After reading up on SCCM and Software Update Points, I understand that if a client fails to connect to a SUP (4 times w/in 30 minutes) then they will 'failover' to an alternate SUP.

In this case the Alternate SUP is our Primary which is over a slow-link/connection (Which is why we have secondaries/remote SUP setup in the first place)

From what I understand is that it will stay on the Alternate SUP until something happens to that one, and it would then failback over to the existing (correct/local) SUP at the location.

Is there any way to change the client to NOT fail over?  In other words - if the local Secondary SUP goes down...  Just keep trying forver and NOT reach out over the WAN to an alternate SUP?

Thanks

Hello,

Client sync metadata from SUP, and download content from DP which will run out your bandwidth. So setup a remote DP should be more helpful.

maybe this:
http://blogs.technet.com/b/configmgrteam/archive/2013/03/27/group-policy-preferences-and-software-updates-in-cm2012sp1.aspx

No, there's nothing in the product to block this. You can block it at a network or DNS level though.

Thanks Jason..

What if... After patching this month, I stopped the wSUS pool in the primary.. In theory - all of the clients currently talking to that server for SUP, would/should revert back to their Secondary?  When most of the clients are back, I would then turn the WSUSpool back on.

Would that work?

It should in theory. The only problem is that the failover only happens for a subset of error codes received by the client. I'm not positive if that will produce the error code or not but you can easily test that.