replace existing replay verify with new one

areshand · areshand · commit c9b71bc156e4 · 2024-12-13T09:16:49.000-08:00
diff --git a/.github/workflows/replay-verify-legacy.yaml b/.github/workflows/replay-verify-legacy.yaml
@@ -0,0 +1,115 @@
+# This defines a workflow to replay transactions on the given chain with the latest aptos node software.
+# In order to trigger it go to the Actions Tab of the Repo, click "replay-verify" and then "Run Workflow".
+#
+# On PR, a single test case will run. On workflow_dispatch, you may specify the CHAIN_NAME to verify.
+
+name: "replay-verify"
+on:
+  # Allow triggering manually
+  workflow_dispatch:
+    inputs:
+      GIT_SHA:
+        required: false
+        type: string
+        description: The git SHA1 to test. If not specified, it will use the latest commit on main.
+      CHAIN_NAME:
+        required: false
+        type: choice
+        options: [testnet, mainnet, all]
+        default: all
+        description: The chain name to test. If not specified, it will test both testnet and mainnet.
+      TESTNET_BUCKET:
+        required: false
+        type: string
+        description: The bucket to use for testnet replay. If not specified, it will use aptos-testnet-backup.
+        default: aptos-testnet-backup
+      MAINNET_BUCKET:
+        required: false
+        type: string
+        description: The bucket to use for mainnet replay. If not specified, it will use aptos-mainnet-backup.
+        default: aptos-mainnet-backup
+
+# cancel redundant builds
+concurrency:
+  # cancel redundant builds on PRs (only on PR, not on branches)
+  group: ${{ github.workflow }}-${{ (github.event_name == 'pull_request' && github.ref) || github.sha }}
+  cancel-in-progress: true
+
+jobs:
+  determine-test-metadata:
+    runs-on: ubuntu-latest
+    steps:
+      # checkout the repo first, so check-aptos-core can use it and cancel the workflow if necessary
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/check-aptos-core
+        with:
+          cancel-workflow: ${{ github.event_name == 'schedule' }} # Cancel the workflow if it is scheduled on a fork
+
+  replay-testnet:
+    if: |
+      github.event_name == 'schedule' ||
+      github.event_name == 'push' ||
+      github.event_name == 'workflow_dispatch' && (inputs.CHAIN_NAME == 'testnet' || inputs.CHAIN_NAME == 'all')
+    needs: determine-test-metadata
+    uses: ./.github/workflows/workflow-run-replay-verify.yaml
+    secrets: inherit
+    with:
+      GIT_SHA: ${{ inputs.GIT_SHA }}
+      # replay-verify config
+      BUCKET: ${{ inputs.TESTNET_BUCKET || 'aptos-testnet-backup' }}
+      SUB_DIR: e1
+      HISTORY_START: 862000000
+      # to see historical TXNS_TO_SKIP, check out ce6158ac2764ee9d4c8738a85f3bcdc6bd0cadc1
+      TXNS_TO_SKIP: "0"
+      # 1195000000-122000000: https://github.com/aptos-labs/aptos-core/pull/13832
+      RANGES_TO_SKIP: "1195000000-1220000000"
+      BACKUP_CONFIG_TEMPLATE_PATH: terraform/helm/fullnode/files/backup/gcs.yaml
+      # workflow config
+      RUNS_ON: "high-perf-docker-with-local-ssd"
+      TIMEOUT_MINUTES: 180
+      MAX_VERSIONS_PER_RANGE: 2000000
+
+  replay-mainnet:
+    if: |
+      github.event_name == 'schedule' ||
+      github.event_name == 'push' ||
+      github.event_name == 'workflow_dispatch' && (inputs.CHAIN_NAME == 'mainnet' || inputs.CHAIN_NAME == 'all' )
+    needs: determine-test-metadata
+    uses: ./.github/workflows/workflow-run-replay-verify.yaml
+    secrets: inherit
+    with:
+      GIT_SHA: ${{ inputs.GIT_SHA }}
+      # replay-verify config
+      BUCKET: ${{ inputs.MAINNET_BUCKET || 'aptos-mainnet-backup' }}
+      SUB_DIR: e1
+      HISTORY_START: 518000000
+      #TXNS_TO_SKIP: 12253479 12277499 148358668
+      TXNS_TO_SKIP: "0"
+      # 1197378568-1198492648: https://github.com/aptos-labs/aptos-core/pull/13832
+      RANGES_TO_SKIP: "1197378568-1198492648"
+      BACKUP_CONFIG_TEMPLATE_PATH: terraform/helm/fullnode/files/backup/gcs.yaml
+      # workflow config
+      RUNS_ON: "high-perf-docker-with-local-ssd"
+      TIMEOUT_MINUTES: 180
+      MAX_VERSIONS_PER_RANGE: 800000
+
+  test-replay:
+    if: ${{ (github.event_name == 'pull_request') && contains(github.event.pull_request.labels.*.name, 'CICD:test-replay')}}
+    needs: determine-test-metadata
+    uses: ./.github/workflows/workflow-run-replay-verify.yaml
+    secrets: inherit
+    with:
+      GIT_SHA: ${{ github.event.pull_request.head.sha }}
+      # replay-verify config
+      BUCKET: ${{ inputs.TESTNET_BUCKET || 'aptos-testnet-backup' }}
+      SUB_DIR: e1
+      HISTORY_START: 862000000
+      # to see historical TXNS_TO_SKIP, check out ce6158ac2764ee9d4c8738a85f3bcdc6bd0cadc1
+      TXNS_TO_SKIP: "0"
+      # 1195000000-1220000000: https://github.com/aptos-labs/aptos-core/pull/13832
+      RANGES_TO_SKIP: "1195000000-1220000000"
+      BACKUP_CONFIG_TEMPLATE_PATH: terraform/helm/fullnode/files/backup/gcs.yaml
+      # workflow config
+      RUNS_ON: "high-perf-docker-with-local-ssd"
+      TIMEOUT_MINUTES: 120 # increase test replay timeout to capture more flaky errors
+      MAX_VERSIONS_PER_RANGE: 2000000
diff --git a/.github/workflows/replay-verify-on-archive.yaml b/.github/workflows/replay-verify-on-archive.yaml
diff --git a/.github/workflows/replay-verify.yaml b/.github/workflows/replay-verify.yaml
@@ -3,37 +3,41 @@
 #
 # On PR, a single test case will run. On workflow_dispatch, you may specify the CHAIN_NAME to verify.
 
