#urlteam 2019-06-27,Thu

↑back Search

Time Nickname Message
00:45 🔗 Jens has quit IRC (Remote host closed the connection)
00:46 🔗 Jens has joined #urlteam
02:15 🔗 Kagee has quit IRC (Quit: WeeChat 2.5)
02:44 🔗 Kagee has joined #urlteam
03:37 🔗 odemg has quit IRC (Ping timeout: 265 seconds)
03:49 🔗 odemg has joined #urlteam
04:14 🔗 mtntmnky_ has joined #urlteam
04:16 🔗 mtntmnky has quit IRC (Remote host closed the connection)
17:32 🔗 Dj-Wawa Error communicating with tracker: 507 Server Error: The tracker needs an operator for manual maintenance. Try again later. for url: https://tracker.archiveteam.org:1338/api/get.
17:32 🔗 Dj-Wawa I think somebody needs to kick the tracker.
18:03 🔗 klg has quit IRC (Quit: brb)
18:05 🔗 Fusl Too many results! Run the export script.
18:05 🔗 Fusl Kaz
18:06 🔗 Kaz yes i know what that is
18:06 🔗 Kaz wait i wonder if that's this thing
18:08 🔗 klg has joined #urlteam
19:24 🔗 JAA The export should run automatically...?
19:24 🔗 JAA At least it normally does.
19:38 🔗 Kaz 2019-06-25 doesn't seem to have uploaded
19:39 🔗 Kaz which is, incidentally, when IA died
19:39 🔗 JAA Ah
19:42 🔗 Kaz right, I'm running ~things~
19:43 🔗 Kaz https://archive.org/details/urlteam_2019-06-25-04-17-08
19:43 🔗 Fusl -> http://xor.meo.ws/SzCeNaJftUiGIK67TMMFxzoop6STjNd8/x.gif
19:45 🔗 Kaz well, that uploaded things but it hasn't fixed the tracker
19:46 🔗 Fusl poke it with a stick?
19:48 🔗 JAA Do we need the one person who knows how all of this works?
19:48 🔗 Fusl it may clear automatically?
19:48 🔗 Fusl if Result.get_count() > DEADMAN_MAX_RESULTS:
19:48 🔗 Fusl return '<div class="alert btn-danger">Too many results! Run the export script.</div>'
19:48 🔗 Kaz sec
19:50 🔗 Kaz it's running a backup job it seems
19:53 🔗 Kaz I'm letting it run.. if things don't pick up in an hour or so we'll poke someone with a brain
19:59 🔗 jut has joined #urlteam
20:06 🔗 Kaz okay had another look through, think I can see what happened
20:07 🔗 Kaz job failed to upload while IA was down, supervisor therefore didn't clear up after itself, so all jobs after that also failed
20:07 🔗 Kaz now running a drain/export & upload
20:08 🔗 Kaz https://tracker.archiveteam.org:1338/ this looks healthier, I guess
22:24 🔗 Kaz OSError: [Errno 28] No space left on device
22:24 🔗 Kaz oh
22:27 🔗 Kaz chfoo: could you have a look please? not sure what is safe to delete at this point.. I would assume ./export/done/ as it looks like that gets pushed elsewhere, but not sure if that'll get us as far as we need
22:33 🔗 chfoo the done directory is a backup directly after it gets uploaded to archive.org so it's safe to delete
22:33 🔗 chfoo directory*
22:34 🔗 chfoo did you run undrain script and delete the working export directory?
22:37 🔗 Kaz I was running export&upload, which is when it ran out of disk
22:37 🔗 Kaz so export folder currently has 1.9G in done (will delete), and 3.7G in 2019-06-27-etc
22:38 🔗 Kaz is current_working_set.txt in there safe to remove? Assuming that'll get recreated?
22:38 🔗 chfoo no, it's taken out of the database
22:38 🔗 Kaz ah, fun
22:39 🔗 chfoo so the undrain script puts the stuff back into the database
22:40 🔗 Kaz so undrain -> clear export/done/, run export&upload again?
22:41 🔗 chfoo clear export/done first to regain disk space, then undrain, and rerun
22:42 🔗 Kaz ok. I also ran a manual upload for a set from the other day (while IA was dead), that's still in the export folder, rather than /done/. Safe to delete?
22:42 🔗 chfoo yes, it should be ok
22:44 🔗 chfoo alternatively, you can download the working set data to a safe place, and just keep export & uploading until the results are cleared out, vacuum the database, then run the undrain script
22:46 🔗 Kaz Just kicked off undrain so will see how that goes, looks like we've had it running away a bit for the last few days https://usercontent.irccloud-cdn.com/file/t9Ywvhcj/image.png

irclogger-viewer