From 725ba43e12666ad31dbbf3efdaacab3d8befd650 Mon Sep 17 00:00:00 2001 From: timvisee Date: Thu, 7 Sep 2023 17:44:17 +0200 Subject: [PATCH] Mention stack trace endpoint in troubleshooting guide --- .../documentation/guides/common-errors.md | 54 +++++++++++++++++++ 1 file changed, 54 insertions(+) diff --git a/qdrant-landing/content/documentation/guides/common-errors.md b/qdrant-landing/content/documentation/guides/common-errors.md index c7c96f710..31f06df1d 100644 --- a/qdrant-landing/content/documentation/guides/common-errors.md +++ b/qdrant-landing/content/documentation/guides/common-errors.md @@ -30,3 +30,57 @@ ulimit -n 10000 ``` Please note, the command should be executed before you run Qdrant server. + +## Qdrant is stuck or frozen + +It might happen that Qdrant is complete stuck. Operations may hang indefinitely, +or an optimization may be stuck forever. Waiting some time does not improve the +situation. + +If you encounter such situation, always check the Qdrant logs first. It may give some +valuable information on what is happening. + +When logs don't show any pointers to the problem, you may extract a stack trace +from the Qdrant server to report the current state of what it is doing. This can +be done by visiting [`/stacktrace`](http://localhost:6333/stacktrace) in your +browser. You can attach this stack trace to an issue report to help Qdrant +developers debug the issue. + +```http +GET /stacktrace + +{ + "result": { + "threads": [ + { + "id": 7, + "name": "qdrant", + "frames": [ + "", + "", + ":0 - qdrant::main::h6b7c98553f8d540b ", + ":0 - std::sys_common::backtrace::__rust_begin_short_backtrace::ha4aa73ba5e009f1f " + ] + }, + { + "id": 8, + "name": "search-0", + "frames": [ + ":0 - epoll_wait ", + ":0 - mio::poll::Poll::poll::h91178360524c8cd2 " + ] + }, + { + "id": 9, + "name": "search-1", + "frames": [ + ":0 - syscall ", + ":0 - parking_lot::condvar::Condvar::wait_until_internal::h26d4f821b8427131 " + ] + } + ] + }, + "status": "ok", + "time": 0.03003201 +} +```