I had a situation where seq was being flooded by a very short interval retry loop;
We could tell by disk space consumption, but the events from the ui seemed normal, so we had no way of telling which service was causing the issue.
When we eventually figured out which service it was and turned it off, all of those messages showed up.
Obviously someone made a pretty big mistake to begin with, but it would have been a lot more helpful to have been able to see those events, is there some reason that it behaved this way? I'm assuming seq was waiting for a break in ingest to process but that break never came?