On Wed, Jan 23, 2002 at 11:46:25PM -0500, Nathan Dorfman wrote:
> On Wed, Jan 23, 2002 at 11:57:18AM -0800, Andrey Petrov wrote:
> > If build.sh could be reduced to a small test which reproduce the problem
> 
> It's not anything specific to build.sh. Running a full build is just
> something to generate a lot of I/O. I've also chanced on these messages
> by doing a find, and in one case just cp'ing one file (~5MB) from /usr
> to /.
> 
> One thing that is now apparent is that it'll only happen once, when 
> the system is freshly booted. It is 100% reproduceable then, AFAIK
> 100% guaranteed to occur after I boot, triggered by some I/O bound
> task. But after that, I have not been able to reproduce it a second
> time, no matter how hard I try -- until the next time I boot.
> 

O'k, I'll try it here. Seems I just need to reboot.
It'll destroy my uptime, you know;-)

> Weird.
> 
> Sorry, but I know next to nothing about the workings of these drivers,
> and thus am at a complete loss as to what to look for or try next on my
> own. Anything else I can do to gather more info on whatever could be
> behind the complaints? Consider me at your disposal.
> 

Thanks, I might need your help. Meanwhile I think no harm could be done for your
file system and/or files, driver resets, i/o retries.

> Privet,
Privet,
	Andrey