locked
Some Windows 10 systems not receiving SCCM client push and some not even being detected by AD System Discovery RRS feed

  • Question

  • If it shows that the SCCM client is NOT installed, then why does it show that there is a version in the "Client Version" column?

    Wednesday, April 26, 2017 10:13 PM

Answers

  • I'm not saying that you shouldn't necessarily delete the resource, I'm saying that if you are deleting the resource, then resetting the client install flag is irrelevant and should be disabled or at least increased from the 2 days you have it configured to address the main issue of this thread.

    Jason | http://blog.configmgrftw.com | @jasonsandys

    • Proposed as answer by Frank Dong Wednesday, May 10, 2017 10:19 AM
    • Marked as answer by I AM Sir Ask Alot Wednesday, May 24, 2017 4:12 AM
    Friday, April 28, 2017 12:48 PM

All replies

  • Was there a client previously installed, and now its not active? Can you check the client logs on one of them? (if they exist) like ccmsetup.log
    Wednesday, April 26, 2017 10:49 PM
  • The Clear Install Flag maintenance task is the only built in process that will change the client from Yes to No. Check to see if you have this site maintenance task enabled and then validate that you do actually want it enabled as it is not enabled by default.

    Jason | http://blog.configmgrftw.com | @jasonsandys

    Thursday, April 27, 2017 2:24 AM
  • From what the SCCM console shows, it does appear that some of the systems did in fact have the SCCM client installed at one point. Most of these systems are offline now, but will be online tomorrow, so I can check the logs then. 

    As for the Clear Install Flag Maint. Task, I do have this enabled and here is how I have configured it:

    Here is how I have Heartbeat Discovery configured:

    Thursday, April 27, 2017 5:08 AM
  • That means that if a system doesn't send a heartbeat within 2 days every 2 days, it will be marked as client No. Is that really what you want?

    Jason | http://blog.configmgrftw.com | @jasonsandys

    Thursday, April 27, 2017 1:10 PM
  • Not sure this is what I want. 

    I am just playing with it right now trying to get it exactly how I want. 

    I don't think the default was working for me too well, because when systems needed to be reimaged, they were just sitting on the bench and never received the client push, so I decreased this to speed that process up. 

    Is there a better way to address this?

    Thanks

    Thursday, April 27, 2017 7:47 PM
  • Not following. What does setting client = NO have to do with your imaging process/flow?

    Jason | http://blog.configmgrftw.com | @jasonsandys

    Thursday, April 27, 2017 8:14 PM
  • Okay, so this is what the process that my employer follows regarding re-imaging:

    • Gets computer from user
    • Technician removes computer object from AD and removes it from SCCM
    • Technician re-images it using Acronis (we do not use OSD here yet)
    • Technician renames the computer to the same name it was before
    • Technician delivers it back to user

    I only use Auto Site Wide Client Push here to get the SCCM client installed on systems. 

    Technicians that re-image systems have notified me stating that they have to wait for a long period of time before they see that the SCCM client gets reinstalled on the newly re-imaged systems. 

    So, to address this, I configured the Heartbeat Discover to 1 day and the Clear Install Flag to 2 days. 

    My thinking is that by decreasing these values for both Heartbeat Discovery and Clear Install Flag, the Site Wide Client Push should deploy the SCCM client to these newly re-imaged systems much faster. 

    However, SCCM sometimes does not pick up the computer object in AD, because I believe it is due to having the same name is it did before, so the Deltas just don't grab it. I usually have to run a Full AD System Discovery in order for these systems, then let Site Wide Client Push do its thing. 

    Maybe I am not on par with how this really works and I a bit confused, due to the fact that this is SO VERY CONVOLUTED AND IT SHOULDN'T BE!

    It's so frustrating. 

    HELP!!

    Thursday, April 27, 2017 11:45 PM
  • Why not include the ConfigMgr agent as part of the deployment process either in the image or part of their manual steps?

    Also, if the techs are deleting the resource in ConfigMgr as their first step then having ConfigMgr clear the install flag plays no part in the process because the resource doesn't exist anymore.


    Jason | http://blog.configmgrftw.com | @jasonsandys

    Friday, April 28, 2017 3:21 AM
  • Well, they do manually install the client, but I wanted a more automated way of doing this. 

    Also, if they don't delete the record from SCCM and a new computer object of the exact same name is recreated in AD, won't this cause some sort of problems?

    Or can the computer record simply stay in SCCM and it will somehow be re-linked to the newly created AD computer object of the same name?

    Friday, April 28, 2017 7:09 AM
  • I'm not saying that you shouldn't necessarily delete the resource, I'm saying that if you are deleting the resource, then resetting the client install flag is irrelevant and should be disabled or at least increased from the 2 days you have it configured to address the main issue of this thread.

    Jason | http://blog.configmgrftw.com | @jasonsandys

    • Proposed as answer by Frank Dong Wednesday, May 10, 2017 10:19 AM
    • Marked as answer by I AM Sir Ask Alot Wednesday, May 24, 2017 4:12 AM
    Friday, April 28, 2017 12:48 PM
  • I configured it to run every 14 days. 

    Thanks

    Wednesday, May 24, 2017 4:16 AM