Closed mbaumanndev closed 6 months ago
Some additionnal details : my compose is running on a Synology NAS and is build on a backup I made a few months ago from a custom system (Ubuntu 22 VM inside Freebox OS).
Since Minecraft process creates the files as user ID 1000, running the backup container with the same user ID will probably resolve this. Adding under your minecraft-backups
service will configure that:
user: "1000"
I will update the examples to do the same.
Hi, thanks for your reply, I now have the following logs :
time="2024-05-25T09:57:05+02:00" level=info msg="Skipping uid/gid change since current user is not root"
2024-05-25T09:57:05+0200 INFO waiting initial delay of 2m...
2024-05-25T09:59:05+0200 INFO waiting for rcon readiness...
2024-05-25T09:59:05+0200 INFO Command executed successfully rcon-cli save-on
2024-05-25T09:59:05+0200 INFO Command executed successfully rcon-cli save-off
2024-05-25T09:59:06+0200 INFO Command executed successfully rcon-cli save-all flush
2024-05-25T09:59:06+0200 INFO Command executed successfully sync
2024-05-25T09:59:06+0200 INFO Backing up content in /data to /backups/world-20240525-095906.tgz
tar (child): /backups/world-20240525-095906.tgz: Cannot open: Permission denied
tar (child): Error is not recoverable: exiting now
tar: /backups/world-20240525-095906.tgz: Cannot write: Broken pipe
tar: Child returned status 2
tar: Error is not recoverable: exiting now
2024-05-25T09:59:06+0200 ERROR tar exited with code 2! Aborting
2024-05-25T09:59:06+0200 INFO Command executed successfully rcon-cli save-on
I'll check later today, but I presume it's an issue with the rights on the folder where my backups are made, I'll keep you in touch
I tried changing the destination folder, it's owner and group are both 1000 and I have the same error
The remainder has to be some kind of ownership mismatch. How about the ownership of the /backups directory itself?
On my disk it has 644 permissions and 1000 as uid/gid, but I didn't delete the existing container, they may not be right, I'll check later today
I did some checks, I found out that my user/group on my Synology were not 1000/1000 but 1026/100, I'll try running the images with thoses uid/gid
After some tests, setting uid/gid with 1000:1000 worked after deleting all my stack and changing the owner on the folder with my ansible script with some changed parameters. I have no clue on why it didn't worked first time i tried it, but it's alright now. Thanks for your help !
Hello, I have an issue and I noticed it because it cause another issue that ignore the backup interval.
Here is what I get in the logs of my backup service :
And it loops on it like this.
Here is an extract of my compose file :
When I connect on a container to check the volume, I have the following permissions set in the
world
folder :If I manage to run
chmod 664
forlevel.dat
andlevel.dat_old
the backup succeed, so I have two hypothsesis :itzg/minecraft-server
create the files with the wrong rightsitzg/mc-backup
has insuficient rights to create the backupI'll continue investigating on what's going on and keep you in touch !