Issue
PXE Boot fails with a "TFTP TIMEOUT" error. This can also present as the following errors:
- PXE T04
- PXE-E36
- PXE-E32
- PXE Error M0F
Cause
- The most likely cause is that the LANDESK PXE MTFTP service is not running or needs to be restarted on the PXE Representative:
- VPN software has changed the MTU setting in the registry on the PXE Representative causing TFTP to fail.
- A network device in the network is dropping or malforming network packets.
- There may not be enough PXE representatives for the incoming network boot traffic. Add more PXE representatives per subnet.
- If this is happening on only one device or one model of device, it could be related to NIC or BIOS firmware
Resolution for TFTP Service not started or needs to be restarted
1. Open the Services applet on the PXE Representative (through Control Panel or type "services.msc" from the Run line"
2. Check the status of the LANDESK PXE TFTP service. If it is not running, start it, if it is already running, restart it.
Resolution for PXE Representative MTU size
On the PXE Representative, do the following:
- Open Regedit
- Navigate to HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\Interfaces
- Search for "MTU" and select "Match whole string only".
- There will be an MTU value for each network interface
- Verify that the value is set to 1500
- The PXE Representative may need to be rebooted
You can also confirm the setting by running the command:
netsh interface ipv4 show subinterfaces
Value at 1500 in Registry for both screenshots but command shows the values are different for the LAN interface.
If you see MTU show a different number than 1500 you will need to run the following command:
netsh interface ipv4 set subinterface “Local Area Connection” mtu=1500 store=persistent
Other Considerations
- What are the specifications of the computer that is hosting the PXE rep?
- What Operating System is the PXE rep running?
- Is this server hosting any other connections?
- How many total network connections does the PXE rep show as active? It's possible the connection limit could be full.
- What is the memory and CPU utilization? Is it getting bogged down by something?
- Does he have any sort of security software on the server that may be analyzing incoming connections and thus slowing down the requests?
- Are there other boot servers or software that may be handling boot requests and causing a mix-up?
- Are the routers configured to forward all the traffic correctly? Are all of the same clients served by one router?
- MTFTP runs on port 69. Is there anything that could be interfering with that?
- What type of hardware are the clients? It is always possible that this is an issue with a particular model or a few various models from a certain vendor.
- Are there clients that work or not completely random?
- Is the BIOS current?
- Is there a firmware update for the NIC available that isn't applied?
- Search for similar issues with that/those models of computers?
- There have been reports that this patch (http://support.microsoft.com/kb/953230) can cause issues with PXE and TFTP.
- Does the PXE rep have multiple network cards? If so, is the bind order correct on them so that the ethernet NIC is first? If they are teamed try breaking the team and setting only one card as active and check the binding.
- Possibly the network card driver on the PXE rep needs updating.
- Possibly the network card driver on the clients needs updating.
- This article may be helpful in understanding DHCP Options relating to PXE: http://www.experts-exchange.com/Networking/Misc/A_2978-PXEClient-what-is-it-for-Can-I-use-PXE-without-it.html
- Is there anything in the event viewer on the PXE rep that could be a clue? Check system and application log.
- Is there anything that could be causing an IP address conflict with the PXE server?
More information on PXE boot errors:
PXE Boot errors and descriptions.
How to troubleshoot LANDESK PXE boot: