Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Blog | Linux Links | Bling | About BLU

BLU Discuss list archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

running degraded RAID to save power



I would suggest doing some rough calculations to see if the proposed $$ 
savings is significant enough.

Also, consider purchasing a 3rd similar drive in 6 months or a year to 
on-purpose get one from another manufacturing run.  Mirror one old and 
the new, and use the 3rd drive to power up and rsync or use raid to sync 
with on occasion.  Or as an 'off site' drive in an external enclosure.

Matthew Gillen wrote:
> So I just ordered the parts to replace my home file server, and as usual I 
> bought two identical drives with the intention of putting them in a RAID-1 
> array.
>
> Given a lot of the anecdotal evidence I've heard about identical drives 
> purchased at the same time (thus likely to come from the same batch) failing 
> very close to each other, I was thinking about a possible solution:
>
> Set up the RAID-1 array with both disks, do the base install/configuration. 
>   Then use hdparm to power-down one of the drives.  Have a script bring the 
> drive up every few days and let RAID re-sync it, then shut it down again.
>
> My thinking is that this will extend the life of one of the drives (in 
> addition to saving some miniscule amount of power).
>
> Has anyone tried something like this?  All the hdparm options dealing with 
> powering on and resetting are marked DANGEROUS.  I'm not even sure that it 
> can be done (specifically, keeping the kernel from trying to wake up the 
> drive except when I want it to).
>
> This is a home server, so a little (ie a week's worth) data loss doesn't 
> worry me too much (compared to losing everything if both drives failed at 
> once).  Or should I not even bother with RAID and use an rsync-based approach?
>
> Thanks,
> Matt
> _______________________________________________
> Discuss mailing list
> Discuss-mNDKBlG2WHs at public.gmane.org
> http://lists.blu.org/mailman/listinfo/discuss
>
>   






BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!



Boston Linux & Unix / webmaster@blu.org