feat: Add Ansible playbook to restart Traefik and verify configuration

- Add restart-traefik.yml playbook to restart Traefik container
- Verify Traefik health after restart
- Check for ACME challenge errors in logs
- Display status summary with next steps
- Useful after Traefik configuration changes

deployment/ansible/playbooks/restart-traefik.yml

@@ -0,0 +1,126 @@
+---
+- name: Restart Traefik and Verify Configuration
+  hosts: production
+  gather_facts: yes
+  become: no
+
+  vars:
+    traefik_stack_path: "{{ stacks_base_path | default('/home/deploy/deployment/stacks') }}/traefik"
+    traefik_container_name: "traefik"
+    traefik_url: "https://traefik.michaelschiemer.de"
+
+  tasks:
+    - name: Check if Traefik stack directory exists
+      stat:
+        path: "{{ traefik_stack_path }}"
+      register: traefik_stack_exists
+
+    - name: Fail if Traefik stack directory does not exist
+      fail:
+        msg: "Traefik stack directory not found at {{ traefik_stack_path }}"
+      when: not traefik_stack_exists.stat.exists
+
+    - name: Check Traefik container status before restart
+      shell: |
+        cd {{ traefik_stack_path }}
+        docker compose ps {{ traefik_container_name }} --format json
+      register: traefik_status_before
+      changed_when: false
+      failed_when: false
+
+    - name: Display Traefik status before restart
+      debug:
+        msg: |
+          ================================================================================
+          Traefik Container Status (Before Restart):
+          {{ traefik_status_before.stdout | default('Container not found or Docker not running') }}
+          ================================================================================
+
+    - name: Restart Traefik container
+      shell: |
+        cd {{ traefik_stack_path }}
+        docker compose restart {{ traefik_container_name }}
+      register: traefik_restart
+      changed_when: traefik_restart.rc == 0
+
+    - name: Wait for Traefik to be ready
+      wait_for:
+        timeout: 30
+      changed_when: false
+
+    - name: Check Traefik container status after restart
+      shell: |
+        cd {{ traefik_stack_path }}
+        docker compose ps {{ traefik_container_name }} --format json
+      register: traefik_status_after
+      changed_when: false
+      failed_when: false
+
+    - name: Check Traefik health endpoint
+      ansible.builtin.uri:
+        url: "{{ traefik_url }}/ping"
+        method: GET
+        status_code: [200]
+        validate_certs: no
+        timeout: 10
+      register: traefik_health
+      ignore_errors: yes
+      changed_when: false
+
+    - name: Get Traefik logs (last 50 lines)
+      shell: |
+        cd {{ traefik_stack_path }}
+        docker compose logs --tail=50 {{ traefik_container_name }}
+      register: traefik_logs
+      changed_when: false
+      failed_when: false
+      ignore_errors: yes
+
+    - name: Check for ACME challenge errors in logs
+      shell: |
+        cd {{ traefik_stack_path }}
+        docker compose logs {{ traefik_container_name }} 2>&1 | grep -i "acme challenge" | tail -10 || echo "No ACME challenge errors found"
+      register: acme_errors
+      changed_when: false
+      failed_when: false
+
+    - name: Display Traefik logs
+      debug:
+        msg: |
+          ================================================================================
+          Traefik Container Logs (last 50 lines):
+          {{ traefik_logs.stdout | default('No logs available') }}
+          ================================================================================
+
+    - name: Display ACME challenge status
+      debug:
+        msg: |
+          ================================
+          ACME Challenge Status:
+          {{ acme_errors.stdout | default('No ACME errors found in recent logs') }}
+          ================================
+
+    - name: Display final status
+      debug:
+        msg: |
+          ========================================
+          ========================================
+          Traefik Restart Summary
+          ========================================
+          Container Status: {% if 'State":"running' in (traefik_status_after.stdout | default('')) %}✅ RUNNING{% else %}❌ NOT RUNNING{% endif %}
+          Health Check: {% if traefik_health.status | default(0) == 200 %}✅ HEALTHY{% else %}❌ UNHEALTHY or TIMEOUT{% endif %}
+          Restart Action: {% if traefik_restart.changed | default(false) %}🔄 Container restarted{% else %}ℹ️ No restart needed{% endif %}
+          ========================================
+          {% if 'State":"running' in (traefik_status_after.stdout | default('')) and traefik_health.status | default(0) == 200 %}
+          ✅ Traefik is running and healthy!
+          
+          Next steps:
+          1. Monitor logs for ACME challenge errors: tail -f {{ traefik_stack_path }}/logs/traefik.log | grep -i acme
+          2. Check certificate status in Traefik dashboard: {{ traefik_url }}
+          3. Wait for next certificate renewal attempt (usually hourly)
+          {% else %}
+          ❌ Traefik is not fully healthy. Check logs for details:
+          docker logs {{ traefik_container_name }}
+          {% endif %}
+          ========================================
+

deployment/gitea-runner/FIX_RUNNER_CONFIG.md

@@ -90,3 +90,17 @@ cd deployment/gitea-runner
 - Prüfe Traefik-Logs: `docker logs traefik`
 - Stelle sicher, dass Gitea nicht überlastet ist
 
