We provide live training programs and we are using a network of Woawza servers on Amazon instances.
I have configured the Origin server to also host one of the Edge application to run the server in single mode, and I have two other edge servers in case the number of attendees goes higher.
We are doing a presentation now and I am testing the load balancing process.
I can see connections alternating between servers, but I get more traffic to the main Origin/ Edge server.
I have 150 connections on that server and only 30 connections on the true Edge server.
The CPU on the Origin/Edge is at 10% and the one on the Edge is at 5%.
How the load balancer works and what is the threshold over it starts moving more traffic towards the edge server?
Thanks’
Hi,
The Dynamic Load Balancing AddOn package provides system load balancing between multiple servers running Wowza server software. Each edge server is configured to use the ServerListenerLoadBalancerSender ServerListener class. We’ll call these servers the “edge” servers. Edge servers will periodically (about every 2.5 seconds) send load and status information over UDP to a single (or multiple) Wowza servers running the ServerListenerLoadBalancerListener ServerListener. We’ll call this server the “load balancer.” The load balancer will keep track of the load and availability of each edge server with which it’s communicating. See the README.html file in the package for detailed instructions.
The load is a connection count and should be equal on all applications within the load balancing pool. If you’re load balancer is sending connections to the edges and origin, the load for the server running the origin and edge will have double the connections other servers will have as it’s using 2 applications within the load balancing pool.
To investigate this issue further I believe we will need to see the configuration of the servers within the load balancing pool. Please open a ticket by emailing support@wowza.com.
Include a ZIP file containing the following directories:
[Wowza-Install]/conf/
[Wowza-Install]/logs/
[Wowza-Install]/transcoder/
[Wowza-Install]/manager/logs/
Please do make sure you provide logs which show Wowza server starting.
If you are not sure how to get this information please see the following tutorial.
https://www.wowza.com/docs/how-to-create-a-compressed-zip-file-in-windows-os-x-and-linux
The above directories should be provided for each machine you’re using.
Regards,
Jason
Problem solved.
I created the Edge server from and EC3 AMI that I saved and when I instantiated new edge server I changed the IP address of the server in the edge application, but forgot to delete the Server.guid file.
So I had two servers, but with the same ID.
I will create a new AMI with the Server.guid file deleted so when I create a new instance from the AMI the server will generate a new ID at the first start-up.
On this note, can I use the ${com.wowza.amazonaws.ec2.AWSEC2_METADATA_PUBLIC_IPV4} variable on my Windows EC2 instance or this work only for the Linux AMI? I can look into switch to the Linux version for the edge servers.
Thanks’