v1.8.0

Added

• Zero-Downtime Rolling Restarts & Traffic Draining Engine (internal/rolling/engine.go, internal/rolling/types.go):
  • Staggered sequential container restarts across replicas with configurable concurrency batch size (1 to 10).
  • Multi-strategy target resolution supporting specific container ID lists, Docker Compose stacks (com.docker.compose.project), and arbitrary label selectors (key=val).
  • Active execution tracking with cancellable context trees, manual abort triggers, and automatic progress timeline updates.
• Graceful Traffic Draining & Reverse Proxy Integration (internal/rolling/drain.go, internal/rolling/drain_test.go):
  • In-flight request evacuation with configurable countdown timeouts (0 to 300 seconds).
  • Outgoing HTTP webhook notifications dispatching drain and undrain signals to reverse proxies (Nginx, Traefik, Envoy, HAProxy) with container ID token substitution.
• Multi-Stage Container Readiness Prober (internal/rolling/probe.go, internal/rolling/probe_test.go):
  • Deep container inspection ensuring containers transition to running state and exit code 0.
  • Native Docker health check monitoring waiting for (healthy) and fast-aborting on (unhealthy).
  • HTTP readiness probes querying configurable application endpoints (/healthz) on target ports.
  • Configurable post-readiness grace periods confirming container stability before advancing to subsequent batches.
• Automated Failure Abort & Rollback Mechanism (internal/rolling/engine.go, internal/rolling/store.go):
  • Instant halt of rolling pipeline upon reaching the failure threshold to safeguard healthy running replicas.
  • Automated rollback triggering container recovery and audit trail event generation.
  • High-priority alert notification dispatch to Discord and Slack webhook channels.
• Rolling Restart REST APIs (internal/api/rolling.go, internal/api/rolling_test.go):
  • POST /api/rolling/start: Launches a rolling restart operation with parameter validation (requires RoleAdmin).
  • GET /api/rolling/status: Returns current engine statistics and active operations in flight (requires RoleViewer).
  • GET /api/rolling/operations: Lists historical operations and provides detailed step timelines (?id=, ?limit=).
  • POST /api/rolling/abort: Cancels an active rolling restart execution (requires RoleAdmin).
• Prometheus Metrics Exporter Extension (internal/metrics/rolling.go, internal/metrics/prometheus.go):
  • Exposes dock_radar_rolling_operations_total, dock_radar_rolling_operations_active, dock_radar_rolling_containers_restarted_total, dock_radar_rolling_containers_failed_total, and dock_radar_rolling_rollbacks_total metrics for Grafana dashboards.
• Embedded Dark-Mode Dashboard Glassmorphism UI (web/dist/index.html):
  • Zero-Downtime Rolling Restarts card with summary KPI counters and real-time execution progress bar.
  • Interactive launcher modal with strategy selector, batch size slider, drain timeout, HTTP probe settings, and auto-rollback toggle.
  • Active container step indicators (Pending ➔ Draining ➔ Restarting ➔ Probing ➔ Healthy).
  • Historical execution audit table with one-click manual abort triggers.