec tests that may trigger this when we are doing a
workload that races with mon upgrades.
Signed-off-by: Sage Weil <sage@redhat.com>
commit ad2393b8b8dc71a8099fa87924dcb6936a27fd84
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Thu Aug 17 09:19:38 2017 -0600
rpm: rm macros in comments
rpm expands all macros in a .spec file, even those in comments. Drop the
percent signs so rpm will not expand these.
This change silences rpmlint's warning about macros in comments.
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
(cherry picked from commit 5c1e9f55d3a144c93c9c18b031db3cdc320561ef)
commit bb03f6b20b4963ad5d51c1c6f45f19bc7eaf220b
Author: Nathan Cutler <ncutler@suse.com>
Date: Tue Nov 21 11:36:02 2017 +0100
tests: ceph-disk: ignore E722 in flake8 test
Very old, and very new, versions of flake8 treat E722 as an error:
flake8 runtests: commands[0] | flake8 --ignore=H105,H405,E127 ceph_disk tests
ceph_disk/main.py:1575:9: E722 do not use bare except'
ceph_disk/main.py:1582:9: E722 do not use bare except'
ceph_disk/main.py:3252:5: E722 do not use bare except'
ceph_disk/main.py:3288:21: E722 do not use bare except'
ceph_disk/main.py:3296:17: E722 do not use bare except'
ceph_disk/main.py:4358:5: E722 do not use bare except'
tests/test_main.py:26:1: E722 do not use bare except'
ERROR: InvocationError: '/opt/j/ws/mkck/src/ceph-disk/.tox/flake8/bin/flake8 --ignore=H105,H405,E127 ceph_disk tests'
Fixes: http://tracker.ceph.com/issues/22207
References: https://gitlab.com/pycqa/flake8/issues/361
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit 3600cd7bfdca76485d2998c5da1a0dd25816d1ce)
commit ecec65906af60c16fc99f669c1591bb821305a64
Merge: 613634c1da aac7a85a3a
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 22 13:36:58 2017 +0800
Merge pull request #19080 from smithfarm/wip-22212-luminous
luminous: tests: ceph-disk: silence deprecation warnings
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit aac7a85a3ae9cce38987ce3f51be4c6cb63c8dd2
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Nov 21 21:47:30 2017 +0800
qa/workunits: silence py warnings for ceph-disk tests
ceph-disk now prints "depreacted" warning message when it starts. but
the tests parses its stdout and stderr for a json string. so we need to
silence the warnings for the tests.
Fixes: http://tracker.ceph.com/issues/22154
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit d44334f31704487ec3574738e75145872d9932cf)
commit 613634c1da4cea975d6e1da7d258600c454c9400
Merge: 5a469c0376 94839deb2f
Author: Andrew Schoen <andrew.schoen@gmail.com>
Date: Mon Nov 20 09:24:41 2017 -0600
Merge pull request #18989 from ceph/wip-luminous-rm22154
luminous: ceph-disk create deprecation warnings
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
commit 5a469c0376e299e5c60d7d9b3c16640515b11f26
Merge: 3532f233f6 37d2726751
Author: Kefu Chai <tchaikov@gmail.com>
Date: Mon Nov 20 22:55:06 2017 +0800
Merge pull request #19025 from tchaikov/wip-pr-19024-luminous
luminous: qa/tasks: prolong revive_osd() timeout to 6 min
Reviewed-by: Sage Weil <sage@redhat.com>
commit 3532f233f691ac22c32c4533a677b95d0e4c6273
Merge: 5e519aebe7 0f7332e925
Author: Sage Weil <sage@newdream.net>
Date: Mon Nov 20 07:23:30 2017 -0600
Merge pull request #18945 from liewegas/wip-22128
mon/OSDMonitor: add option to fix up ruleset-* to crush-* for ec profiles
Reviewed-by: Joao Eduardo Luis <joao@suse.de>
commit 37d2726751c05a0b5a3c1b739bce49995cc72445
Author: Kefu Chai <kchai@redhat.com>
Date: Mon Nov 20 13:40:56 2017 +0800
qa/tasks: prolong revive_osd() timeout to 6 min
see also #17902
Fixes: http://tracker.ceph.com/issues/21474
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 749bbda07522114b99b17a4a01cfcf019520285d)
commit 5e519aebe718bd16d52e974beba8e3bf97b04ae8
Merge: 9abf091336 96c981ce6b
Author: Sage Weil <sage@newdream.net>
Date: Sat Nov 18 20:34:50 2017 -0600
Merge pull request #18983 from pdvian/wip-22150-luminous
luminous: os/bluestore: fix SharedBlob unregistration
commit 0f7332e9259b320777770f0312233a31103bfeda
Author: Sage Weil <sage@redhat.com>
Date: Wed Nov 15 08:55:33 2017 -0600
mon/OSDMonitor: add option to fix up ruleset-* to crush-* for ec profiles
The jewel->luminous upgrade test will fail if we finish the upgrade while
a workload setting old-style ec profiles is running. Add option to
automatically fix them up. Warn to the cluster log when this happens.
For now, enable this option to ease upgrades and whitelist the warning.
Only include this option in luminous so that we implicitly sunset this
compatibility kludge immediately.
Fixes: http://tracker.ceph.com/issues/22128
Signed-off-by: Sage Weil <sage@redhat.com>
commit 9abf0913367a1ad8cf076c79bf3adace8c526853
Merge: c0e032c153 975229c367
Author: Sage Weil <sage@newdream.net>
Date: Sat Nov 18 09:44:46 2017 -0600
Merge pull request #18751 from shinobu-x/wip-21701-luminous
luminous: ceph-kvstore-tool does not call bluestore's umount when exit
commit c0e032c153a36c00109970395fe42cc2013fdb9a
Merge: 81b768ab03 4d4aee14a4
Author: Sage Weil <sage@newdream.net>
Date: Sat Nov 18 09:44:19 2017 -0600
Merge pull request #18750 from shinobu-x/wip-21702-luminous
luminous: BlueStore::umount will crash when the BlueStore is opened by start_kv_only()
commit 81b768ab036e66d3f4d36ddc31febc4bbcb143fb
Merge: f02d7109c5 0fe463b5e1
Author: Sage Weil <sage@newdream.net>
Date: Sat Nov 18 09:41:37 2017 -0600
Merge pull request #18860 from liewegas/wip-filestore-rocksdb-compression-luminous
luminous: os/filestore: disable rocksdb compression
Reviewed-by: Douglas Fuller <dfuller@redhat.com>
commit 94839deb2f32e8059206ad48ad711dd2baf2fef5
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 17 13:10:29 2017 -0500
doc/ceph-volume create a migration section to help users coming from ceph-disk
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit e433efe51481ec63bc9753a0f446fe2b12b5bb6a)
commit b8bf0d047868054135592188c7ebe186181310c5
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 17 13:01:14 2017 -0500
ceph-disk add deprecation warnings in favor of ceph-volume
Signed-off-by: Alfredo Deza <adeza@redhat.com>
Fixes: http://tracker.ceph.com/issues/22154
(cherry picked from commit c9da92e1ec8d60dc40d86630dab95e4fec4d0d30)
commit f02d7109c5f00cd2165082c4336b3623352be7c5
Merge: 9515da2777 ebbe033936
Author: Alfredo Deza <alfredo@deza.pe>
Date: Fri Nov 17 12:19:36 2017 -0500
Merge pull request #18973 from ceph/backport-18924
luminous: ceph-volume: allow using a device or partition for `lvm --data`
Reviewed-by: Alfredo Deza <adeza@redhat.com>
commit ebbe033936c424a0f6ba2d102fc271d0b0e9578d
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 17 08:41:24 2017 -0500
ceph-volume tests.devices.lvm prepare isn't bluestore specific anymore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 2f40dbac386be631ac97e2450117c218d9bf02f9)
commit 96c981ce6b00520a94385228bcf70a90c5ae8ff9
Author: Sage Weil <sage@redhat.com>
Date: Tue Nov 7 22:05:10 2017 -0600
os/bluestore: fix SharedBlob unregistration
We use the SharedBlobSet remove() in three cases:
- from SharedBlob::put(), we try to remove ourselves from the set, but
have to deal with a racing lookup, so the removal is conditional on
nref still being 0.
- from split_cache(), we move the SharedBlob to another collection
- from make_blob_unshared(), we remove the entry when we clear the sbid.
The problem is that the condtiional remove() (for the first case) was being
used for all three cases, and in the second two cases nref is always != 0,
so it doesn't actually happen. This can lead to a crash during cache
shutdown.
Fix by making two variants: remove() that is unconditional, and
try_remove() that is conditional.
Set the sb->coll pointer after because remove() asserts the parent matches
where we are unregistering.
Fixes: http://tracker.ceph.com/issues/22039
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 78a465b90b1c5187cfec9f25eb58b3bf617ca39b)
commit 9515da277737ad81d48f5c032f627f3cb919f746
Merge: 5c9b93dadd d19b740359
Author: Josh Durgin <jdurgin@redhat.com>
Date: Thu Nov 16 13:40:32 2017 -0800
Merge pull request #18957 from tchaikov/wip-pr-18808-luminous
luminous: qa: do not wait for down/out osd for pg convergence
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit ad4668b65cb3e32f1b3afc5ec767004f166856e6
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Nov 16 08:50:34 2017 -0500
ceph-volume tests.functional create a common playbook directory for setups
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit def89a411f2f4bdb23c4bbd3b71c1310a564f07d)
commit d25a04114c98aa1449809248e8b87ecd1124af3d
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 11:01:48 2017 -0500
doc/ceph-volume lvm prepare can accept devices for --data
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 9bfab348b2ff59c0b939a19c22d4a06bc5236b96)
Conflicts:
doc/ceph-volume/lvm/prepare.rst
commit 7f7d62a1848e297153f88a8e476585596114e1d5
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 08:42:27 2017 -0500
ceph-volume tests.functional add setup playbook for xenial filestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 210c1e1d4770dde47f0008cadaaec35deb729272)
commit 210e1b3bf507654167349411c5cc85641b1d8680
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 08:42:02 2017 -0500
ceph-volume tests.functional add data device and journal to xenial filestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d0754e25177b9bf3bce695b7991e99f49312d91c)
commit 211895ac0906747bc448b2326ae5c62f3e6ee4a1
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 08:40:37 2017 -0500
ceph-volume tests.functional add setup playbook for xenial bluestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d0f5623963f58fc4d5727a79d8377b2191b0f9dc)
commit 398a1365a12de95bc9810b89a9aeab9040bb07b8
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 08:40:18 2017 -0500
ceph-volume tests.functional add raw data device to xenial bluestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit fbcfb522981cd4039cfe617609c5f160f0b05c52)
commit 0d082d0dbb43cc418bf0786b1f708968050ae114
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 08:38:23 2017 -0500
ceph-volume tests.functional tox.ini addition of setup playbook for partitions
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d074b534643dde37d732b34c36c5089e41848d6d)
commit a167556eaacb444c41299cd28f1efd4327da4747
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 08:37:17 2017 -0500
ceph-volume tests.functional add raw data device to centos7 bluestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 2d416a849ad7b33dde1386da09c81249e6c2299d)
commit 12e180a84f1456dfe7ab940255c6c47ee8ebddf0
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 08:36:36 2017 -0500
ceph-volume tests.functional add setup playbook for centos7 bluestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 161fd373d2f11f64466927e5d615d039822e2ab5)
commit f06ea77590336d439235bd9acc6028b854e8c181
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 08:35:59 2017 -0500
ceph-volume tests.functional add setup playbook for centos7 filestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit b3bb147298f8816651f231f77bc91a3a1b50d96f)
commit f10bfbeefe06e59629a4e9d904e8035bf060920b
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 14 08:35:42 2017 -0500
ceph-volume tests.functional add raw device to centos7 filestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit c5b8d72570b24dfd0d2f8c1cc8f13812c6c11a69)
commit 968006a271dc82f3dcada4e5de93ecb0e23238c6
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 13 14:29:32 2017 -0500
ceph-volume lvm.common update --data flag to reflect device acceptance
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 4df7ea8d83cfa02c8864d043b5d5fb2f5409dc96)
commit c8f64af027689a1fb1bb916e6d479519a12b764e
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 13 14:28:32 2017 -0500
ceph-volume lvm.prepare add example of raw device or partition in help menu
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 6fb8d566aac0e0fa081e6500852855ad177f784f)
commit f930934159edd947024eef1a33e15a30467cb332
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 13 14:26:44 2017 -0500
ceph-volume lvm.prepare allow partition or raw device as input for --data in filestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 89236ee3f81e50b9b059c10f6d6e7ce048c06474)
commit d19b7403595bef1c3ba73f4a5fb83f5ef2dd3f81
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Nov 8 14:13:54 2017 +0800
qa: do not wait for down/out osd for pg convergence
that osd is not invovlved in the PG state changes.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 7f549af459b40c5bb51a24a3a443ed7a7116915e)
commit 5c9b93dadd8f05b1a146e3a19f7480ae4019815f
Merge: cc41c393b3 2f380b32da
Author: Yuri Weinstein <yuri.weinstein@gmail.com>
Date: Tue Nov 14 08:38:35 2017 -0800
Merge pull request #18908 from smithfarm/wip-luminous-fixup
qa/suites/fs/basic_functional/clusters: more osds
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
commit cc41c393b3a76a5f1a18cbda45acc01e2d6fdfc2
Merge: d69f42e1af 806e49aadb
Author: Alfredo Deza <alfredo@deza.pe>
Date: Tue Nov 14 11:06:00 2017 -0500
Merge pull request #18910 from ceph/backport-18882
luminous: ceph-volume: add functional tests for simple, rearrange lvm tests
Reviewed-by: Alfredo Deza <adeza@redhat.com>
commit 806e49aadb0d832a4bf46fcd681d889d93158232
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 13:18:07 2017 -0500
ceph-volume tests.functional update lvm/tox.ini paths after moving it
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 7d787849556788961155534039886aedfcdb2a88)
commit 724ad01afc0af79fc7f808aae819ff9dc0971b33
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 13:14:55 2017 -0500
ceph-volume tests.functional move top level tox.ini to lvm/tox.ini
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 9b14b96d3bf6c8326cfc91e657161ffab3b8c2b4)
commit 754c237597cc91af4bfe7199f891a9c2a6938ad4
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 13:12:51 2017 -0500
ceph-volume tests.functional move xenial to lvm/xenial
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 00d576d266b63080129e58482bd6fba62534ee97)
commit 762c5f1cc10266d5efa62628c0686b3f8d696bff
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 13:11:58 2017 -0500
ceph-volume tests.functional lvm move xenial/bluestore to xenial/bluestore/create
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d5ec7d1c58141f4593ed9d1e4501c9f3be6d9755)
commit 5b1bde338fbb1e689e3e1305e91f5450ef175edd
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 13:09:48 2017 -0500
ceph-volume tests.functional lvm move xenial/create to xenial/filestore/create
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit c504b8781b3a31475280a032cd75c9e1eaa6d8c9)
commit 5a00604f27acad8cb368495f841ce76cce383414
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 13:02:05 2017 -0500
ceph-volume tests.functional move centos7 to lvm/centos7
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit dc1da3fd0aeafe59a82343fb16fe1fcea66ae3eb)
commit 57d64a86d0758d174c1ca0b432a43194cbe6162c
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 12:58:27 2017 -0500
ceph-volume tests.functional lvm move bluestore to bluestore/create
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit aec1a80a3e1e61720f494980711808bfee783a03)
commit 94a9f3f43d9c3a2ac08be4025e65c47c3767d2c1
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 12:54:01 2017 -0500
ceph-volume tests.functional lvm move create to filestore/create
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d6905f3f620d4c733835e21a275d4127cdcb769b)
commit bc134c20a5409bd6f966d205dd7ba850ff28fc91
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 12:42:19 2017 -0500
ceph-volume tests.functional simple xenial bluestore scenario
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 8f0fdc975951a2a8af31bbaa3abd2b5d807a0add)
commit 004efbfbe4d1eb56190e5e7497a9613776494fae
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 12:40:18 2017 -0500
ceph-volume tests.functional simple xenial filestore scenario
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 7452e1b6f9b82c2ef62bae19c632e81039638ba5)
commit b51744f26f3c6db4d2927a2ac1733f0b80e44010
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 12:30:05 2017 -0500
ceph-volume tests.functional simple tox.ini update for xenial scnearios
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 76ed6a8f2ecab7c58def0453ffed9d8a997e0d5b)
commit 19983b9ba5ed45c8e6fac967fb702fd586132c35
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 09:12:09 2017 -0500
ceph-volume tests.functional simple tox.ini addition for bluestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 02154e445b87a8fdcabbd699a2532833764140b5)
commit 040e7913b028c54aeb9acd854610f6a4f112465d
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 10 09:11:39 2017 -0500
ceph-volume tests.functional add initial files for simple bluestore activate
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 3320f28e799e1d77d1ea69be9ced603bb69cc5fa)
commit 23f6c3e779c743bc1302d1ad542201434ff9431f
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Nov 9 16:55:30 2017 -0500
ceph-volume tests.functional add initial files for simple filestore activate
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 9d18bae52ad77be0e64ef105c33a019446b6f1cf)
commit 8f425b678f6c947a8c83f121be4fa46151fd6664
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Nov 9 16:07:53 2017 -0500
ceph-volume tests.functional create a specific test.yml for simple activate
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 37bde0d3b37a87c456438b0edc0cb9e2dcd39120)
commit 6f1fff00f5210840fe143aac7506ce3773600f05
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Nov 9 15:40:51 2017 -0500
ceph-volume tests.functional create a separate tox.ini environ for
`simple`
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 89757ad12e6938fe2bf997e0f21a5ffac082e30b)
commit d69f42e1afaf8759de1544b3d381671a01cfb55a
Merge: a7c8c8101d 90976ef6b9
Author: Andrew Schoen <andrew.schoen@gmail.com>
Date: Mon Nov 13 10:55:24 2017 -0600
Merge pull request #18907 from ceph/luminous-wip-bz1498200
luminous: ceph-volume: support GPT and other deployed OSDs
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
commit 2f380b32da1b6294b6b0f60622b09b6517f9afdd
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 9 13:44:51 2017 -0500
qa/suites/fs/basic_functional/clusters: more osds
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 998122c324047c96c16d60d7fbb3a21f4dd63c56)
commit 90976ef6b9815b68df620aac46f2c04117766f99
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Nov 9 10:50:43 2017 -0500
doc/ceph-volume include the top level systemd.rst in the toctree
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit e2f281c115c15e820588b478214c3600ff97470f)
commit 9a8a00d3a17f03d2b1b3a92662f7288f7d18fe7a
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Nov 9 06:49:49 2017 -0500
ceph-volume simple.activate add --file in help docstring as an example
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit ff60592897b63581ebce372b3f6ffa0f10d2fe5a)
commit 34194b8d788d91b4f7c969be5f0099a279bc61bf
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Nov 9 06:46:59 2017 -0500
doc/ceph-volume activate has a --file option for json files
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 29913951b71d68b3f3f9b9fa076f309c0252da36)
commit 582262b57bcc959d1698636000d7a804cab9c611
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 17:23:41 2017 -0500
ceph-volume simple.activate enable the osd when activating
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 366fb8c43af54e9c92f4e95d9a0582c1fc378dbc)
commit a32e72cdb0efb55d185488fb02f44055f9d3dffe
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 15:25:05 2017 -0500
ceph-volume systemd.systemctl always force symlink when masking ceph-disk
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit cb470622fd4f13fafe9cb2f38fb25fd1fbd8d89c)
commit 3af40c35f3cf42be722f9d905cb2d0ba2baf1b1b
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 15:21:10 2017 -0500
ceph-volume simple.activate use the contents of the JSON file, fallback to CLI args
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit c954c913395557bcd89f25b0496c4428e6109494)
commit 161e9657d985aa4ceffad2f56e18edd78ca2a81f
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 14:50:16 2017 -0500
ceph-volume simple.activate only check for missing id and uuid if file is not passed in
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 826c2709dd5ab0fbf6eb419a735d15e718a7712c)
commit 585072afc814c6ac87fd64f15c561c51c1e55070
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 14:28:12 2017 -0500
ceph-volume simple.activate allow to pass a file to activate
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit bd3498f3eabf07637cb936e6c036d1d8d75415b0)
commit 01266c193588066425b44fea4c1729692f999692
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 12:49:31 2017 -0500
ceph-volume tests.util add tests for OSDPath validator
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d7fc26ab4e2cebe485647510119ea300d47db184)
commit 56fdc1c973b41d855563e459d23682990e8940bf
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 11:56:36 2017 -0500
ceph-volume systemd.systemctl masking ceph-disk needs to be done by directly symlinking to dev/null
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit cda47c247b929401a7dce3f1ef70371ea80eeceb)
commit 0a0c289de605b6b74b762b2a0a4ed0f3a04de8ad
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 09:47:12 2017 -0500
ceph-volume systemd.systemctl masking needs to use systemctl mask, not disable
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit bad2f62b34243addc854abf691854b886aecdc19)
commit 32913844b36285aae40f9786e5cb035bfca22f2d
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 09:46:20 2017 -0500
ceph-volume simple.scan if mounted device cannot be detected, raise an error
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 6b2fc4c0ba63e4bd51774ea09cc6542738128621)
commit fe73229fb123fdb45f4ef743d8584f6748318868
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 09:17:52 2017 -0500
ceph-volume simple.activate remove str formatting from logging
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 230faf04a185c0b95c8ad497ddd5501dc70c1409)
commit 50c19ef335305215a636595506eda1b5d16f3687
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 09:11:41 2017 -0500
ceph-volume util.arg_validators always convert the path to an absolute path
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit b5acf4d720b5395ae4ccbf007c94c99610e40089)
commit d27c35c972abba2b230e1cff1fa35437500dcdf6
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 09:10:47 2017 -0500
ceph-volume util.arg_validators require superuser privs to check paths
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 376dad74159532f901d0044f95330fc02f93bd25)
commit 0832ddc52496cd81aa7c2e1e85252b50d2f33044
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Nov 8 09:09:06 2017 -0500
ceph-volume simple.scan skip directories to scan
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 6bc1e9ee0b8d7bc475f218ae3dd8c6c7fe5a983a)
commit d6071cb75ba1a702c551d073f032336560654e97
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 15:08:13 2017 -0500
ceph-volume tests.devices.simple add activate CLI unit tests
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit cde0d517563b537319d06c758b93e421cdabe59d)
commit 2bff2e7fdbba37280791d3bc9a363e47aa26ab53
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 15:07:44 2017 -0500
ceph-volume tests.devices add help menu unit test
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit dcc7c72196a327f9501045f025761b228b5379df)
commit ec8624005082913faf89a9b001d4abeb7c27a589
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 13:42:45 2017 -0500
doc/ceph-volume systemd describe the interaction with 'simple' activation
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 7071eeaeb5d77c6ab6e7a24f1b630cb0cf8a8e7c)
commit aacd22681d1e7f5a604025b2019b195041151b3f
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 13:42:11 2017 -0500
doc/ceph-volume index initial description of simple
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f9ee51f9feadb6d0f91b76a2b2ba8a24f1523a77)
commit e9527e65d5da503107e21dc5b695f287c8529049
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 13:41:12 2017 -0500
doc/ceph-volume lvm add back a system.rst with just the lvm portion of it
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f872cd29b93e02b77172f7b382c4746af9ba845d)
commit ef027a737989dca83c0e5810935bad2280146356
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 13:40:44 2017 -0500
doc/ceph-volume make systemd.rst generic enough to be related to other sub-commands
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit cf4e32a0dc3ab4f55fe7857b51afbc63a638ec66)
commit a83b746eaeece5896770b40949804316c29f6bc1
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 13:39:52 2017 -0500
doc/ceph-volume move lvm/systemd.rst to top level, making it generic
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 0672dd608d4601a936d703ea5ba6edc14a161d8e)
commit bc492490e15d3ea91654cd4dabfb3115f835b9ac
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 11:52:17 2017 -0500
doc/ceph-volume simple.activate initial take on documenting activation
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 80c796bd1c43125f5678710b494b03c486b3fce6)
commit 6301c83097729749fc712846954a3c0f89647f07
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 08:35:57 2017 -0500
doc/ceph-volume lvm update systemd anchor to reflect lvm parent
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 53420a4756a5e4fa6b0f2aa74d6ad1e2919f1954)
commit 5fe4122bcfdc9c8e4f69079e45c907148eccdb7d
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 08:35:43 2017 -0500
doc/ceph-volume lvm.activate update systemd anchor to reflect lvm parent
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 7fc0e7ebcae06f28f6ed08c0db14075ada34e8a7)
commit c98886a99914748d73d0f695808a74850bc9e66e
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Nov 7 07:56:35 2017 -0500
doc/ceph-volume lvm.systemd update anchor to reflect lvm parent
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 01a8b7e6a45f6fd85bf53d3cab15acc81d0bae1b)
commit 68f89b52d608c3946be9ad92e733fe9472bd35f0
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 6 15:53:39 2017 -0500
doc/ceph-volume document the scan functionality
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit b30a3eaf17a20a1273724b2cfcaf0a7b36eb08f6)
commit 19d0ca22908fcd3fbe83d39c1b708c5818b34f47
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 6 14:17:20 2017 -0500
doc/ceph-volume update the index to include the simple sub command and files
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d7f85c64e4b409ea95e30ccb5a1388278843c478)
commit 61c90bacc562e1e2d65b767dca97b08973832131
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 6 11:15:28 2017 -0500
ceph-volume tests.simple add tests for trigger parsing from systemd
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 0e2620234e82abf85274b3c08b39cdf76430ce48)
commit 80ae5b1fb024827cceb0594a60e3361fd4a1e7a3
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 6 10:36:14 2017 -0500
ceph-volume tests.simple create etc_path scan tests
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 90ebc4e1ffa7b3273787d708bb2c05180530b5d1)
commit b9ad514b45e0233a17834e19a232bca4467a8140
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 6 10:06:02 2017 -0500
ceph-volume tests.simple add checks for get_contents in scan
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 647bfe0883d2bb05b4ff7f99894075df3e73d2d5)
commit c9308af91a322f92ab7d1f45a423b2a94f64c429
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 6 09:43:58 2017 -0500
ceph-volume tests.util add binary verification tests
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 827aa57557561355dd4068e26801b5f1ed62f295)
commit c94f52dffe0ea4b48c9fb316e8c53fd06a81b870
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Nov 6 09:42:53 2017 -0500
ceph-volume tests create a fixture for temporary files
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 23da3445718c01f71e2a5748fb4fc1c5fdc85ca5)
commit bbb25df806609083efd377befb44d6d3bef3338b
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:36:38 2017 -0400
ceph-volume systemd.systemctl helpers to mask ceph-disk
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 3da23fa4caea8c9dc91da3eee5b5acc1860bf242)
commit c971321430fd1a4de7c6fa2687ecf35a5b3ec71c
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:36:13 2017 -0400
ceph-volume devices include simple at the module import level
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 027a09b3ea6bac416ff632a69a53393badd51708)
commit 1615d947db50ac37e4ef554acc6debe8fd870962
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:35:45 2017 -0400
ceph-volume main add simple to the sub-commands
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 94579c77ffe4d5ba2d457ad3981a2b9ed232a90f)
commit abb77a8d9824eef3bd9c75d17f8df101440c25d1
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:35:03 2017 -0400
ceph-volume util add an OSDPath argument validator
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit a826fdf4a2ed9ff704fb823d05d3cb706ba89e38)
commit ff2a1833da385d6c0c9087e98cb2b93925b06cab
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:34:39 2017 -0400
ceph-volume simple.system add a helper to check if a file is a binary or not
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 7bf8676045b4f966b60cdadf4d5af1c89fca69ae)
commit 84d2a15ce9f1a045ed67f79fa7d1dc99dc1e4b8f
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:34:06 2017 -0400
ceph-volume util.system add a context manager for temporary mounting/unmounting
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 6394cdd41943e3cda6867ecc59ab835cb2151707)
commit 3d8ff08c2d46017749e7ecaa4e4d49dda4d55d74
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:32:42 2017 -0400
ceph-volume simple.scan initial take on directory/device scanning
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 0e54b9be621cc0969a3bbc972d1870080ccdae77)
commit d701569a6486d7ccccf14b5eadf702ad07db4821
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:27:23 2017 -0400
ceph-volume simple.trigger enable systemd triggering for simple setups
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit cbc0c0f775e96a5b0ab5d5bc141acebe4cc48373)
commit 76fe659a19a8a450b4c61f6d2b56e03dc46746ae
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:26:13 2017 -0400
ceph-volume simple.activate initial take on activation
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 5a97374605408d18e9be29bcc051ae7b913d1675)
commit 206c7551435983b4cbdaece61b737e4a1f265dbd
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:23:47 2017 -0400
ceph-volume simple add __init__ for the module
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 285e5bdd28a42d98418ca90ad0cdec6bd87849db)
commit 8a5edb9a2acc3f0b5477f09edae4f97f5ec7f5b5
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Nov 3 14:23:29 2017 -0400
ceph-volume simple create the initial, main module
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f050502425a1993ae8edfdb5785b24129cffeeba)
commit 0fe463b5e1fe62e03b498c32aa708be8c10eb0b8
Author: Sage Weil <sage@redhat.com>
Date: Thu Nov 9 20:10:59 2017 -0600
os/filestore: disable rocksdb compression
Experience working with customer escalations suggests that disabling
compression improves performance, and the storage overhead is generally
not a concern for the metadata and omap data we are storing.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit b878ead071b328e5fe7309a2368383e67679e9f7)
# Conflicts:
# src/common/options.cc
commit a7c8c8101d4b78b4d6e437620b2c1a38cd752c3f
Merge: 1071fdcf73 f56de67980
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Thu Nov 9 18:08:27 2017 +0100
Merge pull request #18801 from linuxbox2/luminous-rgw-readdir2
rgw_file: implement variant offset readdir processing -- pullup 2 of 2
Reviewed-By: Abhishek Lekshmanan <abhishek@suse.com>
commit 1071fdcf73faa387d0df18489ab7b0359a0c0afb
Merge: 4db48ca836 8cacd2a568
Author: Kefu Chai <tchaikov@gmail.com>
Date: Thu Nov 9 22:36:27 2017 +0800
Merge pull request #18734 from shinobu-x/wip-21648-luminous
luminous: mgr[zabbix] float division by zero
Reviewed-by: Wido den Hollander <wido@widodh.nl>
commit 4db48ca8369d1dae021131c93a1c8484512aa99b
Merge: 9cd31820f2 d858570755
Author: Kefu Chai <tchaikov@gmail.com>
Date: Thu Nov 9 22:35:44 2017 +0800
Merge pull request #18851 from tchaikov/wip-18759-pr-luminous
mon/LogMonitor: "log last" should return up to n entries
Reviewed-by: John Spray <john.spray@redhat.com>
commit 9cd31820f22d746e680f4ce3b0c780157d24194f
Merge: 60ff142c8e 8a7f983f33
Author: Kefu Chai <tchaikov@gmail.com>
Date: Thu Nov 9 22:27:47 2017 +0800
Merge pull request #18741 from shinobu-x/wip-22019-luminous
luminous: tests: "ceph osd create" is not idempotent
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit d858570755d88e9a3708952827a99cc24f09f9ab
Author: Kefu Chai <kchai@redhat.com>
Date: Mon Nov 6 16:01:23 2017 +0800
mon/LogMonitor: "log last" should return up to n entries
limit the # of returned entries to "num", and backoff the start iterator
by one if it reaches the rend().
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit de14103f5201da69b360d9b01e249896d59376b3)
commit f56de6798077399f472c8d2df25fb1c3ccaf8090
Author: Matt Benjamin <mbenjamin@redhat.com>
Date: Sun Oct 15 21:48:19 2017 -0400
rgw_file: implement variant offset readdir processing
Introduce new rgw_readdir2(...), which in which continues
from an arbitrary dirent name, which presumably has been
seen in a prior partial enumeration.
Add single-file unit test for READDIR cases, librgw_file_marker.cc.
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
(cherry picked from commit a2c34c597371809bd8fd40ee1fc6b8e6a5145609)
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
commit 60ff142c8e8ed8fb294f7d13a5461f070aa1ff4f
Merge: f1776bc788 7efa91c7ce
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Thu Nov 9 10:10:53 2017 +0100
Merge pull request #18800 from linuxbox2/luminous-rgw_file-fsid-mount
Luminous rgw file fsid mount -- pullup 1 of 2
Reviewed-By: Abhishek Lekshmanan <abhishek@suse.com>
commit f1776bc7887d2245e632d5a89189294a823ff3b2
Merge: 2d3c14e5b8 f7dd3be6b7
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Thu Nov 9 10:04:08 2017 +0100
Merge pull request #18767 from smithfarm/wip-22020-luminous
luminous: multisite: race between sync of bucket and bucket instance metadata
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 2d3c14e5b88c773548af61312438a69dcb454726
Merge: 2bad0ac00c 36e214c67b
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Thu Nov 9 09:56:16 2017 +0100
Merge pull request #18794 from theanalyst/wip-18709
luminous: rgw: Stale bucket index entry remains after object deletion
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 2bad0ac00c898d50c9c036400b4fba09918fb331
Merge: 9787509155 f81e6dc8ab
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 8 18:19:27 2017 +0800
Merge pull request #18726 from shinobu-x/wip-luminous-22035
luminous: Spurious ceph-mgr failovers during mon elections
Reviewed-by: John Spray <john.spray@redhat.com>
commit 9787509155df02bd9df21f389968f77d3de2d988
Merge: 0c3e9e6aab 7e08cdf539
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 8 18:18:39 2017 +0800
Merge pull request #18738 from shinobu-x/wip-21547-luminous
luminous: ceph-mgr gets process called "exe" after respawn
Reviewed-by: John Spray <john.spray@redhat.com>
commit 0c3e9e6aab5cfbd3d8e998b148e19bab30f3fa0e
Merge: 628e76fe2a ed4a692254
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 8 18:17:54 2017 +0800
Merge pull request #18727 from shinobu-x/wip-luminous-22034
luminous: key mismatch for mgr after upgrade from jewel to luminous(dev)
Reviewed-by: John Spray <john.spray@redhat.com>
commit 628e76fe2a939aae4996f43e9e9c9f7d1269e0bb
Merge: ef5264afc9 be5363a771
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 8 14:44:41 2017 +0800
Merge pull request #18723 from shinobu-x/wip-luminous-22023
luminous: osd: make stat_bytes and stat_bytes_used counters PRIO_USEFUL
commit ef5264afc992d78c1b34c8a96a43216191fa5406
Merge: a0a2d27c1e 31c56c6e6b
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 8 14:38:10 2017 +0800
Merge pull request #18737 from shinobu-x/wip-21549-luminous
luminous: the dashboard uses absolute links for filesystems and clients
Reviewed-by: John Spray <john.spray@redhat.com>
commit a0a2d27c1e40280ae214696dbf5c25567c28ba38
Merge: 947cae47db 76ab4eba99
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 8 14:37:19 2017 +0800
Merge pull request #18736 from shinobu-x/wip-21638-luminous
luminous: dashboard OSD list has servers and osds in arbitrary order
Reviewed-by: John Spray <john.spray@redhat.com>
commit 947cae47db9f5dc6d77a85834515670e20b5faf8
Merge: 2e11e6662a e14de28f9e
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 8 14:36:40 2017 +0800
Merge pull request #18728 from shinobu-x/wip-luminous-22032
luminous: dashboard barfs on nulls where it expects numbers
Reviewed-by: John Spray <john.spray@redhat.com>
commit 2e11e6662ae167276ece7a96783f02a24fc1f89f
Merge: 31afe858a9 cd18f84304
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 8 14:34:35 2017 +0800
Merge pull request #18699 from shinobu-x/wip-luminous-21538
luminous: upmap does not respect osd reweights
Reviewed-by: Sage Weil <sage@redhat.com>
commit 31afe858a96f1d5c67124b92f19cf4542df7e647
Merge: d9df23cf3a 388b5016d9
Author: Kefu Chai <tchaikov@gmail.com>
Date: Wed Nov 8 14:30:42 2017 +0800
Merge pull request #18719 from tchaikov/wip-pr-18373-luminous
luminous: osd,mon: add max-pg-per-osd limit
Reviewed-by: Sage Weil <sage@redhat.com>
commit 388b5016d981c9dd98b6505b2d75b0220cfee772
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Oct 18 16:07:50 2017 +0800
doc: fix the desc of "osd max pg per osd hard ratio"
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit d940f6220346ad21dfc4f351c57389e8430c0257)
commit 04852a400483b6459985be80a7773ef153c04b1e
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Oct 17 12:17:09 2017 +0800
mon: change "mon_pg_warn_min_per_osd" to uint64_t
* this silences a warning from -Wsign-compare.
* also switch all its user to the new-style option
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit c3cd0c13a1585388df0f17ab53c170e39c711615)
commit d9df23cf3aa002315e15db46d195be7d5657f9d0
Merge: 567c29f964 53096755c0
Author: Sage Weil <sage@newdream.net>
Date: Tue Nov 7 22:27:15 2017 -0600
Merge pull request #18702 from liewegas/wip-pr-18670-luminous
luminous: qa/tasks/thrashosds: set min_in default to 4
commit 567c29f96461901bd86d4c2d4f84f7b84342837f
Merge: 77c2b0de60 8cbb2eb937
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 22:28:58 2017 +0100
Merge pull request #18596 from smithfarm/wip-21955-luminous
luminous: tests: add EC data pool to testing
Reviewed-By: Patrick Donelly <pdonelly@redhat.com>
commit 7efa91c7ced61f0fde81adfc0540a36fafafb10d
Author: Gui Hecheng <guihecheng@cmss.chinamobile.com>
Date: Sat Jul 22 13:12:06 2017 +0800
rgw_file: introduce rgw_mount2 with a bucket name parameter
This explicitly allow mount a specified bucket in librgw.
Originally, mounting a bucket is implemented like a sub-directory
mount in nfs-ganesha FSAL_RGW with lookup_path.
