Byron Ruth

What did work is this kind of config for a leaf in non-operator mode, which makes sense since its a separate auth model for the leaf. ``` server_name: leaf1 port:...

Hi @ramonberrutti. What concrete use case drove this proposal? You note sourcing. There are messages being sourced that you either want to add or remove headers? My concern is that...

I will fix the test.. it is an existing one that is doing a *contains string* match for the existing accstatz payload.

We have a PR coming this week that addresses a number of issues based on the k8s conformance tests. This may still be the happy path, but it will rule...

I referenced the issue above, #4138. I stumbled upon it again, so I took a few minutes to investigate what it would take to support it (hence the draft PR)...

Ah there are a few other assumptions in the code on implicitly registering and reloading. Will revisit later.

> Because all the replicas do the last sequence and last msg ID checks on their own. @MauriceVanVeen why doesn't just the leader do this? > I think we will...

> I would imagine the use case you mention will be more popular vs staging all the messages in a slice/array first. Be interested in what others think as well....

Annnd.. I just realized I was a bit redundant with the other commentary 🤦. Apologies folks..

Yes. Been delayed with getting back to it, but I will have time at the end of the week.