Hello, the server on Ubuntu shows this message:
An error occurred while accessing or checking the internal database. This means that this database may be corrupted or there is not enough free space. If this error persists, please restore the database (urbackup_server files.db and urbackup_server_settings.db) from the backup. See the log file for details.
There have been no changes in the backup folder since August 6, 2024, and in the /var/urbackup/files folder there are urbackup_server files.db and urbackup_server_settings.db became executable and were assigned other rights: -rwxr-x—
Backups are made in the /Backup folder, which is a REID 5 software array with an EXT4 file system, the rights to the folder are given to the user and the urbackup group, the same rights to the /var/urbackup/ (drwxr-xr-x) folder
At the same time, backups continue to be performed, but the statistics of the occupied disk space have partially disappeared.
There is more than enough free space on all disks!
Tell me what to do?
Try the repair_database script? Or remove_unknown one?
Can’t guarantee the backups will survive intact, but those are what I can think of off the top of my head.
If the database is corrupted, the repair script may work.
tell me where to get this script?
It’s right in the program folder, which on Windows is C:\Program Files\UrBackupServer\ With .bat extensions in that case. On Linux it’ll be wherever it gets installed to… I don’t have a Linux install to go look, but the scripts will be there, I believe there’s no file extension on them under Linux…
I searched Linux by the keywords (find. -type f -name “repair_database*”,“repair_database”,“repair*”,“repair”) starting from the root / and then through all folders. Nothing was found!
Please can someone give me this script for Linux?
I would not like to reinstall the entire system because of this error…
I don’t have a Linux install to play with, all I can tell you is the Administrators Handbook says for Linux to run the remove_unknown one:
…and on Linux while the server is not running:
urbackupsrv remove-unknown
You’ll almost certainly need to be root or use sudo.
I imagine the repair-database function is invoked the same way, also with the server stopped (at least the second thing the Windows batch does after checking you have admin rights is stop the server).
You could just try that in a terminal, I can’t be explicit & can’t offhand tell you how to stop the server under Linux either, if it runs as a service it’ll likely be the same way you’d stop any other service under systemd. You might need to list the services just to get the exact name, but that’s standard Linux system administration, & I’m a bit weak on that when it comes to systemd, such linux installs as I have tend to use sysvinit paired with openrc to manage services.
Since I have no way to test & I’m mostly guessing.
In recent versions of Ubuntu, “systemd” has been replaced with “systemctl”, in order to stop the service, you need to run the “systemctl stop urbackupsrv” command.
The fact is that the database itself is working, backups are performed on a schedule, new settings are applied and most of the functions work.
Only database backup and cleanup does not work. I guess that the backup and cleaning of the database takes place under another user.
Explain to me how UrBackup 2.5.33 works, what type of SQL does it use, what does it have as a WEB service? “Apache” or “Nginx” are missing from the services, what then?
How does it all work?
Older versions of UrBackup were installed separately “sqlite” and “Apache” or “Nginx” to choose from - everything was clear there. And here I don’t understand how it all works.
All I can tell you is what I’ve picked up as a user, the Windows version uses sqlite for the database & comes bundled with its own http daemon, a cut-down version of Apache (I think) built in.
The systemctl command in Linux is one of the ones provided by systemd, seems you know the name of the service to stop it if required. Have you actually tried running remove-unknown or repair-database with it stopped? What was the outcome…
suggested to me that doing so stood some chance of resolving the issue… I just couldn’t tell you the exact commands to issue, not having a Linux install of the server, only the client. If there’s no problems with the database, & no problems in the storage, the only downside of running them will be the time used, remove-unknown in particular can take a long time to run.