To reply to myself, the problem seems to be a delay (possibly due to a bug?) 
whereby it times out. I cannot see any DHCP requests leaving the br0 device. 
However, when I use 'crtl+b' to get into the gPXE prompt, and run 'dhcp net0' 
some DHCP packets are sent and get a reply, and the interface gets an IP 
address.

 Not sure what is going on here!


 -Cam

-




 KVM host has been set up with bridged networking, and I can successfully 
kickstart a host using the command line

 virt-install --name=vtest --ram 512 -f /dev/vg00/vtest --vcpus=2 --network 
bridge=br0 -l http://10.62.128.11/mirrors/scientific/6.1/x86_64/os  -x 
"ks=http://10.62.128.11/kickstart/vm-test.cfg  console=ttyS0"

 However, I cannot get it to work when using virt-manager GUI

 I select 'Network Boot' (PXE), under 'Advanced Options', I select 'Host device 
em1 (Bridge 'br0'), leaving the default Mac address options, however, it fails 
to get a DHCP address, giving the following error in the console:

 ---------------------------8<------------------------

 Waiting for link up on net0....ok
 DHCP (net052:54:00:35:8c:c2 )......................connection timed out 
(0x4c106035)

 No more network devices

---------------------------8<------------------------

 It generates the following network config in the xml:


 ----------------8<----------------

 <interface type='bridge'>
 <source bridge='br0'/>
 <mac address='52:54:00:35:8c:c2'/>
 <model type='virtio'/>
 </interface>

----------------8<----------------

 >From KVM guests I have kickstarted via the command line method, I can get 
 >DHCP addresses.
 I cannot even see any DHCP requests on br0 on the KVM when it is attempting to 
get an address via GUI method


 Any ideas why the virt-manager method does not work?

Cheers for any help,

 Cam
_______________________________________________
virt-tools-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/virt-tools-list

Reply via email to