ceph/doc/changelog/v0.80.8.txt
Sage Weil 5d5b510810 doc/release-notes: v0.80.8 and .9 changelogs
Signed-off-by: Sage Weil <sage@redhat.com>
2015-03-10 12:32:47 -07:00

2548 lines
88 KiB
Plaintext

commit 69eaad7f8308f21573c604f121956e64679a52a7 (tag: refs/tags/v0.80.8)
Author: Jenkins <jenkins@inktank.com>
Date: Tue Jan 13 06:28:08 2015 -0800
0.80.8
commit 3c7cacf00b66a0e17272cf67fe0823ee863dfa7c
Author: Sage Weil <sage@redhat.com>
Date: Tue Aug 19 16:43:02 2014 -0700
mon/OSDMonitor: fix double-free on old MOSDBoot
send_latest() does an m->put().
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 97f9b6df668315fba6a5924b79024c7a986f4110)
commit 852d7b5b3c019c02c042b767fc88916088e1a94d
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 8 11:17:03 2015 -0800
osd: requeue PG when we skip handling a peering event
If we don't handle the event, we need to put the PG back into the peering
queue or else the event won't get processed until the next event is
queued, at which point we'll be processing events with a delay.
The queue_null is not necessary (and is a waste of effort) because the
event is still in pg->peering_queue and the PG is queued.
This is a firefly-specific patch; a (very) similar one will appear in master
in 492ccc900c3358f36b6b14a207beec071eb06707.
Backport: giant, firefly
Signed-off-by: Sage Weil <sage@redhat.com>
commit c26ebd38335bb361aade5aacd05ba3217e602b9c
Merge: b67b7e7 96ba529
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 8 18:31:26 2015 -0800
Merge pull request #3217 from boydc2014/firefly
clear data and payload after removed from ops_in_flight
Reviewed-by: Sage Weil <sage@redhat.com>
commit b67b7e7ad72a1af96f9fb26ade815e65f72b4cb0
Merge: db92718 7faae89
Author: Loic Dachary <ldachary@redhat.com>
Date: Fri Jan 9 01:32:17 2015 +0100
Merge pull request #3127 from ktdreyer/firefly-no-epoch
Revert "ceph.spec.: add epoch"
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit db927186288cd4c63a3483b42e9eb9e016c96156
Merge: 0d4abda 820dbfd
Author: Loic Dachary <ldachary@redhat.com>
Date: Fri Jan 9 01:31:29 2015 +0100
Merge pull request #3128 from dachary/wip-10281-make-check-fedora-20
tests: fixes to run make check on fedora 20
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 0d4abdaf80f1fedff7975d595abaac0a620c8035
Merge: 1fdcb52 11995b3
Author: Loic Dachary <ldachary@redhat.com>
Date: Fri Jan 9 01:30:59 2015 +0100
Merge pull request #3169 from ceph/wip-8797-firefly
Wip 8797 firefly
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 1fdcb524411a02b5627be66d9fd821a473223e9d
Merge: 4b7b1b0 465eede
Author: Loic Dachary <ldachary@redhat.com>
Date: Fri Jan 9 01:28:49 2015 +0100
Merge pull request #3179 from dachary/wip-9998-crush-underfloat-firefly
crush: fix weight underfloat issue (firefly)
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 4b7b1b03bfcb7cb056783555884f211009ea5d46
Merge: 4897ba4 f55b097
Author: Loic Dachary <ldachary@redhat.com>
Date: Fri Jan 9 01:28:11 2015 +0100
Merge pull request #3220 from ceph/wip-mon-backports.firefly
mon: backports for #9987 against firefly
Reviewed-by: Joao Eduardo Luis <joao@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 4897ba4a304bcac548b2121312cd7235c34dd5aa
Merge: efe801f f20225c
Author: Loic Dachary <ldachary@redhat.com>
Date: Fri Jan 9 01:26:30 2015 +0100
Merge pull request #3258 from ceph/wip-10372-firefly
osd: fix librados pool deletion race on firefly
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit efe801ff3a0d25243da18937c07b89227edbaac4
Author: Warren Usui <warren.usui@inktank.com>
Date: Thu Dec 18 20:00:28 2014 -0800
If trusty, use older version of qemu
Fixes #10319
Signed-off-by: Warren Usui <warren.usui@inktank.com>
(cherry-picked from 46a1a4cb670d30397979cd89808a2e420cef2c11)
commit 96ba529ef8ce76e07b8444c94883afb3468d6762
Author: Lei Dong <leidong@yahoo-inc.com>
Date: Tue Dec 30 21:02:45 2014 +0800
clear data and payload inside ops_in_flight_lock
http://tracker.ceph.com/issues/9916 Fixes: #9916
Signed-off-by: Dong Lei <leidong@yahoo-inc.com>
commit e0648e3d30de504b096c4ae3bbe7d9c17652bdb5
Merge: 455f940 3624006
Author: Sage Weil <sage@redhat.com>
Date: Mon Dec 29 14:31:23 2014 -0800
Merge pull request #3264 from dachary/wip-jerasure-firefly
erasure-code: update links to jerasure upstream
commit 455f940908f242b1e34983af61351fd3045ce8ab
Merge: b1ab685 aa95a2d
Author: Sage Weil <sage@redhat.com>
Date: Mon Dec 29 10:55:39 2014 -0800
Merge pull request #3268 from ceph/firefly-10415
libcephfs/test.cc: close fd before umount
commit 362400667aad0b5098fbe8dbec1b0bde059f84a6
Author: Loic Dachary <ldachary@redhat.com>
Date: Sun Dec 28 10:29:54 2014 +0100
erasure-code: update links to jerasure upstream
It moved from bitbucket to jerasure.org
Signed-off-by: Loic Dachary <ldachary@redhat.com>
(cherry picked from commit 8e86f901939f16cc9c8ad7a4108ac4bcf3916d2c)
commit aa95a2d20dbba2f3a775f709493c987d0d001e9c
Author: Yan, Zheng <zyan@redhat.com>
Date: Tue Dec 23 10:22:00 2014 +0800
libcephfs/test.cc: close fd before umount
Fixes: #10415
Signed-off-by: Yan, Zheng <zyan@redhat.com>
(cherry picked from commit d3fb563cee4c4cf08ff4ee01782e52a100462429)
commit f20225cb99a0d2d08fccfdf88dc89d758ecba077 (refs/remotes/gh/wip-10372-firefly)
Author: Sage Weil <sage@redhat.com>
Date: Tue Dec 23 15:49:26 2014 -0800
osdc/Objecter: handle reply race with pool deletion
We need to handle this scenario:
- send request in epoch X
- osd replies
- pool is deleted in epoch X+1
- client gets map X+1, sends a map check
- client handles reply
-> asserts that no map checks are in flight
This isn't the best solution. We could infer that a map check isn't needed
since the pool existed earlier and doesn't now. But this is firefly and
the fix is no more expensive than the old assert.
Fixes: #10372
Signed-off-by: Sage Weil <sage@redhat.com>
commit 820dbfd9947455d07426981b7152861c3c216080
Author: Sage Weil <sage@redhat.com>
Date: Tue Aug 5 15:11:18 2014 -0700
test/ceph-disk.sh: mkdir -p
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c2f58e6694a2457200ab3d59e037ad17b9c82028)
commit 77d393024f9d867b574b8ec8e15ec48a1a291511
Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Date: Thu Sep 4 12:23:27 2014 +0200
test/ceph-disk.sh: resolve symlinks before check
Make sure symlinks are resolved in command_fixture()
before compare result of which command and the current
path.
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
(cherry picked from commit 8ea86dfa7c4a3d7e089cf9d4e49586657875f851)
commit ed6ec2936513d7dd6c45bccd8edf69a12c71dc7b
Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Date: Tue Jun 24 19:54:17 2014 +0200
test/ceph-disk.sh: fix for SUSE
On SUSE 'which' returns always the full path of (shell) commands and
not e.g. './ceph-conf' as on Debian. Add check also for full
path returned by which.
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
(cherry picked from commit 39530536ff923b91899f6303507c283b78040a20)
commit 754363f4563e7dbda1ef23fadc8d6ef1a3fdd0af
Author: Loic Dachary <loic@dachary.org>
Date: Fri Jun 13 14:41:39 2014 +0200
tests: prevent kill race condition
When trying to kill a daemon, keep its pid in a variable instead of
retrieving it from the pidfile multiple times. It prevents the following
race condition:
* try to kill ceph-mon
* ceph-mon is in the process of dying and removed its pidfile
* try to kill ceph-mon fails because the pidfile is not found
* another ceph-mon is spawned and fails to bind the port
because the previous ceph-mon is still holding it
Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit a1c13c57ba20fc329d943ea57523913e11067dc7)
commit 5be6f2f60e3225bf3d214432044721fe474d55d7
Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Date: Wed Jun 25 00:31:48 2014 +0200
osd/OSD.cc: parse lsb release data via lsb_release
Use lsb_release tool to be portable since parsing /etc/lsb-release
is not the same between different distributions. The old code failed
e.g. for SUSE products to parse LSB information.
Fixes: #8654
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
(cherry picked from commit 0b3a3987d382ff33fdf892f189b30df22be80e59)
commit b62187c52324a4489c2fc1385b1d6574a058f7e8
Author: Loic Dachary <loic-201408@dachary.org>
Date: Tue Aug 26 21:59:39 2014 +0200
tests: histogram prevent re-use of local variables
By moving the tests to separate functions.
http://tracker.ceph.com/issues/9235 Fixes: #9235
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit 4b8b25ecd128c34a386ad7c4cc89f323c4d384e1)
commit e2741c8f1d42cfe91b18201a6a49005d90d85d98
Author: Loic Dachary <loic-201408@dachary.org>
Date: Tue Aug 26 21:59:39 2014 +0200
tests: histogram prevent re-use of local variables
By moving the test to a separate function.
http://tracker.ceph.com/issues/9235 Fixes: #9235
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit ee02293ad2ef050672fa8c164ba17b10e8d4ceeb)
commit cf102df7a3f3b38824c26b9e44a21664fcf979a7
Author: Loic Dachary <ldachary@redhat.com>
Date: Wed Dec 10 00:08:57 2014 +0100
tests: avoid tab interpretation problem on fedora 20
Use . instead of tab in echo to avoid variations in how escape sequences
are interpreted by the shell.
http://tracker.ceph.com/issues/10281 Fixes: #10281
Signed-off-by: Loic Dachary <ldachary@redhat.com>
commit f55b097764beb973c12866f2d7161c6bd870aa07 (refs/remotes/gh/wip-mon-backports.firefly)
Author: Sage Weil <sage@redhat.com>
Date: Sun Nov 2 08:50:59 2014 -0800
mon/PGMap and PGMonitor: update last_epoch_clean cache from new osd keys
We were only invalidating the cached value from apply_incremental, which
is no longer called on modern clusters.
Fix this by storing the update epoch in the key as well (it is not part
of osd_stat_t).
Backport: giant, firefly, dumpling(?)
Fixes: #9987
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 093c5f0cabeb552b90d944da2c50de48fcf6f564)
commit 1d314e7a9ab7af5b693583cf2faa5db54f6beb69
Author: Sage Weil <sage@redhat.com>
Date: Sun Nov 2 08:49:48 2014 -0800
mon/PGMap: invalidate cached min_last_epoch_clean from new-style pg keys
We were only invalidating the cache from the legacy apply_incremental(),
which is no longer called on modern clusters.
Fixes: #9987
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3fb731b722c50672a5a9de0c86a621f5f50f2d06)
commit 465eedea9f7411b1e352dc3ccee60a3f1221541d
Author: Sage Weil <sage@redhat.com>
Date: Sun Nov 23 18:50:51 2014 -0800
crush/CrushWrapper: fix create_or_move_item when name exists but item does not
We were using item_exists(), which simply checks if we have a name defined
for the item. Instead, use _search_item_exists(), which looks for an
instance of the item somewhere in the hierarchy. This matches what
get_item_weightf() is doing, which ensures we get a non-negative weight
that converts properly to floating point.
Backport: giant, firefly
Fixes: #9998
Reported-by: Pawel Sadowski <ceph@sadziu.pl>
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 9902383c690dca9ed5ba667800413daa8332157e)
commit cee51af9c1dbde550177c95caf6c93f612442300
Author: Sage Weil <sage@redhat.com>
Date: Fri Nov 21 17:47:56 2014 -0800
crush/builder: prevent bucket weight underflow on item removal
It is possible to set a bucket weight that is not the sum of the item
weights if you manually modify/build the CRUSH map. Protect against any
underflow on the bucket weight when removing items.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 8c87e9502142d5b4a282b94f929ae776a49be1dc)
commit 7ccd5eec11e8cd945d24bf9f6390d6c3fa4a06de
Author: Sage Weil <sage@redhat.com>
Date: Fri Nov 21 17:37:03 2014 -0800
crush/CrushWrapper: fix _search_item_exists
Reported-by: Pawel Sadowski <ceph@sadziu.pl>
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit eeadd60714d908a3a033aeb7fd542c511e63122b)
commit b1ab685e00034751a161a3d5e0325c6581999c75
Merge: dd7c8c2 ef3773a
Author: Sage Weil <sage@redhat.com>
Date: Fri Dec 12 06:19:50 2014 -0800
Merge pull request #3124 from ceph/wip-10194-firefly
rgw: optionally call FCGX_Free() on the fcgi connection
Reviewed-by: Sage Weil <sage@redhat.com>
commit 11995b329045341c17553269267cfd3688a51b0f
Author: Dan Mick <dan.mick@redhat.com>
Date: Wed Dec 10 13:19:53 2014 -0800
Call Rados.shutdown() explicitly before exit
This is mostly a demonstration of good behavior, as the resources will
be reclaimed on exit anyway.
Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit b038e8fbf9103cc42a4cde734b3ee601af6019ea)
commit e00270b51896f168d5013b7dc92ec7f8b9e19da3
Author: Dan Mick <dan.mick@redhat.com>
Date: Wed Dec 10 13:19:16 2014 -0800
rados.py: remove Rados.__del__(); it just causes problems
Recent versions of Python contain a change to thread shutdown that
causes ceph to hang on exit; see http://bugs.python.org/issue21963.
As it turns out, this is relatively easy to avoid by not spawning
threads on exit, as Rados.__del__() will certainly do by calling
shutdown(); I suspect, but haven't proven, that the problem is
that shutdown() tries to start() a threading.Thread() that never
makes it all the way back to signal start().
Also add a PendingReleaseNote and extra doc comments to clarify.
Fixes: #8797
Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit 5ba9b8f21f8010c59dd84a0ef2acfec99e4b048f)
Conflicts:
PendingReleaseNotes
commit 7faae891aefa4c21c50430fa03d9204a86d082f8
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Tue Dec 9 14:52:19 2014 -0700
Revert "ceph.spec.: add epoch"
If ICE ships 0.80.8, then it will be newer than what RHCEPH ships
(0.80.7), and users won't be able to seamlessly upgrade via Yum.
We have three options:
A) Revert the "Epoch: 1" change on the Firefly branch.
B) Revert the "Epoch: 1" change in the ICE packages.
C) Bump the Epoch to "2" in Red Hat's packages.
This commit does Option A.
Option B may or may not be feasible - it would require a "downstream"
change in ICE, and we haven't done that sort of thing before.
Due to the RHEL release schedule, Option C is not available to us at
this point.
This reverts commit b890c1e4706d7cfef7ed24c9df65b439b4f7ff1d.
commit ef3773ac93413c644e056babce1971f846bbc276
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Wed Nov 26 15:18:07 2014 -0800
rgw: optionally call FCGX_Free() on the fcgi connection
Fixes: #10194
A new configurable controls this behavior. This forces disconnection of
the fcgi connection when done with the request.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
commit dd7c8c2bc6b2810ff7f483af940fa09dbe74e83a
Merge: d759467 a597096
Author: Gregory Farnum <greg@gregs42.com>
Date: Mon Dec 8 15:02:52 2014 -0800
Merge pull request #3109 from ceph/firefly-10263
mds: store backtrace for straydir
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
commit d7594672b673796901961cc684c9e7de8dc2c95d
Merge: adf9a75 73b47db
Author: Samuel Just <sam.just@inktank.com>
Date: Mon Dec 8 13:19:44 2014 -0800
Merge pull request #3009 from dachary/wip-10018-primary-erasure-code-hinfo-firefly
osd: deep scrub must not abort if hinfo is missing (firefly)
Reviewed-by: Samuel Just <sjust@redhat.com>
commit a5970963a2148697fc6da64facfbf6ab6686b9cd
Author: Yan, Zheng <zyan@redhat.com>
Date: Fri Nov 7 11:38:37 2014 +0800
mds: store backtrace for straydir
Backport: giant, firefly, emperor, dumpling
Signed-off-by: Yan, Zheng <zyan@redhat.com>
(cherry picked from commit 0d89db5d3e5ae5d552d4058a88a4e186748ab1d2)
commit adf9a758000182d27a6582d516356730d02e4099
Merge: dea38a7 b4a4b75
Author: Sage Weil <sage@redhat.com>
Date: Sat Dec 6 11:06:02 2014 -0800
Merge pull request #3089 from dachary/wip-10063-hobject-shard-firefly
common: do not omit shard when ghobject NO_GEN is set (firefly)
commit dea38a7af638c833304272c324ed2bc386a40f8f
Merge: ccc8b46 5138091
Author: Sage Weil <sage@redhat.com>
Date: Sat Dec 6 10:59:44 2014 -0800
Merge pull request #2480 from dachary/wip-9420-erasure-code-non-regression-firefly
erasure-code: store and compare encoded contents (firefly)
commit ccc8b46b2cdc55c1a861f092259ef36a1296f073
Merge: bef363c cd3447d
Author: Sage Weil <sage@redhat.com>
Date: Fri Dec 5 17:33:05 2014 -0800
Merge pull request #3096 from dachary/wip-9785-dmcrypt-keys-permissions-firefly
ceph-disk: dmcrypt file permissions (firefly)
commit cd3447d04cabf6745001afeef69f25a92400cd0e
Author: Loic Dachary <ldachary@redhat.com>
Date: Thu Dec 4 22:21:32 2014 +0100
ceph-disk: dmcrypt file permissions
The directory in which key files are stored for dmcrypt must be 700 and
the file 600.
http://tracker.ceph.com/issues/9785 Fixes: #9785
Signed-off-by: Loic Dachary <ldachary@redhat.com>
(cherry picked from commit 58682d1776ab1fd4daddd887d921ca9cc312bf50)
commit bef363c1d1fc06fbb315024145a97a809a2471cd
Merge: cb2c83b 9f3b21d
Author: Sage Weil <sage@redhat.com>
Date: Fri Dec 5 09:04:00 2014 -0800
Merge pull request #3086 from dachary/wip-10125-radosgw-init-firefly
rgw: run radosgw as apache with systemd (firefly)
commit b4a4b75e6d4deb6818681902f85baa9f63acdb4f
Author: Loic Dachary <ldachary@redhat.com>
Date: Fri Nov 14 01:16:10 2014 +0100
common: do not omit shard when ghobject NO_GEN is set
Do not silence the display of shard_id when generation is NO_GEN.
Erasure coded objects JSON representation used by ceph_objectstore_tool
need the shard_id to find the file containing the chunk.
Minimal testing is added to ceph_objectstore_tool.py
http://tracker.ceph.com/issues/10063 Fixes: #10063
Signed-off-by: Loic Dachary <ldachary@redhat.com>
(cherry picked from commit dcf09aed121f566221f539106d10283a09f15cf5)
Conflicts:
src/test/ceph_objectstore_tool.py
commit 9f3b21d1b70be591d68bfa57c3393e8f9af8e7df
Author: Loic Dachary <ldachary@redhat.com>
Date: Tue Dec 2 18:10:48 2014 +0100
rgw: run radosgw as apache with systemd
Same as sysv.
http://tracker.ceph.com/issues/10125 Fixes: #10125
Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 7b621f4abf63456272dec3449aa108c89504a7a5)
Conflicts:
src/init-radosgw.sysv
commit cb2c83b2f216e503f7a52115f775bda1dbfe0c6a
Merge: e2ec37b 02d4685
Author: Josh Durgin <jdurgin@redhat.com>
Date: Thu Dec 4 11:32:18 2014 -0800
Merge pull request #3078 from ceph/wip-10030-firefly
librbd: don't close an already closed parent image upon failure
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit e2ec37bf45fa5e7f5e787db9b67dbb2a98f2fbb7
Merge: c4c63e8 af12194
Author: Sage Weil <sage@redhat.com>
Date: Wed Dec 3 23:01:44 2014 -0800
Merge pull request #3063 from ceph/wip-10123-firefly
librbd: protect list_children from invalid child pool IoCtxs
Reviewed-by: Sage Weil <sage@redhat.com>
commit c4c63e82cd7e4716315ca81208293a2567026c72
Author: Samuel Just <sam.just@inktank.com>
Date: Tue Sep 23 15:52:08 2014 -0700
ReplicatedPG: don't move on to the next snap immediately
If we have a bunch of trimmed snaps for which we have no
objects, we'll spin for a long time. Instead, requeue.
Fixes: #9487
Backport: dumpling, firefly, giant
Reviewed-by: Sage Weil <sage@redhat.com>
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit c17ac03a50da523f250eb6394c89cc7e93cb4659)
commit 1b656450ca75b12fb98dee82bace914ef5f45c44
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 23 16:21:33 2014 -0700
osd: initialize purged_snap on backfill start; restart backfill if change
If we backfill a PG to a new OSD, we currently neglect to initialize
purged_snaps. As a result, the first time the snaptrimmer runs it has to
churn through every deleted snap for all time, and to make matters worse
does so in one go with the PG lock held. This leads to badness on any
cluster with a significant number of removed snaps that experiences
backfill.
