cmcld error message

[Ken Stailey]

Can someone please explain what these messages are? They are happening in pairs over and over.

vmunix: Unexpected error from cluster lock query (lock_id=0): Invalid argument

cmcld: Unexpected error from cluster lock query (lock_id=0): Invalid argument

[melvyn burnard]

you appear to be having an issue when talking to your cluster lock mechanism.
Are you using a cluster lock disc? or a quorum server?
and what version of Serviceguard, with what SG patch?

[Juan M Leon · Accepted solution]

can you post the output of cmviewcl -v and
cmviewcl -v -p {package name}

Thank you

[Ken Stailey]

Is this enough for you?

# swlist | grep Service
B3935DA A.11.15.00 MC / ServiceGuard
T1859BA A.11.15.00 ServiceGuard Extension for RAC Bundle

[Ken Stailey]

# cmviewcl -v

CLUSTER STATUS
sysdbx1 up

NODE STATUS STATE
sysdbx11 up running

Network_Parameters:
INTERFACE STATUS PATH NAME
PRIMARY up 0/0/12/0/0 lan2
PRIMARY up 0/0/0/1/0 lan0
PRIMARY up 0/0/4/0/0 lan1

NODE STATUS STATE
sysdbx12 up running

Network_Parameters:
INTERFACE STATUS PATH NAME
PRIMARY up 0/0/12/0/0 lan2
PRIMARY up 0/0/0/1/0 lan0
PRIMARY up 0/0/4/0/0 lan1

[Ken Stailey]

How do I display package names? I can't find that in the cmviewcl man page.

[melvyn burnard]

So you have SG 1.15, but you do not show if t is patched on the nodes.
use the what command, or use swlist -l product |grep PHSS |grep -i service to see if there are indeed patches installed for SG.
Also it seems you have SGeRAC installed for Oracle RAC database.

You have not indicated whehter you have a cluster lock disc, or a quorum server.

cmviewcl -v will show all information including package names.

Out of interest, has this been working for some time and this has just started to happen?
If s, what changes might have been made recently. Also, are these messages on all nodes in the cluster, or just one node?

[Ken Stailey]

There is a third syslog message that always appears after the two:

vmunix: Unexpected error from cluster lock query (lock_id=0): Invalid argument

cmcld: Unexpected error from cluster lock query (lock_id=0): Invalid argument

cmclconfd[15996]: Updated file /var/adm/cmcluster/frdump.cmcld.1 for node sysxdb11 (length = 477281).

[Ken Stailey]

I don't know what to run what(1) on.

sysdbx11# swlist -l product |grep PHSS |grep -i service
sysdbx11#

How can I tell whether I have a cluster lock disc, or a quorum server? The HP admin is out on vacation and I'm filling in.

[Ken Stailey]

There are only two nodes in the cluster and they both have the error message.

I posted the complete "cmviewcl -v" output but I don't see anything that looks like a package name.

[Ken Stailey]

I got an E-mail from the person who did the installation:

MCService guard uses the shared volume group as the quorum device.

[melvyn burnard]

It seems you do not have any patches for Serviceguard or SGeRAC
You need to install these, as there are many bug fixes in the patches.
Having said that, both nodes are reporting the error, which points to a hardware issue with the device you have as your cluster lock.
I suggest that, as you are unfamiliar with this configuration, you should now log a call with your local HP Response Centre to further progress this problem.