1610 boundary group discussion / replacing slow link issue

[sumyun]

I haven't seen much discussion about the new boundary group changes in 1610. I am still testing it in the lab environment, but things aren't happen to what I am expecting.

So after upgrade from 1602 to 1610, boundary group with "slow connection" DP before, is automatically copied to a new boundary group, as described below.

 

https://docs.microsoft.com/en-us/sccm/core/servers/deploy/configure/define-site-boundaries-and-boundary-groups#a-namebkmkboundarygroupsa-boundary-group/

• This boundary group copy does not have boundaries associated with it. However, A fallback link is created between the original group and the new boundary group copy that has the fallback time set to zero.

However, my test clients under that specific boundary group are not able to receive any content from SCCM. Including Pre-prod client upgrade, application deployment etc. I went a bit further to upgrade a SCCM client manually under that subnet (boundary group), same result.

 

Keep this in mind, at post ugprade:
Original BoundaryGroup1 has no more DP associates to it, instead, a fallback link created to the newly created BoundaryGroup1-<xxxxxxxxxx>. BoundaryGroup1-<xxxxxxxxxx> contains the DP (used to be a DP set as slow prior 1610).

 

Any experience can be share about this?

 

 

[sumyun]

Found out the problem for application deployment, the deployment type hasn't set to use fallback DP, issue seems to be resolved with application deployment. However, for client upgrade, I am still investigating. I have not promote the latest client to production yet, which I am using a pre-prod collection to test the latest client. It seems like client installation doesn't work on fallback.