Resolve this by initializing purged_snaps when we finish backfill. The
backfill itself will clear out any stray snaps and ensure the object set
is in sync with purged_snaps. Note that purged_snaps on the primary
that is driving backfill will not change during this period as the
snaptrimmer is not scheduled unless the PG is clean (which it won't be
during backfill).
If we by chance to interrupt backfill, go clean with other OSDs,
purge snaps, and then let this OSD rejoin, we will either restart
backfill (non-contiguous log) or the log will include the result of
the snap trim (the events that remove the trimmed snap).
Fixes: #9487
Backfill: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 255b430a87201c7d0cf8f10a3c1e62cbe8dd2d93)
commit 02d4685c56e129cb179a5ddfb8e87aefc2fce0b5
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Nov 6 05:01:38 2014 -0500
librbd: don't close an already closed parent image upon failure
If librbd is not able to open a child's parent image, it will
incorrectly close the parent image twice, resulting in a crash.
Fixes: #10030
Backport: firefly, giant
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 61ebfebd59b61ffdc203dfeca01ee1a02315133e)
commit af121942d7bdfc59fcfae0429ffb12993e7e019d
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Nov 17 21:49:26 2014 -0500
librbd: protect list_children from invalid child pool IoCtxs
While listing child images, don't ignore error codes returned
from librados when creating an IoCtx. This will prevent seg
faults from occurring when an invalid IoCtx is used.
Fixes: #10123
Backport: giant, firefly, dumpling
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 0d350b6817d7905908a4e432cd359ca1d36bab50)
commit c982da44e0e9e0be3c3d4e8f5e0a186fb2fcebb3
Merge: 4a148df c4e4a31
Author: Sage Weil <sage@redhat.com>
Date: Sun Nov 30 10:12:04 2014 -0800
Merge pull request #3014 from dachary/wip-9665-ceph-disk-partprobe-firefly
ceph disk zap must call partprobe
commit c4e4a310f14ca3049ac90422aea95051fe0d4b15
Author: Loic Dachary <loic-201408@dachary.org>
Date: Fri Oct 10 10:26:31 2014 +0200
ceph-disk: use update_partition in prepare_dev and main_prepare
In the case of prepare_dev the partx alternative was missing and is not
added because update_partition does it.
http://tracker.ceph.com/issues/9721 Fixes: #9721
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit 23e71b1ee816c0ec8bd65891998657c46e364fbe)
Conflicts:
src/ceph-disk
commit e70a81464b906b9a304c29f474e6726762b63a7c
Author: Loic Dachary <loic-201408@dachary.org>
Date: Thu Oct 9 18:52:17 2014 +0200
ceph-disk: run partprobe after zap
Not running partprobe after zapping a device can lead to the following:
* ceph-disk prepare /dev/loop2
* links are created in /dev/disk/by-partuuid
* ceph-disk zap /dev/loop2
* links are not removed from /dev/disk/by-partuuid
* ceph-disk prepare /dev/loop2
* some links are not created in /dev/disk/by-partuuid
This is assuming there is a bug in the way udev events are handled by
the operating system.
http://tracker.ceph.com/issues/9665 Fixes: #9665
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit fed3b06c47a5ef22cb3514c7647544120086d1e7)
commit 5a5f427bc09076ef3fb13a710dede1b47bb232e0
Author: Loic Dachary <loic-201408@dachary.org>
Date: Fri Oct 10 10:23:34 2014 +0200
ceph-disk: encapsulate partprobe / partx calls
Add the update_partition function to reduce code duplication.
The action is made an argument although it always is -a because it will
be -d when deleting a partition.
Use the update_partition function in prepare_journal_dev
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit 922a15ea6865ef915bbdec2597433da6792c1cb2)
Conflicts:
src/ceph-disk
commit 73b47dbee8858f182fd2b4fd8eb5f3c786877bf4
Author: Loic Dachary <loic-201408@dachary.org>
Date: Thu Nov 6 17:11:20 2014 +0100
osd: deep scrub must not abort if hinfo is missing
Instead it should set read_error.
http://tracker.ceph.com/issues/10018 Fixes: #10018
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit 9d84d2e8309d26e39ca849a75166d2d7f2dec9ea)
commit 5138091a4073d966a65f537280f89372e801d019
Author: Loic Dachary <loic-201408@dachary.org>
Date: Tue Sep 23 11:38:09 2014 +0200
erasure-code: add corpus verification to make check
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
commit 8d3d6bf59aec3877c0231e637270e371d9ed3b8c
Author: Loic Dachary <loic-201408@dachary.org>
Date: Sat Sep 13 13:36:09 2014 +0200
erasure-code: workunit to check for encoding regression
Clone the archive of encoded objects and decode all archived objects, up
to and including the current ceph version.
http://tracker.ceph.com/issues/9420 Refs: #9420
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
commit 4f4358708ed3c261ca4027cc9c3dc3f952a99470
Author: Loic Dachary <loic-201408@dachary.org>
Date: Sat Sep 13 10:16:31 2014 +0200
erasure-code: store and compare encoded contents
Introduce ceph_erasure_code_non_regression to check and compare how an
erasure code plugin encodes and decodes content with a given set of
parameters. For instance:
./ceph_erasure_code_non_regression \
--plugin jerasure \
--parameter technique=reed_sol_van \
--parameter k=2 \
--parameter m=2 \
--stripe-width 3181 \
--create \
--check
Will create an encoded object (--create) and store it into a directory
along with the chunks, one chunk per file. The directory name is derived
from the parameters. The content of the object is a random pattern of 31
bytes repeated to fill the object size specified with --stripe-width.
The check function (--check) reads the object back from the file,
encodes it and compares the result with the content of the chunks read
from the files. It also attempts recover from one or two erasures.
Chunks encoded by a given version of Ceph are expected to be encoded
exactly in the same way by all Ceph versions going forward.
http://tracker.ceph.com/issues/9420 Refs: #9420
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
commit 4a148df544978383c1ed7cd8b90f590adb563f3d
Merge: c069bce 01faf13
Author: Loic Dachary <loic-201408@dachary.org>
Date: Wed Nov 19 02:45:26 2014 +0100
Merge pull request #2961 from ceph/wip-10114-firefly
Add annotation to all assembly files to turn off stack-execute bit
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 01faf1356f648ded9acda02e7cc67c1adb9e9ee3
Author: Dan Mick <dan.mick@redhat.com>
Date: Fri Nov 14 17:59:57 2014 -0800
Add annotation to all assembly files to turn off stack-execute bit
See discussion in http://tracker.ceph.com/issues/10114
Building with these changes allows output from readelf like this:
$ readelf -lW src/.libs/librados.so.2 | grep GNU_STACK
GNU_STACK 0x000000 0x0000000000000000 0x0000000000000000 0x000000
0x000000 RW 0x8
(note the absence of 'X' in 'RW')
Fixes: #10114
Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit 06a245a9845c0c126fb3106b41b2fd2bc4bc4df3)
(not-yet-present-in-firefly files in isa-l manually removed)
commit c069bce4e8180da3c0ca4951365032a45df76468
Merge: 0d8ad6a fac1654
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Nov 13 10:36:12 2014 -0800
Merge pull request #2760 from ceph/wip-9835-firefly
osd: fix erasure hung op bug (9835)
Reviewed-by: Samuel Just <sjust@redhat.com>
commit fac165475031efdebbb88898ca5c12cd307a5bc3
Author: Samuel Just <sam.just@inktank.com>
Date: Wed Nov 5 12:12:14 2014 -0800
osd: use OSDMap helper to tell if ops are misdirected
calc_pg_role doesn't actually take into account primary affinity.
Fixes: #9835
Signed-off-by: Samuel Just <sam.just@inktank.com>
commit 588602bf0095de5b59064123ca01345f1364bdde
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 20 13:55:33 2014 -0700
osd: discard rank > 0 ops on erasure pools
Erasure pools do not support read from replica, so we should drop
any rank > 0 requests.
This fixes a bug where an erasure pool maps to [1,2,3], temporarily maps
to [-1,2,3], sends a request to osd.2, and then remaps back to [1,2,3].
Because the 0 shard never appears on osd.2, the request sits in the
waiting_for_pg map indefinitely and cases slow request warnings.
This problem does not come up on replicated pools because all instances of
the PG are created equal.
Fix by only considering role == 0 for erasure pools as a correct mapping.
Fixes: #9835
Signed-off-by: Sage Weil <sage@redhat.com>
commit 0c1c4152e6f402af7612c8c8d4719ab0f4cc6ad9
Author: Sage Weil <sage@redhat.com>
Date: Wed Nov 12 17:04:35 2014 -0800
osd/OSDMap: add osd_is_valid_op_target()
Helper to check whether an osd is a given op target for a pg. This
assumes that for EC we always send ops to the primary, while for
replicated we may target any replica.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 89c02637914ac7332e9dbdbfefc2049b2b6c127d)
commit 0d8ad6ad3c376bcab981bea9a49e1924d7eddb68
Author: Josh Durgin <jdurgin@redhat.com>
Date: Tue Nov 11 18:16:02 2014 -0800
qa: allow small allocation diffs for exported rbds
The local filesytem may behave slightly differently. This isn't
foolproof, but seems to be reliable enough on rhel7 rootfs, where
exact comparison was failing.
Fixes: #10002
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit e94d3c11edb9c9cbcf108463fdff8404df79be33)
commit 0804deeab293e09123d1b58825051ccc4dddbc0e
Author: Sage Weil <sage@inktank.com>
Date: Sun May 25 08:38:38 2014 -0700
osd: fix map advance limit to handle map gaps
The recent change in cf25bdf6b0090379903981fe8cee5ea75efd7ba0 would stop
advancing after some number of epochs, but did not take into consideration
the possibilty that there are missing maps. In that case, it is impossible
to advance past the gap.
Fix this by increasing the max epoch as we go so that we can always get
beyond the gap.
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from commit 1e0a82fd55dede473c0af32924f4bcb5bb697a2b)
commit d30d6b986433eaef920b3703cf5af3c030f8dcf4
Merge: d241aa7 d548431
Author: Gregory Farnum <greg@gregs42.com>
Date: Fri Nov 7 14:10:18 2014 -0800
Merge pull request #2880 from ceph/wip-10025-firefly
#10025/firefly -- tools: fix MDS journal import
Reviewed-by: Greg Farnum <greg@inktank.com>
commit d548431a388da1130564d710e1f006772934224b
Author: John Spray <john.spray@redhat.com>
Date: Fri Nov 7 11:34:43 2014 +0000
tools: fix MDS journal import
Previously it only worked on fresh filesystems which
hadn't been trimmed yet, and resulted in an invalid
trimmed_pos when expire_pos wasn't on an object
boundary.
Fixes: #10025
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit fb29e71f9a97c12354045ad2e128156e503be696)
commit d241aa7a347655242cc71b8fa3d778df6948c494
Merge: 2c85b5d 4afb542
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Nov 6 10:37:42 2014 -0800
Merge remote-tracking branch 'origin/wip-sam-firefly-backports' into firefly
commit 2c85b5d72953d01296213185382707122e06415c
Merge: 23cbffa 1228658
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Nov 6 10:30:20 2014 -0800
Merge pull request #2737 from ceph/wip-9629-firefly
osd: do not clone/preserve snapdir on cache-evict (firefly backport)
Reviewed-by: Samuel Just <sjust@redhat.com>
commit 23cbffaa2936dc2707b5b42f8c0e7ce804324ae2
Merge: 3bba938 e296685
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Nov 6 10:26:21 2014 -0800
Merge pull request #2657 from ceph/wip-9053-9301-firefly
mon: backport two paxos fixes to firefly
Reviewed-by: Joao Luis <joao@redhat.com>
commit 3bba9387eb123c6cf055e874db2925b998dc406c
Merge: 3f9bf73 835f8c6
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Nov 6 10:21:12 2014 -0800
Merge pull request #2656 from ceph/wip-9502-firefly
mon: backport mon disk full check to firefly
Reviewed-by: Samuel Just <sjust@redhat.com>
commit 3f9bf738daf47ff4ff56c9f76d1487a5afc5e30a
Merge: a340e6d 3e17a08
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Nov 6 10:18:27 2014 -0800
Merge pull request #2764 from ceph/wip-9851
osd: bring FileJournal in sync with giant
Reviewed-by: Samuel Just <sjust@redhat.com>
commit a340e6d0b166019f58dca0703faf30dd3178c14f
Merge: b7d5f99 b9450b5
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Nov 6 10:12:17 2014 -0800
Merge pull request #2776 from ceph/wip-9675.firefly
CrushWrapper: pick a ruleset same as rule_id
Reviewed-by: Samuel Just <sam.just@inktank.com>
commit b7d5f99c8f4d751e83dc29305649d7a465c657b1
Author: Sage Weil <sage@redhat.com>
Date: Mon Sep 15 15:29:08 2014 -0700
ceph-disk: mount xfs with inode64 by default
We did this forever ago with mkcephfs, but ceph-disk didn't. Note that for
modern XFS this option is obsolete, but for older kernels it was not the
default.
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 11496399ef318498c11e551f139d96db52d3309c)
commit 1a9d000bb679a7392b9dd115373c3827c9626694
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Thu Oct 9 10:20:27 2014 -0700
rgw: set length for keystone token validation request
Fixes: #7796
Backport: giany, firefly
Need to set content length to this request, as the server might not
handle a chunked request (even though we don't send anything).
Tested-by: Mark Kirkwood <mark.kirkwood@catalyst.net.nz>
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 3dd4ccad7fe97fc16a3ee4130549b48600bc485c)
commit 49d27efde2ce5d282c9ee6ca9c8ea9db8f609392
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Tue Aug 19 13:15:46 2014 -0700
rgw: subuser creation fixes
Fixes: #8587
There were a couple of issues, one when trying to identify whether swift
user exists, we weren't using the correct swift id. The second problem
is that we relied on the gen_access flag in the swift case, where it
doesn't really need to apply.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 1441ffe8103f03c6b2f625f37adbb2e1cfec66bb)
commit 8db2f0969e1715f57088c311a33e7e3499933afb
Merge: 9a15592 2a5d7f0
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 31 08:35:50 2014 -0700
Merge pull request #2847 from dachary/wip-9752-past-intervals-firefly
osd: past_interval display bug on acting
commit 2a5d7f08303eb8b1687c5b58426443b3d40e415c
Author: Loic Dachary <loic-201408@dachary.org>
Date: Fri Oct 31 00:49:21 2014 +0100
osd: past_interval display bug on acting
The acting array was incorrectly including the primary and up_primary.
http://tracker.ceph.com/issues/9752 Fixes: #9752
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit c5f8d6eded52da451fdd1d807bd4700221e4c41c)
commit 9a15592eae6739183049f8376392c5577145871e
Merge: 6fd8879 c20a242
Author: Yan, Zheng <ukernel@gmail.com>
Date: Thu Oct 30 17:01:24 2014 -0700
Merge pull request #2840 from ceph/firefly-9869
Backport "client: cast m->get_client_tid() to compare to 16-bit Inode::flushing_cap_tid"
commit c20a2421f4a5b33407d7f31806dd4587b4e9077c
Author: Greg Farnum <greg@inktank.com>
Date: Wed Oct 22 17:16:31 2014 -0700
client: cast m->get_client_tid() to compare to 16-bit Inode::flushing_cap_tid
m->get_client_tid() is 64 bits (as it should be), but Inode::flushing_cap_tid
is only 16 bits. 16 bits should be plenty to let the cap flush updates
pipeline appropriately, but we need to cast in the proper direction when
comparing these differently-sized versions. So downcast the 64-bit one
to 16 bits.
Fixes: #9869
Backport: giant, firefly, dumpling
Signed-off-by: Greg Farnum <greg@inktank.com>
(cherry picked from commit a5184cf46a6e867287e24aeb731634828467cd98)
commit 4afb54274bb2087da348103d0a7641b3a018d777
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Sep 11 13:46:51 2014 -0700
ReplicatedPG: cancel cb on blacklisted watcher
Fixes: #8315
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 16bd45777166c29c433af3b59254a7169e512d98)
commit 037aa342fa277351fc605da41489a6ceba81ab05
Author: Samuel Just <sam.just@inktank.com>
Date: Sun Sep 21 10:19:43 2014 -0700
ReplicatedPG::on_removal: clear rollback info
Fixes: #9293
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 544b8c7ffb4af01765b87239f2d7ab88479ee779)
commit 8978c8dc1f153a9052b5d967ceff11d4f08d51ea
Merge: 8401e7f 4be53d5
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Oct 30 13:48:18 2014 -0700
Merge remote-tracking branch 'origin/wip-9574' into wip-sam-firefly-backports
commit 8401e7ffa1768770f451143b3c110d1deae1bd40
Author: Samuel Just <sam.just@inktank.com>
Date: Mon Sep 29 15:01:25 2014 -0700
PG: release backfill reservations if a backfill peer rejects
Also, the full peer will wait until the rejection from the primary
to do a state transition.
Fixes: #9626
Backport: giant, firefly, dumpling
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 624aaf2a4ea9950153a89ff921e2adce683a6f51)
commit 5df09fa1b5a42992f9e41aca09e86db0d03d9fbd
Merge: a0937ef a1aa06b
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Oct 30 13:47:22 2014 -0700
Merge remote-tracking branch 'origin/wip-9113' into wip-sam-firefly-backports
commit a0937ef214b9221e7a5e69e7b0f0697471d56293
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 12 10:05:51 2014 -0700
osd/osd_types: consider CRUSH_ITEM_NONE in check_new_interval() min_size check
Fixes: #9718
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d947050c82a511f91c98e1c76e48ffa9e187eee7)
Conflicts:
src/osd/osd_types.cc
commit 35e8e6b2c1733cd421bf6c3916553eea3786e76e
Author: Samuel Just <sam.just@inktank.com>
Date: Mon Oct 20 14:10:58 2014 -0700
PG:: reset_interval_flush and in set_last_peering_reset
If we have a change in the prior set, but not in the up/acting set, we go back
through Reset in order to reset peering state. Previously, we would reset
last_peering_reset in the Reset constructor. This did not, however, reset the
flush_interval, which caused the eventual flush event to be ignored and the
peering messages to not be sent.
Instead, we will always reset_interval_flush if we are actually changing the
last_peering_reset value.
Fixes: #9821
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit d9ff3a6b789c5b9c77aefa3751bd808f5d7b8ca7)
commit 3f35db42977704a12ac4b5bcad6261aaf6b6a88d
Author: Samuel Just <sam.just@inktank.com>
Date: Thu Oct 23 09:11:28 2014 -0700
ReplicatedPG: writeout hit_set object with correct prior_version
Fixes: #9875
Backport: giant, firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 1a3ad307f1a4c0a956d6fd31d13f01ffe411a09d)
commit 6fd88792e77cdc7ad33ff0acf9b3189a7c525430
Merge: 0975ec9 afe6bd8
Author: Sage Weil <sage@redhat.com>
Date: Sun Oct 26 20:37:52 2014 -0700
Merge pull request #2717 from dachary/wip-9747-ceph-spec-firefly
rpm: 95-ceph-osd-alt.rules is not needed for centos7 / rhel7 (firefly)
commit b9450b532ab7ad23ec6e2c22ed7cf55e6e1cc4c0
Author: Xiaoxi Chen <xiaoxi.chen@intel.com>
Date: Wed Aug 20 15:35:44 2014 +0800
CrushWrapper: pick a ruleset same as rule_id
Originally in the add_simple_ruleset funtion, the ruleset_id
is not reused but rule_id is reused. So after some add/remove
against rules, the newly created rule likely to have
ruleset!=rule_id.
We dont want this happen because we are trying to hold the constraint
that ruleset == rule_id.
Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com>
(cherry picked from commit 78e84f34da83abf5a62ae97bb84ab70774b164a6)
Conflicts:
src/test/erasure-code/TestErasureCodeIsa.cc
Fixes: #9675
commit 3e17a0872a3864cb6aee46959afd955ef0cbafeb
Author: Ma Jianpeng <jianpeng.ma@intel.com>
Date: Mon Jul 21 15:08:55 2014 +0800
os/FileJournal: When dump journal, using correctly seq avoid misjudging joural corrupt.
In func FileJournal::dump, it always using seq=0 as last-seq and it can
misjudge the journal corrupt.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
(cherry picked from commit 5f65b4db6d1dad7c2c5a09eab42af63a82ea9e9b)
commit 350da8c98bc1e90cd392992aba290c7478280d88
Author: Loic Dachary <loic-201408@dachary.org>
Date: Fri Sep 26 01:15:53 2014 +0200
os: io_event.res is the size written
And not an error code to be converted with cpp_strerror()
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit 7827e0035e3350ad2d9230f27a1629545f53af5c)
commit ecff3761f3d15061a2cbf0a595ca249a4c424f4c
Author: Ma Jianpeng <jianpeng.ma@intel.com>
Date: Thu Aug 21 15:10:46 2014 +0800
os/FileJournal: For journal-aio-mode, don't use aio when closing journal.
For jouranl-aio-mode when closing journal, the write_finish_thread_entry may exit before
write_thread_entry. This cause no one wait last aios to complete.
On some platform, after that the journal-header on journal corrupted.
To avoid this, when closing jouranl we don't use aio.
