Backup error, how to find the reason

I have the following errors related to backups:



Where can I find the reason of the errors, is there a backup log or something similar?

Thanks a lot.

 

Both devices (sda1 and storagebox) have space left:

 

@till I saw a backup_now.php , would it be adviseable to give manual backup a try?
Otherwise I just can tell, hetzners storageboxes are heavily unreliable, especially on night times it can cause issues.
Getting low speed and connection drops or maintenance windows.
If you can rule out an issue on your end, try adjusting the times when the backup takes place or ask the support to have you moved to another store-unit. It is not uncommon.

 

unless you typed
$screen
which can be installed if needed.
https://gist.github.com/jctosta/af918e1618682638aa82

highly adviseable to use when performing any tasks.
Think of typical putty windows user seeing "connection has been closed" if nothing happened for too long time

 

Thanks a lot to both,
Sorry, I realized that I have not shared complete information. There are nine VPS on my installation (1 master server and 8 slaves).
This issue is happening in just one slave server, and all of them make backups to the same storagebox. Only one site is generating the backup errors, there are other two sites on the same server that are making the backups correctly.
So, about running cron_debug.php, will it launch all sites backup? Cause maybe is a better idea to run form cli just this site backup, can it be done?
Thanks a lot.

 

Or you reached the max 10 connections per storagebox limit maybe?

 

Thanks for your help.

It is possible that I have reached the limit, but I have less than 10 servers, and the backup hours are stepped 30 min. each server. But still being possible because maybe connections are not being closed just when the backup ends, and maybe some servers overlaps their backup windows.

But I would like to investigate to find whether this is the reason or not, so I would like to run a single site backup from cli, or maybe check the backup logs, or whatever other test to determinate what is happening. So, I would like to know if the methods that you @till and @ztk.me proposed, supose running the backup for all sites, and if is there a way to run a single site backup from cli to see what is happening (or just check the logs for scheduled backups).

Thanks a lot.

 

Done, and no errors found.
Also I have seen that the backup of past night did not throw any error :| . But strangely, the database backup of this night has a size of 4.0MB and the manual that I have just done has 5.60MB.
Can I re-enable server.sh and leave the log level for the server in debug to see what happen tonight? Or will they throw too much information and log rotation will overwrite the desired logs?
Thanks a lot.

 

Actually the question was focuses on the part of leaving the log level on debug to check later what happens this night with the backup. Cause I think that maybe it generates a lot of logs and the logs I need to see are overwritten.
Thanks.

 

Ok, I will leave it enabled in this case.

 

Meanwhile, I'm thinking - if it was a mounting issue, it should be in your /var/log/syslog or /var/log/syslog.1
which one could
cat /var/log/syslog | grep mount
or just look for anything interesting around the time of the backup

 

They seem to be ok:

Code:

Jun  9 01:30:02 maria systemd[1]: mnt-storagebox.automount: Got automount request for /mnt/storagebox, triggered by 1018480 (php)
Jun  9 01:30:02 maria kernel: [396588.896970] CIFS: Attempting to mount //u200087.your-storagebox.de/backup
Jun  9 01:34:43 maria systemd[1]: Unmounting /mnt/storagebox...
Jun  9 01:34:43 maria systemd[1]: mnt-storagebox.mount: Succeeded.
Jun  9 01:34:43 maria systemd[1]: Unmounted /mnt/storagebox.
Jun 10 00:00:18 maria systemd[1]: run-user-33.mount: Succeeded.

 

Anything happened at 03:30? Or is it just a timezone setting issue?

 

Yes, it is a timezone issue, this is the hour at which backup is launched.