This message was deleted.

Checked this but no use https://github.com/longhorn/longhorn/issues/6688 Engine is v1 and offline building is disabled Longhorn version 1.5.4

you can force detach

Can you please share command or tell steps how to do it? Or guide to docs?

idk mate looks pretty self explenatory

hmm but that option is disabled

Are you longhorn admin?

yup

no clue why its disabled then

ok

@aloof-branch-69545 Could you help? Thanks

Thanks @little-kangaroo-65735 The issue got resolved https://github.com/longhorn/longhorn/issues/4988#issuecomment-1345676772 It was issue with iscsi-initiator-utils and container-selinux Following this workaround with all longhorn pods restart worked https://longhorn.io/kb/troubleshooting-volume-attachment-fails-due-to-selinux-denials/ Why restart all pods: even after successfully running the daemonset pod was not detaching So decided to restart pods and monitor We scaled down all out workload and did restart

That's great! To improve the troubleshooting experience for others in the future Can you share how did you find the root cause and the Knowledge Base to try?

Yup sure So I started digging into the issue Google searches for "force detach volume longhorn", "longhorn volume not detaching" From this I got to a github issue where it stated it is issue in iscsi https://github.com/longhorn/longhorn/issues/4988#issuecomment-1345676772 But the workaround was for openshift clusters We have the cluster hosted on VMs So after finding more I found this https://longhorn.io/kb/troubleshooting-volume-attachment-fails-due-to-selinux-denials/ So followed these steps

git clone <https://github.com/longhorn/longhorn.git>
cd longhorn
kubectl apply -f deploy/prerequisite/longhorn-iscsi-selinux-workaround.yaml

But still it was not detaching so restarted all pods under longhorn-system namespace Then it worked Basically the issue was with iscsi and selinux and our VMs were Redhat so I found it related

Thanks for the sharing!