Backup import fails on new Cloudron instance

[yusf]

@girish Both Cloudrons are on 6.0.1. I'm not sure how to use dmesg. dmesg | grep oom doesn't yield any results at least.

[yusf]

@yusf From /home/yellowtent/platformdata/logs/box.log but nothing useful from what I can tell.

2021-02-01T09:52:43.771Z box:shell startTask code: 255, signal: null
2021-02-01T09:52:43.774Z box:tasks startTask: 40 completed with code 255 and signal null
2021-02-01T09:52:43.780Z box:tasks setCompleted - 40: {"error":{"message":"Task 40 crashed with code 255 and signal null","code":"crashed"}}
2021-02-01T09:52:43.781Z box:tasks 40: {"percent":100,"error":{"message":"Task 40 crashed with code 255 and signal null","code":"crashed"}}
2021-02-01T09:52:43.781Z box:apps scheduleTask: task 40 of NEW_APP_ID completed
2021-02-01T09:52:43.781Z box:apps Apptask crashed/stopped: Task 40 crashed with code 255 and signal null
2021-02-01T09:52:43.783Z box:locker Released : apptask
2021-02-01T09:52:43.783Z box:sftp rebuilding container
2021-02-01T09:52:43.787Z box:scheduler resumeJobs: NEW_APP_ID
2021-02-01T09:52:43.788Z box:tasks startTask: 40 done
2021-02-01T09:52:43.802Z box:shell inspectSftp exec: docker inspect --format="{{json .Mounts }}" sftp
2021-02-01T09:52:44.211Z box:shell inspectSftp (stdout): [{"Type":"bind","Source":"/home/yellowtent/boxdata/sftp/ssh","Destination":"/etc/ssh","Mode":"ro","RW":false,"Propagation":"rprivate"},{"Type":"volume","Name":"SOURCE_VOL","Source":"/var/lib/docker/volumes/SOURCE_VOL/_data","Destination":"/run","Driver":"local","Mode":"","RW":true,"Propagation":""},{"Type":"volume","Name":"DEST_VOL","Source":"/var/lib/docker/volumes/DEST_VOL/_data","Destination":"/tmp","Driver":"local","Mode":"","RW":true,"Propagation":""},{"Type":"bind","Source":"/home/yellowtent/appsdata/NEW_APP_ID/data","Destination":"/app/data/NEW_APP_ID","Mode":"","RW":true,"Propagation":"rprivate"}]

2021-02-01T09:52:44.211Z box:shell inspectSftp (stderr): 
2021-02-01T09:52:44.213Z box:sftp Skipping rebuild, no changes
2021-02-01T09:52:50.006Z box:apphealthmonitor app health: 0 alive / 1 dead.

[nebulon]

@yusf can you also check the logs of the failing task? Those should be located at /home/yellowtent/platformdata/logs/tasks/40.log

[yusf]

@nebulon Alas there is no 40.log. I ran the restoration task again and the new task 41 didn't show up as 41.log either.

[girish]

@yusf Strange. Do you have enough disk space? Can you give the server a reboot and see if the logs files are getting generated?

[yusf]

@girish I have plenty of disk space, 14 GB. I'll try rebooting first.

[yusf]

No difference in rebooting. This is getting frustrating. Are there other, perhaps manual ways of restoring? For instance I can migrate the app backup to the new server manually if I can import it to the new app somehow.

[girish]

@yusf If you can contact us on support@cloudron.io, we can try to figure out what is happening.

[yusf]

I'm pressed for time to follow up currently, but I managed to solve perform the migration by

1. Configuring backup target back to the local /var/backups
2. Performing a fresh app backup on the old server
3. Add SSH key of new server to old server's authorized_keys
4. Using scp to transfer the app backup from old server to new.
5. Replace the latest local Gitea backup on the new server with the transferred one.
6. Restore to the latest app backup on the new server.

[girish]

@yusf OK, thanks, when you find time, feel free to reach on the support mailbox.