As the entrypoint to Cloud Foundry, routing tier-related errors come from a variety of places. While some of these errors are indications of actual bugs in the Cloud Foundry source code, there can be a number of other root causes: misbehaving applications, misconfigured load balancers, and infrastructure issues, just to name a few. In order to isolate and debug these problems, it becomes important to know which data to collect, how to collect it, and what in the data can indicate the root cause.
In this talk, Angela and Nitya will cover what information is useful to collect and what to look for in the data to systematically eliminate possible causes of the problem. Drawing on their experiences as members of the CF Routing team, they'll go through issues that operators have seen in the past, and how they use tools such as routing logs, pprof, flamegraphs, and wireshark to debug, isolate, and work around these issues.