go/scheduler: Change scheduler functions to return ok

go/consensus/cometbft/apps/roothash/finalization.go

@@ -139,9 +139,10 @@ func (app *rootHashApplication) tryFinalizeRoundInsideTx( //nolint: gocyclo
 	livenessStats.TotalRounds++
 
 	// Record if the highest-ranked scheduler received enough commitments.
-	firstSchedulerIdx, err := rtState.Committee.SchedulerIdx(round, 0)
-	if err != nil {
-		return err
+	firstSchedulerIdx, ok := rtState.Committee.SchedulerIdx(round, 0)
+	if !ok {
+		// Should never happen.
+		return fmt.Errorf("failed to query primary scheduler, no workers in committee")
 	}
 	firstScheduler := rtState.Committee.Members[firstSchedulerIdx]
 
@@ -331,13 +332,12 @@ func (app *rootHashApplication) failRound(
 	)
 
 	// Record that the scheduler did not receive enough commitments.
-	schedulerIdx, err := rtState.Committee.SchedulerIdx(round, 0)
-	if err != nil {
-		// No workers in the committee.
-		return err
+	firstSchedulerIdx, ok := rtState.Committee.SchedulerIdx(round, 0)
+	if !ok {
+		return fmt.Errorf("failed to query primary scheduler, no workers in committee")
 	}
 
-	rtState.LivenessStatistics.MissedProposals[schedulerIdx]++
+	rtState.LivenessStatistics.MissedProposals[firstSchedulerIdx]++
 
 	if err := app.finalizeBlock(ctx, rtState, block.RoundFailed, nil); err != nil {
 		return fmt.Errorf("failed to emit empty block: %w", err)

go/oasis-node/cmd/control/runtime_stats.go

@@ -430,10 +430,11 @@ func doRuntimeStats(cmd *cobra.Command, args []string) { //nolint:gocyclo
 				continue
 			}
 			// Set committee info.
+			var ok bool
 			currentCommittee = state.Committee
-			currentScheduler, err = currentCommittee.Scheduler(currentRound, 0)
-			if err != nil {
-				logger.Error("failed to query transaction scheduler",
+			currentScheduler, ok = currentCommittee.Scheduler(currentRound, 0)
+			if !ok {
+				logger.Error("failed to query primary scheduler, no workers in committee",
 					"err", err,
 					"height", height,
 				)

go/oasis-node/cmd/debug/byzantine/scheduler.go

@@ -76,9 +76,6 @@ func schedulerCheckScheduled(committee *scheduler.Committee, nodeID signature.Pu
 }
 
 func schedulerCheckPrimaryScheduler(committee *scheduler.Committee, nodeID signature.PublicKey, round uint64) bool {
-	rank, err := committee.SchedulerRank(round, nodeID)
-	if err != nil {
-		panic(err)
-	}
-	return rank == 0
+	rank, ok := committee.SchedulerRank(round, nodeID)
+	return ok && rank == 0
 }

go/roothash/api/commitment/pool.go

@@ -241,8 +241,8 @@ func (p *Pool) AddVerifiedExecutorCommitment(c *scheduler.Committee, ec *Executo
 	}
 
 	// Ensure that the scheduler is allowed to schedule transactions.
-	rank, err := c.SchedulerRank(ec.Header.Header.Round, ec.Header.SchedulerID)
-	if err != nil {
+	rank, ok := c.SchedulerRank(ec.Header.Header.Round, ec.Header.SchedulerID)
+	if !ok {
 		// Reject commitments with invalid schedulers.
 		logger.Debug("executor commitment's scheduler is not in the committee",
 			"round", ec.Header.Header.Round,

go/roothash/tests/tester.go

@@ -329,8 +329,8 @@ func (s *runtimeState) generateExecutorCommitments(t *testing.T, consensus conse
 	inMsgsHash.Empty()
 
 	// Gather executor nodes, starting with the scheduler.
-	schedulerIdx, err := executorCommittee.committee.SchedulerIdx(parent.Header.Round, rank)
-	require.NoError(err, "SchedulerIdx")
+	schedulerIdx, ok := executorCommittee.committee.SchedulerIdx(parent.Header.Round, rank)
+	require.True(ok, "SchedulerIdx")
 	schedulerID := executorCommittee.workers[schedulerIdx].Signer.Public()
 
 	executorNodes := make([]*registryTests.TestNode, 0, len(executorCommittee.workers))
@@ -523,8 +523,8 @@ func (s *runtimeState) testSuccessfulRound(t *testing.T, backend api.Backend, co
 		liveRounds[i] = 1
 	}
 
-	schedulerIdx, err := s.executorCommittee.committee.SchedulerIdx(parent.Block.Header.Round, 0)
-	require.NoError(err, "SchedulerIdx")
+	schedulerIdx, ok := s.executorCommittee.committee.SchedulerIdx(parent.Block.Header.Round, 0)
+	require.True(ok, "SchedulerIdx")
 	finalizedProposals[schedulerIdx]++
 
 	require.Equal(uint64(1), livenessStatistics.TotalRounds, "there should be one finalized round")
@@ -565,8 +565,8 @@ func (s *runtimeState) testRoundTimeout(t *testing.T, backend api.Backend, conse
 		missedProposals := make([]uint64, len(livenessStatistics.MissedProposals))
 
