fix: format + tracing

.gitea/workflows/release.yml

@@ -3,7 +3,7 @@ name: Publish & Deploy
 on:
   push:
     tags:
-      - 'v*'
+      - "v*"
 
 permissions:
   contents: write
@@ -86,4 +86,3 @@ jobs:
             REPO_LOWER=$REPO_LOWER TAG=$TAG docker compose up -d
 
             docker image prune -af
-            

readme.md

@@ -3,10 +3,8 @@
 - Race condition on jwks token refresh
 - Rate Limiting
 
-
 git tag -d v1.0.0; git push origin :refs/tags/v1.0.0; git tag -a v1.0.0 -m "Release v1.0.0"; git push origin v1.0.0
 
-
 curl https://chat.iceberg.black/api/v1/models -H "Authorization: Bearer eyJhbGciOiJSUzI1NiIsInR5cCIgOiAiSldUIiwia2lkIiA6ICJublpLek04TkZHVmpWbGFPRXZpMUtFSTVHQWRwaGlsYjh3RHRLeG5JOENZIn0.eyJleHAiOjE3NzU4MTUyODksImlhdCI6MTc3NTgxNDk4OSwianRpIjoiMTgwYzA2NDUtYzZiMC00MDRmLTgyYjEtMzA3YmY4ZmJlNmJlIiwiaXNzIjoiaHR0cHM6Ly9hdXRoLmljZWJlcmcuYmxhY2svcmVhbG1zL2ljZWJlcmciLCJzdWIiOiJmZGRiN2FjZC1kMmE5LTRmMTctOWIxNi1kZjVlN2EzNDI4YjciLCJ0eXAiOiJCZWFyZXIiLCJhenAiOiJjaGF0LWFwaSIsInNjb3BlIjoiIiwiY2xpZW50SG9zdCI6Ijg2LjIxMi44NC4xOTEiLCJjbGllbnRBZGRyZXNzIjoiODYuMjEyLjg0LjE5MSIsImNsaWVudF9pZCI6ImNoYXQtYXBpIn0.abjHABcjCJNiB6vriRw60nfzabEfD7CXwyRhkahFkC8ATgfy4fn0T8PnFfsaRpsXuhamIhWwNskrA7L9V3mbWWKW-JEOvImDhTc8sIX0E5fDTbk8O5wa_2yzNdLpRxdSjLqgL544rB8I-LZ8bl5SxtdN3gHfrnWr5ef8bbLgPRzZIylT3QUpah0uywDM_cfrrve9SMHHOUUItyzmOHLw0Igit1EzyFNyjbWf6OAU6TMjOF_eFTc5sakyBwsdJnGy0nhj5R-wxLpr1ug3iEd3Y-jDzHO4m6jariXVJ7Vvbz71i7sadDEKdSyOKcCeQbU4T0Tf7IxqW4c2DNnk3BFFuw"
 
 curl -X POST "https://auth.iceberg.black/realms/iceberg/protocol/openid-connect/token" -H "Content-Type: application/x-www-form-urlencoded" -d "grant_type=client_credentials" -d "client_id=chat-api" -d "client_secret=5fHUp8Z5GoNM70MVOGuQTfKFkaAE48Za"
@@ -29,19 +27,20 @@ A high-performance Rust API wrapper around Ollama, providing an OpenAI-compatibl
 
 # 🚀 Features
 
-* ✅ OpenAI-compatible API (`/v1/...`)
+- ✅ OpenAI-compatible API (`/v1/...`)
-* ⚡ Streaming (Server-Sent Events)
+- ⚡ Streaming (Server-Sent Events)
-* 🧠 Model lifecycle management (load/unload)
+- 🧠 Model lifecycle management (load/unload)
-* 🔐 API key authentication (optional)
+- 🔐 API key authentication (optional)
-* 📊 Usage tracking & observability
+- 📊 Usage tracking & observability
-* 🔀 Model routing & abstraction
+- 🔀 Model routing & abstraction
-* 🧩 Extensible architecture (multi-provider ready)
+- 🧩 Extensible architecture (multi-provider ready)
 
