Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

jewel: cephfs: Journaler may execute on_safe contexts prematurely #16192

Closed
wants to merge 2 commits into from

Conversation

smithfarm
Copy link
Contributor

John Spray and others added 2 commits July 6, 2017 22:42
This asserted that flush_pos would be ahead of
safe_pos after calling _flush.  However, this
is not guaranteed to be the case because
prezeroing might prevent us from flushing
right now.

Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 58ec1c6)
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
(cherry picked from commit 41618d8)
@smithfarm smithfarm self-assigned this Jul 6, 2017
@smithfarm smithfarm added this to the jewel milestone Jul 6, 2017
@smithfarm smithfarm changed the title jewel: Journaler may execute on_safe contexts prematurely jewel: cephfs: Journaler may execute on_safe contexts prematurely Jul 6, 2017
@smithfarm smithfarm added cephfs Ceph File System and removed core labels Jul 6, 2017
@smithfarm
Copy link
Contributor Author

@jcsp This completes what was started by #15468 - I hope it's not a problem that 10.2.8 contains only the first half of the fix.

@jcsp
Copy link
Contributor

jcsp commented Jul 7, 2017

Hmm, I can't remember how bad the issue was that needed the followup commit: @ukernel ?

@ukernel
Copy link
Contributor

ukernel commented Jul 12, 2017

it's better to also cherry pick 146383a.

@ukernel
Copy link
Contributor

ukernel commented Jul 12, 2017

The whole series fix an issue in purge queue. we haven't back-ported purge queue to jewel. So another option is reverting #15468

@smithfarm
Copy link
Contributor Author

@ukernel @jcsp So this should be considered a blocker for 10.2.8? In other words, we need to quickly do a 10.2.9 to clean this up?

@ukernel
Copy link
Contributor

ukernel commented Jul 12, 2017

the bug can cause damaged mds. it should be blocker for 10.2.8.

@smithfarm
Copy link
Contributor Author

Since we tend towards not backporting features, and the purge queue sounds like a feature, I propose the following course of action:

Alternatively, we could cherry-pick 146383a into this PR, merge it, and cut a 10.2.9 with this PR?

smithfarm added a commit to smithfarm/ceph that referenced this pull request Jul 12, 2017
See the discussion in ceph#16192

Signed-off-by: Nathan Cutler <ncutler@suse.com>
@smithfarm
Copy link
Contributor Author

http://tracker.ceph.com/issues/20536 was rejected

@smithfarm smithfarm closed this Jul 12, 2017
smithfarm added a commit to smithfarm/ceph that referenced this pull request Jul 13, 2017
See the discussion in ceph#16192

Signed-off-by: Nathan Cutler <ncutler@suse.com>
dmick pushed a commit to dmick/ceph that referenced this pull request Jul 18, 2017
See the discussion in ceph#16192

Signed-off-by: Nathan Cutler <ncutler@suse.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug-fix cephfs Ceph File System
Projects
None yet
3 participants