-name: "replay-verify"
+name: "replay-verify-on-archive"
 on:
   # Allow triggering manually
   workflow_dispatch:
     inputs:
-      GIT_SHA:
-        required: false
-        type: string
-        description: The git SHA1 to test. If not specified, it will use the latest commit on main.
-      CHAIN_NAME:
-        required: false
+      NETWORK:
+        required: true
         type: choice
         options: [testnet, mainnet, all]
         default: all
         description: The chain name to test. If not specified, it will test both testnet and mainnet.
-      TESTNET_BUCKET:
+      IMAGE_TAG:
+        required: false
+        type: string
+        description: The image tag of the feature branch to test, if not specified, it will use the latest commit on current branch.
+      START_VERSION:
         required: false
         type: string
-        description: The bucket to use for testnet replay. If not specified, it will use aptos-testnet-backup.
-        default: aptos-testnet-backup
-      MAINNET_BUCKET:
+        description: Optional version to start replaying. If not specified, replay-verify will determines start version itself.
+      END_VERSION:
         required: false
         type: string
-        description: The bucket to use for mainnet replay. If not specified, it will use aptos-mainnet-backup.
-        default: aptos-mainnet-backup
+        description: Optional version to end replaying. If not specified, replay-verify will determines end version itself.
   pull_request:
     paths:
       - ".github/workflows/replay-verify.yaml"
-      - ".github/workflows/workflow-run-replay-verify.yaml"
+      - ".github/workflows/workflow-run-replay-verify-on-archive.yaml"
   schedule:
-    - cron: "0 22 * * 0,2,4" # The main branch cadence. This runs every Sun,Tues,Thurs
+    - cron: "0 8 * * 0,2,4" # The main branch cadence. This runs every Sun,Tues,Thurs UTC 08:00
+
+permissions:
+  contents: read
+  id-token: write #required for GCP Workload Identity federation which we use to login into Google Artifact Registry
+  issues: read
+  pull-requests: read
 
