Open
Description
- We have Add OOMKills to count number of notebook kernel restarts #52, but I don't think that accounts for pod evictions due to memory shortage, and probably also not for pod evictions due to other reasons than memory shortage such as caused by a lack of ephemeral space on the node.
It would be great to have a dashboard conveying if pods have been evicted as well, and perhaps if possibly a category on why. I'm not in any way sure we can get informed about this, but I know its k8s Events emitted associated with this happening for example.