+**Runner überlastet Gitea (DDoS-ähnliches Verhalten):**
+- Der Runner kann Gitea mit zu vielen Anfragen überlasten, besonders bei Fehlern
+- **Lösung**: `fetch_interval` in `config.yaml` erhöhen (Standard: 2s → Empfohlen: 10s oder mehr)
+- **Lösung**: `fetch_timeout` erhöhen (Standard: 5s → Empfohlen: 30s)
+- **Sofortmaßnahme**: Runner deaktivieren/stoppen, bis Gitea wieder stabil ist:
+  ```bash
+  cd deployment/gitea-runner
+  docker compose stop gitea-runner
+  ```
+- Nach Anpassung der Konfiguration Runner neu starten:
+  ```bash
+  docker compose up -d gitea-runner
+  ```
+

deployment/gitea-runner/config.yaml

@@ -18,10 +18,11 @@ runner:
   insecure: true
 
   # Timeout for fetching job from Gitea
-  fetch_timeout: 5s
+  fetch_timeout: 30s
 
-  # Interval for fetching jobs
+  # Interval for fetching jobs (increased to reduce load on Gitea)
-  fetch_interval: 2s
+  # Lower values cause DDoS-like behavior when Gitea is slow or overloaded
+  fetch_interval: 10s
 
 cache:
   # Enable cache server

deployment/stacks/traefik/docker-compose.yml

@@ -5,16 +5,23 @@ services:
     restart: unless-stopped
     security_opt:
       - no-new-privileges:true
-    # Use host network mode to correctly identify client IPs from WireGuard
+    # Use bridge network mode for reliable service discovery
-    # Without this, Traefik sees Docker bridge IPs instead of real client IPs (10.8.0.x)
+    # Service discovery works correctly with Docker labels in bridge mode
-    network_mode: host
+    ports:
-    # When using host network mode, we don't bind ports in docker-compose
+      - "80:80"
-    # Traefik listens directly on host ports 80 and 443
+      - "443:443"
-    # ports:
+      - "2222:2222"  # Gitea SSH
-    #   - "80:80"
+    networks:
-    #   - "443:443"
+      - traefik-public
     environment:
       - TZ=Europe/Berlin
+    command:
+      # Load static configuration file
+      - "--configFile=/traefik.yml"
+      # Increase timeouts for slow backends like Gitea
+      - "--entrypoints.websecure.transport.respondingTimeouts.readTimeout=300s"
+      - "--entrypoints.websecure.transport.respondingTimeouts.writeTimeout=300s"
+      - "--entrypoints.websecure.transport.respondingTimeouts.idleTimeout=360s"
     volumes:
       # Docker socket for service discovery
       - /var/run/docker.sock:/var/run/docker.sock:ro
@@ -30,47 +37,30 @@ services:
       # Enable Traefik for itself
       - "traefik.enable=true"
 
-      # Dashboard - VPN-only access (WireGuard network required)
+      # Dashboard - BasicAuth protected
-      # Accessible only from WireGuard VPN network (10.8.0.0/24)
       - "traefik.http.routers.traefik-dashboard.rule=Host(`traefik.michaelschiemer.de`)"
       - "traefik.http.routers.traefik-dashboard.entrypoints=websecure"
       - "traefik.http.routers.traefik-dashboard.tls=true"
       - "traefik.http.routers.traefik-dashboard.tls.certresolver=letsencrypt"
       - "traefik.http.routers.traefik-dashboard.service=api@internal"
-      # VPN-only + BasicAuth protection (order: vpn-only first, then BasicAuth)
+      - "traefik.http.routers.traefik-dashboard.middlewares=traefik-auth"
-      - "traefik.http.routers.traefik-dashboard.middlewares=vpn-only@file,traefik-auth"
 
-      # BasicAuth for dashboard (user: admin, password: generate with htpasswd)
+      # BasicAuth for dashboard
-      # htpasswd -nb admin your_password
+      - "traefik.http.middlewares.traefik-auth.basicauth.users=admin:$$apr1$$Of2wG3O5$$y8X1vEoIp9vpvx64mIalk/"
-      - "traefik.http.middlewares.traefik-auth.basicauth.users=admin:$$apr1$$8kj9d7lj$$r.x5jhLVPLuCDLvJ6x0Hd0"
 
