[AHM/Staking/VMP] Paginated Offence Reports + Retries for Validator Set (#9619)

* Please see the full design do
[here](https://docs.google.com/document/d/1l2COWct1f-gC8nM0tq7Xs8pBWeAP6pX0LociWC6enUg/edit?tab=t.0)
* closes paritytech/srlabs_findings#520

This PR makes the following changes: 

#### Common 

* `SendToRelayChain` and `SendToAssetHub` traits now return a result,
allowing the caller to know if the underlying XCM was sent or not.
* Adds a number of testing facilities to `pallet-root-offences`, and
`staking-async/papi-tests`. Both of which can be ignored in the review.

#### Offences

* `SendToAssetHub::relay_new_offence` is removed. Instead, we use the
new `relay_new_offence_paged` which is a vector of self-contained
offences, not requiring us to group offences per session in each
message.
* Offences are not sent immediately anymore. 
* Instead, they are stored in a paginated `OffenceSendQueue`.
* `on-init`, we grab one page of this storage map, and sent it.

#### Session Report
* Session reports now also have a retry mechanism.
* Upon each failure, we emit an `UnexpectedEvent`
* If our retries run out and we still can't send the session report, we
will emit a different `UnexpectedEvent`. We also retore the validator
points that we meant to send, and merge them back, so that they are sent
in the next session report.

#### Validator Set
* Similar to offences, they are not sent immediately anymore. 
* Instead, they are stored in a storage item, and are sent on subsequent
on-inits.
* A maximum retry count is added. 

### Review notes

As noted above, ignore all changes in 
* `staking-async/runtimes`
* `staking-async/runtimes/papi-tests`
* `root-offences`

As they are only related to testing.

---------

Co-authored-by: Ankan <[email protected]>
Co-authored-by: cmd[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Author: 3 people

Cargo.lock

@@ -14,6 +14,7 @@ workspace = true
 [dependencies]
 codec = { features = ["derive", "max-encoded-len"], workspace = true }
 hex-literal = { workspace = true, default-features = true }
+log = { workspace = true }
 scale-info = { features = ["derive"], workspace = true }
 serde_json = { features = ["alloc"], workspace = true }
 tracing = { workspace = true }
@@ -137,6 +138,7 @@ alloy-core = { workspace = true, features = ["sol-types"] }
 asset-test-utils = { workspace = true, default-features = true }
 pallet-revive-fixtures = { workspace = true, default-features = true }
 parachains-runtimes-test-utils = { workspace = true, default-features = true }
+sp-tracing = { workspace = true, default-features = true }
 
 [build-dependencies]
 substrate-wasm-builder = { optional = true, workspace = true, default-features = true }
@@ -301,6 +303,7 @@ std = [
 	"frame-system-rpc-runtime-api/std",
 	"frame-system/std",
 	"frame-try-runtime?/std",
+	"log/std",
 	"pallet-ah-ops/std",
 	"pallet-asset-conversion-ops/std",
 	"pallet-asset-conversion-tx-payment/std",

cumulus/parachains/runtimes/assets/asset-hub-westend/Cargo.toml

@@ -304,6 +304,7 @@ impl pallet_staking_async_rc_client::Config for Runtime {
 	type RelayChainOrigin = EnsureRoot<AccountId>;
 	type AHStakingInterface = Staking;
 	type SendToRelayChain = StakingXcmToRelayChain;
+	type MaxValidatorSetRetries = ConstU32<64>;
 }
 
 #[derive(Encode, Decode)]
@@ -350,13 +351,13 @@ pub struct StakingXcmToRelayChain;
 
 impl rc_client::SendToRelayChain for StakingXcmToRelayChain {
 	type AccountId = AccountId;
-	fn validator_set(report: rc_client::ValidatorSetReport<Self::AccountId>) {
+	fn validator_set(report: rc_client::ValidatorSetReport<Self::AccountId>) -> Result<(), ()> {
 		rc_client::XCMSender::<
 			xcm_config::XcmRouter,
 			RelayLocation,
 			rc_client::ValidatorSetReport<Self::AccountId>,
 			ValidatorSetToXcm,
-		>::split_then_send(report, Some(8));
+		>::send(report)
 	}
 }
 

cumulus/parachains/runtimes/assets/asset-hub-westend/src/staking.rs

@@ -793,7 +793,7 @@ enum RcClientCalls<AccountId> {
 	#[codec(index = 0)]
 	RelaySessionReport(rc_client::SessionReport<AccountId>),
 	#[codec(index = 1)]
-	RelayNewOffence(SessionIndex, Vec<rc_client::Offence<AccountId>>),
+	RelayNewOffencePaged(Vec<(SessionIndex, rc_client::Offence<AccountId>)>),
 }
 
 pub struct AssetHubLocation;
@@ -842,42 +842,53 @@ impl sp_runtime::traits::Convert<rc_client::SessionReport<AccountId>, Xcm<()>>
 	}
 }
 
