]> git.apps.os.sepia.ceph.com Git - ceph.git/log
ceph.git
11 years agotask/ceph: separate ceph.conf into function
John Spray [Mon, 12 May 2014 13:25:26 +0000 (14:25 +0100)]
task/ceph: separate ceph.conf into function

This is so that other tasks can modify ceph.conf
as needed.

Signed-off-by: John Spray <john.spray@inktank.com>
11 years agotasks: Use '_' instead of '-' in names
John Spray [Mon, 12 May 2014 09:55:58 +0000 (10:55 +0100)]
tasks: Use '_' instead of '-' in names

Python modules are not allowed to contain hyphens
in the name.  Using hyphens only works here because
we're using low level __import__.

Change run_tasks to replace '-' with '_' when reading
configs, and rename the task modules to have valid
python module names.

Aside from general hygiene, the motivation to do this
is to allow one task to build on code in another
task by importing it.

Signed-off-by: John Spray <john.spray@inktank.com>
11 years agoDon't crash when an invalid branch is passed
Zack Cerza [Thu, 22 May 2014 17:26:50 +0000 (12:26 -0500)]
Don't crash when an invalid branch is passed

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoSet results_server by default
Zack Cerza [Thu, 22 May 2014 13:49:26 +0000 (08:49 -0500)]
Set results_server by default

So that queued jobs get pushed to paddles

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoProcess queued jobs synchronously
Zack Cerza [Wed, 21 May 2014 20:58:27 +0000 (15:58 -0500)]
Process queued jobs synchronously

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoMerge pull request #260 from ceph/wip-8422
Warren Usui [Wed, 21 May 2014 17:24:56 +0000 (10:24 -0700)]
Merge pull request #260 from ceph/wip-8422

Add exception hook to teuthology-worker

11 years agoAdd exception hook to teuthology-worker
Zack Cerza [Wed, 21 May 2014 16:34:11 +0000 (11:34 -0500)]
Add exception hook to teuthology-worker

Workers processes are dying occasionally, and this should cause the
exceptions to be logged.

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoMerge pull request #258 from ceph/wip-fix-format-samba
Zack Cerza [Mon, 19 May 2014 23:33:22 +0000 (18:33 -0500)]
Merge pull request #258 from ceph/wip-fix-format-samba

Fixed formatting and added a better variable for time out

11 years agoFixed broken quotations
Yuri Weinstein [Mon, 19 May 2014 23:18:48 +0000 (16:18 -0700)]
Fixed broken quotations

Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
11 years agoFixed formatting and added a better variable for time out
Yuri Weinstein [Mon, 19 May 2014 23:03:30 +0000 (16:03 -0700)]
Fixed formatting and added a better variable for time out
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
11 years agotask/: add populate_rbd_pool for quickly spinning up a pool with rbd stuff
Samuel Just [Mon, 19 May 2014 22:04:54 +0000 (15:04 -0700)]
task/: add populate_rbd_pool for quickly spinning up a pool with rbd stuff

Signed-off-by: Samuel Just <sam.just@inktank.com>
11 years agoceph_manager: reset timeout if we are making progress in wait_for_recovery
Samuel Just [Mon, 19 May 2014 21:09:51 +0000 (14:09 -0700)]
ceph_manager: reset timeout if we are making progress in wait_for_recovery

Signed-off-by: Samuel Just <sam.just@inktank.com>
11 years agoAvoid calling yaml.safe_load() twice on each job
Zack Cerza [Mon, 19 May 2014 16:05:05 +0000 (11:05 -0500)]
Avoid calling yaml.safe_load() twice on each job

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoSplit progress indicator out to separate functions
Zack Cerza [Mon, 19 May 2014 15:49:24 +0000 (10:49 -0500)]
Split progress indicator out to separate functions

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoincreased sleep to 100 as 90 seemed insufficient
Yuri Weinstein [Mon, 19 May 2014 15:11:20 +0000 (08:11 -0700)]
increased sleep to 100 as 90 seemed insufficient
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
11 years agoAdded more logging
Yuri Weinstein [Sat, 17 May 2014 17:15:57 +0000 (10:15 -0700)]
Added more logging
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
11 years agoAdded sleep log info line
Yuri Weinstein [Sat, 17 May 2014 17:07:52 +0000 (10:07 -0700)]
Added sleep log info line
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
11 years agoMerge pull request #257 from ceph/wip-samba-fix
Zack Cerza [Thu, 15 May 2014 16:48:29 +0000 (11:48 -0500)]
Merge pull request #257 from ceph/wip-samba-fix

increased time wait to 90 sec so tests does not fail

11 years agoincreased time wait to 90 sec so tests does not fail
Yuri Weinstein [Thu, 15 May 2014 16:33:44 +0000 (09:33 -0700)]
increased time wait to 90 sec so tests does not fail

