Waldemar Quevedo
Waldemar Quevedo
Thanks for the info @MoHKale I will try take a look at this during the holidays later this month.
Thanks for the contribution! Apologies on the delay! I seem to have missed the notification from this repo, but now picking up the change into my fork at https://github.com/wallyqs/org-ruby I...
Thanks for the report! Maybe we should as well set via Helm the GOMAXPROCS env var based on the limits definition in case they were set explicitly…
This would be a big change I think, but I would start with fixing this in the monitoring endpoint where this is a common issue.
Hi @Merovius, could you share an example of using `WithCancel` that fails? I think the only APIs that do not support it are `FlushWithContext ` and we can fix that...
Yes it would work with the blocking APIs like Request or NextMsg for example, and most of the JS context APIs too.
Sounds good to me, ok we can fix that
Possible fix for this: https://github.com/nats-io/nats.go/pull/650
OK I see now, because it is using a sync pull subscriber then the client wouldn't be able to let finish processing a message (as it is right now). Something...
@rwrz There is a `nats.FlusherTimeout` option that might help in this case to detect the write error faster instead of waiting for outstanding pings limit to hit first: https://github.com/nats-io/nats.go/blob/274aa57115bb5ba10c3db65a336334d5fdf90b7e/nats.go#L331