Error Indexing Codebase - Handshake Failed

[soumen02]

1. If you can, please include a screenshot of your problem
   Initial error on chat
   
   When I click Open Settings and index, it got stuck on this for around 5-8 minutes
   
   After sometime I see the following,
   
   I continue asking it questions in chat, and when it gives me hallucinated output I go back to the settings page I see the following.
   
2. Please include the name of your operating system
   MacOS Sonoma 14.6.1 (23G93)
3. If you can, steps to reproduce are super helpful

[ascanio]

Same Issue here

[brahyam]

Same issue today on similar platform. could this be due to an outage?

[danperks]

Hi, is this issue still occurring? Seems likely that an outage or temporary bug.

[SeekBytes-AI]

I am having the same issue.
I have reinstalled the cursor after cleanup but did not help.
2024-10-08 19:41:05.109 [info] Handshake timing: 207.13455000007525
2024-10-08 19:41:10.285 [info] Search by sha took 33.64926999993622ms for query ""
2024-10-08 19:41:10.290 [info] Search by message took 39.75151300011203ms for query ""
2024-10-08 19:41:21.111 [info] Handshake start
2024-10-08 19:41:21.318 [info] Handshake timing: 207.79671699996106
2024-10-08 19:41:53.325 [info] Handshake start
2024-10-08 19:41:53.531 [info] Handshake timing: 206.40182100003585
2024-10-08 19:42:04.797 [info] Search by sha took 33.882578999968246ms for query ""
2024-10-08 19:42:04.801 [info] Search by message took 39.56248899991624ms for query ""
2024-10-08 19:42:57.536 [error] Handshake failed:
2024-10-08 19:42:57.536 [error] Error: timeout in handshake with retry
at t.RepoClientMultiplexer.handshakeWithRetry (/Applications/Cursor.app/Contents/Resources/app/extensions/cursor-retrieval/dist/main.js:2:826553)
at async w.getServerStatus (/Applications/Cursor.app/Contents/Resources/app/extensions/cursor-retrieval/dist/main.js:2:841881)
at async w.startIndexingRepository (/Applications/Cursor.app/Contents/Resources/app/extensions/cursor-retrieval/dist/main.js:2:844181)

[daniel-wachira]

I'm also experiencing this problem. Has anyone managed to find a workaround?

[KekmaTime]

dude im tired of it happening i tried reinstalling the whole thing but it still persists

[chriswep]

same, seems to be an issue at their end? https://www.reddit.com/r/cursor/comments/1fytjlg/curosr_forum_multiple_users_started_to_have/

[danperks]

Hey all, devs have been notified and are looking into it - sorry for anyone having issues!

[athmanar]

issue persists:
2024-10-02 10:06:07.652 [info] Setting indexing intent to should-index
2024-10-02 10:06:07.655 [info] Creating merkle client.
2024-10-02 10:06:07.655 [info] Done creating merkle client.
2024-10-02 10:06:07.656 [info] Doing a startup handshake.
2024-10-02 10:06:08.528 [info] Finished initializing merkle tree in 873.6273000240326 ms.
2024-10-02 10:06:08.530 [info] Doing the initial handshake with hash: ba088e651205d23a2561a2a127ba47f54bd00ba8382e36a8877b1168027f860d
2024-10-02 10:06:08.533 [info] Handshake start
2024-10-02 10:06:11.588 [info] Handshake start
2024-10-02 10:06:19.607 [info] Handshake start
2024-10-02 10:06:35.630 [info] Handshake start
2024-10-02 10:07:07.639 [info] Handshake start
2024-10-02 10:08:01.208 [info] Handshake start
2024-10-02 10:09:26.825 [error] Handshake failed:
2024-10-02 10:09:26.827 [error] Error: timeout in handshake with retry
at t.RepoClientMultiplexer.handshakeWithRetry (c:\Users\USER\AppData\Local\Programs\cursor\resources\app\extensions\cursor-retrieval\dist\main.js:2:826553)
at async w.getServerStatus (c:\Users\USER\AppData\Local\Programs\cursor\resources\app\extensions\cursor-retrieval\dist\main.js:2:841881)
at async w.startIndexingRepository (c:\Users\USER\AppData\Local\Programs\cursor\resources\app\extensions\cursor-retrieval\dist\main.js:2:844181)
2024-10-02 10:09:26.829 [error] Handshake failed.
2024-10-02 10:09:26.829 [info] Indexing job successfully done or aborted.

