Skip to content

QA: RTMP ingest restart calls undefined launchPublisher #817

Description

@iamfatness

During #802's actual RTMP CPU-preparation validation, scripts/validate-srt-ingest.mjs reached publisher reconnect and threw launchPublisher is not defined. The script defines only startPublisher; the RTMP reconnect path also needs to assign the returned child to publisher so cleanup and exit/error checks follow the new process.

The prepared application path decoded video/audio, admitted its GPU source and recorded 1,060+ frames with no reported recording loss before this harness exception. This FAIL is retained; it does not qualify RTMP recovery. The undefined call is present in the pre-change script on main.

Done when: the validator uses its actual publisher factory/handle and a real local RTMP restart verifies fresh decoded video/audio and prepared source epoch. This is unranked intake, handled only as the test prerequisite within active #802; it does not create a separate production workstream or promote backlog rank.

Activity

  1. iamfatness commented on Oct 7, 2026

    @iamfatness
    OwnerAuthor

    The undefined RTMP publisher factory was corrected in merged PR #818. Actual local RTMP publisher reconnect restored fresh prepared video and decoded audio; the 1,060-frame recording had zero missing/dropped frames and independent source-pattern/audio checks passed. Original harness FAIL is retained in the #802 evidence.

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

    backlogRanked in docs/BACKLOG.md

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions