# Redundancy and load balancing

**URL:** <https://community.wowza.com/t/redundancy-and-load-balancing/37653>\
**Category:** Wowza Streaming Engine\
**Created:** [December 21, 2011, 12:54pm UTC](https://community.wowza.com/t/redundancy-and-load-balancing/37653 "2011-12-21T12:54:26Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Mark\_Rose](https://avatars.discourse-cdn.com/v4/letter/m/bb73d2/32.png) [@Mark\_Rose](https://community.wowza.com/u/Mark_Rose)\
**Post date:** [December 21, 2011, 12:54pm UTC](https://community.wowza.com/t/redundancy-and-load-balancing/37653/1 "2011-12-21T12:54:26Z")

</div>

I’m trying to figure out a way to get high availability with the Wowza DevPay AMIs. The load balancing is fine (haven’t implemented it yet), but it’s a single point of failure. I figured I’d get around that by putting an Elastic Load Balancer in front of a Wowza load balancer in multiple availability zones. However, I’m unable to add the Wowza DevPay AMI instances to the ELB.

What’s the best way of setting up Wowza to use multiple availability zones, without a single point of failure?

Should I use a regular license for the load balancing instances in each availability zone, then use the DevPay AMIs for handing the recording of streams?

---

<div class="post-metadata">

**Author:** ![Richard\_Lanham](https://avatars.discourse-cdn.com/v4/letter/r/c68b51/32.png) [@Richard\_Lanham](https://community.wowza.com/u/Richard_Lanham)\
**Post date:** [December 21, 2011, 10:46pm UTC](https://community.wowza.com/t/redundancy-and-load-balancing/37653/2 "2011-12-21T22:46:54Z")

</div>

You can configure multiple Load Balancer Listeners for an edge by adding additional lines to the loadbalancertargets.txt file when configuring the Load Balancer sender on each edge.

You can have multiple pipe delimited Liverepeater origin servers defined either in Application.xml /originURL “rtmp://[wowza-address]:1935/origin1|rtmp://[wowza-address]:1935/origin2”, or in a .stream file, also pipe delimited, but including stream name. “rtmp://[wowza-address]:1935/origin1/stream1|rtmp://[wowza-address]:1935/origin2/stream1”

Richard

---

<div class="post-metadata">

**Author:** ![Matt\_Young](https://sea2.discourse-cdn.com/flex002/user_avatar/community.wowza.com/matt_young/32/389_2.png) [@Matt\_Young](https://community.wowza.com/u/Matt_Young)\
**Post date:** [November 25, 2013, 9:58am UTC](https://community.wowza.com/t/redundancy-and-load-balancing/37653/3 "2013-11-25T09:58:44Z")

</div>

For failing over listeners you could look at something like [LVS](http://www.linuxvirtualserver.org/) for a ip based failover. You could likely handle specific redirect IP’s by modifying the existing load balancer (via ModuleLoadBalancerRedirector.java for rtmp connections etc).

---

<div class="post-metadata">

**Author:** ![fei\_long](https://avatars.discourse-cdn.com/v4/letter/f/bb73d2/32.png) [@fei\_long](https://community.wowza.com/u/fei_long)\
**Post date:** [November 25, 2013, 6:45am UTC](https://community.wowza.com/t/redundancy-and-load-balancing/37653/4 "2013-11-25T06:45:30Z")

</div>

Hi, Richard,

Correct me if I am wrong.

Even you configured multiple Load Balancer listeners, the request URL will be different for each Load Balancer listener. Once a listener is down, the client requesting to this listener will not automatically failover to another listener. It is still a single point failure.

Another question is, can I group the edge servers connecting to a listener, for example, 192.168.1.1-10 connection to listener 192.168.1.100, for certain request, listener response with one server from range .1-.5, for certain request, listener responses with one server from range .6-.10. Can this feature be achieved through the Virtual Host?

Thanks.

regards,

long