Re: Checking SMART hard drives with smartctl or smartd

[Martin Pala]

Hello,

you can run the script which will check the SMART status from cron for example 
every minute and if everything is OK, then touch state file (for example 
/var/run/sda_ok) which is checked by monit like this:

check file sda_smart with path /var/run/sda_ok
  if timestamp > 5 minutes then ...

We plan to add native support for check scripts integration in the future.

Regards,
Martin


On Mar 16, 2011, at 2:16 AM, Stephen W. Moore wrote:

> I use the smartmontools (smartctl and smartd) to test the hard drives for
> SMART parameters. Documentation, particularly a Google study of their global
> servers, show that certain SMART values can predict drive failure. The
> interesting parameters are:
> 
> 01   Read Error Rate
> 05   Reallocated Sectors
> 10   Spin Retry
> 194  Temperature
> 196  Reallocation Events
> 198  Uncorrected Sectors
> 
> The smartctl command can be configured the SMART values. Example:
> 
> smartctl -A /dev/sda
> 
> This outputs the table of SMART parameters for /dev/sda. These can be sorted
> to isolate specific values.
> 
> Is there a preferred MONIT implementation of triggering an alert of SMART
> values using smartctl or similar utility?
> 
> 
> --
> To unsubscribe:
> http://lists.nongnu.org/mailman/listinfo/monit-general