]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Yehuda Sadeh [Fri, 21 Dec 2012 18:20:02 +0000 (10:20 -0800)]
task/swift: change upstream repository url
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Samuel Just [Tue, 11 Dec 2012 22:21:48 +0000 (14:21 -0800)]
CephManager: add ability to test split
Signed-off-by: Samuel Just <sam.just@inktank.com>
Joe Buck [Thu, 6 Dec 2012 22:17:16 +0000 (14:17 -0800)]
pexec.py: Parse out role ID from the back.
Also, do not assume that the command needs to run from a specific directory.
Signed-off-by: Joe Buck <jbbuck@gmail.com>
Joe Buck [Thu, 6 Dec 2012 22:19:55 +0000 (14:19 -0800)]
Adding a Hadoop task.
This task configures and starts a Hadoop cluster.
It does not run any jobs, that must be done after
this task runs.
Can run on either Ceph or HDFS.
Signed-off-by: Joe Buck <jbbuck@gmail.com>
Joe Buck [Thu, 6 Dec 2012 22:18:41 +0000 (14:18 -0800)]
New ssh task that adds keys for node -> node ssh.
This generates a new keypair, pushes it to all nodes
in the context and adds all hosts to all other hosts
.ssh/authorized_keys file.
Cleans up all keys and authorized_keys entries
afterwards.
Signed-off-by: Joe Buck <jbbuck@gmail.com>
Reviewed-by: Sam Lang <sam.lang@inktank.com>
Samuel Just [Mon, 10 Dec 2012 22:33:41 +0000 (14:33 -0800)]
ceph.conf: default to smaller recovery chunk
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin [Wed, 5 Dec 2012 22:59:27 +0000 (14:59 -0800)]
qemu: set qemu cache mode based on rbd cache setting
If we don't do this, qemu assumes no caching is used and doesn't send flushes.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
Joao Eduardo Luis [Thu, 29 Nov 2012 00:53:59 +0000 (00:53 +0000)]
Merge branch 'wip-mon-thrasher'
Joao Eduardo Luis [Thu, 22 Nov 2012 16:44:39 +0000 (16:44 +0000)]
task: mon_thrash: thrash monitors while running other tests
Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
Sage Weil [Sun, 25 Nov 2012 16:37:06 +0000 (08:37 -0800)]
run: save original config, too
Sage Weil [Thu, 22 Nov 2012 21:59:58 +0000 (13:59 -0800)]
s3tests: fix typo
Sage Weil [Wed, 21 Nov 2012 16:29:47 +0000 (08:29 -0800)]
workunit: fix indentation
Josh Durgin [Tue, 20 Nov 2012 22:01:03 +0000 (14:01 -0800)]
xfstests: run in parallel on multiple machines
xfstests itself still seems to have some global dependencies that
make it hard to run more than one instance per node, so keep
the one client per node restriction.
Name the image after the client using it, and only run the
nested context managers once, so this task can work with
more than one client.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
Yehuda Sadeh [Tue, 20 Nov 2012 00:19:06 +0000 (16:19 -0800)]
rgw-logsocket: a task to verify opslog socket works
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sam Lang [Tue, 20 Nov 2012 15:53:52 +0000 (09:53 -0600)]
task/pexec: Output stderr to teuthology log
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sam Lang [Fri, 16 Nov 2012 15:58:01 +0000 (09:58 -0600)]
task/ceph-fuse: Add log messages for abort
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sage Weil [Sun, 18 Nov 2012 17:24:10 +0000 (09:24 -0800)]
workunit: fix default subdir
Make subdir argument optional.
Sage Weil [Sun, 18 Nov 2012 00:19:14 +0000 (16:19 -0800)]
valgrind: enumerate warnings in log; check leaks from client, mon only
Mike Ryan [Thu, 15 Nov 2012 21:07:42 +0000 (13:07 -0800)]
task: benchmark recovery
Measures latency before and during recovery using smalliobench.
Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
Sander Pool [Thu, 15 Nov 2012 01:24:50 +0000 (17:24 -0800)]
Starting to auto-document this code.
Sander Pool [Thu, 15 Nov 2012 00:34:11 +0000 (16:34 -0800)]
Merge branch 'master' of https://github.com/ceph/teuthology
Sander Pool [Thu, 15 Nov 2012 00:27:01 +0000 (16:27 -0800)]
Added comment explaining how setup is tied to actual code.
