Merge branch 'master' into feature/stateproofs

algorand · Apr 17, 2022 · c34780a · c34780a
2 parents 1c3f705 + 8a5155f
commit c34780a
Show file tree

Hide file tree

Showing 96 changed files with 6,033 additions and 2,612 deletions.
diff --git a/.circleci/config.yml b/.circleci/config.yml
@@ -138,7 +138,9 @@ workflows:
             - << matrix.platform >>_build
           filters:
             <<: *filters-nightly
-          context: slack-secrets
+          context:
+            - slack-secrets
+            - aws-secrets
 
       - tests_verification_job:
           name: << matrix.platform >>_<< matrix.job_type >>_verification
@@ -196,7 +198,7 @@ commands:
           shell: bash.exe
           command: |
             choco install -y msys2 pacman make wget --force
-            choco install -y golang --version=1.16.11 --force
+            choco install -y golang --version=1.16.15 --force
             choco install -y python3 --version=3.7.3 --force
             export msys2='cmd //C RefreshEnv.cmd '
             export msys2+='& set MSYS=winsymlinks:nativestrict '
@@ -335,9 +337,6 @@ commands:
           key: 'go-cache-v2-{{ .Environment.CIRCLE_STAGE }}-{{ .Environment.CIRCLE_BUILD_NUM }}'
           paths:
             - tmp/go-cache
-      - upload_to_buildpulse:
-          platform: << parameters.platform >>
-          path: << parameters.result_path >>/<< parameters.result_subdir>>
 
   upload_coverage:
     description: Collect coverage reports and upload them
@@ -348,40 +347,6 @@ commands:
           command: |
             scripts/travis/upload_coverage.sh || true
 
-  upload_to_buildpulse:
-    description: Collect build reports and upload them
-    parameters:
-      platform:
-        type: string
-      path:
-        type: string
-    steps:
-      - run:
-          name: Send test results to BuildPulse
-          when: always
-          command: |
-            set -e
-            if ! ls << parameters.path >>/*/*.xml &> /dev/null; then exit 0; fi
-            sed -i"" -e 's/classname="/classname="<< parameters.platform >>-/' << parameters.path >>/*/*.xml
-            case "<< parameters.platform >>" in
-              arm64)
-                URL=https://github.com/buildpulse/test-reporter/releases/download/v0.21.0-pre/test-reporter-linux-arm64
-                SUM=53f94c29ad162c2b9ebb1f4a2f967f5262c0459ee4a0c34332977d8c89aafc18
-                ;;
-              amd64)
-                URL=https://github.com/buildpulse/test-reporter/releases/download/v0.21.0-pre/test-reporter-linux-amd64
-                SUM=4655e54d756580c0de0112cab488e6e08d0af75e9fc8caea2d63f9e13be8beb5
-                ;;
-              mac_amd64)
-                URL=https://github.com/buildpulse/test-reporter/releases/download/v0.21.0-pre/test-reporter-darwin-amd64
-                SUM=2f9e20a6f683c80f35d04e36bc57ecfe605bb48fee5a1b8d8f7c45094028eea3
-                ;;
-            esac
-            curl -fsSL --retry 3 --retry-connrefused $URL > ./buildpulse-test-reporter
-            echo "$SUM *buildpulse-test-reporter" | shasum -a 256 -c --status
-            chmod +x ./buildpulse-test-reporter
-            ./buildpulse-test-reporter submit << parameters.path >> --account-id 23182699 --repository-id 191266671 || true
-
   generic_integration:
     description: Run integration tests from build workspace, for re-use by diferent architectures
     parameters:
@@ -432,6 +397,7 @@ commands:
             export PARTITION_ID=${CIRCLE_NODE_INDEX}
             export PARALLEL_FLAG="-p 1"
             test/scripts/run_integration_tests.sh
+
       - store_artifacts:
           path: << parameters.result_path >>
           destination: test-results
@@ -441,9 +407,6 @@ commands:
           root: << parameters.result_path >>
           paths:
             - << parameters.result_subdir >>
-      - upload_to_buildpulse:
-          platform: << parameters.platform >>
-          path: << parameters.result_path >>/<< parameters.result_subdir>>
 
