Course: Prometheus | The Complete Hands-On for Monitoring & Alerting

udemy.com/course/prometheus-course

Architecture

prometheus pulls metrics

discovery: hardcoded or service discovery

UI

localhost:9090

up - targets scraped by prometheus

config file

prometheus.yml

Plain Text

metric_name{key_name="value", key="value"} metric_value

Exporters

get metrics from third-party systems

TODO: PromQL

client libs

prometheus official

community

metric types

counter - up / restarted service

gauge - up an down

<name>

 

app_requests_inprogress

summary - samples

<name>_count

<name>_sum

 

rate(app_Response_latency_seconds_sum[5m]) / rate(app_Response_latency_seconds_count[5m])

histogram - samples in buckets

<name>_bucket{le="..."}

<name>_bucket{le="..."}

<name>_bucket{le="..."}

<name>_count

<name>_sum

 

 

# le = less then or equal to

# -> every request will be in +Inf bucket

Recording rules

recording rules - pre-cached

stores new time series of the calculated values