John McMeeking

Running on Kubernetes 1.12 Initial etcdcluster yaml: [example-tls-cluster.yaml.txt](https://github.com/coreos/etcd-operator/files/2751130/example-tls-cluster.yaml.txt) Updated etcdcluster yaml: [update-tls-cluster.yaml.txt](https://github.com/coreos/etcd-operator/files/2751132/update-tls-cluster.yaml.txt) End of etcd-operator log showing the update: [etcd-operator.log](https://github.com/coreos/etcd-operator/files/2751153/etcd-operator.log)

@hexfusion Creating a new cluster works fine.

Looks like the existing upgrade code can only change version - patches the image name and version. And after trying to edit the environment via `kubectl edit pod`, I think...

One perspective on this. We use the containerd stream idle timeout to implement a requirement that `kubectl exec` sessions be killed after being idle for 15 minutes. On the surface...

@mikebrow With our 15m stream_idle_timeout exec and logs seem to be behaving without the keep alive requests. Given that, would it be safe / less inadvisable to ignore keep alive...

Hi team. Any prognosis on this? It looks like things kind of came to a halt two week ago.