happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

expose args and exit_code in API

This commit is contained in:
Xuan Son Nguyen 2025-11-23 14:59:04 +01:00
parent 4af1b6cbac
commit f25bfaba4d
4 changed files with 25 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
tools/server/README.md
View File

@ -1455,19 +1455,23 @@ The `status` object can be:
```json
"status": {
"value": "loading"
"value": "loading",
"args": ["llama-server", "-ctx", "4096"]
}
```
```json
"status": {
"value": "failed"
"value": "failed",
"args": ["llama-server", "-ctx", "4096"],
"exit_code": 1
}
```
```json
"status": {
"value": "loaded"
"value": "loaded",
"args": ["llama-server", "-ctx", "4096"]
}
```

6
tools/server/server-models.cpp
View File

@ -368,11 +368,11 @@ void server_models::load(const std::string & name) {
std::vector<std::string> child_env = base_env; // copy
child_env.push_back("LLAMA_SERVER_ROUTER_PORT=" + std::to_string(base_params.port));
// TODO: add logging
SRV_INF("%s", "spawning server instance with args:\n");
for (const auto & arg : child_args) {
SRV_INF(" %s\n", arg.c_str());
}
inst.meta.args = child_args; // save for debugging
std::vector<char *> argv = to_char_ptr_array(child_args);
std::vector<char *> envp = to_char_ptr_array(child_env);
@ -405,7 +405,9 @@ void server_models::load(const std::string & name) {
std::lock_guard<std::mutex> lk(mutex);
auto it = mapping.find(name);
if (it != mapping.end()) {
it->second.meta.status = exit_code == 0
auto & meta = it->second.meta;
meta.exit_code = exit_code;
meta.status = exit_code == 0
? SERVER_MODEL_STATUS_UNLOADED
: SERVER_MODEL_STATUS_FAILED;
}

4
tools/server/server-models.h
View File

@ -58,7 +58,9 @@ struct server_model_meta {
bool in_cache = false; // if true, use -hf; use -m otherwise
int port = 0;
server_model_status status = SERVER_MODEL_STATUS_UNLOADED;
int64_t last_used = 0;
int64_t last_used = 0; // for LRU unloading
std::vector<std::string> args; // additional args passed to the model instance (used for debugging)
int exit_code = 0; // exit code of the model instance process (only valid if status == FAILED)
bool is_active() const {
return status == SERVER_MODEL_STATUS_LOADED || status == SERVER_MODEL_STATUS_LOADING;

13
tools/server/server.cpp
View File

@ -5160,7 +5160,7 @@ public:
std::string name = json_value(body, "model", std::string());
auto model = models->get_meta(name);
if (!model.has_value()) {
res->error(format_error_response("model is not found", ERROR_TYPE_INVALID_REQUEST));
res->error(format_error_response("model is not found", ERROR_TYPE_NOT_FOUND));
return res;
}
if (model->status == SERVER_MODEL_STATUS_LOADED) {
@ -5188,15 +5188,20 @@ public:
json models_json = json::array();
auto all_models = models->get_all_meta();
for (const auto & model : all_models) {
json status {
{"value", server_model_status_to_string(model.status)},
{"args", model.args},
};
if (model.status == SERVER_MODEL_STATUS_FAILED) {
status["exit_code"] = model.exit_code;
}
models_json.push_back(json {
{"name", model.name},
{"id", model.name},
{"in_cache", model.in_cache},
{"path", model.path},
{"status", status},
// TODO: other fields...
{"status", {
{"value", server_model_status_to_string(model.status)}
}},
});
}
res->ok({{"data", models_json}});