Activity
From 01/10/2017 to 02/08/2017
02/08/2017
- 10:30 PM Bug #17981 (Resolved): teuthology.nuke kill_hadoop() function spams teuthology log
- 10:11 PM Revision d12d9bd4 (teuthology): Merge pull request #1018 from jan--f/wip-bootstrap-tumbleweed
- bootstrap: add support for opensuse tumbleweed.
- 10:10 PM Revision 0782513a (teuthology): Merge pull request #1021 from ceph/wip-suite-default
- teuthology-suite: Drop default for --ceph
- 10:00 PM Revision 8b9b406a (teuthology): Merge pull request #1004 from SUSE/wip-17981
- nuke: Use pkill -KILL to unconditionally wipe out hadoop processes
- 07:03 PM Bug #18858 (Resolved): ceph-ansible fails during monitor creation
- ceph branch: jewel
ansible version: 2.2.1
ceph-ansible task running to deploy ceph [jewel branch] fails during mo...
02/07/2017
- 08:22 PM Revision 317b059c (teuthology): Merge pull request #1026 from dmick/master
- prune: use the shortest time of -p or -r to decide on processing
- 08:07 PM Revision fa705c27 (teuthology): prune: use the shortest time of -p or -r to decide on processing
- invoking -p 7 -r 30 was only removing passed jobs 30 days or older.
Signed-off-by: Dan Mick <dan.mick@redhat.com> - 05:06 PM Revision e9f8143b (teuthology): Merge pull request #1025 from ceph/wip-console-force-into-spy
- console: force existing connections into spy mode if !readonly
- 04:54 PM Revision c8bea325 (teuthology): console: force existing connections into spy mode if !readonly
- If someone watching the console didn't think of using "console -s", we
end up power cycling the node in an attempt to... - 11:44 AM Bug #18597 (Resolved): Jobs stall when pcp.front.sepia.ceph.com is unresponsive
- 11:35 AM Bug #18597: Jobs stall when pcp.front.sepia.ceph.com is unresponsive
- https://github.com/ceph/teuthology/pull/1014
02/04/2017
- 10:22 AM Bug #18818 (New): All jobs scheduled using --sha1 show up "branch: master", even if they are {ham...
- When a user schedules a run using ...
02/03/2017
- 08:55 PM Revision dba73ea9 (teuthology): Merge pull request #1024 from ceph/wip-kclient-nuke-nosync
- nuke: work around a reboot -n trouble
- 05:16 PM Revision 004a7556 (teuthology): nuke: Tweak package removal
- Specifically, call remove_ceph_packages() before
remove_installed_packages(). Perhaps we ought to consolidate those
f... - 01:47 PM Revision b853abd0 (teuthology): nuke: improve stale_kernel_mount() check
- Commit 7db9e8b76fd5 ("nuke: bring stale kernel client handling back")
resurrected the check that was removed in commi... - 01:47 PM Revision 21ad5e75 (teuthology): nuke: work around a reboot -n trouble
- Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
- 01:17 AM Revision ee3c862b (teuthology): Add rgw role
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
02/02/2017
- 07:59 PM Revision 93c30f62 (teuthology): nuke: Notice if a remote actually rebooted
- If it didn't, attempt a power cycle.
Signed-off-by: Zack Cerza <zack@redhat.com> - 07:59 PM Revision ab258c93 (teuthology): PhysicalConsole: avoid a race in power_cycle()
- Apparently sometimes the console is still available immediately after
issuing a power cycle command. Sleep magically ... - 07:36 PM Revision f91754d1 (teuthology): Add Remote.uptime
- Signed-off-by: Zack Cerza <zack@redhat.com>
- 01:04 AM Bug #18790 (Won't Fix): ceph-ansible task fails with error "no module named six.moves"
- running ceph-ansible task against kraken branch fails with the following error...
- 01:00 AM Revision 68d2cfa2 (teuthology): changes for daemon-helper to work with systemd
- when ceph-ansible or ceph-deploy is used to setup
ceph cluster, use systemd to work with daemons
and kill/revive daem... - 01:00 AM Revision fad36d81 (teuthology): Add teardown for non vps nodes
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
- 01:00 AM Revision fd0b8127 (teuthology): for old downstream ansible versions add setup-clients
- old versions dont have clients role, use setup-clients
functions for those cases
Signed-off-by: Vasu Kulkarni <vasu@... - 01:00 AM Revision 65038230 (teuthology): Add redhat internal tasks that setup repo
- setup repo based on config items, either use
cdn or compose sytle urls or plain repo's.
