Issue Scheduled backups are not working

[xpaz]

Hi,

I have a server with several domains and scheduled backups for all of them. Scheduled backups are not being done - I have do start them manually. This is what I have checked after some investigation:

* File /etc/cron.d/plesk-backup-manager-task exists, has correct permissions (-rw-r--r--) and is not empty

* File /usr/local/psa/admin/sbin/backupmng exists and looks OK:
[root@ns364063 ~]# ls -la /usr/local/psa/admin/sbin/backupmng
-rwxr-x--- 1 root root 58432 Nov 15 14:02 /usr/local/psa/admin/sbin/backupmng

* Cron jobs are executed regularly, as registered in /var/log/cron, and the task backupmng is executed every 15 minutes, as seen with
grep backupmng /var/log/cron
Feb 6 08:44:01 ns364063 CROND[14960]: (root) CMD ([ -x /usr/local/psa/admin/sbin/backupmng ] && /usr/local/psa/admin/sbin/backupmng >/dev/null 2>&1)
Feb 6 08:59:01 ns364063 CROND[15209]: (root) CMD ([ -x /usr/local/psa/admin/sbin/backupmng ] && /usr/local/psa/admin/sbin/backupmng >/dev/null 2>&1)
Feb 6 09:14:01 ns364063 CROND[22223]: (root) CMD ([ -x /usr/local/psa/admin/sbin/backupmng ] && /usr/local/psa/admin/sbin/backupmng >/dev/null 2>&1)

* File /usr/local/psa/admin/sbin/backupmng exists and looks OK:
[root@ns364063 ~]# ls -la /usr/local/psa/admin/sbin/backupmng
-rwxr-x--- 1 root root 58432 Nov 15 14:02 /usr/local/psa/admin/sbin/backupmng

* The table "BackupsScheduled" in database "psa" exists, and I can open it and see the scheduled backups

* There is enough free room in the server - 74% used of 100 GB

However scheduled backups are not done, if I start them manually there is no issue.

What can be the problem, how can I fix it?

 

[IgorG]

Please check that the following entries appear in /var/log/syslog (for Debian-based systems) or in /var/log/messages (for RHEL-based systems):

server-hostname CRON[10554]: Authentication token is no longer valid; new one required
server-hostname cron[1034]: Authentication token is no longer valid; new one required

If you have it - update root password. Plesk sets up scheduled system tasks in cron of "root" user. "root" Linux user's password is expired. As a result, all scheduled tasks of "root" user stopped working.

Also make sure all post-upgrade tasks were executed correctly with:

# /usr/local/psa/bootstrapper/pp18.0.24-bootstrapper/bootstrapper.sh post-install
# /usr/local/psa/bootstrapper/pp18.0.24-bootstrapper/bootstrapper.sh perform-deferred-actions

 

[xpaz]

Thanks, both checks are negative.

In /var/log/messages there are no mentions of cron or "Authentication token". And the file

/usr/local/psa/bootstrapper/pp18.0.24-bootstrapper/bootstrapper.sh

does not exist - should it? The most recent folder there is pp18.0.21-bootstrapper.

I have executed those post-upgrade tasks, there is an error - although I think it is unrelated:

[root@ns364063 pp18.0.21-bootstrapper]# /usr/local/psa/bootstrapper/pp18.0.21-bootstrapper/bootstrapper.sh post-install
===> Cumulative Plesk upgrade (final stage) has been started.
===> Preparing Plesk upgrade (final stage).
===> Cumulative upgrade of Plesk (final stage) has been completed.
Trying to upgrade Roundcube Web Based mail client configs and DB from '1.3.8' to '1.3.10'... Executing database schema update.
Updating database schema (2013061000)... [FAILED]
Error 500: Error in DDL upgrade 2013061000: [1146] Table 'roundcubemail.cache' doesn't exist

Warning: upgrade Roundcube Web Based mail client configs and DB from '1.3.8' to '1.3.10'. Run /usr/share/psa-roundcube/bin/update.sh manually to complete upgrade.

***** installing problem report *****
Warning: upgrade Roundcube Web Based mail client configs and DB from '1.3.8' to '1.3.10'. Run /usr/share/psa-roundcube/bin/update.sh manually to complete upgrade.
Created symlink from /etc/systemd/system/multi-user.target.wants/postfix.service to /usr/lib/systemd/system/postfix.service.
===> Configuring ProFTPD server
IP addresses were reread from the system.***** installing problem report *****
Warning: upgrade Roundcube Web Based mail client configs and DB from '1.3.8' to '1.3.10'. Run /usr/share/psa-roundcube/bin/update.sh manually to complete upgrade.
STOP plesk-roundcube-1.3.10 installing AT Thu Feb 6 10:31:30 CET 2020: PROBLEMS FOUND

 

[leobaillard]

Hi!

I'm experiencing the same issue on a Debian Jessie server (yes I know, I should upgrade but that is not easy to do on a production server) with Plesk 18.0.21.

Both the post-install and perform-deferred-actions ran fine without any errors on my end.