Hi all, I’m on Cursor 0.44 running on Ubuntu 22.04.
I’m getting this error: Non-turbopuffer vector databases are no longer supported
Any fixes for this?
Hi all, I’m on Cursor 0.44 running on Ubuntu 22.04.
I’m getting this error: Non-turbopuffer vector databases are no longer supported
Any fixes for this?
Same thing is happening to me, tried looking online but I found no reference to this and cursor so it’s obviously a new update overnight with sonnet or cursor.
Seems like an issue with the way RAG is happening with the docs, @deanrie any possible fixes for this?
I have the same problem. Been looking around for fixes but have not found any. Guessing this would have happened due to some update.
I have the same issue. Docs is a key feature of Cursor, I hope they can resolve this ASAP
Same here
same same
Common issues:
* Deprecated APIs no longer work
* Breaking changes in newer versions
* Security vulnerabilities in older implementations
To fix:
* Migrate to a supported vector DB
* Update your dependencies
* Refactor queries to use new syntax
Need the actual error message to give you more specific guidance.
same here 793743fe-ab1e-4dff-acec-439375dd485c Non-turbopuffer vector databases are no longer supported
I’m in the same situation. Does anyone know the solution?
Same issue on Mac.
Faced the same just now.
Same here, very disappointing.
how long will it take to be fixed?
same here
same here, all the prompts that I use from GitHub can not be used right now. also this is happening on both composer and chat. attaching the request id cf88cb83-389d-4415-b987-408a9c5a9b12
same here
I missed these other threads when I was searching earlier. I’m really sorry, Cursor team - but how was this raised 7 hours ago with this much traction with no reply?
Further, is the status page ever updated with actual issues? “Indexing Uptime: 100%” - errr - a quick google of the forum or customer experience feedback would say that almost every metric on this status page is, to put it gently, inaccurate.
Could we please have some acknowledgement? Some understanding whether this is even being investigated?
You guys clearly haven’t been reading the change logs!
Oh, wait…
This was fixed in a backend release yesterday. Apologies for the issues!