Merge pull request #111 from lordmathis/fix/cgo-enabled-build

fix: Add multiplatform CGO_ENABLED=1 build

.github/workflows/release.yaml

@@ -45,15 +45,23 @@ jobs:
   build:
     name: Build Binaries
     needs: build-webui
-    runs-on: ubuntu-latest
+    runs-on: ${{ matrix.runner }}
     strategy:
       matrix:
-        goos: [linux, windows, darwin]
-        goarch: [amd64, arm64]
-        exclude:
-          # Windows ARM64 support is limited
-          - goos: windows
+        include:
+          - goos: linux
+            goarch: amd64
+            runner: ubuntu-latest
+          - goos: linux
             goarch: arm64
+            runner: ubuntu-latest
+            cc: aarch64-linux-gnu-gcc
+          - goos: darwin
+            goarch: arm64
+            runner: macos-latest
+          - goos: windows
+            goarch: amd64
+            runner: windows-latest
     steps:
       - name: Checkout repository
         uses: actions/checkout@v4
@@ -70,11 +78,19 @@ jobs:
           name: webui-dist
           path: webui/dist/
 
+      - name: Install cross-compilation tools (Linux ARM64 only)
+        if: matrix.cc != ''
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y gcc-aarch64-linux-gnu
+
       - name: Build binary
         env:
           GOOS: ${{ matrix.goos }}
           GOARCH: ${{ matrix.goarch }}
-          CGO_ENABLED: 0
+          CGO_ENABLED: 1
+          CC: ${{ matrix.cc }}
+        shell: bash
         run: |
           # Set binary extension for Windows
           BINARY_NAME="llamactl"
@@ -91,8 +107,10 @@ jobs:
             ARCHIVE_OS="macos"
           fi
           ARCHIVE_NAME="llamactl-${{ github.ref_name }}-${ARCHIVE_OS}-${{ matrix.goarch }}"
+          
           if [ "${{ matrix.goos }}" = "windows" ]; then
-            zip "${ARCHIVE_NAME}.zip" "${BINARY_NAME}"
+            # Use 7z on Windows (pre-installed)
+            7z a "${ARCHIVE_NAME}.zip" "${BINARY_NAME}"
             echo "ASSET_PATH=${ARCHIVE_NAME}.zip" >> $GITHUB_ENV
           else
             tar -czf "${ARCHIVE_NAME}.tar.gz" "${BINARY_NAME}"

.vscode/launch.json

@@ -14,6 +14,7 @@
                 "GO_ENV": "development",
                 "LLAMACTL_CONFIG_PATH": "${workspaceFolder}/llamactl.dev.yaml"
             },
+            "console": "integratedTerminal",
         }
     ]
 }

README.md

@@ -183,7 +183,7 @@ data_dir: ~/.local/share/llamactl  # Main data directory (database, instances, l
 
 instances:
   port_range: [8000, 9000]                        # Port range for instances
-  configs_dir: ~/.local/share/llamactl/instances  # Instance configs directory (platform dependent)
+  configs_dir: ~/.local/share/llamactl/instances  # Instance configs directory (platform dependent) [deprecated]
   logs_dir: ~/.local/share/llamactl/logs          # Logs directory (platform dependent)
   auto_create_dirs: true                          # Auto-create data/config/logs dirs if missing
   max_instances: -1                               # Max instances (-1 = unlimited)
@@ -195,6 +195,9 @@ instances:
   default_on_demand_start: true                   # Default on-demand start setting
   on_demand_start_timeout: 120                    # Default on-demand start timeout in seconds
   timeout_check_interval: 5                       # Idle instance timeout check in minutes
+  log_rotation_enabled: true                      # Enable log rotation (default: true)
+  log_rotation_max_size: 100                      # Max log file size in MB before rotation (default: 100)
+  log_rotation_compress: false                    # Compress rotated log files (default: false)
 
 database:
   path: ~/.local/share/llamactl/llamactl.db  # Database file path (platform dependent)
@@ -203,8 +206,7 @@ database:
   connection_max_lifetime: 5m                # Connection max lifetime
 
 auth:
-  require_inference_auth: true   # Require auth for inference endpoints
-  inference_keys: []             # Keys for inference endpoints
+  require_inference_auth: true   # Require auth for inference endpoints, API keys are created in web UI
   require_management_auth: true  # Require auth for management endpoints
   management_keys: []            # Keys for management endpoints
 ```

cmd/server/main.go

@@ -1,6 +1,7 @@
 package main
 
 import (
+	"context"
 	"fmt"
 	"llamactl/pkg/config"
 	"llamactl/pkg/database"
@@ -11,6 +12,7 @@ import (
 	"os"
 	"os/signal"
 	"syscall"
+	"time"
 )
 
 // version is set at build time using -ldflags "-X main.version=1.0.0"
@@ -116,14 +118,23 @@ func main() {
 	<-stop
 	fmt.Println("Shutting down server...")
 
-	if err := server.Close(); err != nil {
+	// Create shutdown context with timeout
+	shutdownCtx, shutdownCancel := context.WithTimeout(context.Background(), 30*time.Second)
+	defer shutdownCancel()
+
+	// Shutdown HTTP server gracefully
+	if err := server.Shutdown(shutdownCtx); err != nil {
 		log.Printf("Error shutting down server: %v\n", err)
 	} else {
 		fmt.Println("Server shut down gracefully.")
 	}
 
-	// Wait for all instances to stop
+	// Stop all instances and cleanup
 	instanceManager.Shutdown()
 
+	if err := db.Close(); err != nil {
+		log.Printf("Error closing database: %v\n", err)
+	}
+
 	fmt.Println("Exiting llamactl.")
 }

cmd/server/migrate_json.go

@@ -13,6 +13,7 @@ import (
 
 // migrateFromJSON migrates instances from JSON files to SQLite database
 // This is a one-time migration that runs on first startup with existing JSON files.
+// Migrated files are moved to a migrated subdirectory to avoid re-importing.
 func migrateFromJSON(cfg *config.AppConfig, db database.InstanceStore) error {
 	instancesDir := cfg.Instances.InstancesDir
 	if instancesDir == "" {
@@ -24,16 +25,6 @@ func migrateFromJSON(cfg *config.AppConfig, db database.InstanceStore) error {
 		return nil // No instances directory, nothing to migrate
 	}
 
-	// Check if database is empty (no instances)
-	existing, err := db.LoadAll()
-	if err != nil {
-		return fmt.Errorf("failed to check existing instances: %w", err)
-	}
-
-	if len(existing) > 0 {
-		return nil // Database already has instances, skip migration
-	}
-
 	// Find all JSON files
 	files, err := filepath.Glob(filepath.Join(instancesDir, "*.json"))
 	if err != nil {
@@ -46,6 +37,12 @@ func migrateFromJSON(cfg *config.AppConfig, db database.InstanceStore) error {
 
 	log.Printf("Migrating %d instances from JSON to SQLite...", len(files))
 
+	// Create migrated directory
+	migratedDir := filepath.Join(instancesDir, "migrated")
+	if err := os.MkdirAll(migratedDir, 0755); err != nil {
+		return fmt.Errorf("failed to create migrated directory: %w", err)
+	}
+
 	// Migrate each JSON file
 	var migrated int
 	for _, file := range files {
@@ -53,6 +50,14 @@ func migrateFromJSON(cfg *config.AppConfig, db database.InstanceStore) error {
 			log.Printf("Failed to migrate %s: %v", file, err)
 			continue
 		}
+
+		// Move the file to the migrated directory
+		destPath := filepath.Join(migratedDir, filepath.Base(file))
+		if err := os.Rename(file, destPath); err != nil {
+			log.Printf("Warning: Failed to move %s to migrated directory: %v", file, err)
+			// Don't fail the migration if we can't move the file
+		}
+
 		migrated++
 	}
 

docs/configuration.md

@@ -74,7 +74,6 @@ database:
 
 auth:
   require_inference_auth: true   # Require auth for inference endpoints
-  inference_keys: []             # Keys for inference endpoints
   require_management_auth: true  # Require auth for management endpoints
   management_keys: []            # Keys for management endpoints
 
@@ -231,6 +230,9 @@ instances:
   default_on_demand_start: true # Default on-demand start setting
   on_demand_start_timeout: 120  # Default on-demand start timeout in seconds
   timeout_check_interval: 5     # Default instance timeout check interval in minutes
+  log_rotation_enabled: true    # Enable log rotation (default: true)
+  log_rotation_max_size: 100    # Max log file size in MB before rotation (default: 100)
+  log_rotation_compress: false  # Compress rotated log files (default: false)
 ```
 
 **Environment Variables:**
@@ -247,6 +249,9 @@ instances:
 - `LLAMACTL_DEFAULT_ON_DEMAND_START` - Default on-demand start setting (true/false)  
 - `LLAMACTL_ON_DEMAND_START_TIMEOUT` - Default on-demand start timeout in seconds
 - `LLAMACTL_TIMEOUT_CHECK_INTERVAL` - Default instance timeout check interval in minutes
+- `LLAMACTL_LOG_ROTATION_ENABLED` - Enable log rotation (true/false)
+- `LLAMACTL_LOG_ROTATION_MAX_SIZE` - Max log file size in MB
+- `LLAMACTL_LOG_ROTATION_COMPRESS` - Compress rotated logs (true/false)
 
 ### Database Configuration
 
@@ -266,17 +271,33 @@ database:
 
 ### Authentication Configuration
 
+llamactl supports two types of authentication:
+
+- **Management API Keys**: For accessing the web UI and management API (creating/managing instances). These can be configured in the config file or via environment variables.
+- **Inference API Keys**: For accessing the OpenAI-compatible inference endpoints. These are managed via the web UI (Settings → API Keys) and stored in the database.
+
 ```yaml
 auth:
   require_inference_auth: true           # Require API key for OpenAI endpoints (default: true)
-  inference_keys: []                     # List of valid inference API keys
   require_management_auth: true          # Require API key for management endpoints (default: true)
   management_keys: []                    # List of valid management API keys
 ```
 
+**Managing Inference API Keys:**
+
+Inference API keys are managed through the web UI or management API and stored in the database. To create and manage inference keys:
+
+1. Open the web UI and log in with a management API key
+2. Navigate to **Settings → API Keys**
+3. Click **Create API Key**
+4. Configure the key:
+   - **Name**: A descriptive name for the key
+   - **Expiration**: Optional expiration date
+   - **Permissions**: Grant access to all instances or specific instances only
+5. Copy the generated key - it won't be shown again
+
 **Environment Variables:**
 - `LLAMACTL_REQUIRE_INFERENCE_AUTH` - Require auth for OpenAI endpoints (true/false)
-- `LLAMACTL_INFERENCE_KEYS` - Comma-separated inference API keys
 - `LLAMACTL_REQUIRE_MANAGEMENT_AUTH` - Require auth for management endpoints (true/false)
 - `LLAMACTL_MANAGEMENT_KEYS` - Comma-separated management API keys
 

docs/docs.go

@@ -2063,20 +2063,19 @@ const docTemplate = `{
         "server.CreateKeyRequest": {
             "type": "object",
             "properties": {
-                "expiresAt": {
-                    "type": "integer",
-                    "format": "int64"
+                "expires_at": {
+                    "type": "integer"
                 },
-                "instancePermissions": {
+                "instance_ids": {
                     "type": "array",
                     "items": {
-                        "$ref": "#/definitions/server.InstancePermission"
+                        "type": "integer"
                     }
                 },
                 "name": {
                     "type": "string"
                 },
-                "permissionMode": {
+                "permission_mode": {
                     "$ref": "#/definitions/auth.PermissionMode"
                 }
             }
@@ -2087,9 +2086,6 @@ const docTemplate = `{
                 "created_at": {
                     "type": "integer"
                 },
-                "enabled": {
-                    "type": "boolean"
-                },
                 "expires_at": {
                     "type": "integer"
                 },
@@ -2116,29 +2112,9 @@ const docTemplate = `{
                 }
             }
         },
-        "server.InstancePermission": {
-            "type": "object",
-            "properties": {
-                "can_infer": {
-                    "type": "boolean"
-                },
-                "can_view_logs": {
-                    "type": "boolean"
-                },
-                "instance_id": {
-                    "type": "integer"
-                }
-            }
-        },
         "server.KeyPermissionResponse": {
             "type": "object",
             "properties": {
-                "can_infer": {
-                    "type": "boolean"
-                },
-                "can_view_logs": {
-                    "type": "boolean"
-                },
                 "instance_id": {
                     "type": "integer"
                 },
@@ -2153,9 +2129,6 @@ const docTemplate = `{
                 "created_at": {
                     "type": "integer"
                 },
-                "enabled": {
-                    "type": "boolean"
-                },
                 "expires_at": {
                     "type": "integer"
                 },

docs/quick-start.md

@@ -17,10 +17,10 @@ Before you start, let's clarify a few key terms:
 
 Llamactl uses two types of API keys:
 
-- **Management API Key**: Used to authenticate with the Llamactl management API (creating, starting, stopping instances).
-- **Inference API Key**: Used to authenticate requests to the OpenAI-compatible endpoints (`/v1/chat/completions`, `/v1/completions`, etc.).
+- **Management API Key**: Used to authenticate with the Llamactl management API and web UI. If not configured, one is auto-generated at startup and printed to the terminal.
+- **Inference API Key**: Used to authenticate requests to the OpenAI-compatible endpoints (`/v1/chat/completions`, `/v1/completions`, etc.). These are created and managed via the web UI.
 
-By default, authentication is required. If you don't configure these keys in your configuration file, llamactl will auto-generate them and print them to the terminal on startup. You can also configure custom keys or disable authentication entirely in the [Configuration](configuration.md) guide.
+By default, authentication is required for both management and inference endpoints. You can configure custom management keys or disable authentication in the [Configuration](configuration.md) guide.
 
 ## Start Llamactl
 
@@ -38,24 +38,17 @@ llamactl
 
     sk-management-...
 
-━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-⚠️  INFERENCE AUTHENTICATION REQUIRED
-━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-🔑  Generated Inference API Key:
-
-    sk-inference-...
-
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 ⚠️  IMPORTANT
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-• These keys are auto-generated and will change on restart
-• For production, add explicit keys to your configuration
-• Copy these keys before they disappear from the terminal
+• This key is auto-generated and will change on restart
+• For production, add explicit management_keys to your configuration
+• Copy this key before it disappears from the terminal
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 Llamactl server listening on 0.0.0.0:8080
 ```
 
-Copy the **Management** and **Inference** API Keys from the terminal - you'll need them to access the web UI and make inference requests.
+Copy the **Management API Key** from the terminal - you'll need it to access the web UI.
 
 By default, Llamactl will start on `http://localhost:8080`.
 
@@ -82,7 +75,7 @@ You should see the Llamactl web interface.
      - **Additional Options**: Backend-specific parameters
 
     !!! tip "Auto-Assignment"
-        Llamactl automatically assigns ports from the configured port range (default: 8000-9000) and generates API keys if authentication is enabled. You typically don't need to manually specify these values.
+        Llamactl automatically assigns ports from the configured port range (default: 8000-9000) and manages API keys if authentication is enabled. You typically don't need to manually specify these values.
 
     !!! note "Remote Node Deployment"
         If you have configured remote nodes in your configuration file, you can select which node to deploy the instance to. This allows you to distribute instances across multiple machines. See the [Configuration](configuration.md#remote-node-configuration) guide for details on setting up remote nodes.
@@ -98,6 +91,24 @@ Once created, you can:
 - **View logs** by clicking the logs button
 - **Stop** the instance when needed
 
+## Create an Inference API Key
+
+To make inference requests to your instances, you'll need an inference API key:
+
+1. In the web UI, click the **Settings** icon (gear icon in the top-right)
+2. Navigate to the **API Keys** tab
+3. Click **Create API Key**
+4. Configure your key:
+   - **Name**: Give it a descriptive name (e.g., "Production Key", "Development Key")
+   - **Expiration**: Optionally set an expiration date for the key
+   - **Permissions**: Choose whether the key can access all instances or only specific ones
+5. Click **Create**
+6. **Copy the generated key** - it will only be shown once!
+
+The key will look like: `llamactl-...`
+
+You can create multiple inference keys with different permissions for different use cases (e.g., one for development, one for production, or keys limited to specific instances).
+
 ## Example Configurations
 
 Here are basic example configurations for each backend:
@@ -246,7 +257,7 @@ print(response.choices[0].message.content)
 ```
 
 !!! note "API Key"
-    If you disabled authentication in your config, you can use any value for `api_key` (e.g., `"not-needed"`). Otherwise, use the inference API key shown in the terminal output on startup.
+    If you disabled authentication in your config, you can use any value for `api_key` (e.g., `"not-needed"`). Otherwise, use the inference API key you created via the web UI (Settings → API Keys).
 
