Infrastructure code is only half of the system; the tool also keeps a record of what it created. This guide explains state, plans, locks, drift, secrets, tests, and rollback, with Terraform examples and the output of local runs.
A new service team needs somewhere safe to run code. This guide builds the account hierarchy, identity model, permission layers, and baselines that let a platform team hand out that place, then shows how to audit it.
Diagnose a timeout by its step: DNS answers and caching, certificate verification, routing, load balancing, retry amplification, deadlines and packet-level evidence.
Learn to read how a process ended, tell CPU, memory and I/O problems apart, understand limits and buffered logs, and reproduce failures with a small harness.
Start from the consumer’s task, not the mechanism: define the capability boundary, assign ownership, write a checkable self-service contract, and validate it with a runnable example.