second lan card and diskless cluster

[MERTENS_1]

Hi alltogether,

I am working on a problem with an diskless cluster on HP-UX 10.20:

We have such a cluster to do data acquisition with VXI-mainframes, controlled by agilent V743 controlers. They boot their OS diskless from a cluster server. All OS are 10.20.

We used to run this cluster from one lan card and a simple hub in front. A few weeks ago we had to add a second lan card to devide the cluster net from the main net in our research center. At this point the problem starts:

First the lan card was not very easy to include in the running system but finally we managed it. This lan card shall controll the cluster now. I recreated the cluster and in the first boot process we found a lot of troubles:

-/etc/ioconfig and /stand/ioconfig were not equal and therefore ioinitrc did and do not continue.

- I forced them to be equal in copying the old files from the recovered cluster to the specific location. Then the error messages from ioinitrc disapear but unfortunately the system stops the boot after the start of the rpc.lockd daemon anywhere inside the ioinit.rc script (or even outside).

Does anybody have some ideas what to check to fix this behaviour. I am very thankfull for some suggestions concerning this matter.

Could there be a io mistake due to the second lan card in the server???

Thank you,
Oliver

[Tim Sanko]

You might want to place the second lan card on a different subnet. I believe it won't work without port aggregation, on the same subnet...