Originally posted by: SystemAdmin

Hi,
Before I start the question i want to confirm:

If cluster is running on a 2 node cluster, to test failover, if i disable both ports of SAN switch to server A, then server A should failover to server B? right....

the problem i have is my disk heart beat is working well, i defined the dhb_network properly, IBM too verified all is correct. but when i disable the port on san switch for the fiber connection to server A, the resource group of A ( i verified that the resource group is proper too) failover doesn't occur. Some say it is not necessary to assign the diskhb-vg to any resource group. Guess what i had it working, but wanted to do a fresh config with properly documenting it...alas it's not workign now. if i unplug the ne cables, it failsover, but not when i unplug fiber cables.

Select a Communication Interface/Device to Change/Show

Move cursor to desired item and press Enter. Use arrow keys to scroll.

Node / Network

Interface/Device IP Label/Device Path IP Address

eclipzconk / net_diskhb_01

hdisk3 conkhb /dev/hdisk3

eclipzconk / net_ether_01

en4 eclipzconk_boot 10.1.204.233

eclipznoc / net_diskhb_01

hdisk3 nochb /dev/hdisk3

eclipznoc / net_ether_01

| en4 eclipznoc_boot 10.1.204.231
Change/Show All Resources and Attributes for a Resource Group

Type or select values in entry fields.
Press Enter AFTER making all desired changes.

Entry Fields
Resource Group Name heartvgg
Participating Nodes (Default Node Priority) eclipzconk eclipznoc

Startup Policy Online On All Available Nodes
Fallover Policy Bring Offline (On Error Node Only)
Fallback Policy Never Fallback

Concurrent Volume Groups diskbeatbtvg +
Use forced varyon of volume groups, if necessary false +
Automatically Import Volume Groups false +

Application Servers [] +

Tape Resources [] +
Raw Disk PVIDs [] +
Disk Fencing Activated false +

Fast Connect Services [] +
Communication Links [] +
Workload Manager Class [] +

Miscellaneous Data []
any help will be appreciated...thanks
#AIX-Forum

Originally posted by: The_Doctor

There are a number of things:

1. I hope your ethernet network has more to it that what is suggested in your example. But since the ethernet network is not your immediate problem, we'll move along.

2. For what you are attempting, the Startup Policy AND Fallover Policy are both questionable. You say your settings are:

- Startup Policy = online on ALL available nodes
- Fallover Policy = bring offline

in which case, there is nothing to fallover since the Resource Group is already running on all nodes. Change these settings (to something a little more appropriate) if you want the Resource Group running only on one node at a time...... and are expecting it to failover to the other node.

3. I NEVER have my disk_hb_vg as part of any Resource Group. I can think of no reason why I would want it as part of a Resource Group. Unless you have some legitimate reason for it...... I'd suggest avoiding the unecessary complication and remove the disk_hb_vg from your Resource Group(s). The next admin person will probably appreciate you more.
#AIX-Forum