[Lustre-discuss] about lustre and virtual ip

[Johann Lombardi]

On Tue, Mar 11, 2008 at 09:54:44AM +0800, Felix New wrote:
>   to make the setup easily, i use a virtual ip(fload ip) between two
> mgs servers(assume mgs0: 10.0.0.1,mgs1:10.0.0.2, vip: 10.0.0.3,
> oss0:10.0.0.11, oss1:10.0.0.12, etc.). mgs0 as the primary mgs, and
> mgs1 as sencondary mgs. when use mgs0, the vip 10.0.0.3 is a alias
> addr on server0, and when mgs0 dead, the vip 10.0.0.3 will move to
> mgs1(controled by heartbeat). so the mgs hight avalibal is transparent
> to all the oss servers, clients, and they can only know the mgs

Lustre servers are stateful. You don't need to use a virtual ip since
the protocol takes care of reconnections/recovery.

> 10.0.0.3, and mkfs with param --mgsnode=10.0.0.3 on oss and mount
> cluster with nid of vip on client.

You just have to list the nids of all the mgs nodes at mkfs time
(i.e. "--mgsnode=10.0.0.1 --mgsnode=10.0.0.2" in your case).
See section "4.2.2.1 Failover" of the Lustre manual:
http://manual.lustre.org/manual/LustreManual16_HTML/LustreInstallation.html#50491441_pgfId-1286309

>   now the problem is: when i add the vip to the mgs0, and mkfs.lustre
> with --mgsnode=10.0.0.3, but the ost can't found the mgs when mount
> the ost.the same error whe mount the cluster,but but when use
> 10.0.0.1.
> client# mount -t lustre 10.0.0.3 at tcp0:/mycfs /cfs/client/
> client# mount.lustre: mount 10.0.0.3 at tcp0:/mycfs at /cfs/client
> failed: Cannot send after transport endpoint shutdown

Then, you should be able to mount the lustre fs on the client side through the
following command:
client# mount -t lustre 10.0.0.1:10.0.0.2:/mycfs /cfs/client/

comma-separation is used to list nids of the same host (a server box can
have multiple network interfaces and the other nodes can then choose the
nid appropriate to their own network interface) whereas nids of different
nodes must be delimited by colons (see the Lustre manual).

Cheers,
Johann