In this quick walkthrough, see how to use io.Insights telemetry to investigate io.Connect Desktop performance in Grafana and Jaeger, then extends the example to application usage and user journeys.
What the walkthrough covers
00:08: Compare startup performance in Grafana. Filter the dashboard by environment and platform version, then compare platform startup times across releases. A slow release stands out in the example, giving the platform team a starting point for investigation. Also look at Workspace, Layout and application startup. The displayed times are demonstration data, and warning thresholds need to reflect your own applications and workloads.
03:06: Find an error with traces in Jaeger. Open a Workspace trace and break it down into spans, which represent individual operations within the trace. Inspect the timing, error information and attributes to see where the operation ran into trouble.
04:30: Explore traces in Grafana. Inspect traces in Grafana using a Desktop startup trace and its component calls. One traced operation’s duration is only part of the overall startup experience, so it should not be treated as the complete startup time.
05:26: View centralized logs. Open logs from the io.Connect Gateway and Desktop, filter by log level and inspect timestamped entries. With log export configured, your team can examine those entries centrally instead of first collecting log files from a user’s machine.
06:28: Add context to application usage. Move from default metrics to custom telemetry. Add a user’s role so that application usage can be compared for groups such as traders and account managers. This requires the relevant custom data and dashboard queries.
08:41: Extend the setup to user journeys. Explore export configuration, user journeys and click stream settings, and see how app-specific instrumentation can help investigate where users spend time or leave a workflow. This section introduces the possibilities; it is not a complete configuration tutorial.