This repository has been archived by the owner on Apr 26, 2024. It is now read-only.
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Outdated suggestions cannot be applied.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.
Fix a bug introduced in Synapse v1.50.0rc1 whereby outbound federation could fail because too many EDUs were produced for device updates. #11730
Fix a bug introduced in Synapse v1.50.0rc1 whereby outbound federation could fail because too many EDUs were produced for device updates. #11730
Changes from 12 commits
e3665f4
9f54d0f
ad8da37
d17de19
a306a89
cc8904e
2cf1691
aceb240
451b4ff
34ae0ba
62e8e45
2107d12
6643b94
8b01179
b5dc363
0476aee
57b3601
File filter
Filter by extension
Conversations
Jump to
There are no files selected for viewing
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If we drop these edus here (or the other one on +329), what ensures they'll be picked up the next time we come to work out which device updates need sending over federation?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We only advance the stream position as far as the loop goes (see a little bit down). I added a comment to it to make it more noticeable.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ahh, I think I see. We tell the caller to update the stream id to
last_processed_stream_id
via return value. This only gets bumped at the end of a the loop body overupdates
.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What's going on here? Could we have problems if the same key occurs twice in
updates
? (Ahh, looks like that'll never happen because of how it's sourced from the DB?)There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think it's intentional: later ones overwrite old ones (hence this
if
-block)There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I was just worried that this might throw away a previous
update_context
, which sounds bad. But we're not changing that soo... maybe we should just leave itThere was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That's 'just' the opentracing span. Maybe I'll put a FIXME note in there about it; not sure what the right answer is at this instant — sounds like it could make it hard to trace things down which is not great, but it's not release blocking considering it was always like this ...