]> git.apps.os.sepia.ceph.com Git - ceph.git/log
ceph.git
9 years agodoc/dev: start Testing in the cloud chapter 7954/head
Nathan Cutler [Mon, 7 Mar 2016 11:04:39 +0000 (12:04 +0100)]
doc/dev: start Testing in the cloud chapter

Signed-off-by: Nathan Cutler <ncutler@suse.com>
9 years agodoc/dev: various refinements
Nathan Cutler [Mon, 7 Mar 2016 09:27:42 +0000 (10:27 +0100)]
doc/dev: various refinements

Signed-off-by: Nathan Cutler <ncutler@suse.com>
9 years agoMerge pull request #7946 from SUSE/wip-doc-dev-teuth
Loic Dachary [Mon, 7 Mar 2016 03:12:28 +0000 (10:12 +0700)]
Merge pull request #7946 from SUSE/wip-doc-dev-teuth

doc/dev: integrate testing into the narrative

Reviewed-by: Loic Dachary <ldachary@redhat.com>
9 years agodoc/dev: integrate testing into the narrative 7946/head
Nathan Cutler [Sat, 5 Mar 2016 21:48:41 +0000 (22:48 +0100)]
doc/dev: integrate testing into the narrative

Signed-off-by: Nathan Cutler <ncutler@suse.com>
9 years agoMerge pull request #7862 from theanalyst/rgw/doc-insecure-keystone
Yehuda Sadeh [Sat, 5 Mar 2016 17:29:46 +0000 (09:29 -0800)]
Merge pull request #7862 from theanalyst/rgw/doc-insecure-keystone

doc: rgw explain keystone's verify ssl switch

Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agoMerge pull request #7887 from cbodley/wip-rgw-test-multi-period
Yehuda Sadeh [Sat, 5 Mar 2016 17:25:27 +0000 (09:25 -0800)]
Merge pull request #7887 from cbodley/wip-rgw-test-multi-period

test/rgw: add multisite test for meta sync across periods

Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agoMerge pull request #7827 from cbodley/wip-rgw-period-meta-logs
Yehuda Sadeh [Sat, 5 Mar 2016 17:25:01 +0000 (09:25 -0800)]
Merge pull request #7827 from cbodley/wip-rgw-period-meta-logs

rgw: fixes for per-period metadata logs

Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agoMerge pull request #7941 from SUSE/wip-teuos-intro
Loic Dachary [Sat, 5 Mar 2016 14:55:13 +0000 (21:55 +0700)]
Merge pull request #7941 from SUSE/wip-teuos-intro

doc: detailed description of bugfixing workflow

Reviewed-by: Loic Dachary <ldachary@redhat.com>
9 years agodoc: detailed description of bugfixing workflow 7941/head
Nathan Cutler [Sat, 5 Mar 2016 04:53:38 +0000 (05:53 +0100)]
doc: detailed description of bugfixing workflow

Signed-off-by: Nathan Cutler <ncutler@suse.com>
9 years agoMerge pull request #7933 from ceph/wip-doc-dist-ceph-detect-init
Loic Dachary [Sat, 5 Mar 2016 01:09:36 +0000 (08:09 +0700)]
Merge pull request #7933 from ceph/wip-doc-dist-ceph-detect-init

doc: add ceph-detect-init(8) source to dist tarball

Reviewed-by: Loic Dachary <ldachary@redhat.com>
9 years agoMerge pull request #7937 from saru95/patch--1
Josh Durgin [Fri, 4 Mar 2016 22:45:12 +0000 (14:45 -0800)]
Merge pull request #7937 from saru95/patch--1

ceph_daemon.py: Resolved ImportError to work with python3

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
9 years agoMerge pull request #7936 from ceph/wip-fix-debian-build
Zack Cerza [Fri, 4 Mar 2016 22:12:37 +0000 (15:12 -0700)]
Merge pull request #7936 from ceph/wip-fix-debian-build

debian/changelog: Remove stray 'v' in version

9 years agoceph_daemon.py: Resolved ImportError to work with python3 7937/head
Sarthak Munshi [Fri, 4 Mar 2016 21:25:17 +0000 (02:55 +0530)]
ceph_daemon.py: Resolved ImportError to work with python3

Signed-off-by: Sarthak Munshi <sarthakmunshi@gmail.com>
9 years agorgw: fix for RGWCoroutinesManager::run() return value 7827/head
Casey Bodley [Fri, 4 Mar 2016 20:22:50 +0000 (15:22 -0500)]
rgw: fix for RGWCoroutinesManager::run() return value

the 'int ret' variable of the inner scope was shadowing an 'int ret'
variable in the outer scope, so we weren't propagating any of the error
codes

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: use current period for InitSyncStatus
Casey Bodley [Fri, 26 Feb 2016 21:23:54 +0000 (16:23 -0500)]
rgw: use current period for InitSyncStatus

the InitSyncStatus coroutine records the position to start incremental
sync after finishing a full sync.  this should be the master's marker
from the current period, rather than its oldest log period

this also adds a check to run_sync() that restarts a full sync if it
sees that our sync period is behind the master's oldest log period

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: meta log rest handlers avoid get_log()
Casey Bodley [Fri, 26 Feb 2016 17:28:41 +0000 (12:28 -0500)]
rgw: meta log rest handlers avoid get_log()

RGWMetadataManager::get_log() will allocate a log and keep it in memory.
this could lead to a potential denial of service by making requests with
lots of different period ids

RGWMetadataLog if effectively stateless (the only state is a set of
modified_shards, which are not touched by any of the rest api calls), so
we can use a temporary instead of calling get_log()

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: pull first log period from master if not found locally
Casey Bodley [Thu, 25 Feb 2016 15:41:01 +0000 (10:41 -0500)]
rgw: pull first log period from master if not found locally

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: implement find_oldest_log_period for RGWMetadataManager
Casey Bodley [Thu, 25 Feb 2016 15:40:09 +0000 (10:40 -0500)]
rgw: implement find_oldest_log_period for RGWMetadataManager

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: period history behaves with no current_period
Casey Bodley [Tue, 23 Feb 2016 20:25:25 +0000 (15:25 -0500)]
rgw: period history behaves with no current_period

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: initialize async_rados before meta_mgr
Casey Bodley [Tue, 23 Feb 2016 18:08:15 +0000 (13:08 -0500)]
rgw: initialize async_rados before meta_mgr

RGWMetadataManager::init() will depend on async_rados

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: make RGWMetadataLog::get_shard_oid public and const
Casey Bodley [Tue, 23 Feb 2016 18:07:36 +0000 (13:07 -0500)]
rgw: make RGWMetadataLog::get_shard_oid public and const

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: add RGWStatObjCR for async raw_obj_stat
Casey Bodley [Mon, 22 Feb 2016 21:42:39 +0000 (16:42 -0500)]
rgw: add RGWStatObjCR for async raw_obj_stat

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: only write meta sync shard markers if can_adjust_marker
Casey Bodley [Fri, 19 Feb 2016 20:24:00 +0000 (15:24 -0500)]
rgw: only write meta sync shard markers if can_adjust_marker

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw: make RGWMetaSyncCR wakeups thread-safe
Casey Bodley [Fri, 19 Feb 2016 20:09:41 +0000 (15:09 -0500)]
rgw: make RGWMetaSyncCR wakeups thread-safe

now that the shards will be coming and going after startup, they need to
be reference counted and protected by a mutex

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agorgw-admin: add read_current_period_id() helper function
Casey Bodley [Fri, 19 Feb 2016 19:19:24 +0000 (14:19 -0500)]
rgw-admin: add read_current_period_id() helper function

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agodebian/changelog: Remove stray 'v' in version 7936/head
Dan Mick [Fri, 4 Mar 2016 19:42:08 +0000 (11:42 -0800)]
debian/changelog: Remove stray 'v' in version

Signed-off-by: Dan Mick <dan.mick@redhat.com>
9 years agoMerge pull request #7918 from ceph/wip-rgw-payer-yehuda
Orit Wasserman [Fri, 4 Mar 2016 18:37:21 +0000 (19:37 +0100)]
Merge pull request #7918 from ceph/wip-rgw-payer-yehuda

rgw: link payer info to usage logging

Reviewed-by: Orit Wasserman <owasserm@redhat.com>
9 years agodoc: add ceph-detect-init(8) source to dist tarball 7933/head
Ken Dreyer [Fri, 4 Mar 2016 16:28:47 +0000 (09:28 -0700)]
doc: add ceph-detect-init(8) source to dist tarball

Prior to this commit, the tarball from "make dist" did not include the
ceph-detect-init(8) man page rST source.

Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
9 years agoMerge pull request #7929 from yehudasa/wip-submodules-fix
Orit Wasserman [Fri, 4 Mar 2016 15:47:22 +0000 (16:47 +0100)]
Merge pull request #7929 from yehudasa/wip-submodules-fix

submodules: revert an accidental change

Reviewed-by: Orit Wasserman <owasserm@redhat.com>
9 years agoMerge pull request #7910 from ceph/wip-14955
Samuel Just [Fri, 4 Mar 2016 15:02:47 +0000 (07:02 -0800)]
Merge pull request #7910 from ceph/wip-14955

buffer: increment history alloc as well in raw_combined

Reviewed-by: Sage Weil <sage@redhat.com>
9 years agosubmodules: revert an accidental change 7929/head
Yehuda Sadeh [Fri, 4 Mar 2016 14:54:49 +0000 (06:54 -0800)]
submodules: revert an accidental change

revert changes that went in in commit f1ac0de5bfd974258c257470feae5ff03054873e.
This affects ceph-erasure-code-corpus, civetweb

Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agoMerge pull request #7916 from ceph/wip-14975
Orit Wasserman [Fri, 4 Mar 2016 11:25:37 +0000 (12:25 +0100)]
Merge pull request #7916 from ceph/wip-14975

rgw: adjust error code when bucket does not exist in copy operation
rgw: don't override error when initializing zonegroup

Fixes: #14975
Reviewed-by: Orit Wasserman <owasserm@redhat.com>
9 years agorgw: avoid showing payer when payer == owner 7918/head
Javier M. Mellid [Wed, 2 Mar 2016 17:43:24 +0000 (18:43 +0100)]
rgw: avoid showing payer when payer == owner

If the payer matches the owner it shows the owner only.

Signed-off-by: Javier M. Mellid <jmunhoz@igalia.com>
9 years agoMerge pull request #7786 from ceph/wip-rgw-indexless
Yehuda Sadeh [Thu, 3 Mar 2016 22:18:25 +0000 (14:18 -0800)]
Merge pull request #7786 from ceph/wip-rgw-indexless

rgw: indexless buckets (Yehuda Sadeh)

 - can define a policy, for which buckets are indexless
 - users can then create buckets under the specified placement target
 - indexless buckets will not be synced across zones
 - does not work with (s3) versioned buckets

Reviewed-by: Matt Benjamin <mbenjamin@redhat.com>
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agorgw: indexless buckets 7786/head
Yehuda Sadeh [Sat, 20 Feb 2016 02:34:46 +0000 (18:34 -0800)]
rgw: indexless buckets

Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agorgw: configurable index type
Yehuda Sadeh [Sat, 20 Feb 2016 01:18:44 +0000 (17:18 -0800)]
rgw: configurable index type

Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agorgw: don't override error when initializing zonegroup 7916/head
Yehuda Sadeh [Thu, 3 Mar 2016 21:04:36 +0000 (13:04 -0800)]
rgw: don't override error when initializing zonegroup

Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agorgw: adjust error code when bucket does not exist in copy operation
Yehuda Sadeh [Thu, 3 Mar 2016 21:48:32 +0000 (13:48 -0800)]
rgw: adjust error code when bucket does not exist in copy operation

Fixes: #14975
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agoMerge pull request #7764 from rzarzynski/wip-rgw-etag-dlo
Yehuda Sadeh [Thu, 3 Mar 2016 21:43:27 +0000 (13:43 -0800)]
Merge pull request #7764 from rzarzynski/wip-rgw-etag-dlo

rgw: multiple improvements regarding etag calculation for SLO/DLO of Swift API.

Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agobuffer: increment history alloc as well in raw_combined 7910/head
Samuel Just [Thu, 3 Mar 2016 19:04:14 +0000 (11:04 -0800)]
buffer: increment history alloc as well in raw_combined

Introduced in 73dcd26fc8798e09d93520c583d63edb28ffc631

Fixes: #14955
Signed-off-by: Samuel Just <sjust@redhat.com>
9 years agoMerge remote-tracking branch 'gh/jewel'
Sage Weil [Thu, 3 Mar 2016 18:35:13 +0000 (13:35 -0500)]
Merge remote-tracking branch 'gh/jewel'

9 years agov10.0.4 v10.0.4
Sage Weil [Thu, 3 Mar 2016 18:34:46 +0000 (13:34 -0500)]
v10.0.4

