Locked snap/db

[t-persson]

I am having a problem with my ETCD cluster, deployed as a statefulset in Kubernetes (running old version v3.3.8) where the "/var/run/etcd/default.etcd/member/snap/db" file seems to be locked by someone else.

It is a 3 node cluster and all 3 nodes have stopped due to this:

2025-03-06 12:41:59.318770 W | etcdserver: another etcd process is using "/var/run/etcd/default.etcd/member/snap/db" and holds the file lock, or loading backend file is taking >10 seconds
2025-03-06 12:41:59.318818 W | etcdserver: waiting for it to exit before starting...

Is it possible to recover from this in any way?
Is this on the etcd side or maybe it is on the CSI(Container Storage Interface) side?
Has this been fixed in later versions of etcd?

Thank you