[Beowulf] how can I know that a hard disk died?

[Velu Erwan]

Le jeudi 11 août 2005 à 13:46 -0400, Dimitri Antoniou a écrit :
>  Hi,
> 
>  We have a 16-node HP LC1000 cluster, with 3 hard disks
>  managed by hardware RAID.
Which hardware raid are you using ? Is it the netraid ?

>  When the disk died, the system didn't notify us,
>  and we haven't found any message in log files,
>  at least not anything obvious.
On many hardware raid you can't :(
On some you can using some proprietary tools
On a very few like the 3ware one you can use smartmontools to reach each
disk of your read to manage its smarts attributes.