On 9/24/06, Darrin Chandler <[EMAIL PROTECTED]> wrote:
On Sun, Sep 24, 2006 at 11:51:18AM +0100, Stuart Henderson wrote:
> On 2006/09/24 09:45, Alf Schlichting wrote:
> > You may be a victim of the "feature" that ( for hardware related reasons)
> > the dmesg buffer doesn't get cleared after reboot so you actually have 2
> > dmesgs, (or maybe even more ?) one after the other, in dmesg buffer.
> > This can easily be overlooked by scrolling back to fast.
>
> It sounds like you don't like it - but it can actually be very useful;
> ddb(4) normally logs commands and responses to the message buffer, so if
> you have a machine that doesn't clear memory at reboot, after a kernel
> panic you can often still retrieve the error, and the commands you
> entered (hopefully to display a backtrace and process list) - if you
> don't have a null nodem cable or pencil handy this is pretty good.
It's a feature I like a lot. The first time I ran into it I was
confused, though.
I'm well aware of it and like it to. From my recollection I scrolled
through dmesg after each boot and every kernel was the 8/21 one. It
wasn't until after the 4th boot that I scrolled to the bottom and the
9/23 one was showing. Disclaimer: I hadn't had coffee yet yesterday
and I was very burnt out from returning home from China last Sunday,
and then having a hellacious week at work, I could have seen anything
and not digested it properly.
Greg
