Cursor seems to be down

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

Cursor seems to be down. When I run a basic command with the agent CLI or in the Cursor IDE, it just spins forever.

Steps to Reproduce

agent "Testing. Is Cursor up? Can you read this prompt?"

Expected Behavior

A response

Operating System

MacOS

Version Information

IDE Version: 3.2.21 (Universal)
CLI 2026.05.01-eea359f

For AI issues: which model did you use?

Auto

For AI issues: add Request ID with privacy disabled

I don’t get a Request ID, because it’s spinning.

Additional Information

Filed a support ticket Ref: th_01KQT21V8YXHG699M4XTQR26EH

Does this stop you from using Cursor

Yes - Cursor is unusable

It’s up now.

Hey, glad it worked. If it happens again, it looks like a known CLI streaming issue with the HTTP/2 transport. It’s being tracked in this thread Cursor cli --connection lost, reconnecting.

Workaround if it hangs again: add this to ~/.cursor/cli-config.json

{ "network": { "useHttp1ForAgent": true } }

That switches the CLI to the HTTP/1.1 fallback.

Let me know if the hang comes back. If it does, it’d help to get a dump from ~/.cursor/cli-logs/ and the CLI version at the time of the issue.

same issue my side.

reconnecting…

and then this image:

@deanrie please help

Hey, it looks like the same HTTP/2 transport issue.

Try these steps:

  • First: Cursor Settings > Network > Run Diagnostics. Check if anything is red.
  • If you’re using the IDE: open App Settings with Cmd/Ctrl+,, search for HTTP/2, turn on Disable HTTP/2, then restart Cursor.
  • If you’re using the CLI agent, add this to ~/.cursor/cli-config.json:
    { "network": { "useHttp1ForAgent": true } }
    

If it still hangs after that, share your Cursor version, OS, and whether you’re using a VPN or a corporate proxy. Zscaler and similar tools often trigger this. Logs from ~/.cursor/cli-logs/ also help if you can reproduce it in the CLI.

everything is in red

Thanks, your screenshots show that DNS resolution is failing for all *.cursor.sh domains, including api2.cursor.sh and authenticator.cursor.sh. So this isn’t an HTTP/2 issue, it’s a DNS issue on your side. The HTTP/2 toggle I shared above won’t help here because DNS resolution fails before it can reach the server.

Try these steps in order:

  1. Change your DNS resolver to Cloudflare 1.1.1.1 / 1.0.0.1 or Quad9 9.9.9.9 / 149.112.112.112. Don’t use 8.8.8.8 for this case, it can cause SSL mismatch issues.
    • Windows: Settings → Network → Change adapter options → connection properties → IPv4 → DNS
    • macOS: System Settings → Network → Details → DNS
  2. If you have a VPN, a corporate proxy like Zscaler, or an antivirus with network filtering, turn it off and test again.
  3. On macOS, disable iCloud Private Relay: System Settings → Apple Account → iCloud → Private Relay. It often breaks CNAME chains.
  4. Check your hosts file: /etc/hosts or C:\Windows\System32\drivers\etc\hosts. There shouldn’t be any entries for cursor.sh.
  5. Quick test: connect to your phone’s mobile hotspot and run diagnostics again. If everything is green, the issue is in your network, ISP, or DNS.

If none of this helps, tell me your OS, country, ISP, and whether you’re using a VPN or corporate network. Also share the output:

  • macOS/Linux: dig api2.cursor.sh +trace
  • Windows: nslookup api2.cursor.sh