-
Notifications
You must be signed in to change notification settings - Fork 6k
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
nautilus: mds: do not defer incoming mgrmap when mds is laggy #36168
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
smithfarm
force-pushed
the
wip-46151-nautilus
branch
2 times, most recently
from
July 19, 2020 06:44
f2b5eb2
to
c660323
Compare
When the mds is laggy, the incoming mgrmap is queued to be processed at a later stage. But, the mds does not handle mgrmap message directly. So, later when the mds is not laggy anymore, the mgrmap message is not handled and is dropped. But, when the mgrmap message was queued up, the mds acknowledges that it has handled the message. This causes the mgr client instance to never process the mgrmap and never connecting to the manager (the receipt of mgrmap drives the connection to the manager). The fix is to not acknowledge messages that the mds cannot handle. In normal cases, the mds does not ack the message but when it's laggy, it just blindly queues up the message -- so, check if the message can be handled (later) even when the mds is laggy. Also, a minor change in a function name -- handle_deferrable_message() is kind of a misnomer since the function is called to process messages that are not deferred. That's changed to handle_message() now. Fixes: http://tracker.ceph.com/issues/44638 Signed-off-by: Venky Shankar <vshankar@redhat.com> (cherry picked from commit beb12fa) Conflicts: src/mds/MDSDaemon.cc src/mds/MDSRank.cc src/mds/MDSRank.h - nautilus has "Message::const_ref" instead of "cref_t<Message>"
smithfarm
force-pushed
the
wip-46151-nautilus
branch
from
July 19, 2020 08:25
c660323
to
1de4565
Compare
@smithfarm is this still DNM? |
test this please |
ajarr
approved these changes
Oct 5, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
backport trackers: https://tracker.ceph.com/issues/46151
backport of #34024
parent trackers: https://tracker.ceph.com/issues/44638
this backport was staged using ceph-backport.sh version 15.1.1.389
find the latest version at https://github.com/ceph/ceph/blob/master/src/script/ceph-backport.sh