Steve Quint wrote:
> Hi all:
> 
> I've been running 1.3 on my Q700 with great results.  This machine is my
> internal network's gateway over PPP.  Until a few weeks ago, I was starting
> ipnat with a script that wrote a new NAT rules file, then start ipnat,
> everytime a PPP link is started (due to a dynamic IP).
> 
> Then I changed the setup to one that's described in the FAQ, where I map
> the entire network to 0/32, then using the command `ipf -y` to re-sync the
> kernal.  The problem is that I started getting crashes with the following
> message:
> 
> -----------------
> Feb 21 00:18:22 homer /netbsd: Data modified on freelist: word 0 of object
> 0x6ca9200 size 128 previous type temp (0xdeadbeeb != 0xdeadbeef)
> Feb 21 00:18:23 homer /netbsd: Data modified on freelist: word 0 of object
> 0x6ca0280 size 128 previous type temp (0xdeadbeeb != 0xdeadbeef)
> Feb 21 00:18:23 homer /netbsd: Data modified on freelist: word 0 of object
> 0x6c9c300 size 128 previous type temp (0xdeadbeeb != 0xdeadbeef)
> -----------------
> 
> These crashes always immediately follow a `ppp-up` and an `ipf -y`, but not
> in a consistant manner.  Maybe once every 7 times or so.  What exactly does
> this message mean?  And where is it coming from?

The last time I saw this message was on my IIci while I was playing with
the write-back enable line on my Daystar accelerator.  It made sense in
that case.  I don't know why your getting it, but it's coming from
kern_malloc.c, either line 217 or 241.
 
> This machine runs headless and without a keyboard, so I was unable to see
> if it drops into the debugger or not, not to mention a trace.

Where do you do the ppp-up and ipf -y?  I do mine on the command line, and
I've never seen anything quite like that....

Later.

-- 
Colin Wood                                 cwood@ichips.intel.com
Component Design Engineer - MD6                 Intel Corporation
-----------------------------------------------------------------
I speak only on my own behalf, not for my employer.