procmail
[Top] [All Lists]

Re: procmail - mutt milebox problem

2001-08-07 22:56:18
Tim Holmes <tim(_at_)unixtechs(_dot_)org> writes:
...
that's just me.  So I never see that message.  But if there's no mail in
$MAIL, there's no file there to read, so that's why it complains.

By default, if you delete all the messages in your mailspool will mutt
delete the file or will it leave a zero-length file?  If the former,
it's broken: mailspool files should never be removed.


However, two or three versions ago, if you left the line:

my_hdr Organization:

Just like that, with nothing there after, it would cause a core dump.  I
banged my head up against the wall trying to figure out why the heck
this was happening!  I sent a few messages to the mutt list and they
basically told me it was because that line was empty.  They'd never
tested something like that so they didn't know it would happen.  Then
told me, "Well if you're smart enough to edit your own .muttrc you
should be smart enough not to leave a line like that, but we'll look
into it."  I just tested it in Mutt 1.2.5i, and 1.3.15i and the problemm
doesn't occur.

mutt seems like a nice MUA from my playing with it, but the attitude
shown by the developers is wrong, potentially dangerous, and should not
be encouraged.  If you can make a program coredump without sending it a
signal explicitly, there's a bug.  As maintainer of procmail I ask that
if you find a way to make procmail dump core without using "kill()",
tell me how and I'll fix it.  To the best of my knowledge, there's no
way to make procmail version 3.15.2 coredump by feeding it any form of
message, putting anything in the rcfile, or giving it any wacky command
line arguments.  It may spew warning messages on the latter two, but it
won't crash.


Philip Guenther
Procmail Maintainer
_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail