Re: another bug

From: Craig A Summerhill <craig_at_cni.org_at_hypermail-project.org>
Date: Tue, 6 Oct 1998 19:20:00 -0400 (EDT)
Message-Id: <9810062320.AA05549_at_a.cni.org>

Kent Landfield <kent_at_landfield.com> wrote:
>
> # This problem has been around since version 1.02, but I forgot about it
> # because it doesn't come up too often...
> #
> # If you run Hypermail against an existing mailbox...
> #
> # hypermail -c <path/config> -m <path/mbox> -l "whatever" -d <path/target_dir>
> #
> # and the mbox file that you are processing only has one message in it...
> #
> # hypermail goes out and creates the HTMLed message (0000.html), but
> # before the date, author, subject, and thread indexes are created it
> # dumps core with this error:
> #
> # Floating exception (core dumped)
>
> I don't get this when I tried it on the latest version. It may be that
> the problem isn't the single message in the mailbox but the message itself.
> It may be triggering a memory corruption bug. If you have a copy of the
> offending message you can send me I'll verify if it is a problem still.

Kent,

Hmm. That's odd. It happens consistently on my system; everytime there is a mailbox with only one message in it. It must be some combination of the defaults that I set when I compiled the code (I changed some of the suggested settings), and the settings in the .hmrc file I am calling.

I'll send you my compile time settings and .hmrc file later, and see if you can find something in there that is causing it to happen. (I have been up and working on a system problem for thirty-some hours, I need to go home, shower, and get in bed -- so it may be tomorrow).

It does dump core on me to, BTW. Is there a core analysis tool I could used that would give you some helpful information?

P.S. One other thought occured to me. I was wondering if it might be related to the parsing problem I wrote about earlier (which you have fixed in the next release -- thank you). Perhaps it is expecting a trailing line with white space on it, or something? The trouble with this idea is that the single message problem was also present in version 1.02, and came along to version 2.x. However, the parsing problem wasn't present in version 1.02...

-- 

   Craig A. Summerhill, Systems Coordinator and Program Officer
   Coalition for Networked Information
   21 Dupont Circle, N.W., Washington, D.C.   20036
   Internet: craig_at_cni.org   AT&Tnet (202) 296-5098

Received on Wed 07 Oct 1998 01:26:56 AM GMT