Subject: Re: Printer Problems, Part 2 ...
To: Brian Buhrow <buhrow@cats.ucsc.edu>
From: Dave Burgess <burgess@s069.infonet.net>
List: current-users
Date: 12/08/1994 17:20:08
> 
> 	The reason you get a clean exit as well as a core file is because lpd
> forks to spawn the filter process.  
> I would note that I was having this problem (core files from lpd) on NetBSD
> 0.9a and traced it down to some bogus logic having to do with which filter
> it wass trying to use, or something like that.  It's been a few months
> since I looked at the problem and my memory isn't real clear.  I do
> remember, however, that ktrace gave me a clue as to what the problem was.
> Oh, yes, it was trying to read a non-existent file or something and because
> the file wasn't there, it wasn't setting the parameters normally set in
> that file to sane values.  In any case, ktrace works well because it traces
> the child as well as the parent process.  So, you'll be able to see what
> it's doing before it crashes.

Well, that was it.

I used ktrace -i and kdump and found out that the daemon was dying on
one of the print files that I had queued up during the NetBSD 1.0 stuff.
I moved all of the print files to a sub-directory and reprinted them.
Everything went great after that.

Spooky.

Thanks for the help.  Everything seems to be working now.