Add GPU monitoring services and configurations

- Create systemd service templates for llama_exporter, nvidia_exporter, and podman.
- Add Prometheus configuration template for GPU metrics scraping.
- Introduce variables for GPU monitoring role in main.yml.
- Implement tasks for syncing llama models, including user setup and package installation.
- Update podman tasks to ensure proper sudo access and lingering for the podman user.
- Modify inventory to include gpu_monitoring group.
- Add Terraform modules for deploying Vaultwarden, including ingress and service configurations.
- Create Grafana dashboard for real-time GPU monitoring metrics.
- Update secrets and environment files to include Vaultwarden admin token.

ansible/roles/gpu_monitoring/defaults/main.yml

@@ -0,0 +1,68 @@
+---
+# gpu_monitoring defaults
+
+# Prometheus configuration
+prometheus_image: "docker.io/prom/prometheus:latest"
+prometheus_image_tag: "latest"
+prometheus_version: "latest"
+prometheus_port: 9091
+prometheus_container_port: 9090
+prometheus_data_dir: "/mnt/data/prometheus"
+prometheus_retention_days: "30"
+prometheus_retention_size: "10GB"
+prometheus_scrape_interval: "15s"
+prometheus_evaluation_interval: "15s"
+prometheus_scrape_timeout: "10s"
+
+# Prometheus container
+prometheus_container_memory: "1g"
+prometheus_user: prometheus
+prometheus_group: prometheus
+prometheus_user_home: "/mnt/data/prometheus"
+prometheus_network_name: prometheus-net
+prometheus_host_port: "{{ prometheus_port }}"
+prometheus_host_gateway: "host.containers.internal"
+prometheus_extra_groups: []
+prometheus_user_folders:
+  - .prometheus
+  - .cache
+
+# NVIDIA GPU Exporter configuration (NVML-based, works with consumer GPUs)
+nvidia_exporter_enabled: true
+nvidia_exporter_port: 9400
+nvidia_exporter_user: nvidia_exporter
+nvidia_exporter_user_home: "/home/nvidia_exporter"
+nvidia_exporter_install_dir: "/opt/nvidia_exporter"
+nvidia_exporter_venv_path: "/opt/nvidia_exporter/.venv"
+
+# AMD GPU Exporter configuration
+amd_exporter_enabled: true
+amd_exporter_port: 9500
+amd_exporter_user: amd_exporter
+amd_exporter_user_home: "/home/amd_exporter"
+amd_exporter_install_dir: "/opt/amdgpu_exporter"
+amd_exporter_venv_path: "/opt/amdgpu_exporter/.venv"
+amd_exporter_python_version: "python3"
+amd_exporter_sysfs_gpu_pattern: "/sys/class/drm/card?/device/"
+
+# Llama Exporter configuration
+llama_exporter_enabled: true
+llama_exporter_port: 9550
+llama_exporter_user: llama_exporter
+llama_exporter_user_home: "/home/llama_exporter"
+llama_exporter_install_dir: "/opt/llama_exporter"
+llama_exporter_venv_path: "/opt/llama_exporter/.venv"
+llama_exporter_scrape_interval: 15
+llama_exporter_scrape_timeout: 5
+
+# Llama.cpp scrape targets
+llama_exporter_targets: []
+
+# Firewall configuration
+prometheus_fw_port: "{{ prometheus_port }}"
+prometheus_fw_proto: "tcp"
+prometheus_fw_allowed_networks:
+  - cidr: "10.19.4.0/24"
+    comment: "local network"
+  - cidr: "10.5.5.5/32"
+    comment: "Laptop Wireguard"