At 13:25 21/11/2002, Barry A. Warsaw wrote:
It happens that I have one I just built, tested and installed on our production MM 2.0.13 installation.>>>>> "RB" == Richard Barrett <[EMAIL PROTECTED]> writes:RB> Is there a particular reason why, on abnormal exit, the exit RB> status returned by these scripts is not drawn from the values RB> with defined meanings in /usr/include/sysexits.h ? Probably no good one. IOW, its historical. Care to provide a patch? -Barry
I'm just testing the 2.1b5 version of it on my test system so it'll probably be tomorrow before I've finished.
The reason I raised it is that, with Sendmail at least, an attempt to deliver to a list alias for a deleted list, where the mail aliases have not been updated, leads to the uninformative message 'unknown mailer error 1' rather than the more accurate/useful 'Unknown user'.
There are also some other issues about how one would want the MTA to deal with uncaught exceptions in a MM script causing a delivery failure. In my case I would prefer the MTA to see these as EX_TEMPFAIL (75) so that I have an opportunity to fix the code and then have re-delivery by the local MTA succeed, rather than immediately hard failing the delivery so that it is rejected back to the sender as it is now.
Anyway, I'll post the patches tomorrow and you can decide if you want to use them.
_______________________________________________
Mailman-Developers mailing list
[EMAIL PROTECTED]
http://mail.python.org/mailman/listinfo/mailman-developers
