Commit Graph

4892 Commits

Author SHA1 Message Date
John Spray
d1c3ef1fee tasks/cephfs: pg_num check in volume_client test
Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
fb2e38f2df tasks/cephfs: fix raising from RemoteProcess.wait()
...in the case where we call wait() on an already-finished
process.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
55a04930e3 tasks/cephfs: fix vstart's LocalDaemon.proc
This class is meant to quack like a DaemonState,
so it needs to expose a .proc attribute.  This
was breaking TestDamageTable.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
5c525219c4 tasks/cephfs: fix wait in test_volume_client
Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
b06818cf4b suites: add an extra client to fs/recovery
The volumeclient test wants 3, and the rest of the tests
can just ignore it.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
5055f1288f suites: add yaml for TestVolumeClient
Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
627c33054c tasks/cephfs: add test_volume_client
Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
523d84b5f0 tasks/cephfs: fix fuse killing in vstart
The "ps au" output was truncating lines which could cause
processes to get missed in some cases, like where there
is a long --client-mountpoint argument.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
cb699cccfe tasks/cephfs: write_n_mb optionally background
Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
b94b35b68f tasks/cephfs: reset IDs of mounts between tests
because volumeclient test twiddles the IDs to
match the auth identities it creates.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
7a19b6dc0d tasks/cephfs: clear extra auth ids on test start
Because new volumeclient tests will be creating IDs
instead of modifying the default ones.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
John Spray
addf313c24 tasks/cephfs: add test case auth_list helper
Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-14 22:55:29 +00:00
Gregory Farnum
3b37c245b3 Merge pull request #743 from ceph/wip-root-metadata
tasks/cephfs: add TestSessionMap.test_session_reject

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2016-01-14 10:37:04 -08:00
Gregory Farnum
7781bf071d Merge pull request #611 from ceph/wip-libcephfs-acl
suites/fs: test ceph-fuse with libcephfs permission check

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2016-01-14 10:34:20 -08:00
Zack Cerza
e4ba609ca7 Merge pull request #796 from ceph/wip-fix-sumblink-distros
Fixed symblink for os configs
2016-01-13 17:10:10 -07:00
Yuri Weinstein
0915747895 Fixed symblink for os configs
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
2016-01-13 16:01:40 -08:00
Zack Cerza
ec44980913 Merge pull request #754 from ceph/wip_add_centos71_master
Removed CentOS 7.1 and added CentOS 7.2 instead
2016-01-13 16:29:00 -07:00
Yuri Weinstein
5a79aa649f Change supported ditros to use latest CentOS 7.2
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
2016-01-13 13:59:55 -08:00
Andrew Schoen
743c8f28c4 Merge pull request #793 from ceph/wip-centos-71-72
suites/teuthology: Add CentOS 7.1 and 7.2
2016-01-12 16:06:25 -06:00
Zack Cerza
be35a7bfc8 suites/teuthology: Add CentOS 7.1 and 7.2
Signed-off-by: Zack Cerza <zack@redhat.com>
2016-01-12 12:20:26 -07:00
John Spray
7d9c9609b0 suites/fs: update log whitelist for test_session_reject
Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-11 14:32:43 +00:00
John Spray
aafa06eb9d tasks/cephfs: generate mount paths on the fly
...because new tests that modify the client ID will
get confused about whether they're mounted or not, otherwise.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-11 14:04:05 +00:00
John Spray
acff449c04 tasks/cephfs: support mount_path in Mount subclasses
For e.g. volumeclient guests that want to mount
a subpath.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-11 13:26:58 +00:00
Yan, Zheng
6082cb2b71 Merge pull request #785 from ceph/wip-14257
tasks/cephfs: fix wait_for_state timing
2016-01-07 22:22:32 +08:00
John Spray
58c13c1a28 tasks/cephfs: fix wait_for_state timing
Return actual elapsed wallclock time instead of
the number of times we polled.

Fixes: #14257
Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-07 11:24:57 +00:00
Greg Farnum
8c23e04b1e Merge remote-tracking branch 'origin/jewel' 2016-01-06 13:50:31 -08:00
Gregory Farnum
a0b365b220 Merge pull request #780 from ceph/jewel-14196
tasks/cephfs: fix race in test_damage
2016-01-06 13:52:09 -08:00
Greg Farnum
eb1d4f607c Merge remote-tracking branch 'origin/jewel-14196' 2016-01-05 23:11:32 -08:00
Greg Farnum
8488a991a9 Merge remote-tracking branch 'origin/wip-13583' 2016-01-05 22:57:51 -08:00
John Spray
26423e9574 tasks/cephfs: fix wait in assert_cluster_log
I had the monc tick interval wrong.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-05 18:58:01 +00:00
John Spray
a6970271be tasks/cephfs: fixes to FuseMount
This was getting stressed in new ways by
TestSessionMap.test_session_reject, which
has a mount that fails initially.

