]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Andrew Schoen [Tue, 19 Jun 2018 19:34:57 +0000 (14:34 -0500)]
lvm: when osd creation fails log the exception
Fixes: http://tracker.ceph.com/issues/24456
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
(cherry picked from commit
d622dadef09a53d24c6a7b9119051594f8da1cae )
Yuri Weinstein [Wed, 13 Jun 2018 22:00:56 +0000 (15:00 -0700)]
Merge pull request #22109 from tchaikov/luminous-24180
luminous: mon/LogMonitor: call no_reply() on ignored log message
Reviewed-by: Sage Weil <sage@redhat.com>
Yuri Weinstein [Wed, 13 Jun 2018 22:00:19 +0000 (15:00 -0700)]
Merge pull request #22290 from pdvian/wip-24216-luminous
luminous: common: partially revert
95fc248 to make get_process_name work
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
Yuri Weinstein [Wed, 13 Jun 2018 21:59:36 +0000 (14:59 -0700)]
Merge pull request #22308 from pdvian/wip-24214-luminous
luminous: Module 'balancer' has failed: could not find bucket -14
Reviewed-by: Sage Weil <sage@redhat.com>
Yuri Weinstein [Wed, 13 Jun 2018 21:59:14 +0000 (14:59 -0700)]
Merge pull request #22414 from wido/luminous-influx
luminous: backport various fixes to mgr/influx module
Reviewed-by: Sage Weil <sage@redhat.com>
Alfredo Deza [Tue, 12 Jun 2018 14:40:35 +0000 (10:40 -0400)]
Merge pull request #22530 from ceph/backport-luminous-22062
luminous: ceph-volume dmcrypt and activate --all documentation updates
Reviewed-by: Alfredo Deza <adeza@redhat.com>
Alfredo Deza [Thu, 17 May 2018 13:40:15 +0000 (09:40 -0400)]
doc/man/ceph-volume explain encryption and activate all with examples
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
0a4cebd1dda876cfc19fbfb97c668d4dae05097c )
Alfredo Deza [Thu, 17 May 2018 13:05:52 +0000 (09:05 -0400)]
doc/ceph-volume activate is able to use --all for existing OSDs
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
d8b9a7c0745d6a901988a08a8218d9da585e5aa8 )
Conflicts:
doc/ceph-volume/lvm/activate.rst
Alfredo Deza [Thu, 17 May 2018 12:46:03 +0000 (08:46 -0400)]
doc/ceph-volume encryption is enabled with the --dmcrypt flag
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
72cef7acf7a9a965960ec3c7c1b99ce8e7fa3922 )
Alfredo Deza [Thu, 17 May 2018 12:45:26 +0000 (08:45 -0400)]
doc/ceph-volume prepapre accepts a --dmcrypt flag
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
e97d0717d323a08a46aa997c63d2416da7e4941f )
Alfredo Deza [Thu, 17 May 2018 12:44:41 +0000 (08:44 -0400)]
doc/ceph-volume activate does not need a dmcrypt flag
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
ac0c925c576868107e585ba458fbff956467bccb )
Alfredo Deza [Tue, 12 Jun 2018 10:47:51 +0000 (06:47 -0400)]
Merge pull request #22510 from ceph/backport-luminous-22399
luminous: ceph-volume tests do not include admin keyring in OSD nodes
Reviewed-by: Alfredo Deza <adeza@redhat.com>
Alfredo Deza [Tue, 12 Jun 2018 10:45:36 +0000 (06:45 -0400)]
Merge pull request #22506 from ceph/backport-luminous-22437
luminous: ceph-volume tests destroy osds on monitor hosts
Reviewed-by: Alfredo Deza <adeza@redhat.com>
Alfredo Deza [Tue, 12 Jun 2018 10:44:54 +0000 (06:44 -0400)]
Merge pull request #22513 from ceph/backport-luminous-21890
luminous: ceph-volume describe better the options for migrating away from ceph-disk
Reviewed-by: Alfredo Deza <adeza@redhat.com>
Alfredo Deza [Tue, 8 May 2018 17:48:20 +0000 (13:48 -0400)]
doc/rados add an anchor to reference replacing osds
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
0cc7bdc67ae3f2b1aeab8cc6046b6d172915e548 )
Alfredo Deza [Tue, 8 May 2018 17:47:48 +0000 (13:47 -0400)]
doc/ceph-volume describe the options for migrating away from ceph-disk
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
537ff9cc3b64c59f9dbb81ac3cf180460ecaa3f3 )
Alfredo Deza [Tue, 8 May 2018 17:47:11 +0000 (13:47 -0400)]
doc/ceph-volume high-level description of simple activate at index.rst
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
