Splunk Forwarders Buffer size and best practices?

[SplunkExplorer]

Hi Splunkers, I have some doubts about forwarder buffer, both universal and heavy.

The starting point is this: I know that, if an indexer goes down and it is receiving data by a UF,  this has a buffering mechanism to store data and send them to proper destination once it is up and running again. If I'm not wrong, the limits of this buffer can be set on a config file (I don't remeber well wich one). Now, the question are:

1. Even if the answer can be obvious, this mechanism is already available for HF?
2. How can I decide the maximum size of my buffer? is there a pre set limit or it depends on my environments?

[SanjayReddy]

Hi @SplunkExplorer 

theses attributes in outputs.conf will take care these 

maxQueueSize
autoLBFrequency
autoLBVolume
useAck

https://docs.splunk.com/Documentation/Splunk/8.2.7/Admin/Outputsconf 

[SplunkExplorer]

Hi @SanjayReddy , thanks a lot. I didn't remember about the auto fields, I got memories only about useack and maxqueue size, very usefull.

About the max queue size, are there any souces I can use to understand the best/max value I can set in my environments? I ask this because my final question is: what is the max value I can set in maxQueueSize It depends by my hardware availability or there a limit I cannot overwhelm?