Two changes here:
 * Raise CommandFailedError instead of RuntimeError when
   a mount fails (i.e. catch process termination instead
   of timing out on /sys/ population)
 * Generalise error handling on umount, so that we only
   raise the exception on an umount failure if the mount
   appears to really not be unmounted.  There is some
   EINVAL corner case that was getting triggered by the test.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-05 18:58:01 +00:00
John Spray
141de6e1b4 tasks/cephfs: add TestSessionMap.test_session_reject
Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-05 18:58:01 +00:00
John Spray
3b9b68b176 tasks/cephfs: add assert_cluster_log
For capturing the cluster log and asserting
that we see a particular expected message.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-05 18:58:00 +00:00
John Spray
d8106fa9e1 tasks: add run_ceph_w to CephManager
Analogous to raw_cluster_command, but instead
of calling blocking CLI command we're invoking
the -w mode.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-05 18:58:00 +00:00
John Spray
8b5edd4fe9 tasks/cephfs: new hooks for auth keys and config
Used when configuring clients with dynamically
generated auth keys, and pointing them at mount paths.

Signed-off-by: John Spray <john.spray@redhat.com>
2016-01-05 18:57:33 +00:00
David Zafman
7d2bbc900d Merge pull request #775 from ceph/wip-stat-error
repair_test: Add whitelist item for new stat error message
2016-01-04 15:25:27 -08:00
Tamilarasi Muthamizhan
af4a5ae4f7 Merge pull request #781 from ceph/wip_remove_dumpling
Removed tests for EOL dumpling clients on jewel clusters
2016-01-04 14:40:35 -08:00
Yuri Weinstein
66d244fe8b Removed tests for EOL dumpling clients on jewel clusters
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
2016-01-04 09:33:46 -08:00
Yan, Zheng
b2a304e651 tasks/cephfs: fix race in test_damage
When stray directory inodes are corrupted, MDS may go to damaged state
after becoming active. (MDCache::open_root/populate_mydir is called by
MDSRank::starting_done).

Fixes: #14196
Signed-off-by: Yan, Zheng <zyan@redhat.com>
2016-01-04 20:42:35 +08:00
Sage Weil
926a8cbb2a tasks/osd_recovery: fewer objects in test_incomplete_pgs
Otherwise we risk filling up a (fast) disk.

Signed-off-by: Sage Weil <sage@redhat.com>
2016-01-02 09:29:28 -05:00
Sage Weil
04e41a608b rados: 5m radosbench (not 10m)
Signed-off-by: Sage Weil <sage@redhat.com>
2016-01-01 09:35:31 -05:00
Sage Weil
6110421bce rados: run radosbench for 10m (not 15m)
Signed-off-by: Sage Weil <sage@redhat.com>
2015-12-30 09:43:59 -05:00
Sage Weil
07232b1e2b rados/singleton/all/pg-removal-interruption: ensure osd.0 is marked down
We were waiting for teh dameon to fail, not for the mons to notice,
which meant that after we restart the daemon the next tell could
try to talk to the old osd.

Signed-off-by: Sage Weil <sage@redhat.com>
2015-12-24 09:57:52 -05:00
Josh Durgin
c7cd4f744b rbd: increase length of non-valgrind fsx workloads
New lab machines make these run much faster. These should take ~1-1.5 hours now.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
2015-12-23 16:59:46 -08:00
Josh Durgin
ef25ba3920 rbd: add helgrind to valgrind subsuite
This checks locking problems. This can be a separate facet for
workunits, but for now it requires a manual 2nd configuration for
rbd_fsx.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
2015-12-23 16:59:46 -08:00
Josh Durgin
888ff87da0 rbd: add a couple new workunits
for rbd-nbd and the journalling cli

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
2015-12-23 16:59:46 -08:00
Josh Durgin
8bd0b9dabf rbd: add some workloads with journalling enabled
Separating features into its own facet would be good,
but requires changing some of the tests a bit.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
2015-12-23 16:59:46 -08:00
Josh Durgin
5864d56592 rbd: only test on xfs
btrfs is not recommended for production

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
2015-12-23 16:59:46 -08:00
Sage Weil
7f446f1a2d Merge remote-tracking branch 'gh/jewel' 2015-12-23 17:07:41 -05:00
Sage Weil
179ff64277 Merge pull request #755 from guangyy/wip-13441-test
pg: add test cases for pg list_missing on EC pool
2015-12-23 17:06:39 -05:00