locked
Noob Question on setup RRS feed

  • Question

  • I have recently been tasked with standing up SCCM 2012 in an environment with nothing set up at all.  We use Amazon Web Services, have two buildings, and we have about 250 end users and just under a 100 servers.  We have also been directed to stand up two SCCM 2012 servers - one in each building.  I have signed up for a Deploying SCCM 12 class, but they want us to play in the sandbox until then.

    My question is hopefully simple for someone who knows.    Should I be setting up a CAS and a Primary or 2 Primary sites?  

    I am reading many things online and cannot seem to grasp what the heck i need to stand up.  Any thoughts appreciated at this time.  


    Matt Dillon

    Tuesday, August 2, 2016 1:22 PM

Answers

  • Yes, that should work assuming all of the pre-reqs (explicitly and implicit) are in place like proper DNS resolution and connectivity. Not sure how pricing for AWS works, but this setup could get expensive quite quickly though since a lot of data will be leaving AWS on a pretty much constant basis.

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

    • Marked as answer by pugmohone Tuesday, August 2, 2016 3:01 PM
    Tuesday, August 2, 2016 2:52 PM

All replies

  • Why two sites at all? That's overkill to the max - especially for such a low client count. A single standalone Primary can handle that perfectly fine. How are those building connected LAN/WAN-wise?

    Torsten Meringer | http://www.mssccmfaq.de

    Tuesday, August 2, 2016 1:45 PM
  • That is what I was thinking... Let me find out exactly why they want that setup and get back to you.  Thanks.  

    Matt Dillon

    Tuesday, August 2, 2016 1:47 PM
  • Yep. No CAS. A CAS is only for a hierarchy consisting of at least two primary sites and the only reason (at least for 99.9% of all scenarios) to have more than one primary site is to support 150,000+ managed systems.

    What you are probably looking at is a single primary site with the primary site server in one of these buildings and a DP in the other.

    If the motivation is for high availability, then the first thing to know if the true HA is not completely achievable at an application level with ConfigMgr. If that is your requirement (complete HA), then you'll have to use something like Hyper-V LiveMotion and host the site server as a virtual guest on Hyper-V.

    If you want application level HA, you can get it by it having three total servers (still a single primary site):

    - Primary site server (in building A)

    - Site system [server] hosting MP, DP, SUP (in building A)

    - Site system [server] hosting MP, DP, SUP (in building B) -- basically identical to the previous system

    If you are only concerned about bandwidth between the two buildings, then you'd go down another route -- I don't think that's the issue here so I won't complicate things by mentioned that route.


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

    Tuesday, August 2, 2016 2:15 PM
  • Looks like, and correct me if this won't work, they want a Primary site with SQL Standard installed in the AWS environment and a Distribution Point in both buildings built onsite virtually.  I think that outta cover it.  

    Matt Dillon

    Tuesday, August 2, 2016 2:45 PM
  • Yes, that should work assuming all of the pre-reqs (explicitly and implicit) are in place like proper DNS resolution and connectivity. Not sure how pricing for AWS works, but this setup could get expensive quite quickly though since a lot of data will be leaving AWS on a pretty much constant basis.

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

    • Marked as answer by pugmohone Tuesday, August 2, 2016 3:01 PM
    Tuesday, August 2, 2016 2:52 PM