Implementing events about block layer errors?

[Stanislaw Gruszka]

Hello.

My boss wants me to add disk errors uevents, to get information on faulty
devices. He told me catching dmesg information using syslog is not reliable
and he would like to get 100% certain information. I know there are currently
implemented uevents in the device mapper layer, but they are limited to
multipath targets. Even if error events were added to the device mapper, 
would it be impossible to find out which physical device generates errors
if the device mapper works on top of the software RAID (correct?). 

Does it make sense to add such a functionality to the block layer 
(or somewhere else), and also could such a modification be included
in the mainline kernel, or  should I rather convince my boss to implement
it via syslog?

Stanislaw Gruszka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/