Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: exim: users

[OT] first mx has no A record

 

 

exim users RSS feed   Index | Next | Previous | View Threaded


hs at schlittermann

May 8, 2008, 12:47 AM

Post #1 of 5 (117 views)
Permalink
[OT] first mx has no A record

Hello,

it's not really an exim issue, but I know we've some mail/RFC gurus here
:)


=== Trying mds-mp02.t-systems.com:25...
=== Connected to mds-mp02.t-systems.com.
<- 220 mds-mp02.sczn.de ESMTP Mirapoint Messaging Server MOS 3.8.4-GA Queueing
-> EHLO pu.schlittermann.de
<- 250-mds-mp02.sczn.de Hello pu.schlittermann.de
<- 250-8BITMIME
<- 250-SIZE 52428800
<- 250-ETRN
<- 250-DSN
<- 250-STARTTLS
<- 250-AUTH LOGIN
<- 250-AUTH=LOGIN
<- 250 PIPELINING
-> MAIL FROM:<info[at]kugelbus.de>
<** 501 Sender domain must exist
-> QUIT
<- 221 mds-mp02.sczn.de Goodbye
=== Connection closed with remote host.


Indeed, the first MX of kugelbus.de (mail.kugelbus.de) doesn't exist (missing
A record).

Does exist some standard (RFC) saying that it's fatal mistake to have no
A record for the first MX? (But such mistake can happen anytime - if my
first MX is outside of my domain, and the MX-Admin decides to remove the
relevant A record. Than I'd rely on my second/third MX.)

(BTW: Exim seems to handle such situation the way *I* expect - the mail
is still routable.)


Best regards from Dresden
Viele Grüße aus Dresden
Heiko Schlittermann
--
SCHLITTERMANN.de ---------------------------- internet & unix support -
Heiko Schlittermann HS12-RIPE -----------------------------------------
gnupg encrypted messages are welcome - key ID: 48D0359B ---------------
gnupg fingerprint: 3061 CFBF 2D88 F034 E8D2 7E92 EE4E AC98 48D0 359B -
Attachments: signature.asc (0.18 KB)


dot at dotat

May 8, 2008, 4:49 AM

Post #2 of 5 (108 views)
Permalink
Re: [OT] first mx has no A record [In reply to]

On Thu, 8 May 2008, Heiko Schlittermann wrote:
>
> Does exist some standard (RFC) saying that it's fatal mistake to have no
> A record for the first MX?

The relevant spec is section 5 of RFC 2821. It isn't very good :-(
The draft revision of the spec is better: see
http://www.ietf.org/internet-drafts/draft-klensin-rfc2821bis-10.txt

I think since only one out of three MX records for your domain are broken,
it is not reasonable to reject the message entirely. RFC 2821 says "If MX
records are present, but none of them are usable, this situation MUST be
reported as an error." (note the "none")

Tony.
--
<fanf[at]exim.org> <dot[at]dotat.at> http://dotat.at/ ${sg{\N${sg{\
N\}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}\
\N}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}

--
## List details at http://lists.exim.org/mailman/listinfo/exim-users
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/


hs at schlittermann

May 8, 2008, 6:11 AM

Post #3 of 5 (108 views)
Permalink
Re: [OT] first mx has no A record [In reply to]

Hello Tony,

Tony Finch <dot[at]dotat.at> (Do 08 Mai 2008 13:49:10 CEST):
> On Thu, 8 May 2008, Heiko Schlittermann wrote:
> >
> > Does exist some standard (RFC) saying that it's fatal mistake to have no
> > A record for the first MX?
>
> The relevant spec is section 5 of RFC 2821. It isn't very good :-(
> The draft revision of the spec is better: see
> http://www.ietf.org/internet-drafts/draft-klensin-rfc2821bis-10.txt
>
> I think since only one out of three MX records for your domain are broken,
> it is not reasonable to reject the message entirely. RFC 2821 says "If MX
> records are present, but none of them are usable, this situation MUST be
> reported as an error." (note the "none")

Thank you.

Best regards from Dresden
Viele Grüße aus Dresden
Heiko Schlittermann
--
SCHLITTERMANN.de ---------------------------- internet & unix support -
Heiko Schlittermann HS12-RIPE -----------------------------------------
gnupg encrypted messages are welcome - key ID: 48D0359B ---------------
gnupg fingerprint: 3061 CFBF 2D88 F034 E8D2 7E92 EE4E AC98 48D0 359B -
Attachments: signature.asc (0.18 KB)


dwmw2 at infradead

May 8, 2008, 7:19 AM

Post #4 of 5 (103 views)
Permalink
Re: [OT] first mx has no A record [In reply to]

On Thu, 2008-05-08 at 09:47 +0200, Heiko Schlittermann wrote:
> Does exist some standard (RFC) saying that it's fatal mistake to have no
> A record for the first MX? (But such mistake can happen anytime - if my
> first MX is outside of my domain, and the MX-Admin decides to remove the
> relevant A record. Than I'd rely on my second/third MX.)

It certainly shouldn't be fatal -- and isn't necessarily even a mistake.
For a long time, the primary MX host for infradead.org was
'phoenix.ipv6.infradead.org', which as you might guess from the name
doesn't have an A record.

There were occasional reports of Postfix getting confused, deciding that
it actually resolved to 0.0.0.0 and thus to the local host, and failing
to deliver the mail -- but other than that, it was fine.

--
dwmw2


--
## List details at http://lists.exim.org/mailman/listinfo/exim-users
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/


hs at schlittermann

May 9, 2008, 3:11 PM

Post #5 of 5 (86 views)
Permalink
Re: [OT] first mx has no A record [In reply to]

Hello David,

David Woodhouse <dwmw2[at]infradead.org> (Do 08 Mai 2008 16:19:15 CEST):
> It certainly shouldn't be fatal -- and isn't necessarily even a mistake.
> For a long time, the primary MX host for infradead.org was
> 'phoenix.ipv6.infradead.org', which as you might guess from the name
> doesn't have an A record.
>
> There were occasional reports of Postfix getting confused, deciding that
> it actually resolved to 0.0.0.0 and thus to the local host, and failing
> to deliver the mail -- but other than that, it was fine.

Thank you for your hint. So the failure is quite clearly on the side of
the "mirapoint appliance". Unfortunely mostly such boxes are really
expensive, so the owner/"admin" will not be willing to confess this bug.
And even worse - it's operated by "T-Systems" (as far as I know) - and
alone the names seems to be the prove enough that it can't be on their side....

Best regards from Dresden
Viele Grüße aus Dresden
Heiko Schlittermann
--
SCHLITTERMANN.de ---------------------------- internet & unix support -
Heiko Schlittermann HS12-RIPE -----------------------------------------
gnupg encrypted messages are welcome - key ID: 48D0359B ---------------
gnupg fingerprint: 3061 CFBF 2D88 F034 E8D2 7E92 EE4E AC98 48D0 359B -
Attachments: signature.asc (0.18 KB)

exim users RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact lists@gossamer-threads.com
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.