Instana — Monitoring the world of Microservices

The legacy ways and toolsets for application and infrastructure monitoring are either no longer useful or usually require too many ad-hoc scripts when we want to configure monitoring for a microservice cloud based architecture, be it a plain docker or orchestrated environment such as Kubernetes or Open Shift. Here the new age tools such as Prometheus and Instana comes for rescue which provides out of the box monitoring targeted for such microservices based environments.

Kubernetes Infrastructure View in Instana

Useful Instana Features for Microservices Environment

1. Application Perspectives : You can co-relate microservices based on Pod properties such as namespace, environment variables, infra variables or simply by names to make a group of services which are serving a specific purpose. This can represent a business flow. Instana adds a graphical representation and a dashboard for these collective services and shows all the useful data for this flow.

Application perspective Dashboards shows all useful trace data for chosen services.

Graphical view shows the communication between services. Not just the selected services but also the upstream and downstream services identified by the auto discovery. Notice the reds where calls have some issues.

2. Infrastructure: Instana provides a holistic view for complete monitored kubernetes environment. However this requires an agent to be installed on all the worker nodes which is a one time task.

Infrastructure view can easily be filtered for just required set of entities such as services, pods, namespaces, hosts etc using a filter which uses Lucene query syntax which is common among Enterprise products such as Elastic Search.

3. Analytics: Anyone who has handled a microservices architecture would agree analyzing intercommunication between services and finding a small issue such as higher latency can take ample amount of time if we just rely on logs or pods infra parameters. Instana brings these together and gives a complete picture for selected services and for a selected time period, which can really pinpoint the issue and also shows the underlying code which is showing the issue or error calls.

Conclusion

We have been so far enjoying our monitoring journey with Instana, Still finding new useful features everyday. Hope this blog helps you as well.