We are using http-log plugin to log all our requests through a http server. The upstream server for logging is an aws lambda function. Its very confusing to understand from the documentation what does the field “latencies.kong” consider to compute the latency? From the documentation ,
kong is the internal Kong latency that it took to run all the plugins.
Does it include the http-log latency itself ? (which is not possible is my understanding).
Can some one please clarify?
Also, if the kong latency is high even though the proxy latency is low, how should I go about debugging the issue ?