From 0cabcbe588ae0055dfc0fc4880797cf05618df49 Mon Sep 17 00:00:00 2001 From: VJHack Date: Fri, 20 Sep 2024 14:54:08 -0500 Subject: [PATCH] fixed server 200 null response when context is exceeded --- examples/server/server.cpp | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/examples/server/server.cpp b/examples/server/server.cpp index 2b10469b1..79acbacca 100644 --- a/examples/server/server.cpp +++ b/examples/server/server.cpp @@ -1464,11 +1464,10 @@ struct server_context { std::vector results(id_tasks.size()); for (size_t i = 0; i < id_tasks.size(); i++) { server_task_result result = queue_results.recv(id_tasks); - if (result.error) { error_handler(result.data); cancel_tasks(id_tasks); - break; + return; } size_t idx = result.data["index"]; @@ -1948,9 +1947,9 @@ struct server_context { continue; } // context shift is disabled and prompt is too large - discard it - if (!params.ctx_shift && slot.n_prompt_tokens > slot.n_ctx ){ + if (!params.ctx_shift && (slot.n_prompt_tokens > slot.n_ctx) ){ slot.release(); - send_error(slot, "input is too large to process. enable context shift or increase the context length", ERROR_TYPE_SERVER); + send_error(slot, "Input is too large to process. Enable context shift or increase the context length", ERROR_TYPE_SERVER); continue; }