Python Log Monitoring with Grafana

I need a lean Python utility that tails live system, application and security logs coming from roughly 6-10 downstream servers, parses them on-the-fly and pushes the results into InfluxDB. From there I want to see real-time Grafana dashboards that summarize error rates, performance metrics and user activity so that issues surface instantly.

What matters most is reliability and low latency: the script should connect (SSH or TCP streaming—whatever you propose) to each server, detect log rotation automatically, and batch data efficiently so Influx stays healthy even during bursts. Field naming, measurement tags and retention policies must map cleanly to Grafana so that new panels can be added without rewriting code.

Deliverables
• Python source (well-commented) with a small config file where I can add/remove servers and choose which log files to follow
• InfluxDB schema setup script or clear instructions
• A starter Grafana JSON that contains sample panels for error counts, response time percentiles and active users per host
• Read-me explaining how to deploy, run and extend the solution

Acceptance criteria
1. Pointing the script at sample log files on two test servers should populate InfluxDB within 30 seconds.
2. Grafana panels must update at ≤5 s intervals with the three metric groups listed above.
3. No data loss during a forced log rotation test.

If you have prior experience with Telegraf, Fluent Bit or Loki and see an advantage in integrating them, feel free to suggest an alternative path, but the core requirement remains a Python-driven pipeline into InfluxDB feeding Grafana.