 ### List Available Models
 

docs/swagger.json

@@ -2056,20 +2056,19 @@
         "server.CreateKeyRequest": {
             "type": "object",
             "properties": {
-                "expiresAt": {
-                    "type": "integer",
-                    "format": "int64"
+                "expires_at": {
+                    "type": "integer"
                 },
-                "instancePermissions": {
+                "instance_ids": {
                     "type": "array",
                     "items": {
-                        "$ref": "#/definitions/server.InstancePermission"
+                        "type": "integer"
                     }
                 },
                 "name": {
                     "type": "string"
                 },
-                "permissionMode": {
+                "permission_mode": {
                     "$ref": "#/definitions/auth.PermissionMode"
                 }
             }
@@ -2080,9 +2079,6 @@
                 "created_at": {
                     "type": "integer"
                 },
-                "enabled": {
-                    "type": "boolean"
-                },
                 "expires_at": {
                     "type": "integer"
                 },
@@ -2109,29 +2105,9 @@
                 }
             }
         },
-        "server.InstancePermission": {
-            "type": "object",
-            "properties": {
-                "can_infer": {
-                    "type": "boolean"
-                },
-                "can_view_logs": {
-                    "type": "boolean"
-                },
-                "instance_id": {
-                    "type": "integer"
-                }
-            }
-        },
         "server.KeyPermissionResponse": {
             "type": "object",
             "properties": {
-                "can_infer": {
-                    "type": "boolean"
-                },
-                "can_view_logs": {
-                    "type": "boolean"
-                },
                 "instance_id": {
                     "type": "integer"
                 },
@@ -2146,9 +2122,6 @@
                 "created_at": {
                     "type": "integer"
                 },
-                "enabled": {
-                    "type": "boolean"
-                },
                 "expires_at": {
                     "type": "integer"
                 },

docs/swagger.yaml

@@ -232,24 +232,21 @@ definitions:
     type: object
   server.CreateKeyRequest:
     properties:
-      expiresAt:
-        format: int64
+      expires_at:
         type: integer
-      instancePermissions:
+      instance_ids:
         items:
-          $ref: '#/definitions/server.InstancePermission'
+          type: integer
         type: array
       name:
         type: string
-      permissionMode:
+      permission_mode:
         $ref: '#/definitions/auth.PermissionMode'
     type: object
   server.CreateKeyResponse:
     properties:
       created_at:
         type: integer
-      enabled:
-        type: boolean
       expires_at:
         type: integer
       id:
@@ -267,21 +264,8 @@ definitions:
       user_id:
         type: string
     type: object
-  server.InstancePermission:
-    properties:
-      can_infer:
-        type: boolean
-      can_view_logs:
-        type: boolean
-      instance_id:
-        type: integer
-    type: object
   server.KeyPermissionResponse:
     properties:
-      can_infer:
-        type: boolean
-      can_view_logs:
-        type: boolean
       instance_id:
         type: integer
       instance_name:
@@ -291,8 +275,6 @@ definitions:
     properties:
       created_at:
         type: integer
-      enabled:
-        type: boolean
       expires_at:
         type: integer
       id:

docs/troubleshooting.md

@@ -115,15 +115,15 @@ vllm serve microsoft/DialoGPT-medium --port 8081
      require_inference_auth: false
    ```
 
-2. **Configure API keys:**
+2. **Configure management API keys:**
    ```yaml
    auth:
      management_keys:
        - "your-management-key"
-     inference_keys:
-       - "your-inference-key"
    ```
 
+   For inference API keys, create them via the web UI (Settings → API Keys) after logging in with your management key.
+
 3. **Use correct Authorization header:**
    ```bash
    curl -H "Authorization: Bearer your-api-key" \

go.mod

@@ -3,13 +3,14 @@ module llamactl
 go 1.24.5
 
 require (
+	github.com/DeRuina/timberjack v1.3.9
 	github.com/go-chi/chi/v5 v5.2.2
 	github.com/go-chi/cors v1.2.2
 	github.com/golang-migrate/migrate/v4 v4.19.1
 	github.com/mattn/go-sqlite3 v1.14.24
 	github.com/swaggo/http-swagger v1.3.4
 	github.com/swaggo/swag v1.16.5
-	golang.org/x/crypto v0.45.0
+	golang.org/x/crypto v0.46.0
 	gopkg.in/yaml.v3 v3.0.1
 )
 
@@ -20,11 +21,12 @@ require (
 	github.com/go-openapi/spec v0.21.0 // indirect
 	github.com/go-openapi/swag v0.23.1 // indirect
 	github.com/josharian/intern v1.0.0 // indirect
+	github.com/klauspost/compress v1.17.11 // indirect
 	github.com/mailru/easyjson v0.9.0 // indirect
 	github.com/swaggo/files v1.0.1 // indirect
 	golang.org/x/mod v0.29.0 // indirect
 	golang.org/x/net v0.47.0 // indirect
 	golang.org/x/sync v0.18.0 // indirect
-	golang.org/x/sys v0.38.0 // indirect
+	golang.org/x/sys v0.39.0 // indirect
 	golang.org/x/tools v0.38.0 // indirect
 )

go.sum

@@ -1,7 +1,11 @@
+github.com/DeRuina/timberjack v1.3.9 h1:6UXZ1I7ExPGTX/1UNYawR58LlOJUHKBPiYC7WQ91eBo=
+github.com/DeRuina/timberjack v1.3.9/go.mod h1:RLoeQrwrCGIEF8gO5nV5b/gMD0QIy7bzQhBUgpp1EqE=
 github.com/KyleBanks/depth v1.2.1 h1:5h8fQADFrWtarTdtDudMmGsC7GPbOAu6RVB3ffsVFHc=
 github.com/KyleBanks/depth v1.2.1/go.mod h1:jzSb9d0L43HxTQfT+oSA1EEp2q+ne2uh6XgeJcm8brE=
 github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc h1:U9qPSI2PIWSS1VwoXQT9A3Wy9MM3WgvqSxFWenqJduM=
 github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/fortytw2/leaktest v1.3.0 h1:u8491cBMTQ8ft8aeV+adlcytMZylmA5nnwwkRZjI8vw=
+github.com/fortytw2/leaktest v1.3.0/go.mod h1:jDsjWgpAGjm2CA7WthBh/CdZYEPF31XHquHwclZch5g=
 github.com/go-chi/chi/v5 v5.2.2 h1:CMwsvRVTbXVytCk1Wd72Zy1LAsAh9GxMmSNWLHCG618=
 github.com/go-chi/chi/v5 v5.2.2/go.mod h1:L2yAIGWB3H+phAw1NxKwWM+7eUH/lU8pOMm5hHcoops=
 github.com/go-chi/cors v1.2.2 h1:Jmey33TE+b+rB7fT8MUy1u0I4L+NARQlK6LhzKPSyQE=
@@ -20,6 +24,8 @@ github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
 github.com/google/go-cmp v0.6.0/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
 github.com/josharian/intern v1.0.0 h1:vlS4z54oSdjm0bgjRigI+G1HpF+tI+9rE5LLzOg8HmY=
 github.com/josharian/intern v1.0.0/go.mod h1:5DoeVV0s6jJacbCEi61lwdGj/aVlrQvzHFFd8Hwg//Y=
+github.com/klauspost/compress v1.17.11 h1:In6xLpyWOi1+C7tXUUWv2ot1QvBjxevKAaI6IXrJmUc=
+github.com/klauspost/compress v1.17.11/go.mod h1:pMDklpSncoRMuLFrf1W9Ss9KT+0rH90U12bZKk7uwG0=
 github.com/kr/pretty v0.3.1 h1:flRD4NNwYAUpkphVc1HcthR4KEIFJ65n8Mw5qdRn3LE=
 github.com/kr/pretty v0.3.1/go.mod h1:hoEshYVHaxMs3cyo3Yncou5ZscifuDolrwPKZanG3xk=
 github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
@@ -45,8 +51,8 @@ github.com/swaggo/swag v1.16.5/go.mod h1:ngP2etMK5a0P3QBizic5MEwpRmluJZPHjXcMoj4
 github.com/yuin/goldmark v1.4.13/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
 golang.org/x/crypto v0.0.0-20210921155107-089bfa567519/go.mod h1:GvvjBRRGRdwPK5ydBHafDWAxML/pGHZbMvKqRZ5+Abc=
-golang.org/x/crypto v0.45.0 h1:jMBrvKuj23MTlT0bQEOBcAE0mjg8mK9RXFhRH6nyF3Q=
-golang.org/x/crypto v0.45.0/go.mod h1:XTGrrkGJve7CYK7J8PEww4aY7gM3qMCElcJQ8n8JdX4=
+golang.org/x/crypto v0.46.0 h1:cKRW/pmt1pKAfetfu+RCEvjvZkA9RimPbh7bhFjGVBU=
+golang.org/x/crypto v0.46.0/go.mod h1:Evb/oLKmMraqjZ2iQTwDwvCtJkczlDuTmdJXoZVzqU0=
 golang.org/x/mod v0.6.0-dev.0.20220419223038-86c51ed26bb4/go.mod h1:jJ57K6gSWd91VN4djpZkiMVwK6gcyfeH4XE8wZrZaV4=
 golang.org/x/mod v0.29.0 h1:HV8lRxZC4l2cr3Zq1LvtOsi/ThTgWnUk/y64QSs8GwA=
 golang.org/x/mod v0.29.0/go.mod h1:NyhrlYXJ2H4eJiRy/WDBO6HMqZQ6q9nk4JzS3NuCK+w=
@@ -66,8 +72,8 @@ golang.org/x/sys v0.0.0-20210615035016-665e8c7367d1/go.mod h1:oPkhp1MJrh7nUepCBc
 golang.org/x/sys v0.0.0-20220520151302-bc2c85ada10a/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.0.0-20220722155257-8c9f86f7a55f/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.5.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.38.0 h1:3yZWxaJjBmCWXqhN1qh02AkOnCQ1poK6oF+a7xWL6Gc=
-golang.org/x/sys v0.38.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
+golang.org/x/sys v0.39.0 h1:CvCKL8MeisomCi6qNZ+wbb0DN9E5AATixKsvNtMoMFk=
+golang.org/x/sys v0.39.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
 golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
 golang.org/x/term v0.0.0-20210927222741-03fcf44c2211/go.mod h1:jbD1KX2456YbFQfuXm/mYQcufACuNUgVhRMnK/tPxf8=
 golang.org/x/term v0.5.0/go.mod h1:jMB1sMXY+tzblOD4FWmEbocvup2/aLOaQEp7JmGp78k=

pkg/config/config.go

@@ -6,233 +6,18 @@ import (
 	"log"
 	"os"
 	"path/filepath"
-	"runtime"
-	"strconv"
-	"strings"
-	"time"
 
 	"gopkg.in/yaml.v3"
 )
 
-// BackendSettings contains structured backend configuration
-type BackendSettings struct {
-	Command         string            `yaml:"command" json:"command"`
-	Args            []string          `yaml:"args" json:"args"`
-	Environment     map[string]string `yaml:"environment,omitempty" json:"environment,omitempty"`
-	Docker          *DockerSettings   `yaml:"docker,omitempty" json:"docker,omitempty"`
-	ResponseHeaders map[string]string `yaml:"response_headers,omitempty" json:"response_headers,omitempty"`
-}
-
-// DockerSettings contains Docker-specific configuration
-type DockerSettings struct {
-	Enabled     bool              `yaml:"enabled" json:"enabled"`
-	Image       string            `yaml:"image" json:"image"`
-	Args        []string          `yaml:"args" json:"args"`
-	Environment map[string]string `yaml:"environment,omitempty" json:"environment,omitempty"`
-}
-
-// BackendConfig contains backend executable configurations
-type BackendConfig struct {
-	LlamaCpp BackendSettings `yaml:"llama-cpp" json:"llama-cpp"`
-	VLLM     BackendSettings `yaml:"vllm" json:"vllm"`
-	MLX      BackendSettings `yaml:"mlx" json:"mlx"`
-}
-
-// AppConfig represents the configuration for llamactl
-type AppConfig struct {
-	Server    ServerConfig          `yaml:"server" json:"server"`
-	Backends  BackendConfig         `yaml:"backends" json:"backends"`
-	Instances InstancesConfig       `yaml:"instances" json:"instances"`
-	Database  DatabaseConfig        `yaml:"database" json:"database"`
-	Auth      AuthConfig            `yaml:"auth" json:"auth"`
-	LocalNode string                `yaml:"local_node,omitempty" json:"local_node,omitempty"`
-	Nodes     map[string]NodeConfig `yaml:"nodes,omitempty" json:"nodes,omitempty"`
-
-	// Directory where all llamactl data will be stored (database, instances, logs, etc.)
-	DataDir string `yaml:"data_dir" json:"data_dir"`
-
-	Version    string `yaml:"-" json:"version"`
-	CommitHash string `yaml:"-" json:"commit_hash"`
-	BuildTime  string `yaml:"-" json:"build_time"`
-}
-
-// ServerConfig contains HTTP server configuration
-type ServerConfig struct {
-	// Server host to bind to
-	Host string `yaml:"host" json:"host"`
-
-	// Server port to bind to
-	Port int `yaml:"port" json:"port"`
-
-	// Allowed origins for CORS (e.g., "http://localhost:3000")
-	AllowedOrigins []string `yaml:"allowed_origins" json:"allowed_origins"`
-
-	// Allowed headers for CORS (e.g., "Accept", "Authorization", "Content-Type", "X-CSRF-Token")
-	AllowedHeaders []string `yaml:"allowed_headers" json:"allowed_headers"`
-
-	// Enable Swagger UI for API documentation
-	EnableSwagger bool `yaml:"enable_swagger" json:"enable_swagger"`
-
-	// Response headers to send with responses
-	ResponseHeaders map[string]string `yaml:"response_headers,omitempty" json:"response_headers,omitempty"`
-}
-
-// DatabaseConfig contains database configuration settings
-type DatabaseConfig struct {
-	// Database file path (relative to the top-level data_dir or absolute)
-	Path string `yaml:"path" json:"path"`
-
-	// Connection settings
-	MaxOpenConnections int           `yaml:"max_open_connections" json:"max_open_connections"`
-	MaxIdleConnections int           `yaml:"max_idle_connections" json:"max_idle_connections"`
-	ConnMaxLifetime    time.Duration `yaml:"connection_max_lifetime" json:"connection_max_lifetime" swaggertype:"string" example:"1h"`
-}
-
-// InstancesConfig contains instance management configuration
-type InstancesConfig struct {
-	// Port range for instances (e.g., 8000,9000)
-	PortRange [2]int `yaml:"port_range" json:"port_range"`
-
-	// Instance config directory override (relative to data_dir if not absolute)
-	InstancesDir string `yaml:"configs_dir" json:"configs_dir"`
-
-	// Logs directory override (relative to data_dir if not absolute)
-	LogsDir string `yaml:"logs_dir" json:"logs_dir"`
-
-	// Automatically create the data directory if it doesn't exist
-	AutoCreateDirs bool `yaml:"auto_create_dirs" json:"auto_create_dirs"`
-
-	// Maximum number of instances that can be created
-	MaxInstances int `yaml:"max_instances" json:"max_instances"`
-
-	// Maximum number of instances that can be running at the same time
-	MaxRunningInstances int `yaml:"max_running_instances,omitempty" json:"max_running_instances,omitempty"`
-
-	// Enable LRU eviction for instance logs
-	EnableLRUEviction bool `yaml:"enable_lru_eviction" json:"enable_lru_eviction"`
-
-	// Default auto-restart setting for new instances
-	DefaultAutoRestart bool `yaml:"default_auto_restart" json:"default_auto_restart"`
-
-	// Default max restarts for new instances
-	DefaultMaxRestarts int `yaml:"default_max_restarts" json:"default_max_restarts"`
-
-	// Default restart delay for new instances (in seconds)
-	DefaultRestartDelay int `yaml:"default_restart_delay" json:"default_restart_delay"`
-
-	// Default on-demand start setting for new instances
-	DefaultOnDemandStart bool `yaml:"default_on_demand_start" json:"default_on_demand_start"`
-
-	// How long to wait for an instance to start on demand (in seconds)
-	OnDemandStartTimeout int `yaml:"on_demand_start_timeout,omitempty" json:"on_demand_start_timeout,omitempty"`
-
-	// Interval for checking instance timeouts (in minutes)
-	TimeoutCheckInterval int `yaml:"timeout_check_interval" json:"timeout_check_interval"`
-}
-
-// AuthConfig contains authentication settings
-type AuthConfig struct {
-
-	// Require authentication for OpenAI compatible inference endpoints
-	RequireInferenceAuth bool `yaml:"require_inference_auth" json:"require_inference_auth"`
-
-	// List of keys for OpenAI compatible inference endpoints
-	InferenceKeys []string `yaml:"inference_keys" json:"inference_keys"`
-
-	// Require authentication for management endpoints
-	RequireManagementAuth bool `yaml:"require_management_auth" json:"require_management_auth"`
-
-	// List of keys for management endpoints
-	ManagementKeys []string `yaml:"management_keys" json:"management_keys"`
-}
-
-type NodeConfig struct {
-	Address string `yaml:"address" json:"address"`
-	APIKey  string `yaml:"api_key,omitempty" json:"api_key,omitempty"`
-}
-
 // LoadConfig loads configuration with the following precedence:
 // 1. Hardcoded defaults
 // 2. Config file
 // 3. Environment variables
 func LoadConfig(configPath string) (AppConfig, error) {
 	// 1. Start with defaults
-	defaultDataDir := getDefaultDataDirectory()
-
-	cfg := AppConfig{
-		Server: ServerConfig{
-			Host:           "0.0.0.0",
-			Port:           8080,
-			AllowedOrigins: []string{"*"}, // Default to allow all origins
-			AllowedHeaders: []string{"*"}, // Default to allow all headers
-			EnableSwagger:  false,
-		},
-		LocalNode: "main",
-		Nodes:     map[string]NodeConfig{},
-		DataDir:   defaultDataDir,
-		Backends: BackendConfig{
-			LlamaCpp: BackendSettings{
-				Command:     "llama-server",
-				Args:        []string{},
-				Environment: map[string]string{},
-				Docker: &DockerSettings{
-					Enabled: false,
-					Image:   "ghcr.io/ggml-org/llama.cpp:server",
-					Args: []string{
-						"run", "--rm", "--network", "host", "--gpus", "all",
-						"-v", filepath.Join(defaultDataDir, "llama.cpp") + ":/root/.cache/llama.cpp"},
-					Environment: map[string]string{},
-				},
-			},
-			VLLM: BackendSettings{
-				Command: "vllm",
-				Args:    []string{"serve"},
-				Docker: &DockerSettings{
-					Enabled: false,
-					Image:   "vllm/vllm-openai:latest",
-					Args: []string{
-						"run", "--rm", "--network", "host", "--gpus", "all", "--shm-size", "1g",
-						"-v", filepath.Join(defaultDataDir, "huggingface") + ":/root/.cache/huggingface",
-					},
-					Environment: map[string]string{},
-				},
-			},
-			MLX: BackendSettings{
-				Command: "mlx_lm.server",
-				Args:    []string{},
-				// No Docker section for MLX - not supported
-			},
-		},
-		Instances: InstancesConfig{
-			PortRange: [2]int{8000, 9000},
-			// NOTE: empty strings are set as placeholder values since InstancesDir and LogsDir
-			// should be relative path to DataDir if not explicitly set.
-			InstancesDir:         "",
-			LogsDir:              "",
-			AutoCreateDirs:       true,
-			MaxInstances:         -1, // -1 means unlimited
-			MaxRunningInstances:  -1, // -1 means unlimited
-			EnableLRUEviction:    true,
-			DefaultAutoRestart:   true,
-			DefaultMaxRestarts:   3,
-			DefaultRestartDelay:  5,
-			DefaultOnDemandStart: true,
-			OnDemandStartTimeout: 120, // 2 minutes
-			TimeoutCheckInterval: 5,   // Check timeouts every 5 minutes
-		},
-		Database: DatabaseConfig{
-			Path:               "", // Will be set to data_dir/llamactl.db if empty
-			MaxOpenConnections: 25,
-			MaxIdleConnections: 5,
-			ConnMaxLifetime:    5 * time.Minute,
-		},
-		Auth: AuthConfig{
-			RequireInferenceAuth:  true,
-			InferenceKeys:         []string{},
-			RequireManagementAuth: true,
-			ManagementKeys:        []string{},
-		},
-	}
+	defaultDataDir := getDefaultDataDir()
+	cfg := getDefaultConfig(defaultDataDir)
 
