Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

#< #5.6.1 smtp;554 5.6.1 Body type not supported by Remote Host> #SMTP#

2,195 views
Skip to first unread message

Mike Kelly

unread,
Jul 30, 2008, 1:21:32 PM7/30/08
to
Not having much luck with this question in the Connectivity News Group, so I
thought I'd try here.

The above error message began to show up a few weeks ago and I have not been
able to satifactorily figure out which of my exchange servers is the
culprit.

Back ground:
Active Directory Domain (2003 version)
Exchange 2003 SP2 with all the latest paches applied on a Windows 2003 (x32)
Ent server. This server has been active for more than 4 years and has never
thrown this message before.
Forestprep and domain prep run both times an exchange server has been
installed.

In order to prepare for the migration to Exchange 2007, we created a virtual
Windows 2008 (x64) server and installed Exchange 2007 on it. Again letting
MS update to it's hearts content, also letting it sync itself to the domain.
Seemed to do alright.

I then moved my in-active acounts to the new server to see how it was done.
After that step I moved a single active maibox (mine) from the 2k3 server to
the 2k7 server and let it sit. Once the initial teething pains had been
dealt with, it appeared to be stable. I could email anyone (internal domain
addresses and external addresses).

A few weeks ago, I started getting messages with the information in the
above subject line in them and my external emails were being rejected. I
could email to anyone internal to our domain, but emails to the outside were
failing.

I rebooted the Exchange 2008 server late Friday and the external emails
worked for about three messages again and then back to failures.

I can restart the SA on either of the servers and external emails will flow,
but it's inconsistant when it fails. Anywhere from 1 to 4 outbound messages.

I came across messages on the web indicating a MIME mismatch, but can't
figure out where to change it, on the Exc2k3 server or the Exc2k7 server.
or for that matter how. I couldn't locate the entries the messages was
indicating.

Anyone come across this yet?

Any help woul dbe appreciated.

Mike Kelly
mkelly at broadviewsoftware.com


Rich Matheisen [MVP]

unread,
Jul 30, 2008, 8:57:11 PM7/30/08
to
On Wed, 30 Jul 2008 13:21:32 -0400, "Mike Kelly"
<dont.wa...@the.office> wrote:

>Not having much luck with this question in the Connectivity News Group, so I
>thought I'd try here.
>
>The above error message began to show up a few weeks ago and I have not been
>able to satifactorily figure out which of my exchange servers is the
>culprit.

Check your SMTP protocol logs. The 5.6.1 is /usually/ the result of an
Exchange server advertising the "8bitmime" keyword in its response to
the EHLO command. If the sending server sends a message usung 8bitmime
and the Exchange server then tries to send (relay) the message to
another SMTP server that *doesn't* advertise the 8bitmime keyword the
Exchange server throws up its figurative hands, says "I'm not gonna
convert this message to 7-bit MIME", and sends a NDR to the sender.

I don't know if that's your problem, but its easy enough to turn off
the 8bitmime advertisement and see if your problem disappears:

http://support.microsoft.com/kb/257569

On the Exchange 2007 HT role you'd use "set-receiveconnector
<connector-name> -EightBitMimeEnabled:$false"
---
Rich Matheisen
MCSE+I, Exchange MVP

Mike Kelly

unread,
Jul 31, 2008, 4:13:20 PM7/31/08
to
Was able to follow both processes to disable 8bitMIME, but I am still
getting this message back after a small random number of externally outbound
emails.

This is driving me nuts, so please keep the ideas/suggestions coming.


"Rich Matheisen [MVP]" <rich...@rmcons.com.NOSPAM.COM> wrote in message
news:f822949ne16tcn409...@4ax.com...