Fixes: 9073
Reported-by: Mark Kirkwood <mark.kirkwood@catalyst.net.nz>
Tested-by: Mark Kirkwood <mark.kirkwood@catalyst.net.nz>
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
(cherry picked from commit e870fd09ce846e5642db268c33bbe8e2e17ffef2)
commit dbc33fbab4b35e2ce1e46a881f6714262502c243
Author: Ma Jianpeng <jianpeng.ma@intel.com>
Date: Thu Aug 21 21:07:51 2014 +0800
os/FileJournal: Only using aio then alloc the related resources.
If define HAVE_LIBAIO, it alloc related resouces. But itt don't check whether
using aio mode. Only using aio it alloc the related resources.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
(cherry picked from commit a66a4931d5be9ee26c0983b3154fdbe37261a51c)
commit 3312c6eeca21dcd566df9bdd8de7b3fe33356b57
Author: Ma Jianpeng <jianpeng.ma@intel.com>
Date: Thu Aug 21 15:49:44 2014 +0800
os/FileJournal: Tune the judge logic for read_header.
When reading journal-header, it should firstly check the result of
pread and then do decoce operation.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
(cherry picked from commit c8e2b89cf6bc36a0ff29887b9e76cbbeceef9f8f)
commit b42107584449e1f85cbee97bfa486ebeb310e6a1
Author: Sage Weil <sage@redhat.com>
Date: Tue Aug 19 20:50:13 2014 -0700
os/FileJournal: signal aio_cond even if seq is 0
This can happen if we write a journal but no events.
Reported-by: Somnath Roy <somnath.roy@sandisk.com>
Reported-by: Ma, Jianpeng <jianpeng.ma@intel.com>
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 57778e2c577c1e1bbf9525232720a2994fa36abc)
commit 4a43ea171d84a9f3a13064030eb386fcfbe3bbb8
Author: Ma Jianpeng <jianpeng.ma@intel.com>
Date: Wed Jul 23 10:10:38 2014 -0700
os/FileJournal: Update the journal header when closing journal
When closing journal, it should check must_write_header and update
journal header if must_write_header alreay set.
It can reduce the nosense journal-replay after restarting osd.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
Reviewed-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 5bf472aefb7360a1fe17601b42e551df120badfb)
commit 0975ec9cec1c466f7b15f5173541a7eab02dae18
Author: Sage Weil <sage@redhat.com>
Date: Tue Oct 21 06:53:36 2014 -0700
Revert "os/FileJournal: stop aio completion thread *after* writer thread"
This reverts commit 334631ae4641824b3df49245f36a8fd4b143bf3f.
commit 5a10b95f7968ecac1f2af4abf9fb91347a290544
Merge: cc69c16 d5bac46
Author: Samuel Just <sam.just@inktank.com>
Date: Fri Oct 17 10:47:22 2014 -0700
Merge pull request #2716 from ceph/wip-firefly-9419
Backport fix for bug #9419
commit cc69c16c2108cb231b2434a53e3eef51b597756b
Merge: f9cdaab 334631a
Author: Samuel Just <sam.just@inktank.com>
Date: Fri Oct 17 10:44:30 2014 -0700
Merge pull request #2724 from dachary/wip-9073-journal-aio-mode-firefly
os/FileJournal: stop aio completion thread *after* writer thread
commit f9cdaabe078415d7927e2618030996f2f24be2f1
Merge: 0b4b34a 412c277
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 17 08:20:53 2014 -0700
Merge pull request #2742 from ceph/firefly-unknown-locktype
mds: reply -EOPNOTSUPP for unknown lock type
commit 412c2770c74abea73a94e10df7b83ebe11ac82ee
Author: Yan, Zheng <zyan@redhat.com>
Date: Tue Oct 14 22:02:41 2014 +0800
mds: reply -EOPNOTSUPP for unknown lock type
Signed-off-by: Yan, Zheng <zyan@redhat.com>
(cherry picked from commit 675392335c53ff7879031fb9184e4f35bcc90fe2)
commit 1228658871e53e350bdab3e72cdefd1caf33c291
Author: Sage Weil <sage@redhat.com>
Date: Sun Sep 21 15:56:18 2014 -0700
osd/ReplicatedPG: do not clone or preserve snapdir on cache_evict
If we cache_evict a head in a cache pool, we need to prevent
make_writeable() from cloning the head and finish_ctx() from
preserving the snapdir object.
Fixes: #8629
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ce8eefca13008a9cce3aedd67b11537145e1fd77)
commit 88e6014463e86e48d78ac419226644209f83f2a0
Author: Sage Weil <sage@redhat.com>
Date: Sun Sep 21 15:54:15 2014 -0700
ceph_test_rados_api_tier: add EvictSnap2 test case
Verify an evict doesn't create a snapdir object. Reproduces #8629
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 398c74eacb1ce4e573aef0d24718a5925d90272b)
commit 0b4b34aac497d17a6474c35891aab2bde962524b
Merge: 322958a 0a72235
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 16 06:09:51 2014 -0700
Merge pull request #2734 from ceph/wip-firefly-undump
mds: fix --undump-journal
Reviewed-by: Sage Weil <sage@redhat.com>
commit 0a72235b0556752fadebc3e155ad41b13a0a15e9
Author: John Spray <john.spray@redhat.com>
Date: Thu Oct 16 11:17:40 2014 +0100
mds: fix --undump-journal
This hadn't worked for a long time. This is a fix
for firefly only, as this code was refactored in giant.
Signed-off-by: John Spray <john.spray@redhat.com>
commit 835f8c6f6121f3ebdec3a0d2d5cb1376301dc03a
Author: Joao Eduardo Luis <joao@redhat.com>
Date: Tue Sep 23 14:02:55 2014 +0100
ceph-mon: check fs stats just before preforking
Otherwise statfs may fail if mkfs hasn't been run yet or if the monitor
data directory does not exist. There are checks to account for the mon
data dir not existing and we should wait for them to clear before we go
ahead and check the fs stats.
Signed-off-by: Joao Eduardo Luis <joao@redhat.com>
(cherry picked from commit 7f71c11666b25e91dd612c58b4eda9ac0d4752f8)
Conflicts:
src/ceph_mon.cc
commit 1ddf435464562f70f63cdb0032da3187f34ce853
Author: Joao Eduardo Luis <joao@redhat.com>
Date: Thu Sep 18 16:53:43 2014 +0100
ceph_mon: check available storage space for mon data dir on start
error out if available storage space is below 'mon data avail crit'
Fixes: #9502
Signed-off-by: Joao Eduardo Luis <joao@redhat.com>
(cherry picked from commit 2da1a2914ac7df18ce842b0aac728fffb5bed2b6)
Conflicts:
src/ceph_mon.cc
commit 112317791b744d9890a65adcc13554c85e90f3af
Author: Joao Eduardo Luis <joao@redhat.com>
Date: Thu Sep 18 16:52:34 2014 +0100
mon: DataHealthService: use get_fs_stats() instead
and relieve the DataStats struct from clutter by using
ceph_data_stats_t instead of multiple fields.
Signed-off-by: Joao Eduardo Luis <joao@redhat.com>
(cherry picked from commit 9996d446988768658db751a7843b13cf3d194213)
Conflicts:
src/mon/DataHealthService.cc
commit f0a92d72fd44542619338db7d6da98e147b6a9fc
Author: Joao Eduardo Luis <joao@redhat.com>
Date: Thu Sep 18 16:32:20 2014 +0100
common: util: add get_fs_stats() function
simplifies the task of obtaining available/used disk space, as well as
used available percentage.
Signed-off-by: Joao Eduardo Luis <joao@redhat.com>
(cherry picked from commit 3d74230d1c0fbfa15487e2a90ac60b883476e840)
commit a8fa009fbe5d5d4d9cfa134d5ecd05c92290a8eb
Author: Joao Eduardo Luis <jeclui@gmail.com>
Date: Thu Sep 18 16:25:44 2014 +0100
include/util.h: prevent multiple inclusion of header
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 76eff9503493312cb97e4a2f9236f4dbcbf931df)
commit e296685e8f3f5158238216eefb76482bd6d55134
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 18 14:23:36 2014 -0700
mon: re-bootstrap if we get probed by a mon that is way ahead
During bootstrap we verify that our paxos commits overlap with the other
mons we will form a quorum with. If they do not, we do a sync.
However, it is possible we pass those checks, then fail to join a quorum
before the quorum moves ahead in time such that we no longer overlap.
Currently nothing kicks up back into a probing state to discover we need
to sync... we will just keep trying to call or join an election instead.
Fix this by jumping back to bootstrap if we get a probe that is ahead of
us. Only do this from non probe or sync states as these will be common;
it is only the active and electing states that matter (and probably just
electing!).
Fixes: #9301
Backport: giant, firefly
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c421b55e8e15ef04ca8aeb47f7d090375eaa8573)
commit 0e57767d5fc524939e8968b506ce2fb3f4f80656
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 18 14:11:24 2014 -0700
mon/Paxos: fix off-by-one in last_ vs first_committed check
peon last_committed + 1 == leader first_committed is okay. Note that the
other check (where I clean up whitespace) gets this correct.
Fixes: #9301 (partly)
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d81cd7f86695185dce31df76c33c9a02123f0e4a)
commit 1f4aaf648f4aa6f6056d0e8ce629eeea05c5424d
Author: Sage Weil <sage@redhat.com>
Date: Wed Aug 13 16:17:02 2014 -0700
mon/Paxos: share state and verify contiguity early in collect phase
We verify peons are contiguous and share new paxos states to catch peons
up at the end of the round. Do this each time we (potentially) get new
states via a collect message. This will allow peons to be pulled forward
and remain contiguous when they otherwise would not have been able to.
For example, if
mon.0 (leader) 20..30
mon.1 (peon) 15..25
mon.2 (peon) 28..40
If we got mon.1 first and then mon.2 second, we would store the new txns
and then boot mon.1 out at the end because 15..25 is not contiguous with
28..40. However, with this change, we share 26..30 to mon.1 when we get
the collect, and then 31..40 when we get mon.2's collect, pulling them
both into the final quorum.
It also breaks the 'catch-up' work into smaller pieces, which ought to
smooth out latency a bit.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c54f1e4d66b22bad715ac17e9baa72ab93e48c46)
commit 6c5b9a666fcd94e175a8b9771368b55246957efe
Author: Sage Weil <sage@redhat.com>
Date: Thu Aug 14 16:55:58 2014 -0700
mon/Paxos: verify all new peons are still contiguous at end of round
During the collect phase we verify that each peon has overlapping or
contiguous versions as us (and can therefore be caught up with some
series of transactions). However, we *also* assimilate any new states we
get from those peers, and that may move our own first_committed forward
in time. This means that an early responder might have originally been
contiguous, but a later one moved us forward, and when the round finished
they were not contiguous any more. This leads to a crash on the peon
when they get our first begin message.
For example:
- we have 10..20
- first peon has 5..15
- ok!
