Since i had a hard time finding out this problem:
My OZW logs where being spammed by a single node for such a big amount of data that the network of z-wave itself was grinding to a halt. A rule of thumb is that your z-wave devices (those that are not battery powered) should say “Ready” at some point. Mine where still in cacheload.
1 Node was constantly aquiring something, the logs was growing so big that journalctl could not keep up (the logging system)
Turns out that in home assistant you could see 1 node with the status “Probe” if i restarted/rebooted it was still in Probe status. I removed the Probe status from the zwcfg xml file and restarted and everything was ok again. Your home assistant will still work after a long bootup but all actions will be delayed/slow and unacceptable when it comes to automation. So its a sneaky one.