Sam Lang 6fd7ebd44d task/mds_thrash: Log mds dump after long delay ... In cases where the mds thrasher continuously loops waiting for an mds to be removed from the map, or for a new mds to become active, we want to start logging the mds state for debugging. Signed-off-by: Sam Lang <sam.lang@inktank.com>		2013-03-27 08:48:45 -05:00
..
locker	locker: make desc optional	2013-03-26 13:27:53 -07:00
orchestra	task/restart: Restart task for testing daemon kill	2013-03-18 11:16:01 -05:00
task	task/mds_thrash: Log mds dump after long delay	2013-03-27 08:48:45 -05:00
test
__init__.py
ceph.conf.template	ceph.conf: osd debug op order = true	2013-03-26 12:21:13 -07:00
contextutil.py
coverage.py	coverage: note db table structure	2012-10-24 16:11:12 -07:00
lock.py	lock: pass desc to lock operation; leave on unlock	2013-03-25 16:47:02 -07:00
misc.py	task/ceph_manager: Only reconnect if powercycled	2013-03-13 11:15:51 -05:00
nuke.py	install, nuke: explicitly purge /var/lib/ceph	2013-03-22 15:22:38 -07:00
parallel.py	Don't lose tracebacks of exceptions raised in a greenlet.	2012-09-11 11:25:21 -07:00
queue.py	add --name option to teuthology	2013-01-31 08:23:34 -06:00
run_tasks.py	Handle interactive-on-error also when error is from contextmanager exit.	2011-11-30 17:07:26 -08:00
run.py	verify /var/lib/ceph not present on start	2013-03-23 20:58:46 -07:00
safepath.py
suite.py	suite: shorten subject a bit	2013-03-19 12:12:56 -07:00