 	// 2. Load from config file
 	if err := loadConfigFile(&cfg, configPath); err != nil {
@@ -300,372 +85,6 @@ func loadConfigFile(cfg *AppConfig, configPath string) error {
 	return nil
 }
 
-// loadEnvVars overrides config with environment variables
-func loadEnvVars(cfg *AppConfig) {
-	// Server config
-	if host := os.Getenv("LLAMACTL_HOST"); host != "" {
-		cfg.Server.Host = host
-	}
-	if port := os.Getenv("LLAMACTL_PORT"); port != "" {
-		if p, err := strconv.Atoi(port); err == nil {
-			cfg.Server.Port = p
-		}
-	}
-	if allowedOrigins := os.Getenv("LLAMACTL_ALLOWED_ORIGINS"); allowedOrigins != "" {
-		cfg.Server.AllowedOrigins = strings.Split(allowedOrigins, ",")
-	}
-	if enableSwagger := os.Getenv("LLAMACTL_ENABLE_SWAGGER"); enableSwagger != "" {
-		if b, err := strconv.ParseBool(enableSwagger); err == nil {
-			cfg.Server.EnableSwagger = b
-		}
-	}
-
-	// Data config
-	if dataDir := os.Getenv("LLAMACTL_DATA_DIRECTORY"); dataDir != "" {
-		cfg.DataDir = dataDir
-	}
-	if instancesDir := os.Getenv("LLAMACTL_INSTANCES_DIR"); instancesDir != "" {
-		cfg.Instances.InstancesDir = instancesDir
-	}
-	if logsDir := os.Getenv("LLAMACTL_LOGS_DIR"); logsDir != "" {
-		cfg.Instances.LogsDir = logsDir
-	}
-	if autoCreate := os.Getenv("LLAMACTL_AUTO_CREATE_DATA_DIR"); autoCreate != "" {
-		if b, err := strconv.ParseBool(autoCreate); err == nil {
-			cfg.Instances.AutoCreateDirs = b
-		}
-	}
-
-	// Instance config
-	if portRange := os.Getenv("LLAMACTL_INSTANCE_PORT_RANGE"); portRange != "" {
-		if ports := ParsePortRange(portRange); ports != [2]int{0, 0} {
-			cfg.Instances.PortRange = ports
-		}
-	}
-	if maxInstances := os.Getenv("LLAMACTL_MAX_INSTANCES"); maxInstances != "" {
-		if m, err := strconv.Atoi(maxInstances); err == nil {
-			cfg.Instances.MaxInstances = m
-		}
-	}
-	if maxRunning := os.Getenv("LLAMACTL_MAX_RUNNING_INSTANCES"); maxRunning != "" {
-		if m, err := strconv.Atoi(maxRunning); err == nil {
-			cfg.Instances.MaxRunningInstances = m
-		}
-	}
-	if enableLRUEviction := os.Getenv("LLAMACTL_ENABLE_LRU_EVICTION"); enableLRUEviction != "" {
-		if b, err := strconv.ParseBool(enableLRUEviction); err == nil {
-			cfg.Instances.EnableLRUEviction = b
-		}
-	}
-	// Backend config
-	// LlamaCpp backend
-	if llamaCmd := os.Getenv("LLAMACTL_LLAMACPP_COMMAND"); llamaCmd != "" {
-		cfg.Backends.LlamaCpp.Command = llamaCmd
-	}
-	if llamaArgs := os.Getenv("LLAMACTL_LLAMACPP_ARGS"); llamaArgs != "" {
-		cfg.Backends.LlamaCpp.Args = strings.Split(llamaArgs, " ")
-	}
-	if llamaEnv := os.Getenv("LLAMACTL_LLAMACPP_ENV"); llamaEnv != "" {
-		if cfg.Backends.LlamaCpp.Environment == nil {
-			cfg.Backends.LlamaCpp.Environment = make(map[string]string)
-		}
-		parseEnvVars(llamaEnv, cfg.Backends.LlamaCpp.Environment)
-	}
-	if llamaDockerEnabled := os.Getenv("LLAMACTL_LLAMACPP_DOCKER_ENABLED"); llamaDockerEnabled != "" {
-		if b, err := strconv.ParseBool(llamaDockerEnabled); err == nil {
-			if cfg.Backends.LlamaCpp.Docker == nil {
-				cfg.Backends.LlamaCpp.Docker = &DockerSettings{}
-			}
-			cfg.Backends.LlamaCpp.Docker.Enabled = b
-		}
-	}
-	if llamaDockerImage := os.Getenv("LLAMACTL_LLAMACPP_DOCKER_IMAGE"); llamaDockerImage != "" {
-		if cfg.Backends.LlamaCpp.Docker == nil {
-			cfg.Backends.LlamaCpp.Docker = &DockerSettings{}
-		}
-		cfg.Backends.LlamaCpp.Docker.Image = llamaDockerImage
-	}
-	if llamaDockerArgs := os.Getenv("LLAMACTL_LLAMACPP_DOCKER_ARGS"); llamaDockerArgs != "" {
-		if cfg.Backends.LlamaCpp.Docker == nil {
-			cfg.Backends.LlamaCpp.Docker = &DockerSettings{}
-		}
-		cfg.Backends.LlamaCpp.Docker.Args = strings.Split(llamaDockerArgs, " ")
-	}
-	if llamaDockerEnv := os.Getenv("LLAMACTL_LLAMACPP_DOCKER_ENV"); llamaDockerEnv != "" {
-		if cfg.Backends.LlamaCpp.Docker == nil {
-			cfg.Backends.LlamaCpp.Docker = &DockerSettings{}
-		}
-		if cfg.Backends.LlamaCpp.Docker.Environment == nil {
-			cfg.Backends.LlamaCpp.Docker.Environment = make(map[string]string)
-		}
-		parseEnvVars(llamaDockerEnv, cfg.Backends.LlamaCpp.Docker.Environment)
-	}
-	if llamaEnv := os.Getenv("LLAMACTL_LLAMACPP_RESPONSE_HEADERS"); llamaEnv != "" {
-		if cfg.Backends.LlamaCpp.ResponseHeaders == nil {
-			cfg.Backends.LlamaCpp.ResponseHeaders = make(map[string]string)
-		}
-		parseHeaders(llamaEnv, cfg.Backends.LlamaCpp.ResponseHeaders)
-	}
-
-	// vLLM backend
-	if vllmCmd := os.Getenv("LLAMACTL_VLLM_COMMAND"); vllmCmd != "" {
-		cfg.Backends.VLLM.Command = vllmCmd
-	}
-	if vllmArgs := os.Getenv("LLAMACTL_VLLM_ARGS"); vllmArgs != "" {
-		cfg.Backends.VLLM.Args = strings.Split(vllmArgs, " ")
-	}
-	if vllmEnv := os.Getenv("LLAMACTL_VLLM_ENV"); vllmEnv != "" {
-		if cfg.Backends.VLLM.Environment == nil {
-			cfg.Backends.VLLM.Environment = make(map[string]string)
-		}
-		parseEnvVars(vllmEnv, cfg.Backends.VLLM.Environment)
-	}
-	if vllmDockerEnabled := os.Getenv("LLAMACTL_VLLM_DOCKER_ENABLED"); vllmDockerEnabled != "" {
-		if b, err := strconv.ParseBool(vllmDockerEnabled); err == nil {
-			if cfg.Backends.VLLM.Docker == nil {
-				cfg.Backends.VLLM.Docker = &DockerSettings{}
-			}
-			cfg.Backends.VLLM.Docker.Enabled = b
-		}
-	}
-	if vllmDockerImage := os.Getenv("LLAMACTL_VLLM_DOCKER_IMAGE"); vllmDockerImage != "" {
-		if cfg.Backends.VLLM.Docker == nil {
-			cfg.Backends.VLLM.Docker = &DockerSettings{}
-		}
-		cfg.Backends.VLLM.Docker.Image = vllmDockerImage
-	}
-	if vllmDockerArgs := os.Getenv("LLAMACTL_VLLM_DOCKER_ARGS"); vllmDockerArgs != "" {
-		if cfg.Backends.VLLM.Docker == nil {
-			cfg.Backends.VLLM.Docker = &DockerSettings{}
-		}
-		cfg.Backends.VLLM.Docker.Args = strings.Split(vllmDockerArgs, " ")
-	}
-	if vllmDockerEnv := os.Getenv("LLAMACTL_VLLM_DOCKER_ENV"); vllmDockerEnv != "" {
-		if cfg.Backends.VLLM.Docker == nil {
-			cfg.Backends.VLLM.Docker = &DockerSettings{}
-		}
-		if cfg.Backends.VLLM.Docker.Environment == nil {
-			cfg.Backends.VLLM.Docker.Environment = make(map[string]string)
-		}
-		parseEnvVars(vllmDockerEnv, cfg.Backends.VLLM.Docker.Environment)
-	}
-	if llamaEnv := os.Getenv("LLAMACTL_VLLM_RESPONSE_HEADERS"); llamaEnv != "" {
-		if cfg.Backends.VLLM.ResponseHeaders == nil {
-			cfg.Backends.VLLM.ResponseHeaders = make(map[string]string)
-		}
-		parseHeaders(llamaEnv, cfg.Backends.VLLM.ResponseHeaders)
-	}
-
-	// MLX backend
-	if mlxCmd := os.Getenv("LLAMACTL_MLX_COMMAND"); mlxCmd != "" {
-		cfg.Backends.MLX.Command = mlxCmd
-	}
-	if mlxArgs := os.Getenv("LLAMACTL_MLX_ARGS"); mlxArgs != "" {
-		cfg.Backends.MLX.Args = strings.Split(mlxArgs, " ")
-	}
-	if mlxEnv := os.Getenv("LLAMACTL_MLX_ENV"); mlxEnv != "" {
-		if cfg.Backends.MLX.Environment == nil {
-			cfg.Backends.MLX.Environment = make(map[string]string)
-		}
-		parseEnvVars(mlxEnv, cfg.Backends.MLX.Environment)
-	}
-	if llamaEnv := os.Getenv("LLAMACTL_MLX_RESPONSE_HEADERS"); llamaEnv != "" {
-		if cfg.Backends.MLX.ResponseHeaders == nil {
-			cfg.Backends.MLX.ResponseHeaders = make(map[string]string)
-		}
-		parseHeaders(llamaEnv, cfg.Backends.MLX.ResponseHeaders)
-	}
-
-	// Instance defaults
-	if autoRestart := os.Getenv("LLAMACTL_DEFAULT_AUTO_RESTART"); autoRestart != "" {
-		if b, err := strconv.ParseBool(autoRestart); err == nil {
-			cfg.Instances.DefaultAutoRestart = b
-		}
-	}
-	if maxRestarts := os.Getenv("LLAMACTL_DEFAULT_MAX_RESTARTS"); maxRestarts != "" {
-		if m, err := strconv.Atoi(maxRestarts); err == nil {
-			cfg.Instances.DefaultMaxRestarts = m
-		}
-	}
-	if restartDelay := os.Getenv("LLAMACTL_DEFAULT_RESTART_DELAY"); restartDelay != "" {
-		if seconds, err := strconv.Atoi(restartDelay); err == nil {
-			cfg.Instances.DefaultRestartDelay = seconds
-		}
-	}
-	if onDemandStart := os.Getenv("LLAMACTL_DEFAULT_ON_DEMAND_START"); onDemandStart != "" {
-		if b, err := strconv.ParseBool(onDemandStart); err == nil {
-			cfg.Instances.DefaultOnDemandStart = b
-		}
-	}
-	if onDemandTimeout := os.Getenv("LLAMACTL_ON_DEMAND_START_TIMEOUT"); onDemandTimeout != "" {
-		if seconds, err := strconv.Atoi(onDemandTimeout); err == nil {
-			cfg.Instances.OnDemandStartTimeout = seconds
-		}
-	}
-	if timeoutCheckInterval := os.Getenv("LLAMACTL_TIMEOUT_CHECK_INTERVAL"); timeoutCheckInterval != "" {
-		if minutes, err := strconv.Atoi(timeoutCheckInterval); err == nil {
-			cfg.Instances.TimeoutCheckInterval = minutes
-		}
-	}
-	// Auth config
-	if requireInferenceAuth := os.Getenv("LLAMACTL_REQUIRE_INFERENCE_AUTH"); requireInferenceAuth != "" {
-		if b, err := strconv.ParseBool(requireInferenceAuth); err == nil {
-			cfg.Auth.RequireInferenceAuth = b
-		}
-	}
-	if inferenceKeys := os.Getenv("LLAMACTL_INFERENCE_KEYS"); inferenceKeys != "" {
-		cfg.Auth.InferenceKeys = strings.Split(inferenceKeys, ",")
-	}
-	if requireManagementAuth := os.Getenv("LLAMACTL_REQUIRE_MANAGEMENT_AUTH"); requireManagementAuth != "" {
-		if b, err := strconv.ParseBool(requireManagementAuth); err == nil {
-			cfg.Auth.RequireManagementAuth = b
-		}
-	}
-	if managementKeys := os.Getenv("LLAMACTL_MANAGEMENT_KEYS"); managementKeys != "" {
-		cfg.Auth.ManagementKeys = strings.Split(managementKeys, ",")
-	}
-
-	// Local node config
-	if localNode := os.Getenv("LLAMACTL_LOCAL_NODE"); localNode != "" {
-		cfg.LocalNode = localNode
-	}
-
-	// Database config
-	if dbPath := os.Getenv("LLAMACTL_DATABASE_PATH"); dbPath != "" {
-		cfg.Database.Path = dbPath
-	}
-	if maxOpenConns := os.Getenv("LLAMACTL_DATABASE_MAX_OPEN_CONNECTIONS"); maxOpenConns != "" {
-		if m, err := strconv.Atoi(maxOpenConns); err == nil {
-			cfg.Database.MaxOpenConnections = m
-		}
-	}
-	if maxIdleConns := os.Getenv("LLAMACTL_DATABASE_MAX_IDLE_CONNECTIONS"); maxIdleConns != "" {
-		if m, err := strconv.Atoi(maxIdleConns); err == nil {
-			cfg.Database.MaxIdleConnections = m
-		}
-	}
-	if connMaxLifetime := os.Getenv("LLAMACTL_DATABASE_CONN_MAX_LIFETIME"); connMaxLifetime != "" {
-		if d, err := time.ParseDuration(connMaxLifetime); err == nil {
-			cfg.Database.ConnMaxLifetime = d
-		}
-	}
-}
-
-// ParsePortRange parses port range from string formats like "8000-9000" or "8000,9000"
-func ParsePortRange(s string) [2]int {
-	var parts []string
-
-	// Try both separators
-	if strings.Contains(s, "-") {
-		parts = strings.Split(s, "-")
-	} else if strings.Contains(s, ",") {
-		parts = strings.Split(s, ",")
-	}
-
-	// Parse the two parts
-	if len(parts) == 2 {
-		start, err1 := strconv.Atoi(strings.TrimSpace(parts[0]))
-		end, err2 := strconv.Atoi(strings.TrimSpace(parts[1]))
-		if err1 == nil && err2 == nil {
-			return [2]int{start, end}
-		}
-	}
-
-	return [2]int{0, 0} // Invalid format
-}
-
-// parseEnvVars parses environment variables in format "KEY1=value1,KEY2=value2"
-// and populates the provided environment map
-func parseEnvVars(envString string, envMap map[string]string) {
-	if envString == "" {
-		return
-	}
-	for _, envPair := range strings.Split(envString, ",") {
-		if parts := strings.SplitN(strings.TrimSpace(envPair), "=", 2); len(parts) == 2 {
-			envMap[parts[0]] = parts[1]
-		}
-	}
-}
-
-// parseHeaders parses HTTP headers in format "KEY1=value1;KEY2=value2"
-// and populates the provided environment map
-func parseHeaders(envString string, envMap map[string]string) {
-	if envString == "" {
-		return
-	}
-	for _, envPair := range strings.Split(envString, ";") {
-		if parts := strings.SplitN(strings.TrimSpace(envPair), "=", 2); len(parts) == 2 {
-			envMap[parts[0]] = parts[1]
-		}
-	}
-}
-
-// getDefaultDataDirectory returns platform-specific default data directory
-func getDefaultDataDirectory() string {
-	switch runtime.GOOS {
-	case "windows":
-		// Try PROGRAMDATA first (system-wide), fallback to LOCALAPPDATA (user)
-		if programData := os.Getenv("PROGRAMDATA"); programData != "" {
-			return filepath.Join(programData, "llamactl")
-		}
-		if localAppData := os.Getenv("LOCALAPPDATA"); localAppData != "" {
-			return filepath.Join(localAppData, "llamactl")
-		}
-		return "C:\\ProgramData\\llamactl" // Final fallback
-
-	case "darwin":
-		// For macOS, use user's Application Support directory
-		if homeDir, _ := os.UserHomeDir(); homeDir != "" {
-			return filepath.Join(homeDir, "Library", "Application Support", "llamactl")
-		}
-		return "/usr/local/var/llamactl" // Fallback
-
-	default:
-		// Linux and other Unix-like systems
-		if homeDir, _ := os.UserHomeDir(); homeDir != "" {
-			return filepath.Join(homeDir, ".local", "share", "llamactl")
-		}
-		return "/var/lib/llamactl" // Final fallback
-	}
-}
-
-// getDefaultConfigLocations returns platform-specific config file locations
-func getDefaultConfigLocations() []string {
-	var locations []string
-	// Use ./llamactl.yaml and ./config.yaml as the default config file
-	locations = append(locations, "llamactl.yaml")
-	locations = append(locations, "config.yaml")
-
-	homeDir, _ := os.UserHomeDir()
-
-	switch runtime.GOOS {
-	case "windows":
-		// Windows: Use APPDATA if available, else user home, fallback to ProgramData
-		if appData := os.Getenv("APPDATA"); appData != "" {
-			locations = append(locations, filepath.Join(appData, "llamactl", "config.yaml"))
-		} else if homeDir != "" {
-			locations = append(locations, filepath.Join(homeDir, "llamactl", "config.yaml"))
-		}
-		locations = append(locations, filepath.Join(os.Getenv("PROGRAMDATA"), "llamactl", "config.yaml"))
-
-	case "darwin":
-		// macOS: Use Application Support in user home, fallback to /Library/Application Support
-		if homeDir != "" {
-			locations = append(locations, filepath.Join(homeDir, "Library", "Application Support", "llamactl", "config.yaml"))
-		}
-		locations = append(locations, "/Library/Application Support/llamactl/config.yaml")
-
-	default:
-		// Linux/Unix: Use ~/.config/llamactl/config.yaml, fallback to /etc/llamactl/config.yaml
-		if homeDir != "" {
-			locations = append(locations, filepath.Join(homeDir, ".config", "llamactl", "config.yaml"))
-		}
-		locations = append(locations, "/etc/llamactl/config.yaml")
-	}
-
-	return locations
-}
-
 // SanitizedCopy returns a copy of the AppConfig with sensitive information removed
 func (cfg *AppConfig) SanitizedCopy() (AppConfig, error) {
 	// Deep copy via JSON marshal/unmarshal to avoid concurrent map access

