README.md

Metrics and logs analysis toolset

NOTICE: Currently scripts are of PoC quality, expect rough edges

install influxdb python package (preferably with pip in virtual environment with python3)
run start_database.sh - this will start docker containers with InfluxDB and Grafana
open localhost:3000 in browser and login to Grafana, then
- add InfluxDB data source pointing at http://indy-load-influxdb:8086
- upload general_metrics.json dashboard
run upload_metrics.py - it will look for logs/metrics/metrics*.csv files
optionally run upload_logs.py - it will look for logs/Node*.log* files
- this can take some time, for example it took 30 minutes to upload 3 Gb of uncompressed logs
- compression looks quite good - aforementioned 3 Gb of logs took about 350 Mbs inside database
look at dashboard, analyze metrics and logs, create new dashboards (don't forget to save and commit them so others can use them as well)
when you're done you can run stop_database.sh to stop and remove all related containers and docker volumes