One client gets PXE-E11, another does not

[moses]

I have two clients on a network with a storage node in it, with the Locations plugin installed and configured. My Windows DHCP server is pointing to the onsite storage node with the correct boot file.

One client is an EliteBook 8540w. Another client is a Hyper-V VM.

The Hyper-V VM boots to the FOG menu without issues.

The Elitebook 8540w simply gets “PXE-E11 Arp Timeout” and reboots.

How can I resolve this issue?

[george1421]

This sounds like a networking issue with Spanning Tree Protocol. Unless you have the port that goes to the device in port fast or RSTP or Fast STP mode, the port doesn’t switch into the forwarding state fast enough so the pxe rom gives up. Can you check with your network group to ensure that a fast reset mode for spanning tree is enabled on this device port, or for a test just turn off spanning tree on the port that goes to this specific device and see if you get a different result.

Is it safe to say this questionable device, the storage node and the hyper-v server are all on the same subnet?

[Wayne Workman]

@moses said:

The Elitebook 8540w simply gets “PXE-E11 Arp Timeout” and reboots.

To me, that error suggests the elitebook is being told to retrieve a boot file on a remote network.
You can use Wireshark on another computer that is connected to the same network the elitebook is - and do a capture. You’ll catch all the DHCP broadcast messages and you can then see exactly what options this elite book is being given when it tries to network boot.

[moses]

@george1421 Yes, they are all on the same subnet. One thing I tried just now: I changed the DHCP server to point to the main server (connected to that site via MPLS) and both devices work successfully.

[george1421]

@moses said:

@george1421 Yes, they are all on the same subnet. One thing I tried just now: I changed the DHCP server to point to the main server (connected to that site via MPLS) and both devices work successfully.

@Wayne-Workman Isn’t this issue closely related to the issue you had with your remote storage node?

[Wayne Workman]

@george1421 Yes, which is why I think it’s trying to get the file from beyond the subnet. Why would there be ARP errors if it weren’t?

[Sebastian Roth]

We need a wireshark dump… Otherwise it’s just reading in the stars.

[moses]

Oh wow…I originally asked the onsite tech to boot a client to see if he got the Fog boot menu…

…he was rebooting the storage node. Sigh…

Never mind, guys. Thanks anyway.