With this change, we allow root_fh to points exactly to the root
of a mounted fs instance, which is a bucket or "/" rather than
always let root_fh points to "/".
Signed-off-by: Gui Hecheng <guihecheng@cmss.chinamobile.com>
(cherry picked from commit 860716e17e840df11a7e1d8b6ca0c1ee694b038b)
commit b53c3948a5e48623bb437b2d36718ce3a799aa54
Author: Gui Hecheng <guihecheng@cmss.chinamobile.com>
Date: Fri May 26 17:45:29 2017 +0800
rgw_file: new fsid for fs instance
We use an in-memory fs_inst_counter for fsid currently, but it
inevitably cause nfs automount problem across a nfs-ganesha crash.
So here we relate the fsid to the s3 uid with a hash.
With this, fsid remains consistent across all fs instances and
across crashes.
We should be able to upgrade from old fsid to new fsid with
update_fh().
Signed-off-by: Gui Hecheng <guihecheng@cmss.chinamobile.com>
(cherry picked from commit 178b547e83a2c0eeb024bbe4ca2ff8d8e7d50572)
commit 77c2b0de6042707b42d6e717e90a525fda8ead76
Merge: 14e0530f50 555746a09e
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 22:07:45 2017 +0100
Merge pull request #18628 from batrick/i21953
luminous: mds: sanitize mdsmap of removed pools
Reviewed-By: Sage Weil <sage@redhat.com>
commit 14e0530f50cd5fba46da914cdd09489f13adb2ec
Merge: bc9f508477 c2237c7c6d
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 18:33:21 2017 +0100
Merge pull request #18626 from dillaman/wip-21968
luminous: cls/journal: fixed possible infinite loop in expire_tags
Reviewed-By: Mykola Golub <trociny@gmail.com>
commit bc9f5084778607c95dcbc93ea60ce296e8e8ee88
Merge: 6bc121ddee 1aeec0c032
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 18:32:20 2017 +0100
Merge pull request #18688 from liewegas/wip-balancer-luminous
luminous: mgr/balancer: backport crush-compat work from master
Reviewed-By: John Spray <jspray@redhat.com>
commit 36e214c67b6982cc7a85e08dedc83ea61e56d6a4
Author: J. Eric Ivancich <ivancich@redhat.com>
Date: Fri Nov 3 09:15:13 2017 -0400
rgw: fix BZ 1500904, Stale bucket index entry remains after object deletion
We have a race condition:
1. RGW client #1: requests an object be deleted.
2. RGW client #1: sends a prepare op to bucket index OSD #1.
3. OSD #1: prepares the op, adding pending ops to the bucket dir entry
4. RGW client #2: sends a list bucket to OSD #1
5. RGW client #2: sees that there are pending operations on bucket
dir entry, and calls check_disk_state
6. RGW client #2: check_disk_state sees that the object still exists, so it
sends CEPH_RGW_UPDATE to bucket index OSD (#1)
7. RGW client #1: sends a delete object to object OSD (#2)
8. OSD #2: deletes the object
9. RGW client #2: sends a complete op to bucket index OSD (#1)
10. OSD #1: completes the op
11. OSD #1: receives the CEPH_RGW_UPDATE and updates the bucket index
entry, thereby **RECREATING** it
Solution implemented:
At step #5 the object's dir entry exists. If we get to beginning of
step #11 and the object's dir entry no longer exists, we know that the
dir entry was just actively being modified, and ignore the
CEPH_RGW_UPDATE operation, thereby NOT recreating it.
Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit b33f529e79b74314a2030231e1308ee225717743)
commit 6bc121ddeec4b12c91ae5ac982ad113bda14d2bf
Merge: ee27efad8b f26f3dc842
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 18:21:41 2017 +0100
Merge pull request #18569 from kmroz/wip-21939-luminous
luminous: list bucket which enable versioning get wrong result when user marker
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit ee27efad8ba4d751828dd73012de12ab16c2f78d
Merge: 601249c6aa a340f1acab
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 18:21:03 2017 +0100
Merge pull request #18591 from cbodley/wip-21938
luminous: rgw: fix error handling in ListBucketIndexesCR
(approved in irc & standups)
Reviewed-By: Abhishek Lekshmanan <abhishek@suse.com>
commit 601249c6aa154d0f9e53332480ac45b5730fb503
Merge: d2226d5adb 15a52ba832
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 18:17:57 2017 +0100
Merge pull request #18539 from linuxbox2/luminous-baixuyeyu-refcnt
luminous: baixuyeyu refcnt
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit d2226d5adbb4752e0180dfa3406e188bd3b37d6e
Merge: 925c68e9e9 902d467a0e
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 18:15:38 2017 +0100
Merge pull request #18516 from ceph/wip-bp-pr18193-luminous
qa/rgw: ignore errors from 'pool application enable'
Reviewed-By: Nathan Cutler <ncutler@suse.com>
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 925c68e9e9538a97f98bef1e5a884d31811aa229
Merge: 2f35da4e60 add8c3db19
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 18:14:21 2017 +0100
Merge pull request #18599 from linuxbox2/lum-rgwfile-21940
rgw_file: set s->obj_size from bytes_written
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 2f35da4e60c2b5fd2f45348587a5b175868d04d7
Merge: b2961db15d 26c523c3be
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Tue Nov 7 13:07:09 2017 +0100
Merge pull request #18566 from kmroz/wip-21914-luminous
luminous: [rbd-mirror] peer cluster connections should filter out command line optionals
Reviewed-By: Jason Dillaman <jdillaman@redhat.com>
commit 8e3227ad51fa4909a39eb0d05798b48200a03fc5
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Oct 17 12:10:55 2017 +0800
osd,mon: change "mon_max_pg_per_osd" to uint64_t
as it should never be a negative number. if this option is disabled,
it's 0.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 0d68197376b6cf48c6679d77ec9f28f279f3da68)
commit ae7823187186310bac117fac437d991398cffd5e
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Sep 26 15:54:14 2017 +0800
osd: add max-pg-per-osd limit
osd will refused to create new pgs, until its pg number is lower
than the max-pg-per-osd upper bound setting.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 4c7df944c7f28232873ba681eedce72cdb062ea5)
commit 1db44c5bef752b3a6ce59ea21548b54b77b587d2
Author: Kefu Chai <kchai@redhat.com>
Date: Sun Oct 15 20:46:28 2017 +0800
qa/tasks/ceph: add "create_rbd_pool" option
True by default, but we can opt not to create the "rbd" pool.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit ac1e13ce63a6f32ae9ab0db8196c8cc8100d94ed)
commit ed34b00566f9b1b65c09989af44ad24cdb27fbdc
Author: Kefu Chai <kchai@redhat.com>
Date: Thu Oct 12 20:07:46 2017 +0800
osd: print osd as "osd.$osdid" in log message
easier to read this way
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit c6b71ebf6b7b7c54560f331d5f8e4e24f5a60633)
commit df611786d2882e346075f5a32aef46bea3c36d47
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Sep 26 15:50:08 2017 +0800
mon/OSDMonitor: bypass checks if pg_temp.forced
originally, monitor ignores the pg_temp messages sent from replica osds,
but to re-trigger peering if replica osd's pg number drops down below
the setting, we need to bypass the checks.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit da6f7a810055a0dc9e364aca8e6efbfbc32881b3)
commit aefc9315ed6ebbf209d4496b0fb33c72dc107961
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Sep 26 15:49:11 2017 +0800
messages/MOSDPGTemp: add `forced` field
the monitor is supposed to bypass some checks if `forced` is true.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 6335bf791003e9e33b2f738ab3da06f3c626537a)
commit b2961db15dd7528019df7990f8f7690ae3a95aa9
Merge: a5899a57e4 4cac43679a
Author: Kefu Chai <tchaikov@gmail.com>
Date: Tue Nov 7 17:48:34 2017 +0800
Merge pull request #18775 from tchaikov/wip-upgrade-luminous-x
qa: add 1-ceph-install to upgrade:luminous-x
Reviewed-by: Yuri Weinstein <yweinste@redhat.com>
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
commit 4cac43679a7688cb4765728e1a75e812d0df22de
Author: Yuri Weinstein <yweinste@redhat.com>
Date: Tue Aug 29 09:50:51 2017 -0700
qa: add 1-ceph-install to upgrade:luminous-x
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
(cherry picked from commit aff342627cf77e2faae6904e626d9fb2c897f155)
commit 1aeec0c0326678bcdba7ee0a6a1204f460e3ddf7
Author: Sage Weil <sage@redhat.com>
Date: Mon Nov 6 21:17:22 2017 -0600
qa/suites/rados/thrash/d-*/*balancer*: debug osd in mgr
Signed-off-by: Sage Weil <sage@redhat.com>
commit f7dd3be6b7804035b47d2e6fb94463f0bc772bbe
Author: Casey Bodley <cbodley@redhat.com>
Date: Tue Oct 31 16:56:01 2017 -0400
rgw: remove placement_rule from cls_user_bucket_entry
placement_rule is no longer needed in cls_user_bucket_entry, because the
only time that it's needed, we can read it from the bucket instance in
RGWRados::update_containers_stats()
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 8e62e3526643da67f5af7daa687120feed469785)
commit 17cc13b87b5378bf03e2b0b52c274772ebaf08ef
Author: Casey Bodley <cbodley@redhat.com>
Date: Tue Oct 31 16:29:31 2017 -0400
rgw: remove placement_rule from rgw_link_bucket()
with the fallback in RGWRados::update_containers_stats(), we no longer
need to pass placement_rule into the cls_user_bucket_entry
this removes the dependency between bucket entrypoint metadata and its
bucket instance metadata during multisite sync
Fixes: http://tracker.ceph.com/issues/21990
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit dd5d278a4f9b080234cbb77e448208803ec0cf93)
commit 338248a8b376074c0f43b626e9f69fc59ee26487
Author: Casey Bodley <cbodley@redhat.com>
Date: Tue Oct 31 16:26:54 2017 -0400
rgw: take placement_rule from bucket info in update_containers_stats
in case the cls_user_bucket_entry doesn't contain a placement_rule, take
it from the bucket instance info
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 531195d80d5bac774d58b210307eb19b57d26345)
commit 975229c3674ce99704aef399ecfb770cc8bcf1da
Author: Chang Liu <liuchang0812@gmail.com>
Date: Tue Oct 3 00:12:43 2017 +0800
tool: ceph-kvstore-tool doesn't umount BlueStore properly
Fixes: http://tracker.ceph.com/issues/21625
Signed-off-by: Chang Liu <liuchang0812@gmail.com>
(cherry picked from commit 81e4560781c4e5a5a033bfd4a4fe4094fc27d964)
commit 4d4aee14a46ed36cdb40c0ff67193c8650b106da
Author: Chang Liu <liuchang0812@gmail.com>
Date: Tue Oct 3 00:01:43 2017 +0800
os/bluestore: make BlueStore opened by start_kv_only umountable
ceph-kvstore-tool use start_kv_only to debug the kvstore. we
will get a crash when we try to umount bluestore in kvstore-tool.
Fixes: http://tracker.ceph.com/issues/21624
Signed-off-by: Chang Liu <liuchang0812@gmail.com>
(cherry picked from commit 2754c5acad442e50d97c8daf648cfcadd38222a6)
commit 3f7eadb79f4bb0a2fe25b229c24b652e7dcbaf00
Author: Sage Weil <sage@redhat.com>
Date: Sat Nov 4 09:32:51 2017 -0500
qa/suites/rados/thrash: combine balancer with require-luminous
We can only do the balancer stuff if we've done the mkfs-time
require luminous because otherwise the mgr commands aren't proxied via
the mon.
Only do balancer for thrash tests; not for basic, monthrash, verify.
Signed-off-by: Sage Weil <sage@redhat.com>
commit 8a7f983f3322b62cb18421c088e652341125eb1f
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Nov 1 10:45:09 2017 +0800
qa: stop testing deprecated "ceph osd create"
"ceph osd create" is not idempotent, and is considered deprecated.
Fixes: http://tracker.ceph.com/issues/21993
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 371d3f40911917704f4992c48236b4c60905bf4e)
commit 7e08cdf53992570d27b47d0028c698b78908ba83
Author: John Spray <john.spray@redhat.com>
Date: Fri Sep 15 12:23:39 2017 -0400
mgr: set explicit thread name
This gets used as our process name in some situations
when respawning. This is the same as what commit 4f177bb6b
did for the MDS.
Fixes: http://tracker.ceph.com/issues/21404
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 44dce0a5664639b2ae0949cc434f1be1176d872d)
commit 31c56c6e6b50b44fef56852e58ab97520ee8614a
Author: Nick Erdmann <n@nirf.de>
Date: Tue Aug 22 16:55:11 2017 +0000
pybind/mgr/dashboard: updated favicon (old one was 404)
Signed-off-by: Nick Erdmann <n@nirf.de>
(cherry picked from commit 7a54520f5c3d4db27fe7e2a12760dcd838ba89aa)
commit d822c15af9b838dc2eea08b5f13f1dba785d776e
Author: Nick Erdmann <n@nirf.de>
Date: Tue Aug 22 08:08:45 2017 +0000
pybind/mgr/dashboard: clean up
Signed-off-by: Nick Erdmann <n@nirf.de>
(cherry picked from commit f512ac3d7712850f274cd01bb589aa9fbc8c8bef)
commit 1f47cc0b53f7e20e156f74b28502c7705ddb4700
Author: Nick Erdmann <n@nirf.de>
Date: Mon Aug 21 17:21:10 2017 +0000
pybind/mgr/dashboard: add url_prefix
This adds a configuration variable url_prefix to the dashboard that
that is prepended to all URLs so you can access the dashboard at
http://$IP:$PORT/$PREFIX/. This is necessary if you wish to use a
reverse http proxy that forwards to the dashboard under a sub-path.
Fixes: http://tracker.ceph.com/issues/20568
Signed-off-by: Nick Erdmann <n@nirf.de>
(cherry picked from commit 6d285fff0df598d66f46d38bf0a6a2cc9c0dc62f)
commit 76ab4eba99785edf01c8ff9464d997b2ec691b31
Author: John Spray <john.spray@redhat.com>
Date: Wed Sep 27 15:13:48 2017 +0100
mgr/dashboard: sort servers and OSDs in OSD list
Fixes: http://tracker.ceph.com/issues/21572
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit f409099fa25b73fe580ee4662aff51636f118ed2)
commit 8cacd2a568fca023a51f3be6e20f90b4ec08c82b
Author: John Spray <john.spray@redhat.com>
Date: Sat Sep 23 15:22:34 2017 +0100
mgr/zabbix: log exceptions with backtrace
Logging just the exception object gave a sometimes-enigmatic
single line. Let's use the logger exception method so that
we get a backtrace.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 316dcc8f0b42b85503632d5472c45c9828882f2c)
commit 06706bea9339e26d1a0ab395e9d42d391ee8c03c
Author: John Spray <john.spray@redhat.com>
Date: Sat Sep 23 15:18:18 2017 +0100
mgr/zabbix: fix div by zero
Fixes: http://tracker.ceph.com/issues/21518
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 9c02738b4ea0303f5b3cbfc4748d6791007be834)
commit a5899a57e4f077f4061fc2aa33a61377de7ac025
Merge: 52f66ef5ba 35de92b259
Author: Sage Weil <sage@newdream.net>
Date: Sat Nov 4 10:52:27 2017 -0500
Merge pull request #18673 from dzafman/wip-21833
osd: build_past_intervals_parallel: Ignore new partially created PGs
commit e14de28f9e524575c633dc98c6ebb9a234c8d1c5
Author: John Spray <john.spray@redhat.com>
Date: Wed Sep 27 14:33:58 2017 +0100
mgr/dashboard: handle null in format_number
Fixes: http://tracker.ceph.com/issues/21570
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit f43859d4dca696ba4cac224e0c7e87e4b2aad40f)
commit ed4a692254e71ce7b30566dfd6d16e8a070b3461
Author: John Spray <john.spray@redhat.com>
Date: Thu Oct 19 09:28:18 2017 -0400
mon: don't blow away bootstrap-mgr on upgrades
Fixes: http://tracker.ceph.com/issues/20950
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 57229ea2a4369518c7a16b7a09b045b7896f5a70)
commit f81e6dc8abe3535a54cbf683992c95c4eaf4e186
Author: John Spray <john.spray@redhat.com>
Date: Thu Oct 12 11:57:50 2017 +0100
mon: handle monitor lag when killing mgrs
Fixes: http://tracker.ceph.com/issues/20629
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 3c3776b30a2da3f5d2ece15c57713c0ce191c778)
commit be5363a771172d3260049b0012519b485e4d728e
Author: Yao Zongyou <yaozongyou@vip.qq.com>
Date: Tue Oct 31 13:23:49 2017 +0800
osd: make stat_bytes and stat_bytes_used counters PRIO_USEFUL
These two counters should be reported to mgr because the dashboard
plugin needing them to display osd detailed information.
Fixes: http://tracker.ceph.com/issues/21981
Signed-off-by: Yao Zongyou <yaozongyou@vip.qq.com>
(cherry picked from commit 40bb3c5f05b6f2a917a7e400f88677f41a9629a3)
commit 77766609d808a974aafdff359dadc46af5bd26f4
Author: Sage Weil <sage@redhat.com>
Date: Thu Nov 2 16:18:03 2017 -0500
pybind/mgr/mgr_module: fix calc_pg_upmaps
This was fixed in the object cleanup.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ed442164d52c803916ca43a5ed76047089d3da66)
commit 56c27e6be813e6a8b972733468eb099596ff7381
Author: Sage Weil <sage@redhat.com>
Date: Thu Nov 2 16:11:26 2017 -0500
mgr/balancer: enable module by default
It will still be "off".
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 26710f0a9b176289237a52e5fa5894342ad163dc)
commit 53096755c02d953867d0fff3e08bc3a62d872ec6
Author: Sage Weil <sage@redhat.com>
Date: Wed Nov 1 08:31:31 2017 -0500
qa/tasks/thrashosds: set min_in default to 4
We have EC tests with k=2,m=2, so we need a min of 4.
Fixes: http://tracker.ceph.com/issues/21997
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d21809b14ea58dc1f44df844e407ebab5a315062)
commit 52f66ef5ba291fccc265fd7a34a4eae201667153
Merge: 1b00bdfeef e58656005d
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Nov 3 10:26:44 2017 +0100
Merge pull request #18446 from theanalyst/wip-21854-luminous
luminous: rgw_file: explicit NFSv3 open() emulation
Reviewed-By: Abhishek Lekshmanan <abhishek@suse.com>
Reviewed-By: Matt Benjamin <mbenjamin@redhat.com>
commit cd18f84304124e1341516194fb73c8bde48db445
Author: Theofilos Mouratidis <t.mour@cern.ch>
Date: Mon Sep 25 16:17:47 2017 +0200
osd: upmap should respect osd reweights
Modify OSDMap::calc_pg_upmaps to take the osd reweight into account when
computing the size of each OSD.
Signed-off-by: Theofilos Mouratidis <t.mour@cern.ch>
Fixes: http://tracker.ceph.com/issues/21538
(cherry picked from commit b8561cefa705d31954d47723125376a3c087f4f4)
commit 03a11694feee0d78dcc1f296a77200635a651707
Author: Theofilos Mouratidis <t.mour@cern.ch>
Date: Mon Sep 25 16:15:45 2017 +0200
test/cli/osdmaptool: test upmap with an out osd
Test upmap with one osd marked out.
Signed-off-by: Theofilos Mouratidis <t.mour@cern.ch>
(cherry picked from commit ff9e024a59d87eda6cbf3be9464ee7ae6f1140d7)
commit 2c7972d3a4928cec039e8637048c26e94c6278d2
Author: Theofilos Mouratidis <t.mour@cern.ch>
Date: Mon Sep 25 16:14:09 2017 +0200
osdmaptool: add --mark-out option
Add a --mark-down <id> option which sets the reweight for a given
osd id to 0.0.
Signed-off-by: Theofilos Mouratidis <t.mour@cern.ch>
(cherry picked from commit bd75a78187143f750f90fb937984eaa6333159d3)
commit 1b00bdfeefd7b724f14634383ae4baaf7a42c996
Merge: c6052f2706 92d0545781
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Thu Nov 2 21:23:50 2017 +0100
Merge pull request #18439 from theanalyst/wip-21655-luminous
luminous: expose --sync-stats via admin api
Reviewed-By: Abhishek Lekshmanan <abhishek@suse.com>
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 282e21bd07c3dfb5d18f338abd3ced4f4eeaa247
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 27 12:53:52 2017 -0500
mgr/balancer: simplify pool_info tracking
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit a59a1df85382d559362c51bb45284b61caaca1d4)
commit e2346f539f477a4b7a4685828876eb89c84ad00d
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 25 22:47:02 2017 -0500
mgr/balancer: less verbose on 'eval' by default; add 'eval-verbose'
The verbose output is helpful for debugging and understanding what is
being measured, but most of the time all that you care about is the score.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2c300bc8d7189b464f59c93deb5eb56915df62b8)
commit 64d75fa26d3ceb6ccdbee3b31b3705c4db2fc59f
Author: Sage Weil <sage@redhat.com>
Date: Sat Oct 21 10:46:27 2017 -0500
mgr/balancer: fix pg vs object terminology
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6513e8b0fc2ebcbbbda937ec6d533ea084ef278f)
commit 6082f53391a46076bda1b5d56685d928acd2cd4b
Author: Sage Weil <sage@redhat.com>
Date: Tue Oct 24 17:12:54 2017 -0500
mgr/balancer: restrict to time of day
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3f0b50b50aeb34ed39767289b8475ec5b877cc27)
commit 91b84cb24466f84ce3e189e08f8cba1e2900a4cc
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 5 17:31:06 2017 -0500
mgr/module: adjust osd_weight min step to .005
That should be ~1 PG or less on average; no real sense it taking a
step smaller than that!
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 297e2d65abcebe45c6be69a768cf3ea98e1b354d)
commit acbf539d326403b6249ed0e2b0ca5ffc4a52152d
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 5 17:26:16 2017 -0500
mgr/balancer: if score regresses, take a few more steps
To avoid getting stuck in a local minima, take a few additional
steps if our score drops and see if it gets better.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 70b503495a0ceb7d1aa4f5fce4403a8eff9a03b5)
commit b9cc962adfd95304cbd56833bb84b827e5eed4d4
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 5 17:25:19 2017 -0500
mgr/balancer: allow 5% misplaced
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 1ad3101818d1855c8eea8386d7682fab2fe9c8b8)
commit 452962767b4f899c8c2589d01a6b8acc3cac9a69
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 5 17:25:06 2017 -0500
mgr/balancer: more aggressive steps
We are smart enough to back off if we misplace too much.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 0dbcf737eaed02b5aa94bc6f27231b4b87f09ce5)
commit 901ae5920c590a2a83f30897b05017e001572e5f
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 1 16:00:06 2017 -0500
qa/suites/rados/thrash/d-balancer: enable balancer in various modes
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2c9c18d1ec3d33d385adf2a30dc1d3b8e987f9a5)
commit a1ebe97886f7a578bc56a109f10a3d6149954087
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 27 17:05:05 2017 -0400
mgr/balancer: crush-compat: phase out osd_weights
Phase out the osd_weight values as we optimize the crush compat
weight set. Allow a small regression in overall score if it means
we are taking a step to remove the osd_weight value.
Fix the 'osd reweightn' invocation.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 861c37fd99a76d5f1665a610224778fcbb41e9e1)
commit 72716f058cd132d836c3251b4e6d4b056bb4c0a8
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 27 15:12:54 2017 -0400
mgr/balancer: crush_compat: cope with 'out' osds
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c2bba5820a0bf2225dfc9ba39cc67c78083aba11)
commit c982848f32ed85639346931e86d331fb52458355
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 27 15:12:36 2017 -0400
mgr/balancer: stop if we get a perfect score
..and fix the 'worse' threshold (we're just working around floating
point imprecision; 1.01 was too coarse).
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit f1f3014ffcc67200c20e6fb2b2e6e8f1141a0a1d)
commit 37ed6f218ba4fafd5360b8f62feb321a1feec56f
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 27 11:40:36 2017 -0400
mgr/balancer: more dead code
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 37bd64405b8a578329b9a903c9a53322ea29caaf)
commit b3dd7abc79fa3dff7768898caf814bcdfb8f8690
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 27 11:27:42 2017 -0400
mgr/balancer: crush-compat: throttle changes based on max_misplaced
Take smaller steps if we overshoot max_misplaced.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit df2e008d3388b657d62ea4547e6b1b7a0251aa04)
commit 25bfd12cf252dbba81f482fe9abe53959f992b4b
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 27 11:19:30 2017 -0400
mgr/balancer: remove dead code
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 020be37fd06a75ec86e05af966be318c2199da9f)
commit 1aff9dfadc148fb1cf9b60400c35bd11011772b2
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 27 11:19:23 2017 -0400
mgr/balancer: include pg up mapping in MappingState
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit f6f2d253c379a6235e810808e7ce3b83cc4e212c)
commit 9121e73f639bd47310ee29dce69a4c105582a3ab
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 26 18:36:06 2017 -0400
mgr/balancer: normalize weight-set weights to sum to target weight
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 99fcd161bd222a4f0d2ddde0f1a4caecd1a78882)
commit 619ae53354aa5682aeb7d77d3795a323188dad1a
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 26 18:35:42 2017 -0400
mgr/balancer: note root id in Eval
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 26a7ec0c8a4c215791369d8650cd17d41317aefe)
commit afcce93591c7e5c53596cec03165033f049a0c63
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 26 18:00:08 2017 -0400
mgr/balancer: make crush-compat mode work!
- it does multiple iterations, like the upmap optimizer.
- it decreases the step size if it isn't improving, in the hope
that it is overshooting the minimum
- debug output is cleaned up a bit (the info level should be
genuinely useful)
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d9a31595ba15de5fda104a0154778e3200fc46a0)
commit c6052f2706ad6b7c92f9d6fa381527e8922b30c6
Merge: 4e1eec777c 56359c367d
Author: Sage Weil <sage@newdream.net>
Date: Thu Nov 2 14:54:50 2017 -0500
Merge pull request #18590 from liewegas/wip-prime-fixes-luminous
luminous: ceph-bluestore-tool: prime-osd-dir: update symlinks instead of bailing
commit 4e1eec777cf194cdf9bd4db1e7c4672d5b2adac7
Merge: 79736e781f 338af16881
Author: Sage Weil <sage@newdream.net>
Date: Thu Nov 2 14:54:18 2017 -0500
Merge pull request #18620 from tchaikov/wip-21527-luminous
luminous: mon/mgr: sync "mgr_command_descs","osd_metadata" and "mgr_metadata" prefixes to new mons
Reviewed-by: Nathan Cutler <ncutler@suse.com>
commit 79736e781fbd36083c639d3f4c22f8f3ae426cce
Merge: 815d8ab573 9d8e5d763b
Author: Sage Weil <sage@newdream.net>
Date: Thu Nov 2 14:53:16 2017 -0500
Merge pull request #18621 from tchaikov/wip-21534-luminous
luminous: mon: update get_store_prefixes implementations
Reviewed-by: Nathan Cutler <ncutler@suse.com>
Reviewed-by: John Spray <john.spray@redhat.com>
commit 815d8ab573215277a5283af1ecabaac84eaf6200
Merge: c23865f07e 6b6eceee02
Author: Sage Weil <sage@newdream.net>
Date: Thu Nov 2 14:52:32 2017 -0500
Merge pull request #18623 from liewegas/wip-pr-18585-luminous
luminous: ceph-bluestore-tool: prime-osd-dir: update symlinks instead of bailing
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit c23865f07e9147f8ac8a8d3837f735558af30dba
Merge: 24bd1de6af 796e336470
Author: Sage Weil <sage@newdream.net>
Date: Thu Nov 2 14:51:58 2017 -0500
Merge pull request #18625 from tchaikov/wip-ceph-disk-unlock-dmcrypted-partitions-luminous
luminous: ceph-disk: unlocks dmcrypted partitions when activating them
Reviewed-by: Nathan Cutler <ncutler@suse.com>
commit 24bd1de6afaa0d3dab763ae946b03ffb7b333430
Merge: 240edcfb18 3c3609b7e3
Author: Sage Weil <sage@newdream.net>
Date: Thu Nov 2 14:51:32 2017 -0500
Merge pull request #18650 from b-ranto/wip-allow-getattr-luminous
luminous: selinux: Allow getattr on lnk sysfs files
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
commit 240edcfb18475bef84bd5d6345dd365411398773
Merge: 98edc3b9f2 deb78542ed
Author: Sage Weil <sage@newdream.net>
Date: Thu Nov 2 14:51:11 2017 -0500
Merge pull request #18675 from jcsp/wip-luminous-mgr
luminous: bulk backport of ceph-mgr improvements
commit 98edc3b9f2012db277c5796055223f6e69b67e66
Merge: 96102df540 0c0d8e6f40
Author: Alfredo Deza <alfredo@deza.pe>
Date: Thu Nov 2 13:39:05 2017 -0400
Merge pull request #18687 from ceph/backport-18656
luminous: "ceph-volume: adds functional testing for bluestore"
Reviewed-by: Alfredo Deza <adeza@redhat.com>
commit 96102df5405fc470696df84996853e254c29fd31
Merge: 172e6c1e0c 2e6b29084a
Author: Kefu Chai <tchaikov@gmail.com>
Date: Fri Nov 3 01:02:26 2017 +0800
Merge pull request #18293 from smithfarm/wip-21795-luminous
luminous: Ubuntu amd64 client can not discover the ubuntu arm64 ceph cluster
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 0c0d8e6f402ae78ae8110a717188f8b366a67696
Author: Andrew Schoen <aschoen@redhat.com>
Date: Tue Oct 31 15:25:23 2017 -0500
ceph-volume tests: adds objecstore as a testing factor
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit f8b114be61d033c853a2e8295b966e30c6a5f721)
commit 9d970edddc1ab453a683bff1dfd19cab46e8cf9a
Author: Andrew Schoen <aschoen@redhat.com>
Date: Tue Oct 31 15:10:28 2017 -0500
ceph-volume: enable the centos7-bluestore and xenial-bluestore tests
This also updates the ansible and testinfra version to match the current
ceph-ansible master branch.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit b49a19a9773599de295197979b498d06ca500cf8)
commit f9554c4e3ef0eea1f4f09828f98c58838390d4fa
Author: Andrew Schoen <aschoen@redhat.com>
Date: Tue Oct 31 15:09:52 2017 -0500
ceph-volume: adds the xenial-bluestore testing scenario
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 38372890c9c40e6f536026f61c5ef6e6d23d7111)
commit b7fef119d0730a3cb8f19a461f773e66ce97af0f
Author: Andrew Schoen <aschoen@redhat.com>
Date: Tue Oct 31 15:05:46 2017 -0500
ceph-volume: adds the centos7-bluestore testing scenario
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 45e4b0c9d7d4fea1d7bd3481d46edf3b42c40d26)
commit deb78542ed45ac402b4492d5b9ae4a9cdf182724
Author: John Spray <john.spray@redhat.com>
Date: Thu Nov 2 07:42:56 2017 -0400
qa: fix mgr _load_module helper
I inadvertently broke this with the latest change
to the module ls output.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 4fb3025682d812f86c50fa36532290fc0f8857ae)
commit d4fcb97839a501094db2e1704c9799a107ad100e
Author: John Spray <john.spray@redhat.com>
Date: Wed Nov 1 19:10:19 2017 +0000
mgr: fix up make_unique usage for backport
This was getting the definition some other way in master,
but in luminous we need to include the backport14 header.
Signed-off-by: John Spray <john.spray@redhat.com>
commit f41cbb72f39359d7efc9feb0825cbad61e51c07c
Author: John Spray <john.spray@redhat.com>
Date: Fri Oct 20 13:54:29 2017 +0100
mon: fix up mgr_inactive_grace for backport
This was converted to options.cc only, but we cannot
backport another commit that removes this legacy health
reporting code, so this commit updates the legacy health
reporting code for the options.cc style config opt.
Signed-off-by: John Spray <john.spray@redhat.com>
commit a80a7a4d7f9ac73b4d3b86f9787dd9de9865530a
Author: John Spray <john.spray@redhat.com>
Date: Wed Nov 1 06:35:14 2017 -0400
mon: include disabled modules in `mgr module ls`
Otherwise, when someone wants to see what's possible
to do with `mgr module enable` they have to trawl
through the whole mgr map dump.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 5861c4c022e825fe28347ba7e61ae86a6f260f24)
commit fad998cb11628db970099cc192277c1824be884a
Author: John Spray <john.spray@redhat.com>
Date: Wed Nov 1 06:34:36 2017 -0400
doc: describe using `mgr module ...` commands
...including the new "mgr services" command.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit d220e1adc337daeac418563dee125f161e85afdf)
commit a46c5733881403f65872ca36d300ab6a4ca9d2cc
Author: John Spray <john.spray@redhat.com>
Date: Wed Nov 1 06:28:30 2017 -0400
doc: describe how to implement standby modules
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit ceb2a91190441b19171e5b5726748e43ee4d1e92)
commit c704b344a4a739e5ba18535a0fd4f05308c89019
Author: John Spray <john.spray@redhat.com>
Date: Mon Oct 23 05:37:49 2017 -0400
qa: fix mgr caps
This was still using Kraken era settings
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 9988ebed9530718b523a23d729d499dedab5eb6d)
commit f58cb7035cf8992b8772f3358f2d67787ab585c3
Author: John Spray <john.spray@redhat.com>
Date: Thu Oct 19 07:50:19 2017 -0400
qa: expand mgr testing
Some extra coverage of the dashboard, including its standby
redirect mode and the publishing of URIs.
Also invoking the command_spam mode of the selftest module.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 05e648be6a51d3efa110ad9826bbdd0adcc4dd4d)
commit a067468b3717dbab20f84996e76e2336d5fb5f30
Author: John Spray <john.spray@redhat.com>
Date: Tue Oct 17 18:39:17 2017 -0400
mgr/selftest: extend test and add background spam mode
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit a382c3f1ca98b9fb7300e2d410bb2a1bb10b35ae)
commit 1430e448ff93f73b8e8fba61f591ce4ec8c02380
Author: John Spray <john.spray@redhat.com>
Date: Tue Oct 17 18:16:22 2017 -0400
mgr: drop GIL around set_uri, set_health_checks
These didn't need to keep the GIL to go and do their
pure C++ parts, and by keeping it they could deadlock
while trying to take ActiveMgrModules::lock.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 27ee148e040ebaf512f8e11f814b3a7c8cf21f8b)
commit 2a94381b53241f9ecb98c4cc605ef6547b910ca2
Author: John Spray <john.spray@redhat.com>
Date: Tue Oct 17 18:14:43 2017 -0400
mgr: fix ~MonCommandCompletion
This was doing a Py_DECREF outside of the Gil.
Fixes: http://tracker.ceph.com/issues/21593
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 58dfa97ba88882fb3540d15e31bcac48a1aef5ef)
commit 045ed0e023ae2d1a567b3426ef2b48980a851b7e
Author: John Spray <john.spray@redhat.com>
Date: Mon Oct 16 10:51:34 2017 -0400
mgr: update for SafeThreadState
A bunch of the previous commits were done
before this class existed, so updating in
one go instead of trying to edit history
in fine detail.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 29193a47e6cf8297d9b1ceecc7695f2c85434999)
commit ef67307966575b618614225d6206019781d71647
Author: John Spray <john.spray@redhat.com>
Date: Fri Oct 13 11:31:22 2017 -0400
mgr: refactor PyOSDMap etc implementation
Implement real python classes from the C side,
rather than exposing only module methods.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 7e61f79f5d56b568103a067d9a1eb87af997ad61)
commit d85f91cf3e2f747cc23218a52d9271dd6e27febb
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 26 18:35:29 2017 -0400
mgr/PyOSDMap: add CRUSH get_item_weight
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit eacc9021459b31e42232bb958536d594d03b07b3)
commit b1307e074b5fb708420686bf814bd4a7aa76f6d8
Author: John Spray <john.spray@redhat.com>
Date: Mon Oct 16 06:33:48 2017 -0400
mgr: fix py_module_registry shutdown
Was calling way too early, which did a
Py_Finalize before the modules had been
joined.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 0d5b1d70e616d7d1c2d6360375770f5c4754649d)
commit 642a26dfcd9bef06fceb119008f5c3078bbf10e7
Author: John Spray <john.spray@redhat.com>
Date: Thu Oct 12 13:14:02 2017 -0400
mgr: fix thread naming
Was passing a reference to a local stringstream into
Thread::create, not realising that it was taking a char*
reference instead of a copy. Result was garbage (or usually,
all threads having the name of the last one created)
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit bb4e71ed2ebdee1ac5e4b3eee390060e19fea0d8)
commit f9223ad56d8e01c1f6e8690a6a186d02eed8d96d
Author: John Spray <john.spray@redhat.com>
Date: Fri Oct 6 11:02:44 2017 -0400
mgr: cut down duplication between active+standby
...by using PyModuleRunner class from ActivePyModule too.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit df8797320bed7ad9f121477e35d7e3862efd89bd)
commit 728e1e4e136aab5b4d185c738de8df7bcdce18ee
Author: John Spray <john.spray@redhat.com>
Date: Wed Oct 4 13:13:25 2017 -0400
mgr: fix os._exit overrides
These would throw an exception when passed
a status code.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit e2442c1e20bf4ff12d58af500b34a18cc60d2de1)
commit ca51a1b9d2cf0a3bf0fd7b69b833fa6696e61888
Author: John Spray <john.spray@redhat.com>
Date: Thu Aug 24 14:07:37 2017 -0400
mon/MgrMonitor: reset services map on drop_active
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 7b629ae46599d79ca1929cfc6637b367c6bb9029)
commit 8d4512b5636eb41b2cbb50a42d071580cff728c6
Author: John Spray <john.spray@redhat.com>
Date: Tue Aug 22 14:47:10 2017 -0400
mgr/dashboard: implement standby mode
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 4f7007d1b0226af3f0cc33627ebf5051975657ac)
commit 3953c0b3b7d0012507ac047085a3f863dc4929d3
Author: John Spray <john.spray@redhat.com>
Date: Tue Aug 22 11:41:26 2017 -0400
pybind/mgr: add MgrStandbyModule
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 3048e85cd712b7da77cf6ac55dd6a689d00e47e5)
commit e37df9459d6d9abdbe310153bd53d1c1c4384a6b
Author: John Spray <john.spray@redhat.com>
Date: Tue Aug 22 14:42:11 2017 -0400
mgr: standby modules come up and run now
...they still don't have access to any config though.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit c1471c7501948004096581ee415ab4a1fa2d9379)
commit 977583eba266df834a7e27516cacb750bf7121ec
Author: John Spray <john.spray@redhat.com>
Date: Wed Aug 16 10:23:59 2017 -0400
mgr: enable running modules in standby mode
Modules can implement a second, separate class
that has access to very little state about the
system and can't implement commands.
They have just enough information to redirect
or forward incoming requests/traffic to the
active instance of the module on the active mgr.
This enables module authors to create modules
that end users can access via any (running) mgr node
at any time, rather than having to first work out
which mgr node is active.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 25566d1edca638bd15b3ba3326ee7e4d3e573cbb)
commit 6a35a96ea838760945073b756ea7ae13e9a68ccf
Author: John Spray <john.spray@redhat.com>
Date: Tue Aug 15 06:53:18 2017 -0400
mgr: clean up python source file naming
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 70d45a6b93c92baf8d6a3b15765110a5384c5e60)
commit d37bf83f419fd5b34f0cbc8ee5313425e5177bc8
Author: John Spray <john.spray@redhat.com>
Date: Mon Aug 14 06:31:18 2017 -0400
mgr: refactor python module management
Separate out the *loading* of modules from
the *running* of modules.
This is a precursor to enabling modules to run
in standby mode.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 9718896c8b844db2f3c07df1d344636da4605e61)
commit b563555edb51b576fcbf94a3b92598a7300aaf09
Author: John Spray <john.spray@redhat.com>
Date: Thu Jul 27 13:49:27 2017 -0400
pybind/mgr: use set_uri hook from dashboard+restful modules
No more guessing the URL!
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 089e105dd7ec762572ac06794caa7f5543075001)
commit 6dd4d0504de9806fcf7c398a8d66b956b45986d9
Author: John Spray <john.spray@redhat.com>
Date: Thu Jul 27 11:50:23 2017 -0400
mgr: enable python modules to advertise their service URI
Fixes: http://tracker.ceph.com/issues/17460
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit a0183a63fa791954d14c57632e184858cefe893d)
commit 778322d0913d4d9c70609a3c40d809f62a366020
Author: John Spray <john.spray@redhat.com>
Date: Thu Jul 27 11:49:45 2017 -0400
mon/MgrMonitor: store services in map and expose with command
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit c3c3e4e90ba6b09e29879b500f211d607ebabb53)
commit a6b87a8e7c859607d4554833c5f9ad4cb2fea187
Author: John Spray <john.spray@redhat.com>
Date: Thu Jul 27 11:46:40 2017 -0400
messages: `services` in MMgrBeacon
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 236841b3b62af92ce0c4852045327fcfbc5c1651)
commit b44cf5f70d0286c9dcdcda8dbb9f014a3386813a
Author: John Spray <john.spray@redhat.com>
Date: Thu Jul 27 11:45:53 2017 -0400
mon/MgrMap: store list of services
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 3f703bd91f07b2fe43a16df0083d7b7c23803fd5)
commit 5b677f3f02500b87757b9965d20151e345b8ba3f
Author: John Spray <john.spray@redhat.com>
Date: Thu Jul 27 06:31:01 2017 -0400
mgr: carry PyModules ref in MonCommandCompletion
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit e938bf9b9d27e192765c805e5f532c9dd4808b21)
commit b69a656ac76fe38431bcd66c626883e324be1ba4
Author: John Spray <john.spray@redhat.com>
Date: Wed Jul 26 12:31:13 2017 -0400
pybind: update MgrModule for ceph_state->ceph_module
& tidy up the places where ceph_state was getting
used outside of MgrModule.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 62cb512e4740f1f78f516b4f2179c1123fae1b36)
commit 8b5f302706e91c5c6a88e918a9d4baafd726c211
Author: John Spray <john.spray@redhat.com>
Date: Wed Jul 26 07:44:00 2017 -0400
mgr: refactor python interface
Expose a python class instead of a module,
so that we have a place to carry our reference
to our MgrPyModule* and to PyModules*, rather than
passing a handle for the former and using
a global pointer for the latter.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 563878ba217491dd0a6fbd588cd56d09e3456c14)
commit 35b4518c4a0c477f0190bfc006434093fe720c05
Author: John Spray <john.spray@redhat.com>
Date: Thu Aug 3 06:22:35 2017 -0400
mgr/dashboard: remove blue highlight on scrubbing pg states
This was kind of unnecessary, highlighting a completely normal
and healthy situation in a different colour. The blue was
also really hard to read against a grey background.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 99fa1fdf4e1be57792f50907147781d12009b32b)
commit 764e7011d0e69e08aa5f56a5c8c56b8875820b7e
Author: John Spray <john.spray@redhat.com>
Date: Thu Jul 27 11:42:16 2017 -0400
mgr/dashboard: clean up fs standby list when empty
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 5e64787c0ae0ac2a365c89bf89dfea425adc17d4)
commit 1d1dce056664a40ea4dc9ebd1dc3826f0e449555
Author: John Spray <john.spray@redhat.com>
Date: Wed Aug 30 13:56:39 2017 +0100
mgr: remove old-style config opt usage
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit ec09a7abc515f802451bf7ef3d22ce8ee6c6c7b3)
commit a0131144036a9ea64c6f8e3289562bf7db18c746
Author: John Spray <john.spray@redhat.com>
Date: Wed Aug 30 12:12:40 2017 +0100
mon: remove old-style mgr config opt usage
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 6af4120d63324150ba19022c41fe4fa8a38cacbb)
commit 526425c63b6a8f2d3f79fdd75c360d7c2ecfeec6
Author: John Spray <john.spray@redhat.com>
Date: Wed Aug 30 11:48:25 2017 +0100
common: populate manager config option metadata
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit eba4c3f2762ae40ba746091e32364c2d68e780d9)
commit 1eee973b90418979c33ca42b0720045f2de670c3
Author: Kefu Chai <kchai@redhat.com>
Date: Thu Jul 13 14:49:48 2017 +0800
common,mds,mgr,mon,osd: store event only if it's added
otherwise
* we will try to cancel it even it's never been added
* we will keep a dangling pointer around. which is, well,
scaring.
* static analyzer will yell at us:
Memory - illegal accesses (USE_AFTER_FREE)
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 2449b3a5c365987746ada095fde30e3dc63ee0c7)
commit 66aa481be983fdd60beddc7ce621dad3edbf5490
Author: John Spray <john.spray@redhat.com>
Date: Tue Oct 3 08:16:10 2017 -0400
mgr: safety checks on pyThreadState usage
Previously relied on the caller of Gil() to
pass new_thread=true if they would be
calling from a different thread.
Enforce this with an assertion, by wrapping
PyThreadState in a SafeThreadState class
that remembers which POSIX thread
it's meant to be used in.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 625e1b5cfb9b8a5843dfe75e97826f70a57d6ebe)
commit 385a6a0e8e6a6dd315ab358dcea7f3c069d9ad4d
Author: John Spray <john.spray@redhat.com>
Date: Tue Aug 22 11:38:25 2017 -0400
mgr: move Gil implementation into .cc
The inclusion of Python.h in the .h was awkward
for other files including Gil.h.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 23c3a075ee1a27e1b57fcb452a4d6ce53080264e)
commit 241d655e127c71e9d9012d9205005c97d61def8a
Author: John Spray <john.spray@redhat.com>
Date: Wed Jul 26 07:21:40 2017 -0400
mgr: reduce Gil verbosity at level 20
Even at 20, it's pretty heavy to be logging
every lock acquire/release.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 987612a97529be7e67b89977c4a0cf47906a5ecb)
commit 30db4f5a54eac28e546c43a2ce099aa89d179495
Author: Jan Fajerski <jfajerski@suse.com>
Date: Wed Oct 11 12:28:19 2017 +0200
pybind/mgr/prometheus: no ports in osd_metadata
Ports might change on a OSD restart and this would create a new metadata
metric for this osd.
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit 48fec7db4b214fe8ef6a04f8cb53fb8a2fb9c2ca)
commit 80fc65cafaad3bb5499a15edbd97ecf483b4ad82
Author: Jan Fajerski <jfajerski@suse.com>
Date: Wed Oct 11 10:59:33 2017 +0200
pybind/mgr/prometheus: add osd_in/out metric; make osd_weight a metric
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit e4c44c1d702ce242f2cb9a58ca7ce1c31fe0a498)
commit db09ea1c75c80dc189042c59e09ebc1c1f8d1249
Author: Jan Fajerski <jfajerski@suse.com>
Date: Wed Oct 11 20:07:19 2017 +0200
pybind/mgr_module: move PRIO_* and PERFCOUNTER_* to MgrModule class
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit f69484debade5f4fa2bd3a0d1badc9291cc9d7b7)
commit bbe8fbe9ca98c1b0c8b5b43c0296fd9cc4d09c77
Author: John Spray <john.spray@redhat.com>
Date: Mon Oct 9 12:10:22 2017 +0100
qa/mgr: fix influx/prometheus test names
This was a typo: they were swapped around.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit d96a59e74b6984b77c9f3b15f702e3bf45053590)
commit be207ede6028e515a0a632e12330848c93d04a34
Author: John Spray <john.spray@redhat.com>
Date: Thu Sep 28 10:50:53 2017 -0400
doc: flesh out prometheus docs
Explain ceph_disk_occupation, importance
of instance labels and honor_labels, provide
example prometheus configuration yaml.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 5227afed5f33fa9487e1bfa3fd8ce0d82eb4a20f)
commit 8a9e1cb00110ed73f492f1f6cf23b3595756efc4
Author: John Spray <john.spray@redhat.com>
Date: Thu Sep 28 10:10:14 2017 -0400
mgr/prometheus: add ceph_disk_occupation series
This is the magic series that enables consumers to
easily get the drive stats that go with their
OSD stats.
Fixes: http://tracker.ceph.com/issues/21594
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 284be75524f7125dc1409b9c05fe47b37484964e)
commit 51d74e41223397abcde006a4ce53e693a2125852
Author: Benjeman Meekhof <bmeekhof@users.noreply.github.com>
Date: Wed Oct 4 10:05:17 2017 -0400
mgr/influx: Correct name of daemon stat measurement to 'ceph_daemon_stats'
Signed-off-by: Benjeman Meekhof <bmeekhof@umich.edu>
(cherry picked from commit f9014a1c75c6a3adf414b48a707fd444e65b3024)
commit 2934dda9133a3a8876ab2701b11d4379440a5e41
Author: Benjeman Meekhof <bmeekhof@users.noreply.github.com>
Date: Tue Oct 3 16:30:43 2017 -0400
mgr/influx: modify module database check to not require admin privileges
- existing check tried to list all DB and fails even if DB exists if user is not admin level
- still tries to create database if not found and user has privs
Signed-off-by: Benjeman Meekhof <bmeekhof@umich.edu>
(cherry picked from commit 06d7d37c7b9a8c3f4435eff04b6f4934be5e676f)
commit 8c816b8e0fc1fd3fe227690287bcc6e6fce7c54d
Author: Jan Fajerski <jfajerski@suse.com>
Date: Tue Oct 10 08:40:31 2017 +0200
pybind/mgr/prometheus: fix metric type undef -> untyped
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit 6306392492d103200b21ea91bce10a315d7c4e16)
commit 2ebab2f19cdd44f4c568eb96388fc4296f0f1814
Author: John Spray <john.spray@redhat.com>
Date: Mon Sep 25 11:14:57 2017 -0400
mgr: respect perf counter prio_adjust in MgrClient
This awkwardly involves re-ordering some definitions
in perf_counters.h in order to refer to the prio
names defined in PerfCountersBuilder.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 88163749b572ffd2bfe0850136fad5dbed2a9180)
commit f8e9c37286d37d43bb768531f9cfbf70f2cb06b6
Author: John Spray <john.spray@redhat.com>
Date: Mon Sep 18 09:06:13 2017 -0400
test: update perfcounters test for priority in output
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 0f531f7871a68db96b2fb66ffdf6fae6935e6107)
commit f073fc4663db17ea44e9c36831d21df1597d193a
Author: John Spray <john.spray@redhat.com>
Date: Wed Sep 13 17:16:54 2017 -0400
qa: add mgr module selftest task
The module self test commands give us a chance to
catch any other ceph changes that change something
that a module was relying on reading.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 99352ceced9d0fe92ddad6b97b1393b41de75d50)
commit 4c22f0f2669d343a8e7e83f0bc2a2dacbe194f34
Author: John Spray <john.spray@redhat.com>
Date: Wed Sep 13 10:46:56 2017 -0400
mgr/prometheus: remove explicit counter list
These have had their priorities bumped up to
USEFUL, so they'll appear in the default
get_all_counters output.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit ad5a31efbea8081f03dd73669e891d03857ef9cc)
commit fec2b3abf9d9abdc12bfeedaf2b22e8b0cc8c9f4
Author: John Spray <john.spray@redhat.com>
Date: Wed Sep 13 10:45:21 2017 -0400
mon: elevate priority of many perf counters
We can be quite liberal here, because mons are
small in number. However, we don't want to expose
KV database counters at this database from OSDs, so
use the prio_adjust mechanism for that.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit ac8320f23dd4c00eb80da0d9837c29744e38bd57)
commit 8b9a18d3a0c40cb0ef10e32018ea1b23ff53f51e
Author: John Spray <john.spray@redhat.com>
Date: Wed Sep 13 07:07:50 2017 -0400
osd: upgrade a bunch of perf counters to PRIO_USEFUL
These are broadly the OSD-wide IO stats, which happen
to also be the ones that were named in the
prometheus plugin until I changed it to be
priority-based.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit a1cc4ba2993de62b60fd1e58a9704877a6da5fe4)
commit 278188c76ebdbaaa36b3d185b1378abf7665f3fc
Author: John Spray <john.spray@redhat.com>
Date: Wed Sep 13 07:06:24 2017 -0400
common: PerfCountersBuilder helper for priorities
Let the caller set a priority as the defaul, to enable them
to create a bunch at a given priority. This is just a
convenience.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 66f61eeda6a2465b5fc0e40a4f1300913db065dc)
commit 3dff5c0f39289e02e882bf86bf3e23e94ee33aa8
Author: John Spray <john.spray@redhat.com>
Date: Tue Sep 12 10:27:12 2017 -0400
mgr/prometheus: add a self-test command
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 76e1ba52b1b95d417cdd04b8fe985acee648f0e9)
commit a6bc96dfe93f7cbcefa8030a3b6830117516931e
Author: John Spray <john.spray@redhat.com>
Date: Tue Sep 12 08:05:28 2017 -0400
mgr/influx: remove file-based config
...and also trim down the configuration to what's really
needed. In general users don't need to pick and choose
metrics. We could add it back if there was a strong
motivation.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 6776d4645afc49a4bfb4b62673c91384239037f4)
commit bda26fe0e707ec9e609be613805340dadd17e8f3
Author: John Spray <john.spray@redhat.com>
Date: Tue Sep 12 06:51:21 2017 -0400
mgr/influx: enable self-test without dependencies
The idea of self-test commands is that they're self
contained and just exercise the module's calls
to the Ceph-side.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 125294ab9d6e99aa4c960fea147a4e86624b869e)
commit 9abd779d3321d1cda9c8677600e1e35af46109e7
Author: John Spray <john.spray@redhat.com>
Date: Tue Sep 12 06:18:15 2017 -0400
mgr/influx: revise perf counter handling
- Use new get_all_perf_counters path
- Consequently get counters for all daemons, not just OSD
- Tag stats with ceph_daemon rather than osd_id, as some
stats appear from more than one daemon type
- Remove summing of perf counters, external TSDB and/or queries
can do this.
- Remove mgr_id tag: this would change depending on which
mgr was active, which is certainly not desirable.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 59b48e7660f4b757804974835027cd08a59843c2)
commit 32f5f0fe515e6890e8aefad6ad8cdc2574ae8a34
Author: John Spray <john.spray@redhat.com>
Date: Thu Aug 3 13:00:56 2017 -0400
mgr: omit module list in beacon logging
This is useful in itself, but awkward when dealing
with logs generally, because it means that when you
grep on the name of a module, you get mostly beacon
messages rather than the log messages from the
module.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 8d1277fa5c578ce0ea23a70cc58c6cf99921ee25)
commit 10ab4f8b6821e2d6593bc09161c23a9163b93611
Author: John Spray <john.spray@redhat.com>
Date: Tue Sep 12 05:42:23 2017 -0400
mgr: define perf counter constants in mgr_module
So that modules can consume perf counter data
intelligently without having to hunt around
in C land for these constants and redefine them.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 39ab28ed47e869e1466cb3a316a2cb11bdedd23a)
commit 290d15ed912285803a40d02308b921b415c580a2
Author: John Spray <john.spray@redhat.com>
Date: Mon Sep 11 09:12:25 2017 -0400
ceph.in: use PRIO_INTERESTING as daemonperf threshold
Using PRIO_USEFUL as the threshold for what goes into
time series databases. I'm claiming that we have
more "useful" counters than fit on the screen,
so daemonperf's "a screen's worth" threshold
should be at the "interesting" level.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 30a74ce343caec2a433cb532ba697fe7013ed05c)
commit 5ee9e15858d5aa2a71a89c13dea9a5db72579726
Author: John Spray <john.spray@redhat.com>
Date: Mon Sep 11 09:12:01 2017 -0400
mon: set some priorities on perf counters
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 29a71c35c39fbe1d4887e3f5ebb93232daab3487)
commit 3bd478794e2626251eda91850974188da208b591
Author: John Spray <john.spray@redhat.com>
Date: Mon Sep 4 05:39:11 2017 -0400
mgr/prometheus: tag stats by daemon name
Using osd=0 or similar tags was problematic because
daemons of different types have some same-named
counters (e.g. MDS and OSD both have objecter
perf counters).
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit eb524c272c89f8f99f22969b78caa016db7c671e)
commit af92c011b20eaef073730be2f8d7112a0f7e99f8
Author: John Spray <john.spray@redhat.com>
Date: Fri Sep 1 12:02:37 2017 -0400
mgr/prometheus: use new get_all_perf_counters interface
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 11137aa269271ad15dcf19a8d51ce6f4acb7a98e)
commit 3ce4a950975188051b6a3e6cb366c2fb23f0e88f
Author: John Spray <john.spray@redhat.com>
Date: Fri Sep 1 12:01:35 2017 -0400
common: used fixed size int for perf counter prio
...to avoid any ambiguity in allowed range and
make clear how to encode it down the wire.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit ba08fc1008d17aa7a5f285ea2705705ce1a0bda0)
commit d4a0c778747970c231a55baea32f59a1009af4dd
Author: John Spray <john.spray@redhat.com>
Date: Fri Sep 1 12:00:59 2017 -0400
mgr: transmit perf counter prio to the mgr
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit f304f84cfbc22c1a54d152cc38227077bc564a7e)
commit cdcac6e92ff57f1a4e5f5cac676049c6584452c4
Author: John Spray <john.spray@redhat.com>
Date: Fri Sep 1 10:46:56 2017 -0400
common: always include priority in perf counter dump
JSON output with inconsistent sets of members is
annoying to use on the receiving side.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit e631f1a72735ec618e2f3012ad7b9c5830d6c0eb)
commit 802cd49bae2cce6c55eddd68feb617df893e1046
Author: John Spray <john.spray@redhat.com>
Date: Tue Aug 29 11:55:28 2017 -0400
mgr: add get_all_perf_counters to MgrModule interface
This is for use by modules that dump counters
in bulk, e.g. to a TSDB.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 9a42d4255d9d968d6162b53b71db292d9d3de2e4)
commit 85a93dcdddc4cd82c8935bfe123cb20f13f8928a
Author: Jan Fajerski <jfajerski@suse.com>
Date: Fri Aug 11 13:09:24 2017 +0200
pybind/mgr/prometheus: export cluster-wide pg stats, not per osd
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit 13b1236b96d4563e0985cad40d3009b60cc475e7)
commit 1d89c0809f64009d646b929fec675e1aaa6c2b1a
Author: Jan Fajerski <jfajerski@suse.com>
Date: Fri Aug 11 12:51:47 2017 +0200
pybind/mgr/prometheus: add more osd metadata
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit e7704fa9cc35549dba526212c2830df589670416)
commit 6a65408c1364669916643329494825aae394b200
Author: Jan Fajerski <jfajerski@suse.com>
Date: Fri Aug 11 12:05:09 2017 +0200
pybind/mgr/prometheus: don't get perf counters that are not in schema
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit d4ba07d04477ccae3a89dcdcafbb7e76149dfd1c)
commit 62092680ad966eabcdc6b557e9858fe0562147ee
Author: Jan Fajerski <jfajerski@suse.com>
Date: Fri Aug 11 12:04:28 2017 +0200
pybind/mgr/prometheus: add mon and osd perf counters to export
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit fa25d31263a26074225e2a00cb82448066b54069)
commit 752c888533aaa9c7ae5781fafe4a6aa3302059fc
Author: Jan Fajerski <jfajerski@suse.com>
Date: Thu Aug 10 19:46:07 2017 +0200
pybind/mgr/prometheus: add index page, export metrics under metrics/
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit d99a506ed37c2d0991d68ecd34ac5fb213a3eea4)
commit b279ae25d95f6c4b8290042fc9f33c4a6f7afac2
Author: Jan Fajerski <jfajerski@suse.com>
Date: Thu Aug 10 18:19:42 2017 +0200
pybind/mgr/prometheus: export selected perf_counters
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit f6e2e36ba72caf6347f3bb6a985925d0e35077a2)
commit 4c092524929db048f6863acf1a9f12281bdc1646
Author: Jan Fajerski <jfajerski@suse.com>
Date: Thu Aug 10 18:18:36 2017 +0200
pybind/mgr/prometheus: export osd and pool metadata
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit 2bea3814699c27baa8f633b56a8800d697685898)
commit e3dafc3dd682a10e2310efdac4615d834933e7d6
Author: Jan Fajerski <jfajerski@suse.com>
Date: Thu Aug 10 18:15:56 2017 +0200
pybind/mgr/prometheus: actually emit reported pg counts
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit c288624eed862559b2c86c5dfc85c837716739ab)
commit 1ef5c88ef08ab9950ee8d7d4acda032b5a54f2f5
Author: Jan Fajerski <jfajerski@suse.com>
Date: Thu Aug 10 18:09:17 2017 +0200
pybind/mgr/prometheus: no need to wait for notify event
If stats or perf counters are not available they won't be emitted.
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit ead0973d7dd12fe985390891c80f1bc15f7b9aec)
commit d0ef1cd2570644451205e5306feca0da94462d1f
Author: Jan Fajerski <jfajerski@suse.com>
Date: Thu Aug 10 18:07:14 2017 +0200
pybind/mgr/prometheus: no need to convert perf_schema to ordered_dict
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit 5e4b4b5ea2a217731691c1c391c252b08452798a)
commit 7f191ff9dff33cac551ae5f64027d566b17b6d98
Author: Jan Fajerski <jfajerski@suse.com>
Date: Wed Aug 9 17:22:49 2017 +0200
pybind/mgr/prometheus: add device_class label to osd metrics
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit 76d1918724320b7d6b1120b57b3002bb24099001)
commit b24263387e46ac537a8ac189f9e0e80699518e0b
Author: Jan Fajerski <jfajerski@suse.com>
Date: Wed Aug 9 16:19:38 2017 +0200
pybind/mgr/prometheus: add cluster wide metrics; no perf counters for now
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit 49b3ff83cd231066d2a8f1809fadbdeb2c0c1f88)
commit 5f0ef3a73a55a3f24997e415a848308afa0263ba
Author: Jan Fajerski <jfajerski@suse.com>
Date: Fri Aug 4 10:23:11 2017 +0200
pybind/mgr/prometheus: prefix metrics with 'ceph'; replace :: with _
Both follow prometheus best practices. While : is a legal metric
character, "Exposed metrics should not contain colons, these are for
users to use when aggregating."
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit 177afcc7886aa3898d092ebd1e101697bc6539fd)
commit a303218c85498ccf472d2f1b35fd391376fb0faf
Author: mhdo2 <mhdo2@users.noreply.github.com>
Date: Mon Aug 21 12:13:01 2017 -0400
doc/mgr: add influx plugin docs
Signed-off-by: My Do <mhdo@umich.edu>
(cherry picked from commit e345fe3c5780976a4e33488b3a75cd24bb2c96c5)
commit de89603e54f550ffc4ff994026f32c5b7d5f6529
Author: mhdo2 <mhdo2@users.noreply.github.com>
Date: Tue Jul 18 18:33:55 2017 -0400
mgr/influx: added influx plugin
Signed-off-by: My Do <mhdo@umich.edu>
(cherry picked from commit 68ae26c014d0471cc3f2f979dc8d822b2e50740f)
commit 2fea47d9710ec1e388db044065bc42b8046dae22
Author: John Spray <john.spray@redhat.com>
Date: Sat Sep 23 11:55:55 2017 -0400
mgr: store declared_types in MgrSession
Because we don't (yet) properly prevent multiple sessions
from daemons reporting the same name (e.g. rgws), storing
it in the DaemonPerfCounters meant that one daemon's report
was referring to another daemon's set of reported types.
This should always have been a property of the session.
The behaviour will still be ugly when multiple daemons
are using the same name (stomping on each other's stats/statsu)
but it shouldn't crash.
Fixes: http://tracker.ceph.com/issues/21197
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit dc415f1ae09a308bd448614934a4c168eb9cf07b)
commit 951d67fc9fee23f3f8f7d7df8fbfb8c29e60f82f
Author: John Spray <john.spray@redhat.com>
Date: Mon Sep 18 10:12:00 2017 +0100
mgr: make pgmap_ready atomic to avoid taking lock
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit d20915741d985e080a723cd6563bc6f4a657276f)
commit 946d1541bd2b14d6f8abbaa201c3521dd117f01b
Author: John Spray <john.spray@redhat.com>
Date: Mon Aug 28 07:29:36 2017 -0400
mgr/DaemonServer: handle MMgrReports in parallel
The DaemonStateIndex locking is sufficient to make all
the report processing safe: holding DaemonServer::lock
through all ms_dispatch was unnecessarily serializing
dispatch.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 64af9d3da0fceff9ad0ff668f60d272c46912f34)
commit d70fae092db920e2db77c7c8b044cfb4d9687992
Author: John Spray <john.spray@redhat.com>
Date: Thu Aug 24 12:53:24 2017 -0400
mgr: clean up DaemonStateIndex locking
Various things here were dangerously operating
outside locks.
Additionally switch to a RWLock because this lock
will be relatively read-hot when it's taken every time
a MMgrReport is handled, to look up the DaemonState
for the sender.
Fixes: http://tracker.ceph.com/issues/21158
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 806f10847cefe5c7a78fc319b1b130d372197dd3)
commit b86cc9a2e49a4d1b53e83f47876a42b42819efcb
Author: John Spray <john.spray@redhat.com>
Date: Thu Aug 31 12:13:23 2017 -0400
mgr: runtime adjustment of perf counter threshold
ceph-mgr has missed out on the `config set` command
that the other daemons got recently: add it here
and hook it all up to the stats period and threshold
settings.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 057b73d641decb9403aba50caae9d139f3a34dd4)
commit c278580d207f0b99daed48afdc689ae77fe73e9b
Author: John Spray <john.spray@redhat.com>
Date: Mon Jul 31 09:24:09 2017 -0400
mgr: apply a threshold to perf counter prios
...so that we can control the level of load
we're putting on ceph-mgr with perf counters. Don't collect
anything below PRIO_USEFUL by default.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit bdc775fdd8acdad5c58ff3065a21396f80ce5db4)
commit 11720b96ec8f9ef683710dffa3da4cda2ecf096d
Author: Sage Weil <sage@redhat.com>
Date: Tue Aug 8 16:36:23 2017 -0400
pybind/mgr/balancer: make auto mode work
(with upmap at least)
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ef1a3be05671ad31907cf8c4beb64a766359bc66)
commit b6a000f0efef3cd6a143b82ae33cacf660cd8e4c
Author: Spandan Kumar Sahu <spandankumarsahu@gmail.com>
Date: Mon Aug 7 04:01:57 2017 +0530
src/pybind/mgr/balancer/module.py: improve scoring method
* score lies in [0, 1), 0 being perfect distribution
* use shifted and scaled cdf of normal distribution
to prioritize highly over-weighted device.
* consider only over-weighted devices to calculate score
Signed-off-by: Spandan Kumar Sahu <spandankumarsahu@gmail.com>
(cherry picked from commit c09308c49ca087fb8c5e7d4261b0234190f863d9)
commit 6090ae6c224904516736c5f1a4ae5bcb6d7e6caa
Author: Sage Weil <sage@redhat.com>
Date: Fri Aug 4 17:59:20 2017 -0400
pybind/mgr/balancer: make 'crush-compat' sort of work
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 7a00e02acd1b2ff21dac829de30f80fd69eae602)
commit add3cd36db7c02d52c2fa429c034c114cac526d8
Author: Sage Weil <sage@redhat.com>
Date: Thu Aug 3 16:23:08 2017 -0400
pybind/mgr/balancer: rough framework
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d5e5c68c374e7d5514f89aac2d3df6008d103a76)
commit 0c73e433ab6583fca6eea7678c23b469d643ae04
Author: Sage Weil <sage@redhat.com>
Date: Thu Jul 27 23:33:06 2017 -0400
mgr/PyOSDMap: OSDMap.map_pool_pgs_up, CRUSHMap.get_item_name
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit a928bf62316c32f37dd1791192fd9a2ddaef0d33)
commit 423947563c8e88f89bef9d71b3116c3302bd40c9
Author: Sage Weil <sage@redhat.com>
Date: Sun Jul 23 00:10:56 2017 -0400
mgr/PyOSDMap: get_crush, find_takes, get_take_weight_osd_map
These let us identify distinct CRUSH hierarchies that rules distribute
data over, and create relative weight maps for the OSDs they map to.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3b8a276c437cfd599c55a935d141375afda676ff)
commit 073f23734c8058e3efcb440df89adab818510695
Author: Sage Weil <sage@redhat.com>
Date: Thu Jul 27 10:07:31 2017 -0400
crush/CrushWrapper: rule_has_take
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ef140de639078b40c05971fb219f7b8c12d83228)
commit df426b5c24e12b1156bccdd8948cbed5977c348a
Author: Sage Weil <sage@redhat.com>
Date: Sat Jul 22 23:50:27 2017 -0400
crush/CrushWrapper: refactor get_rule_weight_osd_map to work with roots too
Allow us to specify a root node in the hierarchy instead of a rule.
This way we can use it in conjunction with find_takes().
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 69454e0570274ff7f252e7f081965dcc9bb04459)
commit 89cac2d5176300838c23a28814257d0f395e39c9
Author: Sage Weil <sage@redhat.com>
Date: Sat Jul 22 23:17:18 2017 -0400
pybind/mgr/balancer: do upmap by pool, in random order
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 028a66d43244c15a77e71f3d3e4f41773837ab02)
commit bfdc955ab6ab424b6a46dc1b0022fb3dc13ce157
Author: Sage Weil <sage@redhat.com>
Date: Tue Jul 11 16:27:08 2017 -0400
pybind/mgr/balancer: add balancer module
- wake up every minute
- back off when unknown, inactive, degraded
- throttle against misplaced ratio
- apply some optimization step
- initially implement 'upmap' only
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 0d9685c50f79fbb53dbc8bd98c95900ef6e902b8)
commit c57a55e8d426e6f87d4649a799d9ddee665618e0
Author: Sage Weil <sage@redhat.com>
Date: Tue Jul 11 16:26:16 2017 -0400
pybind/mgr/mgr_module: add default arg to get_config
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 39c42ddb9339c1950a3a474e8083db8b24e775a6)
commit 4d686ee875d5bc0125426d348e044988aaca9f91
Author: Sage Weil <sage@redhat.com>
Date: Mon Jul 10 23:23:19 2017 -0400
mgr: add trivial OSDMap wrapper class
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2ef005196ba2eb49c34c32def624938c7a8beb03)
commit 1b721632aa8271d74b319714db0a9fedacb2629b
Author: Sage Weil <sage@redhat.com>
Date: Thu Jul 27 10:06:45 2017 -0400
mgr/PyModules: add 'pg_dump' get
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit bfb9286f4212947183c46543d609b664ea13b489)
commit 68d411cf7b934c724f1f8ebeaa7148daebd79599
Author: Sage Weil <sage@redhat.com>
Date: Tue Jul 11 16:25:42 2017 -0400
mgr/PyModules: add 'pg_status' dump
This is summary info, same as what's in 'ceph status'.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 85b5b80906d00e098d4b1af1354c60a357022dd2)
commit 35de92b259f50b52e51ab1daf57041553170406d
Author: David Zafman <dzafman@redhat.com>
Date: Wed Nov 1 09:53:34 2017 -0700
osd: build_past_intervals_parallel: Ignore new partially created PGs
Fixes: http://tracker.ceph.com/issues/21833
Signed-off-by: David Zafman <dzafman@redhat.com>
commit 172e6c1e0c5bb86db7cc0017acaa674aa893e3a0
Merge: 2988a39abe 966683fde6
Author: Casey Bodley <cbodley@users.noreply.github.com>
Date: Wed Nov 1 16:28:25 2017 -0400
Merge pull request #18674 from ceph/wip-rgw-s3-branch
qa/tests: use ceph-luminous branch for s3tests
Reviewed-by: Casey Bodley <cbodley@redhat.com>
commit 966683fde600837eb9a0f82ca6fd00f52e721809
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Wed Nov 1 10:32:07 2017 -0700
qa: use ceph-luminous branch for s3tests
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
commit 2988a39abedc6a092b8b3c2f0b33a3559a55461b
Merge: 4dc0a0e38a 8f87fa2d3a
Author: John Spray <jspray@redhat.com>
Date: Wed Nov 1 15:24:55 2017 +0100
Merge pull request #18412 from kmroz/wip-21659-luminous
luminous: mgr: fix crashable DaemonStateIndex::get calls
Reviewed-by: John Spray <john.spray@redhat.com>
commit 4dc0a0e38a3f1a988c180d47970102df4b326b9e
Merge: 273e035dbe 72c8583107
Author: John Spray <jspray@redhat.com>
Date: Wed Nov 1 12:33:45 2017 +0100
Merge pull request #18113 from jcsp/wip-prometheus-port-backport
luminous: ceph-mgr: can not change prometheus port for mgr
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: John Spray <john.spray@redhat.com>
commit 273e035dbe158f162c2ee559744c456c942f1737
Merge: a7f519913d f463cd84ae
Author: Yuri Weinstein <yuri.weinstein@gmail.com>
Date: Tue Oct 31 14:45:27 2017 -0700
Merge pull request #18655 from ceph/wip-yuriw-bp-PR18634-luminous
qa: add "restful" to ceph_mgr_modules in ceph-ansible suite
commit f463cd84ae1fb38817078967a4a40ba4b600f0e5
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Oct 31 11:21:58 2017 +0800
qa: add "restful" to ceph_mgr_modules in ceph-ansible suite
backport of https://github.com/ceph/ceph/pull/18634
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit caf9ee5c60d999951979d0b67afda8d56e1cd91d)
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit a7f519913df3aba83aa3ead9eee6b94aa8f90ffd
Merge: f2749114c5 b84803d609
Author: Karol Mroz <kmroz@suse.de>
Date: Tue Oct 31 10:26:42 2017 -0700
Merge pull request #17889 from smithfarm/wip-21372-luminous
luminous: core: Improve OSD startup time by only scanning for omap corruption once
Reviewed-by: David Zafman <dzafman@redhat.com>
commit f2749114c596d4016eb14e4777cb93a11b14b705
Merge: 528c3b6d32 1236e0f2e9
Author: Karol Mroz <kmroz@suse.de>
Date: Tue Oct 31 10:18:07 2017 -0700
Merge pull request #18004 from linuxbox2/luminous-rgwf-wr
luminous: rgw_file: fix write error when the write offset overlaps.
Reviewed-by: Casey Bodley <cbodley@redhat.com>
commit 528c3b6d32f9c3319fd429de37bbca578538ed7c
Merge: ec2d294684 d800747fa1
Author: Karol Mroz <kmroz@suse.de>
Date: Tue Oct 31 09:03:48 2017 -0700
Merge pull request #18438 from theanalyst/wip-21696-luminous
luminous: fix a bug about inconsistent unit of comparison
Reviewed-by: Casey Bodley <cbodley@redhat.com>
commit 3c3609b7e3ce7339a845f9191a709d12025fbbb2
Author: Boris Ranto <branto@redhat.com>
Date: Thu Sep 21 17:24:07 2017 +0200
selinux: Allow getattr on lnk sysfs files
This showed up during downstream testing for luminous. We are doing
getattr on the sysfs lnk files and the current policy does not allow
this.
