I am using k3s v1.29.5+k3s1 bundled with etcd v3.5.9-k3s1 And today, for the first time ever, I noticed one of my server nodes was down. After further investigation I am seeing the nodes logs showing: “Panic: runtime error: slice bounds out of range [:8] with capacity 5” This panic function bytesToRev() means the etcd data is corrupted. The slice bounds out of range is a known etcd bug in some 3.5 version where incomplete WAL entry or partial snapshot leads to panic on startup. I am wondering . What version of k3s and etcd should I upgrade to, that has addressed this bug… Thankyou

That is a fairly old version. Youll need to step though 1.30, 1.31, and so on. You will also need to rejoin the node to the cluster, or restore from a snapshot, in order to address the corrupted etcd data.