Under the last version of Checkmk CRE, the Disk IO Summary of every docker container will go into UNKN status after a docker just restarted, until there are actually any IO being made. It used to be that way in Checkmk v2.0, and then it was “fixed” (or at least handled differently) at some point in Checkmk v2.1, and I noticed recently that the issue started occuring again.
I use Checkmk to monitor multiple containers and it’s pretty much the same for every container except the ones that do IO as soon as they start. One solution for me would be to just disable alerts for that service and for the Checkmk discovery service (it wants to remove the service until there are any IO) but that defeats the purpose.