Bug 31297

Created attachment 13585 [details]
console output: systemctl status & start of mdmonitor.serice

Created attachment 13586 [details]
journalctl -b (output)

To me it is unclear whether there is really a fallacy about mdadm or whether it just reports sucess/failure inaccurately (failed with result 'protocol', whatever that means).

Thank you for the report. The attached journal has few relevant lines, search 'mdmonitor' - as in your first attachment.

mdadm: A tool for managing Soft RAID under Linux
Looks a bit basesystem.

Assignee: bugsquad => basesystem

Today btrfs can be used for what mdadm was used instead before, afaik. However as the btrfs support of Mageia is not 100.0% it may be well worthwile to keep mdadm well supported (bug 30178).

additional information:

I also see the message that mdmonitor fails when mga9 starts up (on several computers).

When I check the status after booting, I see:

[root@localhost ~]# systemctl status mdmonitor.service
× mdmonitor.service - Software RAID monitoring and management
     Loaded: loaded (/usr/lib/systemd/system/mdmonitor.service; enabled; preset: enabled)
     Active: failed (Result: protocol) since Wed 2023-10-18 11:39:25 BST; 6min ago
    Process: 776 ExecStart=/sbin/mdadm --monitor --scan -f --pid-file=/run/mdadm/mdadm.pid (code=exited, status=0/SUCCESS)
        CPU: 5ms

Oct 18 11:39:25 localhost systemd[1]: Starting mdmonitor.service...
Oct 18 11:39:25 localhost systemd[1]: mdmonitor.service: Can't open PID file /run/mdadm/mdadm.pid (yet?) after start: No such file>
Oct 18 11:39:25 localhost systemd[1]: mdmonitor.service: Failed with result 'protocol'.
Oct 18 11:39:25 localhost systemd[1]: Failed to start mdmonitor.service.


So the log also shows a message:

Oct 18 11:39:25 localhost systemd[1]: mdmonitor.service: Can't open PID file /run/mdadm/mdadm.pid (yet?) after start: No such file>



As noted in previous comment, the mdmonitor service can be started after booting:


[root@localhost ~]# systemctl start mdmonitor.service && systemctl status mdmonitor.service
○ mdmonitor.service - Software RAID monitoring and management
     Loaded: loaded (/usr/lib/systemd/system/mdmonitor.service; enabled; preset: enabled)
     Active: inactive (dead) since Wed 2023-10-18 11:49:49 BST; 22ms ago
   Duration: 1ms
    Process: 5468 ExecStart=/sbin/mdadm --monitor --scan -f --pid-file=/run/mdadm/mdadm.pid (code=exited, status=0/SUCCESS)
   Main PID: 5469 (code=exited, status=0/SUCCESS)
        CPU: 5ms

Oct 18 11:49:49 localhost systemd[1]: Starting mdmonitor.service...
Oct 18 11:49:49 localhost systemd[1]: Started mdmonitor.service.
Oct 18 11:49:49 localhost systemd[1]: mdmonitor.service: Deactivated successfully.



However, the PID file is not present (after manually staring mdmonitor):

[root@localhost ~]# ls -l /run/mdadm/mdadm.pid
ls: cannot access '/run/mdadm/mdadm.pid': No such file or directory


I do not use RAID on my systems (so I do not think I need mdmonitor).

However, it is always a concern to see a red error message in the boot startup log messages.

CC: (none) => paul.blackburn

I see the same as well and I don't use RAID as well.

CC: (none) => lovaren

My understanding is that it's enabled by default since a system that needs it
will not boot without it, while a system that doesn't need it will just get
the failed to start message, and can easily disable it if not wanted.

I may be wrong about it not booting, but I think that's the case.

CC: (none) => davidwhodgins

I think Dave is correct.
I have no system using RAID.
Here on my workstation with only normal partitions and LVM on LUKS:

$ journalctl -b | grep mdmonitor
maj 20 08:30:04 svarten.tribun systemd[1]: Starting mdmonitor-takeover.service...
maj 20 08:30:05 svarten.tribun systemd[1]: Started mdmonitor-takeover.service.
maj 20 08:30:21 svarten.tribun systemd[1]: Starting mdmonitor.service...
maj 20 08:30:21 svarten.tribun systemd[1]: mdmonitor.service: Can't open PID file /run/mdadm/mdadm.pid (yet?) after start: No such file or directory
maj 20 08:30:21 svarten.tribun systemd[1]: mdmonitor.service: Failed with result 'protocol'.
maj 20 08:30:21 svarten.tribun systemd[1]: Failed to start mdmonitor.service.

CC: (none) => fri