[26738] in Athena Bugs
Re: linux 9.4.22: pine
daemon@ATHENA.MIT.EDU (Jonathan Reed)
Tue Feb 7 11:38:30 2006
Mime-Version: 1.0
Message-Id: <p05230103c00e7ca74908@[18.152.1.192]>
In-Reply-To: <1139329867.8791.282.camel@egyptian-gods.mit.edu>
Date: Tue, 7 Feb 2006 11:36:57 -0500
To: Greg Hudson <ghudson@mit.edu>
From: Jonathan Reed <jdreed@mit.edu>
Content-Type: text/plain; charset="us-ascii" ; format="flowed"
X-Spam-Score: 1.217
X-Spam-Level: * (1.217)
X-Spam-Flag: NO
Cc: astronut@mit.edu, bugs@mit.edu
Errors-To: bugs-bounces@mit.edu
At 11:31 AM -0500 2/7/06, Greg Hudson wrote:
>On Tue, 2006-02-07 at 10:41 -0500, Jonathan D Reed wrote:
>> Writing a simple signal handler in perl indicates
>> that the SIGHUP is in fact sent upon disconnection, however
>> pine never handles it.
>
>I did a little testing, and I'm not sure this is true. I sshed into a
>machine as myself, ran pine, started composing a message, attached to
>the pine process as root from a separate session, and then terminated
>the initial ssh connection. The SIGHUP handler appeared to run.
The way I arrived at that conclusion was to set $HOME to somewhere in
/tmp, and run pine with debug options. It's supposed to log
"Received SIGHUP" in the debug file, but it didn't when I tested it
remotely.
But yeah, you're right, I think it is the tokens. I just tested it
with $HOME set to /tmp, and it created the .pine-interrupted-mail
file. I'm not sure why I never tried that in my testing, but I still
don't understand why pine isn't logging the signal in the debug file.
-Jon
>I suggest an alternate theory: pine can't write the file because sshd or
>login.krb5 has destroyed the user's tokens. Ideally, we would locally
>hack Pine to save the file into /var/tmp when this happens, but I'm not
>sure whether we'll allocate the resources to do that.
--
-------------------
Jonathan Reed
jdreed@mit.edu
-------------------
