Hello,

$ cat /etc/debian_version
12.10

An automated apt update/upgrade failed last night. On trying again, I get:

$ sudo apt update; sudo apt upgrade
<snip>
Preconfiguring packages ...
48 upgraded, 3 newly installed, 0 to remove and 0 not upgraded.
1 not fully installed or removed.
Need to get 0 B/133 MB of archives.
After this operation, 474 MB of additional disk space will be used.
Setting up mariadb-common (1:10.11.11-0+deb12u1) ...
update-alternatives: error: alternative path /etc/mysql/mariadb.cnf doesn't exist
dpkg: error processing package mariadb-common (--configure):
installed mariadb-common package post-installation script subprocess returned error exit status 2
Errors were encountered while processing:
mariadb-common
E: Sub-process /usr/bin/dpkg returned an error code (1)


/etc/mysql/mariadb.cnf doesn't exist on my system, but is provided within the package:

$ ls /etc/mysql/*cnf
/etc/mysql/debian.cnf

$ dpkg -x /var/cache/apt/archives/mariadb-common_1%3a10.11.11-0+deb12u1_all.deb ~/deb
$ dpkg -e /var/cache/apt/archives/mariadb-common_1%3a10.11.11-0+deb12u1_all.deb ~/deb
$ ls ~/deb/etc/mysql/*cnf
/home/user/deb/etc/mysql/mariadb.cnf

$ tail ~/deb/postinst
# Thus, we need to ensure here that mariadb.cnf indeed became the primary
# alternative and override with priority 500 if needed.
if ! update-alternatives --query my.cnf | grep --quiet "Value: /etc/mysql/mariadb.cnf"
then
update-alternatives --install /etc/mysql/my.cnf my.cnf "/etc/mysql/mariadb.cnf" 500 || true
fi
;;
esac

Isn't this supposed to add/replace mariadb.cnf from the package into /etc/mysql/?

FWIW:

$ update-alternatives --query my.cnf | grep --quiet "Value: /etc/mysql/mariadb.cnf"
update-alternatives: warning: alternative /etc/mysql/mariadb.cnf (part of link group my.cnf) doesn't exist; removing from list of alternatives
update-alternatives: warning: alternative /etc/mysql/my.cnf.fallback (part of link group my.cnf) doesn't exist; removing from list of alternatives

Has anyone else seen this?

Thanks,
Gareth

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Sun, 16 Mar 2025 08:56:47 +0000
"Gareth Evans" <[email protected]> wrote:

Hello,

$ cat /etc/debian_version
12.10

An automated apt update/upgrade failed last night. On trying again,
I get:

$ sudo apt update; sudo apt upgrade
<snip>
Preconfiguring packages ...
48 upgraded, 3 newly installed, 0 to remove and 0 not upgraded.
1 not fully installed or removed.
Need to get 0 B/133 MB of archives.
After this operation, 474 MB of additional disk space will be used.
Setting up mariadb-common (1:10.11.11-0+deb12u1) ...
update-alternatives: error: alternative path /etc/mysql/mariadb.cnf
doesn't exist dpkg: error processing package mariadb-common
(--configure): installed mariadb-common package post-installation
script subprocess returned error exit status 2 Errors were
encountered while processing: mariadb-common
E: Sub-process /usr/bin/dpkg returned an error code (1)

/etc/mysql/mariadb.cnf doesn't exist on my system, but is provided
within the package:

$ ls /etc/mysql/*cnf
/etc/mysql/debian.cnf

$ dpkg -x /var/cache/apt/archives/mariadb-common_1%3a10.11.11-0+deb12u1_all.deb
~/deb $ dpkg -e /var/cache/apt/archives/mariadb-common_1%3a10.11.11-0+deb12u1_all.deb
~/deb $ ls ~/deb/etc/mysql/*cnf /home/user/deb/etc/mysql/mariadb.cnf

$ tail ~/deb/postinst
# Thus, we need to ensure here that mariadb.cnf indeed became the
primary # alternative and override with priority 500 if needed.
if ! update-alternatives --query my.cnf | grep --quiet "Value: /etc/mysql/mariadb.cnf" then
update-alternatives --install /etc/mysql/my.cnf my.cnf "/etc/mysql/mariadb.cnf" 500 || true fi
;;
esac

Isn't this supposed to add/replace mariadb.cnf from the package into /etc/mysql/?

