[Lustre-discuss] Network aliasing and HA

[Timh Bergström]

Hi,

I dont get the way lustre figures out which addresses to use, problem
is that i run ha-linux with drbd and all that jazz, i've got the
addresses to fail over, the drbd-disk to failover/resync, i've got the
mount to work (which includes loading and unloading lustre modules) -
but the problem comes when I want everything to work together;

My (current) modprobe:

options lnet networks=tcp0(eth0)10.4.21.50,tcp1(eth1)10.4.22.50

This is the errors i get:
LustreError: 10f-e: Error parsing
'networks="tcp0(eth0)10.4.21.50,tcp1(eth1)10.4.22.50"'
LustreError: 110-0: here...............................|---------|
LustreError: 4527:0:(events.c:707:ptlrpc_init_portals()) network
initialisation failed
(along with a bunch of errors since this module does not load)

The addresses are availible at the time i mount lustre, but it still fails.

I've tried with tcp0(eth0:0) which fails with about the same error,
i've tried tcp0(eth0,eth1) which gives me the wrong addresses (machine
ones) but works.

Anything i've missed here? Do I really need to use dedicated
interfaces, wich is always active or is it some way I can set the nids
to the aliased addresses? I do not want to fail the machine-addresses
over to another server. Im all out of ideas.

-- 
Timh Bergström
System Administrator
Diino AB - www.diino.com
:wq