- 07:43 AM Ceph Fix #49188 (Fix Under Review): mds: speed up the process of looking up a single dentry ,when the ...
- Looking up a single dentry would take a long time, when it was not in the mds cache.The process would load the whole ...
- 06:13 AM CephFS Bug #44316 (Resolved): mds: assert(p != active_requests.end())
10454 1: (ceph::__ceph_assert_fail(char const*, char const*, int, char const*)+0x11 ...
- 01:58 PM CephFS Bug #43090 (Closed): mds:check if oldin is null before accessing its member
- in mds/server : handle_client_rename:
CInode *oldin = 0;
then oldin will be still 0;
- 10:09 AM CephFS Feature #42831 (Resolved): mds: add config to deny all client reconnects
- This helps reduce mds failover time.
- 12:29 PM CephFS Bug #42770 (Closed): Regulary trim inode in memory
- Inode would be trimmed only when cache reached limit or in the bottom lru now. Too many inode in memory would lead to...
- 02:23 AM CephFS Bug #42271 (Resolved): client: ceph-fuse which had been blacklisted couldn't auto reconnect after...
- 02:05 PM CephFS Bug #42107 (Resolved): client: no method to handle SEEK_HOLE and SEEK_DATA in lseek
- There is no method to handle SEEK_HOLE and SEEK_DATA in lseek in ceph-fuse
- 07:35 AM CephFS Bug #40784 (Resolved): mds: metadata changes may be lost when MDS is restarted
- Assumed a client copied some obj to another location in ceph. when early_replied was received ， the cp command would ...
- 02:32 AM Ceph Revision 43ffd5b0 (ceph): mds: destroy reconnect msg when it is from non-existent session
- fixes: http://tracker.ceph.com/issues/40588
This bug is from backporting "https://github.com/ceph/ceph/pull/27256"
- 02:04 AM Ceph Bug #40588: mimic: mds: msg weren't destroyed before handle_client_reconnect returned, if the rec...
Also available in: Atom