Kevin Nisbet

As of Nov 2nd, still needs to be ported to 7.0. Seems to be on all other active release branches.

We're actually moving in the direction of dropping the use of serf entirely for cluster membership: https://github.com/gravitational/satellite/pull/284 Not on 7.0 yet though.

It's in the 7.1 alpha: https://goteleport.com/gravity/docs/changelog/#internal-changes

As slight detail to add, the etcd sync code, atleast at one point, detected whether etcd was running when choosing whether or not to sync the state (to avoid sitting...

I was reading through the ticket Dima referenced, and I think I have a theory about what could've happened. Looking at the screenshot of server -> LB, it looks like...

I don't believe this is currently available in the OSS version through configuration. I'll leave this ticket open so we do address it. As a workaround, you might be able...

So I was reading the Kubernetes issue for this same problem, and they turned up a really interesting concern: https://github.com/kubernetes/kubernetes/pull/96502#issuecomment-734847096 Basically the HSTS header is set at a node level,...

I did a cursory glance at some issues, and the ones I've found appear to suggest that this can occur when the etcd files are legitimately missing or corrupted. So...

Well that panic AFAIK is telling you etcd is about to lose data and some sort of file corruption has occurred, so recovering and ignoring the panic are at best...

At the same time: - should mask the systemd-rfkill.socket / systemd-rfkill.service since they're not used - mask any other services / sockets / resources systemd adds by default that we...