Recurring “Connection stalled / LTe: Connection stalled” on Linux (HTTP version switch only temporary fix)

Where does the bug appear (feature/product)?

Cursor IDE – chat/agent (side chat + main chat)

Environment

  • Cursor Version: 2.4.22

  • VSCode Version: 1.105.1

  • Commit: 618c607a249dd7fd2ffc662c6531143833bebd40

  • Build Type: Stable

  • Release Track: Default

  • Electron: 39.2.7

  • Chromium: 142.0.7444.235

  • Node.js: 22.21.1

  • V8: 14.2.231.21-electron.0

  • OS: Linux x64 6.8.0-90-generic

Description

I repeatedly get the following error when sending prompts to the Cursor AI/agent, even though my internet connection is stable:

The connection stalled. Please try again.
LTe: Connection stalled

Stack trace (example):

text

Request ID: 5d091513-835d-4d56-8c50-b080136960ba
Connection stalled
LTe: Connection stalled
at rpf (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9096:4764)
at fva.run (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9096:8170)
at async Hyt.runAgentLoop (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34196:57047)
at async Zpc.streamFromAgentBackend (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34245:7695)
at async Zpc.getAgentStreamResponse (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:34245:8436)
at async FTe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9170:14575)
at async Object.Ei [as onSubmit] (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:32994:3808)
at async vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:32968:59730

The UI message is the same as in other recent reports: “The connection stalled. Please try again.”cursor+2

Frequency and behavior

  • Issue is recurring, not one‑off.

  • It affects normal prompts to the agent/chat (not just very large or complex ones).

  • When it starts happening, every prompt in that session fails with “Connection stalled / LTe: Connection stalled” until I apply a workaround or wait.

  • My regular internet usage (browser, other tools) works normally during these episodes.

Workarounds tried

These work only temporarily; the error keeps coming back:

  1. Log out of Cursor and log back in

    • Sometimes restores normal behavior for a while, but the “Connection stalled” error comes back after some time or after a few prompts.
  2. Switch HTTP version / disable HTTP/2

    • Toggling HTTP/2 off (forcing HTTP/1.1) initially helped, similar to what is suggested in other posts and docs.dev+2

    • However, on my setup the issue reappears even while staying on HTTP/1.1.

    • Switching between HTTP/2 ↔ HTTP/1.1 (and back) sometimes clears the problem for a short period, but it’s not a stable fix, which seems similar to what other users are reporting.forum.cursor+1

  3. Full restart of Cursor

    • Fully quitting Cursor and restarting occasionally helps for a short time, but the same error comes back later in the same day/session.cursor+1
  4. General network sanity checks

    • Stable non‑VPN connection, other tools work, streaming/web usage fine.

    • This matches other reports where diagnostics and general connectivity look normal but Cursor’s agent requests still stall.cursor+1

I have not noticed a reliable pattern tied to specific projects, file types, or prompt sizes; it feels more like the connection to the agent backend or some cached state gets into a bad condition.

Expected behavior

  • Chat/agent requests should stream responses normally without intermittently stalling.

  • If HTTP/2 is problematic on some networks, switching to HTTP/1.1 should provide a stable fallback, not just a short‑term workaround.qiita+1

Actual behavior

  • The agent/chat becomes unreliable: prompts hang and then fail with “Connection stalled / LTe: Connection stalled” until I log out, restart, or toggle HTTP version.

  • The same problem recurs after some time, even when keeping the same HTTP mode and network setup.

Steps to reproduce (from my experience)

It’s intermittent, but roughly:

  1. Open Cursor IDE.

  2. Open a project and start a chat/agent session.

  3. Send several prompts over time (normal coding/assistant usage).

  4. At some point, new prompts stop working and show:

    The connection stalled. Please try again.

    LTe: Connection stalled

  5. Logging out or switching HTTP version fixes it for a while, then it reappears.

If there is any additional logging or diagnostic flag I can enable to capture more detailed traces around Hyt.runAgentLoop / Zpc.streamFromAgentBackend when this happens, I’m happy to provide more data.