-      # Allow ACME challenges without redirect (higher priority)
+      # Note: ACME challenges are handled automatically by Traefik
-      - "traefik.http.routers.acme-challenge.rule=PathPrefix(`/.well-known/acme-challenge`)"
+      # when httpChallenge.entryPoint: web is set in traefik.yml
-      - "traefik.http.routers.acme-challenge.entrypoints=web"
+      # No explicit router needed - Traefik handles /.well-known/acme-challenge automatically
-      - "traefik.http.routers.acme-challenge.priority=200"
       
       # Global redirect to HTTPS (lower priority, matches everything else)
-      - "traefik.http.routers.http-catchall.rule=HostRegexp(`{host:.+}`)"
+      # ACME challenges are excluded from redirect automatically by Traefik
+      # Explicitly exclude ACME challenge path to be safe
+      - "traefik.http.routers.http-catchall.rule=HostRegexp(`{host:.+}`) && !PathPrefix(`/.well-known/acme-challenge`)"
       - "traefik.http.routers.http-catchall.entrypoints=web"
       - "traefik.http.routers.http-catchall.middlewares=redirect-to-https"
       - "traefik.http.routers.http-catchall.priority=1"
       - "traefik.http.middlewares.redirect-to-https.redirectscheme.scheme=https"
       - "traefik.http.middlewares.redirect-to-https.redirectscheme.permanent=true"
-
-      # Security headers middleware
-      - "traefik.http.middlewares.security-headers.headers.frameDeny=true"
-      - "traefik.http.middlewares.security-headers.headers.contentTypeNosniff=true"
-      - "traefik.http.middlewares.security-headers.headers.browserXssFilter=true"
-      - "traefik.http.middlewares.security-headers.headers.stsSeconds=31536000"
-      - "traefik.http.middlewares.security-headers.headers.stsIncludeSubdomains=true"
-      - "traefik.http.middlewares.security-headers.headers.stsPreload=true"
-
-      # Compression middleware
-      - "traefik.http.middlewares.compression.compress=true"
-
-      # Rate limiting middleware (100 requests per second)
-      - "traefik.http.middlewares.rate-limit.ratelimit.average=100"
-      - "traefik.http.middlewares.rate-limit.ratelimit.burst=50"
     healthcheck:
       test: ["CMD", "traefik", "healthcheck", "--ping"]
       interval: 30s
@@ -78,5 +68,6 @@ services:
       retries: 3
       start_period: 10s
 
-# Note: network_mode: host is used, so we don't define networks here
+networks:
-# Traefik still discovers services via Docker labels using the Docker socket
+  traefik-public:
+    external: true

deployment/stacks/traefik/traefik.yml

@@ -1,4 +1,5 @@
-# Static Configuration for Traefik
+# Static Configuration for Traefik v3.0
+# Minimal configuration - only static settings
 
 # Global Configuration
 global:
@@ -6,12 +7,10 @@ global:
   sendAnonymousUsage: false
 
 # API and Dashboard
-# Note: insecure: false means API is only accessible via HTTPS (through Traefik itself)
-# No port 8080 needed - dashboard accessible via HTTPS at traefik.michaelschiemer.de
 api:
   dashboard: true
   insecure: false
-  # Dashboard accessible via HTTPS router (no separate HTTP listener needed)
+  # Dashboard accessible via HTTPS router
 
 # Entry Points
 entryPoints:
@@ -42,20 +41,20 @@ certificatesResolvers:
       storage: /acme.json
       caServer: https://acme-v02.api.letsencrypt.org/directory
       # Use HTTP-01 challenge (requires port 80 accessible)
+      # Traefik automatically handles /.well-known/acme-challenge requests
       httpChallenge:
         entryPoint: web
-      # Uncomment for DNS challenge (requires DNS provider)
+      # Optional: Increase retry attempts for certificate renewal
-      # dnsChallenge:
+      # This helps when Gitea is temporarily unavailable
-      #   provider: cloudflare
+      preferredChain: ""
-      #   delayBeforeCheck: 30
 
 # Providers
 providers:
   docker:
     endpoint: "unix:///var/run/docker.sock"
     exposedByDefault: false
-    # Network mode is 'host', so we don't specify a network here
+    # Bridge network mode - Traefik uses Docker service discovery via labels
-    # Traefik can reach containers directly via their IPs in host network mode
+    # Services must be on the same network (traefik-public) for discovery
     watch: true
 
   file:
@@ -63,12 +62,12 @@ providers:
     watch: true
 
 # Forwarded Headers Configuration
-# This ensures Traefik correctly identifies the real client IP
 forwardedHeaders:
   trustedIPs:
-    - "127.0.0.1/32"     # Localhost
+    - "127.0.0.1/32"
-    - "172.17.0.0/16"    # Docker bridge network
+    - "172.17.0.0/16"
-    - "172.18.0.0/16"    # Docker user-defined networks
+    - "172.18.0.0/16"
+    - "10.8.0.0/24"
   insecure: false
 
 # Logging