[samgabrail]

yes same issue here running on windows 11 within WSL2

[henriquebs12]

Same issue here, it's a pity to see you can't rely on their services. Their reputation just goes down a bit every time theses outages occurs.

[pinion]

same. mac os x 14.7

Version: 0.41.3
VSCode Version: 1.91.1
Commit: 51c8aff7cb5a89f4a0e462fbacab938bdbfaf140
Date: 2024-09-25T17:13:50.377Z
Electron: 29.4.0
ElectronBuildId: undefined
Chromium: 122.0.6261.156
Node.js: 20.9.0
V8: 12.2.281.27-electron.0
OS: Darwin arm64 23.6.0

[info] Repo client: syncing embeddings with backend: https://repo42.cursor.sh
2024-10-08 22:16:41.105 [info] Handshake start
2024-10-08 22:16:57.698 [info] Handshake start
2024-10-08 22:17:30.360 [info] Handshake start
2024-10-08 22:18:35.031 [error] Handshake failed:
2024-10-08 22:18:35.031 [error] Error: timeout in handshake with retry
at t.RepoClientMultiplexer.handshakeWithRetry (/Applications/Cursor.app/Contents/Resources/app/extensions/cursor-retrieval/dist/main.js:2:826553)
at async w.getServerStatus (/Applications/Cursor.app/Contents/Resources/app/extensions/cursor-retrieval/dist/main.js:2:841881)
at async w.startIndexingRepository (/Applications/Cursor.app/Contents/Resources/app/extensions/cursor-retrieval/dist/main.js:2:844181)
2024-10-08 22:18:35.032 [error] Handshake failed.
2024-10-08 22:18:35.032 [info] Indexing job successfully done or aborted.

curl https://repo42.cursor.sh

<title>503 Service Temporarily Unavailable</title>

503 Service Temporarily Unavailable

[georgeliu1998]

Logging out and back in doesn't solve the issue anymore. Here's my system info:

OS
macOS Sequoia Version 15.0.1 (24A348)

Cursor
Version: 0.41.3
VSCode Version: 1.91.1
Commit: 51c8aff7cb5a89f4a0e462fbacab938bdbfaf140
Date: 2024-09-25T17:13:50.377Z
Electron: 29.4.0
ElectronBuildId: undefined
Chromium: 122.0.6261.156
Node.js: 20.9.0
V8: 12.2.281.27-electron.0
OS: Darwin arm64 24.0.0

[AtlasPilotPuppy]

as a developer willing to help fix how can i help / contribute? @danperks

[danperks]

as a developer willing to help fix how can i help / contribute? @danperks

Hey @AtlasPilotPuppy, thanks for offering to help! Cursor is closed source right now and the dev team are aware of this issue, so for now there's not much that can be done, although I appreciate the offer!

If needed, a dev may ask for more info if they're struggling to reproduce an issue, so keep an eye on this issue and we'll see what happens :)

[cchevall]

I just logged out from my cursor account and logged in with a new account. Right after, this issue showed up on all my repositories. Could it be linked?

[afastlearner]

Same Issue here

same on windows 11

2024-10-08 23:01:09.418 [info] Setting indexing intent to should-index
2024-10-08 23:01:09.423 [info] Creating merkle client.
2024-10-08 23:01:09.423 [info] Done creating merkle client.
2024-10-08 23:01:09.423 [info] Doing a startup handshake.
2024-10-08 23:01:10.125 [info] Finished initializing merkle tree in 703.151900023222 ms.
2024-10-08 23:01:10.127 [info] Doing the initial handshake with hash: 1647c470a870419358f47f50ce90303cc69542bd420670d32bb8f0972469c45c
2024-10-08 23:01:10.129 [info] Handshake start
2024-10-08 23:01:10.276 [info] Handshake timing: 147.45840001106262
2024-10-08 23:01:12.283 [info] Handshake start
2024-10-08 23:01:12.742 [info] Handshake timing: 460.18099999427795
2024-10-08 23:01:16.754 [info] Handshake start
2024-10-08 23:01:17.307 [info] Handshake timing: 555.0661000013351
2024-10-08 23:01:25.322 [info] Handshake start
2024-10-08 23:01:25.477 [info] Handshake timing: 156.43250000476837
2024-10-08 23:01:41.493 [info] Handshake start
2024-10-08 23:01:41.673 [info] Handshake timing: 181.31760001182556
2024-10-08 23:02:13.676 [info] Handshake start
2024-10-08 23:02:13.974 [info] Handshake timing: 301.04829999804497
2024-10-08 23:03:17.975 [error] Handshake failed:
2024-10-08 23:03:17.976 [error] Error: timeout in handshake with retry
at t.RepoClientMultiplexer.handshakeWithRetry (c:\Users\x\AppData\Local\Programs\cursor\resources\app\extensions\cursor-retrieval\dist\main.js:2:826553)
at async w.getServerStatus (c:\Users\x\AppData\Local\Programs\cursor\resources\app\extensions\cursor-retrieval\dist\main.js:2:841881)
at async w.startIndexingRepository (c:\Users\x\AppData\Local\Programs\cursor\resources\app\extensions\cursor-retrieval\dist\main.js:2:844181)
2024-10-08 23:03:17.976 [error] Handshake failed.
2024-10-08 23:03:17.977 [info] Indexing job successfully done or aborted.

