OK NOT A FIX, BUT IT's WORKING (without CiscoWLC or Aruba AP support)
Just got off the phone with support about 30min ago..
Rebuilt the rebuilt nightly job without success.
Ran through the licensing sync wizard (lucky this instance is online)
Created a new job with a single node (router) which worked
Shifted the load around a bit as requested. Random completion between the pollers.
I only have 180 nodes on this instance, over the 3 engines before I go and load over 5,000 in
There was a comment as to whether I had any WLCs (or another hardware vendor I don't recall)
At present I only have the 1 CiscoWLC and that is using the stock standard CiscoWLC NCM Templete.
Shuffled it around and similar random results - including failing on the core server.
Removed the WLC from being a NCM managed node - NCM completed on 2 of the 3 nodes again including the core which it had just failed on.
Ended the call advising I'll upload the diags with the ticket appropriately.
Whilst about to kick off the 3 server diags, I removed the Aruba APs as they reside on the APE which was failing.
It worked.
Re-ran the job I used to use, yup confirmed working.
So, NCM is working again now albeit with no Wireless config support - Caveat for the network guys to manually back it up or I'll script it later if this case is dragging on.
And for those PMs etc: Case # 946301