-pub struct StakingXcmToAssetHub;
-impl ah_client::SendToAssetHub for StakingXcmToAssetHub {
-	type AccountId = AccountId;
-
-	fn relay_session_report(session_report: rc_client::SessionReport<Self::AccountId>) {
-		rc_client::XCMSender::<
-			xcm_config::XcmRouter,
-			AssetHubLocation,
-			rc_client::SessionReport<AccountId>,
-			SessionReportToXcm,
-		>::split_then_send(session_report, Some(8));
-	}
-
-	fn relay_new_offence(
-		session_index: SessionIndex,
-		offences: Vec<rc_client::Offence<Self::AccountId>>,
-	) {
-		let message = Xcm(vec![
+pub struct QueuedOffenceToXcm;
+impl sp_runtime::traits::Convert<Vec<ah_client::QueuedOffenceOf<Runtime>>, Xcm<()>>
+	for QueuedOffenceToXcm
+{
+	fn convert(offences: Vec<ah_client::QueuedOffenceOf<Runtime>>) -> Xcm<()> {
+		Xcm(vec![
 			Instruction::UnpaidExecution {
 				weight_limit: WeightLimit::Unlimited,
 				check_origin: None,
 			},
 			Instruction::Transact {
 				origin_kind: OriginKind::Superuser,
 				fallback_max_weight: None,
-				call: AssetHubRuntimePallets::RcClient(RcClientCalls::RelayNewOffence(
-					session_index,
+				call: AssetHubRuntimePallets::RcClient(RcClientCalls::RelayNewOffencePaged(
 					offences,
 				))
 				.encode()
 				.into(),
 			},
-		]);
-		if let Err(err) = send_xcm::<xcm_config::XcmRouter>(AssetHubLocation::get(), message) {
-			log::error!(target: "runtime::ah-client", "Failed to send relay offence message: {:?}", err);
-		}
+		])
+	}
+}
+
+pub struct StakingXcmToAssetHub;
+impl ah_client::SendToAssetHub for StakingXcmToAssetHub {
+	type AccountId = AccountId;
+
+	fn relay_session_report(
+		session_report: rc_client::SessionReport<Self::AccountId>,
+	) -> Result<(), ()> {
+		rc_client::XCMSender::<
+			xcm_config::XcmRouter,
+			AssetHubLocation,
+			rc_client::SessionReport<AccountId>,
+			SessionReportToXcm,
+		>::send(session_report)
+	}
+
+	fn relay_new_offence_paged(
+		offences: Vec<ah_client::QueuedOffenceOf<Runtime>>,
+	) -> Result<(), ()> {
+		rc_client::XCMSender::<
+			xcm_config::XcmRouter,
+			AssetHubLocation,
+			Vec<ah_client::QueuedOffenceOf<Runtime>>,
+			QueuedOffenceToXcm,
+		>::send(offences)
 	}
 }
 
@@ -893,7 +904,8 @@ impl ah_client::Config for Runtime {
 	type PointsPerBlock = ConstU32<20>;
 	type MaxOffenceBatchSize = ConstU32<50>;
 	type Fallback = Staking;
-	type WeightInfo = ah_client::weights::SubstrateWeight<Runtime>;
+	type MaximumValidatorsWithPoints = ConstU32<{ MaxActiveValidators::get() * 4 }>;
+	type MaxSessionReportRetries = ConstU32<5>;
 }
 
 impl pallet_fast_unstake::Config for Runtime {
@@ -2143,7 +2155,6 @@ mod benches {
 		[pallet_scheduler, Scheduler]
 		[pallet_session, SessionBench::<Runtime>]
 		[pallet_staking, Staking]
-		[pallet_staking_async_ah_client, StakingAhClient]
 		[pallet_sudo, Sudo]
 		[frame_system, SystemBench::<Runtime>]
 		[frame_system_extensions, SystemExtensionsBench::<Runtime>]

polkadot/runtime/westend/src/lib.rs

@@ -0,0 +1,45 @@
+title: '[AHM/Staking/VMP] Paginated Offence Reports + Retries for Validator Set'
+doc:
+- audience: Runtime Dev
+  description: |-
+    This PR makes the following changes:
+
+    #### Common
+
+    * `SendToRelayChain` and `SendToAssetHub` traits now return a result, allowing the caller to know if the underlying XCM was sent or not.
+
+    #### Offences
+
+    * `SendToAssetHub::relay_new_offence` is removed. Instead, we use the new `relay_new_offence_paged` which is a vector of self-contained offences, not requiring us to group offences per session in each message.
+    * Offences are not sent immediately anymore.
+    * Instead, they are stored in a paginated `OffenceSendQueue`.
+    * `on-init`, we grab one page of this storage map, and sent it.
+
+    #### Session Report
+    * Session reports now also have a retry mechanism.
+    * Upon each failure, we emit an `UnexpectedEvent`
+    * If our retries run out and we still can't send the session report, we will emit a different `UnexpectedEvent`. We also retore the validator points that we meant to send, and merge them back, so that they are sent in the next session report.
+
+    #### Validator Set
+    * Similar to offences, they are not sent immediately anymore.
+    * Instead, they are stored in a storage item, and are sent on subsequent on-inits.
+    * A maximum retry count is added.
+crates:
+- name: pallet-offences
+  bump: patch
+- name: pallet-root-offences
+  bump: major
+- name: pallet-staking-async-ah-client
+  bump: major
+- name: pallet-staking-async
+  bump: patch
+- name: westend-runtime
+  bump: minor
+- name: pallet-staking-async-rc-client
+  bump: major
+- name: pallet-staking-async-rc-runtime-constants
+  bump: patch
+- name: asset-hub-westend-runtime
+  bump: minor
+- name: pallet-election-provider-multi-block
+  bump: patch

polkadot/runtime/westend/src/weights/pallet_staking_async_ah_client.rs

@@ -303,7 +303,6 @@ impl<T: crate::Config> Phase<T> {
 
 	fn are_we_done() -> Self {
 		let query = T::AreWeDone::get();
-		log!(debug, "Are we done? {:?}", query);
 		query
 	}