FWIW:

$ update-alternatives --query my.cnf | grep --quiet "Value: /etc/mysql/mariadb.cnf" update-alternatives: warning: alternative /etc/mysql/mariadb.cnf (part of link group my.cnf) doesn't exist;
removing from list of alternatives update-alternatives: warning:
alternative /etc/mysql/my.cnf.fallback (part of link group my.cnf)
doesn't exist; removing from list of alternatives

Has anyone else seen this?

What's the background here? Did you have a working mariadb installation
before, and is it still working?

I have mariadb on sid and it certainly has those files. There is a
somewhat complicated arrangement, and it looks as if a working
configuration file could be in more than one location. It should only
be generic anyway, it includes conf.d and mariabd.conf.d where the
actual configurations are.

--
Joe

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Sun 16/03/2025 at 11:01, Joe <[email protected]> wrote:

What's the background here? Did you have a working mariadb installation before, and is it still working?

Hi Joe,

It was working before and after the failed upgrade, as neither it nor anything else had been upgraded.

I copied mariadb.cnf from the extracted mariadb-common package files to /etc/mysql/ and retried the upgrade, which succeeded, but mariadb then produced "SQLSTATE <some code> no such file or directory" errors in my web apps, so I restarted the mariadb
service, and got:

$ sudo systemctl status mariadb
× mariadb.service - MariaDB 10.11.11 database server
Loaded: loaded (/lib/systemd/system/mariadb.service; enabled; preset: enabled)
Active: failed (Result: exit-code) since Sun 2025-03-16 14:21:46 GMT; 15s ago
Duration: 7h 28min 31.155s
Docs: man:mariadbd(8)
https://mariadb.com/kb/en/library/systemd/
Process: 1777879 ExecStartPre=/usr/bin/install -m 755 -o mysql -g root -d /var/run/mysqld (code=exited, status=0/SUCCESS)
Process: 1777880 ExecStartPre=/bin/sh -c systemctl unset-environment _WSREP_START_POSITION (code=exited, status=0/SUCCESS)
Process: 1777882 ExecStartPre=/bin/sh -c [ ! -e /usr/bin/galera_recovery ] && VAR= || VAR=`/usr/bin/galera_recovery`; [ $? -eq 0 ] && systemctl set-environment>
Process: 1777899 ExecStart=/usr/sbin/mariadbd $MYSQLD_OPTS $_WSREP_NEW_CLUSTER $_WSREP_START_POSITION (code=exited, status=1/FAILURE)
Main PID: 1777899 (code=exited, status=1/FAILURE)
CPU: 102ms

Mar 16 14:21:46 qwerty.local systemd[1]: Starting mariadb.service - MariaDB 10.11.11 database server...
Mar 16 14:21:46 qwerty.local sh[1777896]: [108B blob data]
Mar 16 14:21:46 qwerty.local sh[1777896]: Fatal error in defaults handling. Program aborted
Mar 16 14:21:46 qwerty.local mariadbd[1777899]: [100B blob data]
Mar 16 14:21:46 qwerty.local mariadbd[1777899]: Fatal error in defaults handling. Program aborted
Mar 16 14:21:46 qwerty.local systemd[1]: mariadb.service: Main process exited, code=exited, status=1/FAILURE
Mar 16 14:21:46 qwerty.local systemd[1]: mariadb.service: Failed with result 'exit-code'.
Mar 16 14:21:46 qwerty.local systemd[1]: Failed to start mariadb.service - MariaDB 10.11.11 database server.


Same state reported after a reboot.

Then:

$ sudo mv /etc/mysql/mariadb.cnf /etc/mysql/mariadb.cnf.xxx

$ sudo systemctl restart mariadb

$ sudo systemctl status mariadb

● mariadb.service - MariaDB 10.11.11 database server
Loaded: loaded (/lib/systemd/system/mariadb.service; enabled; preset: enab>
Active: active (running) since Sun 2025-03-16 14:30:37 GMT; 4s ago
...

So working again, but a messy/broken upgrade process for me.

Is this worth reporting?

Thanks,
G

I have mariadb on sid and it certainly has those files. There is a
somewhat complicated arrangement, and it looks as if a working
configuration file could be in more than one location. It should only
be generic anyway, it includes conf.d and mariabd.conf.d where the
actual configurations are.