dd6f368158db7f102b485179166ebe73d6ea622f )
Alfredo Deza [Mon, 4 Jun 2018 20:01:15 +0000 (16:01 -0400)]
ceph-volume tests do not include admin keyring in OSD nodes
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
08760632743040520588e1bc03adc17746ff2eca )
Alfredo Deza [Wed, 6 Jun 2018 15:37:21 +0000 (11:37 -0400)]
ceph-volume tests functional osd destroy on mons only for dmcrypt
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
953c7ef54925059e6de75c2e11d9d40d909bbbca )
Alfredo Deza [Wed, 6 Jun 2018 13:02:42 +0000 (09:02 -0400)]
ceph-volume tests/functional destroy filestore osds on the monitors
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
41277c65b4c745523e3a83bba00eea3482df4100 )
Alfredo Deza [Wed, 6 Jun 2018 12:59:34 +0000 (08:59 -0400)]
ceph-volume tests/functional destroy bluestore osds on the monitors
Signed-off-by: Alfredo Deza <adeza@redhat.com>
(cherry picked from commit
b665e1caa70a8a4b5a8c972b769bf33d5d6edde4 )
Yuri Weinstein [Mon, 11 Jun 2018 14:16:51 +0000 (07:16 -0700)]
Merge pull request #21938 from pdvian/wip-23985-luminous
luminous: librbd: prevent watcher from unregistering with in-flight actions
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Yuri Weinstein [Mon, 11 Jun 2018 14:16:24 +0000 (07:16 -0700)]
Merge pull request #22142 from pdvian/wip-24084-luminous
luminous: rbd-mirror: ensure remote demotion is replayed locally
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Yuri Weinstein [Mon, 11 Jun 2018 14:15:56 +0000 (07:15 -0700)]
Merge pull request #22143 from pdvian/wip-24059-luminous
luminous: Deleting a pool with active notify linger ops can result in seg fault
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Yuri Weinstein [Mon, 11 Jun 2018 14:14:53 +0000 (07:14 -0700)]
Merge pull request #22172 from pdvian/wip-24086-luminous
luminous: [rbd-mirror] potential races during PoolReplayer shut-down
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Yuri Weinstein [Sat, 9 Jun 2018 14:24:39 +0000 (07:24 -0700)]
Merge pull request #21412 from ukernel/luminous-23698
luminous: mds: misc load balancer fixes
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Sat, 9 Jun 2018 14:24:08 +0000 (07:24 -0700)]
Merge pull request #21897 from batrick/i24055
luminous: VolumeClient: allow ceph_volume_client to create 'volumes' without namespace isolation
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Sat, 9 Jun 2018 14:23:02 +0000 (07:23 -0700)]
Merge pull request #22354 from ukernel/luminous-24369
luminous: client: void sending mds request while holding cap reference
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Yuri Weinstein [Fri, 8 Jun 2018 14:34:13 +0000 (07:34 -0700)]
Merge pull request #21968 from ceph/backport-21605
luminous: mon: High MON cpu usage when cluster is changing
Reviewed-by: Sage Weil <sage@redhat.com>
Yuri Weinstein [Fri, 8 Jun 2018 14:33:48 +0000 (07:33 -0700)]
Merge pull request #22253 from jdurgin/wip-stat-decode
luminous - osd/osd_types: fix object_stat_sum_t decode
Reviewed-by: Sage Weil <sage@redhat.com>
Yuri Weinstein [Fri, 8 Jun 2018 14:33:06 +0000 (07:33 -0700)]
Merge pull request #22331 from pdvian/wip-24245-luminous
luminous : mon/MgrMonitor: change 'unresponsive' message to info level
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Yuri Weinstein [Fri, 8 Jun 2018 14:32:21 +0000 (07:32 -0700)]
Merge pull request #22351 from pdvian/wip-24247-luminous
luminous: os/bluestore: simplify and fix SharedBlob::put()
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Yuri Weinstein [Fri, 8 Jun 2018 14:31:35 +0000 (07:31 -0700)]
Merge pull request #22360 from tchaikov/luminous-24361
luminous: mon: enable level_compaction_dynamic_level_bytes for rocksdb
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Josh Durgin [Thu, 7 Jun 2018 23:51:11 +0000 (16:51 -0700)]
Merge pull request #22400 from yuriw/wip-yuriw-p2p-luminous
qa/tests - moved point-to-point suite into own folder