Signed-off-by: Sage Weil <sage@redhat.com>
9 years agoMerge pull request #7897 from branch-predictor/bp-bench-errors-fatal
Loic Dachary [Thu, 3 Mar 2016 16:15:45 +0000 (23:15 +0700)]
Merge pull request #7897 from branch-predictor/bp-bench-errors-fatal

common/obj_bencher.cc: make verify error fatal

Reviewed-by: Loic Dachary <ldachary@redhat.com>
9 years agocommon/obj_bencher.cc: make verify error fatal 7897/head
Piotr Dałek [Thu, 3 Mar 2016 10:30:53 +0000 (11:30 +0100)]
common/obj_bencher.cc: make verify error fatal

When run without "--no-verify", all verification errors are noted,
but they are not forwarded/reported anywhere else but to cerr, which
will cause automated testing to ignore them. Make seq_read_bench and
rand_read_bench return -EIO on any verification error which will,
in turn, return it back to caller.

Fixes: #14971
Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
9 years agoMerge pull request #7757 from ukernel/jewel-14684
Sage Weil [Thu, 3 Mar 2016 14:37:50 +0000 (09:37 -0500)]
Merge pull request #7757 from ukernel/jewel-14684

mon/MDSMonitor.cc: properly note beacon when health metrics changes

Reviewed-by: John Spray <john.spray@redhat.com>
9 years agoMerge pull request #7882 from branch-predictor/bp-unconfuse-health-1
Sage Weil [Thu, 3 Mar 2016 14:15:29 +0000 (09:15 -0500)]
Merge pull request #7882 from branch-predictor/bp-unconfuse-health-1

mon: unconfuse object count skew message

Reviewed-by: Sage Weil <sage@redhat.com>
9 years agoPGMonitor: unconfuse object count skew message 7882/head
Piotr Dałek [Wed, 2 Mar 2016 12:22:38 +0000 (13:22 +0100)]
PGMonitor: unconfuse object count skew message

"Pool <pool> has too few pgs" is okay assuming it does not take other
pools into account. And since it does, it is confusing in the following
scenario:

1. Create two pools, one with small pg count and one with large
   pg count
2. Put a whole lot of objects in smaller pool, resulting in "too few
   pgs" warning on that pool, which is expected behavior.
3. Put a whole lot of objects in larger pool, warning goes away.
   Suddenly smaller pool has plenty of PGs?

Current message suggests adding more nodes (or PGs) to pool, when
actually it's warning about significantly more objects in that
particular pool than in the other pools.

Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
9 years agoMerge pull request #7898 from oritwas/wip-cmake-add-fs_types
John Spray [Thu, 3 Mar 2016 13:54:25 +0000 (13:54 +0000)]
Merge pull request #7898 from oritwas/wip-cmake-add-fs_types

cmake: add common/fs_types.cc to libcommon

Reviewed-by: John Spray <john.spray@redhat.com>
9 years agoMerge pull request #7881 from aclamk/ceph-df-used-fix
Sage Weil [Thu, 3 Mar 2016 13:51:50 +0000 (08:51 -0500)]
Merge pull request #7881 from aclamk/ceph-df-used-fix

mon: fix calculation of %USED

Reviewed-by: Sage Weil <sage@redhat.com>
9 years agoMerge pull request #7892 from majianpeng/bluestore-leak-unwritten-extent
Sage Weil [Thu, 3 Mar 2016 13:49:17 +0000 (08:49 -0500)]
Merge pull request #7892 from majianpeng/bluestore-leak-unwritten-extent

osd: bluestore: fix check for write falling within the same extent

Reviewed-by: Sage Weil <sage@redhat.com>
9 years agoMerge pull request #7886 from stiopaa1/blueStoreOverrides
Sage Weil [Thu, 3 Mar 2016 13:41:21 +0000 (08:41 -0500)]
Merge pull request #7886 from stiopaa1/blueStoreOverrides

osd: bluestore: add 'override' to virtual functions

Reviewed-by: Sage Weil <sage@redhat.com>
9 years agoMerge pull request #7895 from majianpeng/bluestore-trim-overlay
Sage Weil [Thu, 3 Mar 2016 13:40:21 +0000 (08:40 -0500)]
Merge pull request #7895 from majianpeng/bluestore-trim-overlay

os/bluestore/BlueStore: Don't leak trim overlay data before write.

9 years agoos/bluestore/BlueStore: Don't leak trim overlay data before write. 7895/head
Jianpeng Ma [Thu, 3 Mar 2016 13:46:55 +0000 (21:46 +0800)]
os/bluestore/BlueStore: Don't leak trim overlay data before write.

