Unclear Whether the tengu_bridge_env_reregister gate is active has not been read for this account.
What
- The bridge's background polling loop, used for Remote Control/cloud sessions, now treats a 401 (unauthorized) response as a trigger to re-register the environment, in addition to the existing 404 (not found) case, provided the same gates hold.
- The re-registration function now takes a
triggerargument distinguishing 401 from 404, logs the correct status code, and records which trigger fired in telemetry. A 401 uses its own backoff floor (reregister401BackoffFloorMs) and logs a message about a renewed Remote Control credential rather than a re-registered environment. - If repeated 401s exhaust re-registration attempts, or the server issues a replaced environment instead, the bridge now logs a distinct, user-facing fatal error telling the user to restart
claude remote-control.
Why Previously only a 404 triggered recovery, so an expired or renewed credential (401) could leave a Remote Control session stuck polling without a clear path forward. Now it's handled automatically where possible, with a clearer error message when it can't be.
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubt
Whether the `tengu_bridge_env_reregister` gate is active has not been read for this account.