dNation Kubernetes Monitoring allows us to see status of our infrastructure and applications at a glance using semaphore (green/orange/red) principle:
Basic design principles:
- Intuitive – Green, orange and red colors signaling whether or not your action is needed
- Drill – down – if you want details why something is green, orange or red, just click it
- Relevant information only – provide only metrics relevant for this particular area of interest and drill-down level, side-by-side with logs (experimental feature)
Basic use case
- You received 17 alerts at 3 AM
- You look at dashboard and see multiple problems, application A is in red, some PVCs are red, while k8s cluster itself is green; so you know where to start to look for root cause (non-green color)
- You click PVC stat-panel, get a list of PVCs, all of them green, 1 of them red
- It turns out that red PVC is used by application A
- You fix PVC issue, and suddenly whole dashboard becomes green, so you have some indication that things are probably all right now
- As a result, incident response time is improved
This project has been developed, maintained and used in production by professionals to simplify their day-to-day monitoring tasks and reduce incident reaction time.
Commercial support is available, including 24/7, please contact us.
This project is awailable also at AWS Marketplace.
Following projects are used:
- Helm package manager – for easy installation
- Prometheus – for metrics collection
- Grafana – for metrics visualization
- Jsonnet – as a templating language for Prometheus alerts and Grafana dashboards
- Loki – for logs collection
- Docker – for building images
We found about dNation Kubernetes Monitoring from your medium blog. We were facing the same issue with spending so much time on customizing dashboards. The dNation monitoring is really useful for us now and it immensely helps in debugging. Thanks to you and your team for this awesome tool.
Open Source Software, Apache 2.0 License.