Mahdi Gerami

Yeah‌, but not like channels. Like you press forward and then select multiple contacts and then send to them. No need to make a channel only for that one time...

 I also need to change style (overflow) of these children.

I also experienced this problem. In grafana I can see spikes where instance manager takes a lot of cpu cores at random points in time. I should check logs to...

https://drive.google.com/file/d/1Z4DMGnNoClkeYRPQcqSEQfwnyFjPFqPd/view?usp=sharing This is the support bundle I generated from longhorn-ui. The last two spikes were about 1 hour ago and 13 hours ago.

I checked our grafana monitoring around the spike times. One of our worker nodes had cpu, disk io and network bandwidth far below capacity but its memory usage was near...

> One question Do you know which process used up the memory? Your applications or Longhorn's processes? I couldn't find which process caused the problem but right before the problem...

> @mahdiG Can you provide your host infor, e.g. CPU model, ram size, storage configuration? Thank you. This is will be a setting reference in #4516. Our nodes are created...

> But all this means that for your case the cluster will not come back up as the first pod never gets ready without being able to form a quorum....

> As a hack to recover a cluster in this situation you can try to manually edit the statefulset (`cluster-masters` in your case) and temporarily set the `podManagementPolicy` to `Parallel`....