Signed-off-by: Vasu Kulkarni... - 01:00 AM Revision 67b5907d (teuthology): Add internal kernel task to run on updated kernels
- always use updated distro kernel by default for regression
runs
Signed-off-by: Vasu Kulkarni <vasu@redhat.com> - 01:00 AM Revision 5aadb74d (teuthology): Add 'test-mode' config to skip ansible task
- when running with 'test-mode' config skip ansible
task which helps to debug any test task faster on
standalone nodes.... - 01:00 AM Revision cc5d2d54 (teuthology): fix module naming
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
- 01:00 AM Revision e97ef510 (teuthology): import names in init
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
- 01:00 AM Revision aa07ff2d (teuthology): few ceph-ansible fixes
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
- 01:00 AM Revision e1a28833 (teuthology): Fix task overrides for task which use '_' instead of '-'
- task in yaml use '-' instead of original '_'
fix overrides to use same name - 01:00 AM Revision 1ec5ede9 (teuthology): use playbook that doesnt support clients role for old versions
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
- 01:00 AM Revision 37602aea (teuthology): fix typo for config item in ceph-ansible
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
- 01:00 AM Revision 3dad38e7 (teuthology): typo rhbulid to rhbuild
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
- 01:00 AM Revision 7dbe9005 (teuthology): fix installer node name
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
02/01/2017
- 11:58 PM Revision 0e47d0ae (teuthology): Merge pull request #1022 from ceph/wip-unbreak-kclient-nuke
- nuke: bring stale kernel client handling back
- 08:26 PM Revision 7db9e8b7 (teuthology): nuke: bring stale kernel client handling back
- Commit 1d47a121b385 ("Fix nuke, redo some cleanup functions") broke
stale kernel client map/mount handling by droppin... - 08:26 PM Revision 610fdc8e (teuthology): nuke: drop remove_kernel_mounts()
- Calling remove_kernel_mounts() after reboot() is pretty useless. Also,
as explained in the previous commit, there is...
01/27/2017
- 06:42 PM Bug #18575 (Resolved): Queued job fails to start then kills worker
- https://github.com/ceph/teuthology/pull/1020
- 06:19 PM Bug #18701 (New): failed to install new kernel version
- when trying to run ceph-ansible suite in teuthology, the job failed with error "failed to install new kernel version"...
01/26/2017
- 06:17 PM Bug #18688: teuthology-suite rerun option won't work with -S arg
- Nathan, AFAIK you must have @-c@ and I think the reason is that we query yaml configurations from the repo as oppose ...
- 06:14 PM Bug #18688: teuthology-suite rerun option won't work with -S arg
- and that SHA1 matches the one in teuthology-2017-01-25_16:42:38-powercycle-master-testing-basic-mira
Yuri, what ha... - 06:11 PM Bug #18688: teuthology-suite rerun option won't work with -S arg
- Zack, he's passing -S 83af8cdaaa6d94404e6146b68e532a784e3cc99c
- 05:14 PM Bug #18688: teuthology-suite rerun option won't work with -S arg
- Are you saying it's work-as-designed and @-r@ won't work with @-S@? And only work with @-c@ ?
- 04:53 PM Bug #18688: teuthology-suite rerun option won't work with -S arg
- My last comment was incorrect; your variable names are pretty confusing.
The real issue is in the first invocation... - 04:41 PM Bug #18688 (Closed): teuthology-suite rerun option won't work with -S arg
- @-S@ is @--sha1@ - you appear to be passing a branch...
- 04:37 PM Bug #18688 (Closed): teuthology-suite rerun option won't work with -S arg
- compare
@CEPH_BRANCH=83af8cdaaa6d94404e6146b68e532a784e3cc99c; MACHINE_NAME=mira; teuthology-suite -v -S $CEPH_BR...
