7 Replies Latest reply: Feb 19, 2014 12:50 PM by mpare RSS

    Upgrade to ePO 5.1 failing with "FAILURE: In LaunchAppAndWait"...


      Our current environment consists of an ePO 4.0 server (on a VM running W2K3) that we use to manage all VSE 8.7 clients in an environment consisting of approximately 300 workstations/servers/laptops.  The database resides on a dedicated SQL server running 2008 (specifically, 10.50.4000.0).


      We will be implementing HIPS and Application Control over the next several months, and in preparation for these additions, I've been working on upgrading our ePO server to 5.1


      Prior to making changes to production, I set up a test environment to familiarize myself with the upgrade process.


      So far:

      • I successfully upgraded the existing server to ePO 4.0 Patch 7, to allow upgrade to ePO 4.6.6
      • Successfully upgraded to ePO 4.6.6
      • Followed KB71078 to migrate to 64bit platform:
        • Backed up existing DB and files according to KB article, powered off old VM
        • Set up new VM running Windows Server 2008 R2
        • Installed ePO 4.6.6 and followed instructions to migrate database and files over, and change the path names where needed.  Maintained same DB back-end.
        • All services started successfully, and I was able to manage/see my test workstations and they were updating in the console OK.
        • Double-checked logs for errors - none.
      • Followed KB79561 to purge events from tables (since I was getting the database size warning)
      • Removed incompatible extensions


      Then I run the 5.1 upgrade and when it gets to the point of starting the new services, it "hangs" for a long time, then reports "Rolling back action", and the entire install backs out.


      When I enabled debugging, I see the following error:

      FAILURE: In LaunchAppAndWait while trying to run the following program:

      Return code: 1
      Error message: Incorrect Function.

      CustomAction: MerMod_StartCurrentServices


      When I investigated the logs in %temp%, I see the same error.  It attempts this 45 times before rolling back.


      Unfortunately, it seems like this error is somewhat common, and most of my searches have revealed that this occurs when upgrading/migrating prior revisions.  Either way, I've double-checked settings.  DB is in simple recovery mode.  I'm not seeing any errors on the SQL side.  I even re-tried the entire upgrade process without first importing upgraded packages into the repository.


      I also tried the 'upgradecompatibility' method (exporting/importing file) and I get the same error.


      I'm stumped!