Skip to content

Latest commit

 

History

History
21 lines (17 loc) · 1.15 KB

readme.md

File metadata and controls

21 lines (17 loc) · 1.15 KB

ALIVE

Periodically test endpoint in location and ISP aware

Overview

There are 2 part that make Alive, alive-agent run the test cases while alive-server contain information including test cases, agent information and many other relevant configurations. Alive cluster will consist of several alive-agent deployed on various configuration of location and ISP, one alive-server coordinating all the agents.

alive-agent will run test cases and expose the result as prometheus formated metrics ready to be scrapped, by default alive is using fluent-bit to scrape and store the data into prometheus via prometheus_remote_write.

architecture overview of alive cluster

TODO

  • Package for agent with auto update support, probably using snap
  • Automatic provisioning of agent and it's host
  • Account / agent security, probably first combination of simple password and jwt
  • API remove agent
  • multitenancy of test
  • think better test runner
  • think better test result storage and reporting
  • working test report dashboard
  • make the test report dashboard multitenant
  • think about tenant / user of test