Ensure that errors are propagated after retries are exhausted. #258
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
added 5 commits
June 14, 2013 15:21
This prevents exceptions from occuring while trying to create a new transport in the pool, which happens outside a try/except. The result is the same behavior as HTTP, which delays opening a connection until the first request.
This prevents masking of errors when the loop falls through to its completion, all tries raising BadResource. The previous behavior was that the function doesn't return any value, essentially returning None.
Also, don't penalize a node that has some unspecified unretryable exception.
ghost
assigned
|
Looks awesome! 👍 |
|
Try as I might, I can't seem to find problems here. Test even works. 👍 |
seancribbs
pushed a commit
that referenced
this pull request
Jun 20, 2013
Ensure that errors are propagated after retries are exhausted.
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Outdated suggestions cannot be applied.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.
Before this change, as long as the raised error in a client operation was always eligible for retry, the result was that the retried operation would return
Noneafter hitting theRETRY_COUNT. This produced strange masking effects, including returning an empty list from MapReduce operations that failed.Now the last exception will be re-raised when the retries have been exhausted, which was the original desired behavior.
While investigating this behavior, I also discovered that the PB socket would attempt to connect while allocating a new element in the pool, which meant that connection would happen outside the
try ... exceptinsidePool.take(). This meant that failed connections would mean instant propagation of the connection error, avoiding the retry logic. The transport was changed to connect on first request so that it has analogous behavior to the HTTP transport.Closes #218.