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 22:17:06 +0000 (15:17 -0700)]
Merge pull request #21437 from pdvian/wip-23683-luminous
luminous: rgw: radosgw-admin should not use metadata cache for readonly commands
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 22:16:39 +0000 (15:16 -0700)]
Merge pull request #21724 from smithfarm/wip-23231-luminous
luminous: rgw: rgw_statfs should report the correct stats
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 22:15:52 +0000 (15:15 -0700)]
Merge pull request #21738 from cbodley/wip-23906
luminous: rgw: fix use of libcurl with empty header values
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 22:15:11 +0000 (15:15 -0700)]
Merge pull request #22177 from pdvian/wip-24120-luminous
luminous: rgw: trim all spaces inside a metadata value
Reviewed-by: Matt Benjamin <mbenjami@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 22:13:44 +0000 (15:13 -0700)]
Merge pull request #22302 from linuxbox2/wip-luminous-irange
luminous: rgw: add configurable AWS-compat invalid range get behavior
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 22:12:47 +0000 (15:12 -0700)]
Merge pull request #22352 from pdvian/wip-24252-luminous
luminous: rgw: Do not modify email if argument is not set
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 22:10:47 +0000 (15:10 -0700)]
Merge pull request #22363 from mdw-at-linuxbox/wip-luminous-rgw-implt
luminous: rgw: making implicit_tenants backwards compatible
Reviewed-by: Matt Benjamin <mbenjami@redhat.com>
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 12:03:38 +0000 (05:03 -0700)]
Merge pull request #21316 from pdvian/wip-23607-luminous
luminous: rbd: improve 'import-diff' corrupt input error messages
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 12:02:57 +0000 (05:02 -0700)]
Merge pull request #22180 from pdvian/wip-24156-luminous
luminous: [rbd-mirror] potential deadlock when running asok 'flush' command
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Yuri Weinstein [Thu, 7 Jun 2018 12:01:38 +0000 (05:01 -0700)]
Merge pull request #22370 from dillaman/wip-24378-luminous
luminous: librbd: commit IO as safe when complete if writeback cache is disabled
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
Wido den Hollander [Wed, 31 Jan 2018 13:43:54 +0000 (14:43 +0100)]
mgr/influx: Only fetch the current time once when gathering data
By fetching the current time once and storing it into a variable
we save a lot of system calls. On large clusters this can be a lot
of system calls.
In addition we also make sure that all points gathered in the same
loop/run have exactly the same timestamp.
Otherwise there will be a difference in time between the items in
InfluxDB which then causes problems when creating graphs with for
example Grafana.
Signed-off-by: Wido den Hollander <wido@42on.com>
(cherry picked from commit
0857d3bcd4175f39582376fb9980cf3f66fa09ce )
Wido den Hollander [Wed, 31 Jan 2018 12:15:05 +0000 (13:15 +0100)]
mgr/influx: Send more information about pools
In addition to the existing statistics also send Read and Write IOps,
but also quota information.
Signed-off-by: Wido den Hollander <wido@42on.com>
(cherry picked from commit
c803f38a17af625edc1cee29f21e510a03396204 )
Wido den Hollander [Tue, 30 Jan 2018 14:45:40 +0000 (15:45 +0100)]
mgr/influx: Catch ConnectionError if raised
The InfluxDBClient can also re-raise a ConnectionError from the
python requests module if that was caught.
The Exception might be:
ConnectionError: ('Connection aborted.', error(104, 'Connection reset by peer'))
Catch and log this error instead of having it raised which might
cause problems further down.
Signed-off-by: Wido den Hollander <wido@42on.com>
(cherry picked from commit
a0c2c8b900cd2bab2ef8041722036036efd2cb0f )
Tobias Gall [Wed, 13 Dec 2017 18:14:50 +0000 (19:14 +0100)]
Add InfluxDB SSL Option
Add possibility to connect to InfluxDB via https.
Also adding the option for verifying the https cert.