[afastlearner]

I just logged out from my cursor account and logged in with a new account. Right after, this issue showed up on all my repositories. Could it be linked?

I had one repo already indexed and the new ones failed. I closed all the cursor windows and opened that indexed repo again and it was already failed, without loging out.

[Raleose]

There is a strange requests under the hood.

The first one is POST https://repo42.cursor.sh/aiserver.v1.RepositoryService/RepositoryStatus HTTP/1.1.
It always completes by status 404 and body:
{"message":"Route POST:/aiserver.v1.RepositoryService/RepositoryStatus not found","error":"Not Found","statusCode":404}
I tried to modify request host to api2.cursor.sh instead of repo42.cursor.sh, but nothing has changed.

The second one is POST https://api2.cursor.sh/aiserver.v1.RepositoryService/FastRepoInitHandshake HTTP/1.1.
It sometimes completes by status 500, but it's very rare case.
In most cases it completes by 200, but if we look inside the proto body we will find that message contains something like that:
{"status":"STATUS_FAILURE","repoName":"some guid here"}
repoName isn't interesting for us, but STATUS_FAILURE - it is number 3 inside of proto serialized body.

@danperks it may be helpful for devs, i think.

[seqis]

Getting the same error: Running Linux Mint 20.3, MATE Edition.

cat ./remoteTunnelService.log 
2024-10-06 20:49:17.539 [error] Missing 'tunnelApplicationConfig' or 'tunnelApplicationName' in product.json. Remote tunneling is not available.
2024-10-07 08:28:35.825 [error] Missing 'tunnelApplicationConfig' or 'tunnelApplicationName' in product.json. Remote tunneling is not available.
2024-10-07 08:47:37.062 [error] Missing 'tunnelApplicationConfig' or 'tunnelApplicationName' in product.json. Remote tunneling is not available.
2024-10-07 09:02:45.765 [error] Missing 'tunnelApplicationConfig' or 'tunnelApplicationName' in product.json. Remote tunneling is not available.
2024-10-08 18:23:33.506 [error] Missing 'tunnelApplicationConfig' or 'tunnelApplicationName' in product.json. Remote tunneling is not available.

[RyanAD]

I have this issue as well. I used Google Login to create my account if that helps narrow it down.

[Takyon236]

Same issue on 2 differents devices

[ubergeekseven]

same issue. I have tried indexing several times with the handshake failed ultimately occurring. I went through my codebase and added as many of the extraneous portions of my libraries dependency folders to the cursor ignore file. That did not help. I thought it would until searching led me here.

[Raleose]

There is a strange requests under the hood.

The first one is POST https://repo42.cursor.sh/aiserver.v1.RepositoryService/RepositoryStatus HTTP/1.1. It always completes by status 404 and body: {"message":"Route POST:/aiserver.v1.RepositoryService/RepositoryStatus not found","error":"Not Found","statusCode":404} I tried to modify request host to api2.cursor.sh instead of repo42.cursor.sh, but nothing has changed.

The second one is POST https://api2.cursor.sh/aiserver.v1.RepositoryService/FastRepoInitHandshake HTTP/1.1. It sometimes completes by status 500, but it's very rare case. In most cases it completes by 200, but if we look inside the proto body we will find that message contains something like that: {"status":"STATUS_FAILURE","repoName":"some guid here"} repoName isn't interesting for us, but STATUS_FAILURE - it is number 3 inside of proto serialized body.

@danperks it may be helpful for devs, i think.

Now indexing works for me. I looked at the response body of FastRepoInitHandshake and it's status has 2 as value. Great job, thank you

[JoanEsquivel]

I do have the same issue.