Fixes: http://tracker.ceph.com/issues/21523
Signed-off-by: Boris Ranto <branto@redhat.com>
(cherry picked from commit 394c26adb97cd150233fe8760355f486d03624a4)
commit ec2d2946840fac65d9588f225adaa6f51c32f6b0
Merge: 90017e7d93 def3d55eb7
Author: Kefu Chai <tchaikov@gmail.com>
Date: Tue Oct 31 13:04:00 2017 +0800
Merge pull request #18410 from kmroz/wip-21732-luminous
luminous: qa/suites/rest/basic/tasks/rest_test: whitelisting
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 90017e7d933b52794d32fef69d97d5ba529d3936
Merge: 4f024c6831 c603faaf17
Author: Yuri Weinstein <yuri.weinstein@gmail.com>
Date: Mon Oct 30 20:23:39 2017 -0700
Merge pull request #18629 from ceph/wip-yuriw-21978-luminous
qa/suites/upgrade/jewel-x: Changed typo ('hammer' to 'jewel')
commit c603faaf171fd3074ffe5e1342e8b96cd261e097
Author: Yuri Weinstein <yweinste@redhat.com>
Date: Mon Oct 30 14:52:52 2017 -0700
Changed typo ('hammer' to 'jewel')
Fixes http://tracker.ceph.com/issues/21978
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit 555746a09e8bb7bcf99da91f9d00dfab9791d80c
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Fri Oct 27 13:20:07 2017 -0700
MDSMonitor: wait for readable OSDMap before sanitizing
Fixes: http://tracker.ceph.com/issues/21945
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit ca52f3bd93e6c743aa05171108527d877807b426)
commit be13cdc0d4e54ce23c6ecf658100ddd860b1b96f
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Tue Oct 3 12:25:12 2017 -0700
mds: clean up non-existent data pools in MDSMap
Older versions of Ceph weren't strict about preventing pool deletion when the
MDSMap referred to to-be-deleted pool. If we are dealing with a cluster
upgrade, we should try to gracefully handle that by cleaning out data pools
that have been removed.
Reproduced this by allowing CephFS pools to be deleted:
diff --git a/src/mon/OSDMonitor.cc b/src/mon/OSDMonitor.cc
index 85c47c13da6..694b240cb9f 100644
--- a/src/mon/OSDMonitor.cc
+++ b/src/mon/OSDMonitor.cc
@@ -10962,7 +10962,7 @@ int OSDMonitor::_check_remove_pool(int64_t pool_id, const pg_pool_t& pool,
FSMap const &pending_fsmap = mon->mdsmon()->get_pending();
if (pending_fsmap.pool_in_use(pool_id)) {
*ss << "pool '" << poolstr << "' is in use by CephFS";
- return -EBUSY;
+ //return -EBUSY;
}
if (pool.tier_of >= 0) {
pdonnell@icewind ~/ceph/build$ bin/ceph osd pool create derp 4 4
pool 'derp' created
pdonnell@icewind ~/ceph/build$ bin/ceph fs add_data_pool cephfs_a derp
added data pool 3 to fsmap
pdonnell@icewind ~/ceph/build$ bin/ceph osd pool rm derp derp --yes-i-really-really-mean-it
pool 'derp' is in use by CephFSpool 'derp' removed
pdonnell@icewind ~/ceph/build$ bin/ceph fs ls
...
2017-10-03 12:50:48.409561 7f9e2e05b700 -1 /home/pdonnell/ceph/src/osd/OSDMap.h: In function 'const string& OSDMap::get_pool_name(int64_t) const' thread 7f9e2e05b700 time 2017-10-03 12:50:48.407897
/home/pdonnell/ceph/src/osd/OSDMap.h: 1184: FAILED assert(i != pool_name.end())
ceph version 12.1.2-2624-g37884a41964 (37884a419640b446fffc1fa4d6074c97339fdd96) mimic (dev)
1: (ceph::__ceph_assert_fail(char const*, char const*, int, char const*)+0xf5) [0x564ebb5420f5]
2: (()+0x41dade) [0x564ebb3cbade]
3: (MDSMonitor::preprocess_command(boost::intrusive_ptr<MonOpRequest>)+0x1fb9) [0x564ebb4cd119]
Note when testing this fix, use something like this after removing the data pool:
pdonnell@icewind ~/ceph/build$ bin/ceph fs set cephfs_a max_mds 2
Setting max_mds will cause a new FSMap to be created where MDSMap::sanitize is
called; this is simulating the initial load+sanitize of a Hammer legacy MDSMap
by the mons.
Fixes: http://tracker.ceph.com/issues/21568
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 7adf0fb819cc98702cd97214192770472eab5d27)
commit 23fa3b726368f036b31e53a0bec8ad4bc654d993
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Tue Oct 3 12:23:03 2017 -0700
mds: reduce variable scope
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 37884a419640b446fffc1fa4d6074c97339fdd96)
commit 4f024c68319a79bef9d937d7f2363a5b69575773
Merge: 6166148078 5a0016131b
Author: Alfredo Deza <alfredo@deza.pe>
Date: Mon Oct 30 14:05:00 2017 -0400
Merge pull request #18627 from ceph/backport-18513
ceph-volume lvm zap backport
Reviewed-by: Alfredo Deza <adeza@redhat.com>
commit 5a0016131b1ec900255057876c98fc70e00542f0
Author: Andrew Schoen <aschoen@redhat.com>
Date: Fri Oct 27 11:29:50 2017 -0500
ceph-volume: set journal_uuid and journal_device when using a partition
This correctly sets the tags when a partition is used for a filestore
journal.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 1a8561d38dd7363dc920ae82ec1343b2b75a1ca2)
commit 05522e75f8d98658b5440d3c2f70e93ce65c65ce
Author: Andrew Schoen <aschoen@redhat.com>
Date: Tue Oct 24 10:13:40 2017 -0500
docs for ceph-volume lvm zap
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 3e93a31deda96ed6fb23fa19ce3e273de05f7d88)
commit cb1e81447f5c0c4eb8c0fd9fb11d0ceae3e26199
Author: Andrew Schoen <aschoen@redhat.com>
Date: Mon Oct 23 09:51:43 2017 -0500
ceph-volume: add tests for ceph-volume lvm zap
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 3d5e391693fc64747a4774287c968f842294eaa6)
commit 33927c10e78842bb4c4ce4c0e4051ccceb8bb9e8
Author: Andrew Schoen <aschoen@redhat.com>
Date: Mon Oct 23 09:44:07 2017 -0500
ceph-volume: print success message if zap succeeds
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 6d70762dee0cd9047c291cf9869f666a375e956b)
commit d8381ab509d9f60f6fe62c5692a9374b3187d5c9
Author: Andrew Schoen <aschoen@redhat.com>
Date: Fri Oct 20 09:58:48 2017 -0500
ceph-volume: terminal.write only takes one argument
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit a5454eba241abd8cc1b1a660b12a1aec7c3c16e0)
commit 6f2bd88278f2c7dc4cfd156969f98d7cc8c3e763
Author: Andrew Schoen <aschoen@redhat.com>
Date: Thu Oct 19 16:29:39 2017 -0500
ceph-volume: the metadata field for the lv path is lv_path not path
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit e3a9113e7025f1e3a3130ec1e2d565f37bf3d2dc)
commit 1dfe7e9854b00850b86168049821447d33298fff
Author: Andrew Schoen <aschoen@redhat.com>
Date: Thu Oct 19 14:59:26 2017 -0500
ceph-volume: update help text for ceph-volume lvm zap
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit cbc38c4e4a7dcc24b31bcfa6af73eb8cf04f56ad)
commit c5206435a9bae9d93e6cd9a3c2b98127871ad94e
Author: Andrew Schoen <aschoen@redhat.com>
Date: Thu Oct 19 14:55:44 2017 -0500
ceph-volume: remove lvm metadata when zapping an lv
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 5f57fc87205c2d35da9a1f28c72e233ffb3fe4d9)
commit eb53ef67b4e3bf5afb6e414fd42caab67c4cfdf2
Author: Andrew Schoen <aschoen@redhat.com>
Date: Thu Oct 19 14:55:02 2017 -0500
ceph-volume: adds Volume.clear_tags
Will remove all tags for the current lv
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 3c9401f1618f7fcbea827fe7279e8dfde617d957)
commit 4ee13e0587c5e152fdac18004bc4336f8b90e4af
Author: Andrew Schoen <aschoen@redhat.com>
Date: Thu Oct 19 12:01:57 2017 -0500
ceph-volume: zap should leave lvs and partitions intact
This will remove filesystems and wipe data from any lv or partition
given to 'ceph-volume lvm zap' but still leave it intact for further
use.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit caba9fb80b358222192f736bd1b0ab31dca25cec)
commit 8c3ab3459abfebac20bb4ce3ddecf1a7a58cf401
Author: Andrew Schoen <aschoen@redhat.com>
Date: Thu Oct 19 11:44:11 2017 -0500
ceph-volume: adds utilites to wipe the fs and clear data when zapping
These should eventually move to a disk api.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 2f64d4a0abd41afbcb9eba6a237642502d9a9135)
commit 8a0e7fad57e41e10d51e0e167b671668ee1aa286
Author: Andrew Schoen <aschoen@redhat.com>
Date: Thu Oct 19 10:35:51 2017 -0500
ceph-volume: adds tests for api.lvm.remove_lv
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 937b57438147681c2c6e32c6db38d8bea68d4731)
commit c6dd47ab6d1ec6784ec3cef8f65139baeaf4fa3a
Author: Andrew Schoen <aschoen@redhat.com>
Date: Thu Oct 19 10:35:10 2017 -0500
ceph-volume: api.lvm.remove_lv should return True when successful
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 7f055aa6deb3904cf0334e214c13a26098b08aa8)
commit 71ae0b8cabc0a696e6762d7f377f9f4b28d2a963
Author: Andrew Schoen <aschoen@redhat.com>
Date: Wed Oct 18 10:19:25 2017 -0500
ceph-volume: zap logical volumes
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 1d083a2191315ee94301c2f5f102a0906dd05fa8)
commit 7110922c42ee30fa25dc9dacbfc8d0f9ae2f2a79
Author: Andrew Schoen <aschoen@redhat.com>
Date: Wed Oct 18 10:19:00 2017 -0500
ceph-volume: adds a remove_lv command to the lvm api
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit cf98b6971ced0caa29ff4f7ac346df652177fa29)
commit 731610f18b9e4109ce72a4467af209144e4a127d
Author: Andrew Schoen <aschoen@redhat.com>
Date: Tue Oct 17 14:15:18 2017 -0500
ceph-volume: stubs out the ceph-volume lvm zap command
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit 2e64b797ef6ae91623ffba8ae28d3f8ccc7d7b93)
commit c2237c7c6dbcd5420d8ed87c6f2648c3f4e6edae
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Oct 27 09:58:20 2017 -0400
qa/suites/rbd: run cls tests for all dependencies
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 27613a63d1f1ee6fa7327cd1c63b875c0e5247f8)
commit 6b75efab390b3019e404f007090e9031328e1301
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Oct 27 09:57:43 2017 -0400
cls/journal: fixed possible infinite loop in expire_tags
Fixes: http://tracker.ceph.com/issues/21956
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 9e66dca49591e50b9cab5df311f1dc217eb58fcc)
commit 796e33647022de1fd831b31045242fb8e9d0e4d2
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Aug 30 18:46:49 2017 +0800
ceph-disk: unlock all partitions when activate
should unlock all dmcrypted partitions when activating a device.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit ba2aa0cee9e077d8439ba31228b41beb2d827a04)
commit 3e9aad1a116e02bb126010f310ca2267177b5162
Author: Felix Winterhalter <felix@audiofair.de>
Date: Mon Jul 17 02:04:39 2017 +0200
ceph-disk activate unlocks bluestore data partition
Signed-off-by: Felix Winterhalter <felix@audiofair.de>
(cherry picked from commit 1287caf2dbb5ef6e5f243fe1d23633946aef26f9)
commit 6b6eceee027e571d8d3da4af144391f8da85b53e
Author: Yao Zongyou <yaozongyou@vip.qq.com>
Date: Sat Oct 28 18:23:30 2017 +0800
ceph-bluestore-tool: the link target should not ending with new line
Signed-off-by: Yao Zongyou <yaozongyou@vip.qq.com>
(cherry picked from commit 39c68d128cf29f8e0a617009d16c5edf471ead47)
commit cc84812507607d0a91233a513088094533a2d6a7
Author: Yao Zongyou <yaozongyou@vip.qq.com>
Date: Sat Oct 28 18:22:27 2017 +0800
ceph-bluestore-tool: the right action is prime-osd-dir not prime-osd-dev
Signed-off-by: Yao Zongyou <yaozongyou@vip.qq.com>
(cherry picked from commit d418a04e9fed6bf91cc9acc5ac777eadf664c974)
commit 9d8e5d763b173247f95f0769fd52372a8a7a5371
Author: John Spray <john.spray@redhat.com>
Date: Mon Sep 25 10:44:28 2017 +0100
mon: implement MDSMonitor::get_store_prefixes
Fixes: http://tracker.ceph.com/issues/21534
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit a3c317b406c69b9d6a2d7df94806f4c308e6ee6a)
commit 338af1688147babb4e92523972175fe09ebb34b6
Author: huanwen ren <ren.huanwen@zte.com.cn>
Date: Mon Sep 25 15:01:02 2017 +0800
mon/mgr: sync mgr_command_descs table and mgr_metadata table
sync mgr_command_descs table and mgr_metadata table to new mons
Fixes: http://tracker.ceph.com/issues/21527
Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
(cherry picked from commit 13f6aa3aaa6de0aeccec67d7c6f3effe43dcae49)
Conflict: in master PaxosService::get_store_prefixes(..) is marked const
while in luminous, the cleanup commit which added the `const` is not
backported yet, so drop the `const` in the backported commit.
commit de07d2954afb012e51d2eaca12c1763e29a19eca
Author: John Spray <john.spray@redhat.com>
Date: Mon Sep 25 10:37:18 2017 +0100
mon/OSDMonitor: tidy prefix definitions
We should define them in one place to make it easy
when updating get_store_prefixes.
Fixes: http://tracker.ceph.com/issues/21534
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 889ac5166ad8fd30678325ddc2da59f45db53f06)
commit 981e552d72c151b6bf2c9bb64b969f9446ef4789
Author: huanwen ren <ren.huanwen@zte.com.cn>
Date: Mon Sep 25 14:55:55 2017 +0800
mon/osd_metadata: sync osd_metadata table
sync osd_metadata table to new mons when add new mons
Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
(cherry picked from commit 755ec735107c75156ae4935f7255bbfe5dc384d9)
commit 3f0ad5f453a259cc53f50b1a9515e8a720da4a1d
Author: huanwen ren <ren.huanwen@zte.com.cn>
Date: Mon Sep 25 14:55:55 2017 +0800
mon/osd_metadata: sync osd_metadata table
sync osd_metadata table to new mons when add new mons
Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
(cherry picked from commit 6c20433bfb049ac6c69f6f7a979006e8b9ea0e29)
commit 61661480780e555fc501aec7c32163596e1e18d3
Merge: c0ec364ad6 410434b3d2
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sun Oct 29 11:06:28 2017 -0700
Merge PR #18385 into luminous
* refs/pull/18385/head:
mds: fix race in PurgeQueue::wait_for_recovery()
mds: open purge queue when transitioning out of standby replay
mds: always re-probe mds log when standby replay done
Reviewed-by: John Spray <john.spray@redhat.com>
commit c0ec364ad6f54da7efbbb6b84ad34ea9782e3df2
Merge: 5ee57e1b42 695d7ad511
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sun Oct 29 11:06:24 2017 -0700
Merge PR #18316 into luminous
* refs/pull/18316/head:
mds: prevent trim count from underflowing
Reviewed-by: Zheng Yan <zyan@redhat.com>
commit 5ee57e1b42d93fc0dc77b1221c362753684b280e
Merge: 84f4bf40da b99acfd601
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sun Oct 29 11:06:20 2017 -0700
Merge PR #18300 into luminous
* refs/pull/18300/head:
mds: keep CInode::STATE_QUEUEDEXPORTPIN state when exporting inode
Reviewed-by: Zheng Yan <zyan@redhat.com>
Reviewed-by: Amit Kumar <amitkuma@redhat.com>
commit 84f4bf40dade5ef55eba139d63187959b98e84e5
Merge: 2638d5039e 460268b5e9
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sun Oct 29 11:06:16 2017 -0700
Merge PR #17729 into luminous
* refs/pull/17729/head:
ceph.in: validate service glob
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Amit Kumar <amitkuma@redhat.com>
commit 2638d5039e615035563338cd428b44fb23d879ff
Merge: f6f88c2a9b 2973b6d418
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sun Oct 29 11:00:18 2017 -0700
Merge PR #18299 into luminous
* refs/pull/18299/head:
mds: update client metadata for already open session
Reviewed-by: Zheng Yan <zyan@redhat.com>
Reviewed-by: Amit Kumar <amitkuma@redhat.com>
commit f6f88c2a9b40cf84658319a07a2bc26858763093
Merge: 388901f091 f353a1e805
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sun Oct 29 11:00:14 2017 -0700
Merge PR #18298 into luminous
* refs/pull/18298/head:
osdc/ObjectCacher: limit memory usage of BufferHead
Reviewed-by: Zheng Yan <zyan@redhat.com>
commit 388901f0911b2d45f5b5ca2d653da5c605d99e2d
Merge: 45cfeae6eb c5ccbf1108
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sun Oct 29 11:00:10 2017 -0700
Merge PR #18085 into luminous
* refs/pull/18085/head:
ceph_volume_client: fix setting caps for IDs
Reviewed-by: Ramana Raja <rraja@redhat.com>
commit 45cfeae6ebd5a0910c03ff323ecdd5740e569b41
Merge: ecd5b0066f 1021fe235c
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sun Oct 29 11:00:06 2017 -0700
Merge PR #18030 into luminous
* refs/pull/18030/head:
qa: relax cap expected value check
mds: improve cap min/max ratio descriptions
mds: fix whitespace
mds: cap client recall to min caps per client
mds: fix conf types
mds: fix whitespace
doc/cephfs: add client min cache and max cache ratio describe
mds: adding tunable features for caps_per_client
Reviewed-by: Zheng Yan <zyan@redhat.com>
commit ecd5b0066fc091f34a166989b2e68e165c390dd0
Merge: 04acd559db d5e583490d
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sun Oct 29 10:55:32 2017 -0700
Merge PR #17921 into luminous
* refs/pull/17921/head:
ceph_volume_client: perform snapshot operations in
Reviewed-by: Amit Kumar <amitkuma@redhat.com>
Reviewed-by: Ramana Raja <rraja@redhat.com>
commit add8c3db195acfda75a1628196e839b0147e7c3f
Author: Matt Benjamin <mbenjamin@redhat.com>
Date: Thu Oct 26 18:28:10 2017 -0400
rgw_file: set s->obj_size from bytes_written
Required to store correct final accounted_size of objects in
RGWWriteWriteRequest::write_finish.
Fixes: http://tracker.ceph.com/issues/21940
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
(cherry picked from commit b1f528d35aeccabb4e5818aec6feb8e53e562500)
commit 8cbb2eb937cae4e136ef77bc93e6d5346e21c8da
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 9 08:15:21 2017 -0500
qa/cephfs: test ec data pool
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d0732fc96fbc6849dd51b391d85f765c74cfb593)
commit 04acd559db55c6e8040a2ba65a850e72a04ad2ab
Merge: d294493c42 8a87d43db2
Author: Andrew Schoen <andrew.schoen@gmail.com>
Date: Fri Oct 27 11:34:15 2017 -0500
Merge pull request #18593 from ceph/luminous-wip-bz1499840
luminous ceph-volume lvm bluestore support
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
commit d294493c42ea21e85a1fd36eb8623229a2ff6ea6
Merge: 58c1b8c4f2 34cd96d342
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 17:30:35 2017 +0200
Merge pull request #18334 from kmroz/wip-21816-luminous
luminous: rgw: fix bilog entries on multipart complete
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 58c1b8c4f28b73d9b7db1cf3fe79ae5062832c88
Merge: 8742c619b0 d7f6b93af4
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 17:30:07 2017 +0200
Merge pull request #17861 from smithfarm/wip-21441-luminous
luminous: rbd: [cli] mirror getter commands will fail if mirroring has never been enabled
commit 8742c619b03cc5f048dc08f76d67ea64c109b4b3
Merge: 7c03693d0b f8bfd9458d
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 17:29:55 2017 +0200
Merge pull request #18416 from kmroz/wip-21855-luminous
luminous: librbd: object map batch update might cause OSD suicide timeout
Reviewed-By: Jason Dillaman <jdillaman@redhat.com>
commit 7c03693d0bd1a7108bd2d55eda7b7bd1e4e7f904
Merge: fc486a26cb 8b3e9917a7
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 17:29:43 2017 +0200
Merge pull request #18337 from kmroz/wip-21640-luminous
luminous: rbd-mirror: forced promotion can result in incorrect status
Reviewed-By: Jason Dillaman <jdillaman@redhat.com>
commit fc486a26cb12466df6269ec4c7717332be2ed92c
Merge: c944c688e5 4ebd4b3928
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 17:29:22 2017 +0200
Merge pull request #18336 from kmroz/wip-21639-luminous
luminous: librbd: snapshots should be created/removed against data pool
Reviewed-By: Jason Dillaman <jdillaman@redhat.com>
commit c944c688e5a4a7935233e7fb3824b97d9afe44d9
Merge: bc7e648b18 34af07f97a
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 17:29:03 2017 +0200
Merge pull request #17860 from smithfarm/wip-21299-luminous
luminous: rbd: [rbd-mirror] asok hook names not updated when image is renamed
Reviewed-By: Jason Dillaman <jdillaman@redhat.com>
commit 8a87d43db25a364841d969636bd74b7590c3d563
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 26 15:09:18 2017 -0400
ceph-volume lvm.activate remove links on activate before priming
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 634b5caaed7c9188e426727e83a0768bdbc51f0d)
commit 869d13037dbb2b8882c4a4a91ec20c9814bcc438
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 26 07:48:24 2017 -0400
ceph-volume lvm.activate ceph-bluestore-tool needs --cluster too
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit fb36087bc3c7e7bb8358cbf8ad84409fa23fac68)
commit 3baab9f4e286c3b54c179b0f471435c2dbc859b2
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 25 18:57:28 2017 -0400
ceph-volume lvm.activate consume cluster_name from lvm metadata
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 2b7ff497fe9e0e167ab01bf0d8ce5e26f0da836f)
commit 8f41a983356ee9d182f8247f2da8db15848afae0
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 25 18:57:03 2017 -0400
ceph-volume lvm.prepare persist cluster_name on lvm metadata
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 6d38ad987e08791be379f6c006c1eb5bf9464415)
commit 1ae30d3130af845ebf563a74a1e2642f5236f1ed
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 25 15:30:43 2017 -0400
ceph-volume util.prepare bluestore db and wal paths are to devices, not osd dir paths
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit fd616acd6ae0f184c1757e7de38fc746d526908c)
commit fdfbfb8140fc1be3123d1254f4463613dc7c7cb0
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 25 15:30:06 2017 -0400
ceph-volume lvm.prepare bluestore wal and db links get created with mkfs
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 80698d2309a221d2318b11be9b8f8a95b51416ea)
commit 6ac58f3cd5e292153c82d881a30db1f6c4bfd936
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 25 10:28:43 2017 -0400
ceph-volume lvm.prepare use wal and db for mkfs
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 4970ab5c804111ce4b2b57817f1e29d58b18a893)
commit c9ecd14ca220e45960650715703cae3e4376e3c0
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 24 16:28:33 2017 -0400
ceph-volume lvm.prepare ensure tags are set for ceph.type correctly for block and data
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 58dbefd3be7bad51e6c56c50fbfb450e205b3036)
commit 76fb5fec94a9bbfed1163a23faf87e19d5617bd4
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 24 16:26:03 2017 -0400
ceph-volume lvm.activate check if lv device is mounted at path
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 1ead135bef801280e7cec21283c95c7ba440d84b)
commit 343362027368c4170bd785f1633afeb5ed90f3a2
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 24 15:55:21 2017 -0400
ceph-volume lvm.prepare default to using bluestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 39dd0d1d10aba08c3e8299eb1702260274a756a7)
commit 0a27fe88266458eead5a011c27cc323f5b7173a5
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 24 15:54:57 2017 -0400
ceph-volume lvm.create default to using bluestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d3145da6e643dedb2fbed72e7d4d0cddf397c5e3)
commit bd2e0aa78ad1a500935c3134f4d1e6c3413a26ae
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 24 15:54:11 2017 -0400
ceph-volume lvm.common remove boolean defaults for CLI flags
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit a7d5965ab532a621f402bad55d45dfed61d90cc5)
commit c3d474af562ec5b8658bed6362046b0208a2c0eb
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 24 13:29:18 2017 -0400
ceph-volume lvm.prepare update to use create_osd_path
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit bafb72144666f891773bed95e9a0f757673af2e0)
commit bc7e648b182569865d1c2a2aea6da1e52a3d3cb1
Merge: 690be9c711 b015ce1161
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:57:48 2017 +0200
Merge pull request #18138 from kmroz/wip-21684-luminous
luminous: rgw: stop/join TokenCache revoke thread only if started.
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 690be9c711b34be1ec4254ba0cedf7d6755f59f6
Merge: 5727e4a33e 06c31a6cee
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:55:22 2017 +0200
Merge pull request #17994 from ukernel/luminous-21337
luminous: mds: make sure snap inode's last matches its parent dentry's last
Reviewed-By: Patrick Donelly <pdonelly@redhat.com>
commit 5727e4a33e0b82e0f9591251572b61458ec2cfae
Merge: d973d6d7ae 624b3ac505
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:52:43 2017 +0200
Merge pull request #18431 from theanalyst/wip-21857-luminous
luminous: rgw: We cant't get torrents if objects are encrypted using SSE-C
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 02013a714ec5fa6c2ff1073e432b8f6b9ff41305
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Oct 23 09:44:27 2017 -0400
ceph-volume lvm.activate only prime-osd-dir when directory is empty
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 9bb6cfd867c2d8e972093d023622f8c2e5b440d4)
commit 28593dbf1caeeec8f7c69dbd06e2ff9759afd072
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Oct 23 09:20:26 2017 -0400
ceph-volume lvm.activate safeguard against auto detect objectstore flag not being set
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit c9b5f352a8f9e2743e5dca3a0479c8099a59fd12)
commit 8df51c8af5d95d67effece720ec2ff82517155b9
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 15:18:20 2017 -0400
ceph-volume util.system use strings for comparison in py3
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 1b671a8602b036cb8d53bc40eb70198c13143051)
commit 1f09ee754fec40c4b17ea33829e6449d8749690a
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 15:10:08 2017 -0400
f ceph-volume tests update changes from the help menu in lvm
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 3ddcd3c2f19cc848a11a2882f150462ad2f26c65)
commit ec7c42bd59dc05cb8bad082d64ffdc28f13d3dc8
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 14:12:02 2017 -0400
ceph-volume lvm.common --journal-size doesn't accept a volume group
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit a8282e9a580bcfa74abdd33032e2ce2851b5cf29)
commit 243098815d1d49570c193767d008cbc4fa0f174b
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 14:04:31 2017 -0400
ceph-volume lvm.prepare filestore does not require a volume group
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 3653f105927d455b8df143797af0cfe72f8d6834)
commit da5f764a083613e5d74d1401da65c2ba3aff5fb4
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 14:04:05 2017 -0400
ceph-volume lvm.common volume groups are no longer required
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 526c0aa81ac8eed95c2b4664a00ded2118372fff)
commit eaa9150d4a6bacf62fcf485e8d36430743ac06cf
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 12:07:06 2017 -0400
doc/ceph-volume activate workflow mention tmpfs in osd dirs
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 8788fe3721034f390240c01493bc1d27d65a73a5)
commit d65af56eaaab728d8c81ddb69b967cb13f177766
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 12:02:10 2017 -0400
doc/ceph-volume update activation to include bluestore support
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 288a3124793d11d71b540555fd03c2f61a79dd16)
commit 6545b1c14e84af5f25fd4bbd405b56f7b5797113
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 11:58:47 2017 -0400
doc/ceph-volume update create to indicate bluestore support
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 61ce8b4b844fcb64fc95373c5515de902a52fc8c)
commit d973d6d7aefae1c34b664f98955baed4c841579b
Merge: cf621f8b4e 5b5f0b5a33
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:49:03 2017 +0200
Merge pull request #17858 from smithfarm/wip-21448-luminous
luminous: rgw: string_view instance points to expired memory in PrefixableSignatureHelper
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit cf621f8b4e8138284843f6d979aca61651a854bd
Merge: e0c3a05959 0b3a974be7
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:48:54 2017 +0200
Merge pull request #18442 from theanalyst/wip-21637-luminous
luminous: encryption: PutObj response does not include sse-kms headers
Reviewed-By: Casey Bodley <cbodley@redhat.com>
Reviewed-By: Matt Benjamin <mbenjamin@redhat.com>
commit e0c3a05959a70b4cf4213320ba7c990578bf8c19
Merge: dd7aa70301 14a1dcb1ff
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:48:26 2017 +0200
Merge pull request #18437 from theanalyst/wip-21698-luminous
luminous: radosgw-admin usage show loops indefinitly
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit dd7aa7030144d42958f92af5c7f6804a402f3b6d
Merge: 3a8f046716 c3a719da27
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:48:12 2017 +0200
Merge pull request #18434 from theanalyst/wip-21817-luminous
luminous: rgw: zone compression type is not validated
Reviewed-By: Casey Bodley <cbodley@redhat.com>
Reviewed-By: Matt Benjamin <mbenjamin@redhat.com>
commit 3a8f046716456eb0f51db79f152f3dc67d84c319
Merge: 0d9b3a01d0 4c18ac1240
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:47:48 2017 +0200
Merge pull request #18440 from theanalyst/wip-21652-luminous
luminous: policy checks missing from Get/SetRequestPayment operations
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 0d9b3a01d0972a2fc5ae9550ba80c38acd9edbeb
Merge: aa71baeabe a3e2ae4925
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:47:33 2017 +0200
Merge pull request #18441 from theanalyst/wip-21651-luminous
luminous: rgw: avoid logging keystone revocation failures when no keystone is configured
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit aa71baeabef5aa7eb6903593b3dfcc8878227c99
Merge: 1067293a7e 0b84dd07c4
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:47:13 2017 +0200
Merge pull request #18443 from theanalyst/wip-21634-luminous
luminous: s3:GetBucketLocation bucket policy fails with 403
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 1067293a7e21fc835447c9e5e1d8416902593ba1
Merge: f2f8aef89c abefb578c3
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:46:59 2017 +0200
Merge pull request #18444 from theanalyst/wip-21635-luminous
luminous: s3:GetBucketCORS/s3:PutBucketCORS policy fails with 403
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit f2f8aef89c3144ac5d859ea0fe8a7db57cfa8ad6
Merge: 385d84b8eb 0e65ea466f
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:46:44 2017 +0200
Merge pull request #18430 from theanalyst/wip-21695-luminous
luminous: failed CompleteMultipartUpload request does not release lock
Reviewed-By: Casey Bodley <cbodley@redhat.com>
Reviewed-By: Matt Benjamin <mbenjamin@redhat.com>
commit 385d84b8ebb3a4192ce4817aaf6c99b27af0190d
Merge: 4a74a41c47 f9850d2148
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:46:13 2017 +0200
Merge pull request #18445 from theanalyst/wip-21633-luminous
luminous: s3:GetBucketWebsite/PutBucketWebsite fails with 403
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 4a74a41c473fcda61f156d8bf720bf0e4cf0aec0
Merge: 14cb83cd07 8b420c4627
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:45:54 2017 +0200
Merge pull request #18436 from theanalyst/wip-21789-luminous
luminous: user creation can overwrite existing user even if different uid is given
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 14cb83cd077258e061cfa09733cf0b9a82f7507a
Merge: 0339b8b599 fe45086a54
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:45:36 2017 +0200
Merge pull request #18435 from theanalyst/wip-21790-luminous
luminous: RGW: Multipart upload may double the quota
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 0339b8b599f313e6a7b176f0f8b94b8101ab50c3
Merge: 68cc868d1f 248bd94dab
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:45:19 2017 +0200
Merge pull request #18432 from theanalyst/wip-21856-luminous
disable dynamic resharding in multisite enviorment
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 68cc868d1fdca55308fa1aba0bb708d07170e668
Merge: f337bef88a 17df0ba8f0
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:44:40 2017 +0200
Merge pull request #18429 from theanalyst/wip-21792-luminous
luminous: encryption: reject requests that don't provide all expected headers
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 29470eac4ca939c824b56f4fbc03c8b7814f9d02
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 11:51:55 2017 -0400
doc/ceph-volume update prepare with bluestore workflow
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 9f1a5627399e2589fe706b158a32a9fb8642ac23)
commit c90e473ff35d15d309eb92dbdc6b9df2e8bd5ecc
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 10:15:09 2017 -0400
doc/ceph-volume update filestore to indicate it is not the only/default objectstore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit c93603e5764e021305cfeb1c297ee16318126fe1)
commit 3890c47dddd0d028b0c003cb46a9bec670c445f1
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 10:00:17 2017 -0400
doc/ceph-volume elaborate on env vars for interval and tries in activation
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 1dc24d1a7065b4e23b1bfa0e7c6d263b5ebc6906)
commit 88240acf19176ae5c139836b7d46848e15ef5454
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 09:33:55 2017 -0400
ceph-volume lvm.common update cli flags help for bluestore support
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit c9693fb3f8386016aba5d7c8f53f5f71f948c88d)
commit a945e7776309d39955918cec7e0c5f1895abc39b
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 09:23:35 2017 -0400
ceph-volume util.disk st_mode is needed for stat()
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit e3b5663f7968e44ec0cd6db2bfc43acdb233e314)
commit e833916afe22e9c4765344229b5b89cfc9af1b14
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 09:23:02 2017 -0400
ceph-volume decorators always log to the file with exceptions
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit cb3533a72571251d9f1b6171f783246462726857)
commit 53e9cf6fcd7549b4bfa78d7568d56b30b38e81fd
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 08:49:32 2017 -0400
ceph-volume tests verify behavior of get_lv_from_argument
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit adabbd462bcb19a6da65211e24dfb2b9b5043ef4)
commit 40cfec281cc2584b2e2987f356109c6c160d4597
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 08:48:01 2017 -0400
ceph-volume api.lvm return result of get_lv when using arguments
If the argument is an absolute path it is fine to just return whatever
get_lv finds since it is a "safe" call, it will return a None if nothing
is found
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f209c9d4d50bd7abe0f7f6b9fb29359cb7196718)
commit 7139077aab892c3213815d1dcf71d93df80a32c8
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 08:12:01 2017 -0400
ceph-volume lvm.listing support bluestore, not only filestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 6a5b89a475c2a36022ca09dbb7f6b94cc3b44881)
commit 654e06f1202596d5cecef9d03134861eed053882
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 08:10:51 2017 -0400
ceph-volume api.lvm allow full paths to detect lv from arguments
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 8d257101b1556ec5ce805622e9b7b6b74f6b000d)
commit 454d48eb29f4429370b831645a29e0450b5b46b1
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 20 08:10:26 2017 -0400
ceph-volume tests add support for bluestore list verification
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 4674da9c07f871bf1f25bb782c136fe082095a14)
commit 8d72737f90b0cd1db860678ebb66596e233fd646
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 19 13:20:55 2017 -0400
ceph-volume tests activate with fsid works for bluestore and filestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 5b9f0853f780db3b0af6b4b7c1ad1d59184069c9)
commit 846aa2492f6db342d9fb4c477c2e13e61e414aaf
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 19 13:14:05 2017 -0400
ceph-volume tests arg validator can accept absolute paths
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 15586715f506505a57ec5122416541479df74bc8)
commit 2c9ee51a0a7ed5edc900316c19ba27facb9163eb
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 19 12:03:17 2017 -0400
ceph-volume tests update changes from the help menu in lvm
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 39bdc2e3f621a32bcbb3dfe64f156bbfd7e4f5ee)
commit 5724bb04bb74e17b9f7e96b3a3e434cac720c4ed
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 19 11:46:57 2017 -0400
ceph-volume tests update to get_lv from get_journal_lv
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 9be1012072f36e21a3f3cef42a42f7f1eb83c64e)
commit 43e25933701605ac8cd7cb3d6fbd4cbf6705565e
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 19 11:30:29 2017 -0400
ceph-volume tests.util new path/device mount interface
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d5984bad97347e58318793fc05da84832d323113)
commit 9b36a8bf8d471047dc94cdfcac1f84a9802b9fc5
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 19 09:49:25 2017 -0400
ceph-volume tests update api.lvm tests
The create_lv signature changed to require full size description and
tags need to be an actual dictionary (vs. keyword args)
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit e8a18637531cde59d73a97d106958489c1d12117)
commit ea57e70e6f826ac17eae0226b9d916e8bff4d021
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 18 16:06:55 2017 -0400
ceph-volume lvm.activate ceph-bluestore-tool should populate the osd dir
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f3bb616bcf8a4737a3701ebf812de831785d6341)
commit fe3b61ac0da67d9c80ead24c6ccf79e1507fa132
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 18 15:06:01 2017 -0400
ceph-volume lvm.activate auto detect objectstore type
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 472aa0b1997a0b1301b699b7ce30dce728aaea0c)
commit 5da9775184e4bc9bf1b83d18eae602c9a510a89f
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 18 15:00:19 2017 -0400
ceph-volume lvm.trigger enable objectstore auto detection for activate
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f579ef07026564f1a94b2923a2e4640ade2da503)
commit 307c99e4382895e6b42c193e9b98b9206fd57f9a
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 18 12:00:34 2017 -0400
ceph-volume lvm.activate add bluestore support for activation
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 5696fed09b65044142b34f89492a6d91d3243e14)
commit 70632b0fac2fa7e411418c90a50b99391379f436
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 18 11:40:38 2017 -0400
ceph-volume util.system create mappings of mounts
This makes it easier for consumers to detect a path, or a device that
might be mounted or might have more than one mount, including tmpfs and
devtmpfs support, which was just not possible before.
