For an AWS environment (actually the question is valid for Azure) , I was wondering how redundancy works with multiple ActiveGates - I mean beyond OneAGents.
All the documentation states is about OneAgents using multiple ActiveGates for load balancing and redundancy -as long as in the same network zone - but nothing about AWS API connections or Log forwarding.
_ Logs are sent by a forwarder that we install - so I assume this piece of software - like the OneAgents - eventually discovers and is aware of multiple ActiveGates. Problem is as I just stated this is an assumption, nothing is stated in the documentation about it. Not that I need an in depth understanding of it, but at least a general idea: do the log forwarders have the ability to fail over to another active gate?
Or do they forward the logs to all the ActiveGates and Dynatrace ingests from a single source at a time? Does not matter in terms of HA but in terms of traffic potentially this doubles it.
_ AWS Cloudwatch API metrics: ActiveGates do not talk to each other so how do they decide which one pulls the metrics from AWS? I assume it is Dynatrace that decides and 'activates' one or the other, unless they both pull the data and Dynatrace ingest from a single source?
Additional information about aws forwarder here:
Thanks - I didn't find anything in this link about the ability to switch to an alternative ActiveGate? I recall that when we deploy we put the URL pointing to one of the ActiveGate, just as with OneAgents, so I assume as OneAgents eventually learn about other ActiveGates and will use them so does the forwarder? Thing is we are in a design phase and people are asking questions and nowhere do I find a definitve statement about whether ActiveGates can offer redundancy for OneAgents only or for anything transiting via ActiveGates such as logs and cloudwatch metrics.