Can someone please help.

 

We have just upgraded one of our 2007 Exchange clusters to SP1, the upgrade process has failed.

 

These are steps we took to upgrade:

 

1.    Upgrade the CAS/Hubs (worked!)

2.    Upgrade the passive node (setup /m:upgrade) (worked!)

3     reboot passive node

4    stop cluster (stop-clustermailboxServer) (worked!)

5    Move-ClusteredMailboxServer to passive node (worked!)

6    upgrade CMS (setup /upgradecms) on the now Active server (worked!)

7    At this point the cluster came back online.

8    on passive node, ran setup /m:upgrade (failed Prerequisite Check!)

This was the error:

 

    Mailbox Role Checks              ......................... FAILED
     One or more network interface is DHCP enabled. As a best practice, you should s
    tatically assign an IP address to each network interface.
     Before Setup can continue, all cluster managed services owned by this node
    need to be taken offline or preferably moved to another node in the cluster.


    The Exchange Server Setup operation did not complete. For more information, visi
    t http://support.microsoft.com and enter the Error ID.

 

    Exchange Server setup encountered an error.

 

 

The cluster is now in a failed state and will not come online.

 

Can someone please help me to finish the sp1 upgrade.

Incidentally, I have carried out the upgrade on our test environment without any issues.

Is this CCR or SCC? Either way i'm surprised it's left the upgraded node unable to bring the resources online - you've tried moving the resources to the SP1 upgraded node right?

In regard to the failure, either disable the NIC that you've left on DHCP or assign it a static address and then perform the SP1 upgrade on this node again.

Oliver

Thanks Oliver for your reply,

In answer to your question we have CCR.

We can bring the cluster online by moving the Cluster to the Upgraded node, however, we cannot move the cluster group over.
The cluster is now up. But is not replicating.

We still cannot upgrade the last node.




  "Oliver Moazzezi [MVP]"  wrote in message news:%23MgBiT8SIHA.3516@TK2MSFTNGP02.phx.gbl...
  Is this CCR or SCC? Either way i'm surprised it's left the upgraded node unable to bring the resources online - you've tried moving the resources to the SP1 upgraded node right?

  In regard to the failure, either disable the NIC that you've left on DHCP or assign it a static address and then perform the SP1 upgrade on this node again.

  Oliver

Have you fixed the DHCP NIC warning?

Follow this article to upgrade a CCR cluster to SP1

http://technet.microsoft.com/en-us/library/bb676320.aspx

Oliver

Maybe you have network teaming configured on your interfaces; If it's the 
case, i found that sometimes (for some products) the physical interface are 
checked even if TCPIP is not bound to them (TCPIP is only bound to the 
virtual interface). You will have to disable teaming, configure a static IP 
address on every physical interface an then reconfigure teaming (even if it 
seems stupid :-) )

Hope it helps
"Oliver Moazzezi [MVP]"  a écrit dans le 
message de news:%23NbXnFjTIHA.4768@TK2MSFTNGP02.phx.gbl...
Have you fixed the DHCP NIC warning?

Follow this article to upgrade a CCR cluster to SP1

http://technet.microsoft.com/en-us/library/bb676320.aspx

Oliver

Its not a dhcp nic issue.
I know this because I have upgraded other servers with dhcp enabled.
The dhcp message is just a warning and setup can continue with dhcp enabled.

The problem is to do with resources still residing on that node, such as 
cluster and quorum.
Setup will not contine unless these resources are moved over to the other 
node.

My problem is that although I can move the cluster to the other node, I 
cannot move the quorum over.
This is my problem.



"Stephen ROUX"  wrote in message 
news:%23AtQetuTIHA.3532@TK2MSFTNGP04.phx.gbl...
> Maybe you have network teaming configured on your interfaces; If it's the 
> case, i found that sometimes (for some products) the physical interface 
> are checked even if TCPIP is not bound to them (TCPIP is only bound to the 
> virtual interface). You will have to disable teaming, configure a static 
> IP address on every physical interface an then reconfigure teaming (even 
> if it seems stupid :-) )
>
> Hope it helps
> "Oliver Moazzezi [MVP]"  a écrit dans le 
> message de news:%23NbXnFjTIHA.4768@TK2MSFTNGP02.phx.gbl...
> Have you fixed the DHCP NIC warning?
>
> Follow this article to upgrade a CCR cluster to SP1
>
> http://technet.microsoft.com/en-us/library/bb676320.aspx
>
> Oliver

The fact that you cannot move the quorum was not in the initial post. Could 
you please provide us more informations on that new issue.


