fluent / fluent-plugin-prometheus

A fluent plugin that collects metrics and exposes for Prometheus.
Apache License 2.0
257 stars 79 forks source link

What's plugin_id? how to get rid of it #177

Closed flaker closed 4 years ago

flaker commented 4 years ago

Hi!

I am using fluentd receiving logs from multiple fluent-bit instances. I think it's a common scenario. While querying fluentd I see the following results:

fluentd_output_status_buffer_available_space_ratio{endpoint="fluentd-metrics", hostname="fluentd-aggregator-useast2-c1-us-east-2b-0", instance="172.16.171.158:24231", job="fluentd-aggregator-us-east-2b", namespace="default", plugin_id="object:1920508", pod="fluentd-aggregator-useast2-c1-us-east-2b-0", service="fluentd-aggregator-us-east-2b", type="s3", worker_id="0"}
fluentd_output_status_buffer_available_space_ratio{endpoint="fluentd-metrics", hostname="fluentd-aggregator-useast2-c1-us-east-2b-0", instance="172.16.171.158:24231", job="fluentd-aggregator-us-east-2b", namespace="default", plugin_id="object:1c73048", pod="fluentd-aggregator-useast2-c1-us-east-2b-0", service="fluentd-aggregator-us-east-2b", type="s3", worker_id="0"}

they are the same fluentd instance but they differ in "plugin_id". Is there any way to make it more useful for a query? it's current value (object:xxxxxxxx) is not useful as it is.

The docs I could find only mention this label for the in_tail plugin that I am not explicitly using since I only receive using forward.

Thank you!

dimitarKiryakov commented 2 years ago

@flaker Did you manage to find a way?