 ---
 
 # 📡 API Endpoints
 
 ## 1. Core LLM API (OpenAI-compatible)
+
 - [x] `POST /v1/chat/completions` + streaming
 - [x] `POST /v1/completions` + streaming
 - [ ] `POST /v1/embeddings`
@@ -126,16 +125,16 @@ GET /v1/usage
 
 Tracks:
 
-* request count
+- request count
-* latency
+- latency
-* per-model usage
+- per-model usage
 
 ---
 
 ## 🚦 Rate Limiting
 
-* Requests per minute
+- Requests per minute
-* Tokens per minute
+- Tokens per minute
 
 Returns:
 
@@ -158,8 +157,8 @@ Stores conversation history server-side.
 
 ## ⚡ Caching
 
-* Embeddings
+- Embeddings
-* Deterministic prompts (temperature = 0)
+- Deterministic prompts (temperature = 0)
 
 ---
 
@@ -196,8 +195,8 @@ POST /v1/runtime/evict
 
 Strategies:
 
-* LRU
+- LRU
-* memory threshold
+- memory threshold
 
 ---
 
@@ -223,10 +222,10 @@ Client → Rust API → Ollama → Response
 
 ### Layers:
 
-* HTTP (Axum)
+- HTTP (Axum)
-* Service layer (business logic)
+- Service layer (business logic)
-* Provider abstraction
+- Provider abstraction
-* Ollama client
+- Ollama client
 
 ---
 
@@ -241,19 +240,19 @@ trait LlmProvider {
 
 Supports:
 
-* Ollama (current)
+- Ollama (current)
-* OpenAI (future)
+- OpenAI (future)
-* Others
+- Others
 
 ---
 
 # 🎯 Roadmap
 
-* [ ] Full OpenAI compatibility
+- [ ] Full OpenAI compatibility
-* [ ] Multi-node routing
+- [ ] Multi-node routing
-* [ ] GPU-aware scheduling
+- [ ] GPU-aware scheduling
-* [ ] Web UI dashboard
+- [ ] Web UI dashboard
-* [ ] Distributed inference
+- [ ] Distributed inference
 
 ---
 

src/main.rs

@@ -28,13 +28,13 @@ pub fn init_tracing() {
 
 #[tokio::main]
 async fn main() {
-    init_tracing();
-
     #[cfg(debug_assertions)]
     {
         dotenvy::dotenv().ok();
     }
 
+    init_tracing();
+
     let state = AppState {
         ollama: Arc::new(OllamaProvider::new(OLLAMA_URL.as_str())),
     };

src/middlewares/auth/middleware.rs

@@ -41,7 +41,7 @@ pub async fn auth_middleware(mut request: Request, next: Next) -> Result<Respons
                     Ok(next.run(request).await)
                 }
                 Err(e) => {
-                    eprintln!("JWT validation failed: {:?}", e);
+                    tracing::error!("JWT validation failed: {:?}", e);
                     Err(StatusCode::UNAUTHORIZED)
                 }
             }

src/routes/v1/chat.rs

@@ -51,6 +51,7 @@ pub async fn completions(
     State(state): State<AppState>,
     Json(body): Json<api::CompletionRequest>,
 ) -> Result<Response, (axum::http::StatusCode, Json<api::ErrorResponse>)> {
+    tracing::debug!("Received /completion with body {:?}", body);
     if body.base.stream {
         let stream = state.ollama.completions_stream(&body).await.map_err(|e| {
             let (code, msg) = ollama_err(e);
@@ -116,6 +117,7 @@ pub async fn chat_completions(
     State(state): State<AppState>,
     Json(body): Json<api::ChatRequest>,
 ) -> Result<Response, (axum::http::StatusCode, String)> {
+    tracing::debug!("Received /completion with body {:?}", body);
     if body.base.stream {
         let stream = state
             .ollama