Skip to main content Skip to complementary content

Setting up a load balancer on AWS

This section covers main steps of setting up a classic load balancer on AWS. This load balancer will redirect client requests traffic to different Talend Runtime instances managed by Auto Scaling group.

Procedure

  1. Select AWS EC2 services – Load Balancer and choose “classic load balancer” (you may change the load balancer type per your need).
  2. In Step 1, Define Load Balancer section, fill settings as shown below:
    Information noteNote: In this demo, you will redirect the traffic on Load Balancer URL (on http port 80) to Talend Runtime Instances URL (on http port 8040 - as default Talend Runtime services listening port).
  3. In Step 2 and 3 (Assign Security Groups and Configure Security Settings), choose default Security Groups and default Security Settings (make sure you have inbound port 80 opened).
  4. In Step 4, Configure the Health Check and configure the Advanced Details sections:
    1. Set up the ping protocol, port number as shown below.
    2. Make sure the ping path points to Talend Runtime default services page on /services.
    3. Leave Advanced Details by default (change per your need if required).
    Information noteNote: In this setting, the load balancer will request /services web page on each Talend Runtime EC2 instance every 30 seconds to check if the Runtime host is in good health, if not getting health Response (http return code: 200) in 5 seconds with threshold of 2 retries, Load Balancer will deregister unhealthy instance so making sure all traffic can be directed to healthy Talend Runtime host.
  5. In Step 5, leave AZ distribution and you will not need to add EC2 instance in this step (as this Load Balancer will serve Auto Scaling Group in a dynamic way).
  6. In Step 6, add tags as Name – TalendRuntimeAutoScalingLB/
  7. Once created, the new Load Balancer is available to be used.

Did this page help you?

If you find any issues with this page or its content – a typo, a missing step, or a technical error – please let us know!

Leave your feedback here