As stated in the title, setting up indexing takes forever, and fails everytime
I am also having the same issue, cursor unable to index the codebase.
Hi @MrOliverYt
Does your codebase consist of thousands of files? Have you restricted access to directories like node_modules and other package files? If you’re not sure how to do that, you can find an explanation here:
hi, thanks for the response. Indexing seems to work for me again, weird. I’ll let you know if this happens again, but thanks anyways
I’m having the same issue.
I have been using cursor for a week and only since yesterday I started to have regular failure of indexing (“handshake failed”).
Since I haven’t change anything in my settings and my codebase hasn’t significantly grown from 2 days ago when everything was working fine, I have a hard time understanding what’s going on.
I have seen posts in this forum sometimes from a year ago with the exact same problem we are having so it looks like it’s a recurring issue with Cursor.
It’s still usable even without codebase indexing but it make Cursor a much less interesting tool for helping building software.
I know Cursor’s founders are young, but they might want to make sure their current features are robust before working on exciting new stuff.
PS: the other posts with this issue I found.
(post deleted by author)
i am having similar issues (happening since today) i really think its a server side issue on cursor side. wish there was a status page. lots of recent indexing issues too on github:
/getcursor/cursor/issues/1809
/getcursor/cursor/issues/1998
/getcursor/cursor/issues/2035
sorry for an extra post but i cant edit hidden one (ps hiding posts with links is annoying, hope someone approves me already)
ok unfortunately indexing is broken again on my cursor settings
Same problem all day today
I get a handshake failure when trying to index every time since this morning
I have same issue
I had same issue on two different machines and two different networks. Then I read somewhere in this forum - can’t find where now - try having a few files open when indexing. Worked on both machines.
Hi getting the same issue, stuck at setting up indexing, then shows handshake failed
I’m getting the same error. It’s a remote SSH connection, and there is a firewall on the machine, but I have a proxy configured that has cursor domains whitelisted. I found the log ~/.cursor-server/data/logs/20241123T153139/exthost4/anysphere.cursor-retrieval/Cursor Indexing & Retrieval.log
:
2024-11-23 20:06:01.635 [info] Setting indexing intent to should-index
2024-11-23 20:06:01.701 [info] Creating merkle client.
2024-11-23 20:06:01.701 [info] Done creating merkle client.
2024-11-23 20:06:01.701 [info] Doing a startup handshake.
2024-11-23 20:06:05.139 [info] Finished initializing merkle tree in 3360.748883999884 ms.
2024-11-23 20:06:05.141 [info] Doing the initial handshake with hash: 23ed85353c0cdff569ca5b3e077d4bfd7ef6a3cda87c9704f75199b0aa2fbaf9
2024-11-23 20:06:05.144 [info] Handshake start
2024-11-23 20:06:09.145 [info] Handshake start
2024-11-23 20:06:17.147 [info] Handshake start
2024-11-23 20:06:33.153 [info] Handshake start
2024-11-23 20:07:05.153 [info] Handshake start
2024-11-23 20:08:09.156 [info] Handshake start
2024-11-23 20:09:25.819 [error] Handshake failed:
2024-11-23 20:09:25.819 [error] Error: timeout in handshake with retry
at t.RepoClientMultiplexer.handshakeWithRetry (/home/ken/.cursor-server/cli/servers/Stable-001668006cc714afd397f4ef0d52862f5a095530/server/extensions/cursor-retrieval/dist/main.js:2:809591)
at async _.getServerStatus (/home/ken/.cursor-server/cli/servers/Stable-001668006cc714afd397f4ef0d52862f5a095530/server/extensions/cursor-retrieval/dist/main.js:2:825457)
at async _.startIndexingRepository (/home/ken/.cursor-server/cli/servers/Stable-001668006cc714afd397f4ef0d52862f5a095530/server/extensions/cursor-retrieval/dist/main.js:2:828148)
2024-11-23 20:09:25.819 [error] Handshake failed.
2024-11-23 20:09:25.820 [info] Indexing job successfully done or aborted.