DevOps & Release Engineering¶

Grite integrates with CI/CD workflows and release processes.

Why Grite for DevOps?¶

CI-friendly: Non-interactive CLI with JSON output
Git-native: Fits existing git workflows
Traceable: All changes in append-only log
Coordinated: Distributed locks for deployments

CI Failure Tracking¶

Track CI failures and their resolution:

# CI system creates issue on failure
grite issue create --title "CI Failure: test_integration_auth [build #1234]" \
  --body "$(cat <<'EOF'
## Build Info
- Build: #1234
- Branch: feature/oauth
- Commit: abc123

## Failure

test_integration_auth::test_login_success FAILED thread panicked at 'assertion failed: response.status == 200'

## Logs
[Link to full logs]
EOF
)" --label "ci-failure" --label "test"

# Developer investigates and resolves
grite issue comment $ISSUE_ID --body "Flaky test - added retry logic"
grite issue close $ISSUE_ID

CI Integration Script¶

#!/bin/bash
# ci-failure-reporter.sh

if [ "$CI_STATUS" = "failed" ]; then
  grite issue create \
    --title "CI Failure: $CI_JOB_NAME [build #$CI_BUILD_NUMBER]" \
    --body "Build: #$CI_BUILD_NUMBER
Branch: $CI_BRANCH
Commit: $CI_COMMIT

Error: $CI_ERROR_MESSAGE" \
    --label "ci-failure" \
    --json

  grite sync --push
fi

Release Checklist Management¶

Track release checklist items that sync across the team:

grite issue create --title "Release v2.0.0 checklist" \
  --body "$(cat <<'EOF'
## Pre-release
- [ ] All PRs merged
- [ ] Version bumped in Cargo.toml
- [ ] CHANGELOG.md updated
- [ ] Release notes drafted
- [ ] Security scan passed

## Release
- [ ] Tag created: v2.0.0
- [ ] Binaries built
- [ ] Docker images pushed
- [ ] GitHub release published

## Post-release
- [ ] Documentation site updated
- [ ] Announcement posted
- [ ] Monitor error rates for 24h
- [ ] Close release milestone
EOF
)" --label "release" --label "v2.0.0"

Release Workflow¶

# Start release
grite issue create --title "Release v2.0.0" --label "release"
grite lock acquire --resource "repo:release" --ttl 2h

# Update as you go
grite issue comment $RELEASE_ID --body "Tag created"
grite issue comment $RELEASE_ID --body "Binaries uploaded"

# Complete release
grite issue close $RELEASE_ID
grite lock release --resource "repo:release"
grite sync --push

Deployment Coordination¶

Track deployment readiness across environments:

grite issue create --title "Deploy v2.0.0 to production" \
  --body "$(cat <<'EOF'
## Deployment Plan

### Stage 1: Staging
- [x] Deploy to staging
- [x] Run smoke tests
- [x] QA sign-off

### Stage 2: Canary
- [ ] Deploy to 5% of production
- [ ] Monitor metrics for 1h
- [ ] Verify no error spike

### Stage 3: Full Rollout
- [ ] Deploy to remaining 95%
- [ ] Monitor metrics for 4h
- [ ] Update status page

## Rollback Plan
If error rate >1%: `kubectl rollout undo deployment/api`

## Contacts
- Release lead: @alice
- On-call: @bob
EOF
)" --label "deployment" --label "production"

Deployment Lock¶

Prevent concurrent deployments:

# Acquire deployment lock
if ! grite lock acquire --resource "deploy:production" --ttl 1h; then
  echo "Deployment already in progress"
  exit 1
fi

# Deploy
./deploy.sh

# Release lock
grite lock release --resource "deploy:production"

Breaking Change Tracking¶

Track breaking changes and migration status:

grite issue create --title "Breaking change: API v1 deprecation" \
  --body "$(cat <<'EOF'
## Change
API v1 endpoints will be removed in v3.0.0.

