Serious issue with SATA disks again

[Anthony Atkielski]

Erik de Jong writes:

> FWIW, I had WRITE_DMA time outs as well, on a non-SATA disk of about 3
> years old. My instict told me to make that last backup real quick, so
> that's what I did. The disk crashed about a week or two later. I'm an
> absolute FreeBSD newbie, but hardware problems transcend OS
> differences ;-)

What made you think it was a hardware problem?

> I worked in a large software company for a while and found out (much to
> my own surprise at the time) that it's not always that easy. It's not
> always a matter of "I wrote the code so I know what this message means".
> Those messages could be anywhere in any piece of code involved at the
> time of the error. It could be passed on by another piece of code
> (firmware?).

Somebody, somewhere knows what causes this error.

> Second, code comes in many shapes and forms (and I think this is one
> area where open and closed source are probably very similar): neat code,
> sloppy code, quick fixes, things that really shouldn't be there,
> absolute gems and everything in between.

Such as code without any trace of comments, like I see in FreeBSD.

> The BEST troubleshooting tip not matter what you trying to
> resolve is: do not rule out anything.

Including the operating system.

> Obviously, I think you have a hardware issue, but that's only my
> opinion. I'm just writing to share my own experience.

Which hardware issue do I have?

> You're writing that noone knows what these mysterious error messages
> mean? This could very well be the case right now.

Somebody wrote the code.

> Perhaps the issue hasn't been troubleshot enough yet.

It has been around for a least a year, judging by what I've found on the
Net.

> I think that having problems and having difficulty resolving them is far
> from a FreeBSD problem. Report myserious error messages to any software
> company and there's always a risk that they won't be easily resolved.

That doesn't make it okay for FreeBSD.

-- 
Anthony