TrueCloud always fails

TrueNAS General
1

My backup to storj always fails with:

[EFAULT] /mnt/tank/data/syncthing/sync/ modified
/mnt/tank/data/syncthing/ modified
/mnt/tank/data/ modified
/mnt/tank/ modified
/mnt/ modified

Im ONLY backing up /mnt/tank/data/syncthing/

I also get:

unable to create lock in backend: repository is already locked by PID 1988389 on ****** by root (UID 0, GID 0)
lock was created at 2024-12-23 17:04:39 (5h42m45.102018417s ago)
storage ID 1a171ded
the `unlock` command can be used to remove stale locks

Any thoughts on how I can resolve?. Its never finished with success, but it gets to 100%

2

Go into the Storj web interface and manually delete the existing lock file. You might have to hunt around to find it but it’s obvious when you do

1 Like
3

Since I am encrypting file names, how do I find the lock file?

4

Eventually ended up just recreating the backup from scratch which fixed my issue. That’s obviously not a possibility for everyone. If it occurs again I will try removing the lock.

5

I tried that and the lock remained, now I am backing up to an entirely new project to see how it works.

6

The lock file is just a marker file so it’s a so “lock” is the the folder and file name

7

Thanks, we will see if it comes back. I created a new project/paraphrase/api key to start from scratch and see where we get. Thus far it is slower than Backblaze and I am not sure about restic vs rclone.

8

If you have large files that only change a small amount, restic is far, far faster than rclone. I’ve got a number of 50-100GB VMs that take 6 hour to backup to Backblaze but only a few minutes to backup to Storj since restic chunks the files and only has to send chunks that have changed.

Horses for courses as always

9

Well, I figured this one out, it looks like if you abort a backup job it leaves a lock file in the bucketname/lock folder. Delete that and it will work again. This will happen anytime you abort the job, there is no clean up.