pkg/config/config_test.go

@@ -78,8 +78,8 @@ server:
   port: 9090
 instances:
   port_range: [7000, 8000]
-  logs_dir: "/custom/logs"
   max_instances: 5
+  logs_dir: "/custom/logs"
   llama_executable: "/usr/bin/llama-server"
   default_auto_restart: false
   default_max_restarts: 10
@@ -219,7 +219,6 @@ instances:
 	}
 }
 
-
 func TestParsePortRange(t *testing.T) {
 	tests := []struct {
 		name     string
@@ -248,7 +247,6 @@ func TestParsePortRange(t *testing.T) {
 	}
 }
 
-
 func TestGetBackendSettings_NewStructuredConfig(t *testing.T) {
 	bc := &config.BackendConfig{
 		LlamaCpp: config.BackendSettings{
@@ -305,7 +303,6 @@ func TestGetBackendSettings_NewStructuredConfig(t *testing.T) {
 	}
 }
 
-
 func TestLoadConfig_BackendEnvironmentVariables(t *testing.T) {
 	// Test that backend environment variables work correctly
 	envVars := map[string]string{
@@ -375,7 +372,6 @@ func TestLoadConfig_BackendEnvironmentVariables(t *testing.T) {
 	}
 }
 
-
 func TestLoadConfig_LocalNode(t *testing.T) {
 	t.Run("default local node", func(t *testing.T) {
 		cfg, err := config.LoadConfig("nonexistent-file.yaml")

pkg/config/defaults.go

@@ -0,0 +1,154 @@
+package config
+
+import (
+	"os"
+	"path/filepath"
+	"runtime"
+	"time"
+)
+
+func getDefaultConfig(dataDir string) AppConfig {
+	return AppConfig{
+		Server: ServerConfig{
+			Host:           "0.0.0.0",
+			Port:           8080,
+			AllowedOrigins: []string{"*"}, // Default to allow all origins
+			AllowedHeaders: []string{"*"}, // Default to allow all headers
+			EnableSwagger:  false,
+		},
+		LocalNode: "main",
+		Nodes:     map[string]NodeConfig{},
+		DataDir:   dataDir,
+		Backends: BackendConfig{
+			LlamaCpp: BackendSettings{
+				Command:     "llama-server",
+				Args:        []string{},
+				Environment: map[string]string{},
+				Docker: &DockerSettings{
+					Enabled: false,
+					Image:   "ghcr.io/ggml-org/llama.cpp:server",
+					Args: []string{
+						"run", "--rm", "--network", "host", "--gpus", "all",
+						"-v", filepath.Join(dataDir, "llama.cpp") + ":/root/.cache/llama.cpp"},
+					Environment: map[string]string{},
+				},
+			},
+			VLLM: BackendSettings{
+				Command: "vllm",
+				Args:    []string{"serve"},
+				Docker: &DockerSettings{
+					Enabled: false,
+					Image:   "vllm/vllm-openai:latest",
+					Args: []string{
+						"run", "--rm", "--network", "host", "--gpus", "all", "--shm-size", "1g",
+						"-v", filepath.Join(dataDir, "huggingface") + ":/root/.cache/huggingface",
+					},
+					Environment: map[string]string{},
+				},
+			},
+			MLX: BackendSettings{
+				Command: "mlx_lm.server",
+				Args:    []string{},
+				// No Docker section for MLX - not supported
+			},
+		},
+		Instances: InstancesConfig{
+			PortRange: [2]int{8000, 9000},
+			// NOTE: empty string is set as placeholder value since InstancesDir
+			// should be relative path to DataDir if not explicitly set.
+			InstancesDir:         "",
+			AutoCreateDirs:       true,
+			MaxInstances:         -1, // -1 means unlimited
+			MaxRunningInstances:  -1, // -1 means unlimited
+			EnableLRUEviction:    true,
+			DefaultAutoRestart:   true,
+			DefaultMaxRestarts:   3,
+			DefaultRestartDelay:  5,
+			DefaultOnDemandStart: true,
+			OnDemandStartTimeout: 120, // 2 minutes
+			TimeoutCheckInterval: 5,   // Check timeouts every 5 minutes
+			LogsDir:              "",  // Will be set to data_dir/logs if empty
+			LogRotationEnabled:   true,
+			LogRotationMaxSize:   100,
+			LogRotationCompress:  false,
+		},
+		Database: DatabaseConfig{
+			Path:               "", // Will be set to data_dir/llamactl.db if empty
+			MaxOpenConnections: 25,
+			MaxIdleConnections: 5,
+			ConnMaxLifetime:    5 * time.Minute,
+		},
+		Auth: AuthConfig{
+			RequireInferenceAuth:  true,
+			InferenceKeys:         []string{},
+			RequireManagementAuth: true,
+			ManagementKeys:        []string{},
+		},
+	}
+}
+
+// getDefaultDataDir returns platform-specific default data directory
+func getDefaultDataDir() string {
+	switch runtime.GOOS {
+	case "windows":
+		// Try PROGRAMDATA first (system-wide), fallback to LOCALAPPDATA (user)
+		if programData := os.Getenv("PROGRAMDATA"); programData != "" {
+			return filepath.Join(programData, "llamactl")
+		}
+		if localAppData := os.Getenv("LOCALAPPDATA"); localAppData != "" {
+			return filepath.Join(localAppData, "llamactl")
+		}
+		return "C:\\ProgramData\\llamactl" // Final fallback
+
+	case "darwin":
+		// For macOS, use user's Application Support directory
+		if homeDir, _ := os.UserHomeDir(); homeDir != "" {
+			return filepath.Join(homeDir, "Library", "Application Support", "llamactl")
+		}
+		return "/usr/local/var/llamactl" // Fallback
+
+	default:
+		// Linux and other Unix-like systems
+		if homeDir, _ := os.UserHomeDir(); homeDir != "" {
+			return filepath.Join(homeDir, ".local", "share", "llamactl")
+		}
+		return "/var/lib/llamactl" // Final fallback
+	}
+}
+
+// getDefaultConfigLocations returns platform-specific config file locations
+func getDefaultConfigLocations() []string {
+	var locations []string
+	// Use ./llamactl.yaml and ./config.yaml as the default config file
+	locations = append(locations, "llamactl.yaml")
+	locations = append(locations, "config.yaml")
+
+	homeDir, _ := os.UserHomeDir()
+
+	switch runtime.GOOS {
+	case "windows":
+		// Windows: Use APPDATA if available, else user home, fallback to ProgramData
+		if appData := os.Getenv("APPDATA"); appData != "" {
+			locations = append(locations, filepath.Join(appData, "llamactl", "config.yaml"))
+		} else if homeDir != "" {
+			locations = append(locations, filepath.Join(homeDir, "llamactl", "config.yaml"))
+		}
+		locations = append(locations, filepath.Join(os.Getenv("PROGRAMDATA"), "llamactl", "config.yaml"))
+
+	case "darwin":
+		// macOS: Use Application Support in user home, fallback to /Library/Application Support
+		if homeDir != "" {
+			locations = append(locations, filepath.Join(homeDir, "Library", "Application Support", "llamactl", "config.yaml"))
+		}
+		locations = append(locations, "/Library/Application Support/llamactl/config.yaml")
+
+	default:
+		// Linux/Unix: Use ~/.config/llamactl/config.yaml, fallback to /etc/llamactl/config.yaml
+		if homeDir != "" {
+			locations = append(locations, filepath.Join(homeDir, ".config", "llamactl", "config.yaml"))
+		}
+		locations = append(locations, "/etc/llamactl/config.yaml")
+	}
+
+	return locations
+}

