Solaris 10 unresponsive

I am running Solaris 10 on a Sun X4200, and last week the OS froze to the point where I could not open an ssh session to it. It would just hang.
I had to force reboot the server from the ILOM.
This is what I saw in the /var/adm/messages log:

Mar 27 10:31:04 testing in.mpathd[158]: [ID 801587 daemon.error] initifs: ioctl (get interface configuration): Not enough space

Can anyone tell me what this issue is or could be?

you've to tell something more about the configuration...

Not sure exactly what you are looking for. Please let me know and I will be able to post the information.

SunOS x2 5.10 Generic_118855-36 i86pc i386 i86pc Solaris
nge0 is my main interface and nge1 is the failover with static IP.
we have zfs and nfs file systems setup on the server.

looks like a rather old patchlevel... if the problem returns you should maybe add an up2date patch cluster.

Those NVIDIA nics are the worst.

The error from the log suggests a problem with the software that handles IP Multi-Pathing (IPMP). There might be a bug or misconfiguration. Try disabling IPMP for now and see if that solves it. Patch the system when you can, that kernel is very very old.

This is what I have setup in hostname.nge0
somehostname netmask + broadcast + group testing up

hostname.nge1
group testing failover up

Is there anything wrong with this? Does anyone know of any bug in IPMP that could freeze the OS?

From what I can remember- no that is not correct. Your configuration is half way between probe based and link based IPMP.

Read up on IPMP, docs.sun.com has the documentation. In the meantime patch your server!

Right. I will check the link, but I don't think I am that far off from the configuration since we have been using that setup for more than 2 year and never had a problem.

I agree that the Solaris patch is old, and needs to be upgraded but not an easy task on live systems.

Have any of you had any experience with the OS freezing because of that failover configuration? Any ideas if that could actually cause that? Any solaris bug for that?

If this were my system my first step would be going through all of the available patches on SunSolve and see if there are any specific to IPMP. If there were I'd read their details and see if the patches solve issues similar to what happened. I would not rely on forum users to do the work for me. Despite being a non-authoritative source the process would also prevent me from actually learning more about the problem and inhibit my growth as a professional systems administrator.
But hey, do whatever you want.

I agree with you and I will be checking out the patches, but I wanted to know if anyone had had any problems like that, and any possible solutions. Anyway, this is a forum and I don't see a problem in asking questions regardless of your job title, and asking question certainly won't prevent me from learning more about the problem. It is not like I am sitting waiting to get a reply from this post! Cheers. Thanks for the feedback!