 		var schedulerIdx int
-		schedulerIdx, err = s.executorCommittee.committee.SchedulerIdx(blk.Block.Header.Round, 0)
-		require.NoError(err, "SchedulerIdx")
+		schedulerIdx, ok := s.executorCommittee.committee.SchedulerIdx(blk.Block.Header.Round, 0)
+		require.True(ok, "SchedulerIdx")
 		missedProposals[schedulerIdx]++
 
 		require.Zero(livenessStatistics.TotalRounds, "there should be no finalized rounds")

go/scheduler/api/api.go

@@ -194,20 +194,20 @@ func (c *Committee) IsBackupWorker(id signature.PublicKey) bool {
 // Scheduler returns the scheduler with the given rank in the committee's scheduling order
 // for the given round.
 //
-// If no scheduler with the given rank is found, it returns an error.
-func (c *Committee) Scheduler(round uint64, rank uint64) (*CommitteeNode, error) {
-	idx, err := c.SchedulerIdx(round, rank)
-	if err != nil {
-		return nil, err
+// If no scheduler with the given rank is found, it returns false.
+func (c *Committee) Scheduler(round uint64, rank uint64) (*CommitteeNode, bool) {
+	idx, ok := c.SchedulerIdx(round, rank)
+	if !ok {
+		return nil, false
 	}
-	return c.Members[idx], nil
+	return c.Members[idx], true
 }
 
 // SchedulerIdx returns the index of the scheduler with the given rank in the committee's
 // scheduling order for the given round.
 //
-// If no scheduler with the given rank is found, it returns an error.
-func (c *Committee) SchedulerIdx(round uint64, rank uint64) (int, error) {
+// If no scheduler with the given rank is found, it returns false.
+func (c *Committee) SchedulerIdx(round uint64, rank uint64) (int, bool) {
 	var total uint64
 
 	for _, n := range c.Members {
@@ -219,20 +219,20 @@ func (c *Committee) SchedulerIdx(round uint64, rank uint64) (int, error) {
 	}
 
 	if rank >= total {
-		return 0, fmt.Errorf("no worker with the given rank in the committee")
+		return 0, false
 	}
 
 	idx := (rank + total - round%total) % total
 
-	return int(idx), nil
+	return int(idx), true
 }
 
 // SchedulerRank returns the position (index) of a node with the given public key in the committee's
 // scheduling order for the given round. A lower rank indicates higher scheduling priority.
 //
 // If the node is not a worker in the committee and, therefore, not allowed to schedule transactions
-// for the given round, it returns an error.
-func (c *Committee) SchedulerRank(round uint64, id signature.PublicKey) (uint64, error) {
+// for the given round, it returns false.
+func (c *Committee) SchedulerRank(round uint64, id signature.PublicKey) (uint64, bool) {
 	var (
 		total    uint64
 		idx      uint64
@@ -252,12 +252,12 @@ func (c *Committee) SchedulerRank(round uint64, id signature.PublicKey) (uint64,
 	}
 
 	if !isWorker {
-		return 0, fmt.Errorf("node is not a worker in the committee")
+		return 0, false
 	}
 
 	rank := (round + idx) % total
 
-	return rank, nil
+	return rank, true
 }
 
 // String returns a string representation of a Committee.

go/worker/common/committee/node.go

@@ -313,7 +313,7 @@ func (n *Node) GetStatus() (*api.Status, error) {
 		status.ExecutorRoles = cmte.Roles
 
 		// Include scheduler rank.
-		if rank, err := cmte.Committee.SchedulerRank(status.LatestRound+1, epoch.identity.NodeSigner.Public()); err != nil {
+		if rank, ok := cmte.Committee.SchedulerRank(status.LatestRound+1, epoch.identity.NodeSigner.Public()); ok {
 			status.SchedulerRank = rank
 		}
 

go/worker/compute/executor/committee/node.go

@@ -1453,9 +1453,9 @@ func (n *Node) roundWorker(ctx context.Context, bi *runtime.BlockInfo) {
 	// Compute node's rank when scheduling transactions.
 	id := n.commonNode.Identity.NodeSigner.Public()
 	n.committee = n.epoch.GetExecutorCommittee().Committee
-	n.rank, err = n.committee.SchedulerRank(round, id)
-	if err != nil {
-		n.rank = math.MaxUint64
+	n.rank = math.MaxUint64
+	if rank, ok := n.committee.SchedulerRank(round, id); ok {
+		n.rank = rank
 	}
 
 	n.logger.Debug("node is an executor member",
@@ -1546,8 +1546,8 @@ func (n *Node) roundWorker(ctx context.Context, bi *runtime.BlockInfo) {
 				if ec.Header.Header.Round != round {
 					continue
 				}
-				rank, err := n.committee.SchedulerRank(round, ec.Header.SchedulerID)
-				if err != nil {
+				rank, ok := n.committee.SchedulerRank(round, ec.Header.SchedulerID)
+				if !ok {
 					continue
 				}
 				poolRank = rank

go/worker/compute/executor/committee/p2p.go

@@ -74,8 +74,8 @@ func (h *committeeMsgHandler) HandleMessage(_ context.Context, _ signature.Publi
 		// Before opening the signed dispatch message, verify that it was actually signed by one
 		// of the transaction schedulers.
 		committee := epoch.GetExecutorCommittee().Committee
-		rank, err := committee.SchedulerRank(proposal.Header.Round, proposal.NodeID)
-		if err != nil {
+		rank, ok := committee.SchedulerRank(proposal.Header.Round, proposal.NodeID)
+		if !ok {
 			// Invalid scheduler, do not forward.
 			return p2pError.Permanent(errMsgFromNonTxnSched)
 		}