HACMP DARE and Cluster Snapshots Hang at "netstat -i" Command

[kknox]

I have recently implemented ssh 1.2 on my 2 node mutual takeover cluster running AIX 4.3.3.1.
A cluster reconfig became necessary and I discovered that my cluster snapshot would hang on the "netstat -i" command which HACMP uses when creating snapshots.
The same thing occurs when attempting a dynamic reconfig of the cluster.
In addition, a simple command line "netstat -i" would also hang.
I discovered that by flushing the routing table the command line "netstat -i" and HACMP functions would return to normal and operate properly. Then, at some yet to be determined periodicity, "netstat -i" would hang again.

Any suggestions or help will be greatly appreciated.

Signed,
"Chasing my tail"

 

[Yegolev]

Not a real answer, but I would strongly recommend updating to maint 09.

 

[gileb]

Do you use a dns ?
if answer is immediate when you do a netstat -in, it means its a dns problem
If yes you must have a /etc/netsvc.conf, which should contains hosts=local,bind.
And also be sure that you /etc/host contaains all your local definitions (service, standby, and boot address)



hope that helps

 

[PSD]

Follow the advice in the previous post but change your /etc/netsvc.conf file to:

hosts=local,bind4

Also ensure that you have :

Host uses DNS = true

In the HACMP run-time parameters.

Cheers

PSD
IBM Certified Specialist - AIX V4.3 Systems Support
IBM Certified Specialist - AIX V4 HACMP

 

[kknox]

That appears to have solved my problem!

Thank You very much.