Hey, thanks for the report.
After upgrading, the old cursor-server process on the remote host doesn’t shut down automatically, which blocks the install of the new version.
Fix:
On the remote host, run:
pkill -f .cursor-server
rm -rf ~/.cursor-server /tmp/cursor-remote-lock.* /tmp/cursor-server-*.tar.gz
If the 30-second timeout isn’t enough for your network, add this to Cursor Settings (JSON):
"remote.SSH.connectTimeout": 120
A similar issue was fixed here: Connecting to Remote-SSH always fails Connecting to Remote-SSH always fails
The team is aware of this issue and is working on automatically cleaning up old processes during upgrades. Let me know if the workaround helps.