Re: Problem: Random paging request errors

[Jon Eisenstein]

> > (2) Every so often, I get a non-fatal error on my screen about a
> > kernel paging request error.
> 
> If it's usually the same address, we're probably dealing with
> a kernel bug. If you always get different addresses, chances
> are your RAM is broken (you can test this with memtest86).

I tested with memtest86, and luckily found no problems (and good, because
this RAM is new!).
 
> Decoding the oops is always useful, especially if you can find
> a pattern after you've decoded a few. And if you don't manage
> to find any pattern in them, you know the suspicion lies with
> the hardware ...

I decoded the oops, from /var/log/messages, but I'm not sure how to read
it. Do you have any pointers? Where should I be looking for patterns? Do I
just wait until the next time it happens, get a second log, and diff them?
>