CONFIGURE_PROMETHEUS.md

CONFIGURE PROMETHEUS

When Prometheus is enabled, an endpoint is available in port 24321 allowing to see the following metrics:

• fluentd_input_status_num_records_total - the total count of received logs by tag (e.g.: start, chunk, event, postStart, complete and activity)
• fluentd_output_status_emit_count - the total count of forwarded logs by output (e.g.: stdout, remote-syslog, s3, cloudwatch, splunk-hec, datadog, azure-loganalytics, sumologic, kafka, mongo, loki, elasticsearch and bigquery)
• fluentd_output_status_num_errors - the count of total errors by output match

To enable it, you need to set the variable LOG_EXPORT_CONTAINER_ENABLE_MONITORING=true.

To see an example, you can use docker-compose-prometheus.yml to run Log Export Container with Prometheus and Grafana. Then you can access the Log Export Container Metrics dashboard in Grafana (in the port 3000) and see how it's used. There we have the following panels:

1. Received Logs - count of the received logs per tag in the last 5 minutes

2. Forwarded Logs - count of the forwarded logs to the defined output plugins in the last 5 minutes

3. Total Errors Count - total count of errors in the last 5 minutes

4. Last Execution Status By Output - Last Execution Status by output. It'll show 1 if the last forward try failed, otherwise it'll show 0.

Here is a demo video showing the Grafana dashboard:

LEC.prometheus.mp4