librdkafka Grafana dashboard populated from statistics JSON message

stats-dashboard/.gitignore

@@ -0,0 +1 @@
+out.db

stats-dashboard/README.md

@@ -0,0 +1,39 @@
+# Grafana dashboard for librdkafka stats
+
+This Grafana dashboard allows to show metrics created from a series of librdkafka json stats, collected from a text file.
+
+Every line that looks like a librdkafka stats JSON is parsed and the corresponding metrics are put into a SQLite DB.
+
+From there the metrics are ingested into VictoriaMetrics to be queried with PromQL.
+
+Then a Grafana instance is started with a dashboard that queries those metrics, as soon as it's started the corresponding dashboard url
+is opened.
+
+> **_NOTE:_**  When mixing stats from different processes the instance name can be the same. To differentiate them, one can put in the same line a process id enclosed by the `####` separator e.g.: `####process 1####`.
+
+Default password is `librdkafka:librdkafka`
+
+## Requirements
+
+* Python 3
+* Docker
+
+## How to run
+
+1. In a virtual environment install the dependencies with `pip install -r requirements.txt`
+
+2. Generate the DB and open the dashboard
+
+```
+./run.py <text_file_to_parse.log>
+```
+
+2. Later checks
+
+```
+./run.py
+```
+
+## Preview
+
+<img src="img/stats-dashboard.png" width="1200">