6 messages in net.sourceforge.lists.courier-usersRe: [courier-users] Re: RFC/mis-confi...
FromSent OnAttachments
Dan YostApr 1, 2002 10:59 pm 
Sam VarshavchikApr 2, 2002 6:19 am 
Dan YostApr 2, 2002 7:28 am 
Jesse KeatingApr 2, 2002 7:43 am 
Gordon MessmerApr 2, 2002 11:07 am 
Dan YostApr 2, 2002 11:57 am 
Actions with this message:
Paste this link in email or IM:
Paste this link in email or IM:
Atom feed for this thread
Paste this URL into your reader:
Subject:Re: [courier-users] Re: RFC/mis-configured MX questionActions...
From:Dan Yost (da@tri8.com)
Date:Apr 2, 2002 11:57:04 am
List:net.sourceforge.lists.courier-users

On 2 Apr 2002, Gordon Messmer wrote:

Could be. If you look at the part of RFC 974 in question, it's very specific about what records are "irrelevant", and neither of the suggested measures would make an errant record irrelevant.

Indeed, I'd say RFC 974 isn't applicable here. Note, though, that the highest-priority MX record that was returned in this particular case was a completely valid record. The misconfigured record wasn't until further down on the [prioritized] list and, in this particular case, wouldn't have been used at all. So, Courier refused to attempt delivery without even trying the highest-priority (and correctly configured) MX record, which would have succeeded and nobody would have been upset. Obviously, if the domain ever went into backup mode and starting needing the lower-priority (and misconfigured) record (e.g. the primary host was down), then I could understand the refusal to deliver, since the necessary MX record in that case would be a shanked one. Hopefully, the host admin would start noticing, but only in that case, that his backup server wasn't getting the mail and would realize he'd messed up his higher-cost (lower-priority) record. Anyway, I suppose that'd be the beginning of a different thread than my original question.

Thanks a bunch, Dan