11 years agoAdd Pulpito links to teuthology emails
Zack Cerza [Thu, 15 May 2014 16:08:10 +0000 (11:08 -0500)]
Add Pulpito links to teuthology emails

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoDowngrade paramiko to work around an SFTP bug
Zack Cerza [Thu, 15 May 2014 03:43:29 +0000 (22:43 -0500)]
Downgrade paramiko to work around an SFTP bug

See http://tracker.ceph.com/issues/8327

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoMerge pull request #256 from ceph/wip-6542-wusui
Zack Cerza [Tue, 13 May 2014 21:26:12 +0000 (16:26 -0500)]
Merge pull request #256 from ceph/wip-6542-wusui

Add missng docstrings to repair_test.py

11 years agoAdd missng docstrings to repair_test.py
Warren Usui [Tue, 13 May 2014 17:45:09 +0000 (10:45 -0700)]
Add missng docstrings to repair_test.py

Fixes: 6542
Signed-off-by: Warren Usui <warren.usui@inktank.com>
11 years agoFix unit tests under Jenkins
Zack Cerza [Tue, 13 May 2014 18:09:04 +0000 (13:09 -0500)]
Fix unit tests under Jenkins

os.getlogin() was throwing:
  OSError: [Errno 25] Inappropriate ioctl for device

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse VersionNotFoundError packages are missing
Zack Cerza [Tue, 13 May 2014 15:21:18 +0000 (10:21 -0500)]
Use VersionNotFoundError packages are missing

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse config.archive_base if one isn't passed
Zack Cerza [Mon, 12 May 2014 21:25:31 +0000 (16:25 -0500)]
Use config.archive_base if one isn't passed

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoAdd retries to orchestra.connection.connect()
Zack Cerza [Mon, 12 May 2014 21:06:40 +0000 (16:06 -0500)]
Add retries to orchestra.connection.connect()

This is an attempt to fix: http://tracker.ceph.com/issues/8314

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoFix typo
Zack Cerza [Mon, 12 May 2014 20:29:10 +0000 (15:29 -0500)]
Fix typo

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoBetter logging
Zack Cerza [Mon, 12 May 2014 14:30:21 +0000 (09:30 -0500)]
Better logging

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse try_mark_run_dead()
Zack Cerza [Mon, 12 May 2014 14:18:34 +0000 (09:18 -0500)]
Use try_mark_run_dead()

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoAdd try_mark_run_dead()
Zack Cerza [Mon, 12 May 2014 14:16:59 +0000 (09:16 -0500)]
Add try_mark_run_dead()

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoMove list of exceptions to catch
Zack Cerza [Mon, 12 May 2014 14:11:04 +0000 (09:11 -0500)]
Move list of exceptions to catch

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoMove "no results server" warning
Zack Cerza [Mon, 12 May 2014 14:02:27 +0000 (09:02 -0500)]
Move "no results server" warning

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoFix name parsing
Zack Cerza [Mon, 12 May 2014 17:57:21 +0000 (12:57 -0500)]
Fix name parsing

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoMerge pull request #255 from ceph/wip-6921-wusui
Zack Cerza [Mon, 12 May 2014 17:50:56 +0000 (12:50 -0500)]
Merge pull request #255 from ceph/wip-6921-wusui

Allow .teuthology.yaml to set downburst path

11 years agoDon't pass a custom logger anymore
Zack Cerza [Fri, 9 May 2014 20:50:22 +0000 (15:50 -0500)]
Don't pass a custom logger anymore

We already use the hostname in command execution calls

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse 'stderr' and 'stdout' as logger names
Zack Cerza [Fri, 9 May 2014 20:18:50 +0000 (15:18 -0500)]
Use 'stderr' and 'stdout' as logger names

We were using just 'err' and 'out', which isn't very intuitive.

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoCatch any Unicode errors that manage to sneak in
Zack Cerza [Fri, 9 May 2014 19:23:52 +0000 (14:23 -0500)]
Catch any Unicode errors that manage to sneak in

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoExpress hostnames as child logger names
Zack Cerza [Fri, 9 May 2014 17:55:42 +0000 (12:55 -0500)]
Express hostnames as child logger names

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse Remote.user
Zack Cerza [Fri, 9 May 2014 17:07:55 +0000 (12:07 -0500)]
Use Remote.user

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoAdd Remote.user attribute
Zack Cerza [Fri, 9 May 2014 17:07:40 +0000 (12:07 -0500)]
Add Remote.user attribute

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse Remote.shortname in logs
Zack Cerza [Fri, 9 May 2014 16:30:13 +0000 (11:30 -0500)]
Use Remote.shortname in logs

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoMake Remote.shortname actually short
Zack Cerza [Fri, 9 May 2014 16:27:23 +0000 (11:27 -0500)]
Make Remote.shortname actually short

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoRemote.hostname doesn't have to be a property
Zack Cerza [Fri, 9 May 2014 16:25:02 +0000 (11:25 -0500)]
Remote.hostname doesn't have to be a property

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoFix PEP-8 issues
Zack Cerza [Thu, 8 May 2014 23:44:15 +0000 (18:44 -0500)]
Fix PEP-8 issues

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoPass hostname to execute()
Zack Cerza [Thu, 8 May 2014 21:24:57 +0000 (16:24 -0500)]
Pass hostname to execute()

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse 'true' instead of 'echo online'
Zack Cerza [Wed, 7 May 2014 18:36:24 +0000 (13:36 -0500)]
Use 'true' instead of 'echo online'

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoAdd Remote.ensure_online()
Zack Cerza [Wed, 7 May 2014 17:36:06 +0000 (12:36 -0500)]
Add Remote.ensure_online()

If the connection is alive, do nothing. If not, reconnect. Allow any
exceptions to bubble up to the caller. This is intended to fix unhelpful
'Bad File Descriptor' errors we were seeing when VMs go down.

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse Remote.hostname in logs
Zack Cerza [Wed, 7 May 2014 17:29:12 +0000 (12:29 -0500)]
Use Remote.hostname in logs

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoConsolidate log file setup into shared function
Zack Cerza [Wed, 7 May 2014 16:02:52 +0000 (11:02 -0500)]
Consolidate log file setup into shared function

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse Unicode format strings
Zack Cerza [Wed, 7 May 2014 15:59:15 +0000 (10:59 -0500)]
Use Unicode format strings

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoRevert "Revert "Show hostname instead of IP in errors""
Zack Cerza [Tue, 6 May 2014 21:17:34 +0000 (16:17 -0500)]
Revert "Revert "Show hostname instead of IP in errors""

This reverts commit 10fee0e368750cf4cd953db5700df59c7f611119.

Conflicts:
teuthology/orchestra/run.py

11 years agoAllow .teuthology.yaml to set downburst path
Warren Usui [Sat, 10 May 2014 00:20:26 +0000 (17:20 -0700)]
Allow .teuthology.yaml to set downburst path

If .teuthology.yaml defines downburst, _get_downburst_exec()
now returns that value as the path to the downburst executable.

Also cleaned up code in create_if_vm.  ctx.downburst_conf was
never defined, so the code that handled the AttributeError
was always being executed.

Fixes: 6921
Signed-off-by: Warren Usui <warren.usui@inktank.com>
11 years agoMerge pull request #254 from ceph/wip-7707-wusui
Zack Cerza [Fri, 9 May 2014 18:17:31 +0000 (13:17 -0500)]
Merge pull request #254 from ceph/wip-7707-wusui

Use master as default for debian upgrade.

11 years agoForce log lines to be interpreted as UTF-8
Zack Cerza [Fri, 9 May 2014 15:45:01 +0000 (10:45 -0500)]
Force log lines to be interpreted as UTF-8

Any invalid UTF-8 byte will be replaced with a Unicode replacement
character: U+FFFD or '�'

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse binary flag for paramiko ChannelFiles
Zack Cerza [Fri, 9 May 2014 14:24:22 +0000 (09:24 -0500)]
Use binary flag for paramiko ChannelFiles

This works around http://tracker.ceph.com/issues/8313

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse master as default for debian upgrade.
Warren Usui [Fri, 9 May 2014 02:12:16 +0000 (19:12 -0700)]
Use master as default for debian upgrade.

