Fix large download stuck with H2 priority#5835
Closed
shinrich wants to merge 1 commit intoapache:masterfrom
Closed
Fix large download stuck with H2 priority#5835shinrich wants to merge 1 commit intoapache:masterfrom
shinrich wants to merge 1 commit intoapache:masterfrom
Conversation
c3a2a99 to
f9a7014
Compare
Member
Author
|
[approve ci autest] |
Member
Author
|
[approve ci clang-analyzer] |
Member
|
I cannot reproduce the issue on master. Can you confirm this is an issue in public version too? Also, there is a similar function, |
Member
Author
|
Indeed, it looks lovely on master. |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Was trying enabling priorities with HTTP/2 again. Fetching a 1.2MB file from cache would frequently (almost always) fail. It would send the first part and then stall. Eventually things would timeout and abort.
Looking at the debug messages, we saw that many data frames were sent. Then we would hit "No payload". Nothing more would happen on that thread until 30 seconds later when the state machine would cleanup because of inactivity timeout.
Talking with @SolidWallOfCode it appears that we need to signal WRITE_READY to the cache vc after we drain the buffer, so it will write more data into the buffer. This is done in the non-priority case.
I saw all this on our internal branch, so I am not 100% sure this is an issue in open source. But I'm heading out for a week, so I wanted to push this issue up before I left in case it is a general issue on open source master.
I saw some other irregularities with the priority processing which I don't have a fix for. Will file an issue for that.