Two workers update the same resource and one receives a conflict.
Why this happens
Concurrent writes can race: each worker reads an older state, then both attempt an incompatible update. Retrying the same stale request often creates another conflict.
Answer 1: Read fresh state and retry only the safe operation
On conflict, retrieve the current resource, determine whether the desired change is already present, then rebuild the update from current state. Use a retry limit.
