Past Incidents

We are planning to move all AdminDB hosts currently located on the VM host “cocaine” to the newer host “medikinet”. This should improve the performance of the VMs and allows us to decommission the old VM host “cocaine”, which is now known for causing stability and performance problems to it's VMs.

We are currently experiencing problems with new devices interacting with DHCP. This leads to newly registered devices being unable to connect to some websits (i.e. those which don't offer only IPv4 connectivity). We are investigating the problem.

Update 21:47

We were able to pin down the issue to the filesystem of the active DHCP node being corrupted and thus the configuration for newly registered devices unable to be written. For now, we have switched over to the backup DHCP node, so DHCP operation should now be working as intended. We are still working on bringing up the primary node again.

Update 2023-01-03 16:51

The primary node is back online. We have noticed that parts of the filesystem have been corrupted, but it seems the node is working as intended again.