Suppose: bluestore_overlay_max_length=bluestore_min_alloc_size;
bluestore_overlay_max = 2;
For the following ops:
write(off=0, len=4096) --->write into overlay
write(off=4096, len=4096)-->write into overlay
write(off=0, len=bluestore_min_alloc_size)-->because overlay_map.size()
>=2, it allocate a extent.

It should trim overlay data(0,4096) &(4096, 4096),and then write(0,
bluestore_min_alloc_size).
But the original code don't trim overlay data.
This make the later read data is orignal data rather that new data.

Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
9 years agocmake: add common/fs_types.cc to libcommon 7898/head
Orit Wasserman [Thu, 3 Mar 2016 13:17:21 +0000 (14:17 +0100)]
cmake: add common/fs_types.cc to libcommon

Signed-off-by: Orit Wasserman <owasserm@redhat.com>
9 years agoos/bluestore/BlueStore: Fix bug when calc offset & end whether locate in the a extent. 7892/head
Jianpeng Ma [Thu, 3 Mar 2016 10:49:28 +0000 (18:49 +0800)]
os/bluestore/BlueStore: Fix bug when calc offset & end whether locate in the a extent.

Suppose: bluestore_overlay_max_length == bluestore_min_alloc_size
The orignal code which calc content of written whether locate in a
extent:
(offset / min_alloc_size) == (offset + length) /min_alloc_size
This will make the case which offset=0 & length =min_alloc_size locate
in the different extent.
In fact, this content is in the same extent.

Change end = offset + length - 1 make work.

Fixes: #14954
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
9 years agocommon/obj_bencher.cc: use more readable constant instead of magic number
Piotr Dałek [Thu, 3 Mar 2016 10:22:57 +0000 (11:22 +0100)]
common/obj_bencher.cc: use more readable constant instead of magic number

When clean_up_slow() fails, it returns "-5" which is equal to -EIO.
Change it in source, so it's not confusing for someone who does not
remember all error codes (functionality remains the same).

Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
9 years agoMerge pull request #7098 from liewegas/wip-file-layout
Gregory Farnum [Thu, 3 Mar 2016 03:15:18 +0000 (19:15 -0800)]
Merge pull request #7098 from liewegas/wip-file-layout

mds, client: add namespace to file_layout_t (previously ceph_file_layout)

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
9 years agoBluestore.h:Added 'override' to virtual functions 7886/head
Michal Jarzabek [Wed, 2 Mar 2016 20:18:49 +0000 (20:18 +0000)]
Bluestore.h:Added 'override' to virtual functions

Signed-off-by: Michal Jarzabek <stiopa@gmail.com>
9 years agotest/rgw: add multisite test for meta sync across periods 7887/head
Casey Bodley [Wed, 2 Mar 2016 17:15:09 +0000 (12:15 -0500)]
test/rgw: add multisite test for meta sync across periods

Signed-off-by: Casey Bodley <cbodley@redhat.com>
9 years agoMerge pull request #7849 from ceph/wip-cmake
Casey Bodley [Wed, 2 Mar 2016 19:08:53 +0000 (14:08 -0500)]
Merge pull request #7849 from ceph/wip-cmake

Wip cmake

9 years agocmake: simplified heap_profiler_objs dependencies 7849/head
Ali Maredia [Mon, 29 Feb 2016 21:37:57 +0000 (16:37 -0500)]
cmake: simplified heap_profiler_objs dependencies

The heap_profiler object library only needs to be
added to mon,mds, and osd instead of the numerous
targets it is part of

Signed-off-by: Ali Maredia <amaredia@redhat.com>
9 years agoclient/Inode: fix has_dir_layout 7098/head
Sage Weil [Tue, 1 Mar 2016 15:40:41 +0000 (10:40 -0500)]
client/Inode: fix has_dir_layout

Compare to default file_layout_t() -- not zeroed memory.

Signed-off-by: Sage Weil <sage@redhat.com>
9 years agoMerge pull request #7867 from SUSE/wip-14940
branto1 [Wed, 2 Mar 2016 17:26:13 +0000 (18:26 +0100)]
Merge pull request #7867 from SUSE/wip-14940

RPM: move %post(un) ldconfig calls to ceph-base

Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
Reviewed-by: Boris Ranto <branto@redhat.com>
9 years agoMerge pull request #7869 from rzarzynski/wip-rgw-aws4-puthash
Orit Wasserman [Wed, 2 Mar 2016 17:08:48 +0000 (18:08 +0100)]
Merge pull request #7869 from rzarzynski/wip-rgw-aws4-puthash

rgw: calculate payload hash in RGWPutObj_ObjStore only when necessary.

Reviewed-by: Orit Wasserman <owasserm@redhat.com>
9 years agoMerge remote-tracking branch 'gh/jewel'
Sage Weil [Wed, 2 Mar 2016 14:22:28 +0000 (09:22 -0500)]
Merge remote-tracking branch 'gh/jewel'

9 years agoMerge pull request #7612 from liewegas/wip-buffer-combined
Sage Weil [Wed, 2 Mar 2016 13:31:08 +0000 (08:31 -0500)]
Merge pull request #7612 from liewegas/wip-buffer-combined

buffer: raw_combined allocations buffer and ref count together

Reviewed-by: Casey Bodley <cbodley@redhat.com>
9 years agoMerge pull request #7690 from xiexingguo/xxg-wip-14804
Sage Weil [Wed, 2 Mar 2016 13:30:26 +0000 (08:30 -0500)]
Merge pull request #7690 from xiexingguo/xxg-wip-14804

common/TrackedOp: fix inaccurate counting for slow requests

Reviewed-by: Kefu Chai <kchai@redhat.com>
9 years agoMerge pull request #7704 from xiexingguo/xxg-wip-14814
Sage Weil [Wed, 2 Mar 2016 13:29:54 +0000 (08:29 -0500)]
Merge pull request #7704 from xiexingguo/xxg-wip-14814

osd: filestore: fix wrong scope of result code for error cases during mkfs

Reviewed-by: Sage Weil <sage@redhat.com>
9 years agoMerge pull request #7812 from xiexingguo/xxg-wip-14872
Sage Weil [Wed, 2 Mar 2016 13:29:09 +0000 (08:29 -0500)]
Merge pull request #7812 from xiexingguo/xxg-wip-14872

osd: fix wrong return type of find_osd_on_ip()

Reviewed-by: Sage Weil <sage@redhat.com>
9 years agoMerge pull request #7815 from branch-predictor/bp-conditional-crc
Haomai Wang [Wed, 2 Mar 2016 13:28:40 +0000 (21:28 +0800)]
Merge pull request #7815 from branch-predictor/bp-conditional-crc

msg/async: don't calculate msg header crc when not needed

Reviewed-by: Haomai Wang <haomai@xsky.com>
9 years agoMerge pull request #7711 from xiexingguo/xxg-wip-14817
Sage Weil [Wed, 2 Mar 2016 13:28:03 +0000 (08:28 -0500)]
Merge pull request #7711 from xiexingguo/xxg-wip-14817

osd: filestore: fix result code overwritten for clone

Reviewed-by: Sage Weil <sage@redhat.com>
9 years agoMerge pull request #7790 from tchaikov/wip-log-segv
Sage Weil [Wed, 2 Mar 2016 13:27:36 +0000 (08:27 -0500)]
Merge pull request #7790 from tchaikov/wip-log-segv

log: segv in a portable way

Reviewed-by: Nathan Cutler <ncutler@suse.com>
9 years agoMerge pull request #7842 from xiexingguo/xxg-wip-14921
Sage Weil [Wed, 2 Mar 2016 13:26:30 +0000 (08:26 -0500)]
Merge pull request #7842 from xiexingguo/xxg-wip-14921

osd: fix overload of '==' operator for pg_stat_t

Reviewed-by: Sage Weil <sage@redhat.com>
9 years ago[MON] Fixed calculation of %USED. Now it is shows (space used by all replicas)/(raw... 7881/head
Adam Kupczyk [Wed, 2 Mar 2016 11:31:01 +0000 (12:31 +0100)]
[MON] Fixed calculation of %USED. Now it is shows (space used by all replicas)/(raw space available on OSDs). Before it was (size of pool)/(raw space available on OSDs).