pkg/config/env.go

@@ -0,0 +1,325 @@
+package config
+
+import (
+	"os"
+	"strconv"
+	"strings"
+	"time"
+)
+
+// loadEnvVars overrides config with environment variables
+func loadEnvVars(cfg *AppConfig) {
+	// Server config
+	if host := os.Getenv("LLAMACTL_HOST"); host != "" {
+		cfg.Server.Host = host
+	}
+	if port := os.Getenv("LLAMACTL_PORT"); port != "" {
+		if p, err := strconv.Atoi(port); err == nil {
+			cfg.Server.Port = p
+		}
+	}
+	if allowedOrigins := os.Getenv("LLAMACTL_ALLOWED_ORIGINS"); allowedOrigins != "" {
+		cfg.Server.AllowedOrigins = strings.Split(allowedOrigins, ",")
+	}
+	if enableSwagger := os.Getenv("LLAMACTL_ENABLE_SWAGGER"); enableSwagger != "" {
+		if b, err := strconv.ParseBool(enableSwagger); err == nil {
+			cfg.Server.EnableSwagger = b
+		}
+	}
+
+	// Data config
+	if dataDir := os.Getenv("LLAMACTL_DATA_DIRECTORY"); dataDir != "" {
+		cfg.DataDir = dataDir
+	}
+	if instancesDir := os.Getenv("LLAMACTL_INSTANCES_DIR"); instancesDir != "" {
+		cfg.Instances.InstancesDir = instancesDir
+	}
+	if logsDir := os.Getenv("LLAMACTL_LOGS_DIR"); logsDir != "" {
+		cfg.Instances.LogsDir = logsDir
+	}
+	if autoCreate := os.Getenv("LLAMACTL_AUTO_CREATE_DATA_DIR"); autoCreate != "" {
+		if b, err := strconv.ParseBool(autoCreate); err == nil {
+			cfg.Instances.AutoCreateDirs = b
+		}
+	}
+
+	// Instance config
+	if portRange := os.Getenv("LLAMACTL_INSTANCE_PORT_RANGE"); portRange != "" {
+		if ports := ParsePortRange(portRange); ports != [2]int{0, 0} {
+			cfg.Instances.PortRange = ports
+		}
+	}
+	if maxInstances := os.Getenv("LLAMACTL_MAX_INSTANCES"); maxInstances != "" {
+		if m, err := strconv.Atoi(maxInstances); err == nil {
+			cfg.Instances.MaxInstances = m
+		}
+	}
+	if maxRunning := os.Getenv("LLAMACTL_MAX_RUNNING_INSTANCES"); maxRunning != "" {
+		if m, err := strconv.Atoi(maxRunning); err == nil {
+			cfg.Instances.MaxRunningInstances = m
+		}
+	}
+	if enableLRUEviction := os.Getenv("LLAMACTL_ENABLE_LRU_EVICTION"); enableLRUEviction != "" {
+		if b, err := strconv.ParseBool(enableLRUEviction); err == nil {
+			cfg.Instances.EnableLRUEviction = b
+		}
+	}
+	// Backend config
+	// LlamaCpp backend
+	if llamaCmd := os.Getenv("LLAMACTL_LLAMACPP_COMMAND"); llamaCmd != "" {
+		cfg.Backends.LlamaCpp.Command = llamaCmd
+	}
+	if llamaArgs := os.Getenv("LLAMACTL_LLAMACPP_ARGS"); llamaArgs != "" {
+		cfg.Backends.LlamaCpp.Args = strings.Split(llamaArgs, " ")
+	}
+	if llamaEnv := os.Getenv("LLAMACTL_LLAMACPP_ENV"); llamaEnv != "" {
+		if cfg.Backends.LlamaCpp.Environment == nil {
+			cfg.Backends.LlamaCpp.Environment = make(map[string]string)
+		}
+		parseEnvVars(llamaEnv, cfg.Backends.LlamaCpp.Environment)
+	}
+	if llamaDockerEnabled := os.Getenv("LLAMACTL_LLAMACPP_DOCKER_ENABLED"); llamaDockerEnabled != "" {
+		if b, err := strconv.ParseBool(llamaDockerEnabled); err == nil {
+			if cfg.Backends.LlamaCpp.Docker == nil {
+				cfg.Backends.LlamaCpp.Docker = &DockerSettings{}
+			}
+			cfg.Backends.LlamaCpp.Docker.Enabled = b
+		}
+	}
+	if llamaDockerImage := os.Getenv("LLAMACTL_LLAMACPP_DOCKER_IMAGE"); llamaDockerImage != "" {
+		if cfg.Backends.LlamaCpp.Docker == nil {
+			cfg.Backends.LlamaCpp.Docker = &DockerSettings{}
+		}
+		cfg.Backends.LlamaCpp.Docker.Image = llamaDockerImage
+	}
+	if llamaDockerArgs := os.Getenv("LLAMACTL_LLAMACPP_DOCKER_ARGS"); llamaDockerArgs != "" {
+		if cfg.Backends.LlamaCpp.Docker == nil {
+			cfg.Backends.LlamaCpp.Docker = &DockerSettings{}
+		}
+		cfg.Backends.LlamaCpp.Docker.Args = strings.Split(llamaDockerArgs, " ")
+	}
+	if llamaDockerEnv := os.Getenv("LLAMACTL_LLAMACPP_DOCKER_ENV"); llamaDockerEnv != "" {
+		if cfg.Backends.LlamaCpp.Docker == nil {
+			cfg.Backends.LlamaCpp.Docker = &DockerSettings{}
+		}
+		if cfg.Backends.LlamaCpp.Docker.Environment == nil {
+			cfg.Backends.LlamaCpp.Docker.Environment = make(map[string]string)
+		}
+		parseEnvVars(llamaDockerEnv, cfg.Backends.LlamaCpp.Docker.Environment)
+	}
+	if llamaEnv := os.Getenv("LLAMACTL_LLAMACPP_RESPONSE_HEADERS"); llamaEnv != "" {
+		if cfg.Backends.LlamaCpp.ResponseHeaders == nil {
+			cfg.Backends.LlamaCpp.ResponseHeaders = make(map[string]string)
+		}
+		parseHeaders(llamaEnv, cfg.Backends.LlamaCpp.ResponseHeaders)
+	}
+
+	// vLLM backend
+	if vllmCmd := os.Getenv("LLAMACTL_VLLM_COMMAND"); vllmCmd != "" {
+		cfg.Backends.VLLM.Command = vllmCmd
+	}
+	if vllmArgs := os.Getenv("LLAMACTL_VLLM_ARGS"); vllmArgs != "" {
+		cfg.Backends.VLLM.Args = strings.Split(vllmArgs, " ")
+	}
+	if vllmEnv := os.Getenv("LLAMACTL_VLLM_ENV"); vllmEnv != "" {
+		if cfg.Backends.VLLM.Environment == nil {
+			cfg.Backends.VLLM.Environment = make(map[string]string)
+		}
+		parseEnvVars(vllmEnv, cfg.Backends.VLLM.Environment)
+	}
+	if vllmDockerEnabled := os.Getenv("LLAMACTL_VLLM_DOCKER_ENABLED"); vllmDockerEnabled != "" {
+		if b, err := strconv.ParseBool(vllmDockerEnabled); err == nil {
+			if cfg.Backends.VLLM.Docker == nil {
+				cfg.Backends.VLLM.Docker = &DockerSettings{}
+			}
+			cfg.Backends.VLLM.Docker.Enabled = b
+		}
+	}
+	if vllmDockerImage := os.Getenv("LLAMACTL_VLLM_DOCKER_IMAGE"); vllmDockerImage != "" {
+		if cfg.Backends.VLLM.Docker == nil {
+			cfg.Backends.VLLM.Docker = &DockerSettings{}
+		}
+		cfg.Backends.VLLM.Docker.Image = vllmDockerImage
+	}
+	if vllmDockerArgs := os.Getenv("LLAMACTL_VLLM_DOCKER_ARGS"); vllmDockerArgs != "" {
+		if cfg.Backends.VLLM.Docker == nil {
+			cfg.Backends.VLLM.Docker = &DockerSettings{}
+		}
+		cfg.Backends.VLLM.Docker.Args = strings.Split(vllmDockerArgs, " ")
+	}
+	if vllmDockerEnv := os.Getenv("LLAMACTL_VLLM_DOCKER_ENV"); vllmDockerEnv != "" {
+		if cfg.Backends.VLLM.Docker == nil {
+			cfg.Backends.VLLM.Docker = &DockerSettings{}
+		}
+		if cfg.Backends.VLLM.Docker.Environment == nil {
+			cfg.Backends.VLLM.Docker.Environment = make(map[string]string)
+		}
+		parseEnvVars(vllmDockerEnv, cfg.Backends.VLLM.Docker.Environment)
+	}
+	if llamaEnv := os.Getenv("LLAMACTL_VLLM_RESPONSE_HEADERS"); llamaEnv != "" {
+		if cfg.Backends.VLLM.ResponseHeaders == nil {
+			cfg.Backends.VLLM.ResponseHeaders = make(map[string]string)
+		}
+		parseHeaders(llamaEnv, cfg.Backends.VLLM.ResponseHeaders)
+	}
+
+	// MLX backend
+	if mlxCmd := os.Getenv("LLAMACTL_MLX_COMMAND"); mlxCmd != "" {
+		cfg.Backends.MLX.Command = mlxCmd
+	}
+	if mlxArgs := os.Getenv("LLAMACTL_MLX_ARGS"); mlxArgs != "" {
+		cfg.Backends.MLX.Args = strings.Split(mlxArgs, " ")
+	}
+	if mlxEnv := os.Getenv("LLAMACTL_MLX_ENV"); mlxEnv != "" {
+		if cfg.Backends.MLX.Environment == nil {
+			cfg.Backends.MLX.Environment = make(map[string]string)
+		}
+		parseEnvVars(mlxEnv, cfg.Backends.MLX.Environment)
+	}
+	if llamaEnv := os.Getenv("LLAMACTL_MLX_RESPONSE_HEADERS"); llamaEnv != "" {
+		if cfg.Backends.MLX.ResponseHeaders == nil {
+			cfg.Backends.MLX.ResponseHeaders = make(map[string]string)
+		}
+		parseHeaders(llamaEnv, cfg.Backends.MLX.ResponseHeaders)
+	}
+
+	// Instance defaults
+	if autoRestart := os.Getenv("LLAMACTL_DEFAULT_AUTO_RESTART"); autoRestart != "" {
+		if b, err := strconv.ParseBool(autoRestart); err == nil {
+			cfg.Instances.DefaultAutoRestart = b
+		}
+	}
+	if maxRestarts := os.Getenv("LLAMACTL_DEFAULT_MAX_RESTARTS"); maxRestarts != "" {
+		if m, err := strconv.Atoi(maxRestarts); err == nil {
+			cfg.Instances.DefaultMaxRestarts = m
+		}
+	}
+	if restartDelay := os.Getenv("LLAMACTL_DEFAULT_RESTART_DELAY"); restartDelay != "" {
+		if seconds, err := strconv.Atoi(restartDelay); err == nil {
+			cfg.Instances.DefaultRestartDelay = seconds
+		}
+	}
+	if onDemandStart := os.Getenv("LLAMACTL_DEFAULT_ON_DEMAND_START"); onDemandStart != "" {
+		if b, err := strconv.ParseBool(onDemandStart); err == nil {
+			cfg.Instances.DefaultOnDemandStart = b
+		}
+	}
+	if onDemandTimeout := os.Getenv("LLAMACTL_ON_DEMAND_START_TIMEOUT"); onDemandTimeout != "" {
+		if seconds, err := strconv.Atoi(onDemandTimeout); err == nil {
+			cfg.Instances.OnDemandStartTimeout = seconds
+		}
+	}
+	if timeoutCheckInterval := os.Getenv("LLAMACTL_TIMEOUT_CHECK_INTERVAL"); timeoutCheckInterval != "" {
+		if minutes, err := strconv.Atoi(timeoutCheckInterval); err == nil {
+			cfg.Instances.TimeoutCheckInterval = minutes
+		}
+	}
+	// Auth config
+	if requireInferenceAuth := os.Getenv("LLAMACTL_REQUIRE_INFERENCE_AUTH"); requireInferenceAuth != "" {
+		if b, err := strconv.ParseBool(requireInferenceAuth); err == nil {
+			cfg.Auth.RequireInferenceAuth = b
+		}
+	}
+	if inferenceKeys := os.Getenv("LLAMACTL_INFERENCE_KEYS"); inferenceKeys != "" {
+		cfg.Auth.InferenceKeys = strings.Split(inferenceKeys, ",")
+	}
+	if requireManagementAuth := os.Getenv("LLAMACTL_REQUIRE_MANAGEMENT_AUTH"); requireManagementAuth != "" {
+		if b, err := strconv.ParseBool(requireManagementAuth); err == nil {
+			cfg.Auth.RequireManagementAuth = b
+		}
+	}
+	if managementKeys := os.Getenv("LLAMACTL_MANAGEMENT_KEYS"); managementKeys != "" {
+		cfg.Auth.ManagementKeys = strings.Split(managementKeys, ",")
+	}
+
+	// Local node config
+	if localNode := os.Getenv("LLAMACTL_LOCAL_NODE"); localNode != "" {
+		cfg.LocalNode = localNode
+	}
+
+	// Database config
+	if dbPath := os.Getenv("LLAMACTL_DATABASE_PATH"); dbPath != "" {
+		cfg.Database.Path = dbPath
+	}
+	if maxOpenConns := os.Getenv("LLAMACTL_DATABASE_MAX_OPEN_CONNECTIONS"); maxOpenConns != "" {
+		if m, err := strconv.Atoi(maxOpenConns); err == nil {
+			cfg.Database.MaxOpenConnections = m
+		}
+	}
+	if maxIdleConns := os.Getenv("LLAMACTL_DATABASE_MAX_IDLE_CONNECTIONS"); maxIdleConns != "" {
+		if m, err := strconv.Atoi(maxIdleConns); err == nil {
+			cfg.Database.MaxIdleConnections = m
+		}
+	}
+	if connMaxLifetime := os.Getenv("LLAMACTL_DATABASE_CONN_MAX_LIFETIME"); connMaxLifetime != "" {
+		if d, err := time.ParseDuration(connMaxLifetime); err == nil {
+			cfg.Database.ConnMaxLifetime = d
+		}
+	}
+
+	// Log rotation config
+	if logRotationEnabled := os.Getenv("LLAMACTL_LOG_ROTATION_ENABLED"); logRotationEnabled != "" {
+		if b, err := strconv.ParseBool(logRotationEnabled); err == nil {
+			cfg.Instances.LogRotationEnabled = b
+		}
+	}
+	if logRotationMaxSize := os.Getenv("LLAMACTL_LOG_ROTATION_MAX_SIZE"); logRotationMaxSize != "" {
+		if m, err := strconv.Atoi(logRotationMaxSize); err == nil {
+			cfg.Instances.LogRotationMaxSize = m
+		}
+	}
+	if logRotationCompress := os.Getenv("LLAMACTL_LOG_ROTATION_COMPRESS"); logRotationCompress != "" {
+		if b, err := strconv.ParseBool(logRotationCompress); err == nil {
+			cfg.Instances.LogRotationCompress = b
+		}
+	}
+}
+
+// ParsePortRange parses port range from string formats like "8000-9000" or "8000,9000"
+func ParsePortRange(s string) [2]int {
+	var parts []string
+
+	// Try both separators
+	if strings.Contains(s, "-") {
+		parts = strings.Split(s, "-")
+	} else if strings.Contains(s, ",") {
+		parts = strings.Split(s, ",")
+	}
+
+	// Parse the two parts
+	if len(parts) == 2 {
+		start, err1 := strconv.Atoi(strings.TrimSpace(parts[0]))
+		end, err2 := strconv.Atoi(strings.TrimSpace(parts[1]))
+		if err1 == nil && err2 == nil {
+			return [2]int{start, end}
+		}
+	}
+
+	return [2]int{0, 0} // Invalid format
+}
+
+// parseEnvVars parses environment variables in format "KEY1=value1,KEY2=value2"
+// and populates the provided environment map
+func parseEnvVars(envString string, envMap map[string]string) {
+	if envString == "" {
+		return
+	}
+	for _, envPair := range strings.Split(envString, ",") {
+		if parts := strings.SplitN(strings.TrimSpace(envPair), "=", 2); len(parts) == 2 {
+			envMap[parts[0]] = parts[1]
+		}
+	}
+}
+
+// parseHeaders parses HTTP headers in format "KEY1=value1;KEY2=value2"
+// and populates the provided environment map
+func parseHeaders(envString string, envMap map[string]string) {
+	if envString == "" {
+		return
+	}
+	for _, envPair := range strings.Split(envString, ";") {
+		if parts := strings.SplitN(strings.TrimSpace(envPair), "=", 2); len(parts) == 2 {
+			envMap[parts[0]] = parts[1]
+		}
+	}
+}

pkg/config/types.go

@@ -0,0 +1,149 @@
+package config
+
+import "time"
+
+// BackendSettings contains structured backend configuration
+type BackendSettings struct {
+	Command         string            `yaml:"command" json:"command"`
+	Args            []string          `yaml:"args" json:"args"`
+	Environment     map[string]string `yaml:"environment,omitempty" json:"environment,omitempty"`
+	Docker          *DockerSettings   `yaml:"docker,omitempty" json:"docker,omitempty"`
+	ResponseHeaders map[string]string `yaml:"response_headers,omitempty" json:"response_headers,omitempty"`
+}
+
+// DockerSettings contains Docker-specific configuration
+type DockerSettings struct {
+	Enabled     bool              `yaml:"enabled" json:"enabled"`
+	Image       string            `yaml:"image" json:"image"`
+	Args        []string          `yaml:"args" json:"args"`
+	Environment map[string]string `yaml:"environment,omitempty" json:"environment,omitempty"`
+}
+
+// BackendConfig contains backend executable configurations
+type BackendConfig struct {
+	LlamaCpp BackendSettings `yaml:"llama-cpp" json:"llama-cpp"`
+	VLLM     BackendSettings `yaml:"vllm" json:"vllm"`
+	MLX      BackendSettings `yaml:"mlx" json:"mlx"`
+}
+
+// AppConfig represents the configuration for llamactl
+type AppConfig struct {
+	Server    ServerConfig          `yaml:"server" json:"server"`
+	Backends  BackendConfig         `yaml:"backends" json:"backends"`
+	Instances InstancesConfig       `yaml:"instances" json:"instances"`
+	Database  DatabaseConfig        `yaml:"database" json:"database"`
+	Auth      AuthConfig            `yaml:"auth" json:"auth"`
+	LocalNode string                `yaml:"local_node,omitempty" json:"local_node,omitempty"`
+	Nodes     map[string]NodeConfig `yaml:"nodes,omitempty" json:"nodes,omitempty"`
+
+	// Directory where all llamactl data will be stored (database, instances, logs, etc.)
+	DataDir string `yaml:"data_dir" json:"data_dir"`
+
+	Version    string `yaml:"-" json:"version"`
+	CommitHash string `yaml:"-" json:"commit_hash"`
+	BuildTime  string `yaml:"-" json:"build_time"`
+}
+
+// ServerConfig contains HTTP server configuration
+type ServerConfig struct {
+	// Server host to bind to
+	Host string `yaml:"host" json:"host"`
+
+	// Server port to bind to
+	Port int `yaml:"port" json:"port"`
+
+	// Allowed origins for CORS (e.g., "http://localhost:3000")
+	AllowedOrigins []string `yaml:"allowed_origins" json:"allowed_origins"`
+
+	// Allowed headers for CORS (e.g., "Accept", "Authorization", "Content-Type", "X-CSRF-Token")
+	AllowedHeaders []string `yaml:"allowed_headers" json:"allowed_headers"`
+
+	// Enable Swagger UI for API documentation
+	EnableSwagger bool `yaml:"enable_swagger" json:"enable_swagger"`
+
+	// Response headers to send with responses
+	ResponseHeaders map[string]string `yaml:"response_headers,omitempty" json:"response_headers,omitempty"`
+}
+
+// DatabaseConfig contains database configuration settings
+type DatabaseConfig struct {
+	// Database file path (relative to the top-level data_dir or absolute)
+	Path string `yaml:"path" json:"path"`
+
+	// Connection settings
+	MaxOpenConnections int           `yaml:"max_open_connections" json:"max_open_connections"`
+	MaxIdleConnections int           `yaml:"max_idle_connections" json:"max_idle_connections"`
+	ConnMaxLifetime    time.Duration `yaml:"connection_max_lifetime" json:"connection_max_lifetime" swaggertype:"string" example:"1h"`
+}
+
+// InstancesConfig contains instance management configuration
+type InstancesConfig struct {
+	// Port range for instances (e.g., 8000,9000)
+	PortRange [2]int `yaml:"port_range" json:"port_range"`
+
+	// Instance config directory override (relative to data_dir if not absolute)
+	InstancesDir string `yaml:"configs_dir" json:"configs_dir"`
+
+	// Automatically create the data directory if it doesn't exist
+	AutoCreateDirs bool `yaml:"auto_create_dirs" json:"auto_create_dirs"`
+
+	// Maximum number of instances that can be created
+	MaxInstances int `yaml:"max_instances" json:"max_instances"`
+
+	// Maximum number of instances that can be running at the same time
+	MaxRunningInstances int `yaml:"max_running_instances,omitempty" json:"max_running_instances,omitempty"`
+
+	// Enable LRU eviction for instance logs
+	EnableLRUEviction bool `yaml:"enable_lru_eviction" json:"enable_lru_eviction"`
+
+	// Default auto-restart setting for new instances
+	DefaultAutoRestart bool `yaml:"default_auto_restart" json:"default_auto_restart"`
+
+	// Default max restarts for new instances
+	DefaultMaxRestarts int `yaml:"default_max_restarts" json:"default_max_restarts"`
+
+	// Default restart delay for new instances (in seconds)
+	DefaultRestartDelay int `yaml:"default_restart_delay" json:"default_restart_delay"`
+
+	// Default on-demand start setting for new instances
+	DefaultOnDemandStart bool `yaml:"default_on_demand_start" json:"default_on_demand_start"`
+
+	// How long to wait for an instance to start on demand (in seconds)
+	OnDemandStartTimeout int `yaml:"on_demand_start_timeout,omitempty" json:"on_demand_start_timeout,omitempty"`
+
+	// Interval for checking instance timeouts (in minutes)
+	TimeoutCheckInterval int `yaml:"timeout_check_interval" json:"timeout_check_interval"`
+
+	// Logs directory override (relative to data_dir if not absolute)
+	LogsDir string `yaml:"logs_dir" json:"logs_dir"`
+
+	// Log rotation enabled
+	LogRotationEnabled bool `yaml:"log_rotation_enabled" default:"true"`
+
+	// Maximum log file size in MB before rotation
+	LogRotationMaxSize int `yaml:"log_rotation_max_size" default:"100"`
+
+	// Whether to compress rotated log files
+	LogRotationCompress bool `yaml:"log_rotation_compress" default:"false"`
+}
+
+// AuthConfig contains authentication settings
+type AuthConfig struct {
+
+	// Require authentication for OpenAI compatible inference endpoints
+	RequireInferenceAuth bool `yaml:"require_inference_auth" json:"require_inference_auth"`
+
+	// List of keys for OpenAI compatible inference endpoints
+	InferenceKeys []string `yaml:"inference_keys" json:"inference_keys"`
+
+	// Require authentication for management endpoints
+	RequireManagementAuth bool `yaml:"require_management_auth" json:"require_management_auth"`
+
+	// List of keys for management endpoints
+	ManagementKeys []string `yaml:"management_keys" json:"management_keys"`
+}
+
+type NodeConfig struct {
+	Address string `yaml:"address" json:"address"`
+	APIKey  string `yaml:"api_key,omitempty" json:"api_key,omitempty"`
+}

pkg/database/database.go

@@ -107,6 +107,12 @@ func Open(config *Config) (*sqliteDB, error) {
 func (db *sqliteDB) Close() error {
 	if db.DB != nil {
 		log.Println("Closing database connection")
+
+		// Checkpoint WAL to merge changes back to main database file
+		if _, err := db.DB.Exec("PRAGMA wal_checkpoint(TRUNCATE)"); err != nil {
+			log.Printf("Warning: Failed to checkpoint WAL: %v", err)
+		}
+
 		return db.DB.Close()
 	}
 	return nil

pkg/database/instances.go

@@ -45,7 +45,7 @@ func (db *sqliteDB) Create(ctx context.Context, inst *instance.Instance) error {
 		) VALUES (?, ?, ?, ?, ?, ?)
 	`
 
