[BUG] Microshift fails to start on Fedora 36 with LUKS encrypted disk

[jiridanek]

• reopen of https://github.com/openshift/microshift/issues/629

W0810 12:42:19.144863       1 fs.go:588] stat failed on /dev/mapper/luks-63cca6c4-98e1-467a-b8ee-acfac51b19ca with error: no such file or directory
E0810 12:42:19.144874       1 kubelet.go:1389] "Failed to start ContainerManager" err="failed to get rootfs info: failed to get device for dir \"/var/lib/kubelet\": could not find device with major: 0, minor: 34 in cached partitions map"

See various other incarnations of this for different kubernetes distro projects

• https://github.com/kubernetes-sigs/kind/blob/main/site/content/docs/user/known-issues.md#failed-to-get-rootfs-info--stat-failed-on-dev
  • https://github.com/kubernetes-sigs/kind/issues/2411

[jiridanek]

For a workaround, it seems possible to follow instructions in https://github.com/openshift/microshift/blob/main/docs/devenv_rhel8.md#running-microshift which tells to run the microshift binary outside of podman, directly on the machine.

[dhellmann]

We're focusing on RHEL support for the moment and the main branch has moved on quite a bit from the 4.8 release branch. You could try compiling the main branch, but it may be a few months before the team returns to focusing on Fedora support.

[jiridanek]

@dhellmann does CentOS Stream 8 count? I can reproduce the issue on CentOS Stream

[dhellmann]

Are you using a build from the main branch or the 4.8 branch?

[jiridanek]

main branch, here is the issue filled https://github.com/openshift/microshift/issues/880