locked
Distribution Point Best Practice RRS feed

  • Question

  • Hello,

    I have a question about distribution point placement. Our network is technically a MAN consisting of about 13 locations. One location contains our site server and our "nerve center" where all the servers reside. 3 location have 1G connections to the central site; the rest have 100M connections. To main consistency for deployments, I plan to install distribution points at all of the 100M locations. Here are my general requirements/goals:

    • Each location uses it's local distribution point for content. If the content does not exist on that distribution point, it should look for content at the central location.
    • 100M locations should NOT look for content from other 100M locations.
    • The 1G locations should look for content at the central location.

    At this point, I have boundaries defined for each subnet (each location). The 1G location boundaries are in a "1G" boundary group and all the 100M locations are in a "100M" boundary group. The same is true for the distribution points and distribution point groups, respectively.

    The issue I'm having now is that when a client in location "100M-A" is looking for content (which currently does not have a local distribution point), it immediately goes to location "100M-B" instead of "1G". Any suggestions? There is the potential that I'm way off base in what distribution points/boundaries do. Thanks!

    Wednesday, June 19, 2013 7:05 PM

Answers

  • Hi,
    It is explained here: http://technet.microsoft.com/en-us/library/gg712321.aspx , the protected distribution Point setting is replaced by prefered distribution Point and the allow fallback setting.

    Allow fallback source location for content: This distribution point property enables clients to fall back and use the distribution point as the source location for content when the content is not available on a preferred distribution point.

    Regards,
    Jörgen


    -- My System Center blog ccmexec.com -- Twitter @ccmexec

    • Marked as answer by Kevin Surgener Thursday, June 20, 2013 5:45 PM
    • Unmarked as answer by Kevin Surgener Thursday, June 20, 2013 5:46 PM
    • Proposed as answer by Garth JonesMVP Sunday, February 15, 2015 11:50 PM
    • Marked as answer by Garth JonesMVP Saturday, November 14, 2015 4:18 PM
    Wednesday, June 19, 2013 7:45 PM

All replies

  • Hi,
    It is explained here: http://technet.microsoft.com/en-us/library/gg712321.aspx , the protected distribution Point setting is replaced by prefered distribution Point and the allow fallback setting.

    Allow fallback source location for content: This distribution point property enables clients to fall back and use the distribution point as the source location for content when the content is not available on a preferred distribution point.

    Regards,
    Jörgen


    -- My System Center blog ccmexec.com -- Twitter @ccmexec

    • Marked as answer by Kevin Surgener Thursday, June 20, 2013 5:45 PM
    • Unmarked as answer by Kevin Surgener Thursday, June 20, 2013 5:46 PM
    • Proposed as answer by Garth JonesMVP Sunday, February 15, 2015 11:50 PM
    • Marked as answer by Garth JonesMVP Saturday, November 14, 2015 4:18 PM
    Wednesday, June 19, 2013 7:45 PM
  • Jörgen,

    Thank you for the link. There is definitely good information there. I think my problem is that I'm assigning the distribution points to the boundary. If I'm ready correctly, in this configuration, those distribution points become preferred. I do have a question though. If I took out all the distribution points off of the boundary groups and installed local DPs at each location, would the clients use the local DP first then fall back to the DP at my central location? Thanks!

    Wednesday, June 19, 2013 7:55 PM