Fixing OS deployment… again

I already went through THIS when OSD broke down because of a SCCM 1702 bug fixed in Hotfix Rollup 1, that generated multiple “Unknown computer” entities and the task evaluation didn’t manage to find something valuable.

So again I found myself fighting against a THERE ARE NO TASK SEQUENCES AVAILABLE FOR THIS COMPUTER error.

notaskavail

Why? I wish I knew. I checked every log i know of, none of it was helping.

So I started the Google Machine:

https://blogs.technet.microsoft.com/configurationmgr/2010/04/12/osd-task-sequence-fails-with-there-are-no-task-sequences-available-for-this-computer-if-multiple-machines-have-the-same-smbios-guid/
Nope

https://larslohmann.blogspot.it/2015/06/skipping-task-sequence-because-it-is.html Nopenope

https://blog.endpointfocus.com/configmgr-cb-1702-unknown-computer-support-reports-no-task-sequences-available-computer
YEP THAT’S HIM

I was in the exact same situation.
Searched for the unknown x64 entity ID and searched for it in the systems table

select * from UnknownSystem_DISC
select * from System_DISC where SMS_Unique_Identifier0 like 'YOUR_UID'

query

 

That motherf*§&%£&!

That’s the name of a computer that I deployed hours earlier, for some strange issue it got the same UID as the “Unknown Computer x64” entity

Deleted the machine from SCCM Console and from AD, rerun the PXE boot and…

taskavail
Et voilà!

This kind of thing should have been fixed in 1702 hotfix rollup 1. I don’t know what the f@#* is going on.

As I stated before, I already did this https://www.reddit.com/r/SCCM/comments/69ghqk/sccm_1702_osd_issue/dhr9nfv/ to fix the issue about duplicate unknown computer entities EVEN AFTER APPLYING THE 1702 HOTFIX ROLLUP 1 which didn’t delete the unknown entity duplicates automatically, but this one was completely new for me.

EDIT: this didn’t stop it! I ended up having multiple machines with the same ID in the database, despite i deleted them from the console.

A new computer (the same I tried to deploy previously) in System_DISC appeared after i launched a task sequence on it.

query2

I’m gonna try to delete the entries from the database as explained HERE, let’s see if this sh*# goes away. I wonder if this happens because usually I copy task sequences from an existing one when I have a new model to deploy, but it seems unlikely.

DELETE from System_DISC where SMS_Unique_Identifier0 like 'YOUR_ID'

And let’s relaunch a task sequence.

This time the notebook deployed succesfully and the UID is something unique, so I’m gonna assume everything is alright.

The client registered with a normal GUID this time, but yet i don’t understand how all this s@§£ could happen

query3

I will proceed in further testing keeping my eyes open

Leave a comment