Check_MK Agent causing high Disk IO

Good to know. Maybe the newer version of the agent .deb package didn’t remove older configuration properly? We’re using an older version of the official ansible roles that we somewhat fixed in 2020/2021 to work in our env.

Normal operation with systemd. Testing currently in a tmux session with watch -n 60 (local ssh).

Thanks for confirming something is not right. I’ll completely remove all cmk agent files and install the latest agent manually via the .deb package.

We have the issue currently on PVE with local ZFS, local LVM-Thin and Ceph storage. But I suspect something wrong with the agent setup then.
If you have any more hints we should have a look for that would be great.