We have 1 primary site server and 7 2008 r2 servers that act as distribution points. In all but one of our locations when we build a new OS from CM2012 they complete in about 60 min. But in 1 site when we build a new OS they take 120-150 minutes. In this location when I F12 the system it says at the bottom of the screen it is connecting to the correct IP address. This IP is the DP in the same physical building. All the servers are the same hardware, same RAM, save hard-drives etc...

Where on this suspect DP are the log files I need to look at?  C:\windows\ccm\logs?   What logs files do I look at?  What could cause this?  It is almost like this suspect site is going all the way back to our Primary site server to get its files.   how do you confirm where an OS is actually pulling its files?

Thanks.&

This sounds like it could be a boundaries / boundary group issue.

Probably wise to check that the IP addresses or AD Site is correctly allocated to a Boundary Group and that a content location for the boundary group is set.

It's possible that you are PXE booting from the correct DP, but then the client is connecting back to a remote DP or your primary site to get the content.

One way of checking would be to check the logs on the client and see where it is downloading from. Assuming you have command line testing turned on, you should be able to hit F8 during the OSD and check the SMSTS.log file to see where it is downloading the content from.

You can also use the Content Validation feature from the console to verify that the packages on the remote DP are not corrupt.

On the suspect DP check log files in C:\inetpub\logs\LogFiles\W3SVC1

You will see IP addresses of machines that you are imaging

On the target machine check DataTransferService.log - the source DP will be shown there.

ok, it is not boundaries, that was checked. On the suspect DP I did look at the inetpub log files and I see hundreds of references to my workstation that took 2.5 hours to build.     My workstation has an ip of 10.10.241.44 and I see that over and over and over.

I then looked at the DataTransferService on my workstation and this is where I get confused.   I searched for the suspect DP and I do find it.

UpdateURLWithTransportSettings(): OLD URL - http://SCCMDP1.acme.org/SMS_DP_SMSPKG$/Content_cdd0d1d2-ba64-4dd3-b81f-8e4e78ae8a75.1 DataTransferService 3/24/2015 12:50:34 PM 4032 (0x0FC0)
UpdateURLWithTransportSettings(): NEW URL - http://SCCMDP1.acme.org:80/SMS_DP_SMSPKG$/Content_cdd0d1d2-ba64-4dd3-b81f-8e4e78ae8a75.1 DataTransferService 3/24/2015 12:50:34 PM 4032 (0x0FC0)

But then I will see the same thing for other DPs.    so how do I know which DP it is really using?   btw...the time listed above is the actual time I was building this machine yesterday.   12:50:34 PM

Could this just be a network issue?   If it is CM2012 being mis-configured where else should I look and how do I prove it?

So it could be a network issue, what speed is the network interface set to on the problem server? What is the port speed on the switches? What speed is the nic on the client set to?

I've come across 10Mbps switches causing slowness before, and one organisation that set all switch ports to 10Mbps even though it was capable of 1Gbps!