MAAS

  1. Bug #1707999
  2. Comment #28

Comment 28 for bug 1707999

Revision history for this message
Jason Hobbs (jason-hobbs) wrote : Re: pod VM fails to PXE boot after receiving multiple DHCP offers from both primary and secondary rack controllers, for different IPs

Ah, we're using ipxe-qemu 1.0.0+git-20150424.a25a16d-1ubuntu1, which is from ipxe source in april of 2015. They've changed the way the secs field is computed since then - in current source it's '0' for the first request; in the version we're using it's '4' for the first request.

https://github.com/ipxe/ipxe/blob/a25a16d4adf663abd8f6eaab266b9444157357b6/src/net/udp/dhcp.c#L1080

So we are seeing all of the dhcp requests at the dhcp servers, and the first one will always have secs = 4, which will always trigger the secondary dhcp server to respond.