-	_, err = db.DB.ExecContext(ctx, query,
+	result, err := db.DB.ExecContext(ctx, query,
 		row.Name, row.Status, row.CreatedAt, row.UpdatedAt, row.OptionsJSON, row.OwnerUserID,
 	)
 
@@ -53,6 +53,14 @@ func (db *sqliteDB) Create(ctx context.Context, inst *instance.Instance) error {
 		return fmt.Errorf("failed to insert instance: %w", err)
 	}
 
+	// Get the auto-generated ID and set it on the instance
+	id, err := result.LastInsertId()
+	if err != nil {
+		return fmt.Errorf("failed to get last insert ID: %w", err)
+	}
+
+	inst.ID = int(id)
+
 	return nil
 }
 
@@ -263,6 +271,7 @@ func (db *sqliteDB) rowToInstance(row *instanceRow) (*instance.Instance, error)
 
 	// Build complete instance JSON with all fields
 	instanceJSON, err := json.Marshal(map[string]any{
+		"id":      row.ID,
 		"name":    row.Name,
 		"created": row.CreatedAt,
 		"status":  row.Status,

pkg/instance/instance.go

@@ -3,10 +3,11 @@ package instance
 import (
 	"encoding/json"
 	"fmt"
-	"llamactl/pkg/config"
 	"log"
 	"net/http"
 	"time"
+
+	"llamactl/pkg/config"
 )
 
 // Instance represents a running instance of llama server
@@ -68,7 +69,16 @@ func New(name string, globalConfig *config.AppConfig, opts *Options, onStatusCha
 
 	// Only create logger, proxy, and process for local instances
 	if !instance.IsRemote() {
-		instance.logger = newLogger(name, globalInstanceSettings.LogsDir)
+		logRotationConfig := &LogRotationConfig{
+			Enabled:  globalInstanceSettings.LogRotationEnabled,
+			MaxSize:  globalInstanceSettings.LogRotationMaxSize,
+			Compress: globalInstanceSettings.LogRotationCompress,
+		}
+		instance.logger = newLogger(
+			name,
+			globalInstanceSettings.LogsDir,
+			logRotationConfig,
+		)
 		instance.process = newProcess(instance)
 	}
 

pkg/instance/instance_test.go

@@ -27,8 +27,8 @@ func TestNewInstance(t *testing.T) {
 			},
 		},
 		Instances: config.InstancesConfig{
-			LogsDir:             "/tmp/test",
 			DefaultAutoRestart:  true,
+			LogsDir:             "/tmp/test",
 			DefaultMaxRestarts:  3,
 			DefaultRestartDelay: 5,
 		},
@@ -120,8 +120,8 @@ func TestSetOptions(t *testing.T) {
 			},
 		},
 		Instances: config.InstancesConfig{
-			LogsDir:             "/tmp/test",
 			DefaultAutoRestart:  true,
+			LogsDir:             "/tmp/test",
 			DefaultMaxRestarts:  3,
 			DefaultRestartDelay: 5,
 		},

pkg/instance/logger.go

@@ -7,66 +7,117 @@ import (
 	"os"
 	"strings"
 	"sync"
-	"sync/atomic"
 	"time"
+
+	timber "github.com/DeRuina/timberjack"
 )
 
+// LogRotationConfig contains log rotation settings for instances
+type LogRotationConfig struct {
+	Enabled  bool
+	MaxSize  int
+	Compress bool
+}
+
 type logger struct {
 	name        string
 	logDir      string
-	logFile     atomic.Pointer[os.File]
+	logFile     *timber.Logger
 	logFilePath string
 	mu          sync.RWMutex
+	cfg         *LogRotationConfig
 }
 
-func newLogger(name string, logDir string) *logger {
+func newLogger(name, logDir string, cfg *LogRotationConfig) *logger {
 	return &logger{
 		name:   name,
 		logDir: logDir,
+		cfg:    cfg,
 	}
 }
 
-// create creates and opens the log files for stdout and stderr
-func (i *logger) create() error {
-	i.mu.Lock()
-	defer i.mu.Unlock()
+func (l *logger) create() error {
+	l.mu.Lock()
+	defer l.mu.Unlock()
 
-	if i.logDir == "" {
-		return fmt.Errorf("logDir is empty for instance %s", i.name)
+	if l.logDir == "" {
+		return fmt.Errorf("logDir empty for instance %s", l.name)
 	}
 
-	// Set up instance logs
-	logPath := i.logDir + "/" + i.name + ".log"
-
-	i.logFilePath = logPath
-	if err := os.MkdirAll(i.logDir, 0755); err != nil {
+	if err := os.MkdirAll(l.logDir, 0755); err != nil {
 		return fmt.Errorf("failed to create log directory: %w", err)
 	}
 
-	logFile, err := os.OpenFile(logPath, os.O_CREATE|os.O_WRONLY|os.O_APPEND, 0644)
-	if err != nil {
-		return fmt.Errorf("failed to create stdout log file: %w", err)
+	logPath := fmt.Sprintf("%s/%s.log", l.logDir, l.name)
+	l.logFilePath = logPath
+
+	// Build the timber logger
+	t := &timber.Logger{
+		Filename:   logPath,
+		MaxSize:    l.cfg.MaxSize,
+		MaxBackups: 0, // No limit on backups
+		// Compression: "gzip" if Compress is true, else "none"
+		Compression: func() string {
+			if l.cfg.Compress {
+				return "gzip"
+			}
+			return "none"
+		}(),
+		FileMode:  0644,
+		LocalTime: true,
 	}
 
-	i.logFile.Store(logFile)
+	// If rotation is disabled, set MaxSize to 0 so no rotation occurs
+	if !l.cfg.Enabled {
+		t.MaxSize = 0
+	}
 
-	// Write a startup marker to both files
-	timestamp := time.Now().Format("2006-01-02 15:04:05")
-	fmt.Fprintf(logFile, "\n=== Instance %s started at %s ===\n", i.name, timestamp)
+	l.logFile = t
+
+	// Write a startup marker
+	ts := time.Now().Format("2006-01-02 15:04:05")
+	fmt.Fprintf(t, "\n=== Instance %s started at %s ===\n", l.name, ts)
 
 	return nil
 }
 
-// getLogs retrieves the last n lines of logs from the instance
-func (i *logger) getLogs(num_lines int) (string, error) {
-	i.mu.RLock()
-	defer i.mu.RUnlock()
+func (l *logger) readOutput(rc io.ReadCloser) {
+	defer rc.Close()
+	scanner := bufio.NewScanner(rc)
+	for scanner.Scan() {
+		line := scanner.Text()
+		if lg := l.logFile; lg != nil {
+			fmt.Fprintln(lg, line)
+		}
+	}
+}
 
-	if i.logFilePath == "" {
-		return "", fmt.Errorf("log file not created for instance %s", i.name)
+func (l *logger) close() {
+	l.mu.Lock()
+	defer l.mu.Unlock()
+
+	lg := l.logFile
+	if lg == nil {
+		return
 	}
 
-	file, err := os.Open(i.logFilePath)
+	ts := time.Now().Format("2006-01-02 15:04:05")
+	fmt.Fprintf(lg, "=== Instance %s stopped at %s ===\n\n", l.name, ts)
+
+	_ = lg.Close()
+	l.logFile = nil
+}
+
+// getLogs retrieves the last n lines of logs from the instance
+func (l *logger) getLogs(num_lines int) (string, error) {
+	l.mu.RLock()
+	defer l.mu.RUnlock()
+
+	if l.logFilePath == "" {
+		return "", fmt.Errorf("log file not created for instance %s", l.name)
+	}
+
+	file, err := os.Open(l.logFilePath)
 	if err != nil {
 		return "", fmt.Errorf("failed to open log file: %w", err)
 	}
@@ -97,31 +148,3 @@ func (i *logger) getLogs(num_lines int) (string, error) {
 
 	return strings.Join(lines[start:], "\n"), nil
 }
-
-// close closes the log files
-func (i *logger) close() {
-	i.mu.Lock()
-	defer i.mu.Unlock()
-
-	logFile := i.logFile.Swap(nil)
-	if logFile != nil {
-		timestamp := time.Now().Format("2006-01-02 15:04:05")
-		fmt.Fprintf(logFile, "=== Instance %s stopped at %s ===\n\n", i.name, timestamp)
-		logFile.Sync() // Ensure all buffered data is written to disk
-		logFile.Close()
-	}
-}
-
-// readOutput reads from the given reader and writes lines to the log file
-func (i *logger) readOutput(reader io.ReadCloser) {
-	defer reader.Close()
-
-	scanner := bufio.NewScanner(reader)
-	for scanner.Scan() {
-		line := scanner.Text()
-		// Use atomic load to avoid lock contention on every line
-		if logFile := i.logFile.Load(); logFile != nil {
-			fmt.Fprintln(logFile, line)
-		}
-	}
-}

pkg/manager/manager.go

@@ -114,11 +114,6 @@ func (im *instanceManager) Shutdown() {
 		}
 		wg.Wait()
 		fmt.Println("All instances stopped.")
-
-		// 4. Close database connection
-		if err := im.db.Close(); err != nil {
-			log.Printf("Error closing database: %v\n", err)
-		}
 	})
 }
 
@@ -181,6 +176,7 @@ func (im *instanceManager) loadInstance(persistedInst *instance.Instance) error
 	inst := instance.New(name, im.globalConfig, options, statusCallback)
 
 	// Restore persisted fields that NewInstance doesn't set
+	inst.ID = persistedInst.ID
 	inst.Created = persistedInst.Created
 	inst.SetStatus(persistedInst.GetStatus())
 

pkg/manager/manager_test.go

@@ -203,11 +203,11 @@ func createTestAppConfig(instancesDir string) *config.AppConfig {
 		Instances: config.InstancesConfig{
 			PortRange:            [2]int{8000, 9000},
 			InstancesDir:         instancesDir,
-			LogsDir:              instancesDir,
 			MaxInstances:         10,
 			MaxRunningInstances:  10,
 			DefaultAutoRestart:   true,
 			DefaultMaxRestarts:   3,
+			LogsDir:              instancesDir,
 			DefaultRestartDelay:  5,
 			TimeoutCheckInterval: 5,
 		},

pkg/manager/operations.go

@@ -37,7 +37,6 @@ func (im *instanceManager) ListInstances() ([]*instance.Instance, error) {
 		if node := im.getNodeForInstance(inst); node != nil {
 			remoteInst, err := im.remote.getInstance(ctx, node, inst.Name)
 			if err != nil {
-				// Log error but continue with stale data
 				// Don't fail the entire list operation due to one remote failure
 				continue
 			}

pkg/server/handlers_auth.go

@@ -11,17 +11,12 @@ import (
 	"github.com/go-chi/chi/v5"
 )
 
-// InstancePermission defines the permissions for an API key on a specific instance.
-type InstancePermission struct {
-	InstanceID int `json:"instance_id"`
-}
-
 // CreateKeyRequest represents the request body for creating a new API key.
 type CreateKeyRequest struct {
-	Name                string
-	PermissionMode      auth.PermissionMode
-	ExpiresAt           *int64
-	InstancePermissions []InstancePermission
+	Name           string              `json:"name"`
+	PermissionMode auth.PermissionMode `json:"permission_mode"`
+	ExpiresAt      *int64              `json:"expires_at,omitempty"`
+	InstanceIDs    []int               `json:"instance_ids,omitempty"`
 }
 
 // CreateKeyResponse represents the response returned when creating a new API key.
@@ -87,8 +82,8 @@ func (h *Handler) CreateKey() http.HandlerFunc {
 			writeError(w, http.StatusBadRequest, "invalid_permission_mode", "Permission mode must be 'allow_all' or 'per_instance'")
 			return
 		}
-		if req.PermissionMode == auth.PermissionModePerInstance && len(req.InstancePermissions) == 0 {
-			writeError(w, http.StatusBadRequest, "missing_permissions", "Instance permissions required when permission mode is 'per_instance'")
+		if req.PermissionMode == auth.PermissionModePerInstance && len(req.InstanceIDs) == 0 {
+			writeError(w, http.StatusBadRequest, "missing_permissions", "Instance IDs required when permission mode is 'per_instance'")
 			return
 		}
 		if req.ExpiresAt != nil && *req.ExpiresAt <= time.Now().Unix() {
@@ -108,9 +103,9 @@ func (h *Handler) CreateKey() http.HandlerFunc {
 				instanceIDMap[inst.ID] = true
 			}
 
-			for _, perm := range req.InstancePermissions {
-				if !instanceIDMap[perm.InstanceID] {
-					writeError(w, http.StatusBadRequest, "invalid_instance_id", fmt.Sprintf("Instance ID %d does not exist", perm.InstanceID))
+			for _, instanceID := range req.InstanceIDs {
+				if !instanceIDMap[instanceID] {
+					writeError(w, http.StatusBadRequest, "invalid_instance_id", fmt.Sprintf("Instance ID %d does not exist", instanceID))
 					return
 				}
 			}
@@ -142,12 +137,12 @@ func (h *Handler) CreateKey() http.HandlerFunc {
 			UpdatedAt:      now,
 		}
 
-		// Convert InstancePermissions to KeyPermissions
+		// Convert InstanceIDs to KeyPermissions
 		var keyPermissions []auth.KeyPermission
-		for _, perm := range req.InstancePermissions {
+		for _, instanceID := range req.InstanceIDs {
 			keyPermissions = append(keyPermissions, auth.KeyPermission{
 				KeyID:      0, // Will be set by database after key creation
-				InstanceID: perm.InstanceID,
+				InstanceID: instanceID,
 			})
 		}
 

test_client.py

@@ -0,0 +1,136 @@
+#!/usr/bin/env python3
+"""
+Simple Python script to interact with local LLM server's OpenAI-compatible API
+"""
+
+import requests
+import json
+import sys
+
+# Local LLM server configuration
+BASE_URL = "http://localhost:8080"
+API_KEY = None
+MODEL_NAME = None
+
+def get_models():
+    """Fetch available models from /v1/models endpoint"""
+    headers = {}
+    if API_KEY:
+        headers["Authorization"] = f"Bearer {API_KEY}"
+
+    try:
+        response = requests.get(f"{BASE_URL}/v1/models", headers=headers, timeout=10)
+        response.raise_for_status()
+        return response.json()["data"]
+    except Exception as e:
+        print(f"Error fetching models: {e}")
+        return []
+
+def send_message(message):
+    """
+    Send a message to local LLM server API
+
+    Args:
+        message (str): The message to send
+
+    Returns:
+        str: The AI response or error message
+    """
+
+    headers = {
+        "Content-Type": "application/json",
+    }
+
+    if API_KEY:
+        headers["Authorization"] = f"Bearer {API_KEY}"
+
+    data = {
+        "model": MODEL_NAME,
+        "messages": [
+            {
+                "role": "user",
+                "content": message
+            }
+        ],
+        "temperature": 0.7,
+        "max_tokens": 1000,
+        "stream": False,
+    }
+    
+    response = requests.post(f"{BASE_URL}/v1/chat/completions", headers=headers, json=data, timeout=60)
+    response.raise_for_status()
+    return response.json()["choices"][0]["message"]["content"]
+
+def interactive_mode():
+    """Run in interactive mode for continuous conversation"""
+    global BASE_URL, API_KEY, MODEL_NAME
+
+    # Get base URL
+    url_input = input(f"Base URL [{BASE_URL}]: ").strip()
+    if url_input:
+        BASE_URL = url_input
+
+    # Get API key (optional)
+    key_input = input("API key (optional): ").strip()
+    if key_input:
+        API_KEY = key_input
+
+    # Fetch and select model
+    models = get_models()
+    if not models:
+        print("No models available. Exiting.")
+        return
+
+    print("\nAvailable models:")
+    for i, m in enumerate(models, 1):
+        print(f"{i}. {m['id']}")
+
+    while True:
+        try:
+            selection = int(input("\nSelect model: "))
+            if 1 <= selection <= len(models):
+                MODEL_NAME = models[selection - 1]["id"]
+                break
+            print(f"Please enter a number between 1 and {len(models)}")
+        except ValueError:
+            print("Please enter a valid number")
+
+    print(f"\nUsing model: {MODEL_NAME}")
+    print("Type 'quit' or 'exit' to stop")
+    print("-" * 40)
+    
+    while True:
+        try:
+            user_input = input("\nYou: ").strip()
+            
+            if user_input.lower() in ['quit', 'exit', 'q']:
+                print("Goodbye!")
+                break
+                
+            if not user_input:
+                continue
+                
+            print("AI: ", end="", flush=True)
+            response = send_message(user_input)
+            print(response)
+            
+        except KeyboardInterrupt:
+            print("\nGoodbye!")
+            break
+        except EOFError:
+            print("\nGoodbye!")
+            break
+
+def main():
+    """Main function"""
+    if len(sys.argv) > 1:
+        # Single message mode
+        message = " ".join(sys.argv[1:])
+        response = send_message(message)
+        print(response)
+    else:
+        # Interactive mode
+        interactive_mode()
+
+if __name__ == "__main__":
+    main()

