No announcement yet.

Remove Dead Exchange 2003 Cluster

  • Filter
  • Time
  • Show
Clear All
new posts

  • Remove Dead Exchange 2003 Cluster

    Hi all,

    I apologize that my first post is a technical question rather than an introduction; however I need some assistance if anyone is willing.

    We've recently successfully completed a migration from Exchange 2003 to Exchange 2010, and are now in the process of decommissioning our 2003 exchange servers. We have a few clusters placed in strategic geographic locations, and so far I have had no issues properly decommissioning them.

    Unfortunately, during the final stages of the project, BOTH cluster nodes in a two node cluster in the UK failed. Unfortunately due to the lack of monitoring of those particular servers, they were too far gone, and there was no recovering even one of them.

    The good news? We already migrated all of the accounts we cared about from those servers.

    The bad news? They're still stuck in ESM/AD.

    I've seen a few articles/posts on how to remove a single dead Exchange server by ensuring that OAB generation is not set to that server, as well as ensuring that any accounts homed to that server are changed. At that point, you should simply be able to right-click, and delete in ESM. What I have not seen however; is any posting around how to do this for a failed cluster (probably because generally this doesn't happen often!).

    I reach out to you, in the hopes that someone can shed some light, and hopefully it's not much more difficult than removing a single server.

    Thanks in advance (sorry for the long-windedness).


    Toronto, Canada

  • #2
    Re: Remove Dead Exchange 2003 Cluster

    My first thought is:
    Build a new vm with the same cluster name (but just a single host). Do a repair installation of Exchange 2003 so the server is online and then remove it gracefully.

    Please read this before you post:

    Quis custodiet ipsos custodes?


    • #3
      Re: Remove Dead Exchange 2003 Cluster

      Concur with Andy -- if the machine accounts still exist in AD
      Delete in ESM as a last resort as AD will need to be cleaned up properly
      Tom Jones
      MCT, MCSE (2000:Security & 2003), MCSA:Security & Messaging, MCDBA, MCDST, MCITP(EA, EMA, SA, EDA, ES, CS), MCTS, MCP, Sec+
      PhD, MSc, FIAP, MIITT
      IT Trainer / Consultant
      Ossian Ltd

      ** Remember to give credit where credit is due and leave reputation points where appropriate **


      • #4
        Re: Remove Dead Exchange 2003 Cluster

        Great advice, thank you both. I'll let you know what we do, and whether we're successful or not.

        The difficulty will now be trying to figure out whether building it here in our local data center, versus in the UK, will make a difference at all.



        • #5
          Re: Remove Dead Exchange 2003 Cluster

          I just wanted to post a follow-up to let everyone know that we did not have to resort to the methods suggested below.

          What we ended up doing, was sending someone to the UK site, and we systematically pulled pieces of hardware one-by-one until we found the offender, in both boxes (one riser, and one fiber card). We then magically connected the fiber back up to the one remaining card, and got BOTH cluster nodes back up and running. They'd only stay up for 15 to 20 minutes at a time before crashing, so with a lot of coordination, and tons of attempts, we finally were able to get any remaining accounts out of the databases, and decommission both boxes CLEANLY!

          We're ecstatic that we were able to do this properly, but we really and truly appreciate the suggestions that were given to us by those who responded below.

          Thanks again!