RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-02-24 11:37:37 +00:00
Code Issues Packages Projects Releases Wiki Activity
75,078 Commits 216 Branches 395 Tags 1.4 GiB
cb2cc82d93
Commit Graph

2302 Commits

Author SHA1 Message Date
Sage Weil
cb2cc82d93 Merge pull request #16239 from xiexingguo/wip-crush-weight-overflow 
mon: a few more upmap (and other) fixes

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-07-11 12:56:15 -05:00
xie xingguo
5f9de0e611 mon/OSDMonitor: sanity check osd number before applying pg-temp 
In case user did something stupid and causing weird results.

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-07-11 02:14:06 +08:00
Sage Weil
b5f81e7313 qa/tasks/ceph.py: no osd id to 'osd create' command 
This isn't recognized by hammer, and we don't need it for jewel.

Fixes: http://tracker.ceph.com/issues/20548
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-10 10:43:16 -04:00
John Spray
44bce9e02c Merge pull request #15858 from liewegas/wip-mgr-servicemap 
mgr,librados: service map

Reviewed-by: John Spray <john.spray@redhat.com>
 2017-07-10 15:03:07 +01:00
xie xingguo
7d07356383 crush: fix potential weight overflow 
E.g.:
./bin/ceph osd crush reweight osd.0 32768

ID WEIGHT      TYPE NAME                                            UP/DOWN REWEIGHT PRIMARY-AFFINITY
-4 32770.00000 root default~hdd
-3           -     host gitbuilder-ceph-rpm-centos7-amd64-basic~hdd
 0           -         osd.0                                             up  1.00000          1.00000
 1     1.00000         osd.1                                             up  1.00000          1.00000
 2     1.00000         osd.2                                             up  1.00000          1.00000
-1 32770.00000 root default
-2           -     host gitbuilder-ceph-rpm-centos7-amd64-basic
 0           -         osd.0                                             up  1.00000          1.00000
 1     1.00000         osd.1                                             up  1.00000          1.00000
 2     1.00000         osd.2                                             up  1.00000          1.00000

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-07-10 11:23:36 +08:00
Sage Weil
aabe0e35a0 librados: allow service registrations 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-09 14:09:04 -04:00
Sage Weil
8b21c6b6fd Merge pull request #16027 from liewegas/wip-crush-rule-class 
mon,crush: create crush rules using device classes for replicated and ec pools via cli

Reviewed-by: Loic Dachary <ldachary@redhat.com>
 2017-07-08 21:34:13 -05:00
Sage Weil
0c79c4ffac Merge pull request #16228 from smithfarm/wip-rados-upgrade-2 
tests: fix rados/upgrade/jewel-x-singleton and make workunit task handle repo URLs not ending in ".git"

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-07-08 21:32:36 -05:00
Nathan Cutler
98eda9a4dc qa: workunit: handle repo ending in "/ceph-ci" 
If the repo ends in "/ceph-ci" it's the same as if it ended in "/ceph-ci.git"

Before this change, the following command was broken if the workunit specified,
e.g., "branch: jewel":

teuthology-suite --ceph-repo https://github.com/ceph/ceph --ceph master
--suite-repo https://github.com/ceph/ceph-ci --suite-branch wip-foo . . .

Fixes: http://tracker.ceph.com/issues/20554
Signed-off-by: Nathan Cutler <ncutler@suse.com>
 2017-07-08 19:03:30 +02:00
Sage Weil
4bc9f566d0 qa/suites/rados/upgrade: upgrade client.0 node too 
Fixes: http://tracker.ceph.com/issues/20368
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Nathan Cutler <ncutler@suse.com>
 2017-07-08 18:56:09 +02:00
Sage Weil
a0ba660afc Merge pull request #16025 from liewegas/wip-19964 
mon: do crushtool test with fork and timeout, but w/o exec of crushtool

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
 2017-07-08 09:01:10 -05:00
Sage Weil
f17516656e Merge pull request #16212 from dillaman/wip-qa-ceph-task 
qa/tasks/ceph: don't hard-code cluster name when copying fsid

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-07-08 08:54:18 -05:00
Sage Weil
704d13d36f Merge pull request #16213 from liewegas/wip-radosbench-timeout 
qa/tasks/radosbench: longer timeout
 2017-07-08 08:53:55 -05:00
Kefu Chai
fc4d51d603 Merge pull request #16225 from liewegas/wip-mon-auth-caps 
qa/suites/rados/singleton/all/mon-auth-caps: more osds so we can go clean

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-07-08 09:58:21 +08:00
Sage Weil
e30b32bca4 qa/suites/rados/singleton/all/mon-auth-caps: more osds so we can go clean 
and scrub

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-07 17:39:22 -04:00
Vasu Kulkarni
71e0922cfa create rbd pool since its not created by default anymore 
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
 2017-07-07 09:23:43 -07:00
Jason Dillaman
a06f8844c7 qa/tasks/ceph: supply cluster name to ceph CLI 
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2017-07-07 11:21:51 -04:00
Vasu Kulkarni
9b81349985 create mgr nodes explicitly now, previously they where 
started alongside monitor daemons due to systemd rules.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
 2017-07-07 08:16:21 -07:00
Sage Weil
65d9d667ec qa/workunits/cephtool/test.sh: remove two crushtool validation tests 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-07 11:11:25 -04:00
Sage Weil
597a971686 qa/tasks/radosbench: longer timeout 
Hit another timeout due to aggressive thrashing with size==min_size.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-07 11:04:36 -04:00
Jason Dillaman
5a8967cde7 qa/tasks/ceph: don't hard-code cluster name when copying fsid 
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
 2017-07-07 10:44:20 -04:00
Patrick Donnelly
8f121c70d3
Merge remote-tracking branch 'upstream/pull/15844/head' into master 
* commit '47ab651a1f':
  qa: reset journal before cephfs_journal_tool_smoke.sh exits
  qa: remove 'apply' from journal_tool_smoke.sh

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
 2017-07-06 22:16:02 -07:00
Patrick Donnelly
5b87301192
Merge remote-tracking branch 'upstream/pull/15822/head' into master 
* upstream/pull/15822/head:
  qa: add timeout/repeat to pool df

Reviewed-by: John Spray <jspray@redhat.com>
 2017-07-06 22:14:32 -07:00
Patrick Donnelly
77b707a6ea
Merge remote-tracking branch 'upstream/pull/15950/head' into master 
* upstream/pull/15950/head:
  qa: update thrash max mds testing

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
 2017-07-06 22:11:18 -07:00
Patrick Donnelly
64c6079d69
Merge remote-tracking branch 'upstream/pull/15937/head' into master 
* upstream/pull/15937/head:
  qa: remove unused quota config option

Reviewed-by: John Spray <jspray@redhat.com>
 2017-07-06 21:38:45 -07:00
Patrick Donnelly
97cdb1e34a
Merge remote-tracking branch 'upstream/pull/15817/head' into master 
* upstream/pull/15817/head:
  qa: wait for healthy cluster before testing pins

Reviewed-by: Zheng Yan <zyan@redhat.com>
 2017-07-06 21:36:34 -07:00
Patrick Donnelly
2cb42a4dbf
Merge remote-tracking branch 'upstream/pull/13770/head' into master 
* upstream/pull/13770/head:
  tasks/cephfs: add TestStrays.test_replicated_delete_speed

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
 2017-07-06 21:33:03 -07:00
Sage Weil
5870a68c00 Merge pull request #15647 from gmayyyha/format-config-show 
common,config: OPT_FLOAT and OPT_DOUBLE output format in config show

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-07-06 21:25:19 -05:00
Sage Weil
7f18184f63 qa/workunits/mon/crush_ops.sh: require luminous before using device classes 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-06 15:01:03 -04:00
Sage Weil
8fb2517057 erasure-code: implement 'crush-device-class' ec profile property 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-06 15:01:03 -04:00
Sage Weil
dc7a2aaf7a erasure-code: ruleset-* -> crush-* 
1) ruleset is an obsolete term, and
2) crush-{rule,failure-domain,...} is more descriptive.

Note that we are changing the names of the erasure code profile keys
from ruleset-* to crush-*.  We will update this on upgrade when the
luminous flag is set, but that means that during mon upgrade you cannot
create EC pools that use these fields.