"Trejay Rathore"  a écrit dans le message de 
news:14DE94EA-FDA0-4E4D-92B3-1744B7BFED64@microsoft.com...
> Its not a dhcp nic issue.
> I know this because I have upgraded other servers with dhcp enabled.
> The dhcp message is just a warning and setup can continue with dhcp 
> enabled.
>
> The problem is to do with resources still residing on that node, such as 
> cluster and quorum.
> Setup will not contine unless these resources are moved over to the other 
> node.
>
> My problem is that although I can move the cluster to the other node, I 
> cannot move the quorum over.
> This is my problem.
>
>
>
> "Stephen ROUX"  wrote in message 
> news:%23AtQetuTIHA.3532@TK2MSFTNGP04.phx.gbl...
>> Maybe you have network teaming configured on your interfaces; If it's the 
>> case, i found that sometimes (for some products) the physical interface 
>> are checked even if TCPIP is not bound to them (TCPIP is only bound to 
>> the virtual interface). You will have to disable teaming, configure a 
>> static IP address on every physical interface an then reconfigure teaming 
>> (even if it seems stupid :-) )
>>
>> Hope it helps
>> "Oliver Moazzezi [MVP]"  a écrit dans le 
>> message de news:%23NbXnFjTIHA.4768@TK2MSFTNGP02.phx.gbl...
>> Have you fixed the DHCP NIC warning?
>>
>> Follow this article to upgrade a CCR cluster to SP1
>>
>> http://technet.microsoft.com/en-us/library/bb676320.aspx
>>
>> Oliver
>

The dhcp issue is a red herring.
The origanal error was:

***********************************************************************************
Mailbox Role Checks              ......................... FAILED
     One or more network interface is DHCP enabled. As a best practice, you 
should s
    tatically assign an IP address to each network interface.

     Before Setup can continue, all cluster managed services owned by this 
node
    need to be taken offline or preferably moved to another node in the 
cluster.


    The Exchange Server Setup operation did not complete. For more 
information, visi
    t http://support.microsoft.com and enter the Error ID.
***********************************************************************************

The first paragraph warns about the dhcp issue
The second states that setup cannot continue unless all cluster managed 
services are taken offline or moved.

As stated previously I can move the cluster to the upgraded node, however, I 
cannot move the quorum.

Moving the quorum seems to be my problem.




"Stephen ROUX"  wrote in message 
news:%23mYF1jWUIHA.5404@TK2MSFTNGP03.phx.gbl...
> The fact that you cannot move the quorum was not in the initial post. 
> Could you please provide us more informations on that new issue.
>
>
> "Trejay Rathore"  a écrit dans le message de 
> news:14DE94EA-FDA0-4E4D-92B3-1744B7BFED64@microsoft.com...
>> Its not a dhcp nic issue.
>> I know this because I have upgraded other servers with dhcp enabled.
>> The dhcp message is just a warning and setup can continue with dhcp 
>> enabled.
>>
>> The problem is to do with resources still residing on that node, such as 
>> cluster and quorum.
>> Setup will not contine unless these resources are moved over to the other 
>> node.
>>
>> My problem is that although I can move the cluster to the other node, I 
>> cannot move the quorum over.
>> This is my problem.
>>
>>
>>
>> "Stephen ROUX"  wrote in message 
>> news:%23AtQetuTIHA.3532@TK2MSFTNGP04.phx.gbl...
>>> Maybe you have network teaming configured on your interfaces; If it's 
>>> the case, i found that sometimes (for some products) the physical 
>>> interface are checked even if TCPIP is not bound to them (TCPIP is only 
>>> bound to the virtual interface). You will have to disable teaming, 
>>> configure a static IP address on every physical interface an then 
>>> reconfigure teaming (even if it seems stupid :-) )
>>>
>>> Hope it helps
>>> "Oliver Moazzezi [MVP]"  a écrit dans le 
>>> message de news:%23NbXnFjTIHA.4768@TK2MSFTNGP02.phx.gbl...
>>> Have you fixed the DHCP NIC warning?
>>>
>>> Follow this article to upgrade a CCR cluster to SP1
>>>
>>> http://technet.microsoft.com/en-us/library/bb676320.aspx
>>>
>>> Oliver
>>
>

Hi,

I have just read your post about SP1 failure.  I am experiencing the same 
issue - I am at the stage where I have upgraded CA, HT, the passive node and 
the cluster, I am just about to start the upgrade on the was active node but 
it says that all resources have not failed over.  It still thinks that the 
server I am about to upgrade is the quorum owner.

Did you manage to get around the error?
Regards,

Julie

Hi Julie,

Yes, I finally got the resources over to the other node.
I took serveral days and a few reboots and then it allowed me to move over 
the quourm.

Trejay



"Julie Williams"  wrote in message 
news:825D5218-86FD-4FBB-9273-DC9106C0C0A2@microsoft.com...
> Hi,
>
> I have just read your post about SP1 failure.  I am experiencing the same
> issue - I am at the stage where I have upgraded CA, HT, the passive node 
> and
> the cluster, I am just about to start the upgrade on the was active node 
> but
> it says that all resources have not failed over.  It still thinks that the
> server I am about to upgrade is the quorum owner.
>
> Did you manage to get around the error?
> Regards,
>
> Julie