You’ve installed an SCCM primary site server, extended the AD schema, created the container, named it wrong, screwed up the permissions, misconfigured the SQL instance, hosed up the firewall port settings, granted the wrong accounts to the wrong services with the wrong permissions, added too many roles and didn’t configure them right, deployed a bunch of broken things, too many updates, and reimaged half the environment by mistake.

Oh, and you ate the wrong breakfast, drank the wrong juice, and now you and your SCCM site have puked all over the place, something like this.


If you just ate, I apologize. 🙂

Anyhow, moving along… It’s not over.  You still have time to start over, AND screw it up again. 🙂  I would recommend that before you do this again in a production environment, you stand up a test lab, even if it’s just VMware or Hyper-V on your laptop.

Here’s what to do.

  1. Find the person who installed the site and related features.
  2. Cover their face in a soft cotton towel, thoroughly soaked in Chloroform, or one of those cheap-smelled perfumes the teenage girls (or boys) wear these days.
  3. Drag their limp body to a nearby 9 ft-long cushioned sofa. Grab a blanket and a teddy bear.
  4. Cover their body with the blanket, and insert the teddy bear under one arm.
  5. Get a friend to help carry the sofa out to a nearby traffic intersection and run back to the office, fast.
  6. While bystanders figure out what happened, and call the police, you have at least an hour to get your work done.
  7. Log onto the primary site server and stop all ConfigMgr services (you can do this using the PREINST.exe /STOPSITE method, or just go into Services.msc and shut them all down.
  8. Stop SQL Server services on that host as well.
  9. Open AD Users and Computers, browse to the System Management container and delete everything inside of it, including sub containers, attributes, small animals and one-armed circus performers.  But, under no circumstances, should you EVER delete the cup of coffee.
  10. If the SCCM host is/was virtual, shut down the VM.  If it was physical, shut it down too.  Notice the difference?
  11. Delete the AD account for the SCCM host.
  12. Reinstall the VM and give it a new name (just to be safe)
  13. Reconfigure the Windows Server properties, network settings, etc.
  14. Reinstall all of the SCCM prerequisites
  15. Reinstall SCCM and choose a different site code (just to be safe)
  16. Step outside to listen for sirens and bullhorns.  If you hear (or see) anything like police or EMS near the sofa, run over and say “Oh my God.  Jimmy?  I’m sorry officers.  Jimmy keeps doing this.  We’ll take care of him, no worries.”  Then call your friend back to help carry the sofa over to the shoulder and run away again.
  17. If there are no signs of public safety personnel nearby, continue rolling out your site.  Then re-deploy clients.
  18. Relax and enjoy.

Piece of cake.  Jimmy might need a few Tylenol when he wakes up though.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s