Back to bug 2091491

Who When What Removed Added
Xiubo Li 2022-05-30 06:07:18 UTC Target Release 6.1 5.3
Assignee vshankar xiubli
Red Hat One Jira (issues.redhat.com) 2022-05-30 06:09:57 UTC Link ID Red Hat Issue Tracker RHCEPH-4415
Xiubo Li 2022-05-30 06:10:56 UTC Link ID Github ceph/ceph/pull/45955
Xiubo Li 2022-05-30 06:11:28 UTC Link ID Ceph Project Bug Tracker 55240
Veera Raghava Reddy 2022-05-31 21:23:31 UTC CC vereddy
Severity unspecified high
Xiubo Li 2022-07-11 07:36:38 UTC Status NEW POST
Target Release 5.3 6.1
errata-xmlrpc 2022-08-15 21:13:36 UTC Status POST MODIFIED
CC tserlin
Target Release 6.1 6.0
Fixed In Version ceph-17.2.3-1.el9cp
Status MODIFIED ON_QA
Hemanth Kumar 2022-08-18 05:42:33 UTC QA Contact hyelloji ymane
Yogesh Mane 2022-08-18 06:47:03 UTC Flags needinfo?(xiubli)
Xiubo Li 2022-08-22 03:12:20 UTC Flags needinfo?(xiubli)
Yogesh Mane 2022-08-24 07:52:21 UTC Status ON_QA VERIFIED
Eliska 2022-09-07 06:16:57 UTC Flags needinfo?(xiubli)
CC ekristov
Xiubo Li 2022-09-07 08:31:18 UTC Flags needinfo?(xiubli)
Doc Text Cause:

With multiple active mds case, client could send a getattr client request to a replica MDS just after it was created.

Consequence:

For the setattr client request the client will make a path of '#INODE-NUMBER' because the CInode hasn't been linked yet, and the replica MDS will keep retrying until the auth MDS flushes the mdlog and the C_MDS_openc_finish and link_primary_inode are called at most 5 seconds later.

Fix:

When the replica MDS trying to find the ino from auth MDS and if couldn't find it then manually trigger mdslog flush.

Result:

No stuck any more.
Doc Type If docs needed, set a value Bug Fix
Eliska 2022-09-13 08:53:03 UTC Blocks 2126050
Eliska 2022-09-21 13:09:44 UTC Flags needinfo?(xiubli)
Eliska 2022-09-22 14:33:57 UTC Flags needinfo?(xiubli)
Eliska 2022-09-23 07:34:22 UTC Doc Text Cause:

With multiple active mds case, client could send a getattr client request to a replica MDS just after it was created.

Consequence:

For the setattr client request the client will make a path of '#INODE-NUMBER' because the CInode hasn't been linked yet, and the replica MDS will keep retrying until the auth MDS flushes the mdlog and the C_MDS_openc_finish and link_primary_inode are called at most 5 seconds later.

Fix:

When the replica MDS trying to find the ino from auth MDS and if couldn't find it then manually trigger mdslog flush.

Result:

No stuck any more.
.A replica MDS is no longer stuck, if a client sends a getattr client request just after it was created

Previously, if a client sent a getattr client request just after the replica MDS was created, the client would make a path of `#INODE-NUMBER` because the CInode was not linked yet.
The replica MDS would keep retrying until the auth MDS flushed the `mdlog` and the `C_MDS_openc_finish` and `link_primary_inode` were called 5 seconds later at most.

With this fix, the replica MDS trying to find the `handle_find_ino` from auth MDS would manually trigger `mdslog flush`, if it could not find it.
Flags needinfo?(xiubli)
Docs Contact ekristov
Xiubo Li 2022-10-08 02:36:54 UTC Flags needinfo?(xiubli) needinfo?(xiubli) needinfo?(xiubli)
Eliska 2022-10-11 09:40:14 UTC Flags needinfo?(xiubli)
Doc Text .A replica MDS is no longer stuck, if a client sends a getattr client request just after it was created

Previously, if a client sent a getattr client request just after the replica MDS was created, the client would make a path of `#INODE-NUMBER` because the CInode was not linked yet.
The replica MDS would keep retrying until the auth MDS flushed the `mdlog` and the `C_MDS_openc_finish` and `link_primary_inode` were called 5 seconds later at most.

