idiots guid to reverse DNS?

[telecotek1]

My mail server is operational accept for the ocasional bounced mail. users occasionally receive an error that reads..

Diagnostic information for administrators:

Generating server: machinename.domain.local

mike@company.com
s405.sureserver.com #554 "Refused. You have no reverse DNS entry. Contact abuse@suresupport.com for details." ##

Original message headers:

Received: from machinename.domain.local
([fe80::1b7:8ea6:58d7:7fd5]) by machinename.domain.local
([fe80::1b7:8ea6:58d7:7fd5%11]) with mapi; Mon, 1 Mar 2010 18:33:59 -0500
From: john smith <jsmith@domain.com>
To: Mike <mike@company.com
Date: Mon, 1 Mar 2010 18:33:44 -0500
Subject: Re: Payment
Thread-Topic: Payment
Thread-Index: Acq5l6qUoMe9gBW8QyqGKrKVaSrk/A==
Message-ID: <EB8C3654-BA84-4D8B-B8CB-E64D84C7037F@cobaltmed.com>
References: <010301cab98d$9a64e030$cf2ea090$@com>
In-Reply-To: <010301cab98d$9a64e030$cf2ea090$@com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator:
acceptlanguage: en-US
Content-Type: multipart/mixed;
boundary="_002_EB8C3654BA844D8BB8CBE64D84C7037Fcobaltmedcom_"
MIME-Version: 1.0

I asume that I need to make a change to remove machinename.domain.local and insert the FQN of remote.domain.com. Is that correct and if so where do I do that in exchange 2007?

Also the ISP needs to create a PTR. What should that PTR look like - The public ip of the mailserver points to remote.domain.com?

lastly this particular mail server uses app river and when you do a reverse lookup of remote.domain.com it lands on app river. does that mean that a PTR is not needed to be created by the ISP?

Thanks in advance..

 

[ShackDaddy]

Go to your Internet Send Connector. You'll see it under Organization -> Hub Transport. You'll see the URL in there.

Yes, that's what the PTR needs to look like.

The MX record should point to AppRiver, but a reverse lookup of your remote.domain.com IP shouldn't return AppRiver as a result. Does remote.domain.com resolve to the public IP of this office, the address you use to access OWA?

The only thing I can think of is that AppRiver might have had someone create an RDNS\PTR record at the ISP pointing to them.

One thing you could try is putting whatever that RDNS record is (mail23.appriver.com or whatever) into the Send Connector URL. What that would do is map your server's name to what the existing PTR record is, and that may prevent your emails from being dropped at the destination server. But that's not the ideal solution.

Dave Shackelford MVP
ThirdTier.net

 

[telecotek1]

Thanks shackdaddy. But it seems that it's already set correctly at remote.domain.com. Im in Orginization Configuration/Hub Transport - right clicked on windows SBS Internet Send. In the field labeled "Specify the FQDN this connector will provide in response to HELO and EHLO" is remote.domain.com. Am in the wrong place?

 

[telecotek1]

when i telnet to servername 25 i get 220 maichinename.domain.local Microsoft ESMTP.... Shouldn't I get what's in that connecter field? maybe somehow it's corrupted? I should renter something dif then save then renter and save and test? Just a thought.

 

[58sniper]

You need to go to your ISP. They are the only ones who can change your reverse DNS record. That's a record that maps a name to the PUBLIC IP address your server is using to send OUTBOUND mail.

Let's say your send connector is configured for remote.domain.com. Your ISP should set the reverse DNS record (called a PTR record) for your public IP address to the same thing.

To check your record, go to

http://www.whatismyip.com/.

Grab the address it gives you, and paste it in at

http://postmaster.aol.com/tools/rdns.html.

If the result is something like blahblahblah.yourisp.com, it needs to be changed.

Pat Richard MVP
Plan for performance, and capacity takes care of itself. Plan for capacity, and suffer poor performance.

http://www.ucblogs.net/blogs/exchange/

 

[ShackDaddy]

When you telnet to port 25, you're looking at your receive connector, not your send connector. Since you are having trouble with outbound mail, I was wanting you to look at the send connector.