Does this stop you from using Cursor?

  • It doesn’t crash the editor itself, but it significantly disrupts AI workflows.

  • During these episodes, the AI/agent is effectively unusable until I apply a workaround or wait; because it keeps recurring, it’s impacting productivity.

Additional notes / links to similar issues

  • Other users are reporting the same “Connection stalled. Please try again.” message, with similar stack traces and temporary relief from HTTP/2 → HTTP/1.1 workarounds.cursor+3

  • There are also related threads around persistent “Connection failed” / HTTP/2 instability that suggest a broader network/protocol issue on some regions/ISPs.reddit+2

If this is a known regression in 2.4.x, it would be helpful to know:

  • Whether it’s being tracked as a specific bug, and

  • Whether there’s a more permanent workaround (e.g., clearing specific caches, project index reset, specific network setting) I can apply on Linux.


Note: Above text written by AI → context given by human.

Hi there!

We detected that this may be a bug report, so we’ve moved your post to the Bug Reports category.

To help us investigate and fix this faster, could you edit your original post to include the details from the template below?

Bug Report Template - Click to expand

Where does the bug appear (feature/product)?

  • Cursor IDE
  • Cursor CLI
  • Background Agent (GitHub, Slack, Web, Linear)
  • BugBot
  • Somewhere else…

Describe the Bug
A clear and concise description of what the bug is.


Steps to Reproduce
How can you reproduce this bug? We have a much better chance at fixing issues if we can reproduce them!


Expected Behavior
What is meant to happen here that isn’t working correctly?


Screenshots / Screen Recordings
If applicable, attach images or videos (.jpg, .png, .gif, .mp4, .mov)


Operating System

  • Windows 10/11
  • MacOS
  • Linux

Version Information

  • For Cursor IDE: Menu → About Cursor → Copy
  • For Cursor CLI: Run agent about in your terminal
IDE:
Version: 2.xx.x
VSCode Version: 1.105.1
Commit: ......

CLI:
CLI Version 2026.01.17-d239e66

For AI issues: which model did you use?
Model name (e.g., Sonnet 4, Tab…)


For AI issues: add Request ID with privacy disabled
Request ID: f9a7046a-279b-47e5-ab48-6e8dc12daba1
For Background Agent issues, also post the ID: bc-…


Additional Information
Add any other context about the problem here.


Does this stop you from using Cursor?

  • Yes - Cursor is unusable
  • Sometimes - I can sometimes use Cursor
  • No - Cursor works, but with this issue

The more details you provide, the easier it is for us to reproduce and fix the issue. Thanks!

Hey, thanks for the report. This is a known issue, and over the last few weeks several users have reported the same stack trace: LTe: Connection stalled.

The most effective workaround right now is to reset the corrupted workspace index. You have two options:

Option 1: rename the project folder:

MyProject → MyProject_v2

Then open it again in Cursor. This forces a new index to be created. See this resolved thread with the exact same stack trace: Connection Error / Stalled Agent requests for hours - #9 by Fares_Bendhiab

Option 2: delete the project cache directly:

rm -rf ~/.cursor/projects/<your-project-folder>

Then restart Cursor. This is less destructive than renaming, and your chat history will be preserved.

If neither option helps, I’ll need:

  • Network diagnostics: Cursor Settings > Network > Run Diagnostics (screenshot)
  • Console errors: Help > Toggle Developer Tools > Console (screenshot of the red errors when it freezes)
  • Which model are you using?

The team is aware of the issue. Your report helps with prioritization, especially since you’re on Linux, and most previous reports were on macOS.

Let me know how it goes.

1 Like

Update: it seems to be working after resetting Cursor config and Cache (with the loss of MCP config and other things).

command used:
rm -rf ~/.cursor ~/.config/Cursor ~/.local/share/Cursor

P.s. Did not see the advice to delete the project cache haha, had deleted everything until then.

1 Like