I’m trying to use gPxe in our lab environment and running into some issues.


I was getting a DHCP connection timeout error as discussed in the following thread:



After following Thomas’s instructions (building gPxe from source and setting a sleep time) I got over that issue.

I’d to make the changes as highlighted on Pxe chainloading (Etherboot) to break the infinite loop: http://etherboot.org/wiki/pxechaining  (Using windows DHCP server).


Here’s my exact setup:

1.       Built gPxe from source with a sleep time and HTTPS support.

2.       Copied over the undionly.kpxe files to my TFTP server.

3.       Set up DHCP server (Windows) with pxe chainloading to avoid the infinite loop as discussed on http://etherboot.org/wiki/pxechaining.

4.       My DHCP server points to the tramp file which has


set filename https://<server IP>/vmw/rbd/trampchain https://<server IP>/vmw/rbd/tramp

5.       The file at the above url has the following:


set retry-delay 20

post /vmw/rbd/host-register?bootmac=${mac}

6.       Now when my server boots up it gets the DHCP IP during gPxe (which was failing earlier and was resolved by adding the sleep in gPxe).

Then the server displays: Booting from filename “tramp”.

Transfers the file from TFTP server.

Then it tries to execute the post command above and fails with Command not found.

Also get an error : Could not fetch https://<server IP>/vmw/rbd/tramp: Not an executable image.


I’m pretty sure in trying to get this to work I have made some mistakes in my setup. Any help in resolving the above would be really appreciated.


Thanks & regards,