Process MAD.EXE . All Domain Controller Servers in use

[Chaupa]

Source MSExchangeDSAccess
I have multiple DC's and GC's on the same subnet even.
Is there a fix for this that anybody knows of.
I have done a lot of searching but no luck.
Thanks,
Paul

 

[Xybertron]

2102? Can you be a little more specific on the error.

In event viewer you can open an event and click the copy button it makes it real easy to work with. Dan
Microsoft Exchange Support @ Microsoft

 

[Chaupa]

I can't seem copy the message so I copied the text.The error is 2102 and the source is MSExchangeDsAccess and the catagory is Typology.

The Exchange 2000 server is running sp2.
Windows 2000 sp2.


Process MAD.EXE (PID=2004). All Domain Controller Servers in use are not responding:
ntdc01.mycompany.org - This is a DC and GC
NTDNS01.mycompany.org - This is a DC and GC

 

[Xybertron]

How often does this happen?

Are the Directory Service logs clean in event viewer?

Dan
Microsoft Exchange Support @ Microsoft

 

[Chaupa]

The Dcaccess error happens at least every half hour.
There are no error in the Directory Service logs.
I have installed all the Microsoft recommended security and performance fixes.
The ADC is running on the NTDC01 with replication set to always.
I found out that one of my DNS servers seems to be having problems so I statically assigned the good one on the Exchange server. I did note that there is no forwarder on the good DNS server. I don't know if that is a problem, but it is probably worth mentioning.
The Dsaccess Tab of the Exchange 2000 server points to the two Domain controllers and Catalog servers.
Thanks
Paul
Paul

 

[Xybertron]

Do both DC's point to the same DNS Server? Dan
Microsoft Exchange Support @ Microsoft

 

[Chaupa]

The network consultant has the ntdc01 configured to use DHCP for all options. This uses all three DNS servers and one is not a 2000 DNS server.

The ntdns01 is hard coded to use itself as primary and no secondary. This is a DC - DNS - WINS. This is a 2000 Server.

Paul

 

[C5Thunder]

According to MS, this is a normal message and it has to do with kerberos auth. The server with automatically reestablish the connections.

 

[Xybertron]

Yes I said that too... Its not really normal with 2 DC's though.

Point both DC's to the same DNS and only that DNS. Then use forwarders to the ISP. Dan
Microsoft Exchange Support @ Microsoft

 

[Monty87]

I just dicovered this error on a clients server today while investigating another error. The error I was investigating was that back ups (Veritas) was failing every 3 - 4 days.

I don't think the Kerberos Ticket issue is causing the problem but does anyone know if this is possible. My thought was that perhaps the ticket time out was causing a client disconnection resulting in the following error.

Information Store (3312) 74fde1ef-f644-4dd5-9179-de7c76d9acb1: The backup has been stopped because it was halted by the client or the connection with the client failed


The client is a very small business and would not put in a second DC so I wondered if there is another solution.

If not is there any mileage in making the kerberos ticket valid for an extremely long period thereby reducing the problem.

Regards

Monty

 

[Chaupa]

I tried having both DC pointing to the same dns server and no luck.
How does DSAccess work. Is there a time out value, I understand DSaccess looks up information from the Catalog servers and sends e-mails to users. How could it fail if on the DSAccess tab of the server the DC and GC are listed.
Have you seen a lot of these issues at Microsoft?

Paul.

 

[Xybertron]

Not with it happening every 30 mins no.

You must never point Win2k to a non-Win2k box for DNS unless it is at least Bind 8.1.2

DSAccess is how we get to the AD. AD has this setting not Exchange. You can change the Kerberos ticket timeouts but basically if you raise it that is a big security hole.

Once a user logs on they get a ticket. So you disable that account and they can still do everything for as long as the ticket is valid. Dan
Microsoft Exchange Support @ Microsoft

 

[Chaupa]

Thanks,
I am going to try a few more things.
Paul.

 

[Chaupa]

I figured I would provide what seems to be the answer to this issue I was having.
The Exchange Server had a msExchMailboxGUID I could not update.The reason was the allow inheritence box was unchecked. We had seen the error but did not think it was related because we could not find the GUID. You have to perform a little flipping of the GUID to use LDAP to find it in AD.
It has been 24 hours and no errors.
Thanks for the help.
Paul