Sam Lang [Wed, 14 Nov 2012 22:49:24 +0000 (16:49 -0600)]
workunit: Fix indentation
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sam Lang [Wed, 14 Nov 2012 20:07:16 +0000 (14:07 -0600)]
task/ceph-fuse: If umount fails, abort and cleanup
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sam Lang [Wed, 14 Nov 2012 16:45:10 +0000 (10:45 -0600)]
pexec: Logging each command isn't useful
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sam Lang [Wed, 14 Nov 2012 16:11:39 +0000 (10:11 -0600)]
Add task pexec to run bash commands in parallel
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sage Weil [Tue, 13 Nov 2012 23:34:10 +0000 (15:34 -0800)]
schedule_suite.sh: less noise
Sage Weil [Tue, 13 Nov 2012 22:33:38 +0000 (14:33 -0800)]
schedule_suite.sh: less noise
Sam Lang [Mon, 12 Nov 2012 19:16:49 +0000 (13:16 -0600)]
misc: Show url on get failure
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sage Weil [Sat, 10 Nov 2012 16:41:41 +0000 (08:41 -0800)]
ceph-fuse: apply overrides[ceph-fuse] to config
Sage Weil [Sat, 10 Nov 2012 04:23:05 +0000 (20:23 -0800)]
valgrind.supp: deliverate onexit leak
Sage Weil [Sat, 10 Nov 2012 04:22:39 +0000 (20:22 -0800)]
valgrind.supp: ceph-fuse leak from libfuse
Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Fri, 9 Nov 2012 00:37:52 +0000 (16:37 -0800)]
Clean up string interpolation operator spacing ceph_manager.py
Signed-off-by: Samuel Just <sam.just@inktank.com>
Samuel Just [Fri, 9 Nov 2012 00:22:40 +0000 (16:22 -0800)]
Add divergent_priors test
Tests scenario where merge_old_entry encounters a divergent
entry where the prior_version is prior to log_tail. This
is a problem since it will go into the missing set, but won't
be re-added to the missing set during read_log() if the node
restarts prior to recovering the object.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Sam Lang [Thu, 8 Nov 2012 14:55:36 +0000 (08:55 -0600)]
workunit: Move cleanup to separate run
Removing the scratchdir in the remote run command
at the end of the script invocation will do the remove
once the first script finishes. With possibly a shared
scratch dir across workunit clients, we want to wait to
remove the scratch dir once all the workunit scripts have
completed.
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sam Lang [Wed, 7 Nov 2012 19:14:09 +0000 (13:14 -0600)]
workunit: Allow scratch dir to already exist
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sam Lang [Wed, 7 Nov 2012 04:28:56 +0000 (20:28 -0800)]
workunit: Add option to use specified subdir
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Samuel Just [Wed, 7 Nov 2012 20:36:37 +0000 (12:36 -0800)]
ceph_manager: add test_min_size action
Thrasher can now with configurable frequency test min_size by
taking down all but one osd, waiting, killing that osd and bringing
back the others, and verifying that the cluster goes clean.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Josh Durgin [Fri, 2 Nov 2012 18:33:46 +0000 (11:33 -0700)]
schedule: fix var name
Josh Durgin [Fri, 2 Nov 2012 18:08:25 +0000 (11:08 -0700)]
schedule: add option to display jobs in the queue
beanstalkd doesn't let you list jobs in the queue, but you can
inpsect specific job ids.
Alex Elder [Thu, 1 Nov 2012 18:32:56 +0000 (13:32 -0500)]
rbd task: support xfstests repeat count
This adds the ability to use the new repeat count argument to the
run_xfstests.sh script. By default, the test suite will be run
once, but if a count is specified the script will execute the suite
that many times, but will only perform the setup (building the
tests, etc.) once.
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Reviewed-by: Dan Mick <dan.mick@inktank.com>
Josh Durgin [Fri, 26 Oct 2012 01:12:28 +0000 (18:12 -0700)]
Merge pull request #4 from ceph/wip-buck
Use the configured username for _make_scratch_dir
Joe Buck [Thu, 25 Oct 2012 00:33:25 +0000 (17:33 -0700)]
Use the configured username for _make_scratch_dir
David Zafman [Thu, 25 Oct 2012 17:08:05 +0000 (10:08 -0700)]
Add exit to kcon_most script
Signed-off-by: David Zafman <david.zafman@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin [Wed, 24 Oct 2012 23:10:56 +0000 (16:10 -0700)]
coverage: note db table structure
Sage Weil [Mon, 22 Oct 2012 23:51:54 +0000 (16:51 -0700)]
add exec task
Joe Buck [Mon, 22 Oct 2012 19:04:46 +0000 (12:04 -0700)]
Merge pull request #2 from ceph/wip-buck
Wip buck
Joe Buck [Mon, 22 Oct 2012 18:33:44 +0000 (11:33 -0700)]
fixing the tense of my README change
buck [Mon, 22 Oct 2012 18:21:07 +0000 (11:21 -0700)]
Edit to README.bst to note that the ssh keys needed in the targets entries are the hosts public ssh key and not the users defined in the same entry
David Zafman [Tue, 9 Oct 2012 22:50:25 +0000 (15:50 -0700)]
New nfs task that performs NFS client mount of export (see knfsd)
Fixes: 3245
Signed-off-by: David Zafman <david.zafman@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
David Zafman [Sat, 6 Oct 2012 02:26:24 +0000 (19:26 -0700)]
New knfsd task that does an nfs server export
Ability to specify options
By default only export to current hosts
Fixes: 3245
Signed-off-by: David Zafman <david.zafman@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
David Zafman [Tue, 9 Oct 2012 00:59:47 +0000 (17:59 -0700)]
New kcon_most task that enables most ceph kernel logging
Signed-off-by: David Zafman <david.zafman@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
David Zafman [Tue, 9 Oct 2012 00:08:41 +0000 (17:08 -0700)]
Fix ceph-fuse example
Signed-off-by: David Zafman <david.zafman@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
David Zafman [Sat, 6 Oct 2012 02:25:48 +0000 (19:25 -0700)]
Fix typo in README
Signed-off-by: David Zafman <david.zafman@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
tamil [Sat, 6 Oct 2012 00:33:57 +0000 (17:33 -0700)]
Printing the number of tests passed when 'all' tests are successful
Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
Yehuda Sadeh [Tue, 2 Oct 2012 00:17:14 +0000 (17:17 -0700)]
radosgw-admin: usage should time out after 20 minutes
Not 45 seconds.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Mon, 1 Oct 2012 04:08:41 +0000 (21:08 -0700)]
console: add console task
Log the sol console of every target to a file in the archive dir.
Sage Weil [Mon, 1 Oct 2012 04:07:58 +0000 (21:07 -0700)]
internal: stop warning about lockdep circular dependency
This is coming from xfs, currently. Bah.
Sage Weil [Sat, 29 Sep 2012 15:54:08 +0000 (08:54 -0700)]
include newpool in osd cap for client.0
This is needed by the kclient_workunit_kclient task.
Josh Durgin [Fri, 28 Sep 2012 17:07:45 +0000 (10:07 -0700)]
misc: use new syntax for osd caps
pool=pool1,pool2 is not valid for the new grammar
Sage Weil [Sun, 23 Sep 2012 15:03:17 +0000 (08:03 -0700)]
fix lock held when returning to user space typo
Josh Durgin [Fri, 21 Sep 2012 21:54:19 +0000 (14:54 -0700)]
Allow scheduled jobs to use different teuthology branches
teuthology-[schedule|suite] get a parameter to specify the branch,
to put the job in a branch-specific queue. Workers running that
branch of teuthology can pull jobs from that queue.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin [Fri, 21 Sep 2012 23:44:35 +0000 (16:44 -0700)]
Fix errors found by pyflakes
A bunch of unused imports and variables.
Sage Weil [Wed, 19 Sep 2012 21:19:53 +0000 (14:19 -0700)]
ignore 'lock held when returning to user space' from btrfs sb_internal crap
Sam Lang [Tue, 18 Sep 2012 23:31:39 +0000 (16:31 -0700)]
replace tab with spaces
Signed-off-by: Sam Lang <sam.lang@inktank.com>
Sam Lang [Tue, 18 Sep 2012 22:56:08 +0000 (15:56 -0700)]
fix error on teardown failing to unmount /mnt
Sam Lang [Tue, 18 Sep 2012 20:50:36 +0000 (13:50 -0700)]
update doc to be ceph-fuse task (instead of cfuse)
tamil [Fri, 14 Sep 2012 22:04:40 +0000 (15:04 -0700)]
imported subprocess module in nuke script
Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
Josh Durgin [Thu, 13 Sep 2012 21:31:46 +0000 (14:31 -0700)]
nuke: add missing import
Josh Durgin [Thu, 13 Sep 2012 21:29:43 +0000 (14:29 -0700)]
rbd: fix typo and cast to int before comparing format
Josh Durgin [Wed, 12 Sep 2012 18:30:40 +0000 (11:30 -0700)]
rbd: only specify --format if not using the default
This lets older versions that do not support --format still work with
format 1 images.
Tommi Virtanen [Tue, 11 Sep 2012 18:11:39 +0000 (11:11 -0700)]
Don't lose tracebacks of exceptions raised in a greenlet.
Exception objects don't contain the traceback of where they were
raised from (to avoid cyclic data structures wrecking gc and causing
mem leaks), so the singular "raise obj" form creates a new traceback
from the current execution location, thus losing the original location
of the error.
Gevent explicitly wants to throw away the traceback, to release any
objects the greenlet may still be referring to, closing files,
releasing locks etc. In this case, we think it's safe, so stash the
exception info away in a holder object, and resurrect it on the other
side of the results queue.
http://stackoverflow.com/questions/
9268916 /how-to-capture-a-traceback-in-gevent
This can be reproduced easily with
from teuthology.parallel import parallel
def f():
raise RuntimeError("bork")
with parallel() as p:
p.spawn(f)
and looking at the resulting traceback with and without this change.
Alex Elder [Mon, 10 Sep 2012 14:23:08 +0000 (09:23 -0500)]
rbd: allow xfstests task to specify rbd image formats
This adds the ability to specify the rbd image format to use for the
scratch and test devices for the rbd.xfstests task.
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-by: Dan Mick <dan.mick@inktank.com>
Alex Elder [Mon, 10 Sep 2012 14:23:08 +0000 (09:23 -0500)]
rbd: allow image format to be specified
This adds the ability to specify an rbd image format (either 1 or 2)
for an rbd image.
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-by: Dan Mick <dan.mick@inktank.com>
tamil [Mon, 10 Sep 2012 22:02:47 +0000 (15:02 -0700)]
Suppress valgrind error "Invalid write 8"
Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
Yehuda Sadeh [Mon, 10 Sep 2012 18:14:12 +0000 (11:14 -0700)]
radosgw-admin: update task for new usage reporting
Usage reporting output has been modified, also use the new
--categories input param.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Mon, 10 Sep 2012 18:09:25 +0000 (11:09 -0700)]
schedule_suite.sh: try to use same branch for s3-tests.git
If it isn't there, warn and fall back to master.
Sage Weil [Mon, 10 Sep 2012 18:08:57 +0000 (11:08 -0700)]
s3tests: run against arbitrary branch/sha1 of s3-tests.git
Sage Weil [Sun, 9 Sep 2012 21:23:12 +0000 (14:23 -0700)]
debian ntp servers
Mike Ryan [Tue, 4 Sep 2012 16:52:38 +0000 (09:52 -0700)]
task: die on ceph error or coredump
This task allows ceph to signal to teuth that it should die immediately
by touching a file under /tmp/cephtest
Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
Sage Weil [Sat, 25 Aug 2012 02:23:34 +0000 (19:23 -0700)]
disable lockdep recursive warnings until #3040 is fixed
Sage Weil [Thu, 23 Aug 2012 18:00:34 +0000 (11:00 -0700)]
internal: fix escaping of \b in syslog grep
Sage Weil [Wed, 22 Aug 2012 22:29:44 +0000 (15:29 -0700)]
suppress this valgrind error
<error>
<unique>0x4</unique>
<tid>1</tid>
<kind>InvalidWrite</kind>
<what>Invalid write of size 8</what>
<stack>
<frame>
<ip>0x400A299</ip>
<obj>/lib/x86_64-linux-gnu/ld-2.15.so</obj>
<fn>do_lookup_x</fn>
<dir>/build/buildd/eglibc-2.15/elf</dir>
<file>dl-lookup.c</file>
<line>250</line>
</frame>
<frame>
<ip>0x403122F</ip>
</frame>
<frame>
<ip>0x400A522</ip>
<obj>/lib/x86_64-linux-gnu/ld-2.15.so</obj>
<fn>_dl_lookup_symbol_x</fn>
<dir>/build/buildd/eglibc-2.15/elf</dir>
<file>dl-lookup.c</file>
<line>739</line>
</frame>
</stack>
<auxwhat>Address 0x7feffeec8 is on thread 1's stack</auxwhat>
</error>
pops up recently
Sage Weil [Mon, 20 Aug 2012 03:16:43 +0000 (20:16 -0700)]
ceph: fix cpu_profile default
Sage Weil [Sun, 19 Aug 2012 03:10:54 +0000 (20:10 -0700)]
rbd.xfstests: default to 1gb (not 250mb) image
Mike Ryan [Fri, 17 Aug 2012 20:47:13 +0000 (13:47 -0700)]
task: run osd/mds/mon with Google CPU profiler via cpu_profile option
Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
Mike Ryan [Thu, 16 Aug 2012 22:50:10 +0000 (15:50 -0700)]
ceph: support tmpfs_journal option to put journal on tmpfs
Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
Sage Weil [Tue, 14 Aug 2012 22:08:21 +0000 (15:08 -0700)]
queue: fix logging of child return code
Tommi Virtanen [Mon, 13 Aug 2012 23:10:05 +0000 (16:10 -0700)]
Disable asynchronous DNS lookups.
