dears, we are running into the "Slow SQL" query problem on k3s 1.26 and k3s frequently crashes - we searched issues extensively but we do not seem to be able to figure out how to compact/truncate the database. in the syslog there are compact entries, but over time there are more entries, not less

Can you share the specific logs you're seeing about compaction? If it's falling behind that generally means either your DB is too slow, or you have something churning the cluster with object modifications.