Clients have stopped being able to PXE Boot - failed to receive response with winhttp; 80072ee2

Question

_{Thursday, November 12, 2015 10:31 AM}

Hi,

We have a remote site with a 2012 R2 DP configured for PXE Boot which has been working fine for a long time.

Clients are now failing to PXE Boot and the smspxe.log is showing the following error when a client tries to PXE boot.

failed to receive response with winhttp; 80072ee2  SMSPXE  12/11/2015 10:04:16 2348 (0x092C)
End of retries  SMSPXE  12/11/2015 10:04:16 2348 (0x092C)
failed to send the request  SMSPXE  12/11/2015 10:04:16 2348 (0x092C)
PXE::MP_LookupDevice failed; 0x80072ee2 SMSPXE  12/11/2015 10:04:16 2348 (0x092C)
PXE Provider failed to process message. 
Unknown error (Error: 80072EE2; Source: Unknown)    SMSPXE  12/11/2015 10:04:16 2348 (0x092C)
[MAC:ADDRESS], F3B875FF-1083-11E4-BA01-B15A940000FF: Not serviced.  SMSPXE  12/11/2015 10:04:16 2348 (0x092C)

I've been trying to find more info on the winhttp 80072ee2 error but not having much like finding anything thats helped.

So far we have tried removing PXE from the DP, allowing WDS to uninstall, rebooted and then reinstalled PXE and this hasn't helped.

Any suggestions on this error or anywhere else to check for errors?

Thanks.

All replies (8)

_{Friday, November 13, 2015 2:29 PM ✅Answered | 1 vote}

Hi Folks,

Thanks for all the replies. I've managed to sort this now.

Even though all the logs were clear of errors, content was distributing successfully out to the affected DP, state messages etc were being returned from that site to the MP there was obviously something wrong with the passthrough of PXE requests back to the MP. I restarted the MP and the problem resolved immediately.

I had been reluctant to do that as the logs weren't indicating any problems (apart from SMSPXE.log on the DP) so it seems as only the PXE requests were affected.

Thanks again for the suggestions.

_{Thursday, November 12, 2015 10:34 AM}

Is the MP working fine? Plus you can use cmtrace.exe to translate the error code

Torsten Meringer | http://www.mssccmfaq.de

_{Thursday, November 12, 2015 10:38 AM}

Hi Torsten,

Yes the MP seems to be working ok, I can see clients at other sites building ok, just this one is misbehaving.

I already ran the error code through the error code lookup in cmtrace, it came back with this:

The operation timed outSource: Winhttp

Unfortunately I am not sure where to continue tracing that error. I have checked IIS and it seems ok on that server and no errors showing in the logs for IIS.

_{Thursday, November 12, 2015 11:06 AM}

First note that this has nothing to do with PXE. Once WinPE starts and you can get ssts.log, PXE is done.

80072ee2 = "The operation timed out"

This can can be caused by many different things -- all of them external to ConfigMgr. There also is no way for ConfigMgr to know why the query that it is sending is not replied to. You need to check the entire "path" to the MP and back including all networking components. Often, network filters are to blame for this type of error but as mentioned, there's no way ConfigMgr can know this so you need to trace your traffic and everything that may touch it to determine what's going on. You can start by checking the IIS logs on the MP to see if the traffic is even making it to the MP.

Jason | http://blog.configmgrftw.com | @jasonsandys

_{Thursday, November 12, 2015 11:25 AM | 1 vote}

Should have mentioned, its not getting as far as WinPE.

Its just sitting at the PXE - 53: No Boot Filename Recieved.

I'll take a look at the IIS logs on the MP as well for the same time period.

_{Thursday, November 12, 2015 12:51 PM}

Have you checked to see what other services on the site system server hosting the PXE role are running properly? I've experienced this error code when the IIS service on the MP stopped processing requests when its logs were filled.

Are clients on the same network segment as the PXE server able to PXE boot successfully?

Please remember to: "Vote" if my response was helpful. "Mark As Answered" if I answered your question correctly. V/R, Darrick West - Senior Systems Engineer, ConfigMgr: OSD

_{Thursday, November 12, 2015 8:40 PM}

Yeah, sorry, didn't read your post well enough. It's the same answer though just between the DP and MP instead of the client and MP.

Jason | http://blog.configmgrftw.com | @jasonsandys

_{Friday, November 13, 2015 12:04 AM}

Has there been changes to DHCP or networking recently?