Especially on older hosts, we keep triggering errors::
ServerNotFoundError: Unable to find the server at
teuthology.front.sepia.ceph.com: [Errno 3] name does not exist
That comes from libevent's evdns via gevent.dns and httplib2. The rate
of these errors is low enough that they seem to be perhaps timeouts,
or more arbitrary. Busy looping on DNS resolution calls has never
triggered them, so far.
With ``monkey.patch_all(dns=False)``, the teuthology process will
block as a whole whenever doing DNS resolution. This will hopefully be
rare enough that it won't matter.
The only real "fix" seems to be upgrading libraries and hoping for the
best; this commit can be reverted after that is done.
Tommi Virtanen [Thu, 9 Aug 2012 16:42:35 +0000 (09:42 -0700)]
Flush data to temp file before reading it in another process.
Tommi Virtanen [Thu, 9 Aug 2012 16:42:13 +0000 (09:42 -0700)]
Oops tempfile now gives us file objects not fds.
Tommi Virtanen [Wed, 8 Aug 2012 21:48:21 +0000 (14:48 -0700)]
In teuthology-worker, shuffle the child stdout/stderr into our log.
Otherwise, child can suffer a failure that does not get logged by
it's own exception handling machinery, and we have no idea why.
Tommi Virtanen [Wed, 8 Aug 2012 21:45:49 +0000 (14:45 -0700)]
Minimize scope of try-except.
os.write and list.append won't raise CalledProcessError, and now
we don't need to try to contain them for temp file clean up reasons.
Tommi Virtanen [Wed, 8 Aug 2012 21:44:47 +0000 (14:44 -0700)]
Use tempfile.NamedTemporaryFile instead of mkstemp.
Simpler code, no manual cleanup needed. We see a littering of
zero-length temp files from teuthology-worker, and this seems
like a likely source.
Mike Ryan [Thu, 2 Aug 2012 17:58:08 +0000 (10:58 -0700)]
task: verify scrub detects files whose contents changed
Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
Mike Ryan [Wed, 1 Aug 2012 22:35:38 +0000 (15:35 -0700)]
task: scrub OSDs periodically
Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
Sage Weil [Sun, 29 Jul 2012 19:40:06 +0000 (12:40 -0700)]
kernel: push a local .deb instead of using gitbuilder
This lets you specify a path to an existing kernel deb
to be pushed and installed on the remote node.
Limitations:
- We don't build the deb for you. figuring out what
filename 'make deb-pkg' is building is annoying.
- We need to be able to figure out the sha1 from the provided
path. It shouldn't be a problem, given the way make deb-pkg
names the debs.
Sage Weil [Sun, 29 Jul 2012 19:15:06 +0000 (12:15 -0700)]
syslog check: fix false-positive BUG matches in random strings
Sage Weil [Sat, 28 Jul 2012 17:53:09 +0000 (10:53 -0700)]
osd_recovery: also test unfound discovery
This tests for bug #2866.
Sage Weil [Sat, 28 Jul 2012 17:23:02 +0000 (10:23 -0700)]
osd_recovery: test incomplete pg recovery
4-osd test to reproduce #2860 and confirm the fix.