We have some remote sites with limited bandwidth that may be offline for periods of time due to their location. I need a way to make sure that when they come back online the Forwarders don't saturate all of the sites bandwidths trying to send all of the data it has built up whilst offline.
I looked at the maxKBps option but this looks like it is just the processed throughput and if the site is offline then it will keep processing and will not limit the output once connected again.
Is my view on maxKBps correct and if so is there an option to limit output?
Thanks in Advance
As @gcusello already said - this is the way to limit your output because your output starts with the throughput.
But there are some caveats worth being aware of.
1. The maxKBps is per pipeline. So if you have multiple pipelines, your max cumulative throughput will be maxKBps multiplied by number of pipelines.
2. The throughput limit "pushes down" on inputs by filling queues. Depending on the input type (and its settings) it may lead to loss of events if you hit the limit. For example for "push type" inputs (like syslog receiver ports), you might start losing events if your queues get full and inputs have no space to write to.