Make Membership::leader() return a Result<_>

Cargo.toml

@@ -24,6 +24,7 @@ members = [
   "crates/types",
   "crates/builder-api",
   "crates/fakeapi",
+  "crates/utils",
 ]
 resolver = "2"
 

clippy.toml

@@ -0,0 +1 @@
+allowed-wildcard-imports = [ "utils", "hotshot_task_impls", "hotshot_types" ]

crates/example-types/src/node_types.rs

@@ -4,12 +4,6 @@
 // You should have received a copy of the MIT License
 // along with the HotShot repository. If not, see <https://mit-license.org/>.
 
-use crate::{
-    auction_results_provider_types::{TestAuctionResult, TestAuctionResultsProvider},
-    block_types::{TestBlockHeader, TestBlockPayload, TestTransaction},
-    state_types::{TestInstanceState, TestValidatedState},
-    storage_types::TestStorage,
-};
 use hotshot::traits::{
     election::{
         randomized_committee::RandomizedCommittee, static_committee::StaticCommittee,
@@ -18,15 +12,21 @@ use hotshot::traits::{
     implementations::{CombinedNetworks, Libp2pNetwork, MemoryNetwork, PushCdnNetwork},
     NodeImplementation,
 };
-use hotshot_types::data::EpochNumber;
 use hotshot_types::{
-    data::ViewNumber,
+    data::{EpochNumber, ViewNumber},
     signature_key::{BLSPubKey, BuilderKey},
     traits::node_implementation::{NodeType, Versions},
 };
 use serde::{Deserialize, Serialize};
 use vbs::version::StaticVersion;
 
+use crate::{
+    auction_results_provider_types::{TestAuctionResult, TestAuctionResultsProvider},
+    block_types::{TestBlockHeader, TestBlockPayload, TestTransaction},
+    state_types::{TestInstanceState, TestValidatedState},
+    storage_types::TestStorage,
+};
+
 #[derive(
     Copy,
     Clone,

crates/hotshot/Cargo.toml

@@ -58,6 +58,7 @@ sha2 = { workspace = true }
 url = { workspace = true }
 num_enum = "0.7"
 parking_lot = "0.12"
+utils = { path = "../utils" }
 
 [target.'cfg(all(async_executor_impl = "tokio"))'.dependencies]
 tokio = { workspace = true }

crates/hotshot/src/tasks/mod.rs

@@ -11,8 +11,7 @@ pub mod task_state;
 use std::{fmt::Debug, sync::Arc, time::Duration};
 
 use async_broadcast::{broadcast, RecvError};
-use async_compatibility_layer::art::async_sleep;
-use async_compatibility_layer::art::async_spawn;
+use async_compatibility_layer::art::{async_sleep, async_spawn};
 use async_lock::RwLock;
 use async_trait::async_trait;
 use futures::{

crates/hotshot/src/traits/election/randomized_committee.rs

@@ -17,6 +17,7 @@ use hotshot_types::{
     PeerConfig,
 };
 use rand::{rngs::StdRng, Rng};
+use utils::anytrace::Result;
 
 #[derive(Clone, Debug, Eq, PartialEq, Hash)]
 
@@ -142,7 +143,7 @@ impl<TYPES: NodeType> Membership<TYPES> for RandomizedCommittee<TYPES> {
         &self,
         view_number: TYPES::View,
         _epoch: <TYPES as NodeType>::Epoch,
-    ) -> TYPES::SignatureKey {
+    ) -> Result<TYPES::SignatureKey> {
         let mut rng: StdRng = rand::SeedableRng::seed_from_u64(*view_number);
 
         let randomized_view_number: u64 = rng.gen_range(0..=u64::MAX);
@@ -151,7 +152,7 @@ impl<TYPES: NodeType> Membership<TYPES> for RandomizedCommittee<TYPES> {
 
         let res = self.eligible_leaders[index].clone();
 
-        TYPES::SignatureKey::public_key(&res)
+        Ok(TYPES::SignatureKey::public_key(&res))
     }
 
     /// Get the total number of nodes in the committee

crates/hotshot/src/traits/election/static_committee.rs

@@ -16,6 +16,7 @@ use hotshot_types::{
     },
     PeerConfig,
 };
+use utils::anytrace::Result;
 
 #[derive(Clone, Debug, Eq, PartialEq, Hash)]
 