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f61007cc650f6964ef2ff3ea98a1debf02713198)
commit 839dc1f554a12fb109ca1f992784dd4f4933898a
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Oct 16 06:51:05 2017 -0400
ceph-volume util.prepare separate filestore vs. bluestore prepare utils
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 85cdc783008ff0e0566c34aaab7cd5f12db1cbc7)
commit deae88983d80cedcdc842d192f8d659b7533d3dc
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Oct 16 06:50:27 2017 -0400
ceph-volume process allow obfuscating a key or index when logging commands
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit bbcca70d962855d991f0aa8e86a19f1855345668)
commit 18336242b85065d8a2b0a92f22b2e91a81b997ba
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 13 09:22:53 2017 -0400
ceph-volume util.prepare add other ln helpers for bluestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 555b664882db2cd247087f907bdf4d426fb652f8)
commit 37a1564a41ba8ae6a89eb0903f0fdf6b463d90f0
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 12 16:08:52 2017 -0400
ceph-volume util.arg_validators allow to pass a group as --data for lvm
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 211c38f591945dee53196068533784c385bae102)
commit afd1588fe06a80f2277082fd7bd15f43d8e601b7
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 12 16:08:16 2017 -0400
ceph-volume util.disk add utilities for is_device and is_partition
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 6b23fc72240ced158cf5335f6e815dbfd804dfab)
commit 5e8f798d2b8f80b21f3ce4701d22e47992dd0938
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 12 16:05:22 2017 -0400
ceph-volume api.lvm udpate create_lv for bluestore
Require full size notation for LVM (e.g. 50G or 100M) and do not
"translate" tags by pre-fixing them with 'ceph'. That is already how the
rest of the API works, no need to make that translation in one place
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit ac0d1555d3d9caeef9d4fc3e46bc11b7e437413b)
commit f337bef88a31a5d23420665406a06f761e8d4e6e
Merge: 3db3ca9938 d2be1e0ccb
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:43:43 2017 +0200
Merge pull request #17857 from smithfarm/wip-21446-luminous
luminous: rgw:multisite: Get bucket location which is located in another zonegroup, will return 301 Moved Permanently
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 3db3ca99380521c79596b06b695eefbb81d57908
Merge: ced01a5b85 c4d6539374
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:43:00 2017 +0200
Merge pull request #17856 from smithfarm/wip-21444-luminous
luminous: rgw: setxattrs call leads to different mtimes for bucket index and object
Reviewed-By: Casey Bodley <cbodley@redhat.com>
commit 3bbd840fe0d34d42c04fac523aee5fc023d8bd3b
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 12 15:42:22 2017 -0400
ceph-volume lvm.prepare update filestore workflow to use new helpers
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 1e8a79b8f4b9e6360e8e3ace580e46caeedca8c7)
commit ced01a5b85c9ec5657d8f09a447b6308ea597754
Merge: 2c9a9af424 fe0a41b4a9
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:41:15 2017 +0200
Merge pull request #18417 from kmroz/wip-21782-luminous
luminous: cls/journal: possible infinite loop within tag_list class method
Reviewed-By: Jason Dillaman <jdillaman@redhat.com>
commit 2c9a9af424c618c50334c0efa7f5a5f807aeb4fd
Merge: a1d286252a 7d2b7e8e31
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:38:04 2017 +0200
Merge pull request #18287 from smithfarm/wip-retract-assertions-luminous
luminous: rgw: Remove assertions in IAM Policy
Reviewed-By: Adam Emerson <aemerson@redhat.com>
Reviewed-By: Abhishek Lekshmanan <abhishek@suse.com>
commit fb95f3f9fd6122f9b8c3c2215463fc8a1ef62ba2
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 12 14:19:07 2017 -0400
ceph-volume lvm.prepare initial take on bluestore support
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit e4fc3464af472a8dbdf049917eed73519ff82c3b)
commit a1d286252a75c3632a7ecc63a366e5262f9e16cd
Merge: b151619a45 fd207447d1
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 16:29:02 2017 +0200
Merge pull request #17859 from smithfarm/wip-21451-luminous
luminous: rgw: lc process only schdule the first item of lc objects
Reviewed-By: Daniel Gryniewicz <dgryniewicz@redhat.com>
commit 26ef33efc37cdd626c6a7577812bb87947417e01
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Oct 12 14:16:25 2017 -0400
ceph-volume lvm.common update help flags for bluestore usage
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 17028f3a3c1f9394fa82dc6f2bbd596d96acb037)
commit 005359b92a910b72712edbcaa7b2e27a54d3a093
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 11 14:30:41 2017 -0400
ceph-volume api.lvm allow to create a vg
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 16d7feb2b5d5eb5fa0bf23a95b60623ae2e1676f)
commit 795e69673f452227afc12d76ed5aa0d6a8af8665
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 10 12:30:06 2017 -0400
ceph-volume lvm add block CLI flags for bluestore
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 63d2afbc605ac1531bc9f00a1891017bfd42fbce)
commit cbfb082c180e93cd147e164afd1d32ef9234924e
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 10 12:15:14 2017 -0400
ceph-volume lvm make bluestore the default option
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit dc3ced30747312dedd0e895a75ec15b1925ef1ea)
commit a340f1acab04587f5b2c323abc1a14b8dd7de915
Author: Casey Bodley <cbodley@redhat.com>
Date: Mon Oct 9 14:46:47 2017 -0400
rgw: RGWDataSyncControlCR retries on all errors
similar to RGWMetaSyncShardControlCR, we don't want to exit and
stop the data sync processor thread on failures. we want to keep
retrying with backoff
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 065e67bbd312240f5128c31f5525098c62f3afce)
Conflicts:
src/rgw/rgw_data_sync.cc (sync tracing)
commit fdec63a28e24e056938139591e5799a0e3693c20
Author: Casey Bodley <cbodley@redhat.com>
Date: Mon Oct 9 14:09:40 2017 -0400
rgw: fix error handling in ListBucketIndexesCR
the call to set_state() returns 0, when we want operate() to return the
error code instead. use set_cr_error() to do this
Fixes: http://tracker.ceph.com/issues/21735
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit ed6340a47f32472bb1f650407dbe739c44bf4bd6)
commit e711cf319fa12ec7b03edda769685353e24cbfcb
Author: Casey Bodley <cbodley@redhat.com>
Date: Mon Oct 9 14:08:21 2017 -0400
rgw: ListBucketIndexesCR spawns entries_index after listing metadata
if the metadata listing fails, we won't have to clean up entries_index
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 7f127f54406ef0e08a1265b9bc20433fe3f21523)
commit 56359c367d8267281f714dcecefdb50c598718a6
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 26 13:51:40 2017 -0500
ceph-bluestore-tool: prime-osd-dir: update symlinks instead of bailing
If the symlink points to the right location, do nothing. If it doesn't,
replace it. If it's not a symlink, bail with EEXIST.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit de8dc42d42218bc1a1779e1bcc5831c567853c8d)
commit b151619a4523b148026e613860a05add6aff3831
Merge: 84fc30873a 352373ff4a
Author: Abhishek L <abhishek.lekshmanan@gmail.com>
Date: Fri Oct 27 15:11:00 2017 +0200
Merge pull request #18433 from theanalyst/wip-21545-luminous
luminous: rgw file write error
Reviewed-By: Matt Benjamin <mbenjamin@redhat.com>
commit 84fc30873aa6ed3c1bf0c91307c362d27e9bc72a
Merge: 2bc24caa00 bb8586b259
Author: Andrew Schoen <andrew.schoen@gmail.com>
Date: Fri Oct 27 08:03:16 2017 -0500
Merge pull request #18587 from ceph/luminous-wip-volume-api-change
luminous ceph-volume lvm api refactor/move
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
commit 2bc24caa003ea3c4898202a887a5864078c0f3fd
Merge: a5548a8247 5f96edfb41
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 27 06:39:58 2017 -0500
Merge pull request #18568 from liewegas/wip-fix-rop-leak-luminous-2
luminous: osd/PG: fix recovery op leak due to recovery preemption
Reviewed-by: David Zafman <dzafman@redhat.com>
commit bb8586b259726eb0ae72db7c6a238b2d2432fc9b
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 3 16:13:33 2017 -0400
ceph-volume create an api module to hold common api modules
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f72368641d7f38b6395afb7ed70a0a9d0794a03a)
commit 3d32fb30193437b57535c79657ee726704f4b037
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 4 06:44:54 2017 -0400
ceph-volume tests move lvm api tests into its new test module
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit eec0c8a5c27adc9b3fc5e999cbee8165bdb45736)
commit e6145af90c3f1c8c520bb7671194e6d8a13bb5d2
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 4 06:43:36 2017 -0400
ceph-volume tests refactor lvm api import for test_listing
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 66fd41fd4b35e275196bfb94794f3fda9fd77f76)
commit c63d3ba89489cc42aaa364c99b3d3b4c1bba226c
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 4 06:43:18 2017 -0400
ceph-volume tests refactor lvm api import for api tests
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 2a9a36bd24bddb35a7618127a0ec57f661a71566)
commit 6efd25344230c0abb51cd82f99310a4b94f2fcd3
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 4 06:42:47 2017 -0400
ceph-volume tests refactor lvm api import for activate tests
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d81aa8c112312f2fb49d94ce3512b3356b0b8814)
commit 47cd005b672828aedf11aeb67a6099477783f7f0
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Oct 4 06:42:01 2017 -0400
ceph-volume tests refactor lvm api import in conftest
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit a00aff7066fe6a679498ac8a02ec22ef29feced3)
commit 4a6dfa9b6949bdcdba22a7003c451390105d287e
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 3 16:17:08 2017 -0400
ceph-volume devices.lvm refactor importing api to come from api/lvm.py
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit f4bee6cb42a017abce269ed3926deff04038a47e)
commit 9a20f184676973222c931cecff26ca313e636e1d
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 3 16:14:16 2017 -0400
ceph-volume move lvm/api.py to api/lvm.py so disk can consume it
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit bb72480d2bf44ff04ea93c98f4a4e59032a28896)
commit f26f3dc842c04ac70e309266aaedbae9d45eeae8
Author: yuliyang <yuliyang@cmss.chinamobile.com>
Date: Sun Sep 24 08:41:04 2017 +0800
rgw:fix list objects with marker when bucket is enable versioning
fix: http://tracker.ceph.com/issues/21500
Signed-off-by: yuliyang <yuliyang@cmss.chinamobile.com>
(cherry picked from commit bc16b162cf757b9c6ceae30912eb647ef9304f75)
commit a5548a8247cd9b072a2e0357cf6d0358550e7565
Merge: 724609a8a5 1751a094d7
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 26 16:52:28 2017 -0500
Merge pull request #18498 from liewegas/wip-recovery-fixes-luminous
luminous: osd: fix recovery priority and pg state on recovery->backfill transition
Reviewed-by: David Zafman <dzafman@redhat.com.
commit 724609a8a54a7619b48f4608790d81a2a43d4683
Merge: 50c4cc0610 565b5cd1d6
Author: Andrew Schoen <andrew.schoen@gmail.com>
Date: Thu Oct 26 16:50:20 2017 -0500
Merge pull request #18567 from ceph/backport-wip-bz1491250
luminous: ceph-volume lvm list
Reviewed-by: Andrew Schoen <aschoen@redhat.com>
commit 5f96edfb41c4b028732cf12da5ab44322e251b2a
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 26 16:23:41 2017 -0500
osd/PG: fix recovery op leak due to recovery preemption
This was fixed in master in a different patch, but are not yet ready to
backport the bits there that came before this. For now, fix it
specifically for luminous. We can either sort out the conflicts later
or revert this and backport the master parts conflict-free.
Signed-off-by: Sage Weil <sage@redhat.com>
commit 50c4cc06101fd1c8a9323cd3a04c92c8a9ab2580
Merge: a175c0321e 33aa167331
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 26 16:14:43 2017 -0500
Merge pull request #17805 from tchaikov/wip-luminous-pr-17508
luminous: mon: more aggressively convert crush rulesets -> distinct rules
Reviewed-by: Sage Weil <sage@redhat.com>
commit a175c0321e745ac095707d5a2d3cd25232272bfc
Merge: 72b057187a b382db7cfc
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 26 16:14:07 2017 -0500
Merge pull request #18364 from kmroz/wip-21544-luminous
luminous: mon: osd feature checks with 0 up osds
Reviewed-by: Sage Weil <sage@redhat.com>
commit 72b057187aed9b4e5f3a4ecf82fafa6e9cd2d3fd
Merge: f6362e3110 b60d8cf0cb
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 26 16:13:31 2017 -0500
Merge pull request #18413 from kmroz/wip-21693-luminous
luminous: osd: additional protection for out-of-bounds EC reads
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit f6362e3110e4e482ff5a676035b862b1b0b666df
Merge: f36670e190 4c6b1b65f7
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 26 16:11:42 2017 -0500
Merge pull request #18529 from smithfarm/wip-21918-luminous
luminous: Disable messenger logging (debug ms = 0/0) for clients unless overridden.
Reviewed-by: Sage Weil <sage@redhat.com>
commit f36670e190ea96589c373d87d30ee8f4b3c6b20b
Merge: 9a28a5c9d7 cec24f99ab
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 26 16:11:21 2017 -0500
Merge pull request #18540 from liewegas/wip-pr-17894-luminous
luminous: ceph.spec.in,debian/rules: change aio-max-nr to 1048576
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Boris Ranto <branto@redhat.com>
commit 565b5cd1d68437096e283ccb942d8f108c26d3d4
Author: Alfredo Deza <adeza@redhat.com>
Date: Wed Sep 13 08:23:40 2017 -0400
ceph-volume lvm add listing module to main
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit ff72fd0ed52ef2d166007c7769c0140424982dc7)
commit 58095350953140a1fa4b968c2e0a5774a960e739
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 3 11:41:10 2017 -0400
doc/ceph-volume add lvm/list to the toctree
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit d11c80c917f44b33077715b19564aeea30c78e42)
commit 3893947d3e04f049a85245750b841adbbcefeac8
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 3 11:36:06 2017 -0400
doc/ceph-volume create the reference for the list sub-command
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 9a52939ef8fc4abe720d2525c59997f4f9760902)
commit 0992c3d7c136a58ff38f521fc0978ff036843d27
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 3 11:35:39 2017 -0400
doc/ceph-volume add create and list to the index
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit ed5fa519aca1335039bfc32bc060af5181265517)
commit 148537cc333d88b75640506c15a1bc221c9e09d6
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 3 08:59:37 2017 -0400
ceph-volume tests create a unit test module for listing features
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 2c2cf547b07f9510146b86d0746190b6d3fead3d)
commit 285bc06569c8ed06890d23e9537a9d1106825b7f
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Oct 3 08:59:01 2017 -0400
ceph-volume tests create a small factory fixture
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 5a0b17de81770ef3b851a1ff98232d93c78f49f9)
commit 1e4e241d9c77796874a003799a0546fc185c5241
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Oct 2 14:40:41 2017 -0400
ceph-volume tests ensure Volume.as_dict gets populated correctly
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit c5b933a96d067fb8c7f753a67123e5662398f964)
commit a0ab94e8d70ae9fc5d741c47a8f0dcfe033f9a5e
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Oct 2 14:31:57 2017 -0400
ceph-volume lvm.listing initial take on listing command
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 2d80190a039798a637383513f292e6aab62cd3f8)
commit 454c655f2eb0b1e79ca8e1e1d87ff9c0fe4886da
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Oct 2 12:35:13 2017 -0400
ceph-volume lvm.api be able to return a dictionary repr of a Volume object
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 0cae22750f396d1c1664875edb2742c50a41bffd)
commit 1390fb76d76b76428e6a9210005dbf01fd3b3a1a
Author: Alfredo Deza <adeza@redhat.com>
Date: Mon Oct 2 12:12:57 2017 -0400
ceph-volume process allows to fully mute terminal output on commands
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 233f64a4fd90a19c50d57b849ffaa6a25c835f34)
commit 00e9ec1fa25ac3950f43ec3cf84f8321b04507eb
Author: Alfredo Deza <adeza@redhat.com>
Date: Thu Sep 14 07:35:47 2017 -0400
ceph-volume lvm.api use a helper to get an lv from an arg
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 3c5bbbc5822f02a038f093647c5de58bef2fe8f5)
commit 26c523c3beeb5da7e855be64a053ac9dca7d10e2
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Oct 23 14:53:30 2017 -0400
rbd-mirror: strip environment/CLI overrides for remote cluster
Fixes: http://tracker.ceph.com/issues/21894
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 70dc22e03a115b130b9c214030d5996276a1634b)
commit 9a28a5c9d7ae618f5b79d3727ba155b5066a2653
Merge: 42172a4431 18a99f5f6b
Author: Josh Durgin <jdurgin@redhat.com>
Date: Wed Oct 25 11:45:38 2017 -0700
Merge pull request #18456 from liewegas/wip-21882-luminous
luminous: messages/MOSDMap: do compat reencode of crush map, too
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 18a99f5f6b4976f87dcd0d4fe7e34fddd90de22b
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 25 07:00:14 2017 -0500
qa/suites/upgrade/jewel-x: make sure min compat client is hammer for straw2
Signed-off-by: Sage Weil <sage@redhat.com>
commit cec24f99ab5e368d2dae7d53628fee09515bd873
Author: chenliuzhong <liuzhong.chen@easystack.cn>
Date: Tue Oct 24 10:54:33 2017 +0800
ceph.spec.in,debian/rules: change aio-max-nr to 1048576
when osd is more than 14 in one host,it report error that aio is not enough.
As the default aio-max-nr is 65536, one OSD needs 4096 aios and other programs may use aios.
This patch change aio-max-nr to 1048576 when install ceph-osd rpm package and debian package
Signed-off-by: chenliuzhong <liuzhong.chen@easystack.cn>
(cherry picked from commit 36326dc7104fc2f20f19d51b6f618a029ba072d7)
commit 15a52ba8329831ddf350664e6e6612523f98dcf4
Author: baixueyu <baixueyu@inspur.com>
Date: Wed Oct 18 11:16:20 2017 +0800
You can find the problem do like this:
upload obj to bucket1
s3cmd put obj s3://bucket1
cp obj from bucket1 to bucket2
s3cmd cp s3://bucket1/obj s3://bucket2
del obj from bucket1 and bucket2
s3cmd del s3://bucket1/obj
s3cmd del s3://bucket2/obj
you can see the pool 'default.rgw.buckets.data', the data of obj can not deleted.
ceph df | grep default.rgw.buckets.data
Fixes: http://tracker.ceph.com/issues/21819
Signed-off-by: Xueyu Bai baixueyu@inspur.com
(cherry picked from commit 2e11545077c485d5c155ab932b6274a8ac863e61)
commit 4c6b1b65f7337809682bf907be94fe946909959a
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Oct 19 22:13:36 2017 -0400
common/common_init: disable ms subsystem log gathering for clients
The log gathering causes large performance degradation to clients
with high message throughputs. This is hopefully a short-term
workaround until per-message logging can be replaced with an
efficient data recording system for post-incident analysis
use-cases.
Fixes: http://tracker.ceph.com/issues/21860
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit a3a40413f7908b08c40dec4020034cca4a0c4798)
commit 42172a443183ffe6b36e85770e53fe678db293bf
Merge: f2afb7a0d5 9f49698fab
Author: Sage Weil <sage@newdream.net>
Date: Tue Oct 24 21:36:18 2017 -0500
Merge pull request #18503 from liewegas/wip-21878-luminous
luminous: os/bluestore/BlueFS: fix race with log flush during async log compaction
Reviewed-by: Varada Kari <varada.kari@gmail.com>
commit f2afb7a0d590ff55ae11f5bf3b13264a56380ce6
Merge: d0be2106f9 3ad21109ed
Author: Sage Weil <sage@newdream.net>
Date: Tue Oct 24 21:35:29 2017 -0500
Merge pull request #18501 from liewegas/wip-21766-luminous
luminous: os/bluestore: handle compressed extents in blob unsharing checks
commit d0be2106f96377f059ec08800e32e492ec6f82a3
Merge: d52e59bad7 e6eeb796b4
Author: Sage Weil <sage@newdream.net>
Date: Tue Oct 24 21:34:52 2017 -0500
Merge pull request #18459 from liewegas/wip-pr-18457-luminous
luminous: qa/suites/rados/rest/mgr-restful: whitelist more health
commit 902d467a0ec23baebf28a2419a4c0c78e0b1b662
Author: Casey Bodley <cbodley@redhat.com>
Date: Mon Oct 9 10:11:14 2017 -0400
qa/rgw: ignore errors from 'pool application enable'
Fixes: http://tracker.ceph.com/issues/21715
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 160240e0c1c216c950b7210a561b27436198de1f)
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit 1751a094d76ac84fbcecfc9816904cbb496bd765
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 23 17:11:59 2017 -0500
osd/PrimaryLogPG: clear DEGRADED at recovery completion even if more backfill
We may have log recovery *and* backfill to do, but cease to be degraded
as soon as the log recovery portion is done. If that's the case, clear
the DEGRADED bit so that the PG state is not misleading.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3e91fda79d3be771f46394312a6b72ab75660012)
commit d52e59bad700e8add6cd4cf895817aa8fd7a601a
Merge: 724a51cd44 0048e6a58c
Author: Kefu Chai <tchaikov@gmail.com>
Date: Tue Oct 24 11:51:19 2017 +0800
Merge pull request #18491 from liewegas/wip-21573-luminous
luminous: buffer: fix ABI breakage by removing list _mempool member
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 9f49698fab5de9ade79d3aba957afe82daa4fecc
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 20 08:51:17 2017 -0500
os/bluestore/BlueFS: fix race with log flush during async log compaction
During async log compaction we rely on _flush-and_sync_log to update the
log_writer to jump_to. However, if racing threads are also trying to flush
the log and manage to flush our new log events for us, then our flush will
turn into a no-op, and we won't update jump_to correctly at all. This
results in a corrupted log size a bit later one.
Fix by ensuring that there are no in-progress flushes before we add our
log entries. Also, add asserts to _flush_and_sync_log to make sure we
never bail out early if jump_to is set (which would indicate this or
another similar bug is still present).
Fixes: http://tracker.ceph.com/issues/21878
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 4324c8bc7e66633035c15995e3f82ef91d3a5e8c)
commit 3ad21109ed2df068bb113e8551505b850fbfcb30
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 11 16:48:41 2017 -0500
os/bluestore: handle compressed extents in blob unsharing checks
If the blob is compressed, we aren't mapping to a range within
the allocated extents, but rather referencing the entire blob.
Fixes: http://tracker.ceph.com/issues/21766
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit eb26cfbfbb2b9b9fe84dfe9c999e2d67fd2661c0)
commit 8f176f31d091cb53e1baa8c317df9e64bef4d95e
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 22 22:46:00 2017 -0500
osd/PG: on recovery done, requeue for backfill
We were keeping our existing recovery reservation slot (with a high
priority) and going straight to waiting for backfill reservations on
the peers. This is a problem because the reserver thinks we're doing
high priority work when we're actually doing lower-priority backfill.
Fix by closing out our recovery reservation and going to the
WaitLocalBackfillReserved state, where we'll re-request backfill at the
appropriate priority.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 09c50e2fe4fc3e8951a483e4b2693f0ffc008dcf)
commit cc707155ea475836f97791c08c31994fb40e4374
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 22 22:43:18 2017 -0500
osd/PG: move local_reserver recovery cancel to Recovering state transition
This is easier to follow than canceling the reservation in the next state.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit dde007749bb960122f399b4cb8608a8c1ea4eb9c)
commit def29079dc623926fed26a4946d6b9a074dfa2ac
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 20 22:55:29 2017 -0500
qa/suites/upgrade/jewel-x/parallel: run some jewel after completed upgrade
Signed-off-by: Sage Weil <sage@redhat.com>
commit 8697eca5ea24c4724f637e4d4f75d3e0d2b95278
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 20 22:52:45 2017 -0500
qa/suites/upgrade/jewel-x/: set up compat weight-set after cluster upgrade
...but before we run some client workloads. This will hopefully ensure
that older clients will behave.
Signed-off-by: Sage Weil <sage@redhat.com>
commit 89ee508ac51839c3d1f71c5e21443e8f2679c59e
Author: Sage Weil <sage@redhat.com>
Date: Sat Oct 21 13:03:17 2017 -0500
mon/OSDMonitor: add 'osd crush set-all-straw-buckets-to-straw2'
This is a common and recommended step that users should take once they
have clients that support straw2 (hammer!). Let's make it easy.
This may result in a small amount of data movement, but usually it is
very little.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 928770513cbe3a244095e36dcb41c83c0724cfe9)
commit 94a4ca6c8e4d26465cbefd1492ee05f21b08ff1e
Author: Sage Weil <sage@redhat.com>
Date: Sat Oct 21 12:59:16 2017 -0500
crush/CrushWrapper: add bucket_set_alg()
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit edc411ee4cb7b88a970ccfc9d29b3a3ba15d066c)
commit 0048e6a58c7cdf3b3d98df575bc47db8397cd5a9
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 19 16:19:35 2017 -0500
buffer: remove list _mempool member
This broke the C++ ABI by changing the list structure size. Also, it's
not necessary as we can infer the mempool by looking at the other list
contents. We don't (currently) have a need to map an empty list to a
particular mempool and have that state stick.
Fixes: http://tracker.ceph.com/issues/21573
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 9b92d87d37e3892b0096cd728b46154aed1e2d86)
commit 3cd334b6264998490e6fb8321fb7242ff82b36e6
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 19 16:09:10 2017 -0500
buffer: allow mempool to be passed into raw* ctors and create methods
This is more convenient, and also faster than initializing it in
buffer_anon and the immediately moving it elsewhere.
Drop the optionality of the alignment argument.
No users yet.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 481277b8c92368b8cac8fdafe1848720ec1035c2)
commit 724a51cd442d35c950414b86608c48936b01b78b
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 23 12:05:39 2017 -0500
/etc/sysconfig/ceph: remove jemalloc option
This breaks when used with rocksdb, which is now the default.
See http://tracker.ceph.com/issues/20557
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 19efccef248c2d66fe3d7fdac8f56d455f2c7b27)
commit 95b6bc9ac41c4985ad97a515c4105351a0e6b94d
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 23 11:16:26 2017 -0500
etc/default/ceph: remove jemalloc option
This breaks when used with rocksdb, which is now the default.
See http://tracker.ceph.com/issues/20557
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit fba2055b4e54128db1e9a83f27bb3526d65d4db7)
commit e6eeb796b4d24fcf6797519be06aca9771e73028
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 20 23:14:36 2017 -0500
qa/suites/rados/rest/mgr-restful: whitelist more health
The test is fiddling with OSDs.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ddf3e9162da542af0c5f025957b8304e7359c924)
commit 33aa1673316b77b71e473d1ab71cf61587e425a6
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 5 22:46:48 2017 -0400
mon/OSDMonitor: improve crush map validation
- move into OSDMap method
- ensure that rules exist for each pool
- ensure pool type matches rule type
- ensure rule mask min/max size cover the pool size
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 8f8f0b25058ff7319ec9d023f0eceb5832d0ecff)
commit d825de7681ba1cff257ba4108d6c22c81e3a6aab
Author: John Spray <john.spray@redhat.com>
Date: Fri Aug 25 11:06:21 2017 +0100
mon: more forceful renumbering of legacy ruleset IDs
Previously, the rules were only modified in the trivial case,
so we continued to potentially have CRUSH maps with the
legacy ruleset functionality in use.
In order to ultimately remove rulesets entirely, we need
to do this more aggressively, renumbering all the rules
and then updating any pools as needed.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 71d4b2bed54371657693cfb999ade44449be0efd)
Conflicts:
src/mon/OSDMonitor.cc: the check for multiple rules was removed
in master, but not in luminous. once we renumber the legacy ruleset IDs,
it's not need to check for and to warn the user at seeing the case where
1-to-n mapping from ruleset to rule IDs.
commit f52d1ad23d97e1ab8c70f9b17b75daa0773698b7
Author: John Spray <john.spray@redhat.com>
Date: Fri Aug 25 10:42:58 2017 +0100
osd: s/crush_ruleset_in_use/crush_rule_in_use/
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 675483ecca06686fd4c626bfea33748092aef0c6)
commit 17d73e5a2e592719df8099450c9cd5f46d6fd582
Merge: d4158133f7 e25a583323
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 20 23:10:49 2017 -0500
Merge pull request #17773 from xiexingguo/wip-luminous-object-legacy-flag
luminous: osd/PrimaryLogPG: _delete_oid - fix incorrect 'legacy' flag
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit d4158133f7586472d01d9c6e669db92ddcd2e73d
Merge: 84fbb0ff70 8b4fe14ba0
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 20 23:10:06 2017 -0500
Merge pull request #18362 from kmroz/wip-21543-luminous
luminous: qa/tasks: prolong revive_osd() timeout to 6 min
commit 84fbb0ff705df5bc9f164dd5a02bd5a82e6478d7
Merge: 87ab8d5c80 993d098505
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 20 23:09:52 2017 -0500
Merge pull request #18388 from tchaikov/wip-pr-18346-luminous
os/filestore: print out the error if do_read_entry() fails
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
commit 87ab8d5c80374a73ad95f9ed9dd8826c54e64e9f
Merge: 00daa8462a fd42f2a9cc
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 20 23:09:17 2017 -0500
Merge pull request #18398 from smithfarm/wip-21783-luminous
luminous: cli/crushtools/build.t sometimes fails in jenkins' make check run
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 00daa8462a5a337ecc57bd8664bfcc748befeead
Merge: 5a4850c8c6 112069c70d
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 20 23:08:58 2017 -0500
Merge pull request #18401 from liewegas/wip-bluestore-label-luminous
luminous: os/bluestore: ceph-bluestore-tool repair updates and prime-osd-dir support
commit dbddf21bd3aac284122fd5fc84abdd4c89b12b3e
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 20 22:32:33 2017 -0500
messages/MOSDMap: do compat reencode of crush map, too
If we are reencoding an incremental, and it embeds a crush map, we need
to reencode that in a compatible way too. This is especially true now
because we have the compat crush weight-sets. Otherwise, a client may
learn the crush map through an incremental but not understand choose_args,
and not see/understand the alternate weight set. It will send requests
to the wrong OSDs where they will just get dropped.
Fixes: http://tracker.ceph.com/issues/21882
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 9eaafc66278d59447b29ce6cc378d0cd3ad14ad4)
commit c5ccbf1108532547f54f084b8f06c7f5ddf8a4e8
Author: Ramana Raja <rraja@redhat.com>
Date: Sun Sep 24 20:23:12 2017 +0530
ceph_volume_client: fix setting caps for IDs
... that have empty OSD and MDS caps. Don't add a ',' at the
start of OSD and MDS caps.
Fixes: http://tracker.ceph.com/issues/21501
Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit baf3b88800c63ef7467abbc7b54e40c0da669f38)
commit b382db7cfce84adc897cd68e4cd7a4eb1b40a9f1
Author: Brad Hubbard <bhubbard@redhat.com>
Date: Wed Sep 20 13:15:30 2017 +1000
mon/OSDMonitor: mon osd feature checks with 0 up osds
get_up_osd_features() returns 0 if no osds are up which causes feature
checks to fail.
Fixes: http://tracker.ceph.com/issues/21471
Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
(cherry picked from commit 26ba0ba044846d18f9bdfdbee6e2a9c46882c742)
Conflicts:
src/mon/MonCommands.h
- removed mimic reference
src/mon/OSDMonitor.cc
- removed mimic-only command path
commit c845db3f8c2a0c2494fbdc9d6cc63c8e0c3b0563
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 19 18:25:56 2017 -0400
osd/OSDMap: ignore xinfo if features == 0
Some old bug (e.g., http://tracker.ceph.com/issues/20751) could
result in an UP+EXISTS osd having features==0. If that happens,
we shouldn't crash the mon, which (reasonably) does
if (osdmap.get_epoch()) {
if (osdmap.get_num_up_osds() > 0) {
assert(osdmap.get_up_osd_features() & CEPH_FEATURE_MON_STATEFUL_SUB);
check_subs();
}
}
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 8399833f3d221364a8ededb97cd4e1c5a258d924)
commit e58656005d681ee0e35a25b324321cfeea8c7b3a
Author: Matt Benjamin <mbenjamin@redhat.com>
Date: Tue Oct 17 17:45:37 2017 -0400
rgw_file: explicit NFSv3 open() emulation
Expect RGW_OPEN_FLAG_V3 in rgw_write(), and if present, attempt
a stateless open inline.
Required by Ganesha v2.5 and later
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
(cherry picked from commit 6ed23b4a0cea7e41b4743e27319737af693947ec)
commit f9850d2148b396032634f7ffa18ae458bd208195
Author: Adam C. Emerson <aemerson@redhat.com>
Date: Thu Sep 28 13:54:32 2017 -0400
rgw: Check bucket Website operations in policy
Add code to check s3:GetBucketWebsite and s3:PutBucketWebsite
operations against bucket policy.
Fixes: http://tracker.ceph.com/issues/21597
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1493896
Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
(cherry picked from commit ceed535957ac186e241fcff26b103cf7efa959b1)
commit abefb578c345a0ddd7100cae987e582b530d3729
Author: Adam C. Emerson <aemerson@redhat.com>
Date: Wed Sep 27 15:42:27 2017 -0400
rgw: Check bucket CORS operations in policy
Add code to check s3:GetCORS and s3:PutCORS operations against bucket
policy.
Fixes: http://tracker.ceph.com/issues/21578
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1494140
Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
(cherry picked from commit 27eb13fe568cc802feaf69131a21db076bcb6746)
commit 0b84dd07c41a6d69d99672218c72a4f69c21b040
Author: Adam C. Emerson <aemerson@redhat.com>
Date: Wed Sep 27 16:08:56 2017 -0400
rgw: Check bucket GetBucketLocation in policy
Add code to check s3:GetBucketLocation against bucket policy.
Fixes: http://tracker.ceph.com/issues/21582
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1493934
Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
(cherry picked from commit 79188d679edeb6e2f7ca852fdc4224368412cb72)
commit 0b3a974be7bb9cd1d25ee9c2b861f38ea721c66b
Author: Casey Bodley <cbodley@redhat.com>
Date: Wed Sep 27 14:26:25 2017 -0400
rgw: include SSE-KMS headers in encrypted upload response
Fixes: http://tracker.ceph.com/issues/21576
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit a5b3db7acb4150d9d27ba42f72658e418dd444c0)
commit a3e2ae4925e7e25c213ef37072d576d1f1ffdc13
Author: Abhishek Lekshmanan <abhishek@suse.com>
Date: Fri Sep 22 16:11:55 2017 +0200
rgw: defer constructing keystone engine unless url is configured
currently we create a keystone revocation thread even when keystone url
is empty, lets defer the construction of keystone unless the urls are
configured
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
(cherry picked from commit 104c2f59011e6b515e3845cce529ee44334e76c0)
commit 4c18ac124048f72a10d92aecfe1eadabab0a80b4
Author: Adam C. Emerson <aemerson@redhat.com>
Date: Wed Sep 27 14:35:59 2017 -0400
rgw: Check bucket versioning operations in policy
Add code to check s3:GetBucketVersioning and s3:PutBucketVersioning
operations against bucket policy.
Fixes: http://tracker.ceph.com/issues/21389
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1490278
Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
(cherry picked from commit 16de0fc1c5ede961ebd481f13753214a971c206c)
commit be1ded107ac4e1a1b1515d40553aed3581e0afac
Author: Adam C. Emerson <aemerson@redhat.com>
Date: Thu Sep 14 18:07:15 2017 -0400
rgw: Check payment operations in policy
Add code to check s3:GetBucketRequestPayment and
s3:PutBucketRequestPayment operations against bucket policy.
Fixes: http://tracker.ceph.com/issues/21389
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1490278
Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
(cherry picked from commit f9d1ae1d153319e870c3ccaf7afdc92786cdaa3b)
commit 92d0545781ca2bff39f7bd8e9d63bb8c2d8cb997
Author: Nathan Johnson <njohnson@ena.com>
Date: Thu Sep 7 23:27:40 2017 -0500
rgw: admin api - add ability to sync user stats from admin api
Fixes: http://tracker.ceph.com/issues/21301
Signed-off-by: Nathan Johnson <nathan@nathanjohnson.org>
(cherry picked from commit 828412d573a0b31387fc9f6a879f9a6083535730)
commit d800747fa18a1872f1d748119bfcc7eff34ae6cf
Author: gaosibei <gaosb@inspur.com>
Date: Tue Sep 26 10:19:22 2017 +0800
RGW: fix a bug about inconsistent unit of comparison
Fixes：http://tracker.ceph.com/issues/21590
Signed-off-by: gaosibei <gaosb@inspur.com>
(cherry picked from commit cb39f065d6eba87cf1fc2e99334322bf63092df3)
commit 14a1dcb1ff70b2a468cb47e7f239b98e45c63a88
Author: Mark Kogan <mkogan@redhat.com>
Date: Mon Sep 25 09:53:00 2017 +0300
rgw: update the usage read iterator in truncated scenario
Fixes: http://tracker.ceph.com/issues/21196
Signed-off-by: Mark Kogan <mkogan@redhat.com>
(cherry picked from commit 7306514a813661b77bfcbfc6f534dbabbdd3aa78)
commit 8b420c4627faca27f61f472597c6f67cca2efa2f
Author: Casey Bodley <cbodley@redhat.com>
Date: Thu Oct 5 16:39:30 2017 -0400
rgw: RGWUser::init no longer overwrites user_id
if an admin op specifies a user_id and does not find a user with that
id, but does find a user based on a later field (email, access key,
etc), RGWUser::user_id will be overwritten with the existing user's id
when this happens on 'radosgw-admin user create', RGWUser::execute_add()
will modify that existing user, instead of trying to create a new user
with the given user_id (and failing due to the conflicting email,
access key, etc)
by preserving the original user_id (when specified), this uid conflict
is detected in RGWUser::check_op() and a "user id mismatch" error is
returned
Fixes: http://tracker.ceph.com/issues/21685
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 4e65c9e9af86e832cb96d57d487771aa28251e7c)
commit fe45086a54d76cd9b913663d4537b36a8a276ed4
Author: gaosibei <gaosb@inspur.com>
Date: Tue Sep 26 10:44:09 2017 +0800
RGW: Multipart upload may double the quota
Fixes: http://tracker.ceph.com/issues/21586
Signed-off-by: Sibei Gao <gaosb@inspur.com>
(cherry picked from commit 97f95e457fb7f9e36031f41e0f2dd3955daedd87)
commit c3a719da278a830cc1b8101e3d8ef25316da33d7
Author: Casey Bodley <cbodley@redhat.com>
Date: Thu Oct 12 13:26:29 2017 -0400
rgw: 'zone placement' commands validate compression type
Fixes: http://tracker.ceph.com/issues/21775
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 33f8ee8b6fa5605edbbfe12d8e097123e509ad9f)
commit 352373ff4a333c9b3869c9f168ee14530067907a
Author: Yao Zongyou <yaozongyou@vip.qq.com>
Date: Sun Sep 24 22:11:22 2017 +0800
rgw_file: fix write error when the write offset overlaps.
Signed-off-by: Yao Zongyou <yaozongyou@vip.qq.com>
(cherry picked from commit 872d73f945364002f0fa31762e6976db5b4b3c19)
commit 248bd94dab4359dfea34bfa0cbdd16d9a9b76a02
Author: Orit Wasserman <owasserm@redhat.com>
Date: Mon Oct 9 13:25:36 2017 +0300
rgw: disable dynamic resharding in multisite environment
Fixes:http://tracker.ceph.com/issues/21725
Signed-off-by: Orit Wasserman <owasserm@redhat.com>
(cherry picked from commit f1ec7a49bde054a19a0a8fd80f37866e1e40379f)
commit 624b3ac5053dbef008d2e9238c6958d07206f6c3
Author: Zhang Shaowen <zhangshaowen@cmss.chinamobile.com>
Date: Tue Sep 26 10:00:59 2017 +0800
rgw: Torrents are not supported for objects encrypted using SSE-C
Fixes: http://tracker.ceph.com/issues/21720
Signed-off-by: Zhang Shaowen <zhangshaowen@cmss.chinamobile.com>
(cherry picked from commit 9a8ae664da54eb74e74a12cc16a52b0df7df70a0)
commit 0e65ea466f05a2bc8825e51b32901d74082c1b65
Author: Matt Benjamin <mbenjamin@redhat.com>
Date: Tue Oct 3 17:48:29 2017 -0400
rgw: release cls lock if taken in RGWCompleteMultipart
Follows Casey's proposal to conditionally release the lock in
::complete(), in order to avoid duplicated code in various early
return cases.
Fixes: http://tracker.ceph.com/issues/21596
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
(cherry picked from commit 704f793f08a02760d23eb5778b738bb07be0e7cf)
commit 17df0ba8f00abbf4b30e0370090251b0c7f6b643
Author: Enming Zhang <enming.zhang@umcloud.com>
Date: Sat Aug 26 04:46:35 2017 +0800
rgw: encryption add exception handling for from_base64 on bad input
If calling from_base64 with charactors that is not base64 encoded,
rgw will crash.
Signed-off-by: Enming Zhang <enming.zhang@umcloud.com>
(cherry picked from commit f0b88e51be86d84793b794f6fe87fdda62004a4d)
commit 46f09e3803ba3c4b463bfa6a2eeb40a0bd7cc0ef
Author: Enming Zhang <enming.zhang@umcloud.com>
Date: Fri Aug 25 19:48:53 2017 +0800
rgw: encryption fix the issue when not provide encryption mode
Now, in RGW, if someone want to upload an object using server-side
encryption with providing customer key or kms key id, but not
specify the encryption mode in the
"x-amz-server-side-encryption-customer-algorithm" or
"x-amz-server-side-encryption", the object will be uploaded
successfully without encryption.
This is not a correct way to deal with it. It is better to
return error.
Fixes: http://tracker.ceph.com/issues/21581
Signed-off-by: Enming Zhang <enming.zhang@umcloud.com>
(cherry picked from commit f8c4be8a8943096217d89f7c5abd129fa2414b36)
commit fefbc9fec47cc6858f74b7ea372d39efeb95c61f
Author: Enming Zhang <enming.zhang@umcloud.com>
Date: Fri Aug 25 19:47:21 2017 +0800
rgw: encryption SSE-KMS add the details of error msg in response
Signed-off-by: Enming Zhang <enming.zhang@umcloud.com>
(cherry picked from commit 3f16785daf4b3609f2bff7c107a35f908103e32a)
commit 2e493e1f7a1f20d2f33e66816cc72d6b5f193b61
Author: Enming Zhang <enming.zhang@umcloud.com>
Date: Fri Aug 25 19:37:52 2017 +0800
rgw: encryption SSE-C add the details of error msg in response
Signed-off-by: Enming Zhang <enming.zhang@umcloud.com>
(cherry picked from commit 9502549ac2133e969a5f268601b92fd4063c1bd9)
commit fe0a41b4a921ac3510b344a6edf6e930a2fa2465
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Oct 12 08:58:03 2017 -0400
cls/journal: fixed possible infinite loop which could kill the OSD
Fixes: http://tracker.ceph.com/issues/21771
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 9e7475f8fb08b0ef30cad2c0d680cecf331b44ef)
commit e0d57c544d55f3693f858464f7995b4b30cfa88d
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Oct 12 08:56:57 2017 -0400
test: ceph_test_cls_journal was dropped when converting to cmake
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 5d9cfebabac33b645c6199bda63ff4619e6f538f)
commit f8bfd9458d44d96d4ef56a28195a9e36fc0190c6
Author: Jason Dillaman <dillaman@redhat.com>
Date: Sun Oct 15 17:26:25 2017 -0400
librbd: batch large object map updates into multiple chunks
Fixes: http://tracker.ceph.com/issues/21797
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 04c5d12acc045731fbf1d0ddce276b5743d9fd89)
commit 6866abc01be5fd92acc918472a604a3af9894b44
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Oct 13 18:55:47 2017 -0400
test/librbd: initial test cases for trim state machine
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 4674b578d8233da3b3c145606ed9c0e4e88a540a)
commit 3b87dea6584010ae05e540825867c9961708ff36
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Oct 13 15:49:00 2017 -0400
librbd: tweaks to support testing of trim state machine
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 3220480d4a22c81cb74f25e528f2bcf3986342c1)
commit ce628329641ac6322d5c59055e8db9eecafbfef4
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Oct 13 14:44:40 2017 -0400
librbd: combine trim state machine object map batch update states
The PRE/POST states were previously divided into two halves for handling
the copy-up batch and the direct removal batch. This can be simplified by
just using a single PRE/POST for the entire deletion region.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 72ce4576fa2b562799a5bc78bd423cfabe097d67)
Conflicts:
src/librbd/operation/TrimRequest.h
commit 80ff76db5495ae41baf6d4292b6b10c2f789c155
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Oct 13 11:06:38 2017 -0400
cls/rbd: object map update now utilizes constant-time bit vector operations
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit fc99603ea2b5b268181d38507b8b6fb55ae0a6d0)
commit 5dc46378370da8eede9958a88d6764737d55b09b
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Oct 13 11:05:48 2017 -0400
common/bit_vector: provide constant time iteration of underlying bufferlist
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit daa29f7d2b50bc4b44a108a0f91d45ce90fc16e5)
commit ae4ebaca231e4ae94efc8d445fee792f76e656e1
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Oct 13 11:03:25 2017 -0400
common/buffer: expose hidden const deference operator
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit f35947e3c6661dfe9f951896bc54bc8da158b490)
commit b60d8cf0cbe56500af98b3d3857245e94d904b19
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Oct 2 17:28:04 2017 -0400
osd: additional protection for out-of-bounds EC reads
Fixes: http://tracker.ceph.com/issues/21629
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 881f2a02aecef0cd6bf8ec3e5045b76c8441e832)
commit 8f87fa2d3af44840a3044cd736a2f8da0b602b07
Author: John Spray <john.spray@redhat.com>
Date: Sat Sep 23 13:21:47 2017 -0400
mgr: fix crashable DaemonStateIndex::get calls
This function was recently fixed to return null
on missing entries: handle that properly.
Fixes: http://tracker.ceph.com/issues/17737
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit ade4827d86bb2bc79466a2da040475651c2aae0d)
commit def3d55eb71e0c62b7fd523865c2ba196342aa1a
Author: huangjun <huangjun@xsky.com>
Date: Fri Oct 6 11:58:50 2017 +0800
qa/suites/rest/basic/tasks/rest_test: whiltelist OSD_DOWN
Fixes: http://tracker.ceph.com/issues/21425
Signed-off-by: huangjun <huangjun@xsky.com>
(cherry picked from commit 838e12cfc84b386fd4eae09e526b078286f0771e)
commit 20c9e641f7096bf3e208b4d6d28494603eeba891
Author: huangjun <huangjun@xsky.com>
Date: Tue Sep 19 16:13:34 2017 +0800
qa/suites/rest/basic/tasks/rest_test: more whitelisting
Fixes: http://tracker.ceph.com/issues/21425
Signed-off-by: huangjun <huangjun@xsky.com>
(cherry picked from commit 4d30d02e522039461762d51a3a826a337214ca7e)
commit 5a4850c8c66b75d02fc2597e1ca744c6b0303836
Merge: 17c6f35703 03e2a68e43
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 19 16:43:39 2017 -0500
Merge pull request #17692 from smithfarm/wip-21176-luminous
build/ops: python-numpy-devel build dependency for SUSE
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Reviewed-by: Amit Kumar <amitkuma@redhat.com>
commit 17c6f357030315040f8af95e751921c475b32289
Merge: 97b30154f8 5aa445897c
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 19 16:42:45 2017 -0500
Merge pull request #17730 from xiexingguo/wip-pr-17371
luminous: mon, osd: per pool space-full flag support
Reviewed-by: Sage Weil <sage@redhat.com>
commit 97b30154f884f4ea987e3330799bf412b08ba3e4
Merge: 0a8f1540ba fb7e6a581d
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 19 16:42:07 2017 -0500
Merge pull request #17779 from smithfarm/wip-rh-74-luminous
tests: CentOS 7.4 is now the latest
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
commit 0a8f1540ba08bb79e9ec8f3358490a31be734b2b
Merge: 5ca7af785a d7cde5f316
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 19 16:41:34 2017 -0500
Merge pull request #17863 from smithfarm/wip-21343-luminous
luminous: mon: DNS SRV default service name not used anymore
commit 5ca7af785a6695121f841909a6118a5103795d99
Merge: 37d7cfc9f1 49e030da55
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 19 16:40:45 2017 -0500
Merge pull request #17864 from smithfarm/wip-21438-luminous
luminous: core: Daemons(OSD, Mon...) exit abnormally at injectargs command
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 37d7cfc9f136c9604a1e53604a8e0d939eee60db
Merge: 223b0956b9 077b3601d6
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 19 16:40:18 2017 -0500
Merge pull request #17866 from smithfarm/wip-21320-luminous
luminous: mgr: Quieten scary RuntimeError from restful module on startup
commit 223b0956b93504881a5f06d2589a0b138abf37a9
Merge: 4f1a7a82a1 82651da2d5
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 19 16:39:53 2017 -0500
Merge pull request #17951 from smithfarm/wip-21548-luminous
luminous: tests: ceph_manager: bad AssertionError: failed to recover before timeout expired
commit 4f1a7a82a1134d6688537653ac07d924275e1f3a
Merge: 37b13acabe a670921f4f
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 19 16:39:32 2017 -0500
Merge pull request #18342 from xiexingguo/wip-pr-18318
luminous: bluestore: using bluestore_compression_min_blob_size instead of bluestore_compression_max_blob_size to config comp_min_blob_size
Reviewed-by: Sage Weil <sage@redhat.com>
commit 112069c70df5c7719af9291a8de2b4fcf48d049a
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 16 17:46:03 2017 -0500
os/bluestore: fsync on _write_bdev_label
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 7aca8af6625156eeaa936629f83ede33d3bbe866)
commit 1e9aa60e3f6acca6649f5d5d51421cc2bd51e55a
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 16 17:45:54 2017 -0500
ceph-bluestore-tool: implement 'set-label-key' and 'rm-label-key'
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 4b468c989f9e8609e7f640cfa4385c454ae6dc17)
commit 170499a1e570bc110be08ca2a38063229fee75dc
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 16 17:42:29 2017 -0500
os/bluestore: make _write_bdev_label static and public
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 0b0550c967ae0d6ae49641c6859f0b63267e2cd9)
commit 895a277568f902f1c00b9df475ea56b00b115841
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 12 11:40:11 2017 -0500
doc/man/8/ceph-bluestore-tool: add man page
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 7b91e50dbd289e65f10cb98c5eea9f3171c2d1f1)
commit 24b3b7960e5c5d118aacb63753c794466e801f84
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 13 09:20:53 2017 -0500
common/options: document 'key', 'keyfile', 'keyring'
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit cdee8b26b9d93189dcd2875038361fb59ed21ba2)
commit 508889d4a7aa93f3bbe13ad8bfc181f3fc9ed2a9
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 12 13:54:06 2017 -0500
ceph-bluestore-tool: fix show-label to use dev name as key
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 4785345ace7b675a0e1c34c457aad1103aab5a1d)
commit 6e17557f288aa7686b8b68c1df31a9a8c26912ac
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 12 08:48:59 2017 -0500
ceph-bluestore-tool: drop "action ..." print
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 5e486c2bd3414f4ee6519574d4d4584618144867)
commit a128aad320e99c396ff17bcaadb48866667b87df
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 12 08:48:46 2017 -0500
ceph-bluestore-tool: only infer devs if devs.empty()
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2d217367af0df5e18394b1c6179d92db3c0b4448)
commit 24d86603faf4a826684a5e124ed9b447de0b50e8
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 11 17:36:40 2017 -0500
ceph-bluestore-tool: implement prime-osd-dir
This populates an osd dir based on the metadata in a bluestore
device label.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 72ee7e29fb77f581a537b1ff2b274c9f3a7de53f)
commit 0f43e6907cfd27f2d6791b896480f2c537115212
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 11 16:33:54 2017 -0500
os/bluestore: keep meta items for extra bdev paths
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 8d07fa5ad0772b4d979dc10695d230074d1f1e69)
commit bb9b70c33a3e013ef78d931fd865ab44c255d772
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 11 17:33:00 2017 -0500
vstart.sh: specify block files
This is mostly just to test ceph-bluestore-tool prime-osd-dir.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d4e0c46d89c2ebe6ae5f9af1c909a09daaea9687)
commit ce83461a60cea6026a7abdd51b2a7a35302392c6
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 11 16:32:16 2017 -0500
osd: store osd cephx key in "osd_key" meta slot
Note that this *only* works with --mkkey (deprecated except for
vstart.sh) and --key <key>. If you write out a full keyring before
running --mkfs it doesn't get populated. Users should pass --key
instead of --keyring in that case.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 65b707358e5442eab0fd6ba4d05d9f0a7adc969f)
commit 60d43cb51d7b05cc778cda6b5fcd368db618e3db
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 11 16:17:09 2017 -0500
ceph-osd: mkkey before mkfs
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 75400ead3061ba1e2820878783fe7b1ddcd9a5fb)
commit faad6afa3441b5ce9455a4de0535baef78b8e7de
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 11 15:53:56 2017 -0500
os/bluestore: store meta values in main device label
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 33b8e8e86f6bdbee120729561c6cf433d286e2f8)
commit 1027e57d9af374f1fa1dee2a585dfe539dc4f5eb
Author: Sage Weil <sage@redhat.com>
Date: Sun Sep 10 15:37:10 2017 -0400
os/bluestore: fsck: fix error prefixes
__func__ is now _fsck instead of fsck
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 7d381af7aed14e08823eca4d0b12de14456ef6a1)
commit 68642c43384d5a30155cd67129267d38d6cc3a06
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 7 12:29:32 2017 -0400
os/bluestore: allow repair invocation
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 659fcd279ae82c90172fd6ca07a032d75fe50c96)
commit fd42f2a9ccc5016660b9c205f6aa0104ffdd0b10
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Oct 11 17:45:19 2017 +0800
crushtool: print error message to stderr not dout(1)
in hope to fix the mysterious test failure where cli/crushtool/build.t
prints nothing when error message is expected.
Fixes: http://tracker.ceph.com/issues/21758
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit dc78d40af77ff8d5e4f3d39497e6e3c98c309306)
commit 6902627cb3cb41547be5134aba5c10a39b78163a
Author: Sage Weil <sage@redhat.com>
Date: Sat Jul 22 23:51:47 2017 -0400
crush/CrushWrapper: fix output arg for find_{takes,roots}()
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 60b9cfafc3128cc0cb1f89137221fcc46fcd3802)
commit 993d09850580ca7a697c1a221a8e258f4cd64237
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Oct 17 17:08:59 2017 +0800
os/filestore: print out the error if do_read_entry() fails
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 42e85fe35d2fe8f3e99bd110021fd5157cf589d7)
commit 410434b3d23c54971f3e2c4df179b139979a33c0
Author: Yan, Zheng <zyan@redhat.com>
Date: Thu Oct 19 11:32:38 2017 +0800
mds: fix race in PurgeQueue::wait_for_recovery()
After PurgeQueue::wait_for_recovery() locks the mutex, purge queue
may have already been recovered.
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
commit ee96c34a4d1fc48dec6a1afb4fe98634805c02ce
Author: Yan, Zheng <zyan@redhat.com>
Date: Wed Sep 27 19:59:47 2017 +0800
mds: open purge queue when transitioning out of standby replay
MDS opens the purge queue when it starts standby replay. This is
wrong because purge queue may change during standby replay.
Fixes: http://tracker.ceph.com/issues/19593
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit e515e84f69eeab90ea5c5831f7d3e684e48fb62e)
commit 681cddfa420a0f5f4335564843cdfdb1a4f2371d
Author: Yan, Zheng <zyan@redhat.com>
Date: Wed Sep 27 18:49:55 2017 +0800
mds: always re-probe mds log when standby replay done
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit 82c1f5fcfd151aa1e8c39c7f198abec9bf807b65)
commit 37b13acabe3a0cbdcd88c06f49548146d2c69c14
Merge: 90b6067f6f f30e9a5e6b
Author: vasukulkarni <vasu.kulkarni@gmail.com>
Date: Tue Oct 17 15:47:31 2017 -0700
Merge pull request #18363 from ceph/wip-yuriw-21822-luminous
tests - ceph-ansible vars additions
commit f30e9a5e6bd47129f8a35c1f9067216cfe8a6a70
Author: Yuri Weinstein <yweinste@redhat.com>
Date: Tue Oct 17 13:14:36 2017 -0700
tests - ceph-ansible vars additions
added symlinks for distros
Fixes http://tracker.ceph.com/issues/21822
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit 90b6067f6f5cbfaf48afb08bd17cb9e5f46a5af9
Merge: 1718fd969c ab53002186
Author: Sage Weil <sage@newdream.net>
Date: Tue Oct 17 16:31:25 2017 -0500
Merge pull request #17975 from dillaman/wip-systemd-rbd-mirror-luminous
luminous: systemd: rbd-mirror does not start on reboot
Reviewed-by: Sébastien Han <seb@redhat.com>
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
commit 1718fd969c32cc09cfc208a18455291ab2886e93
Merge: 345286b775 9d12fa5e6d
Author: Sage Weil <sage@newdream.net>
Date: Tue Oct 17 16:30:46 2017 -0500
Merge pull request #18234 from jdurgin/wip-filestore-rocksdb-luminous
filestore: set default readahead and compaction threads for rocksdb
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Amit Kumar <amitkuma@redhat.com>
commit 345286b775fd269eb9f13cfcb8d0aab533dbfbc6
Merge: 9223ace029 ab644b89e0
Author: Sage Weil <sage@newdream.net>
Date: Tue Oct 17 16:29:38 2017 -0500
Merge pull request #18236 from tchaikov/wip-21609-luminous
luminous: mon,osd: fix "pg ls {forced_backfill, backfilling}"
Reviewed-by: Sage Weil <sage@redhat.com>
commit 9223ace02909d6551f6126181d11bf00bab1cfe3
Merge: fdefc8094e 099801e616
Author: Sage Weil <sage@newdream.net>
Date: Tue Oct 17 16:29:08 2017 -0500
Merge pull request #18257 from smithfarm/wip-21699-luminous
luminous: mgr status module uses base 10 units
Reviewed-by: John Spray <john.spray@redhat.com>
commit 8b4fe14ba0e9f8e1b09ae4112dbebd370136bd3f
Author: Kefu Chai <kchai@redhat.com>
Date: Fri Sep 22 10:58:40 2017 +0800
qa/tasks: prolong revive_osd() timeout to 6 min
bluestore_fsck_on_mount and bluestore_fsck_on_mount_deep are enabled by
default. and bluestore is used as the default store backend. it takes
longer to perform the deep fsck with verbose log. so prolong the
revive_osd()'s timeout from 150 sec to 360 sec.
Fixes: http://tracker.ceph.com/issues/21474
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 42be200c564184feff1056b7771ce3c1b2cc5527)
commit a670921f4fb2e15f5ea44e9dda238c2eef96897c
Author: linbing <hawkerous@gmail.com>
Date: Mon Oct 16 09:57:23 2017 +0800
bluestore: using bluestore_compression_min_blob_size instead of bluestore_compression_max_blob_size to config comp_min_blob_size
Signed-off-by: linbing <linbing@t2cloud.net>
(cherry picked from commit da8dc4d4c60f5fd66f6eb2ba7257c52948f24681)
commit fdefc8094ed639c859db5b4eba858cfe05329e5c
Merge: bac56cccf9 3a94d69264
Author: Kefu Chai <tchaikov@gmail.com>
Date: Tue Oct 17 15:19:01 2017 +0800
Merge pull request #18339 from tchaikov/wip-gen-state-diagram-luminous
doc: build with multiple-line state transition function signature
Reviewed-By: Nathan Cutler <ncutler@suse.com>
commit bac56cccf9e3fb44a0ed85c151a8b1a0fb5fa4b0
Merge: 61aca44033 37ddfc61c3
Author: Kefu Chai <tchaikov@gmail.com>
Date: Tue Oct 17 12:44:04 2017 +0800
Merge pull request #17835 from rzarzynski/wip-rgw-per-storage-policy-stats-luminous
luminous: rgw: add support for Swift's per storage policy statistics
Reviewed-by: Matt Benjamin <mbenjami@redhat.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>
commit 3a94d6926403ac1f55a9d5723f6b07486f35060b
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 6 08:57:01 2017 -0400
doc parse two lines for getting the context
So that signatures can get parsed when they are split like:
PG::RecoveryState::RepWaitBackfillReserved::react(
const RemoteReservationCanceled &evt)
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 9e2a52ccec4a5c833dd861ae528d52efbc0f9e5f)
commit ddd88ff365c5399d88890abe1608f12c4a960488
Author: Alfredo Deza <adeza@redhat.com>
Date: Fri Oct 6 08:03:36 2017 -0400
doc raise exceptions with a base class
Although valid in very old Python, it is no longer possible to raise
"bare" exceptions without a class.
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 347b7b1f193f97a3577a971e074a23366b0c8617)
commit 8b3e9917a70452cc4182aae469aac331314df4e6
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Sep 26 15:46:28 2017 -0400
rbd-mirror: ensure forced-failover cannot result in sync state
If the local image already exists and the local image isn't
registered in the remote journal, it should be assumed that
it's already in the replaying state so it can verify the
ancestry to detect split-brains.
Fixes: http://tracker.ceph.com/issues/21559
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit bc96c53d947625d12f7b30a45da68a96c09faf70)
commit d7da395e5196eae13b31c018bf503b9e0c169bda
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Sep 26 14:21:42 2017 -0400
rbd-mirror: forced-promotion should interrupt replay delay to shut down
Fixes: http://tracker.ceph.com/issues/21559
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit e3b58312572ceb7abc675128dfb231c9b888655e)
commit 4ebd4b39282f715a30a849c01510a281faec2ca6
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Sep 29 15:11:38 2017 -0400
librbd: snapshots should be created/removed against data pool
Fixes: http://tracker.ceph.com/issues/21567
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 5a3baf1bd852d6c5d0de10a33839658789edc4eb)
commit 34cd96d3425dcea4053ce0514632194dd7b2eaab
Author: Casey Bodley <cbodley@redhat.com>
Date: Thu Oct 12 11:25:13 2017 -0400
cls/rgw: increment header version to avoid overwriting bilog entries
Fixes: http://tracker.ceph.com/issues/21772
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 081ba106a910583851b9acdcaf5971e37a439329)
commit 972916bf69192a2c504bf855e2af0480815ecc91
Author: Casey Bodley <cbodley@redhat.com>
Date: Wed Oct 11 19:27:55 2017 -0400
test/rgw: add test_multipart_object_sync
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 1c14d4daf9201d58b1d0038bdd733a16275e037a)
commit 695d7ad51179d4f1f5e4168dec3758e3ffa4fe4d
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Sat Oct 14 13:47:53 2017 -0700
mds: prevent trim count from underflowing
Fixes: http://tracker.ceph.com/issues/21807
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 452bdd8b35643ed82d8614efcd4ca22688392eb6)
commit 61aca440334219d5d04df632812dbdf164de3fbc
Merge: 462b93ecc5 bb92cc93ac
Author: Matt Benjamin <mbenjamin@redhat.com>
Date: Sun Oct 15 14:30:57 2017 -0400
Merge pull request #17834 from rzarzynski/wip-rgw-21148-luminous
luminous: rgw: add support for Swift's reversed account listings
commit 462b93ecc5e8caf7b0c35dd83dbbe7e33cc015d9
Merge: c4ac0d64a9 119995a8f0
Author: Matt Benjamin <mbenjamin@redhat.com>
Date: Sun Oct 15 12:48:02 2017 -0400
Merge pull request #17811 from smithfarm/wip-21456-luminous
luminous: rgw: wrong error message is returned when putting container with a name that is too long
commit c4ac0d64a95a6f9ec974446a6971f9d60ab67b9f
Merge: 5695829ffd 957e221f1f
Author: Marcus Watts <mwatts@redhat.com>
Date: Fri Oct 13 20:42:16 2017 -0400
Merge pull request #17832 from rzarzynski/wip-rgw-18977-luminous
luminous: rgw: list_objects() honors end_marker regardless of namespace.
commit b99acfd6018178ad8cd7db187d6892780f82a276
Author: Yan, Zheng <zyan@redhat.com>
Date: Thu Oct 12 12:01:35 2017 +0800
mds: keep CInode::STATE_QUEUEDEXPORTPIN state when exporting inode
Fixes: http://tracker.ceph.com/issues/21768
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit 1d160098a93092904026b62326fd9890de48044a)
commit 2973b6d4184cb5c36f8823010ab83a2b8ebb12d0
Author: Yan, Zheng <zyan@redhat.com>
Date: Tue Oct 10 17:47:33 2017 +0800
mds: update client metadata for already open session
session opened by Server::prepare_force_open_sessions() has no
client metadata.
Fixes: http://tracker.ceph.com/issues/21746
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit 9d5c92a3367369789836d8669aa421074c69e4f3)
commit f353a1e805f4be3e31d8c3346f0f41fa2619c770
Author: Yan, Zheng <zyan@redhat.com>
Date: Mon Oct 9 18:00:38 2017 +0800
osdc/ObjectCacher: limit memory usage of BufferHead
when doing small size random writes, size of data structures that used
for tracking dirty data can be larger than the dirty data size.
Fixes: http://tracker.ceph.com/issues/21726
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit 27db0255992354e15b4285891a33fede6849cf62)
commit 5695829ffd72a03f5611393c6037caf73ff1e813
Merge: d3804d216d 6d032e6ce3
Author: Josh Durgin <jdurgin@redhat.com>
Date: Fri Oct 13 15:29:45 2017 -0700
Merge pull request #18275 from ceph/wip-yuriw-21776-luminous
tests - Removed `distors` and added `supported` to run on all support…
Reviewed-by: Vasu Kulkarni <vakulkar@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 6d032e6ce33fe153958a445066a2a7ba00ffd260
Author: Yuri Weinstein <yweinste@redhat.com>
Date: Thu Oct 12 13:36:20 2017 -0700
tests - Removed `distors` and added `supported` to run on all supported OSs
Added `openstack` fragment to run on vps
Forced/hard-coded `machine_type=vps`
Added `print`s
Added only centos and ubuntu latest to support systemd (in distros dir)
Added `ceph osd set-require-min-compat-client luminous`
Fixes http://tracker.ceph.com/issues/21776
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit d3804d216d21d7e2b12e69eae1c1ef293229a6ca
Merge: c9b493d2db 28e7d5645e
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 13 14:34:56 2017 -0500
Merge pull request #18050 from xiexingguo/wip-pr-17610
luminous: os/bluestore: set bitmap freelist resolution to min_alloc_size
Reviewed-by: Sage Weil <sage@redhat.com>
commit c9b493d2db39765c6ecfb31a3c9e8a38465e6ba0
Merge: 09d597d34a b435af1137
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 13 14:34:30 2017 -0500
Merge pull request #18247 from liewegas/wip-seesaw-warning-luminous
qa/suites/rados/singleton/all/mon-seesaw: whitelist MON_DOWN
commit 2e6b29084ae505098fe6889d69133a8527c773be
Author: Kefu Chai <kchai@redhat.com>
Date: Fri Sep 1 17:32:22 2017 +0800
arch/arm: set ceph_arch_aarch64_crc32 only if the build host supports crc32cx
HWCAP_CRC32 is defined by the linux kernel source. so it's defined as
long as the linux kernel source is new enough. but the compiler on the
building host is not necessarily able to build the `crc32cx`
instruction. if we happen to have an incapable compiler on a machine with
recent linux kernel source, the dummy "ceph_crc32c_aarch64()" will be
selected by `ceph_choose_crc32()`. and it always return 0.
See-also: http://tracker.ceph.com/issues/19705
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 8a077bd2a93d3e12e20013e76e57a35240f5f481)
commit 7d2b7e8e3113daf40fd56d8258294d45b854ca55
Author: Adam C. Emerson <aemerson@redhat.com>
Date: Tue Oct 10 16:21:48 2017 -0400
rgw: Remove assertions in IAM Policy
A couple of them could be triggered by user input.
Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
(cherry picked from commit b3118cabb8060a8cc6a01c4e8264cb18e7b1745a)
commit 09d597d34a1d567dc5350570f044eaa3d74a024f
Merge: 7fdd9a6dac 37ea1976b3
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 12 14:57:38 2017 -0500
Merge pull request #18253 from ceph/wip-yuriw-21660_2-luminous_1
tests - Added yaml fragmet to cover testing...
commit 7fdd9a6dac4e047ae710dfaad38d5dbe3caa2634
Merge: 8ad4617f3d 2b7bd94f13
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 12 14:57:23 2017 -0500
Merge pull request #18254 from ceph/wip-yuriw-21660_3-luminous_1
tests - Added yaml fragmet to cover testing...
commit ab644b89e0aad7b0c1a26ed3b9a2f3ea685c7a15
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Oct 11 11:45:33 2017 +0800
qa: s/backfill/backfilling/
it's renamed "backfilling" in 4015343f .
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit e21114274f6a2742748a5a9b965d415241c80f3c)
commit 099801e616c3c9e8501f153501b74d5220383f33
Author: Yanhu Cao <gmayyyha@gmail.com>
Date: Wed Oct 11 09:31:14 2017 +0800
mgr/status: fix ceph fs status returns error
Fixes: http://tracker.ceph.com/issues/21752
Signed-off-by: Yanhu Cao <gmayyyha@gmail.com>
(cherry picked from commit 4a90daa700142ed1f58f5f2a2cc85f3049bccd64)
commit 0618f19128d0eb04004e65e6eab477b7b1b7fb6a
Author: John Spray <jspray@redhat.com>
Date: Wed Aug 30 20:32:11 2017 +0100
mgr/status: format byte quantities in base 2 multiples
Fixes: http://tracker.ceph.com/issues/21189
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 2a5ae210928f8c9f1e3b72863c277ae5655bb771)
commit 2b7bd94f139251b4373a9662e5add86456566486
Author: Yuri Weinstein <yweinste@redhat.com>
Date: Wed Oct 11 13:12:45 2017 -0700
tests - Added yaml fragmet to cover testing...
... for http://tracker.ceph.com/issues/21660#note-17
for luminous-x (see similar #18200, #18226)
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
(cherry picked from commit e7fdda31803cd37604a9296a876320bfcb849d34)
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit 37ea1976b30e4e03ac535ee353c3a5277c53dbd3
Author: Yuri Weinstein <yweinste@redhat.com>
Date: Tue Oct 10 13:48:33 2017 -0700
tests - Added yaml fragmet to cover testing...
... for http://tracker.ceph.com/issues/21660#note-17
for jewel-x (see similar https://github.com/ceph/ceph/pull/18200)
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
(cherry picked from commit b552e636ddd8ce2cfb605b4c124440d8dd1e8e47)
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit 8ad4617f3dea71984d0d2a6119e9e51ef76779fd
Merge: edf28e6912 59a5a10904
Author: Sage Weil <sage@newdream.net>
Date: Wed Oct 11 12:20:04 2017 -0500
Merge pull request #18025 from liewegas/wip-recovery-preemption-luminous
luminous: osd: allow recovery preemption
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
commit edf28e6912c183de4d955f125ef18d8a52594f59
Merge: 91cb216c01 3354937698
Author: Sage Weil <sage@newdream.net>
Date: Wed Oct 11 12:19:45 2017 -0500
Merge pull request #18154 from ktdreyer/luminous-gitignore-debian-patches
luminous: .gitignore: allow debian .patch files
commit 91cb216c01eca89ff7e338328d64216998effd16
Merge: 27d3f8c748 b3f9439cfd
Author: Sage Weil <sage@newdream.net>
Date: Wed Oct 11 12:19:08 2017 -0500
Merge pull request #18189 from tchaikov/wip-retry-oserror-luminous
luminous: ceph-disk: retry on OSError
commit 27d3f8c748de33a35c5006513d198bbcbee19b45
Merge: a33dd01164 f60a942023
Author: Sage Weil <sage@newdream.net>
Date: Wed Oct 11 12:18:27 2017 -0500
Merge pull request #18227 from liewegas/wip-bluestore-mempool-luminous
luminous: os/bluestore: several mempool accounting fixes
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
commit b435af1137ede418740d2dcf39114d8476775fd8
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 11 08:24:50 2017 -0500
qa/suites/rados/singleton/all/mon-seesaw: whitelist MON_DOWN
Mgr can get marked down when mon weirdness is happening.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 0a886bc9f3e9805de04c8c60bd3d78b937737b96)
commit 51c6f0def379ba013a82204398a021663f1b5d56
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Oct 10 20:32:08 2017 +0800
mon/PGMap: let pg_string_state() return boost::optional<>
better to be specific, so we don't run into the same problem even we
are using the MSB of uint64_t for a pg state bit in future. we could,
however use uint64_t(-1) to indicate the pg_string_state()'s failure to
parse the state string, because pg_string_state() only translate a
single state a time. but it's always better to be explicit than
implicit.
Fixes: http://tracker.ceph.com/issues/21609
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 260f87a9f1c4e32ff3a596ea341ad07a901f44a9)
Conflicts:
src/mon/PGMap.cc
src/osd/osd_types.cc
src/osd/osd_types.h: trivial resolution
commit 9d12fa5e6dad5492491f4718ba7c593077d984a6
Author: Josh Durgin <jdurgin@redhat.com>
Date: Tue Oct 10 19:54:31 2017 -0400
common/options: enable multiple rocksdb compaction threads for filestore
One of the major benefits of rocksdb over leveldb is multithreaded
compaction. The default of 1 thread does not provide much benefit, and
is insufficient for heavy rgw workloads.
For high-write and delete omap workloads I've seen up to 8 compaction
threads be used. There's little overhead to having a higher max than
are needed, so set the default to 8.
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 023fa810aa6b3af305e9027e3f717e54d1bb2712)
Conflicts:
src/common/options.cc (trivial)
commit 6f2f8cdc6a6fa64928fd2d5459b4974e281d857e
Author: Mark Nelson <mnelson@redhat.com>
Date: Thu Sep 21 17:21:16 2017 -0500
common/options.cc: Set Filestore rocksdb compaction readahead option.
fixes: http://tracker.ceph.com/issues/21505
Signed-off-by: Mark Nelson <mnelson@redhat.com>
(cherry picked from commit 66567e573836c570040c8d8148c34b0cf9dc9ce2)
commit f60a942023088cbba53a816e6ef846994921cab3
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 28 08:02:50 2017 -0400
os/bluestore: move several buffer{ptr,list}s into cache_other mempool
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 80c60fcde22cf1269ada45d0914543b8f7d49b3e)
commit 37dd32f449bf2b0aaa3dbcba5cc06da076adbb1e
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 28 08:14:37 2017 -0400
os/bluestore: put new attrs in correct mempool too
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit e7762ea1477e861ebc738aa0588a01c0da883d62)
commit dea2c1e6704c40de11c041e9982e9a9f7d9d4cb3
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 22 08:09:37 2017 -0400
os/bluestore: put attrs in mempool
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ef6524436fe095f94fce421ce26ed4c64e1424fe)
commit c2ed1f9b1801e43cf1d345da4d4876de5334cc40
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 22 08:08:52 2017 -0400
buffer: add ptr::[try_]reassign_to_mempool
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 937379d57a0f8097ae713adf714ba0414d36dffe)
commit a33dd01164bb18b24a0909c9419e15fcd2e79446
Merge: 9e48df69de c0a1168ec2
Author: Josh Durgin <jdurgin@redhat.com>
Date: Tue Oct 10 13:17:46 2017 -0700
Merge pull request #18200 from ceph/wip-yuriw-21660-luminous
tests - Added yaml fragmet to cover testing
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
commit c0a1168ec2bacf6d78041f3ae49bfe6fff669913
Author: Yuri Weinstein <yweinste@redhat.com>
Date: Mon Oct 9 12:51:59 2017 -0700
tests - Added yaml fragmet to cover testing on standanone client.4
for http://tracker.ceph.com/issues/21660#note-17
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit b3f9439cfd9141ddde9874c06fd43d346c10c5da
Author: Kefu Chai <kchai@redhat.com>
Date: Sat Oct 7 22:15:11 2017 +0800
ceph-disk: retry on OSError
we are likely to
1) create partition, for instance, sdc1
2) partprobe sdc
3) udevadm settle
4) check the device by its path: /dev/sdc1
but there is chance that the uevent sent from kernel fails to reach udev
before we call "udevadm", hence "/dev/sdc1" does not exist even after
"udevadm settle" returns. so we retry in case of OSError here.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 4f82dfb9e761a09484e6ba3bd027da535162783e)
commit 83b3377c6b04d78b20b7c23d145e015d63e471a4
Author: Kefu Chai <kchai@redhat.com>
Date: Sat Oct 7 21:13:47 2017 +0800
ceph-disk: factor out the retry logic into a decorator
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 02a8620de7f200736900aafc944b2c1ac47d9910)
commit 9e48df69de14d8415eed008b824204313774e2ba
Merge: a81b0da8e4 41af4ca8f3
Author: Alfredo Deza <alfredo@deza.pe>
Date: Sun Oct 8 12:05:25 2017 -0400
Merge pull request #18167 from tchaikov/wip-luminous-sphinx
luminous: admin: bump sphinx to 1.6
Reviewed-by: Alfredo Deza <adeza@redhat.com>
commit 41af4ca8f3cd3db8dc076c07e1a082a053b91b26
Author: Kefu Chai <kchai@redhat.com>
Date: Mon Sep 11 21:54:33 2017 +0800
doc: switch to upstream sphinx-ditaa
it works with setuptools and is now compatible with py3
the py3 branch is created to track the upstream's master branch
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit fda079d5ed4c8a80c677f714a680233a568d512e)
commit e49cae8b03340834e0b44491a4a903c0d61e1f74
Author: Alfredo Deza <adeza@redhat.com>
Date: Tue Aug 15 09:12:27 2017 -0400
admin: bump sphinx to 1.6
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit 88a1114cb3f90e52ba718dc832617fa004d63b3c)
commit a81b0da8e4aca275cc3195c04f3e0d485e4a2f31
Merge: 1e521b5fb4 e4164c43f5
Author: Alfredo Deza <alfredo@deza.pe>
Date: Fri Oct 6 17:42:30 2017 -0400
Merge pull request #17914 from liewegas/wip-21498-luminous
ceph-disk: fix '--runtime' omission for ceph-osd service
Reviewed-by: Alfredo Deza <adeza@redhat.com>
commit 1e521b5fb47de74720d6aab39750c53aa2b2d8e7
Merge: d550610512 ee5b9ea46c
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 6 15:35:06 2017 -0500
Merge pull request #18135 from liewegas/wip-vta-luminous
luminous: cmake: disable VTA on options.cc
commit 33549376987b18a9bdcf5f56e62268c5791124bc
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Thu Sep 7 11:07:59 2017 -0600
.gitignore: allow debian .patch files
The Ubuntu packaging layout with git-buildpackage assumes a
"debian/patches/" directory with several .patch files in it.
When upstream's .gitignore tells Git to ignore .patch files, we have to
edit that line out downstream. When we forget to do that downstream, it
can lead to missing patches and broken downstream builds.
Allow patches in the /debian/patches directory so it's easier to
maintain an Ubuntu package based on upstream's Git repo.
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
(cherry picked from commit c734b0c0296152721b658af7b699a64b3a49d251)
commit d5506105120479c1831ab559b555e2d4ea4c357c
Merge: d46675a1ec 0dc73f1210
Author: Sage Weil <sage@newdream.net>
Date: Fri Oct 6 07:38:53 2017 -0500
Merge pull request #18140 from liewegas/wip-21660-luminous
luminous: src/messages/MOSDMap: reencode OSDMap for older clients
commit d46675a1ecb10f7176265cb1fc5ca971209aecb1
Merge: 39ad203313 8b2bd38785
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 5 22:12:22 2017 -0500
Merge pull request #18037 from ajarr/wip-21602-luminous
luminous: ceph_volume_client: add get, put, and delete object interfaces
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
commit 39ad2033136e32d76e89e86d3f00bd1b928242ce
Merge: 473f9d1317 80f6508fc2
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 5 21:03:42 2017 -0500
Merge pull request #18103 from liewegas/wip-21259-luminous
luminous: osd: dump bluestore debug on shutdown if debug option is set
commit 473f9d13173aa751d54b9a322060111d627f71f1
Merge: c50df6511f 49294965f9
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 5 21:03:28 2017 -0500
Merge pull request #18112 from liewegas/wip-localpool-min-size-luminous
luminous: mgr/localpool: fix min_size, 3x default, crush rule
Reviewed-by: Alexander Marangone <amarango@redhat.com>
commit c50df6511fda7d03c3cfa828acd5e56f2ef45d0f
Merge: 4ff03e5192 16c6dd5129
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 5 21:02:53 2017 -0500
Merge pull request #18126 from smithfarm/wip-sigpipe-porting-luminous
luminous: msg: reimplement sigpipe blocking
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
commit 4ff03e51928c0d7da9a90d50ec504b30eb9e4b32
Merge: 766c6f3647 7986d4566a
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 5 21:02:09 2017 -0500
Merge pull request #18127 from liewegas/wip-21470-luminous
luminous: os/bluestore: fix another aio stall/deadlock
commit 766c6f3647b57b482f76ef254c95a651741d9ad5
Merge: f972702229 3850f8e0b7
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 5 21:01:53 2017 -0500
Merge pull request #18128 from liewegas/wip-bluefs-allocation-luminous
luminous: os/bluestore: make bluefs behave better near enospc
commit f97270222935d1c43a41102c3ada586a7deca190
Merge: e4c51f62ff f99fe83a37
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 5 21:01:27 2017 -0500
Merge pull request #18131 from liewegas/wip-ifname-luminous
luminous: common/pick_address: add {public,cluster}_network_interface option
commit e4c51f62ffc0f2fcc9214f6b7895210aa79c288b
Merge: 66af04cd9b f0a6f616c6
Author: Sage Weil <sage@newdream.net>
Date: Thu Oct 5 21:01:06 2017 -0500
Merge pull request #18132 from gregsfortytwo/wip-20416-bitwise-assert-luminous
osd: make the PG's SORTBITWISE assert a more generous shutdown
Reviewed-by: Sage Weil <sage@redhat.com>
commit 0dc73f12105ec271a43fe176c2f2a22849e8475c
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 5 15:26:16 2017 -0500
src/messages/MOSDMap: reencode OSDMap for older clients
We explicitly select which missing bits trigger a reencode. We
already had jewel and earlier covered, but kraken includes all of
the previously mentioned bits but not SERVER_LUMINOUS. This
prevents kraken clients from decoding luminous maps.
Fixes: http://tracker.ceph.com/issues/21660
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d2664aa34dc4c13da64d1ad187019737080069d7)
commit 3850f8e0b791f2d05906647866b14a390ab2f5da
Author: Sage Weil <sage@redhat.com>
Date: Tue Oct 3 11:18:17 2017 -0500
os/bluestore: add bluestore_bluefs_min_free
We need at least ~1GB free so we can write out new SSTs (which are 256MB
each).
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 65be614f2bfd65a1f13205a3075c87acc72c4836)
(avoided 1_G syntax that luminous doesn't support)
commit b015ce116159ec57d20bcd29306e450a816610f9
Author: Karol Mroz <kmroz@suse.de>
Date: Mon Oct 2 14:01:47 2017 -0700
rgw: stop/join TokenCache revoke thread only if started.
Thread::join triggers an assert otherwise.
Fixes http://tracker.ceph.com/issues/21666
Signed-off-by: Karol Mroz <kmroz@suse.de>
(cherry picked from commit 26f2da083c7dd21b89c1c1e6c498b14e034364a6)
commit ee5b9ea46c8b1e831c60f9cc0b10819b14d5efb8
Author: Kefu Chai <kchai@redhat.com>
Date: Thu Aug 31 18:15:28 2017 +0800
cmake: disable VTA on options.cc
to silence following warning and to avoid compiling this file twice:
ceph/src/common/options.cc: In function ‘std::vector<Option> get_global_options()’:
ceph/src/common/options.cc:151:21: note: variable tracking
size limit exceeded with -fvar-tracking-assignments, retrying without
std::vector<Option> get_global_options() {
^~~~~~~~~~~~~~~~~~
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 4bb56402a6d6f642d54f329aa1e809cbe044e0c6)
commit 59a5a109040884dd2c7c77f3bc84378b54048f1d
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 4 15:28:26 2017 -0500
osd/PG: separate event for RemoteReservationCanceled
Right now we transparently map a RemoteReservationRejected into a
*Canceled event because this what peers send over the wire. Even
once new peers start sending and explicit CANCEL, old peers will
still do so, so we'll maintain this mapping for a while.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 84d71e6a10b02591b5d5e126b346771871eb1575)
commit 206400997daaec97fe9af7e2e735e1b609006107
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 4 14:55:15 2017 -0500
osd/PG: separate verb 'Reject' from passive 'rejected'
This reduces pg->reject_reservation() callsites from 2 to 1 and
makes the state transitions a bit more explicit.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit bf7f101a787652644c27aae4e752fd21f265e866)
commit 8c1716a42def8547f3af446a143715fad24addb5
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 1 15:05:40 2017 -0500
osd: make note about when we get MBackfillReserve REJECT messages
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6e829a3a7c7aaff78142514fde7703cad48953fd)
commit 260ab6009cdce0d846685e4c7789336db22885f1
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 1 15:04:34 2017 -0500
osd/PG: handle RecoveryReservationRejected in RepWaitRecoveryReserved
This state is analogous to RepWaitBackfillReserved; just like we do there
we want to handle the REJECT from the primary by canceling our local
remote_reservation.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ab8f1d291216c07fed7b661896b0a0c1279f50eb)
commit 479b05a8c1659e80028190c7fe90e762084c4c24
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 1 15:03:22 2017 -0500
osd/PG: ignore RemoteReservationRejected if we are RepNotRecoverying
The primary may send us a REJECT (meaning cancel) if recovery/backfill is
preempted there. That can happen even if the recovery isn't reserved or
requested here (e.g., because the primary is still waiting for the local
reservation). Just ignore it and remain in RepNotRecovering.
Fixes: http://tracker.ceph.com/issues/21613
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 1ce235c5dd0b777223f6465e71fb7ee4befd13d8)
commit 2a95156d7be9e6c796d5f0ce7bc0eab89c42230d
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 1 15:01:05 2017 -0500
osd/PG: cancel local reservation in RemoteReservationRejected handler
We can get a RemoteReservationRejected event either because *we* decide
to reject, or because we get a REJECT from the primary that means "cancel"
(e.g., because recovery/backfill was preempted there). In both cases we
want to cancel our remote_reservation.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit f5809afb0f2ca2f1187609e645d6c4c5bd73e39d)
commit 92ff2cbd69c444c6648889a2da7e01be6b00311e
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 1 14:59:31 2017 -0500
osd/PG: move reject_reservation out of RemoteReservationRejected reaction
The RemoteReservationRejected event is also submitted when we are a
replica or backfill target and get a MBackfillReserve REJECT message
because the primary canceled or was preempted. In that case, we don't
want to send a REJECT back to the primary; we only need to send it in the
cases where *we*, locally, decide to reject. Move the call to those call
sites.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 57d18f0e8d9b0428a193c63112a28ddd046337b4)
commit f0a6f616c6c0358c2a7b9eb24fd6ca9c4290be0a
Author: Greg Farnum <gfarnum@redhat.com>
Date: Fri Sep 29 15:18:26 2017 -0700
osd: make the PG's SORTBITWISE assert a more generous shutdown
We want to stop working if we get activated while sortbitwise is not set
on the cluster, but we might have old maps where it wasn't if the flag
was changed recently. And doing it in the PG code was a bit silly anyway.
Instead check SORTBITWISE in the main OSDMap handling code prior to
prepublishing it. Let it go through if we aren't active at the time.
Fixes: http://tracker.ceph.com/issues/20416
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit 0a691b2b1c19fbc30af5c4046736dacb2fb8bfa4)
commit f99fe83a37241fe1c616ecb854b5196112985861
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 28 17:47:17 2017 -0400
common/pick_address: add {public,cluster}_network_interface option
Add _interfaces option to constrain the choice of IPs in the network
list to those on interfaces matching the provided list of interface names.
The _interfaces options only work in concert with the _network options,
so you must also specify a list of networks if you want to use a specific
interface, e.g., by specifying a broad network like "::" or "0.0.0.0/0".
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 95cc790acddb597d6fef2d9e444f0b6e0436f16f)
commit b84803d6092667631527b8b37000cac6447e0b84
Author: David Zafman <dzafman@redhat.com>
Date: Tue Sep 12 17:17:13 2017 -0700
osd: Only scan for omap corruption once
Before
state 2: Can have complete tables (some may be bad)
state 3: Never had complete tables
After
state 2: Can have complete tables (some may be bad)
state 3 with legacy: Can have complete tables (bad ones are cleared)
state 3: Never had complete tables
Once OSDs boot with this change you can't downgrade to a previous release.
If someone does downgrade they could have unstable OSDs that hit assert(state.v < 3).
The following command run after shutting down the cluster but before downgrading
ceph packages would be a way to fix this.
ceph-osdomap-tool --omap-path ... --command resetv2
Fixes: http://tracker.ceph.com/issues/21328
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 8805ef53424e30fd3f24ee38f5a6bdd9e6dd8641)
commit d0ea152865eeff63cf49e137f5ac342ec6725a6a
Author: David Zafman <dzafman@redhat.com>
Date: Tue Sep 12 18:06:10 2017 -0700
tools: Add --backend option to ceph-osdomap-tool default to rocksdb
Fix hard-coded "leveldb" backend. The command is broken in Luminous
now that "rocksdb" is the default.
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit de43493990923bcdd20f88e9d04647e5ba709feb)
commit 94090379a13639ce0a8619dfe178cc62acdd8fc4
Author: Luo Kexue <luo.kexue@zte.com.cn>
Date: Mon Aug 28 09:24:10 2017 +0800
osd, mds, tools: drop the invalid comment and some unused variables
Signed-off-by: Luo Kexue <luo.kexue@zte.com.cn>
(cherry picked from commit 2e93424167de15e91394169f4395f5f446e710e2)
commit 382ff40fb17db2c657318038495ae7140f95b7e6
Author: David Zafman <dzafman@redhat.com>
Date: Tue Sep 12 23:14:15 2017 -0700
tools: Add the ability to reset state to v2
Available for testing and user downgrade.
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 3e4d68640cc43efc0cf10ea8119b3aa583b7f36b)
commit dc2c1d9d2fdd5449dd7c3213e1d2b5cbd928d4d2
Author: David Zafman <dzafman@redhat.com>
Date: Tue Sep 12 22:12:52 2017 -0700
tools: Show DB state information
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 3214882a95f9f70d9f6d28f4e403ee16324530f9)
commit 36675e34b610ed6afa90bdb93da6e49025e4d9e1
Author: Sage Weil <sage@redhat.com>
Date: Tue Oct 3 10:41:33 2017 -0500
os/bluestore/BlueFS: crash on enospc
We were passing this error back to rocksdb, which would then crash (much)
later with a corrupted SST file! No good.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 569e924812bd7cc61e6e755f265890dbd5c95c31)
commit 16c6dd5129f54a6f5eec7fe16719a179c176548e
Author: Greg Farnum <gfarnum@redhat.com>
Date: Tue Oct 3 15:54:06 2017 -0700
msgr: add a mechanism for Solaris to avoid dying on SIGPIPE
This is fairly clean: we define an RAII object in the Messenger.h on
Solaris, and "declare" it with a macro in the implementations. There's
no code duplication and on Linux it's just entirely compiled out.
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit cba20a95e816aaf9f0971b29b14b0be8c524b59d)
Conflicts:
src/msg/async/PosixStack.cc - luminous #include "common/simple_spin.h" is
missing in master (trivial resolution)
commit 7986d4566a1f6d8e22a9acd19608530ce64cbfa1
Author: Sage Weil <sage@redhat.com>
Date: Tue Oct 3 16:48:37 2017 -0500
os/bluestore: use normal Context for async deferred_try_submit
I'm not quite sure why the FunctionContext did not ever execute on the
finisher thread (perhaps the [&] captured some state on the stack that it
shouldn't have?). In any case, using a traditional Context here appears
to resolve the problem (of the async deferred_try_submit() never executing,
leading to a bluestore stall/deadlock).
Fixes: http://tracker.ceph.com/issues/21470
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 67ec75805787ed63b35f8d70478a7a2cd785df06)
commit 9ab0f64bcf74f74838f6066d554f728902c31715
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 29 13:47:19 2017 -0500
os/bluestore: wake kv thread when blocking on deferred_bytes
We need to wake the kv thread whenever setting deferred_aggressive to
ensure that txns with deferred io that have committed but haven't submitted
their deferred writes get submitted. This aligns us with the other
users of deferred_aggressive (e.g., _osr_drain_all).
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 633b17007701d87903fe1d10b19c44210a3326eb)
commit 0251cfcd17111334d3f13dc1e9d52a9aa9302354
Author: Greg Farnum <gfarnum@redhat.com>
Date: Tue Oct 3 15:06:25 2017 -0700
Revert "SIGPIPE suppression for platforms without SO_NOSIGPIPE or MSG_NOSIGNAL"
This reverts commit 131deb39769c1187c334ee84f552d3be01f1751b. It added
code that shouldn't have entered the project repo.
Conflicts:
src/msg/async/AsyncConnection.cc
src/msg/async/AsyncConnection.h
src/msg/simple/Pipe.cc
src/msg/simple/Pipe.h
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit 21bdb730932196a48407be0effab8813a78c4b1b)
commit 72c858310796848128bbccfa328794d504471e5a
Author: wujian <wujian3659@163.com>
Date: Fri Sep 15 11:23:44 2017 +0800
ceph-mgr: can not change prometheus port for mgr
Signed-off-by: wujian <wujian3659@163.com>
(cherry picked from commit 6cc0338ed15a0402471cfd05acfcf00d07517d1a)
commit 49294965f9f92946382949264344c2aa0b6d70bd
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 4 08:25:38 2017 -0500
mgr/localpool: fix rule selection
The 'osd pool create' arg parsing is broken; the rule name for
'ceph osd pool create $name $numpgs replicated $rulename' is passed
via the erasure_code_profile param. Too many req=false options
without a way to disambiguate them.
Work around it by passing both 'rule' and 'erasure_code_profile'
keys, so that if/when the hack in OSDMonitor.cc is removed it will
still work. Blech.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 7623513935525498640defa2064c291fd69a2b76)
commit cf6f90c5e1e17a1f44f6ed80422b5f1c543362ca
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 2 17:11:46 2017 -0500
mgr/localpool: optionally adjust min_size too
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit af72a8932c60a52aad76f1cc94bfbb31048215cc)
commit d18b7b8652e38c679993241ac232bb7310db7e22
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 2 17:11:38 2017 -0500
mgr/localpool: default to 3x
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 9624923f19dc649a8f182f28e9580a3045964d9e)
commit 80f6508fc25018dd9c70d05f99c40325062d129e
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 21 15:19:47 2017 -0400
osd: make shutdown debug conditional (and off by default)
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6169cf64423c7554dc0630d90d8dee27437bf05f)
commit ca177e93d4e52d50a49b76af516a8009153197e7
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 7 17:12:42 2017 -0400
osd: debug_bluestore on shutdown
...just like we do with filestore etc.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 354dccda322b7ac8844b321f603e049128bf4779)
commit 4d931ad0d349a6974f9f7053792e6f826b67169a
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 7 17:11:30 2017 -0400
os/bluestore: dump stray cache content on shutdown
Tracking down http://tracker.ceph.com/issues/21259.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 28d9b6b0e92cf51996a12a43c81f7ac2abcaecaa)
commit 66af04cd9b61d482362361cb3b12af2d8ffa462c
Merge: 57b5f58f5e b064ed172a
Author: Sage Weil <sage@newdream.net>
Date: Tue Oct 3 13:41:21 2017 -0500
Merge pull request #17734 from liewegas/wip-21089-onmount
luminous: os/bluestore: replace 21089 repair with something online (instead of fsck)
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
commit 57b5f58f5e8be116deed59c5c97636eb730c3dc8
Merge: ae9de1673a a069e8ae61
Author: Sage Weil <sage@newdream.net>
Date: Tue Oct 3 13:40:56 2017 -0500
Merge pull request #17814 from liewegas/wip-pg-limits-luminous
luminous: mon: cluster limit on pgs
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit a069e8ae61a452cefe5673f0035251c00857fee5
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 14 16:01:14 2017 -0400
doc/rados/operations/health-checks: fix TOO_MANY_PGS discussion
Fiddling with pgp_num doesn't help with TOO_MANY_PGS.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 027672b777402381f6736e517ed287b38bb17abb)
commit 7320ee7293a34e4ea0ab5254373f51b93249b91d
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 14 16:00:31 2017 -0400
mon: rename mon_pg_warn_max_per_osd -> mon_max_pg_per_osd
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 986b86fbebf9e06f9f841da8ded0bedb310fc69b)
commit 37ddfc61c3376afeccfe7a8029412b9a38401bfc
Author: Casey Bodley <cbodley@redhat.com>
Date: Fri Sep 22 12:03:57 2017 -0400
rgw: dont reuse stale RGWObjectCtx for get_bucket_info()
if the earlier call to store->get_bucket_entrypoint_info() failed with
ENOENT, the obj_ctx will cache exists=false.
put_bucket_entrypoint_info() doesn't invalidate that, so this call to
get_bucket_info() was reading from a stale cache and failing with ENOENT
Fixes: http://tracker.ceph.com/issues/21506
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 3f4f760a85f162262c5b419e9cf78966c7299f0a)
commit 28e7d5645e5fdf5471a2a809bd232521f2d33814
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 20 12:38:46 2017 -0400
os/bluestore: ignore 0x2000~2000 extent oddity from luminous upgrade
Luminous does a block_size granularity freelist, and assumes that
0~ROUND_UP_TO(SUPER_RESERVED,block_size) is used. Current master uses
min_alloc_size granularity and changes that assumption to
0~ROUND_UP_TO(SUPER_RESERVED,min_alloc_size). That means if master
fsck's a luminous-created bluestore, it will think 0x2000~2000 is used
(current baked-in min_alloc_size-based assumption) but the old freelist
says it is free (old mkfs assumption). The disparity is harmless since
the extent is below min_alloc_size, so ignore it.
Fixes: http://tracker.ceph.com/issues/21408
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 4959ad310a96208565b47c035ab5d5dded1b8ff5)
commit ae9de1673a7b4aa032cbde7c5646c4d945ff618a
Merge: eec0edaaa0 3a037dda42
Author: Sage Weil <sage@newdream.net>
Date: Sun Oct 1 15:56:10 2017 -0500
Merge pull request #17867 from smithfarm/wip-21443-luminous
luminous: mgr: Prometheus crash when update
Reviewed-by: Amit Kumar <amitkuma@redhat.com>
commit eec0edaaa031400e7728081c306a1901cb981486
Merge: 98e729a977 ddca5dbe04
Author: Sage Weil <sage@newdream.net>
Date: Sun Oct 1 15:55:59 2017 -0500
Merge pull request #17868 from smithfarm/wip-21452-luminous
luminous: mgr: prometheus module generates invalid output when counter names contain non-alphanum characters
Reviewed-by: Amit Kumar <amitkuma@redhat.com>
commit 98e729a9773acb5200357b3178cf04883383cf4a
Merge: 87e8829605 faf3fec1a6
Author: Sage Weil <sage@newdream.net>
Date: Sun Oct 1 15:55:30 2017 -0500
Merge pull request #17896 from liewegas/wip-bluestore-tool-luminous
ceph-objectstore-tool and ceph-bluestore-tool: backports from master
Reviewed-by: David Zafman <dzafman@redhat.com>
commit 87e8829605adb2609cd4867f3d23489d84e657fb
Merge: a48b2e41e7 1fc6a51a41
Author: Sage Weil <sage@newdream.net>
Date: Sun Oct 1 15:55:11 2017 -0500
Merge pull request #17930 from jcsp/wip-luminous-health-warn
mon: show legacy health warning in `status` output
Reviewed-by: Sage Weil <sage@redhat.com>
commit a48b2e41e78f9190b7f94d11af2a04c919380aae
Merge: ced74c00d6 4f28ef0016
Author: Sage Weil <sage@newdream.net>
Date: Sun Oct 1 15:54:47 2017 -0500
Merge pull request #18031 from xiexingguo/wip-pr-17987
luminous: osd/PrimaryLogPG: kick off recovery on backoffing a degraded object
Reviewed-by: Sage Weil <sage@redhat.com>
commit ced74c00d6de9fbf84ecb4d3064fbaec01ec827f
Merge: 242635d938 f9d0442ca5
Author: Sage Weil <sage@newdream.net>
Date: Sun Oct 1 15:54:17 2017 -0500
Merge pull request #18033 from xiexingguo/wip-pr-17698
luminous: os/bluestore: allocate entire write in one go
Reviewed-by: Sage Weil <sage@redhat.com>
commit 242635d938fcd46d337b7727f0f3f578122b41ad
Merge: 9d8acf6f3f 86d143722e
Author: Sage Weil <sage@newdream.net>
Date: Sun Oct 1 15:53:56 2017 -0500
Merge pull request #18038 from jecluis/wip-21300-luminous
luminous: mon/MgrMonitor: handle cmd descs to/from disk in the absence of active mgr
commit 1021fe235ca756337e3a3ff4804680c1e93b0270
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Fri Sep 29 08:48:14 2017 -0700
qa: relax cap expected value check
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit b37c7f7db7f0d76e524aabdad6dd0f16548268a6)
commit 633b114aca65548b91a898db7bbbef784b76180b
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 8 18:08:51 2017 -0400
os/bluestore: use min_alloc_size for freelist resolution
For HDD with min_alloc_size=64k, this is a 16x reduction in allocation
metadata!
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6b8e4d512604095fb8a209229d4633ac19b499de)
Conflicts:
Slightly conflict with 834542c4027a8cc4d23eff089827d328919372d2, which
drops the literal description of apply().
commit 3decf1c0de9f54b09afc954b13c2c8fac1011de9
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 8 18:08:07 2017 -0400
os/bluestore: align bluefs_extents to min_alloc_size
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 0c777efdcb2ee5a6322f0eb277e681d0f086e0b6)
commit 529c02db4da9b45cc13bedaa20509a896217fd83
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 8 18:07:38 2017 -0400
os/bluestore/FreelistManager: create: accept min alloc size
Accept a block size other than bdev_block_size. Let's call it, oh, I don't
know, min_alloc_size.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 52453d4ca223c8819f8e35f2c0b691803e74537f)
commit 37929849863c60531bcdbf80124e751504ac710f
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 8 18:06:05 2017 -0400
os/bluestore: mkfs: choose min_alloc_size earlier
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3efde01586776b23bbac1e663ae5baf6500acee4)
commit efcf874295db0a68c43a9fd0a3a20cc3dcdf050a
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 8 18:05:29 2017 -0400
os/bluestore: require that bluefs_alloc_size be multiple of min_alloc_size
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 5b47ac59816894e983a98d3da8b5415d569c6663)
commit 58c52baeaed4a0a6dd5aed076db60988845431d0
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 27 17:42:07 2017 -0400
qa/suites/rados/singleton/all/recovery-preemption: add test
This mirrors what I was testing locally.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d7b29acb1952d0b3dfd19326fd3418cfbd66ef3c)
commit 2e28f3e5ae3a4140e802f915b8db5dc434b6a3c2
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 21 12:37:13 2017 -0400
osd/PG: handle racy preemption
If we finish recovery/backfill and go active, but also get
preempted at the same time, we can ignore the event.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d8c3756d26abbaf326f634a494dcb12fe373f68d)
commit ba106b6d32fb3138b7a61fe613b211e02f1e44aa
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 19 15:26:40 2017 -0500
osd/PG: allow local recovery reservations to be preempted
If a PG has a higher recovery priority and a lower-priority item is in
progress, allow it to be preempted. This triggers the RecoveryCancel
or BackfillCancel event with a 0 delay, which means it will immediately
re-request a reservation (and presumably wait).
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit a8534cce1a1661375a93aff2314633bce63695ed)
commit 33ea286d0cedf44b7b3c70bc9f35c7f88fcc0361
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 19 15:25:05 2017 -0500
common/AsyncReserver: support preemption
If an (optional) preemption context is provided, use that to preempt
and existing reservation and grant a higher-priority one.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit dbc002eaa90e952df1acf295a630443ac3ada418)
commit dad61658d1f48f1089e265d7441bf1349fd42f1f
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 21 12:37:49 2017 -0400
common/AsyncReserver: get a cct
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 08d2c8875bba7c41d1f9eeec586f6880f8be99b2)
commit 1b0875ca6d9ab91f59ffe2356113ed2cc643a276
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 21 10:30:57 2017 -0400
osd: PG_STATE_BACKFILL -> PG_STATE_BACKFILLING
Match user-facing string
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 31a34946f7aa0ad2b9848cbc808d209394a9ad44)
- add update to mon/PGMap.cc
commit 9d8acf6f3f85dd1410629cc680474e38f3be9838
Merge: 62ff5e4c98 b54c7ba225
Author: Sage Weil <sage@newdream.net>
Date: Fri Sep 29 07:11:31 2017 -0500
Merge pull request #17862 from smithfarm/wip-21307-luminous
luminous: mon: Client client.admin marked osd.2 out, after it was down for 1504627577 seconds
Reviewed-by: Sage Weil <sage@redhat.com>
commit 62ff5e4c981a9e0d9a92bf3ac85210d98b0db038
Merge: 453858d4b7 4c9d86bd86
Author: Sage Weil <sage@newdream.net>
Date: Fri Sep 29 07:10:55 2017 -0500
Merge pull request #17865 from smithfarm/wip-21465-luminous
luminous: OSD metadata 'backend_filestore_dev_node' is unknown even for simple deployment
Reviewed-by: Sage Weil <sage@redhat.com>
commit 453858d4b777ee81e5af4f443afc07e6af8b1a60
Merge: 2e85e4f0c4 430a140d38
Author: Sage Weil <sage@newdream.net>
Date: Fri Sep 29 07:10:26 2017 -0500
Merge pull request #17936 from liewegas/wip-ruleset-errors-luminous
mon/OSDMonitor: error out if setting ruleset-* ec profile property
Reviewed-by: Sage Weil <sage@redhat.com>
commit 2e85e4f0c4f196dce8115e0a3aecc506caf1c7d1
Merge: 1c228c95dc 481da57cca
Author: Sage Weil <sage@newdream.net>
Date: Fri Sep 29 07:10:06 2017 -0500
Merge pull request #17946 from liewegas/wip-mgr-localpool-luminous
luminous: mgr/localpool: module to automagically create localized pools
Reviewed-by: John Spray <john.spray@redhat.com>
commit 1c228c95dc142c3ca142f039e606e25cb500e53a
Merge: 9915a2f600 9aafc53ed4
Author: Sage Weil <sage@newdream.net>
Date: Fri Sep 29 07:09:27 2017 -0500
Merge pull request #17998 from liewegas/wip-pr-17978-luminous
rbdmap: fix umount when multiple mounts use the same RBD
Reviewed-by: Sage Weil <sage@redhat.com>
commit 86d143722e1f16a19a56ed7a163d1dc9327c0c90
Author: Joao Eduardo Luis <joao@suse.de>
Date: Wed Sep 20 17:53:40 2017 +0100
mon/MgrMonitor: read cmd descs if empty on update_from_paxos()
If the MgrMonitor's `command_descs` is empty, the monitor will not send
the mgr commands to clients on `get_descriptions`. This, in turn, has
the clients sending the commands to the monitors, which will have no
idea how to handle them.
Therefore, make sure to read the `command_descs` from disk if the vector
is empty.
Fixes: http://tracker.ceph.com/issues/21300
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
(cherry picked from commit 3d06079bae0fbc096d6c3639807d9be3597e841a)
commit 8b2bd387859bdc2287ba6dcaba27d8b234984a5a
Author: Ramana Raja <rraja@redhat.com>
Date: Wed Sep 13 19:53:43 2017 +0530
pybind/ceph_volume_client: add get, put, and delete object interfaces
Wrap low-level rados APIs to allow ceph_volume_client to get, put, and
delete objects. The interfaces would allow OpenStack Manila's
cephfs driver to store config data in a shared storage to implement
highly available Manila deployments. Restrict write(put) and
read(get) object sizes to 'osd_max_size' config setting.
Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit d1bd171d6b6eb00c47168f38cec1a30f9c9f02bd)
commit 9421b40f1748c50b7f92abe8615c0cd9208fdbdc
Author: Ramana Raja <rraja@redhat.com>
Date: Mon Sep 18 20:16:30 2017 +0530
pybind/ceph_volume_client: remove 'compat_version'
... class attribute of the 'CephFSVolumeClient' class. It was supposed
to record the earliest version of CephFSVolumeClient that the current
version is compatible with. It's not useful data to be stored as a
class attribute.
Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit 894a734aa5a56b20d3f68bb3ad644b370f193934)
commit dd23023329b2382afef23ee02d7ef9b14baaf07b
Author: Ramana Raja <rraja@redhat.com>
Date: Mon Sep 18 20:09:55 2017 +0530
pybind/ceph_volume_client: set the version
... of on-disk structures to be same as the class attribute 'version'
of the CephFSVolumeClient class.
Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit 8267c2ba09841095f20a8833c155185529e64f46)
commit 9ee3f7ad21049d0692804d4405b4bf0f8c795ae3
Author: Joao Eduardo Luis <joao@suse.de>
Date: Wed Sep 27 17:55:17 2017 +0100
mon/MgrMonitor: populate on-disk cmd descs if empty on upgrade
During kraken, when we first introduced the mgrs, we wouldn't populate
the on-disk command descriptions on create_initial(). Therefore, if we
are upgrading from a cluster that never had a mgr, we may end up
crashing because we have no cmd descs to load from disk.
Fixes: http://tracker.ceph.com/issues/21300
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
commit f9d0442ca50078224b72f9876d2fed41fd409be6
Author: Sage Weil <sage@redhat.com>
Date: Thu Aug 10 16:44:59 2017 -0400
os/bluestore: allocate entire write in one go
On the first pass through the writes, compress data and calculate a final
amount of space we need to allocate. On the second pass, assign the
extents to blobs and queue the writes.
This allows us to do a single allocation for all blobs, which will lead
to less fragmentation and a much better write pattern.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit e200f358499af8e3acb6ac4f675cc167433b53ec)
commit 4f28ef00162692557654d84bc1c0988998a6f2c4
Author: xie xingguo <xie.xingguo@zte.com.cn>
Date: Wed Sep 27 16:05:56 2017 +0800
osd/PrimaryLogPG: kick off recovery on backoffing a degraded object
As we are now blocking frontend ops from accessing that very object!
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit db20328b456d70d6728fd27f17da6f2f3546e84b)
commit 711892dceaa7e4598bffff0edcff45554bc06ffc
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Thu Sep 28 16:01:00 2017 -0700
mds: improve cap min/max ratio descriptions
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit dcf97d17356f59c8870a36a99a77f879a20df348)
commit bef8ded7b82fc6896398bd334bdbd4fc63ebfe51
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Thu Sep 28 15:56:26 2017 -0700
mds: fix whitespace
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit ab69d682937a0a13246d28cfa2bbbf8c1b7ced38)
commit a5c96d4f7a5081242c6abf6c722adfa131030a55
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Wed Sep 27 10:39:01 2017 -0700
mds: cap client recall to min caps per client
Fixes: http://tracker.ceph.com/issues/21575
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 538834171fe4524b4bb7cffdcb08c5b13fe7689f)
commit 913c72d96a6c405b4e4fe57abcfbf58d53a66e63
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Wed Sep 27 09:29:39 2017 -0700
mds: fix conf types
This correct an assertion failure.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit bfc490249566356cff785f2f89dc03d69af322da)
commit 91084538cdfe80230bf346ee9cb6e8bfe080d6d8
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Wed Sep 27 09:29:23 2017 -0700
mds: fix whitespace
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 8a5d71bf4517e6a3c194c3e63e0283747b75d4b1)
commit 5b1306142feb820a6c4f6e180ba68e428985b3f1
Author: ren.huanwen <ren.huanwen@zte.com.cn>
Date: Mon Sep 18 15:51:29 2017 +0800
doc/cephfs: add client min cache and max cache ratio describe
Signed-off-by: ren.huanwen <ren.huanwen@zte.com.cn>
(cherry picked from commit d879ff91e76f66f90364038920fc2c62dd18c449)
commit a39ab633cb97fd4fa0e1ef3e0e2e56709f980857
Author: ren.huanwen <ren.huanwen@zte.com.cn>
Date: Mon Sep 18 15:14:19 2017 +0800
mds: adding tunable features for caps_per_client
Sometimes we need to raise or lower the value of "max_caps_per_client" and
"min_caps_per_client" to improve recall_client_state efficiency
Signed-off-by: ren.huanwen <ren.huanwen@zte.com.cn>
(cherry picked from commit a07b376628505832f5528bd7f2b0faeabe9a5f5d)
commit 4ff28ab144b3a0adfea8ad8ddee750243c824d14
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 21 10:28:59 2017 -0400
osd/osd_types: make BACKFILL <-> "backfilling" for parser
We render BACKFILL as "backfilling"; make sure parse works that
way too.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6fa40e44dc579a009edff7be95d3fd37defbc7f6)
commit 2a9cfa2ad256df11bc1c3d95720ef89d11813846
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 21 10:28:08 2017 -0400
osd/osd_types: remove weird BACKFILL state hack
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 5bcfaf4bd40fee9ea4bfd5ed086b51facc5fe78e)
commit 429804b6bd193611f7474174e45ce94d784fd184
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 19 15:47:23 2017 -0500
osd/PG: Cancel{Recovery,Backfill} -> Defer{Recovery,Backfill}
"Defer" is more accurate here; we aren't canceling anything, just
rescheduling the work.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2e45497a20a0b61975fbf0cf851d417f31b35489)
commit 4cbd62ecdc42aa91045f7f5c7ce5dab0cb59c16e
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 19 14:53:31 2017 -0500
osd/PG: specify delay in Cancel{Recovery,Backfill}
For now it is always the retry interval, but later perhaps not!
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 597dfd11728b55ad181316ac10be7155e37a6ba9)
commit 86778de01166902612b3173f1d0687f13027a21f
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 19 14:49:05 2017 -0500
osd/PG: make some trivial events TrivialEvent
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3eadfa087f3ff5338f87306fd384d1f4749b8005)
commit 6c529092e8f4bd1e2a4a4069bf9f319d49430ccb
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 19 14:44:50 2017 -0500
osd/PG: set {backfill,recovery}_wait when canceling backfill/recovery
The only caller currently is when we get as far as we can with backfill
or recovery but still have unfound objects. In this case, go back into
the *_wait state instead of appearing as though we are still doing
something.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 4b216771bd31e5712471e13ad32ee8a1e519eb30)
commit 06c31a6ceecf1c33e1456c746169961802903001
Author: Yan, Zheng <zyan@redhat.com>
Date: Wed Sep 27 22:19:06 2017 +0800
mds: make sure snap inode's last matches its parent dentry's last
This patch is for luminous only. The issue has been fixed in master
branch in another way (change is much bigger)
Fixes: http://tracker.ceph.com/issues/21337
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
commit 1236e0f2e9e8724569cf5fc7d7d525c7fca6ad77
Author: Yao Zongyou <yaozongyou@vip.qq.com>
Date: Sun Sep 24 22:11:22 2017 +0800
rgw_file: fix write error when the write offset overlaps.
Fixes: http://tracker.ceph.com/issues/21455
Signed-off-by: Yao Zongyou <yaozongyou@vip.qq.com>
(cherry picked from commit 872d73f945364002f0fa31762e6976db5b4b3c19)
commit 9aafc53ed4edca8bebb96bc36a0b37ce32c79a75
Author: Alexandre Marangone <a.marangone@gmail.com>
Date: Tue Sep 26 11:35:04 2017 -0700
rbdmap: fix umount when multiple mounts use the same RBD
When a Kubernetes Pod consumes a RBD it is mounted two
times on the same host. When the host shutdown umount will
fail leading to a hung system
Signed-off-by: Alexandre Marangone <amarango@redhat.com>
(cherry picked from commit 40825daecedb2a3481021e4d36a367c339eb9b62)
commit 9915a2f600788c37fb42f39a766ce93b9ae4d816
Merge: 3e7492b9ad 6b5f212b91
Author: vasukulkarni <vasu.kulkarni@gmail.com>
Date: Wed Sep 27 09:50:44 2017 -0700
Merge pull request #17953 from ceph/wip-ceph-ansible-bport
tests - ceph-ansible backports
commit 6b5f212b9170dbd99779527b41be233b9d40b5e4
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Thu Sep 21 15:34:31 2017 -0700
use ovh instead of vps
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 56d1511bdfaee5a2a0d85bd606348632bfcbf09e)
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
commit ab5300218699c947f380b0437f931d559bdbf074
Author: Sébastien Han <seb@redhat.com>
Date: Tue Sep 26 14:05:37 2017 +0200
rbd-mirorr: does not start on reboot
The current systemd unit file misses 'PartOf=ceph-rbd-mirror.target',
which results in the unit not starting after reboot.
If you have ceph-rbd-mirror@rbd-mirror.ceph-rbd-mirror0, it won't start
after reboot even if enabled.
Adding 'PartOf=ceph-rbd-mirror.target' will enable
ceph-rbd-mirror.target when ceph-rbd-mirror@rbd-mirror.ceph-rbd-mirror0
gets enabled.
Signed-off-by: Sébastien Han <seb@redhat.com>
(cherry picked from commit e6cd9570ba210c5f4501b6f4fa443245a22c4d6c)
commit 481da57cca2f43f3cb14beaaa22eff3b2f62c875
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Sep 26 12:10:28 2017 +0800
doc/mgr: add "local pool" plugin to toc
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 0b831e7886daad5836ca3cb58c01c123f99a1cde)
commit 315e336f38d1bb0e821dbc1943447cc61f203e2c
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Tue Sep 19 17:09:08 2017 -0700
qa/tests: skip test that use public bucket landsat-pds from AWS
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit e11a1f773d8795dd9bd9819aaf229c6598e22ca6)
commit fc7db58fc5ed071436d5f0f6cc9dcd61e3276aea
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Tue Sep 19 17:06:03 2017 -0700
qa/tests: update required ceph-ansbile vars
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 184d5e6bad46451b3a8c5787f56148566bb42f20)
commit 4342d80354c941d47598897ca7ec6a6b4928edf6
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Tue Sep 19 17:03:46 2017 -0700
qa/tests: Fix ceph-ansible upstream vars
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 077904a0b5dffe9d64feade94cf30ffc92f1c0e5)
commit a3d43ad8f3e685a028e943b5aee9ec4720cba60b
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Sun Sep 10 12:42:16 2017 -0700
Fix get_system_type failure due to invalid remote name
recent changes caused the remote name to be invalid, fix the
arg passed to get_system_type
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 9fe58d5ed6bd2338271a9e003f51d80c2e440e23)
commit ea8ddc95a8336f37065edfd7f061601c5c4cac4b
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Fri Sep 8 10:00:49 2017 -0700
Stop the mgr on node which is not client, this will ensure
the client.0's mgr is active mgr for workunit to work.
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit b1fd615e9f507d4243744b34887366086c1890ac)
commit 69bbaf5173c09afbf02b1894843570d403ad4235
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Thu Sep 7 17:10:04 2017 -0700
Test ceph-mgr RESTful api
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit e5b5a1b8c8d0e750d29fe41a16252dcb0adf449f)
commit 581656e3874822f958400c235241cddb07c4aca3
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Thu Sep 7 16:54:51 2017 -0700
Add additional variables as required by ceph-ansible for upstream
installation.
possible revert if those are made default
issue: https://github.com/ceph/ceph-ansible/issues/1834
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit e963bfd6ca0ea1525e37dc35bf3309548c7a3831)
commit 559a47997c7f6c1e8e86e2fd06759ded0e7cccd9
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Fri Aug 18 11:35:54 2017 -0700
Rename folders to fix task order
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 9cc00c5c1a6e99323a4655aad5c093b8d87609f6)
commit 655b0733970c6798af087fdb886bf914711fef1b
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Fri Aug 18 11:09:50 2017 -0700
use bluestore with dmcrypt option
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 1041c803f1f8f7be0edd0e78df352152fa3c4e6f)
commit 244689b36a2beb6dbd9beed429e5b46f4b714bc3
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Fri Aug 18 11:08:00 2017 -0700
Add dmcrypt option
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit f6de5d9f9e2a3666116311c146a8f2a45f3b8dbb)
commit d2ddcc0324a66b1cb67a640df987e5bc1a6f4887
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Fri Aug 18 11:05:01 2017 -0700
Separate the main task from options
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 60d00e0eadda3a7a7cfbac2acdc81a4d55a27570)
commit 05f7ce97a8b94b5abcaddd7d125398c7d0118bbc
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Fri Aug 18 10:47:22 2017 -0700
Catchup with recent changes with ceph-ansible
Adds osd_scenario and ceph_stable_release variables
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 0395b844884a63a2cb7480dcc40762df6915b403)
commit 58ffed6fb9e8695deb53edd7754aed56c73403bd
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Thu Aug 24 11:02:20 2017 -0700
Add workaround for http://tracker.ceph.com/issues/20950
mgr bootstrap key differs on disk, rewrite the new key
using auth get.
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 5c43c5972ddb04724c8659a1b8d71cda1e0acdfb)
commit c2284be89df724bbf1446b28cdf712722d7581a1
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Thu Aug 24 09:24:52 2017 -0700
Add kraken to luminous upgrade case
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 71ea4987e0d55cc49e90dfbaf9722769cc892f3f)
commit 6cc150023f1dc2450bbf83ebdba8360c5693645c
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Mon Aug 7 12:19:41 2017 -0700
Add kernel tar/untar and systemd task
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 526d4aef1cb73ffa6f73e1353dc06eed6cccb9f9)
commit 39eb7cb0bd77925d422c68aaed86fbd39bdfc80e
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Tue Aug 1 16:32:42 2017 -0700
Restart services after upgrade
This seems to be an issue and should be fixed in our systemd scripts,
need to discuss more on this.
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 2116f76a1f1fbb32ca5844c1d051e22b1b657e0f)
commit 566dada1f6a73b8be128167a425becea7ba110fe
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Fri Jul 28 23:11:41 2017 -0700
since the roles are mapped inside ceph-deploy, store the roles that
are mapped and use the new mapped role for upgrades during later
stage.
eg: mon.a is mapped to mon.mira002 during install, store this mapping
and durig upgrade map it back to appropriate name to find the hostname
with that role
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 8a2b9a08095dc97e0e440cc900c650d5efb55a4f)
commit 093a13268dff5b1114294456e83e0e24bfd7be6d
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Fri Jul 28 17:48:09 2017 -0700
Add doc string for missing ceph-deploy-branch
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit e7a0a4f87f2fbea0c9e27341c952245f1ad34ab8)
commit df415e81887b06918731d7d3b7677ecab1ee2db5
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Thu Jul 27 15:24:25 2017 -0700
use mon.a as ceph-admin node, the previous get_first_mon returns
a sorted(mons)[0] as first mon.
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 407ce2613d37252e48c2a648b94b065ab070ec46)
commit 1128ae099cc350f29b243321bc2815aaf7bbfe80
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Thu Jul 27 11:17:31 2017 -0700
use elif instead of else to handle cases for jewel install
jewel needs neither filestore or bluestore as an option, so provide none
when running with jewel branch.
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit f3ec7cabf66c76d712f77a4dfefe96f374507982)
commit 7db1550a55002d28d899aee6ec695604045edabb
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Wed Jul 26 19:18:11 2017 -0700
Jewel to luminous upgrade scenario using ceph-deploy
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 9c85c76a90dcf46ec2bcc6840f4fc57ed86583b5)
commit ab219d2e985cd1c6a7fd17b5ced4db25cb0b7427
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Wed Jul 26 18:56:08 2017 -0700
Add upgrade functionality using ceph-deploy
This is to test for customer like upgrade scenarios and to find
any issues that may be related to systemd, packaging etc
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit 9a73127c16d3395934d003024964ed1373d81083)
commit a145127bde2e2fb96a1fd2b821ed2b647e8ae223
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Wed Jul 26 15:01:15 2017 -0700
Fix old flake8 issues
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit b409c3d163194b0b65e26b0fbc93def8e17a10b4)
commit 28b3c242bdbfc655534cdd037966ed1dcddc978a
Author: Vasu Kulkarni <vasu@redhat.com>
Date: Wed Jul 26 14:58:50 2017 -0700
Add option to skip mgr install for old release
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
(cherry picked from commit a9be0c387ed2ae09227dad833c8647af110bd2de)
commit 82651da2d504913f2145085df8a89484308f3eba
Author: huangjun <huangjun@xsky.com>
Date: Wed Sep 20 00:04:04 2017 +0800
qa/ceph_manager: check pg state again before timedout
Pg state maybe all in active+clean when no recovering going on,
so check it again before timedout.
Fixes: http://tracker.ceph.com/issues/21294
Signed-off-by: huangjun <huangjun@xsky.com>
(cherry picked from commit fa40add7f0123dfeac30986f3d53cdfa77736a87)
commit 4017a26162706f1b6bed3f7b37174c8ca47b54ad
Author: Sage Weil <sage@redhat.com>
Date: Sun Sep 24 14:05:14 2017 -0400
qa/workunits/mgr/test_localpool: simple test for localpool mode
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6383fa5b3004a606c15efd2be1b7ce5d57f05086)
commit 45a804f25bc42f03b0596b1ba394fb6be575d392
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 6 15:34:50 2017 -0400
pybind/mgr/localpool: module to automagically create localized pools
By default, this will create a pool per rack, 3x replication, with a host
failure domain. Those parameters can be customized via mgr config-key
options.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 1a0f42b70a4c9fa68dc47f2f521d0f1e8f5bb220)
commit 430a140d38ba6b9e07c53d5a4a5c6232bef4fb0e
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 20 16:42:01 2017 -0400
mon/OSDMonitor: error out if setting ruleset-* ec profile property
We change ruleset -> crush back in dc7a2aaf7a34b1e6af0c7b79dc44a69974c1da23.
If someone tries to use the old property, error out early, instead of
silently not doing the thing they thought they told us to do.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6376d75eda648789b5e316a1ac2883708db7128e)
commit 1fc6a51a416761d9ccd172937cda1a6b1c11faad
Author: John Spray <john.spray@redhat.com>
Date: Sat Sep 23 13:48:36 2017 +0100
mon: show legacy health warning in `status` output
Previously you only got the text of this if you were
either looking at "health detail" or if you had
already set the preluminous_compat setting (in which
case you presumably were already aware so the message
isn't doing much).
Signed-off-by: John Spray <john.spray@redhat.com>
commit d5e583490d946242bdfd26e863dd0aac27c12e3c
Author: Ramana Raja <rraja@redhat.com>
Date: Wed Sep 20 20:27:22 2017 +0530
ceph_volume_client: perform snapshot operations in
... client configured snap directory name, instead of in hard-coded
'.snap' directory.
Fixes: http://tracker.ceph.com/issues/21476
Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit f4fc1722594ed007706b54901fb07a2a443d1b96)
commit e4164c43f5ec808c0420036d2a647a5a7d38a436
Author: Carl Xiong <xiongc05@gmail.com>
Date: Fri Sep 22 11:55:33 2017 +0800
ceph-disk: fix '--runtime' omission for ceph-osd service
f425a127b introduces a regression that ceph-disk omits "--runtime" when
enabling ceph-osd@$ID.service units for device-backed OSDs.
Fixes: http://tracker.ceph.com/issues/21498
Signed-off-by: Carl Xiong <cxiong@suse.com>
(cherry picked from commit a385b5b0c35106c9b44d81655983b2f7566b21cd)
commit faf3fec1a628c9c4f4791e33d7795ef719267cbc
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 13 18:33:21 2017 -0400
ceph-bluestore-tool: better default logging; --log-file and --log-level options
- keep derr going to stderr (so we see 'fsck error' messages)
- hide the rest
- add friendly --log-file option
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 9bd98b42fea1791935280f5ad19e5420d5303e03)
commit 2184e3077caa9de5f21cc901d26f6ecfb76de9e1
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 7 18:27:20 2017 -0400
ceph-bluestore-tool: add 'bluefs-bdev-expand' to expand wal or db usage
If you are using the wal or db devices, this will expand bluefs's usage to
include the entire block device.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit aa4baad529835c1999ff9cc1a2f509c52a0cc699)
commit af9a43a47b1dd0d8bb9946d4b4a0ab9b5c39e427
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 7 18:20:27 2017 -0400
ceph-bluestore-tool: add 'bluefs-bdev-sizes' command
Show bdev sizes vs owned extents.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 9e492f2a2217ee94670e9e0d6cf28e4b7bff5ad0)
commit 6a6678c52ac4aba7fd313ddc2cb3c013819fe28f
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 7 18:12:21 2017 -0400
ceph-bluestore-tool: factor out bluefs mount
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 86db2d7b09de4362d90fba834d61978860d73309)
commit 53f9941f28ac1435bf14f48d369b8ea03336c9d2
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 7 12:27:35 2017 -0400
os/ObjectStore: add repair interface
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c7b7a1f04f78fa62890c567d0ca53874c8d75eb7)
commit edd4d98fefe63ad31bf96fa61716b4ee079c0498
Author: David Zafman <dzafman@redhat.com>
Date: Fri Sep 8 17:53:07 2017 -0700
ceph-objectstore-tool: Make pg removal require --force
Add new export-remove to combine the 2 operations
Fixes: http://tracker.ceph.com/issues/21272
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 3bb20f6d750915d176c7a34367dac966a20faa76)
commit bbe0f34cc81a8db873ef23cca0cdea8651411b86
Author: David Zafman <dzafman@redhat.com>
Date: Fri Sep 8 17:09:48 2017 -0700
ceph-objectstore-tool: Better messages for bad --journal-path
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 49ca1fff7fc4360d2f3a9cac60c0ba651cbc4750)
commit fd7d53bacb10d5b2094f1f579c96899d8a0a4b4f
Author: David Zafman <dzafman@redhat.com>
Date: Wed Sep 6 20:41:50 2017 -0700
test: Fix ceph-objectstore-tool test for standalone and latest code
vstart.sh now defaults to bluestore, so specify filestore
Set environment for run-standalone.sh and cmake build
Create td/cot_dir as test directory
Crush output format change
Change dir into test directory
Give a little time after pool creation
Check for core files as ceph-helpers.sh does
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 3ac219df2d635a08be52df3ad44cf0683de46af2)
commit 3e20e0f3071188f52b7a7c3f9e36b09305431ed0
Author: David Zafman <dzafman@redhat.com>
Date: Wed Sep 6 20:40:37 2017 -0700
test: Move ceph-objectstore-tool test to standalone
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 495c32fd31b311d22f0d4509e896916bd2671615)
commit ddca5dbe04a2647218fabff6098691edd6b7a157
Author: John Spray <john.spray@redhat.com>
Date: Mon Aug 28 19:55:06 2017 +0100
mgr/prometheus: only turn - into _minus at end
...of perf counter name.
So that mds_mem_dir- becomes mds_mem_dir_minus, but
throttle-filestore_bytes becomes throttle_filestore_bytes.
At some point once this is all settled we should
probably just change Ceph's internal perf counter
naming to satisfy the major TSDB naming rules.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 29ac9270d43a225bb2336adaaf813c12e12a715a)
commit 8375251d150397c0097718f6e93ed38e9e89cfbb
Author: jermudgeon <jhaustin@gmail.com>
Date: Sun Aug 27 21:26:28 2017 -0800
mgr/prometheus: Fix for MDS metrics
MDS metrics come in these forms:
mds_mem_dir #Directories
mds_mem_dir+ #Directories opened
mds_mem_dir- #Directories closed
In this case, continuing the trend of replacing all illegal characters with '_' results in…
mds_mem_dir #Directories
mds_mem_dir_ #Directories opened
mds_mem_dir_ #Directories closed
which is palpably a bad idea.
Suggested replacement for '+' = '_plus' seems fine, and a perusal of all metrics indicate that only MDS metrics end in '-' or '+' at this time.
Replacing '-' with '_minus' is probably less good for the general case, if anyone has a better idea…
I suppose another alternative would be to change MDS metrics so they don't use 'illegal' characters, but this also seems cumbersome and would break more third parties.
Fixes: http://tracker.ceph.com/issues/20899
Signed-off-by: Jeremy H Austin <jhaustin@gmail.com>
(cherry picked from commit d719cd04b294e90ab9d440ba7d033826c069a2de)
commit 3a037dda4273cea289641e93a006b2857147ee73
Author: John Spray <john.spray@redhat.com>
Date: Fri Sep 8 11:33:02 2017 -0400
mgr: fix py calls for dne service perf counters
Fixes: http://tracker.ceph.com/issues/21253
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit f9a4ca07acecd15986cbce61a6e118a6cb05af29)
commit 077b3601d6bf68167d1857b79bb3782e23bc8ccc
Author: John Spray <john.spray@redhat.com>
Date: Thu Jul 27 11:48:42 2017 -0400
mgr/restful: cleaner message when not configured
RuntimeError+backtrace prints should be for
genuinely exceptional exceptions. For these
known cases, we should just print the message cleanly.
http://tracker.ceph.com/issues/21292
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit f7a1d57dd0d6c29b8c57a5de624da33777403fe5)
commit 4c9d86bd86954cbee25a3ea2a60af063114f7790
Author: Sage Weil <sage@redhat.com>
Date: Tue Aug 8 13:59:41 2017 -0400
os/filestore: fix device/partition metadata detection
The UUID thing (a) relies on partition labels to work, which isn't
always true (and won't be true for ceph-volume going forward), and
(b) reportedly doesn't work anyway. The fd-based helper works
just fine (even for vstart).
Fixes: http://tracker.ceph.com/issues/20944
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit e62862be0b68a17763e19ed8453436c457294505)
commit 49e030da55030068c35790f751ea6631a324e5c2
Author: Yan Jun <yan.jun8@zte.com.cn>
Date: Wed Sep 13 18:15:33 2017 +0800
test/config: regulate parameter order of ASSERT_EQ
Signed-off-by: Yan Jun <yan.jun8@zte.com.cn>
(cherry picked from commit ccaba817dcd2c84c75cf0f784d90b78527c9f92b)
commit 85ab5c86aafa92a36078133d7da07db54f064f4b
Author: Yan Jun <yan.jun8@zte.com.cn>
Date: Tue Sep 12 11:20:34 2017 +0800
common: fix daemon abnormal exit at parsing invalid arguments
Signed-off-by: Yan Jun <yan.jun8@zte.com.cn>
(cherry picked from commit 3f075372d1400806ae32e0e9210b9c15316dfe66)
commit d7cde5f316242caa86ef62690104228aafc58a42
Author: Kefu Chai <kchai@redhat.com>
Date: Thu Sep 7 13:12:15 2017 +0800
mon,monmap: use new style config opts
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 3124eb2148f5c6703c75b9e556ddb773898a6e55)
commit fe039e41d78f7d74c39cdc9034cadccf632e9ac2
Author: Kefu Chai <kchai@redhat.com>
Date: Thu Sep 7 12:47:36 2017 +0800
mon,option: set default value for mon_dns_srv_name
Fixes: http://tracker.ceph.com/issues/21204
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 50207769f0c187cc8caf2bed3a1c94df409718e5)
commit b54c7ba225314cc3e67707ec8dbac4a7f8381929
Author: John Spray <john.spray@redhat.com>
Date: Wed Sep 6 12:57:40 2017 -0400
mon: fix `osd out` clog message
This was printing the absolute time instead of the period.
Fixes: http://tracker.ceph.com/issues/21249
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit f07480fec729d66f4cdc19a1c3bde3d185df1289)
commit d7f6b93af4ce1115a1bde5883421c8ad860aa53f
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Sep 8 16:43:58 2017 -0400
rbd: mirror "get" actions now have cleaner error messages
Fixes: http://tracker.ceph.com/issues/21319
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 053be9847ff0052348aa259520d641923e57537d)
commit bb0874c9db50bbfbf84396a4f1308282102a1694
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Sep 8 15:41:36 2017 -0400
cls/rbd: avoid recursively listing the watchers on rbd_mirroring object
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 79be496ac098859304efcabc7b5eb8afca6eca91)
commit 34af07f97a67ead7d00b4352aea6300ac737ad9b
Author: Mykola Golub <mgolub@mirantis.com>
Date: Fri Aug 18 20:08:12 2017 +0200
rbd-mirror: potential lockdep issue
(cycle between ImageReplayerAdminSocketHook and ImageReplayer locks)
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
(cherry picked from commit 226b1a3be16915c79e16106d42260687683e7a92)
commit 925f223b7a57eae382c168d4bc4f05f50dd7bf6b
Author: Mykola Golub <mgolub@mirantis.com>
Date: Fri Aug 11 14:27:41 2017 +0200
rbd-mirror: update asok hook name on image rename
Fixes: http://tracker.ceph.com/issues/20860
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
(cherry picked from commit 9ddb5da3377029d11cb1089102d7146e9ee2537a)
commit fd207447d1acebbf7a62c0d03b3ec9025ac19fbd
Author: lu.shasha <lu.shasha@eisoo.com>
Date: Thu Aug 17 16:02:02 2017 +0800
rgw: fix lc process only schdule the first item of lc objects
If more than two bucket hash to the same lc object, only the first bucket will do lifecycle operation. Other buckets lifecycle will never be schduled.
Fixes: http://tracker.ceph.com/issues/21022
Signed-off-by: Shasha Lu <lu.shasha@eisoo.com>
(cherry picked from commit 9561dc0a903d9021f64857762e73887ffe46c28e)
commit 5b5f0b5a33c638ec6f52d6936c60e5efd76eacab
Author: Radoslaw Zarzynski <rzarzyns@redhat.com>
Date: Wed Aug 23 21:31:11 2017 +0200
rgw: fix accessing expired memory in PrefixableSignatureHelper.
Fixes: http://tracker.ceph.com/issues/21085
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit 9260d39ceb1ef39a32c8f1742c3069bb83d628f8)
commit d2be1e0ccb64021f770f28a6097f5468d375dbec
Author: lu.shasha <lu.shasha@eisoo.com>
Date: Fri Aug 25 15:05:52 2017 +0800
rgw: return bucket's location no matter which zonegroup it located in.
Get bucket location which is created in another zonegroup, will return "301 Moved Permanently".
Fixes: http://tracker.ceph.com/issues/21125
Signed-off-by: Shasha Lu <lu.shasha@eisoo.com>
(cherry picked from commit 71d2eca3e00de19419fb07bf9346f5863b20d40f)
commit c4d6539374a02378209688dfa829caa39af0a214
Author: Abhishek Lekshmanan <abhishek@suse.com>
Date: Thu Aug 31 13:39:13 2017 +0200
rgw: rgw_rados: set_attrs now sets the same time for BI & object
`RGWRados::set_attrs()` used to set slightly different mtimes for the
object and in the bucket dir entry as we do an object write and set the
time at bucket index as the time at that point. Fix this by using the
setting the same mtime for the object as well.
Fixes: http://tracker.ceph.com/issues/21200
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
(cherry picked from commit 87e8e89fe11fa805739a2b85d1b009424641fa4e)
commit dfe42384fbce90ec4c315e10ebd6b4fc4234827b
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Wed Dec 28 19:22:32 2016 +0100
rgw: calculate and print Swift's X-Account-Storage-Policy-* headers.
Fixes: http://tracker.ceph.com/issues/17932
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 5e5c0677ba69f0d162df1da9a22b1fb56028baa4)
commit 6e4c63fb136d958e0b4fe279419978a322043201
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Wed Dec 28 17:14:35 2016 +0100
rgw: bucket linking stores also the info about a placement rule.
Fixes: http://tracker.ceph.com/issues/17932
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit de7a8679af93457c10bf514c17a7cfc5327e745c)
commit 933f840094c3614466904f0fce32c2346a8f1709
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Wed Dec 28 17:12:50 2016 +0100
rgw: convey placement rule in RGWBucketEnt and cls_user_bucket_entry.
Fixes: http://tracker.ceph.com/issues/17932
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 5eca89f08d29ac6aa7b55a3b4fb5b4183bdc0496)
commit 5a1d0c42854fb35d44e63859d81074867f1a3295
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Wed Dec 28 15:54:42 2016 +0100
rgw: clean-up around and implement the move semantics in RGWBucketEnt.
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit 545dabe31a9ac231db4b8e3ce674ddfd9c91ee20)
commit bb92cc93ac32c1f98d7cfae778d684b94dd2de58
Author: Radoslaw Zarzynski <rzarzyns@redhat.com>
Date: Tue Aug 29 19:16:20 2017 +0200
rgw: enforce the std::move semantic across the path of RGWUserBuckets.
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit 4a7a4344840c873189331f22c8143ba031622d20)
commit 395e9a48d32db7dfe0419316772886b939a44233
Author: Radoslaw Zarzynski <rzarzyns@redhat.com>
Date: Tue Aug 29 19:19:59 2017 +0200
rgw: {end_}marker params are handled during Swift's reversed account listing.
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit 2f1f7a179d6a031c77929359313fc1fcc9f75bef)
commit 8582664267b2e72a836e27df859910481371dee0
Author: Radoslaw Zarzynski <rzarzyns@redhat.com>
Date: Mon Aug 28 21:54:56 2017 +0200
rgw: add basic support for Swift's reversed account listings.
Fixes: http://tracker.ceph.com/issues/21148
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit 870f54806b7c1f570d2e82d3d2aa308b2178c6f0)
commit d61de49cf270fedb206530a9d3cc1b4970f2b484
Author: Radoslaw Zarzynski <rzarzyns@redhat.com>
Date: Mon Aug 28 21:53:28 2017 +0200
rgw: abstract partial data processing in RGWListBuckets.
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit a842dd09d584ff7636ec972646caff218df59c4b)
commit 957e221f1ffc73726187aba99e9e103291b0ba6f
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Wed May 24 16:50:26 2017 +0200
rgw: list_objects() honors end_marker regardless of namespace.
This patch fixes a regression related to handling of the end_marker
parameter during Swift's container listing operation. It has been
introduced in a5d1fa0587184f43c69d8e03114b58d43f320781 and causes
Tempest's test_list_container_contents_with_end_marker to fail.
Fixes: http://tracker.ceph.com/issues/18977
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 09db1e412ef0853e7f345f813c8d452ec8354c36)
commit 004240eea79b02eb830f3c9e72de8e0216ff7557
Author: Sage Weil <sage@redhat.com>
Date: Sat Sep 9 22:47:06 2017 -0400
qa/standalong/mon/osd-pool-create: fewer pgs in test
This runs afoul of the new max pg per osd limit.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c9ffeeebebe84111ce3be51c04170804bf2dbc74)
commit 54ba2dda2034dd80594f17261451832744d2c7bf
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 8 08:08:30 2017 -0400
mon/OSDMonitor: assume a minimum cluster size of 3
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 1010761e6ad392fcaa647ec0d1b3d071535adfb3)
commit f581731eaeb0f0c625f62928bf3724f8e705132c
Author: Sage Weil <sage@redhat.com>
Date: Fri Sep 1 14:45:12 2017 -0400
mon/OSDMonitor: prevent pg_num from exceeding mon_pg_warn_max_per_osd
Check total pg count for the cluster vs osd count and max pgs per osd
before allowing pool creation, pg_num change, or pool size change.
"in" OSDs are the ones we distribute data too, so this should be the right
count to use. (Whether they happen to be up or down at the moment is
incidental.)
If the user really wants to create the pool, they can change the
configurable limit.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3ea2e518d27e6c06182c2cb3d9c0b9a0dab8dd22)
commit acb2f280f03324e2ddb9d359e18ff2f0a3f897e6
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 6 11:41:07 2017 -0400
common/options: reduce mon_pg_warn_max_per_osd to 200
This is 2x the recommended target (100 per OSD).
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit b7fa440a9802005c167f854e1d19f9dd4fa8dfff)
commit 119995a8f0ed70fb3e9efb5511576c44cfd6dbaa
Author: Radoslaw Zarzynski <rzarzyns@redhat.com>
Date: Tue Sep 19 14:46:53 2017 +0200
rgw: rename the configurables for metadata limits to start with rgw_.
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit 9b06985caec27fc102769e46799ca6608e80eb85)
commit 6080248f12e05a7478af1f8f31747da1aa9b881c
Author: Radoslaw Zarzynski <rzarzyns@redhat.com>
Date: Tue Aug 29 12:16:30 2017 +0200
rgw: return proper message when deleting non-empty Swift's container.
The bug that is fixed in this patch has been responsible for failing
the Tempest's test_delete_non_empty_container test case.
The investigation has been made by: Marcus Watts <mwatts@redhat.com>.
Fixes: http://tracker.ceph.com/issues/21169
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
(cherry picked from commit 308c8d3ab7e3e4978beccba2c27beb95e75cde22)
commit 71e37b1a10796ddfe786a2d389b0497ce1dc1bf4
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Sun Nov 20 21:49:57 2016 +0100
rgw: seed::get_torrent_file returns errors in the usual way.
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 21ad80a1c237f8d243efaacd8218b29719a0a6b0)
commit 429f5e1d8b7ec114289ba45e7c896e264b15ec7c
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Tue May 30 16:36:06 2017 +0200
rgw: add support for max_meta_count of Swift API's /info.
Fixes: http://tracker.ceph.com/issues/17934
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 9e53c14a3398dfe55f6b816bb57c4b1b6d01c8ec)
commit 70e150cdfa108f469d85a27ad1a93952f71347e0
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Tue May 30 22:19:33 2017 +0200
rgw: Swift API returns 400 Bad Request on too long container names.
Fixes: http://tracker.ceph.com/issues/17935
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 1845e41292696da9e20ecd87b9260b44ee312aed)
commit 02c208a6b1ae335ab15c0fab3d9d6dba803037fa
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Mon May 29 19:08:21 2017 +0200
rgw: honor custom rgw_err::message in Swift's error handling.
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit e8516147d2926631fed98a793606bcbb0e6db9e1)
commit 912132b272dbf534e7594b40a78c5efcd51c9466
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Mon May 29 18:02:23 2017 +0200
rgw: add support for max_meta_value_length of Swift API's /info.
Fixes: http://tracker.ceph.com/issues/17936
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 06b1f25021924e0c55da0bc22216f31344735dfe)
commit cd219153978974a3711093912ca329ae5d1d997c
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Mon Nov 21 19:34:48 2016 +0100
rgw: refactor rgw_get_request_metadata to reduce the number of dynallocs.
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 64b792326350c4c09ed5f829d082d4050bfb0408)
commit 4b82d1e25bf8f7be1915e7467cb822d5beea96a1
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Tue May 30 22:19:58 2017 +0200
rgw: add support for max_meta_name_length of Swift API's /info.
Fixes: http://tracker.ceph.com/issues/17938
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 593d656dffd477943b8bd15153004d905b4b0d73)
commit fb7e6a581d063db10bb7e0cce4ae358068c7a2bd
Author: Nathan Cutler <ncutler@suse.com>
Date: Mon Sep 18 18:01:17 2017 +0200
tests: CentOS 7.4 is now the latest
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit 2311b64025cdb6131035aaf01e7c97486da12e15)
commit e25a583323a225dcfab0567c872247b33bf36f8f
Author: xie xingguo <xie.xingguo@zte.com.cn>
Date: Mon Sep 18 14:44:36 2017 +0800
osd/PrimaryLogPG: _delete_oid - fix incorrect 'legacy' flag
For pre-Luminous created objects, we shall default 'legacy' flag
to true, so we can still create a snapdir object properly if necessary
for Luminous backward compatibility.
Note that this patch is not going to land on master first
(and hence can not be cherry-picked from master) because it will
finally be deprecated by https://github.com/ceph/ceph/pull/17579,
in which we are going to kill the snapdir object completely for Mimic.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
commit 5aa445897c293cae407454fadbbc47116a2cbe76
Author: xie xingguo <xie.xingguo@zte.com.cn>
Date: Sat Sep 16 10:04:56 2017 +0800
qa: fix pool-quota related tests
https://github.com/ceph/ceph/pull/17371 introduces support of
per-pool space-full flag, which turns out to set both
full and full_no_quota flags now if a pool is currently running out
of quota.
Actually this test is fragile as long as we keep appending new flags
at pool granularity, but let's not bother with that complexity now.
Fixes: http://tracker.ceph.com/issues/21409
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit 5dd3458eec7d6c56a71ee6f0fcdd45aa15a58ec7)
commit 00b44ecf4415710150e0646174174fbde0167f42
Author: xie xingguo <xie.xingguo@zte.com.cn>
Date: Mon Aug 28 15:51:28 2017 +0800
mon, osd: per pool space-full flag support
The newly introduced 'device-class' can be used to separate
different type of devices into different pools, e.g, hdd-pool
for backup data and all-flash-pool for DB applications.
However, if any osd of the cluster is currently running out
of space (exceeding the predefined 'full' threshold), Ceph
will mark the whole cluster as full and prevent writes to all pools,
which turns out to be very wrong.
This patch instead makes the space 'full' control at pool granularity,
which exactly leverages the pool quota logic but shall solve
the above problem.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit b4ca5ae462c6f12ca48b787529938862646282cd)
Conflicts:
slight confilicts in src/osd/OSDMap.h because
e71626732452951ed9e6c489f15618424d15acf2 removed get_full_osd_util().
commit b064ed172a309aad4e8850110be4311466bf5894
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 14 10:02:33 2017 -0400
os/bluestore: repair 21089 on freelist init
Fix up the size inconsistency on freelist init. This way it will always
happen after an upgrade... and before the user moves to something
post-luminous.
Signed-off-by: Sage Weil <sage@redhat.com>
commit 1bf89aa3ce2f1fdb7979e536f6788c8dae275d47
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 14 10:01:25 2017 -0400
os/bluestore: fsck: remove fsck repair for 21089
This requires the user to run fsck, which they likely won't do.
Signed-off-by: Sage Weil <sage@redhat.com>
commit e9f143c681ab1a79b9c726cab55282d71ecfe1c7
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 13 23:03:07 2017 -0400
os/bluestore/KernelDevice: hack to inject bad device size
This conditionally reintroduces bug http://tracker.ceph.com/issues/21089,
fixed by f6f1ae3724d593d3709d982c973ec18a25a47b6e, to aid in testing
repair.
Signed-off-by: Sage Weil <sage@redhat.com>
commit 460268b5e9c9a77aae134049aa320977f02fc911
Author: Patrick Donnelly <pdonnell@redhat.com>
Date: Wed Aug 30 15:28:11 2017 -0700
ceph.in: validate service glob
Fixes: http://tracker.ceph.com/issues/21191
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
(cherry picked from commit 59bee1e7274934cebe0dcefd9fcedb6886a5e2d0)
commit 03e2a68e431eaa737b9e66c4d3b289e58d6cf19b
Author: Nathan Cutler <ncutler@suse.com>
Date: Wed Aug 30 10:17:31 2017 +0200
build/ops: python-numpy-devel build dependency for SUSE
Fixes: http://tracker.ceph.com/issues/21176
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit 1586f2ca9ab94af85682945a3c7c7ebbd82c6e03)
Reference in New Issue View Git Blame Copy Permalink