Signed-off-by: Tobias Gall <tobias.gall@mailbox.org>
(cherry picked from commit
f5efaa2f522b7cc7497db1d5a82785e7f08b7bb0 )
Wido den Hollander [Mon, 11 Dec 2017 14:44:05 +0000 (15:44 +0100)]
mgr/influx: Print (in debug) time it takes to send data to Influx
On large clusters it might take a very long time to send data to Influx
due to the gathering and parsing of statistics.
By keeping a counter and printing it admins can adjust the interval if it
becomes to heavy for their cluster.
Signed-off-by: Wido den Hollander <wido@42on.com>
(cherry picked from commit
3f07636bc235ace8da3f2f07aefd56e53085ea0e )
Wido den Hollander [Mon, 11 Dec 2017 14:12:51 +0000 (15:12 +0100)]
mgr/influx: Revise configuration of module
It's a lot like the Zabbix module and allows for setting configuration
options on run-time and also fetch them from the module.
A few additional commands have been registered to make sure it is easy
to interact with the module.
Signed-off-by: Wido den Hollander <wido@42on.com>
(cherry picked from commit
429718af1cb30de49e8c7550b04bd788010ba1a4 )
Wido den Hollander [Wed, 29 Nov 2017 08:07:23 +0000 (09:07 +0100)]
mgr/influx: Send fsid as a tag to Influx to identify cluster
This allows for multiple Ceph clusters to send their data to the
same Influx database.
Using the fsid values for different clusters can be queried from
Influx
Signed-off-by: Wido den Hollander <wido@42on.com>
(cherry picked from commit
2b71a8919c705834d9e669731c2ac83c52715b5b )
Wido den Hollander [Wed, 29 Nov 2017 07:18:35 +0000 (08:18 +0100)]
mgr/influx: PEP-8 fixes to InfluxDB module
The code now scores:
Your code has been rated at 8.94/10
Various indentation fixes, whitespaces and other PEP-8 related changes
Signed-off-by: Wido den Hollander <wido@42on.com>
(cherry picked from commit
9440be3da292b98739bc8362d82b1e86b831d50b )
Abhishek L [Wed, 6 Jun 2018 12:57:00 +0000 (14:57 +0200)]
Merge pull request #21699 from jcsp/wip-mgr-doc-bal-luminous
luminous: doc/mgr/balancer: document
Reviewed-By: Sage Weil <sage@redhat.com>
Marcus Watts [Wed, 30 May 2018 20:37:31 +0000 (16:37 -0400)]
rgw: making implicit_tenants backwards compatible.
In jewel, "rgw keystone implicit tenants" only applied to swift. As of
luminous), this option applies to s3 also.
Sites that used this feature with jewel now have outstanding data that
depends on the old behavior.
The fix here is to expand "rgw keystone implicit tenants" so that it
can be set to any of "none", "all", "s3" or "swift" (also 0=false=none,
1=true=all). When set to "s3" or "swift", the actual id lookup
is also partitioned.
Formerly "rgw keystone implicit tenants" was a legacy opt.
This change converts it to the new style of option,
including support for dynamically changing it.
Fixes: http://tracker.ceph.com/issues/24348
Signed-off-by: Marcus Watts <mwatts@redhat.com>
(cherry picked from commit
a28a38f6e91da3abe59c34fad0e059eeaf29a65f )
David Zafman [Tue, 5 Jun 2018 18:11:23 +0000 (11:11 -0700)]
Merge pull request #22375 from dzafman/wip-scrub-omap-luminous
Reviewed-by: Kefu Chai <kchai@redhat.com>
Yan, Zheng [Tue, 5 Jun 2018 00:49:18 +0000 (08:49 +0800)]
Merge pull request #21187 from smithfarm/wip-23157-luminous
luminous: mds: underwater dentry check in CDir::_omap_fetched is racy
Yan, Zheng [Tue, 5 Jun 2018 00:44:24 +0000 (08:44 +0800)]
Merge pull request #21278 from batrick/i22696
luminous: client: dirty caps may never get the chance to flush
Yan, Zheng [Sat, 17 Feb 2018 01:37:48 +0000 (09:37 +0800)]
mds: fix check of underwater dentries
Underwater dentry is dentry that is dirty in our cache from journal
replay, but had already been flushed to disk before the mds failed.
