From hsu@clinet.fi  Tue Feb 27 12:01:44 1996
Received: from hauki.clinet.fi (root@hauki.clinet.fi [194.100.0.1])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id MAA19356
          for <FreeBSD-gnats-submit@freebsd.org>; Tue, 27 Feb 1996 12:01:38 -0800 (PST)
Received: from plentium.clinet.fi (root@plentium.clinet.fi [194.100.0.7]) by hauki.clinet.fi (8.7.3/8.6.4) with ESMTP id WAA28556 for <FreeBSD-gnats-submit@freebsd.org>; Tue, 27 Feb 1996 22:01:27 +0200 (EET)
Received: (hsu@localhost) by plentium.clinet.fi (8.7.3/8.6.4) id WAA12047; Tue, 27 Feb 1996 22:01:27 +0200 (EET)
Message-Id: <199602272001.WAA12047@plentium.clinet.fi>
Date: Tue, 27 Feb 1996 22:01:27 +0200 (EET)
From: Heikki Suonsivu <hsu@clinet.fi>
Reply-To: hsu@clinet.fi
To: FreeBSD-gnats-submit@freebsd.org
Subject: send-pr: Aborting...
X-Send-Pr-Version: 3.2

>Number:         1047
>Category:       gnu
>Synopsis:       send-pr: Aborting...
>Confidential:   no
>Severity:       non-critical
>Priority:       high
>Responsible:    freebsd-bugs
>State:          closed
>Quarter:
>Keywords:
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Tue Feb 27 12:10:01 PST 1996
>Closed-Date:    Sun Oct 20 09:50:23 MET DST 1996
>Last-Modified:  Sun Oct 20 09:53:26 MET DST 1996
>Originator:     Heikki Suonsivu
>Release:        FreeBSD 2.2-CURRENT i386
>Organization:
Clinet, Espoo, Finland
>Environment:

-current from 4 days back

>Description:

send-pr sometimes after you have written your bug report and exit the
editor, will say

send-pr: Aborting...

and deletes your bug report without sending it.  Enormously irritating!

This actually was happening often long time ago, but it went away for
at least several months.  Now its back, or is happening less often.
It first ate my previous PR, then I rewrote it and it worked, and
tried to PR this problem, and again send-pr ate the report (but
learning from experience, I had written the buffer to a safe place
before exiting the editor!).

>How-To-Repeat:

Author lots of bug reports.

It may or may not help to use emacs to edit the pr's.

>Fix:

As a workaround I would suggest:

< # Catch some signals. ($xs kludge needed by Sun /bin/sh)
< xs=0
< trap 'rm -f $REF $TEMP; exit $xs' 0
< trap 'echo "$COMMAND: Aborting ..."; rm -f $REF $TEMP; xs=1; exit' 1 2 3 13 15
--
> # Catch some signals. ($xs kludge needed by Sun /bin/sh)
> xs=0
> trap 'rm -f $REF $TEMP; exit $xs' 0
> trap 'echo "$COMMAND: Aborting ..."; echo $REF $TEMP 'left around, clean up manually'; xs=1; exit' 1 2 3 13 15

This does not fix the actual problem, why does it get a signal?  A bug in sh?

-- 
Heikki Suonsivu, T{ysikuu 10 C 83/02210 Espoo/FINLAND, hsu@clinet.fi
mobile +358-40-5519679 work +358-0-4375360 fax -4555276 home -8031121
>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->feedback 
State-Changed-By: joerg 
State-Changed-When: Sat Oct 19 23:44:02 MET DST 1996 
State-Changed-Why:  
This most likely looks like a local configuration or hardware 
problem.  Does it still persist at all? 

send-pr itself DTRT, thus it cannot be fixed.  Nobody else seems 
to experience such troubles. 

State-Changed-From-To: feedback->closed 
State-Changed-By: joerg 
State-Changed-When: Sun Oct 20 09:50:23 MET DST 1996 
State-Changed-Why:  
The problem is not inside send-pr.  Maybe it's a shell problem (see 
PR bin/1206). 

The originator agreed to close the PR due to its relative low 
importance and sporadic (unreproducible) behaviour of the problem 
itself. 
>Unformatted:
 

