Unload taskQueueManager by instance, not queue ID #1917
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
yiminc
approved these changes
Sep 13, 2021
yiminc
reviewed
Sep 13, 2021
| @@ -201,7 +201,7 @@ func (w *taskWriter) appendTasks( | |||
| return resp, nil | |||
|
|
|||
| case *persistence.ConditionFailedError: | |||
| w.tlMgr.signalFatalProblem(w.tlMgr.taskQueueID) | |||
| w.tlMgr.signalFatalProblem(w.tlMgr) | |||
There was a problem hiding this comment.
BTW, if we were just call w.tlMgr.Unload() here we won't have to deal with this bug.
There was a problem hiding this comment.
Not sure what you mean - taskQueueManager doesn't have an Unload?
There was a problem hiding this comment.
tlMgr.Stop maybe? that would work because it dedups on a per-taskQueueManager instance basis.
There was a problem hiding this comment.
I meat tlm.Stop(), but yes we still need to remove it from the collection of matching engine, which is keyed on taskQueueID, and dedupe would still needed. So, maybe it does not matter.
Matching engine can incorrectly unload a taskQueueManager instance because the fatal signals from a previous instance are still being delivered and reference the same queue ID. To address we unload by taskQueueManager object instance rather than simply by queue ID.
yiminc
pushed a commit
that referenced
this pull request
Sep 13, 2021
Matching engine can incorrectly unload a taskQueueManager instance because the fatal signals from a previous instance are still being delivered and reference the same queue ID. To address we unload by taskQueueManager object instance rather than simply by queue ID.
|
Russian Revolution PR: https://en.wikipedia.org/wiki/Russian_Revolution |
|
☮️ 🏞️ 🍞 |
alexshtin
pushed a commit
that referenced
this pull request
Sep 16, 2021
Matching engine can incorrectly unload a taskQueueManager instance because the fatal signals from a previous instance are still being delivered and reference the same queue ID. To address we unload by taskQueueManager object instance rather than simply by queue ID.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
What changed?
Unload taskQueueManager by instance, not queue ID
Why?
Matching engine can incorrectly unload a taskQueueManager instance
because the fatal signals from a previous instance are still being
delivered and reference the same queue ID. To address we unload by
taskQueueManager object instance rather than simply by queue ID.
How did you test it?
New unit test
Potential risks
Minimal - unload occurs in strictly fewer cases
Is hotfix candidate?
no