The background queue for ActivityPub-related tasks in Micro.blog seems to have accumulated 5 million jobs overnight. Some replies to the fediverse will be delayed until I sort out what happened or it eventually catches up.
I don’t know the size of your user base, but that seems like a lot.

is this unique to Mastodon (have heard it is not the most efficient), or the protocol broadly?

@bax The protocol is generally considered very chatty. It sends a lot of messages back and forth between instances. (I’m still puzzled where this big increase came for my servers this week, though.)