When the upgrade completes (users sets require_osd_release = luminous)
existing ec profiles are updated automatically.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-06 15:01:03 -04:00
Sage Weil
5a4ba443c4 mon,crush: 'osd crush rule create-replicated ...' 
New command to create crush rule that specifies a class of device.  Plus
all of the fallout in other callers to the CrushWrapper helpers, the
crushtool cli change, and cli test.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-06 15:00:37 -04:00
Sage Weil
d02c9a5b01 Merge pull request #16122 from tchaikov/wip-qa-warning-msg 
qa/tasks/ceph: should be "Waiting for all PGs", not "all osds"

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-07-05 10:26:56 -05:00
Mykola Golub
2a9f56f818 Merge pull request #15860 from dillaman/wip-20168 
librbd: fail IO request when exclusive lock cannot be obtained

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
 2017-07-05 14:52:55 +03:00
Kefu Chai
2cae544cf2 qa/tasks/ceph: should be "Waiting for all PGs", not "all osds" 
amend the warning message

Signed-off-by: Kefu Chai <kchai@redhat.com>
 2017-07-05 18:58:58 +08:00
Kefu Chai
04e0ef541d Merge pull request #15754 from tchaikov/wip-test-auth-caps 
qa/suites: add test exercising workunits/mon/auth_caps.sh

Reviewed By: Brad Hubbard <bhubbard@redhat.com>
Reviewed-by: Joao Eduardo Luis <joao@suse.de>
 2017-07-05 15:05:21 +08:00
John Spray
b363e58530 Merge pull request #15958 from liewegas/wip-mgr-modules 
mgr,mon: enable/disable mgr modules via 'ceph mgr module ...' commands

Reviewed-by: John Spray <john.spray@redhat.com>
 2017-07-05 00:04:59 +01:00
Mykola Golub
866cf72440 Merge pull request #15956 from dillaman/wip-librbd-devstack 
test: fix failing rbd devstack teuthology test

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
 2017-07-01 15:02:29 +03:00
Mykola Golub
19fbe87791 Merge pull request #15923 from dillaman/wip-qa-rbd-memcheck-python 
qa/suites/rbd: restrict python memcheck validation to CentOS

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
 2017-07-01 11:24:05 +03:00
Sage Weil
c9c697225f Merge pull request #16044 from liewegas/wip-i-cant-believe-its-not-btrfs 
qa: stop testing btrfs

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2017-06-30 21:17:49 -05:00
Sage Weil
aa986f58ef Merge pull request #16046 from liewegas/wip-valgrind-on-centos 
do all valgrind runs on centos

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2017-06-30 21:17:21 -05:00
Sage Weil
1025d7a5aa Merge pull request #15959 from smithfarm/wip-rgw-singleton-cleanup 
tests: rgw/singleton: drop duplicate filestore-xfs.yaml

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2017-06-30 09:09:53 -05:00
Sage Weil
c7893283cd do all valgrind runs on centos 
We are fighting two issues with valgrind on ubuntu (xenial, yakkety,
and z):

	http://tracker.ceph.com/issues/18126
	http://tracker.ceph.com/issues/20360

Revert this when it is fixed.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-30 09:33:18 -04:00
Sage Weil
5a2d78ad85 qa: stop testing btrfs 
This is too noisy and we don't recommend it anyway.

See http://tracker.ceph.com/issues/20169

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-30 09:28:03 -04:00
Sage Weil
d953198aff mon,mgr: track available modules in MgrMap too 
Require --force to enable a module that isn't reported as available.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-29 23:10:51 -04:00
Sage Weil
16fcee1f71 mon,mgr: manage mgr module list in mgrmap 
Kill old mgr_modules option.

Add new mgr_initial_modules option, on the mon, for the initial cluster
mgrmap.

Add ls, enable, disable commands.

Respawn mgr if the module list changes.  In the future we could enable
new modules without a full restart, but disabling probably requires (and
is best handled by) a respawn.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-06-29 23:10:51 -04:00
Sage Weil
7fe818990a Merge pull request #15968 from xiexingguo/wip-poolopts-fix 
mon/OSDMonitor: two pool opts related fix

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-06-29 22:03:02 -05:00
Kefu Chai
7ee4907698 Merge pull request #15925 from jcsp/wip-fsstatus-cleanup 
mgr: clean up fsstatus module

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-06-30 09:31:02 +08:00
Josh Durgin
4bcd6f666b Merge pull request #14760 from dzafman/wip-19657 
osd: On EIO from read recover the primary replica from another copy

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2017-06-29 14:00:18 -07:00
John Spray
623f1240a2 tasks/cephfs: add TestStrays.test_replicated_delete_speed 
Reproducer for http://tracker.ceph.com/issues/16914

Signed-off-by: John Spray <john.spray@redhat.com>
 2017-06-29 17:21:57 +01:00