Understanding GitLab CI
summary
Explains how GitLab CI runs on runners with executors (shell, Docker, Kubernetes) and why choosing the right executor and image matters: switching to Alpine cuts a test run from about 5 minutes to 3 minutes. Covers local debugging with runner exec, Docker-in-Docker tradeoffs and using --cache-from to speed builds, cache limitations because GitLab cache stores zip archives that degrade past ~100MB, artifact vs cache misuse, stage gating and scaling to handle peaks of up to 2,000 concurrent jobs.
AI-written summary · about
chapters (10)
- 0:00Start
- 3:27PIPELINES
- 7:53What is the executor?
- 11:57gitlab-runner exec
- 20:10Parallelism
- 20:31Scale vertically
- 25:35Cache vs Artifacts
- 27:21How we can docker build?
- 30:38Enable Docker-in-Docker
- 31:33Maybe docker
related talks
35mGitLab CI and Kubernetes, Build Test and Deploy Your Projects
26mOptimizing Your CI Pipelines
22mEmbedded Software Development Efficiency Improvement with Docker and Kubernetes
36mPractical Tips for Build Optimization for GitLab and Docker Users
38mFrom dev to prod with GitLab CI
32mEnhance Your Kubernetes CI/CD Pipelines With Open Source
15mDocker image testing in GitLab CI
35mGitlab Runners at Scale as a Service