Make sure to iterate over all keys when cancelling keys in a SelectorPool #4471
+5
−2
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Outdated suggestions cannot be applied.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 we updated our server from 9.1.2.0 to 9.1.7.0 we started seeing a file descriptor leak which, if left for long enough, would run the machine out of file descriptors.
We tracked down the problem to #3952 which changed how SelectionKeys were closed.
It looks like if exceptions aren't caught while calling
cancel
on the SelectionKeys then the loop can exit early and leave some keys un-cancelled.When we went back to the same key cancelling code that was used in 9.1.2.0 the problem went away.