Cancel jobs

How do I cancel moving to file system jobs?

1 Like

Hi,

Unfortunately it’s not possible yet.

1 Like

Let’s say, hypothetically, you accidentally issued a command to move all videos to filesystem when you intended to move them all to storage.

You quickly notice the problem and issue the correct command. But it looks like every video is going to bounce to the server and back to storage.

Problem: The local drive can only hold a tiny fraction of the remote storage and it’s quickly filling up.

Question: Should I stop PeerTube before it reaches ~95% disk capacity to prevent a system crash? Or will it handle things intelligently? What’s my best course of acti----I mean----hypothetically the best course of action? :fire: :woozy_face: :fire: :fire:


Edit: I fixed it by loading a database snapshot taken (I think) just before the fiasco. I probably need to clean the filesystem now.


Edit2: Nope, it’s still filling up for some reason. Even after a reboot.


Edit3: Ran the prune-storage server command and it has helped. Could be solved…


Edit4: Some videos in object storage are not recognized as being there. They say « failed to play ». And I’m getting this error on some videos.

Cross-Origin Request Blocked: The Same Origin Policy disallows reading the remote resource at https://medias.tankie.tube/streaming-playlists/hls/52facabc-b820-465f-93cc-0ba903cf950e/57fed105-f1c1-47ca-8489-547b4a5f13a7-1080.m3u8. (Reason: CORS header ‘Access-Control-Allow-Origin’ missing). Status code: 404.

The number of borked videos is small enough to manually fix. Backblaze still has all the video files. Now I just need to know how to fix it.


Edit5: Drive is still filling up. :sob: :sob: I’m going to get some rest and reassess later.

Disk is full and website is down. I’m helpless. Everything should be backed up though.


Edit: I created a separate topic for my problem: Recovering instance from crash due to full filesystem