Unable to Ignore 'unreadable (pending) sectors' Error

[seanfulton]

I have many, many machines running SSDs and they periodicaly generate errors like this:

Device: /dev/sdc [SAT], 1 Currently unreadable (pending) sectors

The SSDs tend to reman these failures because they will go away over time. But how to I stop smartd from e-mailing these alerts. Because of the number of machines we have, we can get up to 100 e-mails a day. It makes it difficult to weed out more serious errors.

I have tried the following:
DEVICESCAN -I 197 -m gcn-alerts@… -M exec /usr/libexec/smartmontools/smartdnotify -n standby,10,q

and

DEVICESCAN -m gcn-alerts@… -M exec /usr/libexec/smartmontools/smartdnotify -n standby,10,q -t -I 197

But the e-mails keep coming. Any idea how to get rid of errors for type 197 failures?

[Christian Franke]

Warnings for unreadable sectors are configured separately. Try -C 0 to disable these.

Remove -I 197 and add -R 197! -r 197! to receive emails when this attribute changes.

See the smartd.conf man page​ for details.

[seanfulton]

I am trying to not to report changes to 197 at all.

Does -C 0 turn off all reporting or just unreadable sectors? I still want to get drive failure, pending failure messages, etc.

[Christian Franke]

The directive -C 0 overrides the default -C 197 and therefore only affects this attribute. See man page for more details.

[Christian Franke]

These warnings could be disabled with -C 0.