[Beowulf] Big storage

Loic Tortay tortay at cc.in2p3.fr
Sat Sep 15 00:47:59 PDT 2007


According to Chris Samuel:
> On Friday 14 September 2007 17:32:54 Loic Tortay wrote:
> 
> > During the last HEPiX meeting, Peter Kelemen mentionned something told
> > to him by a ZFS developer (Jeff Bonwick, if I'm not mistaken) about
> > data corrupted by a Fibre Channel HBA during transfer between disk and
> > host.  ZFS, reportedly, detected (and corrected) the corruption.
> > Of course a ZFS developer may be biased.
> 
> Could it have been this story by Eric Lowe ?
> 
> http://blogs.sun.com/elowe/entry/zfs_saves_the_day_ta
> 
I specifically remember it was a story involving a Fibre Channel HBA,
so this must be another story.

Your friend blog entry is also very interesting, although maybe a bit 
optimistic on the "no data corruption guarantee" unless he has 
pre-existing data checksums to confirm that.  That would be some 
"experimental" proof of ZFS error correction effectiveness.

>
> > I emailed the internal ZFS interest list with my saga, and quickly got a
> > response. Another user, also running a Tyan 2885 dual-Opteron workstation
> > like mine, had experienced data corruption with SATA disks. The root cause?
> > A faulty power supply.   
> 
Interestingly enough, we had lots of silent data corruptions with Tyan 
288x motherboards and 3ware 8506 SATA RAID controllers (in late 2003 or
early 2004).

This was due to a bug in the 3ware controllers, that did not work 
reliably when installed in the PCI-X slots of the mainboard (it's in 
3ware publically available support knowledge base).


Loïc.
-- 
| Loïc Tortay <tortay at cc.in2p3.fr> -     IN2P3 Computing Centre     |



More information about the Beowulf mailing list