[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Xen-devel] RE: new /etc/xen/network script



> One potential issue is the link set of peth0, which creates 
> an interface route for peth0, at least on my currently broken 
> box.  The ifup of eth0 (veth0) will try to add a duplicate 
> route for the primary subnet via eth0. Can't validate or test 
> any of the above since my box is crashing at boot, unrelated 
> issue, will dig into this tmrw further when I get onto a box..
> 
> This in no way explains why having the del_addrs works for 
> him when it's prior to the link set up of peth0 (??), so this 
> isn't the fatal error.

No, I messed up the inclusion of Edwin's patch as I failed to understand
the root cause.
 
What I don't understand is why the link set up on peth0 causes an
address to appear. 

I thought it might be an old address from when it was eth0 getting
resurected, hence the line I added. I wander if 'ifconfig ${netdev}
0.0.0.0 down' would work better?

Ian 

> Would this make more sense: to have the route removal happen as so:
> 
> --- network-bridge.orig 2005-08-17 03:38:13.000000000 -0700
> +++ network-bridge      2005-08-17 04:08:33.298424365 -0700
> @@ -197,6 +197,7 @@
>          ip link set ${bridge} up
>          ip link set vif0.0 up
>          ip link set p${netdev} up
> +       del_addrs p${netdev}
>          if ! ifup ${netdev} ; then
>                  if [ ${kip} ] ; then
>                          # use the addresses we grocked from 
> proc/cmdline
> 
> Just for testing..
> 
> signed-off-by Nivedita Singhvi niv#us.ibm.com
> 
> thanks,
> Nivedita
> 
> 

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.