Rich Matheisen [MVP]

unread,
Jul 31, 2008, 8:30:59 PM7/31/08
to
On Thu, 31 Jul 2008 16:13:20 -0400, "Mike Kelly"
<dont.wa...@the.office> wrote:

>Was able to follow both processes to disable 8bitMIME, but I am still
>getting this message back after a small random number of externally outbound
>emails.
>
>This is driving me nuts, so please keep the ideas/suggestions coming.

Well, is the 5.6.1 error coming from specific domains? If so, create a
SMTP Connector (E2K3) and check the box that says to use HELO. For
E2K7 create a send connector and set parameter -forceHELO:$true. For
both, put the domain names into the connector's address space.

If you never send EHLO you'll never see the 8bitmime keyword and your
servers will never send 8bitmime messages to those domains.

Mike Kelly

unread,
Aug 1, 2008, 10:01:19 AM8/1/08
to
"> Well, is the 5.6.1 error coming from specific domains? "
If you mean external domains, then no. Any domain I try to send to (outside
of our company domains) gets me that message. It appears to come from the
Exc2K3 server which is for all intents and purposes next inline for message
handling. It is acting like a bridgehead but we will eventually get away
from that type of config when the Exc2K3 server is retired.


> If so, create a
> SMTP Connector (E2K3) and check the box that says to use HELO. For
> E2K7 create a send connector and set parameter -forceHELO:$true. For
> both, put the domain names into the connector's address space.

When Exc2K7 installed, it added connectors on both Exchange servers that
point to each other and now it wont let me "recreate" them and it wont let
me edit the Exc2K3 connector on its server. I have to do it from the Exc2K7
server, which I have no idea how to do.

This brings me to what is probably becoming on obvious confession. I have no
idea how to run these commands you fire off that are partials.

For example, I can open the Exchange Power Shell and was able to execute the
last command, but only beause it was complete. "-forceHELO:$true" isn't and
I am at a loss as to what should go in front of it to enabel the command to
complete. Could you expand on it please.

>
> If you never send EHLO you'll never see the 8bitmime keyword and your
> servers will never send 8bitmime messages to those domains.

I have no problem turning to HELO instead of EHLO, but is there any
ramifications to Mailing the Internet in general doing this?

Rich Matheisen [MVP]

unread,
Aug 1, 2008, 8:14:30 PM8/1/08
to
On Fri, 1 Aug 2008 10:01:19 -0400, "Mike Kelly"
<dont.wa...@the.office> wrote:

>"> Well, is the 5.6.1 error coming from specific domains? "
>If you mean external domains, then no. Any domain I try to send to (outside
>of our company domains) gets me that message. It appears to come from the
>Exc2K3 server which is for all intents and purposes next inline for message
>handling. It is acting like a bridgehead but we will eventually get away
>from that type of config when the Exc2K3 server is retired.

Well, it either does, or it doesn't. Check the SMTP protocol logs on
the E2K3 server. It may be getting the 5.6.1 from an upstream server
(your ISP, maybe?) and sending the NDR to the originator. Knowing
which machine is complaining would certainly help you find the
problem.

Is Exchange 2003 your outflow server? Or do you have it send the
outbound mail to another MTA for delivery?

I run a mixed-mode organization and, in the beginning, the E2K7 HT
servers sent the mail to the E2K3 servers, which sent the mail to a
pair of Ironmail servers (for policy enforcement). I didn't have to
change anything except server names in the InterOrg Routing Group
Connector.

>> If so, create a
>> SMTP Connector (E2K3) and check the box that says to use HELO. For
>> E2K7 create a send connector and set parameter -forceHELO:$true. For
>> both, put the domain names into the connector's address space.
>
>When Exc2K7 installed, it added connectors on both Exchange servers that
>point to each other and now it wont let me "recreate" them and it wont let
>me edit the Exc2K3 connector on its server. I have to do it from the Exc2K7
>server, which I have no idea how to do.

The Exchange Management Shell.

