In order to keep an eye on the state and response of the application have there been added monitoring.
//TODO: Add image of how Prometheus work.
And from the dataflow and the provisioing files to grafana have we created the following dashboards. Which tracks different parts of the system.
This dashboard shows:
- How many users are registered to the platform
- How many messages have been posted
- Who follows the most people
- Who are followed by the most people
- Who have posted the most messages.
- What request id are the latest from the simulator
Fig.1 - Business intelligence dashboard
This dashboard shows:
- Different endpoints and their response times
- How summery of different response codes and distributions.
Fig.2 - Endpoint dashboard (The image, is not so good since it was taken after the services was stopped)
This dashboard shows:
- Average (CPU, RAM, IO, Network) consumption.
- Very detailed information about stats of the Linux host machine.
Fig.3 - Node exporter dashboard
This dashboard shows:
- Errors that the simulator sees.
It does this by polling the metrics from the application Data Scraper which scrape the data from the simulator
Fig.4 - Simulator dashboard