Emil Burzo

I've ran into the same issue and I suspect it's a regression, since an older version works.

@derekbit I'll try, I don't have any monitoring setup which is so granular or is there a native kubernetes way of getting this I might not know about?

## CPU -- longhorn - solid lines: `rate(container_cpu_usage_seconds_total{pod=~"longhorn-manager.*|instance-manager.*"}[5m])` - dashed lines: `node_load5`  ## I/O -- longhorn - solid lines: `rate(container_blkio_device_usage_total{pod=~"longhorn-manager.*|instance-manager.*"}[5m])` - dashed lines: `node_load5`  ## CPU - all...

> Is the load still high after detaching all Longhorn volumes? Unfortunately yes, even after stopping all resources and making sure all volumes are detached, load spikes still happen. >...

Alright, some initial results / observations: ## Spikes It seems I was wrong, the load spikes are **not** coming from longhorn, as they exist even in the clean state (just...

> synchronize the cache with etcd Not sure if it makes a difference, but I forgot to mention: I'm using a postgres instance on a non-k3s node, instead of etcd.