Hello.
Weight is set in Pool of load balancer
Please tell me what action will be taken if Weight is set.
I confirmed the operation
We had a movement that was frequently allocated to servers with high Weight numbers
Is this movement correct?
I'd like to do traffic forwarding to the intended server.
(BIG-IP - Priority Group Activation) operation
For example
· There are four web servers (web1, web2, web3, web4).
- When normal, only load 1 is set for web 1 and web 2.
· When web 1 fails, only web2 communication
- When web1 and web2 fail, they communicate with web3 and web4.
To do this
Weight setting below.
web1 weight1
web2 weight1
web3 weight2
web4 weight2
Argorithm: Round-Robin
But the operation
Communication is possible in all situations.
The "weight" parameter is just a way of setting a proportion of the traffic that will be delivered to a given pool member so regardless of what you do, all servers in a given pool will always receive some traffic (some just receive more than others and higher weight receives more traffic so the way you've got it configured now, web 3/4 will receive twice as much as web 1/2).
Just to make sure I understand, it sounds like what you're after is to have web 1 and web 2 receive all the traffic unless both are down, and only then start sending traffic to web 3 and web 4. If that's the case, you'll need to put web 1 and 2 in their own pool (let's call that 'active_pool' for the sake of this example) and web 3 and web 4 in a separate pool ('standby_pool' for example) and use an application rule to fail over to the 'standby_pool' when both servers in the 'active_pool' are down. To do so, you can set the default pool on your virtual server to the 'active_pool' and then apply the following application rule:
acl active_down nbsrv(active_pool) eq 0
use_backend standby_pool if active_down
The "weight" parameter is just a way of setting a proportion of the traffic that will be delivered to a given pool member so regardless of what you do, all servers in a given pool will always receive some traffic (some just receive more than others and higher weight receives more traffic so the way you've got it configured now, web 3/4 will receive twice as much as web 1/2).
Just to make sure I understand, it sounds like what you're after is to have web 1 and web 2 receive all the traffic unless both are down, and only then start sending traffic to web 3 and web 4. If that's the case, you'll need to put web 1 and 2 in their own pool (let's call that 'active_pool' for the sake of this example) and web 3 and web 4 in a separate pool ('standby_pool' for example) and use an application rule to fail over to the 'standby_pool' when both servers in the 'active_pool' are down. To do so, you can set the default pool on your virtual server to the 'active_pool' and then apply the following application rule:
acl active_down nbsrv(active_pool) eq 0
use_backend standby_pool if active_down
Hello, I wrote a blog post on that here: Creating an Active-Passive Pool on VMware NSX Load Balancer
lhoffer>
Thanks for your comment.
understood.
Bayu Wibowo>
Thanks for your comment.
I will look at the blog.