Signed-off-by: Adam Kupczyk akupczyk@mirantis.com
9 years agoMerge pull request #7390 from badone/wip-13829
Kefu Chai [Wed, 2 Mar 2016 03:22:24 +0000 (11:22 +0800)]
Merge pull request #7390 from badone/wip-13829

common: Allow OPT_INT settings with negative values

Reviewed-by: Joao Eduardo Luis <joao@suse.de>
9 years agoMerge pull request #7854 from yehudasa/wip-authors-update
Kefu Chai [Wed, 2 Mar 2016 02:18:25 +0000 (10:18 +0800)]
Merge pull request #7854 from yehudasa/wip-authors-update

AUTHORS: update email

Reviewed-by: Abhishek Lekshmanan <abhishek@suse.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
9 years agoMerge pull request #7846 from dillaman/wip-14551
Josh Durgin [Wed, 2 Mar 2016 01:18:04 +0000 (17:18 -0800)]
Merge pull request #7846 from dillaman/wip-14551

rbd: update default image features

Conflicts:
PendingReleaseNotes (trivial)

9 years agoMerge pull request #7841 from dillaman/wip-14550
Josh Durgin [Wed, 2 Mar 2016 01:18:18 +0000 (17:18 -0800)]
Merge pull request #7841 from dillaman/wip-14550

rbd: deprecate image format 1

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
9 years agoMerge pull request #7839 from dillaman/wip-14809
Josh Durgin [Wed, 2 Mar 2016 01:18:05 +0000 (17:18 -0800)]
Merge pull request #7839 from dillaman/wip-14809

librbd: remove last synchronous librados calls from open/close state machine

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
9 years agoMerge pull request #7785 from dillaman/wip-14822
Josh Durgin [Wed, 2 Mar 2016 01:17:50 +0000 (17:17 -0800)]
Merge pull request #7785 from dillaman/wip-14822

librbd: journal replay needs to support re-executing maintenance ops

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
9 years agoMerge pull request #7776 from dillaman/wip-librbd-journal-tweaks
Josh Durgin [Wed, 2 Mar 2016 01:17:31 +0000 (17:17 -0800)]
Merge pull request #7776 from dillaman/wip-librbd-journal-tweaks

journal: improve commit position tracking

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
9 years agocmake: Turned LTTng OFF
Ali Maredia [Mon, 29 Feb 2016 21:11:04 +0000 (16:11 -0500)]
cmake: Turned LTTng OFF

This is temporary, until a way to stop repetitive
unecessary generation of headerfiles is figured out.

Signed-off-by: Ali Maredia <amaredia@redhat.com>
9 years agoRPM: move scriptlets from ceph to ceph-base 7867/head
Nathan Cutler [Tue, 1 Mar 2016 20:25:11 +0000 (21:25 +0100)]
RPM: move scriptlets from ceph to ceph-base

This addresses the following RPMLINT error:

ceph-base.x86_64: E: library-without-ldconfig-postun (Badness:
300) /usr/lib64/libosd_tp.so.1.0.0
ceph-base.x86_64: E: library-without-ldconfig-postun (Badness:
300) /usr/lib64/libos_tp.so.1.0.0
This package contains a library and provides no %postun scriptlet
containing a call to ldconfig.

ceph-base.x86_64: E: library-without-ldconfig-postin (Badness:
300) /usr/lib64/libosd_tp.so.1.0.0
ceph-base.x86_64: E: library-without-ldconfig-postin (Badness:
300) /usr/lib64/libos_tp.so.1.0.0
This package contains a library and provides no %post scriptlet
containing a call to ldconfig.

http://tracker.ceph.com/issues/14940 Fixes: #14940

Signed-off-by: Nathan Cutler <ncutler@suse.com>
9 years agoAUTHORS: update email 7854/head
Yehuda Sadeh [Tue, 1 Mar 2016 04:13:11 +0000 (20:13 -0800)]
AUTHORS: update email

Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
9 years agorgw: calculate payload hash in RGWPutObj_ObjStore only when necessary. 7869/head
Radoslaw Zarzynski [Tue, 1 Mar 2016 20:52:44 +0000 (21:52 +0100)]
rgw: calculate payload hash in RGWPutObj_ObjStore only when necessary.

Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
9 years agoMerge pull request #7767 from athanatos/wip-sam-journal-throttle-4
Samuel Just [Tue, 1 Mar 2016 20:51:15 +0000 (12:51 -0800)]
Merge pull request #7767 from athanatos/wip-sam-journal-throttle-4

osd: filestore: restructure journal and op queue throttling

Reviewed-by: Samuel Just <sjust@redhat.com>
9 years agomds: deny access to file with pool_ns layout if feature is missing
Yan, Zheng [Sun, 14 Feb 2016 09:27:26 +0000 (17:27 +0800)]
mds: deny access to file with pool_ns layout if feature is missing

