EXTERNAL Konold Martin (Firma, RtP2/TEF72) wrote:
Hi Dave,
I hooked up another WD drive WDC WD5002ABYS-02B1B0 from a
of the failing drive WDC WD3200AAJS-60Z0A0 to one of the failing
Is this the WD TLER problem/ripoff? The originals are 'desktop' while the replacement is 'enterprise'.
I have no clue about a desktop/enterprise difference. Where is such a list maintained? The faulty drives are directly shipped from HP with the z400 and xw4600 workstations.
I don't want to say too much, because my opinions as to WD's behaviour are unprintable! 'TLER' is WD's name for a mandatory ATA-8 feature called 'Error Recovery Control' that they do not implement in some drives. So I suggest you google for more details using the words I quoted.
He had no problems at all on 10.2 .
I verified that the problem is not triggered with 11.0 either.
Martin, at this point I really would suggest taking it to the linux-ide list.
You mean it is not appropriate for opensuse-kernel anymore?
No, I'm sorry I didn't state it well; I meant no criticism of this list. It's just that I think there are several people on the linux-ide list who are very familiar with the libata subsystem development and familiar with many different drives who may recognize your exact symptoms and so provide a definitive solution quickly and easily.
Did you also try a new kernel?
Yes, I tried with OpenSUSE 11.4 DVD install just this morning.
During the very first installation of OpenSUSE 11.4 on a problematic machine I got the following crash: [snip] [ 75.422596] BUG: unable to handle kernel NULL pointer dereference at 00000004 [ 75.422607] IP: [<c031fe99>] shrink_dcache_for_umount_subtree+0xf9/0x1b0 [ 75.422617] *pde = 00000000 [ 75.422623] Oops: 0002 [#1] SMP [snip]
I then clean the hdd using dd and then tried a second install. This time it went well and I am currently compiling the kernel in an endless loop (since 45 min).
OK. There was no ata error in the log you posted, just the kernel oops. So perhaps the drive is working correctly with the new kernel whilst a previous failure had corrupted the data, causing the oops. That would be good news :) I believe the current 11.4 kernel is also the current stable one, which I think is 2.6.37.3. Though your subject line says 2.6.37.1-1.2; I guess that is what is on the DVD. Did you run an initial upgrade? So the implication would seem to be that your problem has probably been fixed by some patch between 2.6.34.8-15 and 2.6.37.1-1.2. Which means somebody on the linux-ide list probably can suggest exactly which patch solved the problem and so you can be confident of the fix. Cheers, Dave -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-kernel+help@opensuse.org