Okay so it looks like there's a race condition in the "destroy old servers" system, where the image is being held open because it's linked to an instance that no longer exists.
Which is of course, irritating.
so I'm going through and updating the images that have instances that no longer exist associated with them
and continually crashing into rate limiting ... which is *really* frustrating
adding a bunch of "sleep" statements to my code so I can stop getting told to fuck off by the server
Okay well I don't know why this is happening
fuchsiaaaaaaaaaaaaaaaaa
@aurynn Mastodon's ratelimits are very restrictive, even the normal (mobile) webclient runs into them quite a lot. You can check the headers to see how quick you're going through the 300 requests / 5 minutes
@f0x oh in this case it's my cloud provider's rate limiting