Skip to content

Home Assistant: delete ring_push_guard once ring_doorbell ships the ding.id fix (ADR-0101) #1240

Open

Description

@Gerrrt

ADR-0101 adds ring_push_guard, our custom component that patches ring_doorbell 0.9.14's RingEventListener._get_ring_event. Some Ring pushes have no ding.id, three in a row make firebase_messaging shut the push receiver down, and after that no Ring motion or doorbell event reaches Home Assistant (python-ring-doorbell/python-ring-doorbell#537). On trinity it died at 13:12 and at 14:09 on 2026-10-21.

The patch is meant to be temporary. It applies only to 0.9.14, and it logs a warning when Home Assistant's ring_doorbell is any other version.

Activity

  1. added this to the Nothing blocks these milestone on Oct 21, 2026
  2. added
    featureA new capability or a change
    sensitivestacks/sensitive on trinity (VLAN 99)
    on Oct 21, 2026
  3. Gerrrt commented on Nov 10, 2026

    @Gerrrt
    OwnerAuthor

    Status on 2026-11-10:

    Upstream: ring_doorbell's last release is still 0.9.14. python-ring-doorbell/python-ring-doorbell#540 and #556 are open, with merge blocked and no activity since 2026-09-15. Home Assistant 2026.11.1 and dev both pin ring-doorbell==0.9.14. So the removal step stays blocked.

    The first checkbox is not met, for a reason this patch doesn't cover. On trinity, the Ring event.* entities recorded no real motion or doorbell event in the recorder's ten days (since 2026-10-31). There was no KeyError: 'id'. Instead, firebase_messaging 0.4.5 logged Unknown error: Incorrect padding, shutting down FcmPushClient. It slices the crypto-key header by position, and some Ring pushes now carry dh=…; p256ecdsa=… (sdb9696/firebase-messaging#44). This happens before _get_ring_event runs, and the push is never acknowledged, so it comes back at every login.

    #1386 adds a second version-gated patch for that to ring_push_guard (ADR-0110). #1383 tracks its removal separately. This issue's first checkbox can be re-checked once #1386 is deployed.

  4. Gerrrt commented on Nov 11, 2026

    @Gerrrt
    OwnerAuthor

    Status on 2026-11-11:

    #1386 is now running. It merged 2026-11-10 14:58, but Home Assistant had last started at 04:31 that day. The read-only mount already had fcm_patch.py, but the module wasn't loaded. Home Assistant was restarted at 02:49 and 02:50 UTC on 2026-11-11, and the current process started at 02:50:55.

    In the minutes after, it looks clean but nothing is proven yet. There's no Incorrect padding, no shutting down FcmPushClient, no KeyError: 'id', no Dropped a push and no version-mismatch warning. Before this, the padding crash came within a minute of each start. The recorder has no real Ring event.* state since the restart, but it's the middle of the night. Checkbox 1 stays open until a day of real motion events shows up in the recorder. That check is set for about 2026-11-12 03:00 UTC, and #1383 is waiting on the same evidence.

    Upstream hasn't changed. ring_doorbell is still at 0.9.14, python-ring-doorbell/python-ring-doorbell#540 and #556 are still open with no activity since 2026-09-15, and Home Assistant 2026.11.1 still pins ring-doorbell==0.9.14. The removal step stays blocked.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    featureA new capability or a changesensitivestacks/sensitive on trinity (VLAN 99)

    Projects

    No projects

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions