From ca2c8c3d36a08c5e38de43bb1e084c2022ab6f55 Mon Sep 17 00:00:00 2001 From: Alexander_Kabui Date: Mon, 8 Jan 2024 14:16:30 +0300 Subject: update issue:cd-gn3-configurations --- issues/cd-gn3-configurations.gmi | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/issues/cd-gn3-configurations.gmi b/issues/cd-gn3-configurations.gmi index a9e2360..e4df2e0 100644 --- a/issues/cd-gn3-configurations.gmi +++ b/issues/cd-gn3-configurations.gmi @@ -9,15 +9,21 @@ ## Description -The issue is that there is no straightforward way of knowing how to call the GN3 API from GN2. Multiple configurations, such as GN_SERVER_URL and GN3_LOCAL_URL, exist, making it difficult to understand how to debug. +The issue arises in GN3 when the Gunicorn worker dies due to the default 30-second time limit being exceeded. This occurs while polling the Fahamu API from GN3, where response times can occasionally take up to 1 minute. -The `GN_SERVER_URL` which from config is running on port 8880 results into this error: +See error from GN2: ``` -"HTTPConnectionPool(host='localhost', port=8880): Max retries exceeded with URL: /api/llm/gnqna (Caused by NewConnectionError(': Failed to establish a new connection: [Errno 111] Connection refused'))" +('Connection aborted.', RemoteDisconnected('Remote end closed connection without response')). + + ``` -## Notes 08/01/2024 +Error for worker timeout in GN3: + +``` +[2024-01-08 13:35:17 +0300] [32527] [CRITICAL] WORKER TIMEOUT (pid:32661) +``` -The problem arises from GN3's default Gunicorn timeout setting of 30 seconds, causing worker termination if exceeded. Fixing the issue involves adjusting the timeout to 1200, as done in GN2. \ No newline at end of file +Fixing the issue involves adjusting the timeout to 1200, as done in GN2. \ No newline at end of file -- cgit v1.2.3