>This brings me to what is probably becoming on obvious confession. I have no
>idea how to run these commands you fire off that are partials.
>
>For example, I can open the Exchange Power Shell and was able to execute the
>last command, but only beause it was complete. "-forceHELO:$true" isn't and
>I am at a loss as to what should go in front of it to enabel the command to
>complete. Could you expand on it please.

set-sendconnector <name> -forceHELO:$true

To see what's been set on the same connector:

get-sendconnector <name> | fl

>> If you never send EHLO you'll never see the 8bitmime keyword and your
>> servers will never send 8bitmime messages to those domains.
>
>I have no problem turning to HELO instead of EHLO, but is there any
>ramifications to Mailing the Internet in general doing this?

There are some drawbacks. But SMTP operated (and still does) for years
without turning to ESMTP.

Mike Kelly

unread,
Aug 5, 2008, 1:40:10 PM8/5/08
to
I was looking through the logs on the Exc2K3 server and found that the
Exc2K7 server is advertising itself under one domain but the AD is resolving
to another domain. Would this cause the kind of behaviour I have been
seeing?


I turned off the EHLO on the Exc2K7 server Send Connector and restarted the
SA. Got the same error.

Mike Kelly

unread,
Aug 5, 2008, 3:58:22 PM8/5/08
to
How can edit the settings of the connector that Exc2K7 made but is placed on
the Exc2K3 server?

"Mike Kelly" <dont.wa...@the.office> wrote in message
news:%23TPCHJy...@TK2MSFTNGP05.phx.gbl...

Rich Matheisen [MVP]

unread,
Aug 5, 2008, 8:42:45 PM8/5/08
to
On Tue, 5 Aug 2008 13:40:10 -0400, "Mike Kelly"
<dont.wa...@the.office> wrote:

>I was looking through the logs on the Exc2K3 server and found that the
>Exc2K7 server is advertising itself under one domain but the AD is resolving
>to another domain.

Huh? Your servers are all in the same AD forest, right? And they use
the same DNS? While you can have multiple domains in the forest your
DNS should be resolving names to IP addresses in domains for which
it's authoritative. IOW, the domains in the AD forest.

What do you mean by "advertising itself"?

>Would this cause the kind of behaviour I have been
>seeing?

You have multiple AD forests and sepaate Exchange organizations in
each one, each of them named identically??

>I turned off the EHLO on the Exc2K7 server Send Connector and restarted the
>SA. Got the same error.

You shouldn't be using a send connector between servers in the same
Exchange organization. Mail between the Exchange 2003 and Exchange
2007 servers should be moving over the InterOrg Routing Group that was
created from your E2K7 HT server and an E2K3 bridgehead server.

Mike Kelly

unread,
Aug 6, 2008, 8:41:54 AM8/6/08
to

> Huh? Your servers are all in the same AD forest, right? And they use
> the same DNS? While you can have multiple domains in the forest your
> DNS should be resolving names to IP addresses in domains for which
> it's authoritative. IOW, the domains in the AD forest.
>
> What do you mean by "advertising itself"?
>

Both servers are in the same domain, but that AD domain isnt the one we want
to advertise to the world. We actually host about six mailing domains, one
of which is the AD domain but we are trying to migrate from that domain for
Internet communicatiuons.

>
> You have multiple AD forests and sepaate Exchange organizations in
> each one, each of them named identically??

Nope, one forest and a child domain, but it isnt hosting an exchange server.

>
>>I turned off the EHLO on the Exc2K7 server Send Connector and restarted
>>the
>>SA. Got the same error.
>
> You shouldn't be using a send connector between servers in the same
> Exchange organization. Mail between the Exchange 2003 and Exchange
> 2007 servers should be moving over the InterOrg Routing Group that was
> created from your E2K7 HT server and an E2K3 bridgehead server.

The send connector in the in Exc2K7 server is "Internet", but the Exc2k3
server has an Internet conenctor and a connector that is for inter-excange
communications to the Exc2k7 server.