- second peon has 18..30
- we apply this state
- we are now 18..30
- we finish the round
- send commit to first peon (empty.. we aren't contiguous)
- send no commit to second peon (we match)
- we send a begin for state 31
- first peon crashes (it's lc is still 15)
Prevent this by checking at the end of the round if we are still
contiguous. If not, bootstrap. This is similar to the check we do above,
but reverse to make sure *we* aren't too far ahead of *them*.
Fixes: #9053
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 3e5ce5f0dcec9bbe9ed4a6b41758ab7802614810)
commit 11d2c9dd4aeb835ca73bfb41fb15b1038547adf6
Author: Sage Weil <sage@redhat.com>
Date: Wed Aug 13 16:01:01 2014 -0700
mon/Paxos: put source mon id in a temp variable
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit bb046ed01ecf58b8c87eeeee2e00a476e6fba467)
commit 322958a5aad82c031b54592b372aa053e8993be4
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 15 12:26:00 2014 -0700
qa/workunits/rbd/import_export.sh: be case insensitive
Stop tripping over this change (from dumpling).
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 5558afa03dbd1b20766b76e9410ef5bc3e73784f)
commit 334631ae4641824b3df49245f36a8fd4b143bf3f
Author: Sage Weil <sage@redhat.com>
Date: Fri Aug 29 19:40:29 2014 -0700
os/FileJournal: stop aio completion thread *after* writer thread
The writer thread may submit a new aio to update the header in its
final moments before shutting down. Do not stop the aio thread until after
that has happened or else we may not wait for those aio completions.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c776a89880fdac270e6334ad8e49fa616d05d0d4)
Conflicts:
src/os/FileJournal.cc
commit 111eec9ff6325a12fcbf066ae08f27919aeae5d8
Merge: cf4e300 6c0127f
Author: Sage Weil <sage@redhat.com>
Date: Tue Oct 14 14:57:42 2014 -0700
Merge remote-tracking branch 'gh/firefly' into firefly-next
commit cf4e30095e8149d1df0f2c9b4c93c9df0779ec84
Author: Xiaoxi Chen <xiaoxi.chen@intel.com>
Date: Tue Aug 5 16:12:22 2014 +0800
mon/OSDMonitor : Use user provided ruleset for replicated pool
When creating a replicated pool, currently ceph ignore the ruleset
name provided by user but use a global default ruleset.
This patch fix this bug, so the rulset specified by
ceph osd pool create replicated
can be properly set.
Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com>
(cherry picked from commit bf9726a294abd32b429170284ac328a592802648)
commit bfd7da10e708a6eefc6d992b2b6337b7f06fd5ed
Author: Loic Dachary <loic@dachary.org>
Date: Tue Jun 3 13:05:19 2014 +0200
documentation: update osd pool create erasure
The properties are replaced with erasure code profiles. Remove the
reference to properties and the documentation of each erasure-code
related property.
Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 8ff4edda73abb920c91e1226a330e3659def1fbe)
commit afe6bd89f8a1588fb67063d1a08a4be8c1ab2ce6
Author: Loic Dachary <loic-201408@dachary.org>
Date: Sat Oct 11 18:20:36 2014 +0200
rpm: 95-ceph-osd-alt.rules is not needed for centos7 / rhel7
The || instead of && had it always installed. That was fixed in EPEL
already.
http://tracker.ceph.com/issues/9747 Fixes: #9747
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit 5ff4a850a0d809b3f25988c6cceb82c35095ef84)
commit d5bac46e06c5420f29a021b294e391b2c6694cbd
Author: David Zafman <dzafman@redhat.com>
Date: Wed Sep 24 16:02:21 2014 -0700
osd: Return EOPNOTSUPP if a set-alloc-hint occurs with OSDs that don't support
Add CEPH_FEATURE_OSD_SET_ALLOC_HINT feature bit
Collect the intersection of all peer feature bits during peering
When handling CEPH_OSD_OP_SETALLOCHINT check that all OSDs support it
by checking for CEPH_FEATURE_OSD_SET_ALLOC_HINT feature bit.
Fixes: #9419
Backport: firefly
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 9b39033f2b2bcdd2be0f6da4dff06023d0f77499)
Conflicts:
src/include/ceph_features.h
src/osd/PG.cc
src/osd/PG.h
src/osd/ReplicatedPG.cc
commit de08802dcf35aea516d013d3d6116aaa7707b923
Author: David Zafman <dzafman@redhat.com>
Date: Fri Sep 19 15:12:55 2014 -0700
osd: Remove unused PG functions queue_notify(), queue_info(), queue_log()
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 70ef4c11cbae669799c30c7592073ad7aa11dcd6)
commit 5b5aba73031e901457ca27cf15600ce1ca90e258
Merge: 345714b a1ae7f4
Author: Gregory Farnum <greg@inktank.com>
Date: Fri Oct 10 06:57:06 2014 -0700
Merge pull request #2691 from ceph/firefly-unused-variable
Firefly unused variable
commit a1ae7f471c809e69d363b9145e70160533bfa48c
Author: Yan, Zheng <zyan@redhat.com>
Date: Fri Oct 10 21:36:39 2014 +0800
mds: Locker: remove unused variable
Signed-off-by: Yan, Zheng <zyan@redhat.com>
commit 345714b6b4d004ad03cc7952dc56c6db87664ee4
Merge: fd20a1d 2afb6fe
Author: Yan, Zheng <ukernel@gmail.com>
Date: Fri Oct 10 09:37:53 2014 +0800
Merge pull request #2681 from ceph/firefly-locker-null
mds: Locker: fix a NULL deref in _update_cap_fields
commit 2afb6febdd8482b8fec5890d79944d656faf1382
Author: Greg Farnum <greg@inktank.com>
Date: Thu Oct 9 15:12:19 2014 -0700
mds: Locker: fix a NULL deref in _update_cap_fields
The MClientCaps* is allowed to be NULL, so we can't deref it unless
the dirty param is non-zero. So don't do the ahead-of-time lookup;
just call it explicitly in the if block.
Signed-off-by: Greg Farnum <greg@inktank.com>
(cherry picked from commit 3cd8a7fb9683577a7d6e934f18c29b7e84415be6)
commit fd20a1d01bde67fb1edc6058e38435af9d5d6abc
Merge: e1bd1b2 86926c6
Author: Loic Dachary <loic-201408@dachary.org>
Date: Wed Oct 8 08:44:46 2014 +0200
Merge pull request #2662 from dachary/wip-9677-ioprio-class-firefly
common: ceph_ioprio_string_to_class always returns -EINVAL
commit 86926c6089d63014dd770b4bb61fc7aca3998542
Author: Loic Dachary <loic-201408@dachary.org>
Date: Tue Oct 7 14:06:38 2014 +0200
common: ceph_ioprio_string_to_class always returns -EINVAL
The l string is always empty because std::transform needs a
pre-allocated string. Replace with the in-place version. Add unit tests.
http://tracker.ceph.com/issues/9677 Fixes: #9677
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit 3535b7aba3df8b54fa5117b8a9c2f52b8f0f118b)
Conflicts:
src/test/Makefile.am
commit 5f2eec5036a2910aca1e8ce2d94444d3ed0477df
Author: Loic Dachary <loic-201408@dachary.org>
Date: Tue Oct 7 14:05:08 2014 +0200
osd: log error if set_ioprio fails to parse class
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit 5088e0d49332d579ba7e33c2c9baee3d5f701a3e)
commit 2796d5151df4dcde324a4d09a83c9a779cece00e
Author: Loic Dachary <loic-201408@dachary.org>
Date: Tue Oct 7 14:03:39 2014 +0200
common: set_ioprio debug message including pid
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit 33339c7754875eb7e513345ee6b26a9b2b4d2707)
commit d5ed6b0587b9999b2fd41377b0426e3b09ef8ab9
Author: Loic Dachary <loic-201408@dachary.org>
Date: Tue Oct 7 14:02:09 2014 +0200
common: do not set ioprio if pid is not set
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit c7e4c0bfe70bf29d3b8fe4df4e4b934853e33d26)
commit e1bd1b2774f4eae5444b5f7b984193cb91a2dd98
Merge: 726c6a1 459dca1
Author: Sage Weil <sage@redhat.com>
Date: Tue Oct 7 09:58:03 2014 -0700
Merge pull request #2632 from ceph/wip-9039-firefly
rgw: copy object data if target bucket is in a different pool
commit 726c6a147a14c00cf12eb6c6561655475282419f
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 6 15:50:51 2014 -0700
debian/control: fix python-ceph -> ceph file move to allow upgrades
This is a backport of 5c6c366d2abe771c581690270c2d176ebb30c571 with the
version numbers changed, to compensate for the change in
fe3434f41cd09433975d7d0f9dbb2fae662e4a1b (backported in
bf1933e5c184476a354664c42fec834e9f59067c).
Tested-by: Tamil Muthamizhan <tamil.muthamizhan@inktank.com>
Signed-off-by: Sage Weil <sage@redhat.com>
commit 884f7c40c4a28d519847d3995c8d98e5837ceaf0
Merge: 31d57c9 c8a8e47
Author: Sage Weil <sage@redhat.com>
Date: Mon Oct 6 07:01:50 2014 -0700
Merge remote-tracking branch 'gh/wip-rpm-epoch-firefly' into firefly
Reviewed-by: Boris Ranto <branto@redhat.com>
commit 31d57c9a28502a4a72f8aa141f7ed63ffe1e0192
Merge: 9a3bac0 548be0b
Author: Loic Dachary <loic-201408@dachary.org>
Date: Mon Oct 6 09:50:33 2014 +0200
Merge pull request #2643 from johnugeorge/wip-9492-crush-firefly
Crush: Backporting fixes for #9492 to firefly
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
commit 548be0b2aea18ed3196ef8f0ab5f58a66e3a9af4
Author: Johnu George <johnugeo@cisco.com>
Date: Mon Sep 29 10:07:44 2014 -0700
Crush: Ensuring at most num-rep osds are selected
Crush temporary buffers are allocated as per replica size configured
by the user.When there are more final osds (to be selected as per
rule) than the replicas, buffer overlaps and it causes crash.Now, it
ensures that at most num-rep osds are selected even if more number of
osds are allowed by indep rule. The fix for firstn rules is already
merged as part of bug #9492. Required test files are added.
Fixes: #9492
Signed-off-by: Johnu George johnugeo@cisco.com
(cherry picked from commit 234b066ba04976783d15ff2abc3e81b6cc06fb10)
commit e30c570ce691a994898b4a933c57e7ae014cdc30
Author: Johnu George <johnugeo@cisco.com>
Date: Wed Sep 24 09:32:50 2014 -0700
Crush: Ensuring at most num-rep osds are selected
Crush temporary buffers are allocated as per replica size configured
by the user.When there are more final osds (to be selected as per
rule) than the replicas, buffer overlaps and it causes crash.Now, it
ensures that at most num-rep osds are selected even if more number of
osds are allowed by the rule.
