SDLC blueprint

Handbook

DevOps — major processes & flow maps

```mermaid

DevOps — major processes & flow maps

1. CI/CD pipeline (conceptual)

flowchart LR CO[Commit] --> BU[Build] BU --> UT[Unit tests] UT --> SA[Static analysis] SA --> IT[Integration tests] IT --> SEC[Security scan] SEC --> STG[Stage deploy] STG --> AT[Acceptance tests] AT --> PRD[Production deploy] PRD --> MON[Monitor]

2. DevOps infinity loop

flowchart LR PL[Plan] --> CD[Code] CD --> BU[Build] BU --> TE[Test] TE --> RE[Release] RE --> DP[Deploy] DP --> OP[Operate] OP --> MN[Monitor] MN --> PL

3. Incident response flow

flowchart TD AL[Alert / detection] --> TR[Triage] TR --> SEV{Severity?} SEV -->|High| IC[Incident commander assigned] SEV -->|Low| OC[On-call resolves] IC --> DG[Diagnose] OC --> DG DG --> FX[Fix / mitigate] FX --> VR[Verify resolution] VR --> PM[Post-mortem]

4. Deployment strategy decision

flowchart TD DS{Deployment strategy?} DS -->|Low risk| RO[Rolling update] DS -->|Medium risk| CN[Canary release] DS -->|High risk| BG[Blue-green deploy] DS -->|Feature control| FF[Feature flag] RO --> MN[Monitor] CN --> MN BG --> MN FF --> MN MN --> RB{Issues?} RB -->|Yes| RBK[Rollback] RB -->|No| OK[Stable]

5. Phases A–F (DevOps emphasis)

Blueprint phase DevOps emphasis Key practice
A Shape Include operability requirements (SLOs, monitoring needs) Shift-left on operational thinking
B Plan Pipeline design; deployment strategy; environment planning IaC from the start
C Build CI: automated build, test, scan on every commit Trunk-based development; small, frequent commits
D Verify Automated test pyramid; security scanning; performance testing Quality gates in the pipeline
E Release CD: automated deployment; canary/blue-green; feature flags Make deployments boring
F Learn Monitoring, alerting, incident response, post-mortems Production feedback closes the loop

6. DORA metrics collection

flowchart LR VC[Version control] --> DF[Deployment frequency] VC --> LT[Lead time for changes] PRD[Production monitoring] --> CFR[Change failure rate] PRD --> MTTR[Mean time to restore] DF --> DB[DORA dashboard] LT --> DB CFR --> DB MTTR --> DB

7. Flow details (walkthrough)

CI/CD pipeline — Every commit triggers an automated pipeline. Build, test, scan, deploy in sequence. Failures at any stage block progression. The pipeline is the single source of truth for deployment readiness.

Infinity loop — DevOps is not a linear process; it is a continuous cycle. Planning incorporates operational learnings. Code includes infrastructure. Testing includes production-like environments. Monitoring informs the next planning cycle.

Incident response — Structured, not ad-hoc. Severity determines response level. Incident commanders coordinate complex responses. Post-mortems extract learning from every significant incident.

Deployment strategies — Choose based on risk tolerance. Rolling updates are simplest. Canary releases test with a subset of traffic. Blue-green provides instant rollback. Feature flags decouple deployment from user-visible release.

8. Authoritative sources & further reading

Full curated list: REFERENCE-LINKS.md.

Canonical source

Edit https://github.com/autowww/blueprints/blob/main/sdlc/methodologies/devops/process-and-flows.md first; regenerate with docs/build-handbook.py.