Seagate disk problems (NCQ bug???)

D. Hugh Redelmeier hugh at mimosa.com
Mon May 11 19:44:02 UTC 2009


| From: Wolfgang S. Rupprecht <wolfgang.rupprecht+gnus200905 at gmail.com>
| "D. Hugh Redelmeier" <hugh at mimosa.com> writes:

|   I'll see if I can find a way to report this to seagate.  They
| don't seem to make it very easy by not having a prominent support@
| address that I can find documented anywhere.

Folks on the forum whine a lot about Seagate support.  I've had mixed
but mostly poor results.

Phone support sometimes has long hold or queue times.

"email" support is really a web form.

I've had OK results with their online chat support.

In any case, if the frontline folks know the answer, you'll probably
get it.  I have had no luck getting beyond them.  But I've only tried
a little bit.

If you can get a problem that others can duplicate, and you post it on
the forum, perhaps that will get their attention.

| Their web page had some convoluted sign up to get a seagate approved
| identity.  The sign up refused me enough times that I just figured it
| was broken.

Their forum software is as bad as I've seen.

| Well, the test case that works for me is to copy a half dozen 1GB files
| from a sata dvd reader to the sata seagate disk.
| 
| It might be very kernel dependent though.  I hadn't seen the disk act up
| before a few weeks ago.  It also takes hours for the system to wedge up
| solidly.  This isn't going to be a fun bug for Seagate to find.

Could you dd an equivallent volume from /dev/zero to see if it also
causes a problem?  That eliminates one more variable from the problem.

| I don't know that it is NCQ related, only that other reports of similar
| lockups under streaming conditions claimed it was related to a known
| Seagate bug related to their NCQ implementation.  I might be
| misremembering or misunderstanding though...

Why do you call it "streaming conditions"?  Do you think that the fact
that the DVD delivers bytes more slowly than the hard drive would accept
them is important?  Or is there some other aspect that you think is
relevant?

NCQ has been around long enough that I'd expect/hope that the nits
have been picked.  Of course I could be wrong.




More information about the fedora-list mailing list