Commit Graph

82470 Commits

Author SHA1 Message Date
Mike Christie
4ca0d508bb doc: add daemon-reload for rbd-target-api setup
Signed-off-by: Mike Christie <mchristi@redhat.com>
2018-01-16 12:48:51 -06:00
Mike Christie
f45aafed1c doc: add gwcli manual install instructions
Signed-off-by: Mike Christie <mchristi@redhat.com>
2018-01-16 12:48:41 -06:00
Yuri Weinstein
9c2278bbbc
Merge pull request #19613 from qrGitHub/wip-rgw-optimize-bucketLCPrepare
rgw: optimize function bucket_lc_prepare

Reviewed-by: Casey Bodley <cbodley@redhat.com>
2018-01-16 09:06:21 -08:00
Yuri Weinstein
afb35bafec
Merge pull request #19878 from cbodley/wip-rgw-get-omap-keys-cr
rgw: fix handling of ENOENT in RGWRadosGetOmapKeysCR

Reviewed-by: Adam Emerson <aemerson@redhat.com>
2018-01-16 09:05:28 -08:00
Yuri Weinstein
36f1dc0f04
Merge pull request #19897 from cbodley/wip-rgw-put-system-obj
rgw: rgw_put_system_obj takes bufferlist

Reviewed-by: Matt Benjamin <mbenjami@redhat.com>
Reviewed-by: Orit Wasserman <owasserm@redhat.com>
2018-01-16 09:04:36 -08:00
Andrew Schoen
21f459437b ceph-volume: docs for --crush-device-class
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2018-01-16 09:20:34 -06:00
Andrew Schoen
3404d8bba5 ceph-volume: adds crush_device_class to json metadata for 'osd new'
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2018-01-16 09:20:34 -06:00
Andrew Schoen
36a388d72a ceph-volume: adds crush_device_class to the functional tests
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2018-01-16 09:20:33 -06:00
Andrew Schoen
11da218563 ceph-volume: add crush_device_class to lvm metadata
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2018-01-16 09:20:33 -06:00
Andrew Schoen
bf468bc373 ceph-volume: adds a --crush-device-class flag to lvm create and prepare
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2018-01-16 09:20:33 -06:00
Kefu Chai
31f9af83b3
Merge pull request #19945 from tchaikov/wip-deprecated-in-cxx17
librbd,librados: do not include stdbool.h in C++ headers

Reviewed-by: Adam C. Emerson <aemerson@redhat.com>
2018-01-16 22:16:43 +08:00
Kefu Chai
cff35427ef
Merge pull request #19261 from dillaman/wip-22286
mgr: disconnect unregistered service daemon when report received

Reviewed-By: Kefu Chai <kchai@redhat.com>
2018-01-16 22:15:47 +08:00
Sage Weil
5f9911bc4a
Merge pull request #19942 from shinobu-x/wip-sk-minmax-ms_async_op_threads
common: Add min/max of ms_async_op_threads

Reviewed-by: Sage Weil <sage@redhat.com>
2018-01-16 07:33:22 -06:00
Sage Weil
fdc3f5c24c
Merge pull request #19397 from xxhdx1985126/master
mon: update PaxosService::cached_first_committed in PaxosService::maybe_trim()

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
2018-01-16 07:33:09 -06:00
Sage Weil
bf16f59887 osd: piecewise scrub
Perform scrub in stages, with each unit of work requeuing an item in the
work queue.

Signed-off-by: Sage Weil <sage@redhat.com>
2018-01-16 06:54:06 -06:00
runsisi
08a4de142f cmake: do not find bzip2/lz4 for rocksdb
RocksDB has compression disabled by default[1] and it's never been enabled[2],
so do not link these compression libraries implicitly

[1] https://github.com/ceph/rocksdb/blob/master/CMakeLists.txt#L76
[2] https://github.com/ceph/ceph/blob/master/cmake/modules/BuildRocksDB.cmake

Signed-off-by: runsisi <runsisi@zte.com.cn>
2018-01-16 19:44:44 +08:00
Piotr Dałek
0230fe6733 osd_types.cc: don't store 32 least siognificant bits of state twice
a25221e551 introduced a new, 64bit
pg state field, retaining encoding and decoding of least significant
32 bits for compatibility reasons.
This commit also adds encoding of entire 64bit state field, meaning
that the least significant bits are encoded and decoded twice.
Fix this by encoding each half of 64bit field separately.