   tests_verification_command:
     description: Check if all tests were run at least once and only once across all parallel runs
@@ -651,6 +614,7 @@ jobs:
     working_directory: << pipeline.parameters.build_dir >>/project
     environment:
       E2E_TEST_FILTER: "SCRIPTS"
+      E2E_PLATFORM: << parameters.platform >>
     steps:
       - prepare_build_dir
       - prepare_go

diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
@@ -20,7 +20,7 @@ jobs:
       - name: Install golang
         uses: actions/setup-go@v2
         with:
-          go-version: '1.16.11'
+          go-version: '1.16.15'
       - name: Build Test
         run: |
           export ALGORAND_DEADLOCK=enable

diff --git a/.github/workflows/reviewdog.yml b/.github/workflows/reviewdog.yml
@@ -44,7 +44,7 @@ jobs:
       - name: Install specific golang
         uses: actions/setup-go@v2
         with:
-          go-version: '1.16.11'
+          go-version: '1.16.15'
       - name: Create folders for golangci-lint
         run: mkdir -p cicdtmp/golangci-lint
       - name: Check if custom golangci-lint is already built

diff --git a/agreement/cryptoVerifier.go b/agreement/cryptoVerifier.go
@@ -22,8 +22,12 @@ import (
 
 	"github.com/algorand/go-algorand/logging"
 	"github.com/algorand/go-algorand/protocol"
+	"github.com/algorand/go-algorand/util/metrics"
 )
 
+var voteVerifierOutFullCounter = metrics.MakeCounter(
+	metrics.MetricName{Name: "algod_agreement_vote_verifier_responses_dropped", Description: "Number of voteVerifier responses dropped due to full channel"})
+
 // TODO put these in config
 const (
 	voteParallelism     = 16
@@ -210,6 +214,7 @@ func (c *poolCryptoVerifier) voteFillWorker(toBundleWait chan<- bundleFuture) {
 				select {
 				case c.votes.out <- asyncVerifyVoteResponse{index: votereq.TaskIndex, err: err, cancelled: true}:
 				default:
+					voteVerifierOutFullCounter.Inc(nil)
 					c.log.Infof("poolCryptoVerifier.voteFillWorker unable to write failed enqueue response to output channel")
 				}
 			}

diff --git a/agreement/pseudonode.go b/agreement/pseudonode.go
@@ -29,6 +29,7 @@ import (
 	"github.com/algorand/go-algorand/logging/logspec"
 	"github.com/algorand/go-algorand/logging/telemetryspec"
 	"github.com/algorand/go-algorand/protocol"
+	"github.com/algorand/go-algorand/util/metrics"
 )
 
 // TODO put these in config
@@ -43,6 +44,9 @@ var errPseudonodeVerifierClosedChannel = errors.New("crypto verifier closed the
 var errPseudonodeNoVotes = errors.New("no valid participation keys to generate votes for given round")
 var errPseudonodeNoProposals = errors.New("no valid participation keys to generate proposals for given round")
 
