fix: consume part of StreamingResponseIterator to support failure while under a retry context #10206
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
googlebot
added
the
cla: yes
crwilcox
added
the
do not merge
| try: | ||
| self._first_result = six.next(self._wrapped) | ||
| self._stored_first_result = True | ||
| except StopIteration: |
There was a problem hiding this comment.
Will this work for the WatchStream? For Watch, the client has to send the first request. We will not receive a response until after this request has been processed.
There was a problem hiding this comment.
Watch uses ResumableBidiRpc and has its own way of managing recovery. This will have an effect on things like query though.
tseaver
previously requested changes
Jan 24, 2020
|
NOTE: 3 tests look to need refactoring, also should likely improve coverage given this new behavior. |
crwilcox
commented
Jan 27, 2020
| # to retrieve the first result, in order to fail, in order to trigger a retry. | ||
| try: | ||
| self._stored_first_result = six.next(self._wrapped) | ||
| except TypeError: |
There was a problem hiding this comment.
@tseaver Under test we have a non-iterable. While it is odd that you would call this without it does seem possible? So added a guard here
crwilcox
changed the title
crwilcox
removed
the
do not merge
busunkim96
approved these changes
Jan 29, 2020
This was referenced Jan 31, 2020
Closed
Closed
[CHANGE ME] Re-generated firestore to pick up changes in the API or client library generator.
#10291
Closed
Closed
Closed
This was referenced Feb 4, 2020
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
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.
In the Node.js SDK, there's code that explicitly waits until at least one result has been observed and retries the underlying stream until then. This doesn't handle the case where the connection is interrupted mid-stream but has been good enough to resolve this class of issues. https://github.com/googleapis/nodejs-firestore/blob/master/dev/src/index.ts#L1127
The Firestore Python SDK needs equivalent code to make it resilient to this class of failure.
This PR causes init of the streaming response iterator to consume the first result immediately which will cause the error to occur while under the retry context. Once the iterator is created it is returned and future access is no longer in a retry context.