01/25/2017
- 10:35 PM Revision ab0bf27b (teuthology): teuthology-suite: Drop default for --ceph
- Signed-off-by: Zack Cerza <zack@redhat.com>
- 09:23 PM Revision b91a954b (teuthology): Merge pull request #1020 from ceph/wip-waitpid
- Tell gevent not to patch os.waitpid()
Reviewed-by: Dan Mick <dmick@redhat.com> - 09:08 PM Revision a324f457 (teuthology): Tell gevent not to patch os.waitpid()
- Signed-off-by: Zack Cerza <zack@redhat.com>
- 01:38 PM Revision fbc17402 (teuthology): bootstrap: add support for opensuse tumbleweed.
- Signed-off-by: Jan Fajerski <jfajerski@suse.com>
01/24/2017
- 11:07 PM Revision 0858e0df (teuthology): Pin gevent to < 1.2
- On the off-chance that it's responsible for the "phantom job" issue
Signed-off-by: Zack Cerza <zack@redhat.com> - 09:59 PM Bug #18656 (Resolved): broken teuthology - "No module named manhole"
- 09:59 PM Bug #18656: broken teuthology - "No module named manhole"
- and fixed @teuthology@ user too
- 09:47 PM Bug #18656: broken teuthology - "No module named manhole"
- deleted @virtualenv@ and run @bootstrap@ for @yuriw@ user and it worked
- 09:39 PM Bug #18656 (Resolved): broken teuthology - "No module named manhole"
- As @teuthology@ and as @yuriw@ user =>...
- 07:27 PM Revision 92201ccd (teuthology): Merge pull request #1017 from ceph/wip-manhole
- Use manhole to provide a way to debug hung jobs
Reviewed-by: Dan Mick <dmick@redhat.com> - 06:56 PM Revision 4a064bbb (teuthology): Use manhole to provide a way to debug hung jobs
- https://pypi.python.org/pypi/manhole
Signed-off-by: Zack Cerza <zack@redhat.com> - 05:42 PM Revision edcda794 (teuthology): Merge pull request #1016 from jcsp/wip-jewel-valgrind
- jewel: backport valgrind suppression file updates from master
- 04:47 PM Revision 9990e1f4 (teuthology): Merge pull request #1014 from jcsp/wip-18594
- pcp: use a timeout when downloading graphite graphs
- 04:32 PM Revision a728ad62 (teuthology): valgrind.supp: xenial libstdc++ leak
- Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit a2b149be17b56276c6715bcf9c49a93b9bda118e) - 04:32 PM Revision 4d9d7ccd (teuthology): valgrind: suppression for md_config_t string leak
- Fixes: http://tracker.ceph.com/issues/17924
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from com... - 04:32 PM Revision 0840bfeb (teuthology): valgrind: xenial msync whitelist
- Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 25e496922b1c56193b40acae75e970cb2dd2c249)
01/23/2017
01/20/2017
- 10:28 PM Bug #17986 (Resolved): orchestra.console.spawn_sol_log expects to find pexpect in system Python
- https://github.com/ceph/teuthology/pull/1003
- 09:22 AM Revision adc1beff (teuthology): pcp: use a timeout when downloading graphite graphs
- Fixes: http://tracker.ceph.com/issues/18597
Signed-off-by: John Spray <john.spray@redhat.com> - 09:21 AM Bug #18597 (Resolved): Jobs stall when pcp.front.sepia.ceph.com is unresponsive
- We should be applying a timeout so that we skip past attempting to download charts when the server is unhealthy.
- 01:17 AM Bug #18575: Queued job fails to start then kills worker
- Found some more
teuthology-2017-01-16_11:00:03-rbd-kraken-distro-basic-smithi/721193...
01/17/2017
- 06:19 PM Bug #18575: Queued job fails to start then kills worker
- May be a residual failure from this problem: http://tracker.ceph.com/issues/18482
Will leave this ticket open for ... - 05:36 PM Bug #18575: Queued job fails to start then kills worker
- I touched ...
- 05:33 PM Bug #18575 (Resolved): Queued job fails to start then kills worker
- I noticed the number of workers drop on the Grafana dashboard. I correlated the time the worker count decreased to t...
