We are having an issue during OSD

[rthorn8]

We have 2 servers - One in our corporate Office, Server02, and the other across the WAN in our west coast office Server01.  It takes a few minutes, around 5, for the Task Sequence wizard to show up when imaging a computer in our west coast office. While the computer being imaged is at the (blank) configuration Manager Screen if I go in to CMD and look at the SMSTS.log File I see request going to Server02 with errors before it tries server01 and actually brings of the task sequence wizard. Once it stops trying to connect over the wan to server02 the task sequence wizard does show up and everything works after that.

The computer (s) being imaged are server01 are on the same network/vlan.  So why is the computer even trying requesting from Server02 before  server01?  Boundary groups are set correctly .

Server 01 is our primary server.

I see this in the smstslog a few times on the computer being imaged.

Current time info: .                                         

Getting MP time information                      

Requesting client identity                            

Setting the authenticator.

CLibSMSMessageWinHttpTransport::Send: WinHttpOpenRequest - URL: SERVER02.stella-jones.com:443  CCM_POST /ccm_system_AltAuth/request

SSL, using authenticator in request.

In SSL, but with no client cert.

Error. Received 0x80072ee2 from WinHttpSendRequest.

connect (sock, (struct sockaddr *) &SockAddrIn, sizeof (struct sockaddr_in)) == 0, HRESULT=8007274c (C:\__w\1\s\src\Framework\OSDMessaging\libsmsmessaging.cpp,944)

socket 'connect' failed; 8007274c

hr, HRESULT=80072ee2 (C:\__w\1\s\src\Framework\OSDMessaging\libsmsmessaging.cpp,10293)             

Sending with winhttp failed; 80072ee2

Will retry in 4 second(s)               

Retrying...                                                                                                                                                                                                                                                                                                                                                          

Then after a few retries I see the exact same request to Server01 that succeeds.

Current time info: .                                                                                                                                                                                                         

Getting MP time information

Requesting client identity

Setting the authenticator.

CLibSMSMessageWinHttpTransport::Send: WinHttpOpenRequest - URL: SERVER01.stella-jones.com:443  CCM_POST /ccm_system_AltAuth/request 

SSL, using authenticator in request.         

In SSL, but with no client cert.    

Request was successful.

Everything works after the request to server01 succeeds

[#2]

The issue you are experiencing with the Task Sequence wizard taking a few minutes to show up and attempting to connect to Server02 before Server01 could be due to the client's initial management point (MP) being set to Server02. This can cause the client to first attempt to communicate with Server02, resulting in delays and errors before it retries and successfully connects to Server01.

The "socket 'connect' failed; 8007274c" error in the smsts.log indicates a failure to connect to Server02, which triggers a retry and eventually leads to a successful connection to Server01. This behavior is likely because the client is initially configured to use Server02 as its MP.

To resolve this issue, you may need to ensure that the client's initial MP is set to Server01. This can be done by checking the client's assigned site and boundary group configurations in Configuration Manager. Additionally, reviewing the boundary group relationships and the boundaries associated with Server01 and the west coast office clients is important to ensure that the clients are correctly assigned to the appropriate boundary group.

By verifying and adjusting the client's site assignment, boundary group configurations, and boundary relationships, you can help ensure that the client communicates with the nearest and most appropriate management point, which in this case is Server01, and reduce the delay in displaying the Task Sequence wizard.

@rthorn8 

[#3]

@rthorn8Are you still having this issue?

I'm administering a CM 2309 environment and we randomly started seeing the same exact errors.

Boundary groups are configured correctly and haven't changed recently. The PXE Distribution points are also have the 'preferred MP' set to the respective MP for that site/boundary group.

However, the clients seem to be picking random MP's to download their policies. This sometimes causes delays when retrieving the task sequences that are available.  

[#4]

@Tyler_StoneHi

I am seeing the exact same thing in CM2309, PXE gives in SSL, but no client cert, and it takes 4 to 5 minutes to get to the select a task sequence window, and then everything is normal speed wise during the osd.

Did you ever resolve this?