HPUX Cluster Problem

[vinay_26]

Hi,

We have a single node cluster on HP UX 11.0. When i try to activate a VG manually when the packages are down it says this error.

vgchange -a e vg_p04_ss01
vgchange: Volume group "/dev/vg_p04_ss01" belongs to a different Cluster. Activation failed.

I am not sure whether its a single node cluster or its apart of metro cluster. there is a cluster with threee node in a different location with the identical VG configuratiosn.


I would like to know is there any command to identify whether its a part of metro cluster?

[melvyn burnard]

no, there is not. WHat you can do is see if the MC software is installed
swlist -l product | grep -i metro
Also, take a look at the cluster configuration of the other 3 node cluster to make sure this node does NOT belong to that cluster.
If not, then be VERY careful in confirming these disks do NOT belong to the other cluster before attempting to use them on this cluster.

[vinay_26]

When i see the package it says unowned what does it mean.

Its a single node cluster. pls find the Cmviewcl output also.

# /usr/sbin/cmviewcl -v

CLUSTER STATUS
clusterukl2_rf up

NODE STATUS STATE DLM/GMS STATE
mary up running halted

Network_Parameters:
INTERFACE STATUS PATH NAME
PRIMARY up 4/0/0 lan0
PRIMARY up 1/0/0 lan1
STANDBY up 2/0/0 lan2
STANDBY up 3/0/0 lan4

UNOWNED_PACKAGES

PACKAGE STATUS STATE PKG_SWITCH NODE
ukl1inv_rf down halted disabled unowned

Policy_Parameters:
POLICY_NAME CONFIGURED_VALUE
Failover configured_node
Failback manual

Script_Parameters:
ITEM STATUS NODE_NAME NAME
Subnet up mary 10.221.51.0

Node_Switching_Parameters:
NODE_TYPE STATUS SWITCHING NAME
Primary up enabled mary

PACKAGE STATUS STATE PKG_SWITCH NODE
ukl1sam_rf down halted disabled unowned

Policy_Parameters:
POLICY_NAME CONFIGURED_VALUE
Failover configured_node
Failback manual

Script_Parameters:
ITEM STATUS NODE_NAME NAME
Subnet up mary 10.221.51.0

Node_Switching_Parameters:
NODE_TYPE STATUS SWITCHING NAME
Primary up enabled mary

PACKAGE STATUS STATE PKG_SWITCH NODE
ukl2bck_rf down halted disabled unowned

Policy_Parameters:
POLICY_NAME CONFIGURED_VALUE
Failover configured_node
Failback manual

Script_Parameters:
ITEM STATUS NODE_NAME NAME
Subnet up mary 10.221.51.0

Node_Switching_Parameters:
NODE_TYPE STATUS SWITCHING NAME
Primary up enabled mary

[Geoff Wild]

Is the cluster down?

IE - cmhaltcl?

if not - try:

vgchange -c n /dev/vg_p04_ss01

then:

vgchange -a e vg_p04_ss01

Again - make sure the disks are not already running on another cluster - IE - if they sync with say the Metro cluster - make sure they are split before vgchanging them...

Rgds...Geoff

[melvyn burnard]

It means no node owns this package at the time of running the command, usually because the package is DOWN (been halted, or not started).

It may help you to review the Managing Serviceguard manual at:
http://docs.hp.com/en/ha