Rich Matheisen [MVP]

unread,
Aug 6, 2008, 12:59:14 PM8/6/08
to
On Wed, 6 Aug 2008 08:41:54 -0400, "Mike Kelly"
<dont.wa...@the.office> wrote:

>
>> Huh? Your servers are all in the same AD forest, right? And they use
>> the same DNS? While you can have multiple domains in the forest your
>> DNS should be resolving names to IP addresses in domains for which
>> it's authoritative. IOW, the domains in the AD forest.
>>
>> What do you mean by "advertising itself"?
>>
>
>Both servers are in the same domain, but that AD domain isnt the one we want
>to advertise to the world. We actually host about six mailing domains, one
>of which is the AD domain but we are trying to migrate from that domain for
>Internet communicatiuons.

That doesn't really answer the question, though. Does "advertise" mean
it shows something in the 220 SMTP banner? Or in the HELO\EHLO command
data? Or something else? And, if that's what you mean, can whatever
the names is, can it be resolved in DNS?

>> You have multiple AD forests and sepaate Exchange organizations in
>> each one, each of them named identically??
>
>Nope, one forest and a child domain, but it isnt hosting an exchange server.
>
>>
>>>I turned off the EHLO on the Exc2K7 server Send Connector and restarted
>>>the
>>>SA. Got the same error.
>>
>> You shouldn't be using a send connector between servers in the same
>> Exchange organization. Mail between the Exchange 2003 and Exchange
>> 2007 servers should be moving over the InterOrg Routing Group that was
>> created from your E2K7 HT server and an E2K3 bridgehead server.
>
>The send connector in the in Exc2K7 server is "Internet", but the Exc2k3
>server has an Internet conenctor and a connector that is for inter-excange
>communications to the Exc2k7 server.

Okay, so mail sent to the Internet leaves the organization through the
E2K3 server's SMTP virtual server for users with mailboxes on the E2K3
side of the InterOrg Routing Group Connector, and mail to the Internet
leaves the organization through the Send Connector on the E2K7 side of
the InterOrg Routing Group Connector. Which of the servers, E2K3 or
E2K7, gets the 5.6.1 status, and when they do, what was the server
that responded with that status? Was the server that responded with
that status one of your servers, or was it some server outside your
organization?

Mike Kelly

unread,
Aug 6, 2008, 2:03:56 PM8/6/08
to
Here is the message I get in totality: (I changed the @ symbol in the email
addresses to # to prevent harvesting)

Diagnostic information for administrators:

Generating server: mail.broadviewsoftware.com

mi...@mekjr.ca
#< #5.6.1 smtp;554 5.6.1 Body type not supported by Remote Host> #SMTP#

Original message headers:

Received: from BVSMail.atkin.com ([172.20.1.42]) by
mail.broadviewsoftware.com
with Microsoft SMTPSVC(6.0.3790.3959); Wed, 6 Aug 2008 13:45:02 -0400
X-GFI-TNEF2MIME: 1.2
Received: from BVSMail.atkin.com ([fe80::66:761f:d6a8:bb5a]) by
BVSMail.atkin.com ([fe80::66:761f:d6a8:bb5a%10]) with mapi; Wed, 6 Aug 2008
13:44:54 -0400
From: "Kelly, Michael" <MKelly#BroadViewSoftware.com>
To: "mi...@mekjr.ca" <mike#mekjr.ca>
Date: Wed, 6 Aug 2008 13:44:47 -0400
Subject: test 1
Thread-Topic: test 1
thread-index: Acj37B5lJ7nZ5ns+QQeK8mx1pbft8g==
Message-ID: <F051BB82787E0D41A88C0...@BVSMail.atkin.com>
Accept-Language: en-US, en-CA
Content-Language: en-US
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
<F051BB82787E0D41A88C0...@BVSMail.atkin.com>
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.3790.4133
MIME-Version: 1.0
Content-Type: multipart/mixed;
boundary="part__050D2663_1856_47B8_98D8_43825880D"
Return-Path: <MKelly#BroadViewSoftware.com>
X-OriginalArrivalTime: 06 Aug 2008 17:45:02.0281 (UTC)
FILETIME=[273E8390:01C8F7EC]


I am still working my way through the logs to see if I can get any kind of
sense out of them.

"Rich Matheisen [MVP]" <rich...@rmcons.com.NOSPAM.COM> wrote in message

news:jklj94t3t4icshm5h...@4ax.com...


> On Wed, 6 Aug 2008 08:41:54 -0400, "Mike Kelly"
> <dont.wa...@the.office> wrote:
>
>>
>>> Huh? Your servers are all in the same AD forest, right? And they use
>>> the same DNS? While you can have multiple domains in the forest your
>>> DNS should be resolving names to IP addresses in domains for which
>>> it's authoritative. IOW, the domains in the AD forest.
>>>
>>> What do you mean by "advertising itself"?
>>>
>>
>>Both servers are in the same domain, but that AD domain isnt the one we
>>want
>>to advertise to the world. We actually host about six mailing domains, one
>>of which is the AD domain but we are trying to migrate from that domain
>>for
>>Internet communicatiuons.
>
> That doesn't really answer the question, though. Does "advertise" mean
> it shows something in the 220 SMTP banner? Or in the HELO\EHLO command
> data? Or something else? And, if that's what you mean, can whatever
> the names is, can it be resolved in DNS?
>

In this context, Advertise means the hostname.domain that the server
provides when identifying itself to other SMTP servers. Our Exchange 2003
server is setup with a RDNS entry. The Exchange 2007 server is currently
not, but will take over the RDNS name and IP when we retire the 2003.

> Okay, so mail sent to the Internet leaves the organization through the
> E2K3 server's SMTP virtual server for users with mailboxes on the E2K3
> side of the InterOrg Routing Group Connector, and mail to the Internet
> leaves the organization through the Send Connector on the E2K7 side of
> the InterOrg Routing Group Connector. Which of the servers, E2K3 or
> E2K7, gets the 5.6.1 status, and when they do, what was the server
> that responded with that status? Was the server that responded with
> that status one of your servers, or was it some server outside your
> organization?


Current connection layout:

Internet
|
|
Firewall (SMTP rules and NAT)
|
|
Exchange 2003 Server - most of the mailboxes are hosted here
|
|
Echange 2007 Server - Currenty, my mailbox is the only active one.

For Internet email to get to me, it has to pass through the 2003 server. For
my outbound email, to external domains, I am guessing the 2007 server sends
it straight out. but it looks like the 2003 server is somehow filtering it
first.


Mike Kelly

unread,
Aug 6, 2008, 2:10:26 PM8/6/08
to
And found this bit in the 2007 server log (again with the # instead of @)

BVSMail is the 2k7 server and mdant is the 2k3 server.


2008-08-06T17:45:02.387Z,172.x.x.x,BVSMail,172.x.x.y,mdant.atkin.com,08CAC45B8F70B413,Intra-Organization
SMTP Send Connector,SMTP,SEND,334,,mike#mekjr.ca,250 2.1.5 mike#mekjr.ca
,4066,1,,,,,MKelly#BroadViewSoftware.com,2008-08-06T17:44:51.573Z

2008-08-06T17:45:07.231Z,172.x.x.x,,172.x.x.y,BVSMail,08CAC45B8F70B414;2008-08-06T17:45:05.293Z;0,BVSMAIL\Default
BVSMAIL,SMTP,RECEIVE,335,<bfa5rZAk...@mail.broadviewsoftware.com>,MKelly#BroadViewSoftware.com,,6020,1,,,Delivery
Status Notification (Failure),postmaster#BroadViewSoftware.com,<>,0bI:

2008-08-06T17:45:13.575Z,,BVSMail,,bvsmail,,,STOREDRIVER,DELIVER,335,,MKelly#BroadViewSoftware.com,,6110,1,,,,,<>,2008-08-06T17:45:06.953Z


Rich Matheisen [MVP]

unread,
Aug 6, 2008, 8:52:20 PM8/6/08
to
On Wed, 6 Aug 2008 14:03:56 -0400, "Mike Kelly"
<dont.wa...@the.office> wrote:

>Here is the message I get in totality: (I changed the @ symbol in the email
>addresses to # to prevent harvesting)
>
>Diagnostic information for administrators:
>
>Generating server: mail.broadviewsoftware.com

^^^^^^^^^^^^^^^^^^^^^^^^^^
So, who is this? ---------------+

This machine claims it will accept email delivered with 8bitmime
(check the results of EHLO to their machine). My gues is that it does,
but then tries to deliver the message to a machine that cannot accept
email in that format. At that point the mail.broadviewsoftware.com
machine returns a NDR saying that the body tryp isn't supported.

Is mail.broadviewsoftware.com your machine? If not, the problem lies
with someone else's machine. If you use SMTP (sendin HELO) instead of
ESMTP (sending EHLO) when you send to that domain do you get the same
failure?

Mike Kelly

unread,
Aug 7, 2008, 11:11:19 AM8/7/08
to

>>
>>Generating server: mail.broadviewsoftware.com
> ^^^^^^^^^^^^^^^^^^^^^^^^^^
> So, who is this? ---------------+
>

That's my Exc2K3 server.

> This machine claims it will accept email delivered with 8bitmime
> (check the results of EHLO to their machine). My gues is that it does,
> but then tries to deliver the message to a machine that cannot accept
> email in that format. At that point the mail.broadviewsoftware.com
> machine returns a NDR saying that the body tryp isn't supported.
>
> Is mail.broadviewsoftware.com your machine? If not, the problem lies
> with someone else's machine. If you use SMTP (sendin HELO) instead of
> ESMTP (sending EHLO) when you send to that domain do you get the same
> failure?

The Send connector on the 2k7 server is set to HELO (as listed in the
"get-sendconnector <connector name> | flbut) but the logs on the 2k3 server
show it still connecting as EHLO.

I also just discovered that if I reset the SA on the 2k3 server I can sneak
a message out then it stops the messages again.

Rich Matheisen [MVP]

unread,
Aug 7, 2008, 1:58:35 PM8/7/08
to
On Thu, 7 Aug 2008 11:11:19 -0400, "Mike Kelly"
<dont.wa...@the.office> wrote:

>
>>>
>>>Generating server: mail.broadviewsoftware.com
>> ^^^^^^^^^^^^^^^^^^^^^^^^^^
>> So, who is this? ---------------+
>>
>
>That's my Exc2K3 server.
>
>> This machine claims it will accept email delivered with 8bitmime
>> (check the results of EHLO to their machine). My gues is that it does,
>> but then tries to deliver the message to a machine that cannot accept
>> email in that format. At that point the mail.broadviewsoftware.com
>> machine returns a NDR saying that the body tryp isn't supported.
>>
>> Is mail.broadviewsoftware.com your machine? If not, the problem lies
>> with someone else's machine. If you use SMTP (sendin HELO) instead of
>> ESMTP (sending EHLO) when you send to that domain do you get the same
>> failure?
>
>The Send connector on the 2k7 server is set to HELO (as listed in the
>"get-sendconnector <connector name> | flbut) but the logs on the 2k3 server
>show it still connecting as EHLO.

Try restarting the transport service on the E2K7 server. But I don't
think that's going to work becasue you're really using the RGC to send
the mail between the two Exchange servers.

Check the SMTP log on the E2K3 server and see whether it's your E2K3
server that's sending the status, or if the E2K3 server' receiving the
status from the server it's sending the mail to.

Message has been deleted
0 new messages