server: add /v1/metrics endpoint

tools/server/server-context.cpp

@@ -2912,6 +2912,27 @@ void server_routes::init_routes() {
         return res;
     };
 
+    this->get_v1_metrics = [this](const server_http_req &) {
+        auto res = std::make_unique<server_res_generator>(ctx_server);
+        
+        // Calculate uptime in seconds
+        int64_t uptime_sec = (llama_time_us() - t_server_start) / 1000000;
+        
+        json data = {
+            {"status", "online"},
+            {"uptime_sec", uptime_sec}
+        };
+        
+        // Include system_info if available
+        if (!system_info_str.empty()) {
+            data["system_info"] = system_info_str;
+        }
+        
+        res->ok(data);
+        return res;
+    };
+
+
     this->get_metrics = [this](const server_http_req &) {
         auto res = std::make_unique<server_res_generator>(ctx_server);
         if (!params.endpoint_metrics) {

tools/server/server-context.h

@@ -51,8 +51,9 @@ struct server_context {
 struct server_res_generator;
 
 struct server_routes {
-    server_routes(const common_params & params, server_context & ctx_server, std::function<bool()> is_ready = []() { return true; })
+    server_routes(const common_params & params, server_context & ctx_server, std::function<bool()> is_ready = []() { return true; }, int64_t t_start = 0)
-            : params(params), ctx_server(*ctx_server.impl), is_ready(is_ready) {
+            : params(params), ctx_server(*ctx_server.impl), is_ready(is_ready), t_server_start(t_start),
+              system_info_str(common_params_get_system_info(params)) {
         init_routes();
     }
 
@@ -60,6 +61,7 @@ struct server_routes {
     // handlers using lambda function, so that they can capture `this` without `std::bind`
     server_http_context::handler_t get_health;
     server_http_context::handler_t get_metrics;
+    server_http_context::handler_t get_v1_metrics;
     server_http_context::handler_t get_slots;
     server_http_context::handler_t post_slots;
     server_http_context::handler_t get_props;
@@ -90,4 +92,6 @@ private:
     const common_params & params;
     server_context_impl & ctx_server;
     std::function<bool()> is_ready;
+    int64_t t_server_start;
+    std::string system_info_str;
 };

tools/server/server.cpp

@@ -108,12 +108,15 @@ int main(int argc, char ** argv, char ** envp) {
         return 1;
     }
 
+    // Capture server start time for metrics
+    int64_t t_server_start = llama_time_us();
+
     //
     // Router
     //
 
     // register API routes
-    server_routes routes(params, ctx_server, [&ctx_http]() { return ctx_http.is_ready.load(); });
+    server_routes routes(params, ctx_server, [&ctx_http]() { return ctx_http.is_ready.load(); }, t_server_start);
 
     bool is_router_server = params.model.path.empty();
     std::optional<server_models_routes> models_routes{};
@@ -153,6 +156,7 @@ int main(int argc, char ** argv, char ** envp) {
 
     ctx_http.get ("/health",              ex_wrapper(routes.get_health)); // public endpoint (no API key check)
     ctx_http.get ("/v1/health",           ex_wrapper(routes.get_health)); // public endpoint (no API key check)
+    ctx_http.get ("/v1/metrics",          ex_wrapper(routes.get_v1_metrics)); // public endpoint (no API key check)
     ctx_http.get ("/metrics",             ex_wrapper(routes.get_metrics));
     ctx_http.get ("/props",               ex_wrapper(routes.get_props));
     ctx_http.post("/props",               ex_wrapper(routes.post_props));