07-20-2008 11:56 PM
Hi!
The serverfarm contains 2 rserver.
We use default round-robin algorithm for load-balance.
Our experience ACE send every second connection to same server.
2 here 2 there etc.
Is it normally ?
Regards,
07-21-2008 12:05 AM
Is stickiness configured?
If yes then you wont get an even distribution.
I would recommend clearing the serverfarm statistics (clear serverfarm
connections" column and not the "Current conns" column.
Round robin algorithm doesn't care about the current number of connections, it only takes into account the total number of connections sent. Current connections can be more on 1 server vs other just because of HTTP1.1 persistence (where multiple GETS are sent on a single TCP connection).
If you want to loadbalance on "current connection" basis then change the predictor to
Least connection.
Syed Iftekhar Ahmed
07-21-2008 12:15 AM
Hi!
It was a exactly and accurate answer for me.
Very thanks.
Regards,
There wis not stickiness.
07-21-2008 12:18 AM
Hi,
On a lightly loaded ACE you may see this. The ACE has two network processors which work independently. Both implement the RR algorithm on the same servers in the serverfarm. Hence you may see what looks like an error in the distribution.
HTH
Cathy
07-21-2008 12:25 AM
Hi Cathy!
It was very intresting remarks.
when i clear the serverfarm counter RR work correctly. ( 1-1 in teh test environment )
Earlier i see 2-2 distribution.
where is the true ???
which algorithm gives 1-1 distribution ?
Regards,
07-21-2008 12:43 AM
Hi,
Over the long term, in the absence of stickiness then the load balance algorithms will even out. The key in my previous reply was that the NPs act independently. There is no way of forcing an exact 1-1 distribution. Whether you see 1-1 or 2-2 is a matter of chance - depending on how the NPs see the serverfarms and make their decisions.
Kind Regards
Cathy
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: