"SuperGumby [SBS MVP]" <> wrote in message
news:...
> Let's keep it simple with two sites, HQ and Location1 (L1).
>
> WSUS is installed on HQWSUS.our.lan and also on L1WSUS.our lan with L1WSUS
> being a replica of HQWSUS. A DNS entry exists at each location pointing
> LOCALWSUS to the local WSUS at each location and group policy tells all
> PC's to use LOCALWSUS.
>
> The user of PC37 is at HQ and connects for less time than a download
> takes, unplugs the PC and takes it to L1. It would not surprise me to find
> the PC 'resuming' the download but it may start from beginning again, no
> real problem either way.
This is where understanding the functionality of BITS is important, as well
as the benefits of using Range Protocol Headers in HTTP v1.1. Suffice it to
say that BITS is capable of resuming the download from where it's left off.
In fact, this is no different a solution than if this were a desktop PC and
simply powered off whilst a download were in progress -- a scenario that
happens a lot more often than is likely thought of.
> Problem is we now have PC37 registered on both servers (even if it wasn't
> partway through a download).
Actaully PC37 will only *register* if an actual detection takes place. The
download, which is a pure HTTP-based file transfer from
http://LocalWSUS/Content/* will not cause a registration with the WSUS
Server to take place; that's done by a call to the webservice
http://LocalWSUS/ClientWebService/* -- but your point is, nonetheless,
valid. As some point PC37 *will* become registered with both machines, and
this is a management issue that must be worked out.
> Does the WSUS 'uniqueID' given each workstation allow such movement to be
> tracked between cascaded servers?
If you're doing reporting rollup from the replica server, presumably all of
the "events" recorded against the SusClientID of PC37 will be rolled up, and
the upstream server will have a full record of activity against this client.
The consideration here is that the replica server(s) will *not* have a
complete record of activity against this client.
> Is the entry on HQWSUS deleted or moved so that it is recognised as
> 'belonging' to L1WSUS? or do we end up with different status of the
> machine (and hence in reports) on the 2 servers?
The whole key is the GUID, known as the SusClientID, combined with (added in
WSUS v3), the FQDN of the machine.
The other consideration, even more critical, is understanding that the TTL
on the DNS records must be abnormally short in a scenario where there is
high mobility, in order to prevent the client machine from accessing the
"last accessed" server across the WAN, because of cached DNS 'A' records,
pointing to the wrong (not local) WSUS Server.
--
Lawrence Garvin, M.S., MCITP(x2), MCTS(x5), MCP(x7), MCBMSP
Principal/CTO, Onsite Technology Solutions, Houston, Texas
Microsoft MVP - Software Distribution (2005-2009)
MS WSUS Website:
http://www.microsoft.com/wsus
My Websites:
http://www.onsitechsolutions.com;
http://wsusinfo.onsitechsolutions.com
My MVP Profile:
http://mvp.support.microsoft.com/pro...awrence.Garvin