+var pseudonodeBacklogFullByType = metrics.NewTagCounter("algod_agreement_pseudonode_tasks_dropped_{TAG}", "Number of pseudonode tasks dropped per type")
+var pseudonodeResultTimeoutsByType = metrics.NewTagCounter("algod_agreement_pseudonode_tasks_timeouts_{TAG}", "Number of pseudonode task result timeouts per type")
+
 // A pseudonode creates proposals and votes with a KeyManager which holds participation keys.
 //
 // It constructs these messages as if they arrived from an external source and were verified.
@@ -176,6 +180,7 @@ func (n asyncPseudonode) MakeProposals(ctx context.Context, r round, p period) (
 		return proposalTask.outputChannel(), nil
 	default:
 		proposalTask.close()
+		pseudonodeBacklogFullByType.Add("proposal", 1)
 		return nil, fmt.Errorf("unable to make proposal for (%d, %d): %w", r, p, errPseudonodeBacklogFull)
 	}
 }
@@ -193,6 +198,7 @@ func (n asyncPseudonode) MakeVotes(ctx context.Context, r round, p period, s ste
 		return proposalTask.outputChannel(), nil
 	default:
 		proposalTask.close()
+		pseudonodeBacklogFullByType.Add("vote", 1)
 		return nil, fmt.Errorf("unable to make vote for (%d, %d, %d): %w", r, p, s, errPseudonodeBacklogFull)
 	}
 }
@@ -474,6 +480,7 @@ verifiedVotesLoop:
 				return
 			case <-outputTimeout:
 				// we've been waiting for too long for this vote to be written to the output.
+				pseudonodeResultTimeoutsByType.Add("vote", 1)
 				t.node.log.Warnf("pseudonode.makeVotes: unable to write vote to output channel for round %d, period %d", t.round, t.period)
 				outputTimeout = nil
 			}
@@ -577,6 +584,7 @@ verifiedVotesLoop:
 				return
 			case <-outputTimeout:
 				// we've been waiting for too long for this vote to be written to the output.
+				pseudonodeResultTimeoutsByType.Add("pvote", 1)
 				t.node.log.Warnf("pseudonode.makeProposals: unable to write proposal vote to output channel for round %d, period %d", t.round, t.period)
 				outputTimeout = nil
 			}
@@ -597,6 +605,7 @@ verifiedPayloadsLoop:
 				return
 			case <-outputTimeout:
 				// we've been waiting for too long for this vote to be written to the output.
+				pseudonodeResultTimeoutsByType.Add("ppayload", 1)
 				t.node.log.Warnf("pseudonode.makeProposals: unable to write proposal payload to output channel for round %d, period %d", t.round, t.period)
 				outputTimeout = nil
 			}

diff --git a/cmd/algod/main.go b/cmd/algod/main.go
@@ -84,12 +84,12 @@ func run() int {
 	}
 
 	version := config.GetCurrentVersion()
-	heartbeatGauge := metrics.MakeStringGauge()
-	heartbeatGauge.Set("version", version.String())
-	heartbeatGauge.Set("version-num", strconv.FormatUint(version.AsUInt64(), 10))
-	heartbeatGauge.Set("channel", version.Channel)
-	heartbeatGauge.Set("branch", version.Branch)
-	heartbeatGauge.Set("commit-hash", version.GetCommitHash())
+	var baseHeartbeatEvent telemetryspec.HeartbeatEventDetails
+	baseHeartbeatEvent.Info.Version = version.String()
+	baseHeartbeatEvent.Info.VersionNum = strconv.FormatUint(version.AsUInt64(), 10)
+	baseHeartbeatEvent.Info.Channel = version.Channel
+	baseHeartbeatEvent.Info.Branch = version.Branch
+	baseHeartbeatEvent.Info.CommitHash = version.GetCommitHash()
 
 	if *branchCheck {
 		fmt.Println(config.Branch)
@@ -339,12 +339,11 @@ func run() int {
 			defer ticker.Stop()
 
 			sendHeartbeat := func() {
-				values := make(map[string]string)
+				values := make(map[string]float64)
 				metrics.DefaultRegistry().AddMetrics(values)
 
-				heartbeatDetails := telemetryspec.HeartbeatEventDetails{
-					Metrics: values,
-				}
+				heartbeatDetails := baseHeartbeatEvent
+				heartbeatDetails.Metrics = values
 
 				log.EventWithDetails(telemetryspec.ApplicationState, telemetryspec.HeartbeatEvent, heartbeatDetails)
 			}