Activity
From 02/10/2020 to 03/10/2020
03/10/2020
- 03:24 AM Feature #44533: kclient: add min/max perf latency support
- The patch work: https://patchwork.kernel.org/patch/11428321/
- 12:55 AM Feature #44533 (In Progress): kclient: add min/max perf latency support
- 12:55 AM Feature #44533 (Resolved): kclient: add min/max perf latency support
- Split the code and task into small ones.
This will focus on the min/max perf latency support for the read/write/met... - 03:23 AM Feature #44534 (In Progress): kclient: add standard deviation support for the read/write/metadata...
- 03:23 AM Feature #44534 (Resolved): kclient: add standard deviation support for the read/write/metadata la...
- We need one way to calculate the standard deviation for the read/write/metadata latency.
- 12:51 AM Feature #43215: kclient: send usage metrics to MDS
- For the patch series [1] it only support the debugfs showing, will add sending the metric to ceph stuff after the dep...
03/03/2020
- 06:47 PM Bug #44398 (Resolved): fs: improve wanted handling so we don't request unused caps (active-standb...
- This ticket is to track progress on the CephFS kernel client towards dropping "wanted" caps when they aren't in use, ...
03/02/2020
- 11:04 PM Feature #42098 (Duplicate): Add support for RADOS compressible/incompressible allocation hints
- Oops -- looks like I opened it twice (#43636)
02/27/2020
- 11:31 PM Feature #43794: Add support for replica localized or balanced read policy
- Config options and librbd support is merged in Octopus
- 11:31 PM Feature #43636: Add support for RADOS compressible/incompressible allocation hints
- Config options and librbd support is merged in Octopus
- 05:20 PM Bug #44340 (Resolved): add a way to get clientid for a mount
- We currently don't have a way to tell what the client ID of a particular mount is, in situations where there are mult...
02/23/2020
- 12:19 PM Bug #44215 (In Progress): fs/ceph: show helpful dmesg output on connectivity issues during mount
- Patchwork: https://patchwork.kernel.org/patch/11398773/
02/19/2020
- 05:04 PM Bug #44215 (Resolved): fs/ceph: show helpful dmesg output on connectivity issues during mount
- Right now, mounting CephFS gives unhelpful errors if there are connectivity issues:...
- 02:24 PM Bug #40339 (Resolved): kernel BUG at fs/ceph/mds_client.c:600! invalid opcode: 0000 [#1] SMP
- 09:45 AM Bug #44202 (Rejected): kclient: BUG: ceph_inode_cachep and ceph_dentry_cachep caches are not clea...
- I hit this call traces by running some test cases when unmounting the fs mount points.
It seems there still have s...
02/18/2020
- 10:35 PM Feature #44189 (Resolved): support asynchronous unlink
- kernel side of https://tracker.ceph.com/issues/38951
Jeff has been working on this all along. - 12:48 PM Bug #44166: Client fails to release capabilities, lazy collective shared file I/O (simplified PoC...
- I can reproduce the same issue on the following kernel:
kernel-ml-5.5.4-1.el7.elrepo.x86_64 - 03:33 AM Bug #44177 (Fix Under Review): kclient: BUG: KASAN: use-after-free in __ceph_open_session+0x2a9/0...
- 03:33 AM Bug #44177: kclient: BUG: KASAN: use-after-free in __ceph_open_session+0x2a9/0x370 [libceph]
- The fixing patch: https://patchwork.kernel.org/patch/11387803/
- 03:28 AM Bug #44177 (In Progress): kclient: BUG: KASAN: use-after-free in __ceph_open_session+0x2a9/0x370 ...
- When there has new osdmap comes, it will replace and release the old osdmap memory, but if the mount is still on the ...
- 03:26 AM Bug #44177 (Fix Under Review): kclient: BUG: KASAN: use-after-free in __ceph_open_session+0x2a9/0...
- When mounting the cephfs, hit the call trace once:...
02/17/2020
- 09:38 AM Bug #44166 (New): Client fails to release capabilities, lazy collective shared file I/O (simplifi...
- Hello,
when doing collective shared file I/O on a file opened in Lazy mode, the clients seem to fail to release ca...
02/12/2020
- 05:20 PM Support #43809: ceph osd server issues (page allocation failure)
- Ah thanks, I'll move away the mounts from the osds :)
- 02:59 AM Bug #44091 (Fix Under Review): mount.ceph: some options will be truncated and couldn't see any er...
- 02:59 AM Bug #44091: mount.ceph: some options will be truncated and couldn't see any error in dmesg
- The fixing PR:
https://github.com/ceph/ceph/pull/33197
It dues to the mount.ceph was trying to parse the mount op... - 02:56 AM Bug #44091 (In Progress): mount.ceph: some options will be truncated and couldn't see any error i...
- 02:56 AM Bug #44091 (Resolved): mount.ceph: some options will be truncated and couldn't see any error in d...
- I was trying to add some new options to the kclient and testing it with "remount", but I couldn't see them was passed...
02/11/2020
- 06:58 AM Bug #44061 (Fix Under Review): kclient: new mount API will always see the posix acl is enabled
- 06:58 AM Bug #44061: kclient: new mount API will always see the posix acl is enabled
- The patchwork: https://patchwork.kernel.org/patch/11374885/
02/10/2020
- 02:16 PM Bug #44061 (In Progress): kclient: new mount API will always see the posix acl is enabled
- 02:14 PM Bug #44061 (Resolved): kclient: new mount API will always see the posix acl is enabled
- The posix acl is not settable after new mount API is introduced.
Also available in: Atom