Replies: 16 comments
|
hey <@1359229044578582699> cognee has a lock mechanism that prevents multiple cognify operations from running simultaneously it is designed to prevent database corruption and ensure data integrity you can use existing |
|
But after the tool call what If I want to start the process again |
|
after it completes you can restart the cognify process by checking the status first and then initiating a new cognify operation |
|
Let me try |
|
╰────────────────────────────────╯ 2025-12-26T08:09:22.159585 [error ] Failed to initialize Kuzu database: IO exception: Could not set lock on file : /home/chinu_tensor/OpsRabbit-v2/.venv/lib/python3.12/site-packages/cognee/.cognee_system/databases/cognee_graph_kuzu.lock Pipeline failed: IO exception: Could not set lock on file : /home/chinu_tensor/OpsRabbit-v2/.venv/lib/python3.12/site-packages/cognee/.cognee_system/databases/cognee_graph_kuzu.lock I get this when i do cognify after tool call |
|
can you add this to your .env |
|
what's the error now |
|
SHARED_KUZU_LOCK=true |
|
Yes kuzu lock feature uses redis |
|
it worked without redis |
|
how did it work without redis |
|
What is you env setup |
|
If you uses 'SHARED_KUZU_LOCK=false' it uses local threading lock instead of redis |
|
It is true |
|
idk Ig i did the cognifying process earlier due to that it worked |
|
hi <@1373135538147692604> |
Uh oh!
There was an error while loading. Please reload this page.
=Hi,
I am using LangGraph with Cognee. There is a tool that uses cognee.search to get some results.
Before starting the FastAPI server, I cognify the documents that I have. However, when a tool call is made to cognee.search and I try to run cognify again, it throws an error due to a lock.
I want the process to work like this:
If the cognifying process is already going on, the tool should return that the process is in progress.
Even when the tool call is made, the cognifying process should start.
This discussion was automatically pulled from Discord.
All reactions