Make sure that uri is defined for debian upgrades.
Use master as default.

Added _get_uri_() which consolidates checking and assignment of
sha1, tags, and branch values.

Fixes: 7707
Signed-off-by: Warren Usui <warren.usui@inktank.com>
11 years agoMerge pull request #253 from ceph/wip-7510-wusui
Zack Cerza [Thu, 8 May 2014 18:47:35 +0000 (13:47 -0500)]
Merge pull request #253 from ceph/wip-7510-wusui

Further clarify 'too many values to unpack' error.

11 years agoFurther clarify 'too many values to unpack' error.
Warren Usui [Wed, 7 May 2014 21:05:56 +0000 (14:05 -0700)]
Further clarify 'too many values to unpack' error.

Many errors in yaml configurations cause ValueError to get thrown
with the message 'too many values to unpack.'  A previously reverted
change tried to handle all these situations and print an appropriate
message.

The current behavior of throwing the ValueError exception and exiting
is probably what we still want in these cases.  So instead of handling
the error, the code now checks for the exception at the top-most call
and displays appropriate messages in log.error and in the ctx.summary
data itself.

Fixes: 7510
Signed-off-by: Warren Usui <warren.usui@inktank.com>
11 years agoMerge pull request #252 from ceph/wip-fsx-krbd
Josh Durgin [Thu, 8 May 2014 01:14:54 +0000 (18:14 -0700)]
Merge pull request #252 from ceph/wip-fsx-krbd

rbd_fsx: expose krbd and related fsx options

Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
11 years agoMerge pull request #249 from ceph/wip-8284
Zack Cerza [Wed, 7 May 2014 18:20:40 +0000 (13:20 -0500)]
Merge pull request #249 from ceph/wip-8284

rados.py: Add pool_snaps option for ceph_test_rados test command

11 years agorbd_fsx: expose krbd and related fsx options
Ilya Dryomov [Tue, 6 May 2014 04:45:22 +0000 (08:45 +0400)]
rbd_fsx: expose krbd and related fsx options

Expose

-K (enable krbd mode) through 'krbd',
-Z (use direct IO) through 'direct_io',
-U (disable randomized striping) through 'randomized_striping',
-H (disable discard ops) through 'punch_holes',

-r readbdy (read alignment) through 'readbdy',
-w writebdy (write alignment) through 'writebdy',
-h holebdy (discard alignment) through 'holebdy'.

Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
11 years agoRewrite most file-retrieval functions
Zack Cerza [Mon, 5 May 2014 21:03:39 +0000 (16:03 -0500)]
Rewrite most file-retrieval functions

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoFix linter errors
Zack Cerza [Mon, 5 May 2014 17:25:17 +0000 (12:25 -0500)]
Fix linter errors

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoFIx mktemp dir and redundant Paramiko connecting.
Warren Usui [Thu, 1 May 2014 19:52:45 +0000 (12:52 -0700)]
FIx mktemp dir and redundant Paramiko connecting.

Use previously initialized connection for sftp_get calls.
Use local directory for tarball temp file location.

11 years agoUse SFTPClienti get for long reads/writes
Warren Usui [Thu, 1 May 2014 01:09:12 +0000 (18:09 -0700)]
Use SFTPClienti get for long reads/writes

Modified remote.py to use the paramiko SFTPClient get
method to extract long files (mostly tar files) from
the remote host.  The code formerly saved the data
in a long local string which was very inefficient.

Fixes: 8261
Signed-off-by: Warren Usui <warren.usui@inktank.com>
11 years agoRevert "Revert "Clean up remote.py and misc.py changes.""
Zack Cerza [Mon, 5 May 2014 21:57:13 +0000 (16:57 -0500)]
Revert "Revert "Clean up remote.py and misc.py changes.""

This reverts commit 02504c3fd27d788e2e446369015b14cbf259a8d2.

11 years agoRevert "Revert "Handle raw data I/O.""
Zack Cerza [Mon, 5 May 2014 21:57:01 +0000 (16:57 -0500)]
Revert "Revert "Handle raw data I/O.""

This reverts commit 0c8a3e2334631f5fc20cb7933d0005353ea6a46e.

11 years agoRevert "Restrict paramiko to old versions for now"
Zack Cerza [Mon, 5 May 2014 21:56:50 +0000 (16:56 -0500)]
Revert "Restrict paramiko to old versions for now"

This reverts commit c86df77aa68dc5331b98df3fa038faee77c578ad.

11 years agoUse longer varchar for locked_by in DB.
Sandon Van Ness [Tue, 6 May 2014 19:06:56 +0000 (12:06 -0700)]
Use longer varchar for locked_by in DB.

Signed-off-by: Sandon Van Ness <sandon@inktank.com>
11 years agorados.py: Add pool_snaps option for ceph_test_rados test command
David Zafman [Sat, 3 May 2014 00:30:36 +0000 (17:30 -0700)]
rados.py: Add pool_snaps option for ceph_test_rados test command

Fixes: #8284
Signed-off-by: David Zafman <david.zafman@inktank.com>
11 years agoMerge pull request #251 from ceph/wip-8295
Zack Cerza [Tue, 6 May 2014 16:34:04 +0000 (11:34 -0500)]
Merge pull request #251 from ceph/wip-8295

rgw: fix indentation for cache_pools

11 years agorgw: fix indentation for cache_pools
Sage Weil [Tue, 6 May 2014 16:22:49 +0000 (09:22 -0700)]
rgw: fix indentation for cache_pools

Fixes: #8295
Signed-off-by: Sage Weil <sage@inktank.com>
11 years agoMerge pull request #250 from ceph/wip-fix-thrasher
Josh Durgin [Tue, 6 May 2014 00:20:19 +0000 (17:20 -0700)]
Merge pull request #250 from ceph/wip-fix-thrasher

ceph_manager: reset osd weights to 1 when waiting for clean

Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
11 years agoceph_manager: reset osd weights to 1 when waiting for clean
Sage Weil [Tue, 6 May 2014 00:05:10 +0000 (17:05 -0700)]
ceph_manager: reset osd weights to 1 when waiting for clean

If we leave the weights adjusted, we can get PGs stuck in a remapped state
because we are probabilistically rejecting placement.  Avoid this by
setting on in osds to 1.0 before waiting for clean.

Signed-off-by: Sage Weil <sage@inktank.com>
11 years agoRestrict paramiko to old versions for now
Zack Cerza [Mon, 5 May 2014 18:10:42 +0000 (13:10 -0500)]
Restrict paramiko to old versions for now

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoRevert "Handle raw data I/O."
Zack Cerza [Mon, 5 May 2014 15:03:29 +0000 (10:03 -0500)]
Revert "Handle raw data I/O."

This reverts commit 257e1459fa064701d789f0ad54384bb80b45e6d9.

11 years agoRevert "Clean up remote.py and misc.py changes."
Zack Cerza [Mon, 5 May 2014 15:03:18 +0000 (10:03 -0500)]
Revert "Clean up remote.py and misc.py changes."

This reverts commit 74eff43ee1a2b73159277370cfa9d194e42bf49c.

11 years agoceph_manager: fix float stringification
Sage Weil [Sat, 3 May 2014 14:24:32 +0000 (07:24 -0700)]
ceph_manager: fix float stringification

Signed-off-by: Sage Weil <sage@inktank.com>
11 years agoceph_manager: fix typo
Sage Weil [Sat, 3 May 2014 04:06:48 +0000 (21:06 -0700)]
ceph_manager: fix typo

From ce7fa1839f4b3e3db675b2d68a2bb57849f58c1e. Tested this time.

Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
11 years agoMerge pull request #248 from ceph/wip-thrash-osd-weights
Josh Durgin [Sat, 3 May 2014 00:06:54 +0000 (17:06 -0700)]
Merge pull request #248 from ceph/wip-thrash-osd-weights

ceph_manager: randomly reweight in osds

Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
11 years agoceph_manager: randomly reweight in osds
Sage Weil [Fri, 2 May 2014 23:32:53 +0000 (16:32 -0700)]
ceph_manager: randomly reweight in osds

Signed-off-by: Sage Weil <sage@inktank.com>
11 years agoBump paramiko to 1.12.0
Zack Cerza [Fri, 2 May 2014 16:48:08 +0000 (11:48 -0500)]
Bump paramiko to 1.12.0

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agorgw: option to create a cache pool
Sage Weil [Thu, 1 May 2014 21:36:08 +0000 (14:36 -0700)]
rgw: option to create a cache pool

64mb for now!

Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
11 years agoMerge pull request #247 from ceph/requests-sessions
Alfredo Deza [Thu, 1 May 2014 17:17:00 +0000 (13:17 -0400)]
Merge pull request #247 from ceph/requests-sessions

Use a requests.Session object for retries instead of safe_while

11 years agoAdd branch name to job config
Zack Cerza [Thu, 1 May 2014 16:48:54 +0000 (11:48 -0500)]
Add branch name to job config

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoAdd suite name to job config
Zack Cerza [Thu, 1 May 2014 16:21:08 +0000 (11:21 -0500)]
Add suite name to job config

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoAdd /build and /*.yaml to gitignore
Zack Cerza [Thu, 1 May 2014 16:25:19 +0000 (11:25 -0500)]
Add /build and /*.yaml to gitignore

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoCalculate a timeout to use based on queue size
Zack Cerza [Tue, 29 Apr 2014 16:30:50 +0000 (11:30 -0500)]
Calculate a timeout to use based on queue size

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoAdd --runs, to print only run names
Zack Cerza [Tue, 29 Apr 2014 16:26:39 +0000 (11:26 -0500)]
Add --runs, to print only run names

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoRefactor teuthology.beanstalk
Zack Cerza [Mon, 28 Apr 2014 19:35:35 +0000 (14:35 -0500)]
Refactor teuthology.beanstalk

This architecture will make it easier to add new functionality.

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoDrop usage of safe_while
Zack Cerza [Tue, 29 Apr 2014 21:12:49 +0000 (16:12 -0500)]
Drop usage of safe_while

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoUse the new ResultsReporter.session object
Zack Cerza [Tue, 29 Apr 2014 21:10:51 +0000 (16:10 -0500)]
Use the new ResultsReporter.session object

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoAdd a requests.Session object to ResultsReporter
Zack Cerza [Tue, 29 Apr 2014 21:07:59 +0000 (16:07 -0500)]
Add a requests.Session object to ResultsReporter

By default it is set up to retry requests 10 times

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoRemove unused timeout arg to ResultsReporter init
Zack Cerza [Tue, 29 Apr 2014 21:02:08 +0000 (16:02 -0500)]
Remove unused timeout arg to ResultsReporter init

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoMerge pull request #243 from ceph/wip-8116-wusui
Zack Cerza [Tue, 29 Apr 2014 21:16:27 +0000 (16:16 -0500)]
Merge pull request #243 from ceph/wip-8116-wusui

Wip 8116 wusui

11 years agoClean up remote.py and misc.py changes.
Warren Usui [Thu, 24 Apr 2014 22:56:41 +0000 (15:56 -0700)]
Clean up remote.py and misc.py changes.

Fixed method names to be non-redundant (remote_mktemp in remote is
now just mktemp, for example), and made some parameters be more
self descriptive.  Added some docstrings.  Fixed sudo setting in
get_file method.

Made chmod independent of the actual sftp file getting.
Do not do extra file copy if non-sudo read is needed.
Fixed some parameter names. Made sure temp files are removed.

11 years agoHandle raw data I/O.
Warren Usui [Tue, 22 Apr 2014 18:14:51 +0000 (11:14 -0700)]
Handle raw data I/O.

Paramiko 1.13.0 checks data in the Channel and fails if
invalid UTF-8 characters are sent.  The teuthology/misc.py
functions that piped cat output (get_file) and piped tar
output (pull_directory and pull_directory_tarball) formerly
did not work for Paramiko 1.13.0.

Code was changed to use SFTPClient to copy raw data.  The
remote_mktemp and function was changed to be a method of the
remote object.  Remote object methods to copy files and write
tar files were also added.

In misc.py, get_file() was changed to be a wrapper around the
remote object call.  Pull_directory() and pull_directory_tarball
were changed to use the remote object methods as well.

The remote methods used to send the data makes use of the
SFTPClient interface in Paramiko.

The remote_mktemp function was also modified slightly to insure
that new-lines do not appear in temp file names.

Fixes: 8116
Signed-off-by: Warren Usui <warren.usui@inktank.com>
11 years agoWhen deleting all of a run's jobs, delete the run
Zack Cerza [Tue, 29 Apr 2014 19:52:31 +0000 (14:52 -0500)]
When deleting all of a run's jobs, delete the run

Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
11 years agoChanges so these are not installed and still removed
Gregory Meno [Tue, 29 Apr 2014 17:44:27 +0000 (10:44 -0700)]
Changes so these are not installed and still removed