Re: [libesmtp-devel] libESMTP 0.8.6

[Brian Stafford <brian stafford uklinux net>]

On Thu, 18 October 22:24 Lucas Adamski wrote:
> I'm getting a coredump on the example program when I try to use the -S flag,
> on RedHat 7.1.  Tried different recipients, options, etc.  -s seems to be the
> cause.  Core attached, output below:
> 
<snip>
> $ ./mail-file-a -s Test -n success -n failure -n delay -d -m test-mail
> wakked1@yahoo.com
> S: 220 localhost.localdomain ESMTP Sendmail 8.11.2/8.11.2; Thu, 18 Oct 2001
> 21:20:44 GMT
> C: EHLO mwwdev1
> S: 250-localhost.localdomain Hello mwwdev1 [127.0.0.1], pleased to meet you
> 250-ENHANCEDSTATUSCODES
> 250-8BITMIME
> 250-SIZE
> 250-DSN
> 250-ONEX
> 250-ETRN
> 250-XUSR
> 250 HELP
> C: XUSR
> S: 250 2.0.0 Initial submission
> C: MAIL FROM:<>
> S: 250 2.1.0 <>... Sender ok
> C: RCPT TO:<wakked1@yahoo.com> NOTIFY=SUCCESS,FAILURE,DELAY
> S: 250 2.1.5 <wakked1@yahoo.com>... Recipient ok
> C: DATA
> S: 354 Enter mail, end with "." on a line by itself
> Segmentation fault (core dumped)

Thanks to all for spotting this one.  Line 691 in headers.c is wrong :-

1031 $ diff headers.c.orig headers.c
691c691
<             (*print) (message, message->current_header);
---
>             (*print) (message, info->hdr);

I moved the block of code that calls this when I made the change to preserve 
the order of replaced headers in a message (0.8.4 -> 0.8.5pre1).  
Unfortunately in its new context message->current_header is always NULL.

The problem is not actually with the Subject: header, it was just the bug's 
victim in this particular case.

Brian Stafford