Fallback to next urls if download fails in HttpDownloader #9015
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
jin
added
the
team-ExternalDeps
katre
pushed a commit
that referenced
this pull request
Aug 6, 2019
Fixes #8974. `HttpDownloader` never retried `IOException` that could have occurred during `ByteStreams.copy(payload, out)`, HttpConnector would have retried connection phase errors but not payload ones. This chageset adds fallback to next url(s) if present for both payload read errors and connection errors and adds (completely) missing tests for `HttpDownloader`. Note, that this PR technically disables questionable `HttpConnectorMultiplexer` optimization that attempts to connect to multiple urls at the same time (by starting threads that race with each other) and picking the one that connected faster. There is a way to keep that optimization while falling back to next urls, but it would require all exceptions to contain url that caused it so that `HttpDownloader` could retry download on other urls. I think making `HttpDownloader` more reliable and actually using provided `urls` as fallback is much more important than mentioned optimization. Closes #9015. PiperOrigin-RevId: 261702678
katre
pushed a commit
that referenced
this pull request
Aug 7, 2019
Fixes #8974. `HttpDownloader` never retried `IOException` that could have occurred during `ByteStreams.copy(payload, out)`, HttpConnector would have retried connection phase errors but not payload ones. This chageset adds fallback to next url(s) if present for both payload read errors and connection errors and adds (completely) missing tests for `HttpDownloader`. Note, that this PR technically disables questionable `HttpConnectorMultiplexer` optimization that attempts to connect to multiple urls at the same time (by starting threads that race with each other) and picking the one that connected faster. There is a way to keep that optimization while falling back to next urls, but it would require all exceptions to contain url that caused it so that `HttpDownloader` could retry download on other urls. I think making `HttpDownloader` more reliable and actually using provided `urls` as fallback is much more important than mentioned optimization. Closes #9015. PiperOrigin-RevId: 261702678
katre
pushed a commit
that referenced
this pull request
Aug 12, 2019
Fixes #8974. `HttpDownloader` never retried `IOException` that could have occurred during `ByteStreams.copy(payload, out)`, HttpConnector would have retried connection phase errors but not payload ones. This chageset adds fallback to next url(s) if present for both payload read errors and connection errors and adds (completely) missing tests for `HttpDownloader`. Note, that this PR technically disables questionable `HttpConnectorMultiplexer` optimization that attempts to connect to multiple urls at the same time (by starting threads that race with each other) and picking the one that connected faster. There is a way to keep that optimization while falling back to next urls, but it would require all exceptions to contain url that caused it so that `HttpDownloader` could retry download on other urls. I think making `HttpDownloader` more reliable and actually using provided `urls` as fallback is much more important than mentioned optimization. Closes #9015. PiperOrigin-RevId: 261702678
philwo
added
the
team-OSS
philwo
removed
the
team-OSS
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.
Fixes #8974.
HttpDownloadernever retriedIOExceptionthat could have occurred duringByteStreams.copy(payload, out), HttpConnector would have retried connection phase errors but not payload ones.This chageset adds fallback to next url(s) if present for both payload read errors and connection errors and adds (completely) missing tests for
HttpDownloader.Note, that this PR technically disables questionable
HttpConnectorMultiplexeroptimization that attempts to connect to multiple urls at the same time (by starting threads that race with each other) and picking the one that connected faster. There is a way to keep that optimization while falling back to next urls, but it would require all exceptions to contain url that caused it so thatHttpDownloadercould retry download on other urls. I think makingHttpDownloadermore reliable and actually using providedurlsas fallback is much more important than mentioned optimization.