Re: I/O errors during RAID check but no SMART errors

[Franco Martelli <martellif67@gmail.com>]

On 09/10/24 at 21:10, Jochen Spieker wrote:
> Andy Smith:
> > Hi,
> >
> > On Wed, Oct 09, 2024 at 08:41:38PM +0200, Franco Martelli wrote:
> > > Do you know whether MD is clever enough to send an email to root when it
> > > fails the device? Or have I to keep an eye on /proc/mdstat?
> >
> > For more than a decade mdadm has shipped with a service that runs in
> > monitor mode to do this.
> >
> >      https://manpages.debian.org/bookworm/mdadm/mdadm.8.en.html#MONITOR_MODE
>
> And this is configured here:
>
> # grep -B1 MAILADDR /etc/mdadm/mdadm.conf
> # instruct the monitoring daemon where to send mail alerts
> MAILADDR root

Yes, I think I have mdadm running properly to achieve that:

~$ pstree -cpan|grep mdadm
  |-mdadm,459 --monitor --scan

~$ grep -B1 MAILADDR /etc/mdadm/mdadm.conf
# instruct the monitoring daemon where to send mail alerts
MAILADDR root

I asked this question because I was unsure. In the past it happened 
system freeze so I needed to hard reboot the the system, sometime this 
caused a "resync" of the array and no email was sent to root. A "resync" 
activity falls in the "RebuildStarted" mdadm's event. The man-page says:

...
Only Fail, FailSpare, DegradedArray, SparesMissing and TestMessage cause 
Email to be sent.
...

This explain why an email hadn't been sent from mdadm. Thank you very 
much Andy and Jochen.

Cheers,
--
Franco Martelli