20

I just configured a Debian server as follows
apt-get install smartmontools
vi /etc/default/smartmontools - and here uncommented these 2 lines
# uncomment to start smartd on system startup
start_smartd=yes
# uncomment to pass additional options to smartd on startup
smartd_opts="--interval=1800"

Then I edited /etc/smartd.conf and uncommented only this line
DEVICESCAN -S on -o on -a -m [email protected] -s (S/../.././02|L/../../6/03)

I was wondering if this is enough for the system to send me an alert message when something goes wrong and/or if there is a way to simulate a problem and see if smartd sends me warnings by email.

1
  • Really you should configure the alias for root in /etc/aliases (if using postfix) so that any server generated emails are forwarded to your email account, not just smartd.
    – Geoffrey
    Aug 1, 2018 at 1:31

1 Answer 1

30

You can test the configuration by adding -M test to the line that begins with DEVICESCAN, immediately after that keyword. The next time you restart smartd, it will send out an email notification.

2
  • 1
    I found that the -M option has to be immediately after DEVICESCAN, not at the end. This works: DEVICESCAN -M test -H -s S/../../7/02 -m [email protected] but this doesn't (illegal -M Directiive: test): DEVICESCAN -H -s S/../../7/02 -m [email protected] -M test Aug 3, 2017 at 15:01
  • -M test at the end of the DEVICESCAN works ok on Debian Jessie / smartctl 6.6 Jul 22, 2019 at 13:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .