It appears to have recovered now. From the Grafana dash, it's unclear to me what led to this issue. I don't see e.g. a influx of many or slow jobs that would explain a backlog in processing.

I do see that the 20-25 min outage affected all the low-traffic jobs, which supports the theory that it wasn't caused by something in that job group blocking the processing of other jobs in that group:

Is there more to do in this task? Should this still be open? Will there be an incident report? If yes, who is planning to work on it?

It is pending incident documentation by Core Platform, in particular to have documented what caused it, what we learned from investigation/mitigation, and to file tasks for any follow-up/prevention. I'll move it to their Inbox as I now understand Clinic Duty Inbox is not used.

Moving onto the board for indicent docs (Wikimedia-Incident is only for the incident itself and/or action items to improve stuff). Note that a document for this one has not yet been created yet.