1

We've had an issue that has persisted for about 10 days now, which has gone unnoticed by myself because we haven't imaged a whole lot of PCs lately.

Essentially, we get the "E53 - No boot filename received" error when trying to boot clients. Wireshark shows that the distribution points receive BOOTP traffic, so I am inclined to think that this is not a network issue (and nothing else has changed)

Moving on - both our WDS enabled hosts that we have seem to be functioning fine. No changes have been made, there's no errors on the servers and the event logs show that PXE services start up fine, although two days ago, both have had the following entries spam up 8 times at once in their WDS logs every 15-25 minutes:

An endpoint was opened by provider WDSTFTP.

followed by another 8 of

An endpoint was closed by provider WDSTFTP.

Type: UDP

Address: (random UDP ports)

I then tried to look at the PXE logs on the distribution points, but they only go back a day. I'll post those in a second, but here's what we usually see in smsdpusage.log when everything works OK:

  • Unable to open Registry key Software\Microsoft\CCM. Return Code [80070002]. Client HTTPS state is Unknown.
  • Gathering statistics from C:\inetpub\logs\LogFiles\W3SVC1\ex170410.log
  • Report state message 0x00000000 to MP
  • Report Body: ID="0" Type="903" IDType="0"/> Criticality="0"/> Count="4">41635515797198720 SMS_Distribution_Point_Monitoring

What we started seeing 10 days ago, was the following:

  • Unable to open Registry key Software\Microsoft\CCM. Return Code [80070002]. Client HTTPS state is Unknown. Gathering statistics from C:\inetpub\logs\LogFiles\W3SVC1\u_ex170415.log
  • Report state message 0x00000003 to MP
  • Report Body: 0000
  • reply has no message header marker
  • Error sending DP state message to site server from remote DP http://CM12-DP.domain. Code 0x80004005

I have no idea what the code of 3 means, but here's what our SMSPXE.log looks like continually, repeated over and over again..

  • RequestMPKeyInformation: Send() failed.
  • Failed to get information for MP: http://CM12-MP.domain. 80004005.
  • reply has no message header marker
  • PXE::MP_LookupDevice failed; 0x80004005
  • RequestMPKeyInformation: Send() failed.
  • Failed to get information for MP: http://CM12-MP.domain. 80004005.
  • reply has no message header marker
  • Failed to send status message (80004005)
  • Failed to send the status message
  • PXE::MP_ReportStatus failed; 0x80004005
  • PXE Provider failed to process message.
  • Unspecified error (Error: 80004005; Source: Windows)
  • D4:3D:7E:6B:0E:9B, 00000000-0000-0000-0000-D43D7E6B0E9B: Not serviced.

We've tried restarting services and systems but no luck so far. My predecessor set our SCCM site up before I arrived, so although I am happy to try and troubleshoot this more myself, I could spends hours tangling myself up in this with no real direction of where I should go to next!

Any help is very much appreciated!

Cheers,

Oliver

  • That issue is actually a DHCP issue. Make sure that the primary DHCP server has a proxy DHCP server set up and pointing to your WDS server. Check your WDS server's properties to make sure that it is configured properly. (WDS is actually a completely separate DHCP server and TFTP server, that is why proxy dhcp is set up and not the pxe file setting in the main DHCP server). – Elliot Labs Apr 27 '17 at 12:10
  • Cheers for the reply and this was actually my first thought; so, we're not using the DHCP options for TFTP boot server magic, its all done at the switch level with IP helpers. We've already checked the switch configs and it looks fine - but ultimately, we get the traffic come through to the DPs so I am fairly sure the network side is ok. FInally, our issues came about around the time that the management point crashed (the VM ran out of space and couldn't write anymore, despite Windows telling it it could). – Trevelyan Apr 28 '17 at 12:54
  • @Trevelyan Please don't add solved in your question ... post your solution as an answer and accept it. only with an accepted answer the question will be marked as solved in the system. – Gerald Schneider May 4 '17 at 10:07
0

Looks like the SCCM installation was a bit broken. CCMExec.exe was corrupted and even the site reset wasn't helping. I restored a backup of the installation disk from a few months ago and all works fine now

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.