This project is a new rewrite of Vivo which provides a visualization interface for logs, metrics, and traces.
-
I don't know Go or JS/React, but I made my best effort to put the basics of the infrastructure into the next chapter of the Vivo project.
-
VALIDATE Fluent Bit Version
Make sure the Fluent Bit running inside service is at least the version NIGHTLY_BUILD=2023-03-26-06_04_22...
- We need to support not only Logs type, also Metrics and Traces
- Previous implementation in the containers it needs between 500MB-1GB in space, it takes a lot to deploy it. Let's keep it under 80MB.
Another important factor, is that the previous version worked fine for logs, but now we ned support metrics and traces, so we needed to extend the project scope. The previous version used to depend on Fluent Bit stdout to trap records output which is complex to extend for different types of telemetry data.
Well, this is not a full rewrite, I need help so others can step up and bring the missing functionality and styles to this new version.
This version uses a different approach and components:
- [component] service/: Golang service that starts/manages Fluent Bit
- [component] frontend/: Frontend React App that pulls telemetry data from Fluent Bit (new approach)
- [component] Fluent Bit: telemetry agent with new export/streaming capabilities
To simplify the data management per type, Fluent Bit now supports a new output plugin called Vivo Exporter which buffers every telemetry data in a queue of a fixed size and exposes the content through HTTP endpoints: /logs, /metrics, /traces. The UI part pulls data from there.
All the data retrieved from the Fluent Bit Vivo Exporter comes in JSON format, and each event type has its specific structure.
Make sure you have Docker in your environment, then inside this repository, start the services with docker-compose:
docker-compose up
You can access the Web interface by using the following address:
After a successful start, the following end-points will be available:
| Port | Interface | Description |
|---|---|---|
| 8000 | Web UI | The web interface for Vivo |
| 9000 | Forward Input | Data ingestion by using Fluent Forward protocol |
| 9010 | HTTP Input | Data ingestion through HTTP |
| 2025 | Vivo Exporter | Fluent Bit Vivo streams where the UI pulls data from. This is always exposed. |
Let's use curl:
curl -XPOST -H "Content-Type: application/json" -d '{"hello": "Calyptia!"}' http://127.0.0.1:9010
-
When ingesting 1 record with Curl, the UI is not rendering the content, but when sending a second record it gets rendered properly. This is a bug in the UI code.
-
The UI tabs for metrics and traces should print just raw JSON. Note that for logs there is a expected schema where the UI use that to render the content. If metrics and traces are received, the rendering will fail due to the unexpected schema.
-
UI parity level with original Vivo UI version.
| Vivo version | Uncompressed image sizes |
|---|---|
| vivo-next | service: 84.38MB, frontend: 22.46MB, Total: 106.84MB |
| vivo (old) | Total: 991.47MB |