Originally posted by: SystemAdmin

Is the RSCT 2.4.12.0 that comes with the latest AIX 5300-11-03-1013 stable ?

I updated my system AIX 5300-11-03-1013 with HACMP 5.4.1.7. I did a failover test with halt -q on the Node1. Node2 crashed with the error: TS_DMS_EXPIRING_EM (Dead Man Switch) and dump.

There is no activity on my database cluster. It's is in testing.

Attached: errpt output
#PowerHA-(Formerly-known-as-HACMP)-Technical-Forum
#PowerHAforAIX

Originally posted by: Casey_B

There is an rsct apar that sounds applicable.

IZ66768: SERIAL NETWORK THREAD MONITORING ERROR CAN CAUSE DMS TIMEOUT DURING HACMP FAILOVER

Upgrade to include at least this APAR...Or to the latest RSCT level.

Hope this helps,
Casey
#PowerHAforAIX
#PowerHA-(Formerly-known-as-HACMP)-Technical-Forum

Originally posted by: SystemAdmin

Thank's Casey B

I upgraded to RSCT 2.4.12.2
The problem is solved.
#PowerHA-(Formerly-known-as-HACMP)-Technical-Forum
#PowerHAforAIX

Originally posted by: parsona

Hi

Having the same problem. Have the following installed:

AIX 5.3 ML11 SP3
PowerHA 5.4 with SP8
rsct 2.4.13.1

Cluster has a shared concurrent volume group that is mirrored between two buildings. (AIX mirrorvg).

If i issue halt -q, failover works but if I physically remove power from one site, i.e server and storage, I get DMS on the second node while it is aquiring the resources.
Cluster has two disk heartbeats, one in each building.
Is this normal behaviour.

Thank for your assistance.
#PowerHA-(Formerly-known-as-HACMP)-Technical-Forum
#PowerHAforAIX

Originally posted by: Casey_B

Hello,

The problem described above is very specific. It only occurs during a node down, and at specific
levels of RSCT.

Thanks,
Casey
#PowerHAforAIX
#PowerHA-(Formerly-known-as-HACMP)-Technical-Forum