Alice Frosi

It would make sense that it is a bug in this PR as I changed the `internal/lib/stats/stats_server.go`. I really need to reproduce the issue locally :/

@saschagrunert We might still rely on cadvisor as I can see in kubelet log the error and events `failed to get imageFs info: non-existent label "crio-images` and this correspond to...

ok, reading a bit cadvisor code, I'm worried that the multi storage won't work. It seems they get the containers/pod information directly from the filesystem :/ not form CRI APIs....

I think the bug comes from some wrong info sent by crio. Cadvisor has a crio client where [it should get the root of the storage driver](https://github.com/google/cadvisor/blob/master/fs/fs.go#L299), but this is...

The last commit finally fixes the issue for the autoscale test :relieved: /test e2e-gcp

@saschagrunert yes, I can there is some mismatch how the documentation is generated on my fedora laptop. However, I'd like to discuss about the problem with cadvisor I mentioned in...

/retest

@haircommander many thanks for the review! I'll try to address your comments asap :)

> Is the iterator pattern (initialize(), next(), ...) being made safe somehow, or will two goroutines calling into the ListAllImages() method on the same multiStoreServer at the same time be...

I added 2 commits that: 1. a4537e16fa23508097563f39fdd4d3e4ba5a40c4 : fix @giuseppe 's [comment](https://github.com/cri-o/cri-o/pull/5624#discussion_r914800770) about the image deletion 2. https://github.com/cri-o/cri-o/pull/5624/commits/576bd4108202df01549ebb7a905c62383b0ee040 : fix @nalind 's [comment](https://github.com/cri-o/cri-o/pull/5624#issuecomment-1176644801) about the iterator The remaining points to...