To decide if an dentry is underwater, original code compares dirty
dentry's version to on-disk dirfrag's version. This method is racy
because CDir::log_mark_dirty() can increase dirfrag's version without
adding log event. After mds failover, version of dirfrag from journal
replay can be less than on-disk dirfrag's version. So newly dirtied
dentry can be equal to or less than the on-disk dirfrag's version.
The race can cause incorrect fragstat/rstat
Fixes: http://tracker.ceph.com/issues/23032
Signed-off-by: Yan, Zheng <zyan@redhat.com>
(cherry picked from commit
9d271696b7735ab5b7384cdd386d6ac3eaafe437 )
Conflicts:
src/mds/CDir.cc - conditional being replaced by "if (!dn)" is different
in luminous
Yuri Weinstein [Mon, 4 Jun 2018 20:27:42 +0000 (13:27 -0700)]
qa/tests - moved point-to-point suite into own folder
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
dongdong tao [Thu, 4 Jan 2018 07:05:41 +0000 (15:05 +0800)]
client: make mark_caps_clean and mark_caps_dirty as member function to Inode
Signed-off-by: dongdong tao <tdd21151186@gmail.com>
(cherry picked from commit
06ce613779a810dd979cb2eb15b510131a81fd74 )
dongdong tao [Wed, 27 Dec 2017 15:47:16 +0000 (23:47 +0800)]
client: fix dirty caps might never be flushed
Fixes: http://tracker.ceph.com/issues/22546
Signed-off-by: dongdong tao <tdd21151186@gmail.com>
(cherry picked from commit
aeb920be8ab5f0e5e47d82943c214b012bb8ec5c )
Conflicts:
src/client/Client.cc
Yuri Weinstein [Mon, 4 Jun 2018 14:32:52 +0000 (07:32 -0700)]
Merge pull request #22258 from yuriw/wip-yuriw-fix-luminous
luminous: qa/tests: added missed ubuntu_latest
Yan, Zheng [Fri, 1 Jun 2018 02:21:20 +0000 (10:21 +0800)]
client: void sending mds request while holding cap reference
Client::_write() calls is_quota_bytes_approaching() while holding Fw
cap reference. The later function may send lookup_name request to mds.
This can cause deadlock (mds delays handling the lookup_name request
because subtree is freezing. The subtree stays in freezing state because
mds is revoking the Fw cap and client does not release it)
This patch isn't cherry-picked from master because quota implementation
has been changed in master (introduced quota realm).
Fixes: http://tracker.ceph.com/issues/24369
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Yan, Zheng [Mon, 20 Nov 2017 10:15:39 +0000 (18:15 +0800)]
mds: don't decay load counters when printing load
Changing load counters while printing load results confusing output.
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit
831c5d91cc3863c8ec6ef76ff71f4b7d237ee089 )
Yan, Zheng [Sat, 2 Jun 2018 14:24:24 +0000 (22:24 +0800)]
Merge pull request #21475 from pdvian/wip-23704-luminous
luminous: fuse: wire up fuse_ll_access
Yan, Zheng [Sat, 2 Jun 2018 14:21:56 +0000 (22:21 +0800)]
Merge pull request #22119 from pdvian/wip-24049-luminous
luminous: ceph-fuse: missing dentries in readdir result
Yan, Zheng [Sat, 2 Jun 2018 01:14:41 +0000 (09:14 +0800)]
Merge branch 'luminous' into wip-24049-luminous
Yan, Zheng [Sat, 2 Jun 2018 01:06:54 +0000 (09:06 +0800)]
Merge branch 'luminous' into wip-23704-luminous
Yan, Zheng [Sat, 2 Jun 2018 00:36:48 +0000 (08:36 +0800)]
Merge pull request #21495 from pdvian/wip-23770-luminous
luminous: ceph-fuse: return proper exit code
Yan, Zheng [Sat, 2 Jun 2018 00:35:33 +0000 (08:35 +0800)]
Merge pull request #22176 from pdvian/wip-24107-luminous
luminous: mds: set could_consume to false when no purge queue item actually exe…
Yan, Zheng [Sat, 2 Jun 2018 00:34:47 +0000 (08:34 +0800)]
Merge pull request #22310 from ukernel/luminous-24341
luminous: mds: fix some memory leak
Yan, Zheng [Sat, 2 Jun 2018 00:34:19 +0000 (08:34 +0800)]
Merge pull request #22271 from pdvian/wip-24205-luminous
luminous: mds: broadcast quota to relevant clients when quota is explicitly set
Yan, Zheng [Sat, 2 Jun 2018 00:33:28 +0000 (08:33 +0800)]
