PVF: Don't dispute on missing artifact (#7011)

* PVF: Don't dispute on missing artifact

A dispute should never be raised if the local cache doesn't provide a certain
artifact. You can not dispute based on this reason, as it is a local hardware
issue and not related to the candidate to check.

Design:

Currently we assume that if we prepared an artifact, it remains there on-disk
until we prune it, i.e. we never check again if it's still there.

We can change it so that instead of artifact-not-found triggering a dispute, we
retry once (like we do for AmbiguousWorkerDeath, except we don't dispute if it
still doesn't work). And when enqueuing an execute job, we check for the
artifact on-disk, and start preparation if not found.

Changes:

- [x] Integration test (should fail without the following changes)
- [x] Check if artifact exists when executing, prepare if not
- [x] Return an internal error when file is missing
- [x] Retry once on internal errors
- [x] Document design (update impl guide)

* Add some context to wasm error message (it is quite long)

* Fix impl guide

* Add check for missing/inaccessible file

* Add comment referencing Substrate issue

* Add test for retrying internal errors

---------

Co-authored-by: parity-processbot <>

polkadot/node/core/candidate-validation/src/tests.rs

@@ -672,6 +672,62 @@ fn candidate_validation_multiple_ambiguous_errors_is_invalid() {
 	assert_matches!(v, ValidationResult::Invalid(InvalidCandidate::ExecutionError(_)));
 }
 
+// Test that we retry on internal errors.
+#[test]
+fn candidate_validation_retry_internal_errors() {
+	let validation_data = PersistedValidationData { max_pov_size: 1024, ..Default::default() };
+
+	let pov = PoV { block_data: BlockData(vec![1; 32]) };
+	let validation_code = ValidationCode(vec![2; 16]);
+
+	let descriptor = make_valid_candidate_descriptor(
+		ParaId::from(1_u32),
+		dummy_hash(),
+		validation_data.hash(),
+		pov.hash(),
+		validation_code.hash(),
+		dummy_hash(),
+		dummy_hash(),
+		Sr25519Keyring::Alice,
+	);
+
+	let check = perform_basic_checks(
+		&descriptor,
+		validation_data.max_pov_size,
+		&pov,
+		&validation_code.hash(),
+	);
+	assert!(check.is_ok());
+
+	let candidate_receipt = CandidateReceipt { descriptor, commitments_hash: Hash::zero() };
+
+	let pool = TaskExecutor::new();
+	let (mut ctx, ctx_handle) =
+		test_helpers::make_subsystem_context::<AllMessages, _>(pool.clone());
+	let metrics = Metrics::default();
+
+	let v = test_with_executor_params(ctx_handle, || {
+		validate_candidate_exhaustive(
+			ctx.sender(),
+			MockValidateCandidateBackend::with_hardcoded_result_list(vec![
+				Err(ValidationError::InternalError("foo".into())),
+				// Throw an AWD error, we should still retry again.
+				Err(ValidationError::InvalidCandidate(WasmInvalidCandidate::AmbiguousWorkerDeath)),
+				// Throw another internal error.
+				Err(ValidationError::InternalError("bar".into())),
+			]),
+			validation_data,
+			validation_code,
+			candidate_receipt,
+			Arc::new(pov),
+			PvfExecTimeoutKind::Backing,
+			&metrics,
+		)
+	});
+
+	assert_matches!(v, Err(ValidationFailed(s)) if s == "bar".to_string());
+}
+
 #[test]
 fn candidate_validation_timeout_is_internal_error() {
 	let validation_data = PersistedValidationData { max_pov_size: 1024, ..Default::default() };