Nir Soffer

Is this a real issue? do we need timeouts or cancellation support in the tests? If not the best way is to disable the linter setting or the entire linter....

> Why not wait until we switch to Go 1.24, apply this PR, and be done with it? Sounds good but it does not solve the linter issue.

> > Sounds good but it does not solve the linter issue. > > Why wouldn't it? That is the main point of the PR... How waiting silences the linter?...

Also reproduce in our testing environment by replacing ceph image: https://github.com/RamenDR/ramen/pull/1836/commits/7c42caefafd0741030efcab0da297a79b6cdcea9

> What type of OSDs are you creating? Are you creating the OSDs on raw devices or partitions? We reproduce in 2 environments: - linux: minikube vm (kvm2 driver) with...

We tested now also ceph:main - the same issue is reproduced: ``` % kubectl rook-ceph --context dr1 ceph status Info: running 'ceph' command with args: [status] cluster: id: df4b469b-f857-48f3-b1d8-600740038790 health:...

More info from the vm with this issue: ``` # lsblk --json -O /dev/vdb { "blockdevices": [ { "alignment": 0, "id-link": null, "id": null, "disc-aln": 0, "dax": false, "disc-gran": "512B",...

Comparing to ceph:v19, we see similar "unable to decode label" errors in osd-prepare pod, but it continues. ceph:main ``` Running command: /usr/bin/ceph-osd --cluster ceph --osd-objectstore bluestore --mkfs -i 0 --monmap...

The log change comes from this commit: ``` commit 8c28695cde065235e23dbd7af827df22965cb76e Author: Adam Kupczyk Date: Tue Jan 21 17:12:01 2025 +0000 os/bluestore: Improve dout when bdev label cannot be decoded Remove...

> What exactly is your cephcluster CR? We used this kustomization to adapt the cluster to our environment. We added another patch to replace the ceph image. ```yaml --- resources:...