webui/package.json

@@ -18,30 +18,29 @@
     "lint:fix": "eslint . --ext .ts,.tsx --fix"
   },
   "dependencies": {
-    "@radix-ui/react-checkbox": "^1.3.2",
-    "@radix-ui/react-dialog": "^1.1.14",
-    "@radix-ui/react-label": "^2.1.7",
+    "@radix-ui/react-checkbox": "^1.3.3",
+    "@radix-ui/react-dialog": "^1.1.15",
+    "@radix-ui/react-label": "^2.1.8",
     "@radix-ui/react-radio-group": "^1.3.8",
-    "@radix-ui/react-slot": "^1.2.3",
+    "@radix-ui/react-slot": "^1.2.4",
     "@tailwindcss/vite": "^4.1.11",
     "class-variance-authority": "^0.7.1",
     "clsx": "^2.1.1",
     "date-fns": "^4.1.0",
-    "lucide-react": "^0.555.0",
+    "lucide-react": "^0.561.0",
     "react": "^19.2.0",
     "react-dom": "^19.2.0",
     "sonner": "^2.0.7",
     "tailwind-merge": "^3.4.0",
     "tailwindcss": "^4.1.11",
-    "zod": "^4.1.12"
+    "zod": "^4.2.0"
   },
   "devDependencies": {
     "@eslint/js": "^9.39.1",
     "@testing-library/jest-dom": "^6.9.1",
     "@testing-library/react": "^16.3.0",
     "@testing-library/user-event": "^14.6.1",
-    "@types/eslint__js": "^9.14.0",
-    "@types/node": "^24.10.1",
+    "@types/node": "^25.0.2",
     "@types/react": "^19.2.4",
     "@types/react-dom": "^19.2.3",
     "@vitejs/plugin-react": "^5.1.1",
@@ -50,11 +49,11 @@
     "eslint-plugin-react": "^7.37.5",
     "eslint-plugin-react-hooks": "^7.0.1",
     "eslint-plugin-react-refresh": "^0.4.20",
-    "jsdom": "^27.2.0",
+    "jsdom": "^27.3.0",
     "tw-animate-css": "^1.4.0",
     "typescript": "^5.9.3",
-    "typescript-eslint": "^8.48.0",
-    "vite": "^7.2.2",
+    "typescript-eslint": "^8.50.0",
+    "vite": "^7.3.0",
     "vitest": "^4.0.8"
   }
 }

webui/src/components/__tests__/InstanceList.test.tsx

@@ -4,8 +4,7 @@ import userEvent from '@testing-library/user-event'
 import InstanceList from '@/components/InstanceList'
 import { InstancesProvider } from '@/contexts/InstancesContext'
 import { instancesApi } from '@/lib/api'
-import type { Instance } from '@/types/instance'
-import { BackendType } from '@/types/instance'
+import { BackendType, type Instance } from '@/types/instance'
 import { AuthProvider } from '@/contexts/AuthContext'
 
 // Mock the API

webui/src/components/apikeys/CreateApiKeyDialog.tsx

@@ -8,9 +8,9 @@ import { Checkbox } from "@/components/ui/checkbox";
 import { Alert, AlertDescription } from "@/components/ui/alert";
 import { Loader2 } from "lucide-react";
 import { apiKeysApi } from "@/lib/api";
-import { CreateKeyRequest, PermissionMode, InstancePermission } from "@/types/apiKey";
+import { PermissionMode, type CreateKeyRequest } from "@/types/apiKey";
 import { useInstances } from "@/contexts/InstancesContext";
-import { format, addDays } from "date-fns";
+import { format } from "date-fns";
 
 interface CreateApiKeyDialogProps {
   open: boolean;
@@ -61,22 +61,19 @@ function CreateApiKeyDialog({ open, onOpenChange, onKeyCreated }: CreateApiKeyDi
     }
 
     // Build request
-    const permissions: InstancePermission[] = [];
+    const instanceIds: number[] = [];
     if (permissionMode === PermissionMode.PerInstance) {
-      Object.entries(instancePermissions).forEach(([instanceId, canInfer]) => {
-        if (canInfer) {
-          permissions.push({
-            InstanceID: parseInt(instanceId),
-            CanInfer: true,
-          });
+      Object.entries(instancePermissions).forEach(([instanceId, hasPermission]) => {
+        if (hasPermission) {
+          instanceIds.push(parseInt(instanceId));
         }
       });
     }
 
     const request: CreateKeyRequest = {
-      Name: name.trim(),
-      PermissionMode: permissionMode,
-      InstancePermissions: permissions,
+      name: name.trim(),
+      permission_mode: permissionMode,
+      instance_ids: instanceIds,
     };
 
     // Add expiration if provided
@@ -87,7 +84,7 @@ function CreateApiKeyDialog({ open, onOpenChange, onKeyCreated }: CreateApiKeyDi
         setError("Expiration date must be in the future");
         return;
       }
-      request.ExpiresAt = Math.floor(expirationDate.getTime() / 1000);
+      request.expires_at = Math.floor(expirationDate.getTime() / 1000);
     }
 
     setLoading(true);
@@ -107,10 +104,10 @@ function CreateApiKeyDialog({ open, onOpenChange, onKeyCreated }: CreateApiKeyDi
   };
 
   const handleInstancePermissionChange = (instanceId: number, checked: boolean) => {
-    setInstancePermissions({
-      ...instancePermissions,
+    setInstancePermissions(prev => ({
+      ...prev,
       [instanceId]: checked,
-    });
+    }));
   };
 
   return (
@@ -119,7 +116,7 @@ function CreateApiKeyDialog({ open, onOpenChange, onKeyCreated }: CreateApiKeyDi
         <DialogHeader>
           <DialogTitle>Create API Key</DialogTitle>
         </DialogHeader>
-        <form onSubmit={handleSubmit} className="space-y-4">
+        <form onSubmit={(e) => void handleSubmit(e)} className="space-y-4">
           {error && (
             <Alert variant="destructive">
               <AlertDescription>{error}</AlertDescription>
@@ -172,25 +169,30 @@ function CreateApiKeyDialog({ open, onOpenChange, onKeyCreated }: CreateApiKeyDi
                   <p className="text-sm text-muted-foreground">No instances available</p>
                 ) : (
                   <div className="space-y-2">
-                    {instances.map((instance) => (
-                      <div key={instance.id} className="flex items-center space-x-2">
-                        <Checkbox
-                          id={`instance-${instance.id}`}
-                          checked={instancePermissions[instance.id] || false}
-                          onCheckedChange={(checked) =>
-                            handleInstancePermissionChange(instance.id, checked as boolean)
-                          }
-                          disabled={loading}
-                        />
-                        <Label
-                          htmlFor={`instance-${instance.id}`}
-                          className="font-normal cursor-pointer flex-1"
+                    {instances.map((instance, index) => {
+                      const isChecked = !!instancePermissions[instance.id];
+                      return (
+                        <div
+                          key={`${instance.name}-${index}`}
+                          className="flex items-center space-x-2"
                         >
-                          {instance.name}
-                        </Label>
-                        <span className="text-sm text-muted-foreground">Can Infer</span>
-                      </div>
-                    ))}
+                          <Checkbox
+                            id={`instance-${instance.id}`}
+                            checked={isChecked}
+                            onCheckedChange={(checked) => {
+                              handleInstancePermissionChange(instance.id, checked as boolean);
+                            }}
+                            disabled={loading}
+                          />
+                          <Label
+                            htmlFor={`instance-${instance.id}`}
+                            className="font-normal cursor-pointer flex-1"
+                          >
+                            {instance.name}
+                          </Label>
+                        </div>
+                      );
+                    })}
                   </div>
                 )}
               </div>

webui/src/components/form/KeyValueInput.tsx

@@ -59,7 +59,7 @@ const KeyValueInput: React.FC<KeyValueInputProps> = ({
       // Reset to single empty row if value is explicitly undefined/null
       setPairs([{ key: '', value: '' }])
     }
-    // eslint-disable-next-line react-hooks/exhaustive-deps
+     
   }, [value])
 
   // Update parent component when pairs change

webui/src/components/instance/AutoRestartConfiguration.tsx

@@ -5,7 +5,7 @@ import NumberInput from '@/components/form/NumberInput'
 
 interface AutoRestartConfigurationProps {
   formData: CreateInstanceOptions
-  onChange: (key: keyof CreateInstanceOptions, value: any) => void
+  onChange: <K extends keyof CreateInstanceOptions>(key: K, value: CreateInstanceOptions[K]) => void
 }
 
 const AutoRestartConfiguration: React.FC<AutoRestartConfigurationProps> = ({

webui/src/components/instance/BackendConfiguration.tsx

@@ -3,9 +3,11 @@ import type { CreateInstanceOptions } from '@/types/instance'
 import { getBasicBackendFields, getAdvancedBackendFields } from '@/lib/zodFormUtils'
 import BackendFormField from '@/components/BackendFormField'
 
+type BackendFieldValue = string | number | boolean | string[] | Record<string, string> | undefined
+
 interface BackendConfigurationProps {
   formData: CreateInstanceOptions
-  onBackendFieldChange: (key: string, value: any) => void
+  onBackendFieldChange: (key: string, value: BackendFieldValue) => void
   showAdvanced?: boolean
 }
 
@@ -26,7 +28,7 @@ const BackendConfiguration: React.FC<BackendConfigurationProps> = ({
         <BackendFormField
           key={fieldKey}
           fieldKey={fieldKey}
-          value={(formData.backend_options as any)?.[fieldKey]}
+          value={(formData.backend_options as Record<string, BackendFieldValue> | undefined)?.[fieldKey]}
           onChange={onBackendFieldChange}
         />
       ))}
@@ -41,7 +43,7 @@ const BackendConfiguration: React.FC<BackendConfigurationProps> = ({
               <BackendFormField
                 key={fieldKey}
                 fieldKey={fieldKey}
-                value={(formData.backend_options as any)?.[fieldKey]}
+                value={(formData.backend_options as Record<string, BackendFieldValue> | undefined)?.[fieldKey]}
                 onChange={onBackendFieldChange}
               />
             ))}
@@ -53,7 +55,7 @@ const BackendConfiguration: React.FC<BackendConfigurationProps> = ({
         <BackendFormField
           key="extra_args"
           fieldKey="extra_args"
-          value={(formData.backend_options as any)?.extra_args}
+          value={(formData.backend_options as Record<string, BackendFieldValue> | undefined)?.extra_args}
           onChange={onBackendFieldChange}
         />
       </div>

webui/src/components/settings/ApiKeysSection.tsx

@@ -1,10 +1,10 @@
-import { useEffect, useState } from "react";
+import { useEffect, useState, Fragment } from "react";
 import { Button } from "@/components/ui/button";
 import { Badge } from "@/components/ui/badge";
 import { Alert, AlertDescription } from "@/components/ui/alert";
 import { Trash2, Copy, Check, X, ChevronDown, ChevronRight } from "lucide-react";
 import { apiKeysApi } from "@/lib/api";
-import { ApiKey, KeyPermissionResponse, PermissionMode } from "@/types/apiKey";
+import { type ApiKey, type KeyPermissionResponse, PermissionMode } from "@/types/apiKey";
 import CreateApiKeyDialog from "@/components/apikeys/CreateApiKeyDialog";
 import { format, formatDistanceToNow } from "date-fns";
 
@@ -20,7 +20,7 @@ function ApiKeysSection() {
   const [loadingPermissions, setLoadingPermissions] = useState<Record<number, boolean>>({});
 
   useEffect(() => {
-    fetchKeys();
+    void fetchKeys();
   }, []);
 
   const fetchKeys = async () => {
@@ -52,7 +52,7 @@ function ApiKeysSection() {
 
   const handleKeyCreated = (plainTextKey: string) => {
     setNewKeyPlainText(plainTextKey);
-    fetchKeys();
+    void fetchKeys();
     setCreateDialogOpen(false);
   };
 
@@ -75,7 +75,7 @@ function ApiKeysSection() {
 
     try {
       await apiKeysApi.delete(id);
-      fetchKeys();
+      void fetchKeys();
     } catch (err) {
       alert(err instanceof Error ? err.message : "Failed to delete API key");
     }
@@ -87,7 +87,7 @@ function ApiKeysSection() {
     } else {
       setExpandedRowId(key.id);
       if (key.permission_mode === PermissionMode.PerInstance) {
-        fetchPermissions(key.id);
+        void fetchPermissions(key.id);
       }
     }
   };
@@ -136,7 +136,7 @@ function ApiKeysSection() {
               <code className="flex-1 p-3 bg-white dark:bg-gray-900 border border-green-300 dark:border-green-800 rounded font-mono text-sm break-all">
                 {newKeyPlainText}
               </code>
-              <Button onClick={handleCopyKey} variant="outline" size="sm">
+              <Button onClick={() => void handleCopyKey()} variant="outline" size="sm">
                 {copiedKey ? <Check className="h-4 w-4" /> : <Copy className="h-4 w-4" />}
               </Button>
             </div>
@@ -175,9 +175,8 @@ function ApiKeysSection() {
             </thead>
             <tbody>
               {keys.map((key) => (
-                <>
+                <Fragment key={key.id}>
                   <tr
-                    key={key.id}
                     className="border-t hover:bg-muted/50 cursor-pointer"
                     onClick={() => handleRowClick(key)}
                   >
@@ -217,7 +216,7 @@ function ApiKeysSection() {
                         size="icon"
                         onClick={(e) => {
                           e.stopPropagation();
-                          handleDeleteKey(key.id, key.name);
+                          void handleDeleteKey(key.id, key.name);
                         }}
                         title="Delete key"
                       >
@@ -236,25 +235,15 @@ function ApiKeysSection() {
                           <p className="text-sm text-muted-foreground">Loading permissions...</p>
                         ) : permissions[key.id] ? (
                           <div className="space-y-2">
-                            <p className="text-sm font-semibold">Instance Permissions:</p>
-                            <table className="w-full text-sm">
-                              <thead>
-                                <tr className="border-b">
-                                  <th className="text-left py-2">Instance Name</th>
-                                  <th className="text-left py-2">Can Infer</th>
-                                </tr>
-                              </thead>
-                              <tbody>
-                                {permissions[key.id].map((perm) => (
-                                  <tr key={perm.instance_id} className="border-b">
-                                    <td className="py-2">{perm.instance_name}</td>
-                                    <td className="py-2">
-                                      <Check className="h-4 w-4 text-green-600" />
-                                    </td>
-                                  </tr>
-                                ))}
-                              </tbody>
-                            </table>
+                            <p className="text-sm font-semibold">Allowed Instances:</p>
+                            <ul className="text-sm space-y-1">
+                              {permissions[key.id].map((perm) => (
+                                <li key={perm.instance_id} className="flex items-center gap-2">
+                                  <Check className="h-3 w-3 text-green-600" />
+                                  {perm.instance_name}
+                                </li>
+                              ))}
+                            </ul>
                           </div>
                         ) : (
                           <p className="text-sm text-muted-foreground">No permissions data</p>
@@ -262,7 +251,7 @@ function ApiKeysSection() {
                       </td>
                     </tr>
                   )}
-                </>
+                </Fragment>
               ))}
             </tbody>
           </table>

webui/src/components/settings/SettingsDialog.tsx

@@ -1,4 +1,4 @@
-import { Dialog, DialogContent, DialogHeader, DialogTitle } from "@/components/ui/dialog";
+import { Dialog, DialogContent, DialogDescription, DialogHeader, DialogTitle } from "@/components/ui/dialog";
 import ApiKeysSection from "./ApiKeysSection";
 
 interface SettingsDialogProps {
@@ -12,6 +12,9 @@ function SettingsDialog({ open, onOpenChange }: SettingsDialogProps) {
       <DialogContent className="sm:max-w-5xl max-h-[90vh] overflow-y-auto">
         <DialogHeader>
           <DialogTitle>Settings</DialogTitle>
+          <DialogDescription>
+            Manage your application settings and API keys.
+          </DialogDescription>
         </DialogHeader>
         <ApiKeysSection />
       </DialogContent>

webui/src/components/ui/alert.tsx

@@ -4,13 +4,13 @@ import { cva, type VariantProps } from "class-variance-authority"
 import { cn } from "@/lib/utils"
 
 const alertVariants = cva(
-  "relative w-full rounded-lg border p-4",
+  "relative w-full rounded-lg border px-4 py-3 text-sm grid has-[>svg]:grid-cols-[calc(var(--spacing)*4)_1fr] grid-cols-[0_1fr] has-[>svg]:gap-x-3 gap-y-0.5 items-start [&>svg]:size-4 [&>svg]:translate-y-0.5 [&>svg]:text-current",
   {
     variants: {
       variant: {
-        default: "bg-background text-foreground",
+        default: "bg-card text-card-foreground",
         destructive:
-          "border-destructive/50 text-destructive dark:border-destructive [&>svg]:text-destructive",
+          "text-destructive bg-card [&>svg]:text-current *:data-[slot=alert-description]:text-destructive/90",
       },
     },
     defaultVariants: {
@@ -19,41 +19,48 @@ const alertVariants = cva(
   }
 )
 