Dave Shackelford MVP
ThirdTier.net

 

[telecotek1]

Thanks all for the support so far. A quick update the ISP claims that a PTR should be complete within 4 to 6 hours. The one thing that has me worried is the SMTP Banner. If it reads machinename.domain.com then the recieving server will try to RDNS something other then a FQDN.

ShackDaddy - I did check what you suggested and it is indeed the FQDN. I'm thinking if thats the only place that I should be checking then something else is wrong. I'm not affraid of the shell is there a command that I can run to look at what it is currently using for a banner and another commenad to change it?

Much much appreciated btw.

 

[telecotek1]

Set-sendConnector Default SERVERNAME -Banner "220 remote.domain.com"

Would this work assuming that "Default SERVERNAME" was the name of the one that I wanted to modify? What else could this possibly impact?

 

[58sniper]

Don't use the default send connector for outbound email. Create a dedicated send connector, and a dedicated receive connector for Internet related email. You get much more granularity.

Pat Richard MVP
Plan for performance, and capacity takes care of itself. Plan for capacity, and suffer poor performance.

http://www.ucblogs.net/blogs/exchange/

 

[telecotek1]

Granularity? You mean from a historical/trouble shooting stand point?

 

[58sniper]

and flexibility.

Pat Richard MVP
Plan for performance, and capacity takes care of itself. Plan for capacity, and suffer poor performance.

http://www.ucblogs.net/blogs/exchange/

 

[telecotek1]

Still can't send email to AOL. Believe it or not my client sneds a ton of mail to aol. Still not working take a look at the latest results..

mxtool passed OK

220 remote.domain.com Microsoft ESMTP MAIL Service ready at Fri, 5 Mar 2010 13:17:04 -0500


Not an open relay.
0 seconds - Good on Connection time
5.195 seconds - Warning on Transaction time
OK - xxx.xxx.xxx.xxx resolves to remote.domain.com
OK - Reverse DNS matches SMTP Banner

Session Transcript:
HELO please-read-policy.mxtoolbox.com
250 remote.domain.com Hello [64.20.227.133] [47 ms]
MAIL FROM: <supertool@mxtoolbox.com>
250 2.1.0 Sender OK [47 ms]
RCPT TO: <test@example.com>
550 5.7.1 Unable to relay [5054 ms]
QUIT
221 2.0.0 Service closing transmission channel [47 ms]

How-ever from aol..

Diagnostic information for administrators:

Generating server: servername.domain.local

fsmith@aol.com
#550 4.4.7 QUEUE.Expired; message expired ##

Original message headers:

Received: from servername.domain.local
([fe80::1b7:8ea6:58d7:7fd5]) by servername.domain.local
([fe80::1b7:8ea6:58d7:7fd5%11]) with mapi; Tue, 2 Mar 2010 19:58:55 -0500
From: name@acme.com <name@acme.com>
To: "fsmith@aol.com" <fsmith@aol.com>
Date: Tue, 2 Mar 2010 19:58:55 -0500
Subject: test 2 reverse dns
Thread-Topic: test 2 reverse dns
Thread-Index: AQHKumyyHOpmkqOV7kiMZCOODFDl4g==
Message-ID: <9ACDD0A1567FF447A8D2F973CDC842932DB74A30@servername.domain.local>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator:
acceptlanguage: en-US
Content-Type: multipart/mixed;
boundary="_004_9ACDD0A1567FF447A8D2F973CDC842932DB74A30Servername_"
MIME-Version: 1.0

Now I have deleted the Send Connector and created a new one. Does anyone have an idea why my server is still sending servername.domain.local?? this is driving me nutz!

 

[telecotek1]

Scratch that it's working now. Stil not sure why tho if my smtp banner is clearly wrong as per aol's error message. If anyone has any info on what that was all about please share.

Thanks to everyone who helped me out much much appreciated!

j