- 05:41 PM Revision bf2dda6d (teuthology): Merge pull request #1011 from ceph/wip-kernel-kill-vsplitter
- kernel: be more flexible about sha1 matching
- 02:51 PM Revision a1807631 (teuthology): kernel: be more flexible about sha1 matching
- Some rpm scripts don't allow dashes in the Release field, so let's
accept both -g and _g. Kill _vsplitter() as Calxe... - 12:45 AM Revision 88c22159 (teuthology): Dont set qa-suite repo during worker startup
- There is no need to initialize qa-suite repo during worker startup,
moreover it fails with default values trying to c... - 12:25 AM Revision f5d9712f (teuthology): Use clock by default (instead of clock.check)
- We're seeing clocks desynchronized. My theory is that this might
be because ntp can take five minutes or more to act... - 12:25 AM Revision ca2643e2 (teuthology): clock: Fix clock's use of Remote.run()
- Signed-off-by: Dan Mick <dan.mick@redhat.com>
- 12:19 AM Revision ed1a1e9a (teuthology): Merge pull request #1009 from ceph/wip-archive
- Avoid a race condition with job archive creation
Reviewed-by: Dan Mick <dmick@redhat.com>
01/16/2017
- 11:52 PM Revision cf07393d (teuthology): Fix a conflict with openstack requirements
- Signed-off-by: Zack Cerza <zack@redhat.com>
- 11:16 PM Revision d59c7221 (teuthology): worker: Create job archive directories
- ... not just run archive directories. This is to resolve a race
condition between the job creating its archive direct... - 11:14 PM Revision 85888aef (teuthology): run: Don't fail if the archive dir exists
- Signed-off-by: Zack Cerza <zack@redhat.com>
01/14/2017
- 01:07 AM Revision 45dcba28 (teuthology): for old downstream ansible versions add setup-clients
- old versions dont have clients role, use setup-clients
functions for those cases
Signed-off-by: Vasu Kulkarni <vasu@... - 01:07 AM Revision 37dd17c3 (teuthology): Add teardown for non vps nodes
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
- 01:03 AM Revision 3fb81198 (teuthology): Add redhat internal tasks that setup repo
- setup repo based on config items, either use
cdn or compose sytle urls or plain repo's.
Signed-off-by: Vasu Kulkarni...
01/13/2017
- 11:18 PM Bug #18510 (Closed): ceph-ansible task needs fix in the way key creation is done
- issue 18528 explains it better
- 09:34 PM Bug #18510: ceph-ansible task needs fix in the way key creation is done
- currently, the task ceph_ansible has client.admin hardcoded and so it fails to create keyring files for client.*
- 11:16 PM Bug #18528 (New): ceph-ansible task does not create client keyrings yet
- ceph-ansible task currently assumes only client.admin.keyring as the client keyring file.
rather, it should be abl... - 07:10 PM Revision 0981a7f1 (teuthology): Changes for daemon-helper to work with systemd
- when cluster is setup using ceph-ansible or ceph-deploy
use systemd commands to kill/revive deamons.
Signed-off-by: ... - 06:36 PM Bug #18420: Wrong interaction of os_version provided on command line with os_type test definition...
- Trying with ...
01/12/2017
- 11:55 PM Bug #18510 (Closed): ceph-ansible task needs fix in the way key creation is done
- ceph-ansible task should be able to add key creation, the way ceph task currently does it.
currently, it works for... - 02:01 AM Bug #18492: upgrade:hammer-jewel-x fails on xenial b/c of different packages
- packages like rbd-mirror, ceph-mgr, etc. don't exist in hammer. We need to mention them explicitly on xenial to be ab...
- 12:57 AM Bug #18492 (New): upgrade:hammer-jewel-x fails on xenial b/c of different packages
- Logs: http://qa-proxy.ceph.com/teuthology/yuriw-2017-01-10_23:01:29-upgrade:hammer-jewel-x-kraken-distro-basic-smithi...
01/10/2017
- 10:47 PM Revision 7062a4c9 (teuthology): Merge pull request #1006 from ceph/wip-cd-debug
- Remove high level of debug logging which is not required
- 05:45 PM Revision c79dff7c (teuthology): Remove high level of debug logging which is not required
- Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
Also available in: Atom