Files
VoiceStudio/tests/test_worker_errors.py
T
velixio aa1d739843 feat(workers): dubbing goes remote, and the protocol stops lying to old workers
The remote-GPU line, verified on hardware rather than asserted.

**Dubbing renders on the worker.** dub_generate.py dispatches the coarse
`dub_segments` operation through the gateway, following the audiobook
pattern: per-unit local fallback after consecutive remote failures, one
aggregated notice rather than one per segment. A 40-minute dub that loses
its worker at segment 200 degrades instead of producing 200 error rows.

**An out-of-date worker is now refused by name.** This was the worst
defect in the plan and it was silent: an un-upgraded worker registered
cleanly, then ignored `inputs` and rendered a clone with NO reference
audio — returned as success. A plausible wrong result with nothing
anywhere to surface it. Workers now declare features, and one missing
them is turned away with the features named and `no task was run`.
Verified live: a worker one commit behind was correctly refused.

**"Offline" and "cannot run this" are different facts.** Asking a live
worker for an engine it lacks answered "is offline or cannot be reached.
Wake the selected worker" — while that worker reported ready, one free
slot and 3.6 ms latency. The user was sent to wake a machine that was
already awake. The scheduler now distinguishes absent from present-but-
incapable, and names the engine rather than the operation, because the
engine is the thing a user can install.

**An engine with no catalog entry is no longer hidden.** A `repo_ids`
non-emptiness check had been implemented as a runtime filter, so a worker
silently refused to advertise any engine lacking a models.yaml entry —
which is four registered engines, including CosyVoice. Users with those
already installed would have lost remote support with only a log line.
Empty `repo_ids` now means "not downloadable here", never "not runnable".

**And a script so this stops being done by hand.**
scripts/verify-remote-worker.sh runs the per-phase acceptance checks
against a live worker, non-destructively. Its preconditions are the
mistakes that cost the most time: exactly one listener on the control
port (two instances silently shared it), and never detecting the worker
with a pgrep pattern that matches the ssh shell running it.

Its first real run found the dubbing picker claiming remote placement.
That turned out to be the CHECK being stale, not the picker — the port
had landed since it was written. It now asserts self-consistency instead:
the picker may claim remote only for an operation the control plane
actually advertises as remotely producible, which cannot rot the next
time an op is ported.

Backend 5291 passed, frontend 1812 passed. Acceptance script: no
automated failures across Phases 4-8 on an RTX 4090. Four checks remain
MANUAL by design — true airplane mode, concurrent downloads, killing a
worker mid-audiobook, and the model-list UI — and are reported as
unverified rather than passed.
2026-08-11 17:39:46 +05:30

114 lines
3.8 KiB
Python

"""Error taxonomy: would retrying somewhere else help, and who pays for it.
The scheduler needs two independent answers per failure, and conflating them is
what produces the two worst behaviours: a poison task that rotates through
every worker in the fleet, and a healthy fleet that quarantines itself for
being busy.
"""
from __future__ import annotations
import pytest
from worker import errors
from worker.errors import ErrorClass, WorkerError, from_exception, from_reason
@pytest.mark.parametrize(
"cls,retryable",
[
(ErrorClass.TRANSIENT, True),
(ErrorClass.CAPABILITY, True),
(ErrorClass.CAPACITY, True),
(ErrorClass.TIMEOUT, True),
(ErrorClass.TERMINAL, False),
(ErrorClass.PROTOCOL, False),
],
)
def test_retryability_matrix(cls, retryable):
assert cls.retryable is retryable
@pytest.mark.parametrize(
"cls,charges",
[
(ErrorClass.TRANSIENT, True),
(ErrorClass.TIMEOUT, True),
# Declining work you cannot or should not take is not misbehaviour.
(ErrorClass.CAPACITY, False),
(ErrorClass.CAPABILITY, False),
(ErrorClass.TERMINAL, False),
(ErrorClass.PROTOCOL, False),
],
)
def test_chargeability_matrix(cls, charges):
assert cls.charges_worker is charges
def test_retryable_and_chargeable_are_independent():
"""Capability failures retry elsewhere yet never charge the worker — the
two questions must not collapse into one flag."""
assert ErrorClass.CAPABILITY.retryable is True
assert ErrorClass.CAPABILITY.charges_worker is False
def test_protocol_codes_classify():
assert errors.classify_code("WORKER_AT_CAPACITY") is ErrorClass.CAPACITY
assert errors.classify_code("INSUFFICIENT_MEMORY") is ErrorClass.CAPABILITY
assert errors.classify_code("EXECUTION_TIMEOUT") is ErrorClass.TIMEOUT
assert errors.classify_code("STALE_EPOCH") is ErrorClass.PROTOCOL
assert errors.classify_code("INVALID_TASK_PARAMS") is ErrorClass.TERMINAL
def test_docs_taxonomy_keys_are_reused_not_reinvented():
"""A failure that already has a user-facing hint keeps it on the wire."""
assert errors.classify_code("HF_AUTH_FAILED") is ErrorClass.TERMINAL
assert errors.classify_code("BROKEN_VENV") is ErrorClass.CAPABILITY
assert errors.classify_code("VIDEO_DOWNLOAD_NETWORK") is ErrorClass.TRANSIENT
def test_unknown_codes_default_to_retryable():
"""One wasted retry beats permanently failing work that would have run."""
assert errors.classify_code("SOMETHING_NEW") is ErrorClass.TRANSIENT
def test_secrets_never_reach_the_wire():
"""The worker is a remote machine whose logs the user may never read, so
scrubbing has to happen before the error is sent, not at display time."""
err = from_reason("failed with token hf_ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789")
assert "hf_ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789" not in err.message
def test_hint_is_carried_when_the_taxonomy_has_one():
err = from_reason("boom", code="HF_AUTH_FAILED")
assert err.error_class is ErrorClass.TERMINAL
assert err.hint
def test_every_protocol_code_has_an_actionable_hint():
from core.failure import _HINTS
assert errors._PROTOCOL_CODES.keys() <= _HINTS.keys()
assert all(_HINTS[code].strip() for code in errors._PROTOCOL_CODES)
def test_from_exception_never_produces_an_empty_message():
"""Empty str(e) was the root of the 'unknown error' reports (#122/#63)."""
class Silent(Exception):
pass
assert from_exception(Silent()).message
def test_wire_shape_is_complete():
payload = WorkerError(
error_class=ErrorClass.TIMEOUT, code="EXECUTION_TIMEOUT", message="m", hint="h"
).to_dict()
assert payload == {
"error_class": "timeout",
"code": "EXECUTION_TIMEOUT",
"message": "m",
"hint": "h",
"retryable": True,
}