// the find
yunlzheng/prometheus-book
Prometheus操作指南
A Chinese-language GitBook that covers Prometheus from first install through PromQL, Alertmanager, Grafana, clustering, service discovery, Kubernetes and the Prometheus Operator. It suits Chinese-speaking ops engineers who want one ordered read instead of piecing together the official docs and blog posts.
The order is sound: PromQL comes before alerting, and the alerting chapter follows the sequence you would actually configure things in, from rule syntax to Alertmanager routing, inhibition and receivers. The receiver section covers WeChat Work and DingTalk webhooks, which most English-language Prometheus material skips, and the examples/ directory has real config files for federation, file-based service discovery and Operator-managed monitors rather than only prose snippets. The best-practice chapter on the four golden signals and the USE method is short but tells you what to alert on, which is the question most readers end up with.
The last push was February 2024 and the content is older than that. It teaches Grafana Singlestat and Graph panels, which Grafana has since replaced with Stat and Time series, and the Kubernetes and Operator chapters were written against older manifest layouts that will probably need rework on a current cluster. It also predates Prometheus 3.0, so anything version-specific needs checking against the current docs. Treat it as a map of the concepts, not a deployment guide. The repo also looks unfinished in places: a draft/ folder of chapters that never made it into the book, a stray examples/cluster/xxx.log, and misspelled filenames like READMD.md in ha/ and kubernetes/ and practicase in the best-practice chapter. The links still resolve because the table of contents uses the same typos, but nobody has tidied the tree.