--
Joe

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Sun, 16 Mar 2025 14:38:45 +0000
"Gareth Evans" <[email protected]> wrote:

On Sun 16/03/2025 at 11:01, Joe <[email protected]> wrote:

What's the background here? Did you have a working mariadb
installation before, and is it still working?

Hi Joe,

It was working before and after the failed upgrade, as neither it nor anything else had been upgraded.

I copied mariadb.cnf from the extracted mariadb-common package files
to /etc/mysql/ and retried the upgrade, which succeeded, but mariadb
then produced "SQLSTATE <some code> no such file or directory" errors
in my web apps, so I restarted the mariadb service, and got:

$ sudo systemctl status mariadb
× mariadb.service - MariaDB 10.11.11 database server
Loaded: loaded (/lib/systemd/system/mariadb.service; enabled;
preset: enabled) Active: failed (Result: exit-code) since Sun
2025-03-16 14:21:46 GMT; 15s ago Duration: 7h 28min 31.155s
Docs: man:mariadbd(8)
https://mariadb.com/kb/en/library/systemd/
Process: 1777879 ExecStartPre=/usr/bin/install -m 755 -o mysql -g
root -d /var/run/mysqld (code=exited, status=0/SUCCESS) Process:
1777880 ExecStartPre=/bin/sh -c systemctl unset-environment _WSREP_START_POSITION (code=exited, status=0/SUCCESS) Process:
1777882 ExecStartPre=/bin/sh -c [ ! -e /usr/bin/galera_recovery ] &&
VAR= || VAR=`/usr/bin/galera_recovery`; [ $? -eq 0 ] && systemctl set-environment> Process: 1777899 ExecStart=/usr/sbin/mariadbd
$MYSQLD_OPTS $_WSREP_NEW_CLUSTER $_WSREP_START_POSITION (code=exited, status=1/FAILURE) Main PID: 1777899 (code=exited, status=1/FAILURE)
CPU: 102ms

Mar 16 14:21:46 qwerty.local systemd[1]: Starting mariadb.service -
MariaDB 10.11.11 database server... Mar 16 14:21:46 qwerty.local
sh[1777896]: [108B blob data] Mar 16 14:21:46 qwerty.local
sh[1777896]: Fatal error in defaults handling. Program aborted Mar 16 14:21:46 qwerty.local mariadbd[1777899]: [100B blob data] Mar 16
14:21:46 qwerty.local mariadbd[1777899]: Fatal error in defaults
handling. Program aborted Mar 16 14:21:46 qwerty.local systemd[1]: mariadb.service: Main process exited, code=exited, status=1/FAILURE
Mar 16 14:21:46 qwerty.local systemd[1]: mariadb.service: Failed with
result 'exit-code'. Mar 16 14:21:46 qwerty.local systemd[1]: Failed
to start mariadb.service - MariaDB 10.11.11 database server.

Same state reported after a reboot.

Then:

$ sudo mv /etc/mysql/mariadb.cnf /etc/mysql/mariadb.cnf.xxx

$ sudo systemctl restart mariadb

$ sudo systemctl status mariadb

● mariadb.service - MariaDB 10.11.11 database server
Loaded: loaded (/lib/systemd/system/mariadb.service; enabled;
preset: enab> Active: active (running) since Sun 2025-03-16 14:30:37
GMT; 4s ago ...

So working again, but a messy/broken upgrade process for me.

Is this worth reporting?

Yes, I would think so, most Debian servers will be running mariadb on
bookworm, either for production use or for internal purposes or both.

Sid is on mariadb 11.4.5, so while it would have used the current
bookworm version at some time, the fact that there has just been an
upgrade in stable means a security fix for a bug that was not known
when it was in sid.

--
Joe

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 16 Mar 2025, at 16:06, Joe <[email protected]> wrote:

On Sun, 16 Mar 2025 14:38:45 +0000
"Gareth Evans" <[email protected]> wrote

So working again, but a messy/broken upgrade process for me.

Is this worth reporting?

Yes, I would think so, most Debian servers will be running mariadb on bookworm, either for production use or for internal purposes or both.

Sid is on mariadb 11.4.5, so while it would have used the current
bookworm version at some time, the fact that there has just been an
upgrade in stable means a security fix for a bug that was not known
when it was in sid.

OK thanks very much
G

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Reported at

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1100655

for anyone interested.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)