linux/fs/ceph
Yan, Zheng fc927cd32f ceph: always re-send cap flushes when MDS recovers
commit e548e9b93d makes the kclient
only re-send cap flush once during MDS failover. If the kclient sends
a cap flush after MDS enters reconnect stage but before MDS recovers.
The kclient will skip re-sending the same cap flush when MDS recovers.

This causes problem for newly created inode. The MDS handles cap
flushes before replaying unsafe requests, so it's possible that MDS
find corresponding inode is missing when handling cap flush. The fix
is reverting to old behaviour: always re-send when MDS recovers

Signed-off-by: Yan, Zheng <zyan@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
2015-07-31 11:38:53 +03:00
..
acl.c ceph: switch some GFP_NOFS memory allocation to GFP_KERNEL 2015-06-25 11:49:31 +03:00
addr.c ceph: fix ceph_writepages_start() 2015-06-25 18:30:53 +03:00
cache.c
cache.h
caps.c ceph: always re-send cap flushes when MDS recovers 2015-07-31 11:38:53 +03:00
ceph_frag.c
debugfs.c VFS: normal filesystems (and lustre): d_inode() annotations 2015-04-15 15:06:57 -04:00
dir.c ceph: rework dcache readdir 2015-06-25 11:49:32 +03:00
export.c VFS: normal filesystems (and lustre): d_inode() annotations 2015-04-15 15:06:57 -04:00
file.c Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2015-07-04 19:36:06 -07:00
inode.c Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client 2015-07-02 11:35:00 -07:00
ioctl.c ceph: fix divide-by-zero in __validate_layout() 2014-10-14 12:57:05 -07:00
ioctl.h
Kconfig
locks.c ceph: fix ceph_encode_locks_to_buffer() 2015-07-31 11:38:47 +03:00
Makefile
mds_client.c ceph: switch some GFP_NOFS memory allocation to GFP_KERNEL 2015-06-25 11:49:31 +03:00
mds_client.h ceph: rework dcache readdir 2015-06-25 11:49:32 +03:00
mdsmap.c
snap.c ceph: make sure syncfs flushes all cap snaps 2015-06-25 11:49:29 +03:00
strings.c ceph: rename snapshot support 2015-04-22 18:33:41 +03:00
super.c ceph: pre-allocate data structure that tracks caps flushing 2015-06-25 11:49:31 +03:00
super.h ceph: always re-send cap flushes when MDS recovers 2015-07-31 11:38:53 +03:00
xattr.c ceph: pre-allocate data structure that tracks caps flushing 2015-06-25 11:49:31 +03:00