Fixes: #9492
Signed-off-by: Johnu George <johnugeo@cisco.com>
(cherry picked from commit 6b4d1aa99718e3b367496326c1e64551330fabc0)
commit 9a3bac0c1a7a42cdf7ba846c9ad0a3ae0f15b4bb
Merge: 0b978fb 5a5e7e7
Author: Sage Weil <sage@redhat.com>
Date: Fri Oct 3 11:58:41 2014 -0700
Merge pull request #2634 from dachary/wip-9653-ceph-disk-bootstrap-osd-firefly
ceph-disk: bootstrap-osd keyring ignores --statedir (firefly)
commit c8a8e4763a55dec44836bc679254ee7dcc448567
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 1 12:33:38 2014 -0700
ceph.spec: fix typo
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit da9ae5c92ecb4059e0ec6be5ce03af46430e76a5)
commit b890c1e4706d7cfef7ed24c9df65b439b4f7ff1d
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 1 06:02:02 2014 -0700
ceph.spec.: add epoch
This is done in fedora packaging. Do it here too so that you can move
between upstream packages (from ceph.com) and fedora and other derivatives
will builds.
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 83888362089346e473d6fd6e1d366b826d7bd739)
Conflicts:
ceph.spec.in
commit d01db8080d97bfae34dcee3d153bf10e6f5327dd
Author: Sage Weil <sage@inktank.com>
Date: Tue May 20 13:41:35 2014 -0700
ceph.spec.in: remove BuildRoot
Deprecated
Fixes: #8143
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from commit 401319a1527dd9cb5398916105d31e7ec065763d)
commit 3c2b5c440863df548afc2bd8aa5440f15a44ac02
Author: Dan Mick <dan.mick@inktank.com>
Date: Tue Aug 12 16:31:22 2014 -0700
ceph.spec.in: tests for rhel or centos need to not include _version
rhel_version and centos_version are apparently the OpenSUSE Build
names; the native macros are just "rhel" and "centos" (and contain
a version number, should it be necessary).
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit 7474f720c2418cf3d52b755f2b60c524e413570a)
commit c82c29d6174022be45929fe9ba8a84993eef974a
Author: Dan Mick <dan.mick@inktank.com>
Date: Tue Aug 12 14:46:52 2014 -0700
ceph.spec.in: Add a small comment on the empty %files section
as suggested by Dan Mick.
Signed-off-by: Erik Logtenberg <erik@logtenberg.eu>
(cherry picked from commit e37b262c7928934530c5bb09fe56f83eb61f4244)
commit a4f748aa906fc65b14f65515721bc3a815c18fb8
Author: Dan Mick <dan.mick@inktank.com>
Date: Tue Aug 12 14:39:18 2014 -0700
ceph.spec.in: Obsolete all older versions.
Now this changeset can be used on all current ceph releases that already
have the package split.
Signed-off-by: Erik Logtenberg <erik@logtenberg.eu>
(cherry picked from commit 875a99e25f0ad2cb47149a3b5a28b4771a09125c)
commit 74c7f3caec1cc7a5da89ef33de36a8b59249cfcd
Author: Dan Mick <dan.mick@inktank.com>
Date: Tue Aug 12 14:09:43 2014 -0700
ceph.spec.in: No version on ceph-libs Obsoletes.
If we are installing with the new package structure we don't ever want the
new package to co-exist with the old one; this includes the mistakenly-
released v0.81 on Fedora, which should be removed in favor of this
version.
Signed-off-by: Sandon Van Ness <sandon@inktank.com>
Reviewed-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit 8f95daf66b5fdb2a8141988480f984c1249599c5)
commit 561261b6efb7ea442686bb8fce387c4de2482067
Author: Sandon Van Ness <sandon@inktank.com>
Date: Fri Aug 8 18:01:30 2014 -0700
ceph.spec.in: Obselete ceph-libcephfs (not libcephfs)
I am guessing that because it was a sub-package libcephfs was mistakenly
used instead of ceph-libcephfs.
Signed-off-by: Sandon Van Ness <sandon@inktank.com>
(cherry picked from commit 75985024bd30ca6fbe4c61aa7f7cbe5306c9a988)
commit 107bfd9ee7dbf360561187b9e0946964d40b9b1c
Author: Erik Logtenberg <erik@logtenberg.eu>
Date: Fri Aug 1 14:20:18 2014 +0200
ceph.spec.in: We need those nice recent changes for rhel7 in Fedora too.
Signed-off-by: Erik Logtenberg <erik@logtenberg.eu>
(cherry picked from commit 00877ae502ac52613bcd5c5c834d72787d668dca)
commit 7946c5e5de8d6cd25d20beee15f3489113e51539
Author: Dan Mick <dan.mick@inktank.com>
Date: Wed Aug 27 12:56:43 2014 -0700
Move fedora patch file (used by ceph.spec.in) to rpm/ subdir
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit 06b92cee621cbe33a6f17e8c64169db4453a5160)
commit cb2ae9afa611175226efb5544f7d2aa705d55ece
Author: Erik Logtenberg <erik@logtenberg.eu>
Date: Fri Aug 1 00:13:50 2014 +0200
ceph.spec.in, init-ceph.in: Don't autostart ceph service on Fedora.
This patch is taken from the current Fedora package and makes the upstream
ceph.spec compliant with Fedora policy. The goal is to be fully compliant
upstream so that we can replace current Fedora package with upstream
package to fix many bugs in Fedora.
Addition from Dan Mick <dan.mick@inktank.com>:
Do this for RHEL and Centos as well, since they surely will benefit
from the same policy. Note: this requires changes to
autobuild-ceph and ceph-build scripts, which currently copy
only the dist tarball to the rpmbuild/SOURCES dir.
Signed-off-by: Erik Logtenberg <erik@logtenberg.eu>
Signed-off-by: Dan Mick <dan.mick@inktank.com>:
(cherry picked from commit 461523b06cdf93e32f1d8b354ac3799e73162d33)
commit 2b11376f1ee8925ab16065ebda912b11d3d7be59
Author: Erik Logtenberg <erik@logtenberg.eu>
Date: Thu Jul 31 23:54:03 2014 +0200
ceph.spec.in: Add obsoletes for libcephfs
This fixes a bug for Fedora:
https://bugzilla.redhat.com/show_bug.cgi?id=1116614
Signed-off-by: Erik Logtenberg <erik@logtenberg.eu>
(cherry picked from commit e9da2d8f2142771f206ef67f19e7f194855275d0)
commit eefc62e3b85f402a7e1ae31c272c8a432d979379
Author: Erik Logtenberg <erik@logtenberg.eu>
Date: Thu Jul 31 23:49:56 2014 +0200
ceph.spec.in: add ceph-libs-compat
Added a ceph-libs-compat package in accordance with Fedora packaging
guidelines [1], to handle the recent package split more gracefully.
In Fedora this is necessary because there are already other packages
depending on ceph-libs, that need to be adjusted to depend on the new
split packages instead. In the mean time, ceph-libs-compat prevents
breakage.
[1] http://fedoraproject.org/wiki/Upgrade_paths_%E2%80%94_renaming_or_splitting_packages
Signed-off-by: Erik Logtenberg <erik@logtenberg.eu>
(cherry picked from commit 6c264f2204cbd54d90b02101e40ac9aa5aa72d7c)
Conflicts:
ceph.spec.in
commit 0b978fb15a1307644aba3119419bb7386f98ee04
Author: Sage Weil <sage@redhat.com>
Date: Sun Aug 10 14:41:19 2014 -0700
mon/Paxos: add perfcounters for most paxos operations
I'm focusing primarily on the ones that result in IO here.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit b09b8563d35dda23faed43afef2a983e93a879c5)
commit 74aa7afc719e517dbed300f802c1bc2dafe43ee0
Author: Sage Weil <sage@redhat.com>
Date: Sun Aug 10 14:00:11 2014 -0700
mon/MonitorDBStore: add get_{keys,bytes}() accounting to Transaction
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit fd421b26748e872ddf8e0f068dda2106853edff1)
commit 5a5e7e7bcedbccbe4ae8aab159af6d8615eb3887
Author: Loic Dachary <loic-201408@dachary.org>
Date: Fri Oct 3 14:08:57 2014 +0200
ceph-disk: bootstrap-osd keyring ignores --statedir
The STATEDIR variable is used to initialize the bootstrap-osd keyring
before it gets a chance to be overriden by --statedir. Replace it with
{statedir} so that it can be substituted after all options have been
parsed.
http://tracker.ceph.com/issues/9653 Fixes: #9653
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from commit fa0bd06b4657e5b84e237b76033ac3d3478b6a1f)
commit 459dca1613a14cfad8d3afd7e3c783d825573a42
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Tue Aug 12 13:36:11 2014 -0700
rgw: copy object data if target bucket is in a different pool
Fixes: #9039
Backport: firefly
The new manifest does not provide a way to put the head and the tail in
separate pools. In any case, if an object is copied between buckets in
different pools, we may really just want the object to be copied, rather
than reference counted.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 5d3a7e595f47455896304bf358e5251915d0f16f)
commit 711a7e6f81983ff2091caa0f232af914a04a041c
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Sep 15 00:53:50 2014 -0400
rbd: ObjectCacher reads can hang when reading sparse files
The pending read list was not properly flushed when empty objects
were read from a space file.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit cdb7675a21c9107e3596c90c2b1598def3c6899f)
commit b7784dc1baa47560a733fe9dcd2acec51bc93165
Author: Jason Dillaman <dillaman@redhat.com>
Date: Sat Sep 6 22:59:40 2014 -0400
Enforce cache size on read requests
In-flight cache reads were not previously counted against
new cache read requests, which could result in very large
cache usage. This effect is most noticeable when writing
small chunks to a cloned image since each write requires
a full object read from the parent.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 4fc9fffc494abedac0a9b1ce44706343f18466f1)
commit ac4fca065a12f3b00a47bf5ec37983696255174b
Author: Alexandre Marangone <alexandre.marangone@inktank.com>
Date: Fri Sep 5 10:36:24 2014 -0700
rgw: add .log to default log path
Fixes: #9353
Signed-off-by: Alexandre Marangone <alexandre.marangone@inktank.com>
(cherry picked from commit 46732420897a2619059050044f4980a4737df43e)
commit f03ae7b00e5694f1670493396a1cee195fcc6b35
Merge: 78c3ef9 ee02cfd
Author: Yehuda Sadeh <yehuda@inktank.com>
Date: Thu Oct 2 15:28:40 2014 -0700
Merge pull request #2565 from ceph/wip-rgw-firefly-backports
Wip rgw firefly backports
commit 78c3ef90604ca117255cefe232771a7564fed8b1
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 25 13:16:52 2014 -0700
osdc/Objecter: only post_rx_buffer if no op timeout
If we post an rx buffer and there is a timeout, the revocation can happen
while the reader has consumed the buffers but before it has decoded and
constructed the message. In particular, we calculate a crc32c over the
data portion of the message after we've taken the buffers and dropped the
lock.
