Exercise 3: A redis Scrape Job, and Why redis_exporter Is Needed — Possible Solution
====================================================================

scrape_configs:
  - job_name: 'redis'
    scrape_interval: 30s
    static_configs:
      - targets: ['redis1:9121', 'redis2:9121']

-- Why redis_exporter is needed, rather than scraping Redis directly --
--
-- Prometheus can only scrape targets that expose a /metrics endpoint
-- in its own specific text-based exposition format (the # HELP / #
-- TYPE / metric-name-and-value lines from obs1-2). Redis itself has
-- no built-in HTTP endpoint speaking that format -- it exposes its
-- own internal statistics through its native protocol (via commands
-- like INFO), a completely different interface Prometheus's scraper
-- has no way to understand directly.
--
-- redis_exporter is the adapter this chapter describes: it's a
-- separate small process that connects to Redis using Redis's own
-- native protocol, queries the same statistics a human operator
-- would get from running INFO, and then re-exposes those same
-- numbers as an ordinary Prometheus-format /metrics endpoint -- which
-- is exactly why the two scrape targets in this config point at port
-- 9121 (redis_exporter's own default port), not Redis's own default
-- port 6379. Without the exporter sitting in between, there would be
-- no /metrics endpoint at either redis1 or redis2 for Prometheus to
-- scrape at all.

WHY THIS WORKS AS AN ANSWER
------------------------------
This writes a correctly structured scrape_configs entry matching the
chapter's own example format, then explains the exporter requirement
by contrasting Redis's own native protocol against Prometheus's
specific exposition-format requirement, and explicitly ties the
9121-vs-6379 port distinction back to the exporter's own separate
process running alongside Redis.