You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We have a WAF that has been deployed to dev for quite a while, but as we move it to sandbox, we noticed that it causes the envoy admin server to start becoming unresponsive. We use our admin endpoint to serve /ready and /server_info for liveness and readiness healthchecks, and when this happens, Kubernetes starts killing the healthy pods.
The container doesn't seem to be resource constrained when this is happening, but I wonder if it's something like envoyproxy/envoy#16425.
We are still in the midst of debugging, but posting here in case someone has ran into this before.