@[email protected] @[email protected] @[email protected] hopefully you’ve seen my other messages on Fedia. I think I’ve fixed the problem by deleting /m/fediverse and purging the backlog of deliveries. The threads mentioned above were delivered but due to the issues with /m/fediverse (and possibly /m/random) the server was running way behind because it was spending all its time retrying deliveries that would never be accepted by other instances. Sadly I think I’m going to have to leave /m/fediverse deleted till the underlying mbin bug gets fixed.
jerry
joined 8 years ago
@[email protected] I removed the block. I’ve been having some luck in reducing the server load today with the help of the mbin team.
@[email protected] messages to/from threads has been causing some major problems with the message processing queues on fedia.io. To keep the instance running, I had to block it until I can sort out why.
Being mentioned in the same breath as the others in the post above is wholly inappropriate. Those are the MVPs. I just pretend to know what I’m doing.
@IlliteratiDomine @DarkThoughts I found the problem - it was an error I introduced when moving fedia,io to containers recently. it works now.
@IlliteratiDomine @DarkThoughts Thanks for tagging me. I wasn't aware there was a problem. I'll take a look.
@Cavalarrr @ernest I did what all the cool kids are doing and disabled federation on fedia.io. The problem isn't really the local users as much as all the content being federated in.
@[email protected] @[email protected] @[email protected] I had to remove /m/test too, and now the delivery queues are holding at zero, as they should be.