Justin Piszcz wrote:
On Fri, 21 Nov 2008, Alan Cox wrote:
Oct 27 11:33:41 Arzamas kernel: ata6.00: error: { IDNF ABRT }
Different error. IDNF -> ID not found
So the only thing in common is WD disks. I don't know at this point..
NCQ with
WD disks (at least raptor150s/300s) is completely broken and I have to
disable
it to use the disks without NCQ errors and dropping out of arrays (on
3 diff)
systems with different chipsets, p35, 965.
This guy has the same errors I do when I enable NCQ. It seems like WD
raptors
should be blacklisted from using NCQ in the kernel.
NCQ: http://lkml.org/lkml/2007/10/17/380
Concerning my error specifically:
How much time have you spent replacing and RMA'ing disks? Personally
for me
I spent $3600 on new velociraptors but after having spend hours, days
with
doing testing, replacing, I have spent much more in time than the
drives are
worth. I am really getting sick and tired of it, every few days or
every week
its another drive failure (or more). What are you plans, just keep
RMA'ing and hopefully they'll find the bug and fix it or just give up
and build a new system for production and the RMA'ing can be a hobby
in another system?
What I meant when I asked did you buy them new, was, did you buy them
recently,
because with the raptor150s I have in another system, all 1+ year old,
bought
new but I wonder if they changed something in their firmware to cause
this
problem, when I opened a case with WD about the issue and spoke with
someone
in India they are only setup to answer basic questions and anything
complicated
you will get canned answer: "RMA."
Time to dump WD raptor drives?
I will note I am using several 750GiB drives without any issue whether
NCQ is enabled or disabled in various raid and non-raid configurations.
Comments--buggy raptors?
Well I guess I will have some data points after the weekend... I just
got a new WD 1TB on sale from Newegg, and I'm going to be using it
raid10 with a raid0 of a pair of Seagate 500GB, and expect to beat it
somewhat severely moving data from place to place.
--
Bill Davidsen <davidsen@xxxxxxx>
"Woe unto the statesman who makes war without a reason that will still
be valid when the war is over..." Otto von Bismark
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
