Although it’s not recommended to upgrade, I thought I’d risk it and try since it’s worked ok in the past with Symboisis.
The steps that I took:
First I ran a small VPS on Mythic Beasts with on demand pricing with a Bullseye Sympl install. I copied the updated the sources in /etc/apt/sources.list
and the files in /etc/apt/sources.list.d
on the VPS to upgrade.
sudo apt update
sudo apt upgrade
Asked to restart services during package upgrades, I said yes.
Asked about /etc/tunserver.conf
, I took the package maintainer’s version:
Configuration file '/etc/turnserver.conf'
==> Modified (by you or by a script) since installation.
==> Package distributor has shipped an updated version.
What would you like to do about it ? Your options are:
Y or I : install the package maintainer's version
N or O : keep your currently-installed version
D : show the differences between the versions
Z : start a shell to examine the situation
The default action is to keep your current version.
*** turnserver.conf (Y/I/N/O/D/Z) [default=N] ?
Y
Upgrade round cube database: Yes
Roundcube config.inc.php: I kept the local version
I ran: sudo apt full-upgrade
.
Asked about MariaDB/MySQL config file, I said install package maintainers version:
Configuration file '/etc/mysql/mariadb.conf.d/50-server.cnf'
==> Modified (by you or by a script) since installation.
==> Package distributor has shipped an updated version.
What would you like to do about it ? Your options are:
Y or I : install the package maintainer's version
N or O : keep your currently-installed version
D : show the differences between the versions
Z : start a shell to examine the situation
The default action is to keep your current version.
*** 50-server.cnf (Y/I/N/O/D/Z) [default=N] ?
Y
Restart the VM: sudo systemctl reboot
Remove the stale packages: sudo apt autoremove
Checked Nectcloud which highlighted the memory limit being below the recommended 512MB that’s recommended, so I edited the config and restarted apache2:
sudo vim /etc/php/7.4/apache2/php.ini
# Find memory_limit and increase to 512M
sudo service apache2 restart
All Nectcloud checks in the web admin passed
The cron for Nextcloud was erroring, so I added --define apc.enable_cli=1
to the cron job call:
sudo -u www-data crontab -e
One final issue that I found is with sympl-monit
where I’ve been getting the following errors to root@server:
Started Sympl service monitor.
INFO Runner: mysqld: Checking service is enabled
INFO Runner: mysqld: Connection test errored - Refusing to operate on alias name or linked unit file: mysqld.service; caused by 3 sender=:1.2 -> dest=:1.1198 serial=84316 reply_serial=33 path=; interface=; member= error_name=org.freedesktop.systemd1.UnitLinked
WARN Runner: mysqld: FAILED: Service unavailable
INFO Runner: RESULT: 9/10 passed.
I tried running the check it was doing: /usr/bin/mysqladmin --defaults-extra-file=/etc/mysql/debian.cnf ping
, however that returned ok, so not sure of the cause.
As a workaround I’ve removed the mysqld check for now to avoid being spammed: sudo rm /etc/sympl/monit.d/mysqld
.
Other than that I’ve not spotted another issues yet.
In a trial upgrade from a clean buster Sympl install to bullseye I had noticed an issue with the old version of php still being used so Apache2 wouldn’t start. Enabling PHP7.4 and disabling PHP7.3 resolved it in that case. I’m not sure what the difference was compared to my main setup.