4 messages in net.sourceforge.lists.courier-usersRE: [courier-users] Re: rfcviolation ...| From | Sent On | Attachments |
|---|---|---|
| Vittorio Ballestra | Sep 28, 2001 2:20 am | |
| Lars Christensen | Sep 28, 2001 3:02 am | |
| Vittorio Ballestra | Sep 28, 2001 3:27 am | |
| Dirk Kulmsee | Oct 4, 2001 9:47 am |
| Subject: | RE: [courier-users] Re: rfcviolation problem | |
|---|---|---|
| From: | Dirk Kulmsee (kulm...@netgroup.de) | |
| Date: | Oct 4, 2001 9:47:43 am | |
| List: | net.sourceforge.lists.courier-users | |
-----Ursprüngliche Nachricht----- Von: cour...@lists.sourceforge.net [mailto:cour...@lists.sourceforge.net]Im Auftrag von Vittorio Ballestra Gesendet: Freitag, 28. September 2001 12:28 An: Lars Christensen Cc: cour...@lists.sourceforge.net Betreff: [courier-users] Re: rfcviolation problem
Il 12:02, venerdì 28 settembre 2001, hai scritto:
The problem is, that OE is full of bugs. Sometimes OE doesn't encode 8bit characters in the right way. The Courier mail server is very
strict about
They definitely sucks.... it's a shame.
The right solution would be for MS to clean up their code, but since this is not likely to happen anytime soon, you will have to go into submit*.C
How about having two different configuration options? 1. i do / don't want to accept incorrect mails 2. i do / don't want to send a notification for each malformed mail e.g. to "outl...@microsoft.com"
True - we would pay the additional traffic, but there is the slim chance that a continuous stream of error messages gets even Microsoft people to work.
Another remark why users are not to blame but suffer from tricky software: i have seen mails rejected by courier, because an antivirus software (Trend Micro's Viruswall NT) thought it would be nice to include a timestamp with "Westeuropäische Sommerzeit" in the headers. In this case it is not MS and not the user. Even the network or mailserver administrator probably has little chance to change the behaviour of the AV-software.
I don't have the skills myself to work on the courier sources (and get a _presentable_ result). The best thing for me would be a Debian package of courier with a choice of behaviour in the conf-files.
Regards Dirk