It seems it may be related to this new bug report: https://github.com/LemmyNet/lemmy/issues/4529

Basically when you get a steady stream of federation requests, at some point Lemmy can’t keep up because of how each inbound federation triggers outbound requests to get other details. And they are run sequentially, so slow requests block the rest of the inbound federation requests as there’s no queue.

It’s slowly crawling along sending us updates but it’s about 12 hours behind and may not catch up until the other side of the world goes to bed.

Plus it seems it has been getting much worse each day, now it has reached the point of being a problem, I’m hoping tomorrow is not a disaster 😐