@@ -140,11 +141,11 @@ impl<TYPES: NodeType> Membership<TYPES> for StaticCommittee<TYPES> {
         &self,
         view_number: TYPES::View,
         _epoch: <TYPES as NodeType>::Epoch,
-    ) -> TYPES::SignatureKey {
+    ) -> Result<TYPES::SignatureKey> {
         #[allow(clippy::cast_possible_truncation)]
         let index = *view_number as usize % self.eligible_leaders.len();
         let res = self.eligible_leaders[index].clone();
-        TYPES::SignatureKey::public_key(&res)
+        Ok(TYPES::SignatureKey::public_key(&res))
     }
 
     /// Get the total number of nodes in the committee

crates/hotshot/src/traits/election/static_committee_leader_two_views.rs

@@ -16,6 +16,7 @@ use hotshot_types::{
     },
     PeerConfig,
 };
+use utils::anytrace::Result;
 
 #[derive(Clone, Debug, Eq, PartialEq, Hash)]
 
@@ -140,11 +141,12 @@ impl<TYPES: NodeType> Membership<TYPES> for StaticCommitteeLeaderForTwoViews<TYP
         &self,
         view_number: TYPES::View,
         _epoch: <TYPES as NodeType>::Epoch,
-    ) -> TYPES::SignatureKey {
+    ) -> Result<TYPES::SignatureKey> {
         let index =
             usize::try_from((*view_number / 2) % self.eligible_leaders.len() as u64).unwrap();
         let res = self.eligible_leaders[index].clone();
-        TYPES::SignatureKey::public_key(&res)
+
+        Ok(TYPES::SignatureKey::public_key(&res))
     }
 
     /// Get the total number of nodes in the committee

crates/hotshot/src/traits/networking/libp2p_network.rs

@@ -1025,7 +1025,15 @@ impl<K: SignatureKey + 'static> ConnectedNetwork<K> for Libp2pNetwork<K> {
     {
         let future_view = <TYPES as NodeType>::View::new(view) + LOOK_AHEAD;
         let epoch = <TYPES as NodeType>::Epoch::new(epoch);
-        let future_leader = membership.leader(future_view, epoch);
+        let future_leader = match membership.leader(future_view, epoch) {
+            Ok(l) => l,
+            Err(e) => {
+                return tracing::info!(
+                    "Failed to calculate leader for view {:?}: {e}",
+                    future_view
+                );
+            }
+        };
 
         let _ = self
             .queue_node_lookup(ViewNumber::new(*future_view), future_leader)

crates/hotshot/src/types/handle.rs

@@ -8,7 +8,7 @@
 
 use std::sync::Arc;
 
-use anyhow::{anyhow, Ok, Result};
+use anyhow::{anyhow, Context, Ok, Result};
 use async_broadcast::{InactiveReceiver, Receiver, Sender};
 use async_lock::RwLock;
 use committable::{Commitment, Committable};
@@ -315,16 +315,20 @@ impl<TYPES: NodeType, I: NodeImplementation<TYPES> + 'static, V: Versions>
     }
 
     /// Wrapper for `HotShotConsensusApi`'s `leader` function
+    ///
+    /// # Errors
+    /// Returns an error if the leader cannot be calculated
     #[allow(clippy::unused_async)] // async for API compatibility reasons
     pub async fn leader(
         &self,
         view_number: TYPES::View,
         epoch_number: TYPES::Epoch,
-    ) -> TYPES::SignatureKey {
+    ) -> Result<TYPES::SignatureKey> {
         self.hotshot
             .memberships
             .quorum_membership
             .leader(view_number, epoch_number)
+            .context("Failed to lookup leader")
     }
 
     // Below is for testing only:

crates/libp2p-networking/src/network/behaviours/dht/store.rs

@@ -37,8 +37,16 @@ impl<R: RecordStore, K: SignatureKey> RecordStore for ValidatedStore<R, K>
 where
     K: 'static,
 {
-    type ProvidedIter<'a> = R::ProvidedIter<'a> where R: 'a, K: 'a;
-    type RecordsIter<'a> = R::RecordsIter<'a> where R: 'a, K: 'a;
+    type ProvidedIter<'a>
+        = R::ProvidedIter<'a>
+    where
+        R: 'a,
+        K: 'a;
+    type RecordsIter<'a>
+        = R::RecordsIter<'a>
+    where
+        R: 'a,
+        K: 'a;
 
     // Delegate all `RecordStore` methods except `put` to the inner store
     delegate! {

crates/task-impls/Cargo.toml

@@ -38,6 +38,7 @@ tagged-base64 = { workspace = true }
 time = { workspace = true }
 tracing = { workspace = true }
 url = { workspace = true }
+utils = { path = "../utils" }
 vbs = { workspace = true }
 vec1 = { workspace = true }
 

crates/task-impls/src/consensus/handlers.rs

@@ -6,7 +6,6 @@
 
 use std::{sync::Arc, time::Duration};
 
-use anyhow::{ensure, Context, Result};
 use async_broadcast::Sender;
 use async_compatibility_layer::art::{async_sleep, async_spawn};
 use chrono::Utc;
@@ -19,7 +18,8 @@ use hotshot_types::{
     },
     vote::HasViewNumber,
 };
-use tracing::{debug, error, instrument};
+use tracing::instrument;
+use utils::anytrace::*;
 
 use super::ConsensusTaskState;
 use crate::{
@@ -44,9 +44,9 @@ pub(crate) async fn handle_quorum_vote_recv<
     ensure!(
         task_state
             .quorum_membership
-            .leader(vote.view_number() + 1, task_state.cur_epoch)
+            .leader(vote.view_number() + 1, task_state.cur_epoch)?
             == task_state.public_key,
-        format!(
+        info!(
             "We are not the leader for view {:?}",
             vote.view_number() + 1
         )
@@ -63,7 +63,7 @@ pub(crate) async fn handle_quorum_vote_recv<
         sender,
         &task_state.upgrade_lock,
     )
-    .await;
+    .await?;
 
     Ok(())
 }
@@ -83,9 +83,9 @@ pub(crate) async fn handle_timeout_vote_recv<
     ensure!(
         task_state
             .timeout_membership
-            .leader(vote.view_number() + 1, task_state.cur_epoch)
+            .leader(vote.view_number() + 1, task_state.cur_epoch)?
             == task_state.public_key,
-        format!(
+        info!(
             "We are not the leader for view {:?}",
             vote.view_number() + 1
         )
@@ -102,7 +102,7 @@ pub(crate) async fn handle_timeout_vote_recv<
         sender,
         &task_state.upgrade_lock,
     )
-    .await;
+    .await?;
 
     Ok(())
 }
@@ -124,7 +124,7 @@ pub(crate) async fn handle_view_change<
     );
 
     let old_view_number = task_state.cur_view;
-    debug!("Updating view from {old_view_number:?} to {new_view_number:?}");
+    tracing::debug!("Updating view from {old_view_number:?} to {new_view_number:?}");
 
     // Move this node to the next view
     task_state.cur_view = new_view_number;
@@ -138,7 +138,7 @@ pub(crate) async fn handle_view_change<
         .clone();
     if let Some(cert) = decided_upgrade_certificate_read {
         if new_view_number == cert.data.new_version_first_view {
-            error!(
+            tracing::error!(
                 "Version upgraded based on a decided upgrade cert: {:?}",
                 cert
             );
@@ -177,7 +177,7 @@ pub(crate) async fn handle_view_change<
     let cur_view_time = Utc::now().timestamp();
     if task_state
         .quorum_membership
-        .leader(old_view_number, task_state.cur_epoch)
+        .leader(old_view_number, task_state.cur_epoch)?
         == task_state.public_key
     {
         #[allow(clippy::cast_precision_loss)]
@@ -228,7 +228,7 @@ pub(crate) async fn handle_timeout<TYPES: NodeType, I: NodeImplementation<TYPES>
         task_state
             .timeout_membership
             .has_stake(&task_state.public_key, task_state.cur_epoch),
-        format!("We were not chosen for the consensus committee for view {view_number:?}")
+        debug!("We were not chosen for the consensus committee for view {view_number:?}")
     );
 
     let vote = TimeoutVote::create_signed_vote(
@@ -239,7 +239,8 @@ pub(crate) async fn handle_timeout<TYPES: NodeType, I: NodeImplementation<TYPES>
         &task_state.upgrade_lock,
     )
     .await
-    .context("Failed to sign TimeoutData")?;
+    .wrap()
+    .context(error!("Failed to sign TimeoutData"))?;
 
     broadcast_event(Arc::new(HotShotEvent::TimeoutVoteSend(vote)), sender).await;
     broadcast_event(
@@ -251,7 +252,7 @@ pub(crate) async fn handle_timeout<TYPES: NodeType, I: NodeImplementation<TYPES>
     )
     .await;
 
-    debug!(
+    tracing::debug!(
         "We did not receive evidence for view {} in time, sending timeout vote for that view!",
         *view_number
     );
@@ -274,7 +275,7 @@ pub(crate) async fn handle_timeout<TYPES: NodeType, I: NodeImplementation<TYPES>
         .add(1);
     if task_state
         .quorum_membership
-        .leader(view_number, task_state.cur_epoch)
+        .leader(view_number, task_state.cur_epoch)?
         == task_state.public_key
     {
         task_state