Re: Huge unreliability - does Linux have something to do with it?

[Horst von Brand]

jerome lacoste <jerome.lacoste@xxxxxxxxx> said:
> Bernd Eckenfels <ecki-news2005-01@xxxxxxxxxxxx> said:
> >> Could a hardware failure look like bad sectors to fsck?

> > A failure of the bus or a former sporadic error can cause defective fs, but
> > normally you have a read error in fsck no structure error.
> > 
> > Are you using hdparm? is the system perhaps overheating or overclocked?

> no overclock
> hdparm is used but I cannot tell you exactly what the config is (now
> machine has been running memtest for 1.5 hour). I don't think I use
> special option: probably the defaults in my config file (mult_sect 16,
> dma on, write_cache off).

There are combinations of IDE + disk that slowly corrupt filesystems with
DMA on, if the default setting is DMA off _don't touch it_. Not all bad
combinations are catched by the code in the kernel (intel + some Western
Digital disk is what drove me up the wall until I disabled DMA).

What machine is this, what disk?

> overheating: perhaps. The machine is hot and running many hours per
> day (usually 12-16). It s running the fans very often, but it's always
> been like that. I've tried to control the fan, but then the
> temperature goes high very quickly. So I let the fans run.

Wise decision.
-- 
Dr. Horst H. von Brand                   User #22616 counter.li.org
Departamento de Informatica                     Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria              +56 32 654239
Casilla 110-V, Valparaiso, Chile                Fax:  +56 32 797513
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/