ReconfigureAsync: fix re-entrancy issue #1772
Merged
Conversation
This file contains hidden or 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
When this was refactored long ago, we went from a counter starting at 1 (which handled the first run) to a string reason (which defaults to empty). Combined with the || clause format, this subtly introduced a re-entrancy bug where a subsequent run can enter during the first run (where first was true, so no exchange for reason occurred). This was causing several Sentinel issues due to on connect handlers setting config as nodes returned and ultimately the Set<T> triggering a ReconfigureIfNeeded....and actually running, while the original config was still in progress. This led to all sorts of race oddness especially if one endpoint was any kind of significantly different in timing than another.
Collaborator
Author
|
cc @TimLovellSmith on this fix - not the only Sentinel issue but it was one of the NoConnectionAvailable sources, because a re-entrancy race improperly set both nodes as master yielding an unselectable replica in the DemandReplica path. In reality, this probably caused far more production connection problems. |
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Outdated suggestions cannot be applied.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.
When this was refactored long ago, we went from a counter starting at 1 (which handled the first run) to a string reason (which defaults to empty). Combined with the
||clause format, this subtly introduced a re-entrancy bug where a subsequent run can enter during the first run (where first was true, so no exchange for reason occurred).This was causing several Sentinel issues due to on connect handlers setting config as nodes returned and ultimately the
Set<T>triggering aReconfigureIfNeeded....and actually running, while the original config was still in progress. This led to all sorts of race oddness especially if one endpoint was any kind of significantly different in timing than another.Note: the other changes are only labels on the bool arguments for clarity at the call sites.