Repair Background Service, Worker & Lifecycle Failures

The likely issue is a contract mismatch around background services, workers, daemons, startup, supervision, heartbeats, duplicate instances, leases, crash loops, retry/backoff, health state, dependency readiness, graceful shutdown, and operational ownership. I would first define the invariant: Each intended background role has one authoritative owner and lifecycle; startup, supervision, retry, lease/heartbeat, and shutdown transitions must not create duplicate workers, infinite crash loops, or silent work loss. Then I would trace the smallest owning boundary, keep VERIFIED facts separate from LIKELY/UNKNOWN state, apply or propose the narrowest repair, and add deterministic normal, boundary, failure, retry/cancellation, and recovery tests. The workflow preserves authorization, privacy, security policy, public interfaces, user-visible behavior, and documented ownership. With repository/workspace access, findings can map to exact paths and focused local validation; with partial evidence, the result remains an explicit diagnosis/evidence map with no invented execution claims.