This repository has been archived by the owner on Nov 25, 2024.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Outdated suggestions cannot be applied.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.
The internal NATS instance is definitely convenient but it does have one problem: its lifecycle is tied to the Dendrite process. That means if Dendrite panics or OOMs, it takes out NATS with it. I suspect this is sometimes contributing to what people see with stuck streams, as some operations or state might not be written to disk fully before it gets interrupted.
Using
SyncAlways
means that NATS will effectively useO_SYNC
and block writes on flushes, which should improve resiliency against this kind of failure considerably. It might affect performance a little but shouldn't be significant.Also updates NATS to 2.10.20 as there have been all sorts of fixes since 2.10.7, including better
SyncAlways
handling.Signed-off-by: Neil Alexander git@neilalexander.dev