Sage Weil 3d10b34074 mds: fix null_snapflush with multiple intervening snaps ... The client is allowed to not send a snapflush if there is no dirty metadata to write for a given snap. However, the mds can only look up inodes by the last snapid in the interval. So, when doing a null_snapflush (filling in for snapflushes the client didn't send), we have to walk forward through intervening snaps until we find the right inode. Note that this means we will call _do_snap_update multiple times on the same inode, but with different snapids. Add unit test to check this. Signed-off-by: Sage Weil <sage@newdream.net>		2010-11-10 20:58:49 -08:00
..
client	qa: test mdstable failures	2009-12-01 16:19:19 -08:00
mds	qa: another mds test fix... wait long enough for the previous death to trigger	2010-06-17 13:44:12 -07:00
rbd	qa: add rbd test	2010-10-28 15:49:21 -07:00
rgw	qa: improved rgw tests	2010-09-24 10:16:14 -07:00
workunits	mds: fix null_snapflush with multiple intervening snaps	2010-11-10 20:58:49 -08:00
loopall.sh	qa: run tests in random order	2010-04-02 20:07:34 -07:00
runallonce.sh	Revert "qa: remove random ordering from runallonce, for predictable resource matching"	2010-08-24 13:55:30 -07:00
runoncfuse.sh	qa: runoncfuse shouldn't start cfuse in background any more.	2010-07-23 14:37:48 -07:00
runonkclient.sh	qa: Whoops, never added the runonkclient script	2009-12-14 15:46:06 -08:00
setup-chroot.sh	Add the setup-chroot.sh script	2010-09-28 19:25:10 -07:00