Instead of fixing this race (for example, by reverifying rx_buffers under
the lock while calculating the crc.. bleh), just skip the rx buffer
optimization entirely when a timeout is present.
Note that this doesn't cover the op_cancel() paths, but none of those users
provide static buffers to read into.
Fixes: #9582
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
backport of 126d0b30e990519b8f845f99ba893fdcd56de447
commit a261b4952056aab7b067453930342960bbe55089
Author: Sage Weil <sage@redhat.com>
Date: Mon Sep 29 14:28:32 2014 -0700
debian: move ceph_rest_api.py into ceph
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit fe3434f41cd09433975d7d0f9dbb2fae662e4a1b)
commit eb0f6e347969b40c0655d3165a6c4531c6b595a3
Author: Sage Weil <sage@redhat.com>
Date: Mon Sep 29 14:24:01 2014 -0700
ceph.spec.in: move ceph_rest_api.py into ceph
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 8cda623e0ba34a48a70e9ea988d619b15605c4fd)
commit c4188e31f7bc8f3c337e637cd99c41d5ee4b6787
Author: Sage Weil <sage@redhat.com>
Date: Mon Sep 29 13:44:03 2014 -0700
ceph.spec: fix python-flask dependency
This is needed by ceph-rest-api, which is in ceph.rpm; it's not related to
python-ceph (except that ceph-rest-api happens to require that too).
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit b2416240b88b2e067dfc79a2723335f1584562d0)
commit bf1933e5c184476a354664c42fec834e9f59067c
Author: Sage Weil <sage@redhat.com>
Date: Mon Sep 29 13:40:18 2014 -0700
debian: python-flask is needed by ceph, not python-ceph
It's used by ceph-rest-api which is in the 'ceph' (server) package.
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit e42424e777e4f7d8b03650482253734c1fa8709d)
Conflicts:
debian/control
commit 94a7fbaa11c51db294dce0dc1df728f69aef5bf8
Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Date: Fri Sep 19 12:25:07 2014 +0200
rgw_main.cc: add missing virtual destructor for RGWRequest
CID 1160858 (#1 of 1): Non-virtual destructor (VIRTUAL_DTOR)
nonvirtual_dtor: Class RGWLoadGenRequest has a destructor
and a pointer to it is upcast to class RGWRequest which doesn't
have a virtual destructor.
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
(cherry picked from commit b82ceda7775ff85943d9143b73789eb37b09bfa9)
commit 9fee8de25ab5c155cd6a3d32a71e45630a5ded15
Author: Greg Farnum <greg@inktank.com>
Date: Mon Sep 29 16:10:36 2014 -0700
Locker: accept ctime updates from clients without dirty write caps
The ctime changes any time the inode does. That can happen even without
the file itself having changed, so we'd better accept the update whenever
the auth caps have dirtied, without worrying about the file caps!
Fixes: #9514
Backport: firefly
Signed-off-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 0ea20a668cf859881c49b33d1b6db4e636eda18a)
commit 461ece5e9fb1d4994a6214a3b6bdae136773629d
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 1 18:01:51 2014 -0700
doc/release-notes: fix attributions for 8702 fix
Oops!
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 188370a94353e29fcb8981699022803e23f3fedd)
commit 917529a78e5046f621df5c48fe5d50d2f7e56560
Author: Sage Weil <sage@redhat.com>
Date: Wed Oct 1 17:48:12 2014 -0700
doc/release-notes: v0.80.6
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c0dc3a56974a469b61523b67cc032cc5726a3a5f)
Conflicts:
doc/release-notes.rst
commit 060a5b1422fcdfb8e84636579a2f0c2f1ec14300
Author: Adam Crume <adamcrume@gmail.com>
Date: Thu Sep 18 16:57:27 2014 -0700
common: Add cctid meta variable
Fixes: #6228
Signed-off-by: Adam Crume <adamcrume@gmail.com>
(cherry picked from commit bb45621cb117131707a85154292a3b3cdd1c662a)
commit a1aa06b7fb30e509193e1b1bb7355b5f21aedc82
Author: Samuel Just <sam.just@inktank.com>
Date: Wed Oct 1 14:30:59 2014 -0700
ReplicatedPG: dump snap_trimq on pg query
Signed-off-by: Samuel Just <sam.just@inktank.com>
commit 34f38b68d89baf1dcbb4571d4f4d3076dc354538
Author: Samuel Just <sam.just@inktank.com>
Date: Mon Sep 29 16:26:54 2014 -0700
ReplicatedPG: do not queue the snap trimmer constantly
Previously, we continuously requeued the snap trimmer while in
TrimmingObjects. This is not a good idea now that we try to
limit the number of snap trimming repops in flight and requeue
the snap trimmer directly as those repops complete.
Fixes: #9113
Backport: giant, dumpling, firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
commit b29bf00f68cf133151c98db06e9498b3e8be22ed
Author: Samuel Just <sam.just@inktank.com>
Date: Wed Sep 24 13:55:47 2014 -0700
ReplicatedPG: clean out completed trimmed objects as we go
Also, explicitely maintain a max number of concurrently trimming
objects.
Fixes: 9113
Backport: dumpling, firefly, giant
Signed-off-by: Samuel Just <sam.just@inktank.com>
commit ee02cfd23facb3404fc377f643b213c2f498474d
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Thu Sep 18 20:53:10 2014 -0700
rgw: calculate hash after writing data
Since data is written asynchronously, we should do the hash calculation
while it's pending.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 5bb94ede19a50543a02a8019ed6c9680b3852d4e)
commit 216730221575d88a72b06ed3d71c9a54cffc5719
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Thu Sep 18 20:51:02 2014 -0700
crypto: don't hash zero sized buffer
libnss returns an error and we assert in that case.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 7b137246b49a9f0b4d8b8d5cebfa78cc1ebd14e7)
commit dab7a4f6ad054b53cedca76ee329a6395918b1ab
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Fri Sep 12 14:07:44 2014 -0700
rgw: push hash calculater deeper
This might have been the culprit for #9307. Before we were calculating
the hash after the call to processor->handle_data(), however, that
method might have spliced the bufferlist, so we can't be sure that the
pointer that we were holding originally is still invalid. Instead, push
the hash calculation down. Added a new explicit complete_hash() call to
the processor, since when we're at complete() it's too late (we need to
have the hash at that point already).
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit d41c3e858c6f215792c67b8c2a42312cae07ece9)
Conflicts:
src/rgw/rgw_rados.h
commit bd0a91343ce70d71acced753688a502b7e8b552e
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Thu Aug 21 16:30:10 2014 -0700
rgw: separate civetweb log from rgw log
The civetweb log now is independent from the rgw log.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 850242cad749e33e1e6bc008baa75c8ea7eda0c1)
Conflicts:
src/civetweb
src/rgw/rgw_main.cc
commit a777562b780e7ab312f881c38b1db26983a1ac47
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Tue Sep 23 13:40:39 2014 -0700
civetweb: update submodule
Update submodule to include multiple fixes.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
commit 1bdcc079d79d8211b44e2a46511cd2240f71744b
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Fri Aug 22 15:12:16 2014 -0700
rgw: convert header field underscores into dashes
Fixes: 9206
Backport: firefly
Certain web servers filter out underscores in the header field name.
Convert them into dashes.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 11acb7097ce21c6218dd48d0c21e0e04a361eb9a)
commit b8fa2ed60b6cce51701df972dbb6f5e02e0d84ba
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Mon Aug 25 10:38:42 2014 -0700
rgw: fix test to identify whether object has tail
Fixes: #9226
Reported-by: Sylvain Munaut <s.munaut@whatever-company.com>
Backport: firefly
We need to identify whether an object is just composed of a head, or
also has a tail. Test for pre-firefly objects ("explicit objs") was
broken as it was just looking at the number of explicit objs in the
manifest. However, this is insufficient, as we might have empty head,
and in this case it wouldn't appear, so we need to check whether the
sole object is actually pointing at the head.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 751b3e26532932a42ca34f9c062a0a3e29a58cff)
commit 6fee71154d838868807fd9824d829c8250d9d2eb
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Wed Aug 27 17:44:18 2014 -0700
rgw: don't try to authenticate a CORS preflight request
Fixes: #8718
Backport: firefly
CORS preflight requests don't need to be authenticated. Treat them as
coming from anonymous user.
Reported-by: Robert Hubbard <bobby.hubbard@garmin.com>
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 848fcf7871e07fc689bdcd18943ace36b2f4906e)
commit c75a79cbac30cd14d37d89217113824d98693572
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Thu Aug 14 13:35:12 2014 -0700
rgw: fix compilation
RGWRadosPutObj couldn't refer to the ceph context.
Reviewed-by: Sage Weil <sage@redhat.com>
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 6a555434ee3edaf742ee7e5910bcba8dd0de46dd)
commit b0d08aab837808f18708a4f8ced0503c0fce2fec
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Tue Aug 12 11:17:47 2014 -0700
rgw: call throttle_data() even if renew_state() failed
Otherwise we're going to leak the aio callback handle.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 7998c322179dd69a1250937321c3c2bb023e0e57)
commit a953b313f1e2f884be6ee2ce356780f4f70849dd
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Wed Jul 30 21:32:48 2014 -0700
rgw: disable civetweb url decoding
Fixes: #8621
We want to have the raw request uri, as we do the decoding ourselves.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit ffac52b316e7022796d44ae58804d9c20b9c3df9)
commit ba5357714a19b8af989fef1c75ef775837c6a9d6
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Tue Aug 12 14:23:46 2014 -0700
rgw: copy_obj_data() uses atomic processor
Fixes: #9089
copy_obj_data was not using the current object write infrastructure,
which means that the end objects weren't striped.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 800eff24824c0083b8e2441fc34e0bdca5da36dc)
commit d73dbc3a39117eddaaabb2c25d9238cd7c51711b
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Thu Aug 21 21:53:38 2014 -0700
rgw: clear bufferlist if write_data() successful
Fixes: #9201
Backport: firefly
We sometimes need to call RGWPutObjProcessor::handle_data() again,
so that we send the pending data. However, we failed to clear the buffer
that was already sent, thus it was resent. This triggers when using non
default pool alignments.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 9181114d6f6062c55ee4b351fc3495345e545c36)
commit 4be53d5eebbc294878ba38050d841359b6c8e19e
Author: Samuel Just <sam.just@inktank.com>
Date: Tue Sep 23 12:16:55 2014 -0700
PG: check full ratio again post-reservation
Otherwise, we might queue 30 pgs for backfill at 0.80 fullness
and then never check again filling the osd after pg 11.
Fixes: #9574
Backport: dumpling, firefly, giant
Signed-off-by: Samuel Just <sam.just@inktank.com>