Signed-off-by: Piotr Dałek <piotr.dalek@corp.ovh.com>
2018-01-16 11:04:24 +01:00
Piotr Dałek
ae7472fb07 osd_types.cc: reorder encoding/decoding of fields in pg_stat_t
Fields state, purged_snaps and snaptrimq_len are new to Mimic.
Reorder them in a way that newest field (snaptrimq_len) is before
two others and uses other encoding version, so the pull request
https://github.com/ceph/ceph/pull/19520 can be backported to Luminous
without breaking Luminous -> Mimic upgrade.
This also changes encoding/decoding version back to 24 as both state
and purged snaps were added post-Luminous and pre-Mimic, so we can
push it into a single struct version and keep snaptrimq_len into other
version.

Signed-off-by: Piotr Dałek <piotr.dalek@corp.ovh.com>
2018-01-16 10:28:58 +01:00
Mike Christie
f80f083d0a doc: separate gwcli install from post install setup steps
Signed-off-by: Mike Christie <mchristi@redhat.com>
2018-01-15 13:02:24 -06:00
Mike Christie
3cc367dc6e doc: add note to clarify trusted_ip_list use
Signed-off-by: Mike Christie <mchristi@redhat.com>
2018-01-15 13:02:20 -06:00
Mike Christie
3806f9750d doc: update esx instructions
Signed-off-by: Mike Christie <mchristi@redhat.com>
2018-01-15 13:02:13 -06:00
Jason Dillaman
aab2defaca
Merge pull request #19822 from trociny/wip-ggate-ceph-globals
rbd-ggate: fix parsing ceph global options

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Reviewed-by: Willem Jan Withagen <wjw@digiware.nl>
2018-01-15 08:51:34 -05:00
Kefu Chai
0c38eb87e2
Merge pull request #19673 from zouaiguo/wip_erase_code
mon/OSDMonitor.cc : set erasure-code-profile to "" when create replicated pools.

Reviewed-by: Joao Eduardo Luis <joao@suse.de>
Reviewed-by: Kefu Chai <kchai@redhat.com>
2018-01-15 18:34:34 +08:00
Nathan Cutler
2049155ea5 doc: mention rcceph drop in PendingReleaseNotes
Signed-off-by: Nathan Cutler <ncutler@suse.com>
2018-01-15 11:24:05 +01:00
Nathan Cutler
a24afcdcd3 build/ops: rpm: rip out rcceph script
"rcceph" is a SysVinit-style command-line interface for stopping, starting,
enabling, etc. all ceph-osd and ceph-mon systemd units on a machine, in one go.

Since the same functionality is provided by ceph-{osd,mon}.target, the script
is obsolete. It is also unmaintained. Judging from the absence of recent
mentions of the script online, I guess it is no longer used.

Leaving dead code in the tree can cause confusion, especially when the code is
packaged and shipped to customers. Therefore I propose to rip it out.

Signed-off-by: Nathan Cutler <ncutler@suse.com>
2018-01-15 11:22:07 +01:00
huanwen ren
771ecd55f6 common/time: add time.h for Alpine build
We create the Alpine apk package to view "https://github.com/ceph/ceph/blob/master/README.alpine.md",
at compile time found the following error:
"/src/common/ceph_time.h:209:4: error: invalid use of incomplete type 'struct ceph :: time_detail :: timeval'.

Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
2018-01-15 14:23:55 +08:00
Kefu Chai
9a9a00cb3d
Merge pull request #19896 from tchaikov/wip-vive-cxx-17
msg/simple: s/ceph::size/std::size/

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Adam C. Emerson <aemerson@redhat.com>
2018-01-15 11:27:28 +08:00
Kefu Chai
1f21f2f3a3
Merge pull request #19870 from liewegas/wip-strip-ceph-base
debian/rules: strip ceph-base libraries

Reviewed-By: Kefu Chai <kchai@redhat.com>
2018-01-15 11:24:08 +08:00
Kefu Chai
17dcd13ee5
Merge pull request #18755 from aclamk/wip-reverted-approx-size
osd/PG: revert approx size