-const Alert = React.forwardRef<
-  HTMLDivElement,
-  React.HTMLAttributes<HTMLDivElement> & VariantProps<typeof alertVariants>
->(({ className, variant, ...props }, ref) => (
-  <div
-    ref={ref}
-    role="alert"
-    className={cn(alertVariants({ variant }), className)}
-    {...props}
-  />
-))
-Alert.displayName = "Alert"
+function Alert({
+  className,
+  variant,
+  ...props
+}: React.ComponentProps<"div"> & VariantProps<typeof alertVariants>) {
+  return (
+    <div
+      data-slot="alert"
+      role="alert"
+      className={cn(alertVariants({ variant }), className)}
+      {...props}
+    />
+  )
+}
 
-const AlertTitle = React.forwardRef<
-  HTMLParagraphElement,
-  React.HTMLAttributes<HTMLHeadingElement>
->(({ className, ...props }, ref) => (
-  <h5
-    ref={ref}
-    className={cn("mb-1 font-medium leading-none tracking-tight", className)}
-    {...props}
-  />
-))
-AlertTitle.displayName = "AlertTitle"
+function AlertTitle({ className, ...props }: React.ComponentProps<"div">) {
+  return (
+    <div
+      data-slot="alert-title"
+      className={cn(
+        "col-start-2 line-clamp-1 min-h-4 font-medium tracking-tight",
+        className
+      )}
+      {...props}
+    />
+  )
+}
 
-const AlertDescription = React.forwardRef<
-  HTMLParagraphElement,
-  React.HTMLAttributes<HTMLParagraphElement>
->(({ className, ...props }, ref) => (
-  <div
-    ref={ref}
-    className={cn("text-sm [&_p]:leading-relaxed", className)}
-    {...props}
-  />
-))
-AlertDescription.displayName = "AlertDescription"
+function AlertDescription({
+  className,
+  ...props
+}: React.ComponentProps<"div">) {
+  return (
+    <div
+      data-slot="alert-description"
+      className={cn(
+        "text-muted-foreground col-start-2 grid justify-items-start gap-1 text-sm [&_p]:leading-relaxed",
+        className
+      )}
+      {...props}
+    />
+  )
+}
 
 export { Alert, AlertTitle, AlertDescription }

webui/src/components/ui/badge.tsx

@@ -5,7 +5,7 @@ import { cva, type VariantProps } from "class-variance-authority"
 import { cn } from "@/lib/utils"
 
 const badgeVariants = cva(
-  "inline-flex items-center justify-center rounded-md border px-2 py-0.5 text-xs font-medium w-fit whitespace-nowrap shrink-0 [&>svg]:size-3 gap-1 [&>svg]:pointer-events-none focus-visible:border-ring focus-visible:ring-ring/50 focus-visible:ring-[3px] aria-invalid:ring-destructive/20 dark:aria-invalid:ring-destructive/40 aria-invalid:border-destructive transition-[color,box-shadow] overflow-hidden",
+  "inline-flex items-center justify-center rounded-full border px-2 py-0.5 text-xs font-medium w-fit whitespace-nowrap shrink-0 [&>svg]:size-3 gap-1 [&>svg]:pointer-events-none focus-visible:border-ring focus-visible:ring-ring/50 focus-visible:ring-[3px] aria-invalid:ring-destructive/20 dark:aria-invalid:ring-destructive/40 aria-invalid:border-destructive transition-[color,box-shadow] overflow-hidden",
   {
     variants: {
       variant: {

webui/src/components/ui/button.tsx

@@ -9,14 +9,13 @@ const buttonVariants = cva(
   {
     variants: {
       variant: {
-        default:
-          "bg-primary text-primary-foreground shadow-xs hover:bg-primary/90",
+        default: "bg-primary text-primary-foreground hover:bg-primary/90",
         destructive:
-          "bg-destructive text-white shadow-xs hover:bg-destructive/90 focus-visible:ring-destructive/20 dark:focus-visible:ring-destructive/40 dark:bg-destructive/60",
+          "bg-destructive text-white hover:bg-destructive/90 focus-visible:ring-destructive/20 dark:focus-visible:ring-destructive/40 dark:bg-destructive/60",
         outline:
           "border bg-background shadow-xs hover:bg-accent hover:text-accent-foreground dark:bg-input/30 dark:border-input dark:hover:bg-input/50",
         secondary:
-          "bg-secondary text-secondary-foreground shadow-xs hover:bg-secondary/80",
+          "bg-secondary text-secondary-foreground hover:bg-secondary/80",
         ghost:
           "hover:bg-accent hover:text-accent-foreground dark:hover:bg-accent/50",
         link: "text-primary underline-offset-4 hover:underline",
@@ -26,6 +25,8 @@ const buttonVariants = cva(
         sm: "h-8 rounded-md gap-1.5 px-3 has-[>svg]:px-2.5",
         lg: "h-10 rounded-md px-6 has-[>svg]:px-4",
         icon: "size-9",
+        "icon-sm": "size-8",
+        "icon-lg": "size-10",
       },
     },
     defaultVariants: {

webui/src/components/ui/card.tsx

@@ -20,7 +20,7 @@ function CardHeader({ className, ...props }: React.ComponentProps<"div">) {
     <div
       data-slot="card-header"
       className={cn(
-        "@container/card-header grid auto-rows-min grid-rows-[auto_auto] items-start gap-1.5 px-6 has-data-[slot=card-action]:grid-cols-[1fr_auto] [.border-b]:pb-6",
+        "@container/card-header grid auto-rows-min grid-rows-[auto_auto] items-start gap-2 px-6 has-data-[slot=card-action]:grid-cols-[1fr_auto] [.border-b]:pb-6",
         className
       )}
       {...props}

webui/src/components/ui/checkbox.tsx

@@ -19,7 +19,7 @@ function Checkbox({
     >
       <CheckboxPrimitive.Indicator
         data-slot="checkbox-indicator"
-        className="flex items-center justify-center text-current transition-none"
+        className="grid place-content-center text-current transition-none"
       >
         <CheckIcon className="size-3.5" />
       </CheckboxPrimitive.Indicator>

webui/src/components/ui/input.tsx

@@ -8,7 +8,7 @@ function Input({ className, type, ...props }: React.ComponentProps<"input">) {
       type={type}
       data-slot="input"
       className={cn(
-        "file:text-foreground placeholder:text-muted-foreground selection:bg-primary selection:text-primary-foreground dark:bg-input/30 border-input flex h-9 w-full min-w-0 rounded-md border bg-transparent px-3 py-1 text-base shadow-xs transition-[color,box-shadow] outline-none file:inline-flex file:h-7 file:border-0 file:bg-transparent file:text-sm file:font-medium disabled:pointer-events-none disabled:cursor-not-allowed disabled:opacity-50 md:text-sm",
+        "file:text-foreground placeholder:text-muted-foreground selection:bg-primary selection:text-primary-foreground dark:bg-input/30 border-input h-9 w-full min-w-0 rounded-md border bg-transparent px-3 py-1 text-base shadow-xs transition-[color,box-shadow] outline-none file:inline-flex file:h-7 file:border-0 file:bg-transparent file:text-sm file:font-medium disabled:pointer-events-none disabled:cursor-not-allowed disabled:opacity-50 md:text-sm",
         "focus-visible:border-ring focus-visible:ring-ring/50 focus-visible:ring-[3px]",
         "aria-invalid:ring-destructive/20 dark:aria-invalid:ring-destructive/40 aria-invalid:border-destructive",
         className

webui/src/components/ui/radio-group.tsx

@@ -1,42 +1,43 @@
 import * as React from "react"
 import * as RadioGroupPrimitive from "@radix-ui/react-radio-group"
-import { Circle } from "lucide-react"
+import { CircleIcon } from "lucide-react"
 
 import { cn } from "@/lib/utils"
 
-const RadioGroup = React.forwardRef<
-  React.ElementRef<typeof RadioGroupPrimitive.Root>,
-  React.ComponentPropsWithoutRef<typeof RadioGroupPrimitive.Root>
->(({ className, ...props }, ref) => {
+function RadioGroup({
+  className,
+  ...props
+}: React.ComponentProps<typeof RadioGroupPrimitive.Root>) {
   return (
     <RadioGroupPrimitive.Root
-      className={cn("grid gap-2", className)}
+      data-slot="radio-group"
+      className={cn("grid gap-3", className)}
       {...props}
-      ref={ref}
     />
   )
-})
-RadioGroup.displayName = RadioGroupPrimitive.Root.displayName
+}
 
-const RadioGroupItem = React.forwardRef<
-  React.ElementRef<typeof RadioGroupPrimitive.Item>,
-  React.ComponentPropsWithoutRef<typeof RadioGroupPrimitive.Item>
->(({ className, ...props }, ref) => {
+function RadioGroupItem({
+  className,
+  ...props
+}: React.ComponentProps<typeof RadioGroupPrimitive.Item>) {
   return (
     <RadioGroupPrimitive.Item
-      ref={ref}
+      data-slot="radio-group-item"
       className={cn(
-        "aspect-square h-4 w-4 rounded-full border border-primary text-primary ring-offset-background focus:outline-none focus-visible:ring-2 focus-visible:ring-ring focus-visible:ring-offset-2 disabled:cursor-not-allowed disabled:opacity-50",
+        "border-input text-primary focus-visible:border-ring focus-visible:ring-ring/50 aria-invalid:ring-destructive/20 dark:aria-invalid:ring-destructive/40 aria-invalid:border-destructive dark:bg-input/30 aspect-square size-4 shrink-0 rounded-full border shadow-xs transition-[color,box-shadow] outline-none focus-visible:ring-[3px] disabled:cursor-not-allowed disabled:opacity-50",
         className
       )}
       {...props}
     >
-      <RadioGroupPrimitive.Indicator className="flex items-center justify-center">
-        <Circle className="h-2.5 w-2.5 fill-current text-current" />
+      <RadioGroupPrimitive.Indicator
+        data-slot="radio-group-indicator"
+        className="relative flex items-center justify-center"
+      >
+        <CircleIcon className="fill-primary absolute top-1/2 left-1/2 size-2 -translate-x-1/2 -translate-y-1/2" />
       </RadioGroupPrimitive.Indicator>
     </RadioGroupPrimitive.Item>
   )
-})
-RadioGroupItem.displayName = RadioGroupPrimitive.Item.displayName
+}
 
 export { RadioGroup, RadioGroupItem }

webui/src/contexts/__tests__/InstancesContext.test.tsx

@@ -3,8 +3,7 @@ import { render, screen, waitFor } from "@testing-library/react";
 import type { ReactNode } from "react";
 import { InstancesProvider, useInstances } from "@/contexts/InstancesContext";
 import { instancesApi } from "@/lib/api";
-import type { Instance } from "@/types/instance";
-import { BackendType } from "@/types/instance";
+import { BackendType, type Instance } from "@/types/instance";
 import { AuthProvider } from "../AuthContext";
 
 // Mock the API module
@@ -71,37 +70,37 @@ function TestComponent() {
 
       {/* Action buttons for testing with specific instances */}
       <button
-        onClick={() => createInstance("new-instance", { backend_type: BackendType.LLAMA_CPP, backend_options: { model: "test.gguf" } })}
+        onClick={() => void createInstance("new-instance", { backend_type: BackendType.LLAMA_CPP, backend_options: { model: "test.gguf" } })}
         data-testid="create-instance"
       >
         Create Instance
       </button>
       <button
-        onClick={() => updateInstance("instance1", { backend_type: BackendType.LLAMA_CPP, backend_options: { model: "updated.gguf" } })}
+        onClick={() => void updateInstance("instance1", { backend_type: BackendType.LLAMA_CPP, backend_options: { model: "updated.gguf" } })}
         data-testid="update-instance"
       >
         Update Instance
       </button>
       <button
-        onClick={() => startInstance("instance2")}
+        onClick={() => void startInstance("instance2")}
         data-testid="start-instance"
       >
         Start Instance2
       </button>
       <button
-        onClick={() => stopInstance("instance1")}
+        onClick={() => void stopInstance("instance1")}
         data-testid="stop-instance"
       >
         Stop Instance1
       </button>
       <button
-        onClick={() => restartInstance("instance1")}
+        onClick={() => void restartInstance("instance1")}
         data-testid="restart-instance"
       >
         Restart Instance1
       </button>
       <button
-        onClick={() => deleteInstance("instance2")}
+        onClick={() => void deleteInstance("instance2")}
         data-testid="delete-instance"
       >
         Delete Instance2

webui/src/lib/healthService.ts

@@ -156,11 +156,14 @@ class HealthService {
       this.callbacks.set(instanceName, new Set())
     }
 
-    this.callbacks.get(instanceName)!.add(callback)
+    const callbacks = this.callbacks.get(instanceName)
+    if (callbacks) {
+      callbacks.add(callback)
 
-    // Start health checking if this is the first subscriber
-    if (this.callbacks.get(instanceName)!.size === 1) {
-      this.startHealthCheck(instanceName)
+      // Start health checking if this is the first subscriber
+      if (callbacks.size === 1) {
+        this.startHealthCheck(instanceName)
+      }
     }
 
     // Return unsubscribe function
@@ -214,22 +217,24 @@ class HealthService {
     }
 
     // Start new interval with appropriate timing
-    const interval = setInterval(async () => {
-      try {
-        const health = await this.performHealthCheck(instanceName)
-        this.notifyCallbacks(instanceName, health)
+    const interval = setInterval(() => {
+      void (async () => {
+        try {
+          const health = await this.performHealthCheck(instanceName)
+          this.notifyCallbacks(instanceName, health)
 
-        // Check if state changed and adjust interval
-        const previousState = this.lastHealthState.get(instanceName)
-        this.lastHealthState.set(instanceName, health.state)
+          // Check if state changed and adjust interval
+          const previousState = this.lastHealthState.get(instanceName)
+          this.lastHealthState.set(instanceName, health.state)
 
-        if (previousState !== health.state) {
-          this.adjustPollingInterval(instanceName, health.state)
+          if (previousState !== health.state) {
+            this.adjustPollingInterval(instanceName, health.state)
+          }
+        } catch (error) {
+          console.error(`Health check failed for ${instanceName}:`, error)
+          // Continue polling even on error
         }
-      } catch (error) {
-        console.error(`Health check failed for ${instanceName}:`, error)
-        // Continue polling even on error
-      }
+      })()
     }, pollInterval)
 
     this.intervals.set(instanceName, interval)

webui/src/main.tsx

@@ -6,7 +6,10 @@ import './index.css'
 import { AuthProvider } from './contexts/AuthContext'
 import { ConfigProvider } from './contexts/ConfigContext'
 
-ReactDOM.createRoot(document.getElementById('root')!).render(
+const rootElement = document.getElementById('root')
+if (!rootElement) throw new Error('Failed to find the root element')
+
+ReactDOM.createRoot(rootElement).render(
   <React.StrictMode>
     <AuthProvider>
       <ConfigProvider>

webui/src/test/setup.ts

@@ -1,10 +1,44 @@
 import '@testing-library/jest-dom'
-import { afterEach, vi } from 'vitest'
+import { afterEach, beforeEach } from 'vitest'
 
-// Mock fetch globally since your app uses fetch
-global.fetch = vi.fn()
+// Create a working localStorage implementation for tests
+// This ensures localStorage works in both CLI and VSCode test runner
+class LocalStorageMock implements Storage {
+  private store: Map<string, string> = new Map()
 
-// Clean up after each test
-afterEach(() => {
-  vi.clearAllMocks()
+  get length(): number {
+    return this.store.size
+  }
+
+  clear(): void {
+    this.store.clear()
+  }
+
+  getItem(key: string): string | null {
+    return this.store.get(key) ?? null
+  }
+
+  key(index: number): string | null {
+    return Array.from(this.store.keys())[index] ?? null
+  }
+
+  removeItem(key: string): void {
+    this.store.delete(key)
+  }
+
+  setItem(key: string, value: string): void {
+    this.store.set(key, value)
+  }
+}
+
+// Replace global localStorage
+global.localStorage = new LocalStorageMock()
+
+// Clean up before each test
+beforeEach(() => {
+  localStorage.clear()
+})
+
+afterEach(() => {
+  localStorage.clear()
 })

webui/src/types/apiKey.ts

@@ -15,14 +15,10 @@ export interface ApiKey {
 }
 
 export interface CreateKeyRequest {
-  Name: string
-  PermissionMode: PermissionMode
-  ExpiresAt?: number
-  InstancePermissions: InstancePermission[]
-}
-
-export interface InstancePermission {
-  InstanceID: number
+  name: string
+  permission_mode: PermissionMode
+  expires_at?: number
+  instance_ids: number[]
 }
 
 export interface CreateKeyResponse extends ApiKey {

webui/tsconfig.node.json

@@ -4,7 +4,8 @@
     "skipLibCheck": true,
     "module": "ESNext",
     "moduleResolution": "bundler",
-    "allowSyntheticDefaultImports": true
+    "allowSyntheticDefaultImports": true,
+    "types": ["node"]
   },
   "include": ["vite.config.ts"]
 }