Incoming federated messages not being processed by mastodon instance

[theothermattm]

Hi there,

I'm struggling to get federation working between servers.

The situation I'm in is that occasionally (but not always) @ messages to other servers work (not always). However, follow requests do not seem to work either way.

I do see /inbox calls being made in my puma logs, which I'm assuming are from the remote server?

[b1a593b8-e9e6-4fbc-83d1-f366126ba867] method=POST path=/api/v1/accounts/109485949763964285/follow format=html controller=Api::V1::AccountsController action=follow status=200 duration=91.11 view=0.81 db=38.95
[8fd02a4b-b6cb-4a85-8869-c47f829b5ef0] method=POST path=/users/mattm/inbox format=html controller=ActivityPub::InboxesController action=create status=202 duration=17.93 view=0.00 db=2.82 key=https://mastodon.social/users/nixCraft#main-key

This is the database for that user:

Yet I do not see the user as being followed (still see the "Withdraw Follow Request") button. I give it some time (sometimes hours) and it still does not actually follow the user.

It's also worth noting that an @ mention from another server doesn't make it to my server.

I'm running a mastodon instance on AWS with Elastic Container Service, fronted by Cloudfront, if it matters.

Happy to provide more information and thanks for reading.

[nightpool]

Not really sure there's enough information here for us to debug this. What is the domain of your server? What do the sidekiq jobs look like? Are there any errored/failed jobs there? Are there any other messages in the sidekiq container logs? From the logs you sent, it does seem like https://mastodon.social/users/nixCraft successfully sent the "follow response" to your server, and it was so now it would up to your server to process that response and handle it appropriately (so that https://mastodon.social/users/nixCraft would show up as "Followed" in your local server interface).

[theothermattm]

Thank you for responding! The domain is https://mastodon.deptagency.com. And here are some of the sidekiq logs from around the time of that follow:

2022-12-14T11:30:57.885-05:00
2022-12-14T16:30:57.885Z pid=7 tid=2egr class=ActivityPub::DeliveryWorker jid=20e3d0adc1c0536052f6709d elapsed=0.136 INFO: done
2022-12-14T11:30:57.749-05:00
2022-12-14T16:30:57.749Z pid=7 tid=2egr class=ActivityPub::DeliveryWorker jid=20e3d0adc1c0536052f6709d INFO: start
2022-12-14T11:30:34.290-05:00
2022-12-14T16:30:34.289Z pid=7 tid=2f2z class=ActivityPub::DeliveryWorker jid=cbde867f827ee5875638bfee elapsed=0.516 INFO: done
2022-12-14T11:30:33.773-05:00
2022-12-14T16:30:33.773Z pid=7 tid=2f2z class=ActivityPub::DeliveryWorker jid=cbde867f827ee5875638bfee INFO: start
2022-12-14T11:28:30.561-05:00
2022-12-14T16:28:30.560Z pid=7 tid=2ion class=ActivityPub::DeliveryWorker jid=10e4d53dd46719ee8a263078 elapsed=0.456 INFO: done
2022-12-14T11:28:30.104-05:00
2022-12-14T16:28:30.104Z pid=7 tid=2ion class=ActivityPub::DeliveryWorker jid=10e4d53dd46719ee8a263078 INFO: start
2022-12-14T11:23:34.699-05:00
2022-12-14T16:23:34.699Z pid=7 tid=2f2z class=ActivityPub::DeliveryWorker jid=5cfb0990b811200669829b82 elapsed=0.467 INFO: done
2022-12-14T11:23:34.232-05:00
2022-12-14T16:23:34.232Z pid=7 tid=2f2z class=ActivityPub::DeliveryWorker jid=5cfb0990b811200669829b82 INFO: start

[nightpool]

What actually appears in the https:​//mastodon.deptagency.com/sidekiq dashboard though?

[nightpool]

All of those jobs are "outgoing" jobs, it sounds like maybe you're not processing any incoming requests at all. Are you sure your container configuration includes all Sidekiq queues? See e.g. https://docs.joinmastodon.org/admin/scaling/#sidekiq-queues for more info. It sounds like maybe you're not processing the ingress queue.

[theothermattm]

@nightpool thank you! Sure enough I have TONS of enqueued jobs...

It turned out that the command I was using to execute the sidekiq container only contained a few of the necessary queues. once I enabled that I was in business.