Initial public release — Inspector v0.7.0

Docker-watchdog with per-container restart rules, Gotify/Pushover
notifications, and an embedded dashboard + web settings page.

See README.md for an overview and INSTALL.md for the Docker quickstart.

config.example.yml

@@ -0,0 +1,137 @@
+# Inspector example configuration.
+# Copy to config.yml, edit, and mount into the container at /app/config.yml.
+
+inspector:
+  interval_seconds: 60
+  log_level: INFO
+  dry_run: false
+  log_dir: /app/logs
+  log_filename: inspector.log
+  log_retention_days: 30
+  heartbeat: false
+
+notifications:
+  # Gotify: https://gotify.net
+  gotify:
+    enabled: false
+    url: ""
+    token: ""
+    priority: 5
+
+  # Pushover: https://pushover.net
+  # Independent of Gotify — enable either, both, or neither. If both are
+  # enabled, every alert is sent to both. Inspector's priority is an
+  # integer on a Gotify-style scale (default 5) and is mapped to
+  # Pushover's -2..2 range at send time: <=2 -> -1, 3..4 -> 0,
+  # 5..6 -> 1, >=7 -> 2.
+  pushover:
+    enabled: false
+    token: ""           # Pushover application API token
+    user_key: ""        # Pushover user or group key
+    priority: 5
+    device: ""          # optional target device; blank = all devices
+    sound: ""           # optional sound name; blank = user default
+
+# Persist a durable per-container event history to a local SQLite DB.
+# Enables the /events page and the Health summary digest.
+database:
+  enabled: false
+  path: /app/data/events.db
+  retention_days: 30
+
+# Periodic digest of per-container events, delivered through the enabled
+# notification channels. Requires database.enabled: true to have data.
+health_summary:
+  enabled: false
+  interval_hours: 24
+  priority: 3
+
+# Recurring daily maintenance window. Containers that opt in with
+# `maintenance_window: true` are not acted on during this window.
+maintenance:
+  enabled: false
+  start: "00:00"
+  end: "06:00"
+  dir: /app/data/maintenance
+  max_hours: 8
+
+# Manual notification muting per container (toggled from the dashboard).
+notification_mute:
+  enabled: false
+  dir: /app/data/muted
+  default_hours: 12
+
+# Embedded web dashboard. Enable with a non-empty password.
+# Visit http://<host>:<port>/ for status, /events for history,
+# /settings to edit this file from the browser.
+dashboard:
+  enabled: false
+  host: 0.0.0.0
+  port: 8080
+  username: admin
+  password: ""
+
+# Optional external integrations.
+integrations:
+  # Uptime Kuma: optionally use a public status page as an extra
+  # reachability check per container.
+  uptime_kuma:
+    enabled: false
+    url: "https://status.example.com"
+    status_page_slug: "default"
+
+  # Portainer API integration: used by the advanced recovery actions
+  # `portainer_stack_redeploy` and `restart_then_redeploy`, which can
+  # recover containers whose network namespace is tied to a VPN sidecar.
+  # Leave disabled unless you need it.
+  #
+  # Token options (pick one per environment):
+  #   api_token_env: "ENV_VAR_NAME"  — reads token from that env var at
+  #                                    runtime (preferred; not stored in config)
+  #   api_token: "ptr_..."           — embed token directly in this file
+  portainer:
+    enabled: false
+    default_environment: local_docker
+    stack_redeploy_cooldown_seconds: 300
+    post_redeploy_wait_seconds: 90
+    environments:
+      local_docker:
+        base_url: "http://portainer.example.local:9000"
+        api_token_env: "PORTAINER_API_TOKEN"
+        endpoint_id: 1
+        verify_ssl: false
+
+containers:
+  # Minimal entry: Inspector only acts on containers listed here.
+  # Each key is the Docker container name on your host.
+  example-web:
+    enabled: true
+    max_error_count: 5
+    cooldown_seconds: 300
+    grace_period_seconds: 120
+    max_restarts_per_hour: 3
+    restart_on:
+      - UNHEALTHY
+      - NOT RUNNING
+      - ERROR
+
+  # Example with an opt-in HTTP reachability check. Useful when Docker
+  # reports a container as healthy but it's not actually serving
+  # requests (common with VPN-routed containers whose tunnel drops).
+  example-api:
+    enabled: true
+    max_error_count: 5
+    cooldown_seconds: 300
+    grace_period_seconds: 120
+    max_restarts_per_hour: 3
+    color: "#60a5fa"
+    restart_on:
+      - UNHEALTHY
+      - NOT RUNNING
+      - ERROR
+      - UNREACHABLE
+    http_check:
+      enabled: true
+      url: "http://example-api:8080/health"
+      timeout_seconds: 5
+      expected_status: 200