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Reviewed-by: Kefu Chai <kchai@redhat.com>
2018-01-15 11:14:45 +08:00
Kefu Chai
7aba57b9b4
Merge pull request #18191 from hjwsm1989/osd-mark-down
qa/standalone/osd/osd-mark-down: create pool to get updated osdmap faster

Reviewed-by: Kefu Chai <kchai@redhat.com>
2018-01-15 11:09:02 +08:00
Kefu Chai
35a57fab84
Merge pull request #19877 from adamemerson/wip-dencish
denc cleanups and other fixes

Reviewed-by: Willem Jan Withagen <wjw@digiware.nl>
Reviewed-by: Kefu Chai <kchai@redhat.com>
2018-01-15 11:06:12 +08:00
David Zafman
88ce0c1a91 test: Verify stat calculations during backfill
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 18:17:23 -08:00
David Zafman
f5af1af6d3 test: Verify stat calculations during recovery
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 18:17:23 -08:00
David Zafman
aeba36a660 ceph-helpers.sh: Add flush_pg_stats() to wait_for_clean() to make it reliable
osd-scrub-repair.sh: Fixes for omap keys landing on different OSDs due to flush

Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 18:17:23 -08:00
David Zafman
64047e1bac osd: Don't start recovery for missing until active pg state set
I was seeing recovery hang when it is started before _activate_committed()
The state machine passes into "Active" but this transitions to activating
pg state and after commmitted into "active" pg state.

Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 18:17:23 -08:00
David Zafman
c77941f593 qa: Ignore degraded PGs when injecting random eio errors
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 18:17:23 -08:00
David Zafman
91d1487ecf osd: Improve the way insufficient targets is handled to be compatible with EC
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 18:17:23 -08:00
David Zafman
74f9e70903 osd: Improve pg degraded state setting based on _update_calc_stats() degraded count
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 18:16:20 -08:00
Sage Weil
df5a37091a
Merge pull request #19946 from liewegas/wip-22677
qa/workunits/rados/test_rados_tool: fix stray |, race
2018-01-14 13:40:54 -06:00
David Zafman
b769b98686 osd: Handling when recovery sources have missing
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 10:38:17 -08:00
David Zafman
5ffde0002f osd: Base pg degraded state on num_degraded_objects
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 10:38:17 -08:00
David Zafman
e1075f1073 osd: Rewrite _update_calc_stats() to make it cleaner and more accurate
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 10:38:17 -08:00
David Zafman
9a66a6f517 osd: cleanup: Remove unused const vars
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 10:37:22 -08:00
David Zafman
01ac714aa3 osd: cleanup: Fix log message
Signed-off-by: David Zafman <dzafman@redhat.com>
2018-01-14 10:37:22 -08:00
Sage Weil
dca1257ab1 osd/PG: flush osr before listing scrub objects
We need to make sure previous transactions are readable before we list
objects.  This avoids races like, say,

- ecbackend rolls forward and removes a rollback object
- bluestore delays the txn apply due to some prior op
- scrub lists objects, sees rollback object it doesn't expect

Signed-off-by: Sage Weil <sage@redhat.com>
2018-01-14 10:47:25 -06:00
Sage Weil
588f0643f1 osd/ECBackend: turn be_deep_scrub options into legacy
We don't have a lightweight mechanism for doing trivial config options
that is better than legacy_config_opts.h yet.  Until then,

Signed-off-by: Sage Weil <sage@redhat.com>
2018-01-14 10:47:24 -06:00
Sage Weil
3f922e79c3 qa/tasks/ceph: disable osd_debug_deep_scrub_sleep in case it is set
Otherwise the final scrub may take too long.

Signed-off-by: Sage Weil <sage@redhat.com>
2018-01-14 10:47:24 -06:00
Sage Weil
3e66d88f30 osd/*Backend: debug: inject sleep during deep scrub
Signed-off-by: Sage Weil <sage@redhat.com>
2018-01-14 10:47:24 -06:00
Sage Weil
004ee202fa osd/PG: drop waiting_on, use waiting_on_whom
Signed-off-by: Sage Weil <sage@redhat.com>
2018-01-14 10:47:24 -06:00
Sage Weil
ff9fa835da qa/workunits/rados/test_rados_tool: remove stray |, fix race
Fixes: http://tracker.ceph.com/issues/22676
Signed-off-by: Sage Weil <sage@redhat.com>
2018-01-14 10:35:14 -06:00