Advice for validating JMX exporter rules

[AagonP]

I have read through the documentation but I still encounter some challenges when working JMX exporter rules. So I'm opening this issue only to ask for advice on how we can better handle these scenarios when working with JMX exporter:

1. When I was writing a rule like:

   - pattern: "kafka.consumer<type=consumer-fetch-manager-metrics, client-id=(.+), topic=(.+), partition=(.+)><>records-lag"
     name: kafka_consumer_fetch_manager_records_lag

   As I forget that JMX exporter rule regex is not anchored, kafka_consumer_fetch_manager_records_lag is actually presenting 3 metrics ( records-lag, records-lag-avg, records-lag-max). How can I avoid situation like this when certainly my goal was just to get the records-lag attribute.

2. Is there a way we can do linting for the configuration file?

Thanks for your time.

[dhoard]

Currently, there is no linter.

An out-of-band linting tool is hard to implement since the rules are based on the dynamically scraped MBeans.