> > The same [hang while fscking a dirty filesystem] happened when I
> > tried 2.1.131. 2.0.36 works allright on my box.
>
> So 2.2.xpre will fail on the fsck, but 2.0.36 will then fsck the same
> disk just fine ? What sort of disk/controller ?
I've had the same problem as a result of enabling the "Use DMA by
default if available" configuration option in recent kernels. The
problem, at least in my case, seems to be that using the default DMA
configuration with the IDE controller on a motherboard with a VP-2
chipset results in a system that will work fine for quite some time
but hang solidly on extended periods of heavy disk activity.
Using the "Use DMA by default..." option with this chipset will work
fine for several hours at a time under normal conditions, but will
reliably hang during fsck, as described above. The amount of time
that goes by between enabling this option and the hang during fscking
following the first dirty shutdown makes it difficult to associate the
fsck hang with the selection of this configuration option.
I'd suggest changing the line in the configuration help file from
"It is nearly always safe to say Y to this question." to something
like "It is nearly always safe to say Y to this question, unless
you'll be using a VIA VP-2 chipset."
-- John Kodis.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/