HACMP warning can't understand it 3

[alexia32]

Hello

I have a new cluster hacmp that I have configure the same way as the production one.
HA Configuration Verification: from node2 & node1nodes:
Verification to be performed on the following:
Cluster Topology
Cluster Resources
Retrieving data from available cluster nodes. This could take a few minutes.....
Verifying Cluster Topology...
WARNING: node1: Read on disk /dev/sandisk1 failed.
Check cables and connections.
A reserve may be set on that disk by another node.
WARNING: node2: Read on disk /dev/sandisk1 failed.
Check cables and connections.
A reserve may be set on that disk by another node.
WARNING: The mode of volume group vg200 cannot be determined on node node1
WARNING: The mode of volume group vg200 cannot be determined on node node2
WARNING: The mode of volume group vg201 cannot be determined on node node1
WARNING: The mode of volume group vg201 cannot be determined on node node2
Verifying Cluster Resources...
processing RG: node1_rg
processing RG: node2_rg
processing RG: node1_rg
processing RG: node2_rg
processing RG: diskhb_rg
Remember to redo automatic error notification if configuration has changed.
Verifying additional pre-requisites for Dynamic Reconfiguration...
...completed.
Updating ODM errnotify on node node2.
Updating ODM errnotify on node node1.
Verification has completed normally.


Any idea what could mean the message:

WARNING: The mode of volume group vg200cannot be determined on node node1

I am on AIX 5.3TL8SP3 and HACMP v5.2 (I know it is on old version but we will organise to upgrade to5.4 as now the applciation has certified this one...)

Thanks in advance for any help

Cheers
Al

 

[khalidaaa]

Could be a bug in HACMPv5.2!

http://rootvg.net/component/option,com_smf/Itemid,91/topic,117.0

What is the /dev/sandisk1? It complains about this first!

Regards,
Khalid

 

[alexia32]

Thanks Khalid.

What is interesting I don;t have these warnings on the other clusters... and there all the same version and configured the same way.
Anyway that's good to know it is a known bug and that can be ignore!

Cheers
Al

 

[p5wizard]

I'd still schedule a cluster failover test next time you have a maintenance window to make double sure this is a bug...


HTH,

p5wizard

 

[ogniemi]

can be ignored if "odmget -q attribute=conc_capable CuAt" returns affected VGs and you the ODM stanza as on:

http://www-01.ibm.com/support/docview.wss?uid=isg1IY75595

 

[alexia32]

Hi

Thanks a lot. I have checked and can see all my VG correctly and checked the APAR.
So yeah I can defenitly ignore the warning. But I will have an outage this week on production and I will take the opportunity to check if I have the same warnings also as all cluster are configured the same way.

cheers
al

 

[alexia32]

Hello
I have checked the prod servers and they all have the same warning.
So those messages can be ignored as the vg are here and cluster working perfectly.

Thanks again

This thread can be close.

Cheers
Al