HACMP 5.1 and Cascading without fallback

[made13]

Hi,

I have a 2 node cluster running HACMP 5.1 on aix5.2. NodeA has two resources setup as Cascading without FALLBACK set to 'true'. NodeB also has two resources setup as Cascading without FALLBACK set to 'true'.

When I did a cluster stop on NodeA using the 'takeover' option, the resources from NodeA successfully started up on NodeB. When HA was re-started on NodeA, to my surprise, it re-acquired its resources from NodeB, despite having the CWOF option enabled.

On doing a 'resource move' from NodeA to NodeB and then stopping/starting HA on NodeA, it did not re-acquire its resources, which is what CWOF is all about.

Has anyone seen this in HACMP5.1?

Thanks

 

[DSMARWAY]

what maintenance level are on ? i think 5 is the latest
( lslpp -l | grep -i cluster show something like 5.x.X.5)

check the smitty setting on the other node to check if cwof is set ?

Try a resync and see if that solves it ?

We are runnign HACMP5.1 and have set CWOF and it works as it should do.