Prometheus

 Prometheus

• Created to monitor highly dynamic container environments
• Constantly monitors all the services
• alert when crash
• identify problems before

Architecture

•  Prometheus server has 3 pods
• Storage -> stores metric data
  
• Retrieval -> pulls metric data
  
• Http Server -> accepts queries

Targets

• Components that Prometheus monitor are called Targets
• Linux/Windows server
• single appln
• database
  

Units

• Each target has unit of monitoring
• cpu status
• memory/disk space
• exception count
• request count
• request duration
•  

Metric

• Units that one would like to monitor is called metric
• Metrics are saved into Storage component
• It is human readable text based 

 Configuration

• Global
• scrape & evaluation intervals to define interval to scrape.
  
• Rulefiles
•  Aggregating metric value or creating alerts when condition is met
  
• scrape configs 
• What resources promethues monitors

PromQL

•