Merge pull request #22221 from pdvian/wip-24201-luminous
luminous: client: fix issue of revoking non-auth caps
Yan, Zheng [Sat, 2 Jun 2018 00:32:54 +0000 (08:32 +0800)]
Merge pull request #22208 from pdvian/wip-24188-luminous
luminous: kceph: umount on evicted client blocks forever
Yan, Zheng [Sat, 2 Jun 2018 00:32:03 +0000 (08:32 +0800)]
Merge pull request #22171 from ukernel/luminous-24108
luminous: mds: avoid calling rejoin_gather_finish() two times successively
Yan, Zheng [Sat, 2 Jun 2018 00:31:22 +0000 (08:31 +0800)]
Merge pull request #22168 from ukernel/luminous-24207
luminous: client: avoid freeing inode when it contains TX buffer head
Yan, Zheng [Sat, 2 Jun 2018 00:30:30 +0000 (08:30 +0800)]
Merge pull request #22118 from pdvian/wip-24050-luminous
luminous: mds: include nfiles/nsubdirs of directory inode in MClientCaps
Yan, Zheng [Sat, 2 Jun 2018 00:30:11 +0000 (08:30 +0800)]
Merge pull request #22018 from batrick/i23991
luminous: client: hangs on umount if it had an MDS session evicted
Yan, Zheng [Sat, 2 Jun 2018 00:28:56 +0000 (08:28 +0800)]
Merge pull request #21990 from batrick/i23935
luminous: mds: don't discover inode/dirfrag when mds is in 'starting' state
Yan, Zheng [Sat, 2 Jun 2018 00:28:02 +0000 (08:28 +0800)]
Merge pull request #21989 from batrick/i24130
luminous: mds: handle imported session race
Yan, Zheng [Sat, 2 Jun 2018 00:27:43 +0000 (08:27 +0800)]
Merge pull request #21922 from pdvian/wip-23984-luminous
luminous: mds: mark new root inode dirty
Yan, Zheng [Sat, 2 Jun 2018 00:27:16 +0000 (08:27 +0800)]
Merge pull request #21921 from pdvian/wip-23982-luminous
luminous: qa: fix blacklisted check for test_lifecycle
Yan, Zheng [Sat, 2 Jun 2018 00:26:49 +0000 (08:26 +0800)]
Merge pull request #21901 from pdvian/wip-23951-luminous
luminous: mds: kick rdlock if waiting for dirfragtreelock
Yan, Zheng [Sat, 2 Jun 2018 00:26:28 +0000 (08:26 +0800)]
Merge pull request #21900 from pdvian/wip-23946-luminous
luminous: mds: crash when failover
Yan, Zheng [Sat, 2 Jun 2018 00:25:33 +0000 (08:25 +0800)]
Merge pull request #21874 from pdvian/wip-23936-luminous
luminous: cephfs-journal-tool: wait prezero ops before destroying journal
Yan, Zheng [Sat, 2 Jun 2018 00:24:58 +0000 (08:24 +0800)]
Merge pull request #21841 from pdvian/wip-23931-luminous
luminous: qa: remove racy/buggy test_purge_queue_op_rate
Yan, Zheng [Sat, 2 Jun 2018 00:24:31 +0000 (08:24 +0800)]
Merge pull request #21730 from joscollin/wip-23933-luminous
luminous: client: avoid second lock on client_lock
Yan, Zheng [Sat, 2 Jun 2018 00:23:21 +0000 (08:23 +0800)]
Merge pull request #21617 from pdvian/wip-23835-luminous
luminous: mds: fix occasional dir rstat inconsistency between multi-MDSes
Yan, Zheng [Sat, 2 Jun 2018 00:22:30 +0000 (08:22 +0800)]
Merge pull request #21600 from joscollin/wip-23475-luminous
luminous: ceph-fuse: trim ceph-fuse -V output
Yan, Zheng [Sat, 2 Jun 2018 00:21:41 +0000 (08:21 +0800)]
Merge pull request #21616 from joscollin/wip-23308-luminous
luminous: doc: Fix -d description in ceph-fuse
Yan, Zheng [Sat, 2 Jun 2018 00:19:58 +0000 (08:19 +0800)]
Merge pull request #21899 from pdvian/wip-23950-luminous
luminous: mds: trim log during shutdown to clean metadata
Yan, Zheng [Sat, 2 Jun 2018 00:15:10 +0000 (08:15 +0800)]
Merge pull request #21589 from pdvian/wip-23818-luminous
luminous: client: add client option descriptions
Yan, Zheng [Sat, 2 Jun 2018 00:13:38 +0000 (08:13 +0800)]
Merge pull request #21687 from batrick/i23638
luminous: ceph-fuse: getgroups failure causes exception
Jason Dillaman [Thu, 31 May 2018 13:29:00 +0000 (09:29 -0400)]
librbd: commit IO as safe when complete if writeback cache is disabled
We do not need to flush IO to ensure its safe if the writeback cache is
disabled when performing a journal replay. Instead, immediately mark the
IO as safe and let the journal's periodic commit throttle handle updating
the position.
Fixes: http://tracker.ceph.com/issues/23516
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit
a27269ab95e4cadcade2131367e40ccf1011316a )
Conflicts:
src/test/rbd_mirror/test_ImageReplayer.cc: removed "md_config_t::get_val" usage
Kefu Chai [Thu, 31 May 2018 15:04:14 +0000 (23:04 +0800)]
mon: enable level_compaction_dynamic_level_bytes for rocksdb
Fixes: http://tracker.ceph.com/issues/24361
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit
60f074fb0639d807e75b6a7858c5e13b5dc7c082 )
Volker Theile [Wed, 16 May 2018 07:52:51 +0000 (09:52 +0200)]
rgw: Do not modify email if argument is not set
The RGW Admin OPS API overwrites the users email either it is not set via argument, e.g. when executing /{admin}/user?uid=test5&suspended=true.
Signed-off-by: Volker Theile <vtheile@suse.com>
(cherry picked from commit
c6792a8be2fc1496e7a76b0e6b2579d32556b6b3 )
David Zafman [Thu, 31 May 2018 00:18:03 +0000 (17:18 -0700)]
osd: Handle omap and data digests independently
Caused by:
be078c8b7b131764caa28bc44452b8c5c2339623
The original attempt above to fix the omap_digest handling when
data_digest isn't present had 2 errors. First, it checked
is_data_digest() and is_omap_digest() instead of digest_present and
omap_digest_present which indicate the source digest is available.
Second, MAYBE could only be set if both digests are available.
Fixes: http://tracker.ceph.com/issues/24366
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit
01f9669928abd571e14421a51a749d44fa041337 )
David Zafman [Wed, 30 May 2018 18:47:04 +0000 (11:47 -0700)]
cleanup: Remove debug option osd_debug_scrub_chance_rewrite_digest
This option seems pointless and there are no test cases that use it.
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit
6adeaed32f70923d012bf9410bfa8651694be3cf )
xie xingguo [Sat, 30 Sep 2017 08:49:20 +0000 (16:49 +0800)]
osd/PrimaryLogPG: do not set data/omap digest blindly
As bluestore has bulitin csum, we generally no longer generate
object data digest for now. The consequence is that we should
handle data/omap digest more carefully to make certain ops,
such as copy_from/promote, to work properly since they heavily
relies on data digest for data transfer correctness.
Example of failure:
http://pulpito.ceph.com/xxg-2017-09-30_11:46:34-rbd-master-distro-basic-mira/
1690609 /
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit
be078c8b7b131764caa28bc44452b8c5c2339623 )
Conflicts:
src/osd/PrimaryLogPG.cc (still have snapdir)
Sage Weil [Mon, 21 May 2018 15:06:37 +0000 (10:06 -0500)]
os/bluestore: simplify and fix SharedBlob::put()
There is a narrow race possible:
A: lookup foo
A: put on foo
A: foo --nref == 0
B: lookup foo
B: put foo
B: foo --nref == 0
B: try_remove() succeeds, removes
A: try_remove() tries to remove foo again, probably crashes
We could fix this by flagging the object in some way to indicate it was
removed (maybe clearing parent?), but then we need to be careful about
dereferencing foo to get parent from put().
Fix this by moving to a simpler model: make lookup fail if nref == 0.
This eliminates the races around put() entirely because once nref reaches
0 it never goes up again.
Fixes: http://tracker.ceph.com/issues/24211
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit
8c8944b2c45ca9dc5b8fd4db1590e1d24206c0b3 )