With this fix, the replica MDS trying to find the `handle_find_ino` from auth MDS would manually trigger `mdslog flush`, if it could not find it.
.A replica MDS is no longer stuck, if a client sends a getattr client request just after it was created

Previously, if a client sent a getattr client request just after the replica MDS was created, the client would make a path of `#INODE-NUMBER` because the `CInode` was not linked yet.
The replica MDS would keep retrying until the auth MDS flushed the `mdlog` and the `C_MDS_openc_finish` and `link_primary_inode` were called 5 seconds later at most.

With this fix, the replica MDS trying to find the `CInode` from auth MDS would manually trigger `mdslog flush`, if it could not find it.
Eliska 2022-10-11 09:43:16 UTC Doc Text .A replica MDS is no longer stuck, if a client sends a getattr client request just after it was created

Previously, if a client sent a getattr client request just after the replica MDS was created, the client would make a path of `#INODE-NUMBER` because the `CInode` was not linked yet.
The replica MDS would keep retrying until the auth MDS flushed the `mdlog` and the `C_MDS_openc_finish` and `link_primary_inode` were called 5 seconds later at most.

With this fix, the replica MDS trying to find the `CInode` from auth MDS would manually trigger `mdslog flush`, if it could not find it.
.A replica MDS is no longer stuck, if a client sends a getattr client request just after it was created

Previously, if a client sent a getattr client request just after the replica MDS was created, the client would make a path of `#INODE-NUMBER` because the `CInode` was not linked yet.
The replica MDS would keep retrying until the auth MDS flushed the `mdlog` and the `C_MDS_openc_finish` and `link_primary_inode` were called 5 seconds later at most.

With this fix, the replica MDS trying to find the `CInode` from auth MDS would manually trigger `mdslog flush`, if it could not find it.
Xiubo Li 2022-10-11 12:20:39 UTC Flags needinfo?(xiubli)
Red Hat Bugzilla 2022-12-31 16:21:24 UTC Assignee xiubli vshankar
Red Hat Bugzilla 2022-12-31 19:46:56 UTC QA Contact ymane hyelloji
Red Hat Bugzilla 2022-12-31 19:50:33 UTC QA Contact hyelloji
Red Hat Bugzilla 2023-01-01 05:39:36 UTC CC tserlin
Red Hat Bugzilla 2023-01-01 08:48:15 UTC CC vereddy
Red Hat Bugzilla 2023-01-01 08:49:26 UTC Assignee vshankar nobody
Alasdair Kergon 2023-01-04 04:55:29 UTC QA Contact ymane
Alasdair Kergon 2023-01-04 04:55:53 UTC Assignee nobody xiubli
Alasdair Kergon 2023-01-04 06:25:53 UTC CC tserlin
Alasdair Kergon 2023-01-04 06:59:12 UTC CC vereddy
Red Hat Bugzilla 2023-01-09 08:30:48 UTC CC ceph-eng-bugs
Alasdair Kergon 2023-01-09 19:43:36 UTC CC ceph-eng-bugs
errata-xmlrpc 2023-03-20 18:37:49 UTC Status VERIFIED RELEASE_PENDING
errata-xmlrpc 2023-03-20 18:56:39 UTC Status RELEASE_PENDING CLOSED
Resolution --- ERRATA
Last Closed 2023-03-20 18:56:39 UTC
errata-xmlrpc 2023-03-20 18:57:06 UTC Link ID Red Hat Product Errata RHBA-2023:1360

Back to bug 2091491