[ On Thu, October 22, 1998 at 05:41:24 (-0400), Charles M. Hannum wrote: ]
> Subject: Re: stability problems with NetBSD/sparc 1.3.2
>
> This makes no sense.  If `foo' is likely to crash the machine, then it
> sounds to me like you have a way to reproduce the problem, and you
> should be able to get some information.

Like I said, the crashes are random and I'm treating things with Kidd
Gloves until I can find the true cause....  FYI the grepping through all
directories in my PATH did not crash the system (nor did the /etc/daily
run), and I did find the naughty error message deep in the guts of ctwm.

> Regarding rarpd, I asked a very specific question, and it would have
> been nice if you had simply said:
> 
>   `Yes, I'm using the release version, and it doesn't add ARP
>   entries.'
> 
> rather than writing a page of verbiage with *no* additional useful
> content.  Supplying additional information is okay -- to an extent --
> but if I have to wade through it to get to the answer I wanted, then
> you're wasting my time and my inclination is to just punt.

The point I was trying to make is that the answer isn't all that simple.
*no* version of rarpd that I've tried has worked, including the two most
recently released versions on the 1.3 release branch.

Instead of playing chicken and egg with you I gave you all of the
information I thought you could possibly use.  If you don't want to read
it, that's your problem.  If you didn't find the answer you wanted,
and/or you didn't find any additionally useful information in the answer
I gave, then I don't think I can help any further.  I didn't ask you
specifically to help and I didn't send my reply to you specifically.  I
appreciate your questions and I appologize for not writing down all the
information in the first place -- the truth is your questions prompted
me to think even more deeply about an issue that I attached to another
message as more of an afterthought the first time around.

What would help in situations like this would be a development server I
could quickly build stuff on, and a test environment to play with, and
of course the time it would take to work through all the combinations
and permutations until I found a solution or at least a cause.  It would
also help if there was a way to get anonymous access to the CVS tree in
which I could separate out the deltas and comments to see if any changes
might have been intended to solve, or in any way looked as if they
should have solved a problem like this.  But these are all secondary
issues....  ;-)

-- 
							Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods@acm.org>      <robohack!woods>
Planix, Inc. <woods@planix.com>; Secrets of the Weird <woods@weird.com>