Re: making raid5 more robust against block errors

[Brad Campbell <brad@wasp.net.au>]

Neil Brown wrote:
> On Thursday March 30, swmike@swm.pp.se wrote:
>> Is there any work going on to handle readerrors on a raid5 disk being 
>> handled by recreating the faulty block from the other disks and just 
>> rewriting the block, instead of kicking the disk out?
> 
> It's done. 2.6.15 I think, but definitely in 2.6.16.

And thanks to a dodgy drive, I can confirm it works for RAID-5 at least..

I wonder Neil, would it be possible to have a corrected error type message added somewhere to 
mdadm's monitor mode.. so you at least get an E-mail to let you know something dodgy happened and 
it's all good now, but perhaps you should take a periodic look at that drive to make sure it's not 
flaking out ?

Or have I just asked for another feature that's already there ?

I've been doing a cat `find /sys/block -name errors` periodically to make sure the counters are not 
incrementing.

Brad
-- 
"Human beings, who are almost unique in having the ability
to learn from the experience of others, are also remarkable
for their apparent disinclination to do so." -- Douglas Adams