## Migration Guide
Replace `/api/v1/*` with `/api/v2/*`. See docs/migration-v2.md.

## Consumer Status
| Consumer | Status | Contact |
|----------|--------|---------|
| Mobile app | Migrated | @mobile-team |
| Partner A | In progress | [email protected] |
| Partner B | Not started | [email protected] |
| Internal dashboard | Migrated | @frontend |

## Timeline
- Deprecation notice: 2024-01-01
- Final warning: 2024-03-01
- Removal: 2024-04-01
EOF
)" --label "breaking-change" --label "api"

Feature Flag Coordination¶

Track feature flag states and ownership:

grite issue create --title "[Feature Flags] Active flags inventory" \
  --body "$(cat <<'EOF'
## Active Feature Flags

| Flag | Purpose | Owner | Status |
|------|---------|-------|--------|
| `new_checkout_flow` | Redesigned checkout | @alice | 50% rollout |
| `dark_mode` | UI dark theme | @bob | 100% (ready to remove) |
| `experimental_search` | New search algorithm | @carol | Internal only |
| `rate_limit_v2` | New rate limiting | @dave | 10% rollout |

## Cleanup Queue
Flags at 100% for >30 days should be removed:
- [ ] `dark_mode` - remove flag, keep feature
- [ ] `new_login_page` - remove flag, keep feature
EOF
)" --label "feature-flags" --label "inventory"

Infrastructure Tracking¶

Track infrastructure changes:

grite issue create --title "Infra: Upgrade Kubernetes to 1.28" \
  --body "$(cat <<'EOF'
## Reason
Security patches and new features needed.

## Plan
1. [ ] Update staging cluster
2. [ ] Run integration tests
3. [ ] Update production cluster (maintenance window)

## Rollback
kubectl can downgrade within 24h window.

## Dependencies
- [ ] Update kubectl locally
- [ ] Update CI runners
- [ ] Update helm charts
EOF
)" --label "infrastructure" --label "kubernetes"

On-Call Handoff¶

Document on-call status:

grite issue create --title "On-call handoff: Week of 2024-01-15" \
  --body "$(cat <<'EOF'
## Outgoing: @alice
## Incoming: @bob

## Active Issues
1. Elevated error rate on /api/search - monitoring
2. Database slow queries - investigating

## Recent Incidents
- 2024-01-14: Brief API outage, resolved (see incident-123)

## Upcoming Maintenance
- 2024-01-17 02:00 UTC: Database maintenance window

## Notes
- Deploy freeze until 2024-01-18
- New runbook for auth issues: docs/runbooks/auth.md
EOF
)" --label "oncall" --label "handoff"

Monitoring Alerts¶

Track alert status:

grite issue create --title "Alert: High CPU on prod-api-3" \
  --body "$(cat <<'EOF'
## Alert Details
- Host: prod-api-3
- Metric: CPU > 90% for 5m
- Started: 2024-01-15 14:30 UTC

## Investigation
- Checked recent deploys: none
- Checked traffic: normal
- Found: Runaway process in container

## Resolution
Restarted container, CPU normalized.

## Follow-up
- [ ] Add memory limits to container spec
- [ ] Create alert for container restarts
EOF
)" --label "alert" --label "resolved"

CI/CD Pipeline Integration¶

GitHub Actions Example¶

- name: Report CI Failure
  if: failure()
  run: |
    grite issue create \
      --title "CI Failure: ${{ github.job }} [#${{ github.run_number }}]" \
      --body "Workflow: ${{ github.workflow }}
    Branch: ${{ github.ref_name }}
    Commit: ${{ github.sha }}
    Run: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}" \
      --label "ci-failure" \
      --json
    grite sync --push

GitLab CI Example¶

report_failure:
  stage: .post
  when: on_failure
  script:
    - grite issue create --title "CI Failure: $CI_JOB_NAME" --label "ci-failure"
    - grite sync --push

Best Practices¶

Use Locks for Critical Operations¶

# Deployment
grite lock acquire --resource "deploy:$ENV" --ttl 1h

# Database migrations
grite lock acquire --resource "migration:$DB" --ttl 30m

# Release creation
grite lock acquire --resource "repo:release" --ttl 2h

Automate Issue Creation¶

Integrate with your CI/CD to create issues automatically:

On build failure
On deployment start/complete
On alert firing

Keep Issues Updated¶

# Deployment progress
grite issue comment $DEPLOY_ID --body "Canary at 5%"
grite issue comment $DEPLOY_ID --body "Canary healthy, proceeding to 100%"
grite issue comment $DEPLOY_ID --body "Deployment complete"
grite issue close $DEPLOY_ID

Archive Old Issues¶

# Monthly cleanup
grite issue list --state closed --label "ci-failure" | \
  xargs -I {} grite issue label add {} --label "archived"

Next Steps¶

Distributed Locks - Deployment coordination
CLI Reference - CI automation
JSON Output - Parsing in scripts