listen-to clusterIP address

Abdul Khader akhader at ies.etisalat.ae
Wed Jun 5 11:39:42 UTC 2013


Better to write a script which would first check the availability of 
Virtual IP before doing "rndc reconfig" during a failover. In case the 
script does not find the VIP in the first run, you can put in a loop to 
check for VIP for N number of times with N number of seconds interval.
The failover time depends mostly on the resources being transferred. If 
VIP is the only resource, then the script should pick up the VIP in 
about 60-80 seconds.

Regards

Abdul Khader

On 05/06/2013 1:57 PM, Phil Mayers wrote:
> On 05/06/13 20:06, paul wrote:
>> Thanks for the quick reply. rndc reconfig has the same problem as a
>> restart. I need to automatically listen to the new ip address without
>> manual intervention.
>
> "rndc reconfig" need not be manual - surely your cluster software can 
> execute a script on IP failover?
>
> Anyway, as you've spotted, lowering the listen interval can emulate 
> this. Personally I'd want bind to respond a bit quicker than 0-60 
> seconds when a failover occurs.
>
> The other alternative under Linux would be a long-running process 
> listening to a netlink socket for address changes and exec'ing a 
> reconfig. I wonder if there is such a beast?
> _______________________________________________
> Please visit https://lists.isc.org/mailman/listinfo/bind-users to 
> unsubscribe from this list
>
> bind-users mailing list
> bind-users at lists.isc.org
> https://lists.isc.org/mailman/listinfo/bind-users
>



More information about the bind-users mailing list