We have an environment which has 2 collectors. Our environment is set such that when a service is being deployed, it gets the dynatrace collector details from a single common environment file. Initially when we had a few services, we had entered the details of 1 collector in the environment file. But once more services were brought in, we observed an imbalance in the agent distribution across the 2 collectors since most of them were pointing to collector1. (Few of them had manual config to collector2).
We tried creating a common dns name for the 2 collectors and adding this name in the environment file. But this did not help us in loadbalancing the agents across the collectors.
On looking up the options for this, I came across Collector group load balancing. But I found the following info in the Dynatrace community which shows that this is not a foolproof method.
Is there a way to ensure loadbalancing of the agents?
Solved! Go to Solution.
There is failover/load balancing built into the functionality of the agents and collector groups. The mechanism ensures that one collector in a group does not get overloaded. If a collector in a group gets more than 50% of the overall agents in the group, agents will automatically move to the other collectors.
Also, be careful not to point all agents to one collector. While the mechanism will balance, there is increased load on the collector when applications start up. If all of the agents/apps are restarted at the same time, you run the risk of overloading the collector.
I recommend reading through the following doc for more info and to ensure collector best practice: https://www.dynatrace.com/support/doc/appmon/admin...
Hope this helps!