Signed-off-by: Yan, Zheng <zyan@redhat.com>
9 years agomds: fix inodestat size calculation
Yan, Zheng [Sun, 14 Feb 2016 08:59:31 +0000 (16:59 +0800)]
mds: fix inodestat size calculation

Signed-off-by: Yan, Zheng <zyan@redhat.com>
9 years agomds: properly purge backtrace object
Yan, Zheng [Sun, 14 Feb 2016 08:42:56 +0000 (16:42 +0800)]
mds: properly purge backtrace object

backtrace objects are always in default namespace

Signed-off-by: Yan, Zheng <zyan@redhat.com>
9 years agomessages/MClientCaps: fix decoding
Yan, Zheng [Mon, 15 Feb 2016 15:28:32 +0000 (23:28 +0800)]
messages/MClientCaps: fix decoding

Signed-off-by: Yan, Zheng <zyan@redhat.com>
9 years agomds/Server: fix whitespace
Sage Weil [Fri, 26 Feb 2016 15:32:20 +0000 (10:32 -0500)]
mds/Server: fix whitespace

Signed-off-by: Sage Weil <sage@redhat.com>
9 years agomds/Locker: fix Session leak on access denial
Sage Weil [Fri, 26 Feb 2016 15:32:15 +0000 (10:32 -0500)]
mds/Locker: fix Session leak on access denial

Signed-off-by: Sage Weil <sage@redhat.com>
9 years agofix init of new fields in MCLientCaps
Sage Weil [Tue, 23 Feb 2016 15:26:15 +0000 (10:26 -0500)]
fix init of new fields in MCLientCaps

9 years agomds/MDSMap: add file layout v2 incompat feature
Sage Weil [Wed, 10 Feb 2016 14:43:32 +0000 (09:43 -0500)]
mds/MDSMap: add file layout v2 incompat feature

The existing MDSMonitor machinations already add this to the MDSMap to
boot out old MDS's.

Signed-off-by: Sage Weil <sage@redhat.com>
9 years agomds: deny access to directories with pool_ns layouts if feature is missing
Sage Weil [Wed, 10 Feb 2016 14:38:33 +0000 (09:38 -0500)]
mds: deny access to directories with pool_ns layouts if feature is missing

If the client feature is missing, given them EIO.  Change check_access to
return an error code to distinguish this from EACCES.

Signed-off-by: Sage Weil <sage@redhat.com>
9 years agomon/MDSMonitor: prevent pool 0 from being used as a data pool
Sage Weil [Mon, 1 Feb 2016 18:01:32 +0000 (13:01 -0500)]
mon/MDSMonitor: prevent pool 0 from being used as a data pool

Pool 0 means no change or default in the legacy ceph_file_layout in the
layout ioctl and file create arguments. Prevent it from being used to avoid
putting users in an awkward situation later.

Signed-off-by: Sage Weil <sage@redhat.com>
9 years agomds: don't break compatibility of MClientCaps/MClientReply
Yan, Zheng [Mon, 1 Feb 2016 14:17:04 +0000 (22:17 +0800)]
mds: don't break compatibility of MClientCaps/MClientReply

Signed-off-by: Yan, Zheng <zyan@redhat.com>
9 years agofs_types: file_layout_t: convert pool -1 (undefined) to 0 in legacy encoding
Sage Weil [Tue, 12 Jan 2016 14:57:06 +0000 (09:57 -0500)]
fs_types: file_layout_t: convert pool -1 (undefined) to 0 in legacy encoding

Old code assumes that fl_pg_pool == 0 means the pool is not defined, while
file_layout_t uses -1.  Translate between the two.

Note that this means a valid file_layout_t with pool_id == 0 cannot be
accurately translated to a legacy file_layout_t.  That is somewhat
unavoidable, and should not be a problem since real clusters create 'rbd'
as pool 0 and it does not use any file layouts.

Signed-off-by: Sage Weil <sage@redhat.com>
9 years agoqa/workunits/fs/misc/layout_vxattrs: test layout.pool_namespace
Sage Weil [Mon, 4 Jan 2016 15:51:36 +0000 (10:51 -0500)]
qa/workunits/fs/misc/layout_vxattrs: test layout.pool_namespace

Signed-off-by: Sage Weil <sage@redhat.com>
9 years agoclient: expose layout.pool_namespace vxattr
Sage Weil [Mon, 4 Jan 2016 14:15:34 +0000 (09:15 -0500)]
client: expose layout.pool_namespace vxattr

Signed-off-by: Sage Weil <sage@redhat.com>