fix: Gitea Traefik routing and connection pool optimization

- Remove middleware reference from Gitea Traefik labels (caused routing issues)
- Optimize Gitea connection pool settings (MAX_IDLE_CONNS=30, authentication_timeout=180s)
- Add explicit service reference in Traefik labels
- Fix intermittent 504 timeouts by improving PostgreSQL connection handling

Fixes Gitea unreachability via git.michaelschiemer.de

deployment/ansible/playbooks/monitor-traefik-continuously.yml

@@ -0,0 +1,141 @@
+---
+# Monitor Traefik Continuously
+# Überwacht Traefik-Logs und Docker Events in Echtzeit um Restart-Quelle zu finden
+- name: Monitor Traefik Continuously
+  hosts: production
+  gather_facts: yes
+  become: no
+  vars:
+    traefik_stack_path: "{{ stacks_base_path }}/traefik"
+    monitor_duration_minutes: 30  # Standard: 30 Minuten, kann überschrieben werden
+
+  tasks:
+    - name: Display monitoring information
+      ansible.builtin.debug:
+        msg: |
+          ================================================================================
+          TRAEFIK CONTINUOUS MONITORING
+          ================================================================================
+          
+          Überwachungsdauer: {{ monitor_duration_minutes }} Minuten
+          
+          Überwacht:
+          1. Traefik-Logs auf "Stopping server gracefully" / "I have to go"
+          2. Docker Events für Traefik-Container
+          3. Docker Daemon Logs für Container-Stops
+          
+          Starte Monitoring...
+          ================================================================================
+
+    - name: Get initial Traefik status
+      ansible.builtin.shell: |
+        docker inspect traefik --format '{{ '{{' }}.State.Status{{ '}}' }} {{ '{{' }}.State.StartedAt{{ '}}' }}' 2>/dev/null || echo "UNKNOWN"
+      register: initial_status
+      changed_when: false
+
+    - name: Start monitoring Traefik logs in background
+      ansible.builtin.shell: |
+        cd {{ traefik_stack_path }}
+        timeout {{ monitor_duration_minutes * 60 }} docker compose logs -f traefik 2>&1 | grep --line-buffered -iE "stopping server gracefully|I have to go" | while read line; do
+          echo "[$(date '+%Y-%m-%d %H:%M:%S')] $line"
+        done > /tmp/traefik_monitor_$$.log 2>&1 &
+        echo $!
+      register: log_monitor_pid
+      changed_when: false
+      async: "{{ monitor_duration_minutes * 60 + 60 }}"
+      poll: 0
+
+    - name: Start monitoring Docker events in background
+      ansible.builtin.shell: |
+        timeout {{ monitor_duration_minutes * 60 }} docker events --filter container=traefik --filter event=die --format "[{{ '{{' }}.Time{{ '}}' }}] {{ '{{' }}.Action{{ '}}' }} {{ '{{' }}.Actor.Attributes.name{{ '}}' }}" 2>&1 | tee /tmp/traefik_docker_events_$$.log &
+        echo $!
+      register: docker_events_pid
+      changed_when: false
+      async: "{{ monitor_duration_minutes * 60 + 60 }}"
+      poll: 0
+
+    - name: Wait for monitoring period
+      ansible.builtin.pause:
+        minutes: "{{ monitor_duration_minutes }}"
+
+    - name: Stop log monitoring
+      ansible.builtin.shell: |
+        pkill -f "docker compose logs.*traefik" || true
+        sleep 2
+      changed_when: false
+      failed_when: false
+
+    - name: Stop Docker events monitoring
+      ansible.builtin.shell: |
+        pkill -f "docker events.*traefik" || true
+        sleep 2
+      changed_when: false
+      failed_when: false
+
+    - name: Read Traefik log monitoring results
+      ansible.builtin.slurp:
+        src: "{{ item }}"
+      register: log_results
+      changed_when: false
+      failed_when: false
+      loop: "{{ log_monitor_pid.stdout_lines | map('regex_replace', '^.*', '/tmp/traefik_monitor_' + ansible_date_time.epoch + '.log') | list }}"
+
+    - name: Read Docker events monitoring results
+      ansible.builtin.slurp:
+        src: "{{ item }}"
+      register: docker_events_results
+      changed_when: false
+      failed_when: false
+      loop: "{{ docker_events_pid.stdout_lines | map('regex_replace', '^.*', '/tmp/traefik_docker_events_' + ansible_date_time.epoch + '.log') | list }}"
+
+    - name: Get final Traefik status
+      ansible.builtin.shell: |
+        docker inspect traefik --format '{{ '{{' }}.State.Status{{ '}}' }} {{ '{{' }}.State.StartedAt{{ '}}' }} {{ '{{' }}.RestartCount{{ '}}' }}' 2>/dev/null || echo "UNKNOWN"
+      register: final_status
+      changed_when: false
+
+    - name: Check Traefik logs for stop messages during monitoring
+      ansible.builtin.shell: |
+        cd {{ traefik_stack_path }}
+        docker compose logs traefik --since {{ monitor_duration_minutes }}m 2>&1 | grep -iE "stopping server gracefully|I have to go" || echo "Keine Stop-Meldungen gefunden"
+      register: traefik_stop_messages
+      changed_when: false
+      failed_when: false
+
+    - name: Summary
+      ansible.builtin.debug:
+        msg: |
+          ================================================================================
+          MONITORING ZUSAMMENFASSUNG ({{ monitor_duration_minutes }} Minuten):
+          ================================================================================
+          
+          Initial Status: {{ initial_status.stdout }}
+          Final Status: {{ final_status.stdout }}
+          
+          Traefik Stop-Meldungen während Monitoring:
+          {% if traefik_stop_messages.stdout and 'Keine Stop-Meldungen' not in traefik_stop_messages.stdout %}
+          ❌ STOP-MELDUNGEN GEFUNDEN:
+          {{ traefik_stop_messages.stdout }}
+          
+          ⚠️  PROBLEM BESTÄTIGT: Traefik wurde während des Monitorings gestoppt!
+          
+          Nächste Schritte:
+          1. Prüfe Docker Events Log: /tmp/traefik_docker_events_*.log
+          2. Prüfe Traefik Log Monitor: /tmp/traefik_monitor_*.log
+          3. Prüfe wer den Stop-Befehl ausgeführt hat:
+             - journalctl -u docker.service --since "{{ monitor_duration_minutes }} minutes ago"
+             - docker events --since "{{ monitor_duration_minutes }} minutes ago" --filter container=traefik
+          {% else %}
+          ✅ KEINE STOP-MELDUNGEN GEFUNDEN
+          
+          Traefik lief stabil während des {{ monitor_duration_minutes }}-minütigen Monitorings.
+          
+          {% if initial_status.stdout != final_status.stdout %}
+          ⚠️  Status hat sich geändert:
+          - Vorher: {{ initial_status.stdout }}
+          - Nachher: {{ final_status.stdout }}
+          {% endif %}
+          {% endif %}
+          
+          ================================================================================
+