[lug] Reliable SATA server?
Andrew Hoffman
sixgod at gmail.com
Sun Apr 22 17:14:47 MDT 2012
If I remember correctly we ran across an issue using consumer sata drives
in a raid array. With a dell controller and segate sata drives.
I did some looking around and I recall this being the issue:
These threads talk about modifying the drive's time limit on error
recovery. This may well be relevant: a long delay caused by a drive's
error recovery procedure may fool the RAID system into thinking that the
drive is dead.
http://forums.seagate.com/stx/board/message?board.id=ata_drives&thread.id=10695
http://forums.seagate.com/stx/board/message?board.id=ata_drives&thread.id=11830
That is most likely the issue you are seeing with your drives. Its not that
they are dead or dying its that they are timing out and causing the
raid controller to flip out.
Western digital makes a class of hard drives similar to the black series
tiltled re2, re3, re4 etc.
This drive the RE4 works with raid.
http://www.newegg.com/Product/Product.aspx?Item=N82E16822136579
This does not:
http://www.newegg.com/Product/Product.aspx?Item=N82E16822136792
I'll leave it up to you to research further but this sounds close.
-Andy
On Sun, Apr 22, 2012 at 4:15 AM, <stimits at comcast.net> wrote:
> ....
>
> SAS some advantages over SATA: higher spindle rotation speed of 10K to
> 15K, and much better internal error handling. If you're saying your
> satisfied with old SCSI drives and SATA has issues, then perhaps SAS
> would be a good move. But right now I think lots more memory might be
> the answer.
> ....
>
> Not to mention that some SAS bays allow dual channel to the bay...which
> improves performance when they work, and are a safer redundant
> system when one channel fails.
>
> _______________________________________________
> Web Page: http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> Join us on IRC: irc.hackingsociety.org port=6667 channel=#hackingsociety
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lug.boulder.co.us/pipermail/lug/attachments/20120422/4d4f1669/attachment.html>
More information about the LUG
mailing list