@davidaugust Should be restored now. Some errant logs escaped my rotator and filled up the disk, whoops!
@davidaugust Should be restored now. Some errant logs escaped my rotator and filled up the disk, whoops!
If you were using file.garden links in your #CBTS bot, please go ahead and re-set your preferred posting frequency and save your changes. The issue was resolved by the domain owners earlier this week.
also I might need to temporarily disable any bots that are trying to pull from the file.garden domain because the failed Request calls are doing something I don't have time to properly fix.
@Zero_heavysnow There are a few problems I'm seeing, not sure which is the cause. Will monitor and should have fix by tomorrow, thank you.
It looks like there is an issue with file.garden, one of the sites I've recommended to #CBTS users for image hosting in the somewhat recent past.
Per their Blue Sky post (https://bsky.app/profile/filegarden.com/post/3li2t6ljobs2x), GoDaddy has terminated their .garden TLD registration and they are at filegarden.com for now, but direct links are not working.
@Zero_heavysnow i will try to look into this. What is the bot account that is having trouble?
I wrote a little botsin.space postmortem here: https://muffinlabs.com/posts/2024/12/21/12-21-botsin-space-post-mortem/ i might tweak it a bit but anyway there it is
@nora OK actually I think my manual intervention might've worked? give it a shot.
@nora I've got one or two more manual things I can try, but confidence is low. I will let you know if they do turn out, though.
@nora Are you seeing anything in your logs on the GTS side? Do you have any kind of firewall or allow/deny-list for access? I've tried a couple novel instances to make sure new app creation isn't broken and so far everything else goes smoothly (but I do get the same broken behavior as you do trying that specific instance.)
@nora oh I've never seen this page appear for any meaningful amount of time, and it really shouldn't. Sorry to "turn it off an on again" but can you try a fresh re-auth and see if it hangs again? Just ping me before you do so I can follow logs.
@nora My GTS install (randomrul.es) is on 0.17 and it worked totally normally, standard flow. If you can provide some screenshots or context I can try harder, but it doesn't seem to be back-end specific.
@nora oh! it's GTS. I don't know I've tested against one of those before, this might not be simple but I have my own I can dogfood with.
@settima @lexipenia I have resolved the issue with authorizing to zirk.us instance. See https://github.com/mastodon/mastodon/issues/27740 for the underlying issue; I resolved from the app side by wiping the stored credentials and applying for fresh authorization from the server. Should be all good now.
MythBot has been migrated over to @mythologybot with botsin.space having its retirement.
All #CBTS bots that have botsin.space as their instance will have their post frequency changed to 0 (Never Post) programmatically on Saturday.
@settima @lexipenia first time I've seen this one, dayjob is heavy this week but I'll look into it this weekend if not sooner. Same with your message about the scheduling issue on the other bot - - thanks for your patience!
@egallager I defer to the instance and its application/admin standards. it is not a guarantee of safe passage
@davidaugust testing suggests the 500 limit is not hard enforced. If you're previewing the output on the website it will turn text red if it's >500, but it won't stop you from saving the grammar and should try to post whatever is generated regardless of length and let the server decide to accept it or not.