On Thu, Nov 21, 2002 at 09:43:44AM +0530, Arvind wrote:
> actually Binand, this dude is _probably_ getting concerned with the mx
> entries where this works well but sadly cannot be applied to normal A type
> records.

Well it can be done, but not in the way MX records work. And its not
a solution to the OP's problem either (who wants a backup webserver to
take over in case the main webserver goes down).

DNS round robin is meant for load balancing (that too, its a bit too
primitive, IMO). What the OP needs is failover. There is a difference,
which this dude seems not to have grasped (eg: Many MX records with the
same priority is load balancing, many MX records with different priorities
is failover). Cisco Local Director provides both functionalities, IIRC
(ie, the failed server is removed from the load balance pool, and two
Local Directors can work in a stateful failover mode). I have heard of
LinuxDirector too recently.

Keepalived (which I suggested in another post in this thread) provides
failover for me. YMMV.

Binand

-- 
If you found this helpful, please take some time off to rate it:
http://svcs.affero.net/rm.php?r=binand


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
linux-india-help mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/linux-india-help

Reply via email to