no, there was a change, quoting the release notes
and changelog:Code:Reduce amplification when writing to the cluster filesystem (pmxcfs), by adapting the fuse setup and using a lower-level write method (issue 5728).
Thanks for the clarification and the change. So this would explain the observation of Brobert, right?
Nontheless the basic design of pmxcfs is still that it needs to write back to the sqlite file on a regular schedule to ensure configuration consistency or did I missunderstood something? I'm asking since (including in the thread which resulted in the change) people tend to think that PVE "trashes ssds" because it's buggy or misdesigned while in reality it's just not designed for consumer hardware (which I'm fine with btw) due to it's intended usage in enterprise environments.