 # cancel redundant builds
 concurrency:
@@ -43,7 +47,7 @@ concurrency:
 
 jobs:
   determine-test-metadata:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-latest-32-core
     steps:
       # checkout the repo first, so check-aptos-core can use it and cancel the workflow if necessary
       - uses: actions/checkout@v4
@@ -54,68 +58,25 @@ jobs:
   replay-testnet:
     if: |
       github.event_name == 'schedule' ||
-      github.event_name == 'push' ||
-      github.event_name == 'workflow_dispatch' && (inputs.CHAIN_NAME == 'testnet' || inputs.CHAIN_NAME == 'all')
+      github.event_name == 'workflow_dispatch' && (inputs.NETWORK == 'testnet' || inputs.NETWORK == 'all')
     needs: determine-test-metadata
-    uses: ./.github/workflows/workflow-run-replay-verify.yaml
+    uses: ./.github/workflows/workflow-run-replay-verify-on-archive.yaml
     secrets: inherit
     with:
-      GIT_SHA: ${{ inputs.GIT_SHA }}
-      # replay-verify config
-      BUCKET: ${{ inputs.TESTNET_BUCKET || 'aptos-testnet-backup' }}
-      SUB_DIR: e1
-      HISTORY_START: 862000000
-      # to see historical TXNS_TO_SKIP, check out ce6158ac2764ee9d4c8738a85f3bcdc6bd0cadc1
-      TXNS_TO_SKIP: "0"
-      # 1195000000-122000000: https://github.com/aptos-labs/aptos-core/pull/13832
-      RANGES_TO_SKIP: "1195000000-1220000000"
-      BACKUP_CONFIG_TEMPLATE_PATH: terraform/helm/fullnode/files/backup/gcs.yaml
-      # workflow config
-      RUNS_ON: "high-perf-docker-with-local-ssd"
-      TIMEOUT_MINUTES: 180
-      MAX_VERSIONS_PER_RANGE: 2000000
+      NETWORK: "testnet"
+      IMAGE_TAG: ${{ inputs.IMAGE_TAG }}
+      START_VERSION: ${{ inputs.START_VERSION }}
+      END_VERSION: ${{ inputs.END_VERSION }}
 
   replay-mainnet:
     if: |
       github.event_name == 'schedule' ||
-      github.event_name == 'push' ||
-      github.event_name == 'workflow_dispatch' && (inputs.CHAIN_NAME == 'mainnet' || inputs.CHAIN_NAME == 'all' )
-    needs: determine-test-metadata
-    uses: ./.github/workflows/workflow-run-replay-verify.yaml
-    secrets: inherit
-    with:
-      GIT_SHA: ${{ inputs.GIT_SHA }}
-      # replay-verify config
-      BUCKET: ${{ inputs.MAINNET_BUCKET || 'aptos-mainnet-backup' }}
-      SUB_DIR: e1
-      HISTORY_START: 518000000
-      #TXNS_TO_SKIP: 12253479 12277499 148358668
-      TXNS_TO_SKIP: "0"
-      # 1197378568-1198492648: https://github.com/aptos-labs/aptos-core/pull/13832
-      RANGES_TO_SKIP: "1197378568-1198492648"
-      BACKUP_CONFIG_TEMPLATE_PATH: terraform/helm/fullnode/files/backup/gcs.yaml
-      # workflow config
-      RUNS_ON: "high-perf-docker-with-local-ssd"
-      TIMEOUT_MINUTES: 180
-      MAX_VERSIONS_PER_RANGE: 800000
-
-  test-replay:
-    if: ${{ (github.event_name == 'pull_request') && contains(github.event.pull_request.labels.*.name, 'CICD:test-replay')}}
+      github.event_name == 'workflow_dispatch' && (inputs.NETWORK == 'mainnet' || inputs.NETWORK == 'all' )
     needs: determine-test-metadata
-    uses: ./.github/workflows/workflow-run-replay-verify.yaml
+    uses: ./.github/workflows/workflow-run-replay-verify-on-archive.yaml
     secrets: inherit
     with:
-      GIT_SHA: ${{ github.event.pull_request.head.sha }}
-      # replay-verify config
-      BUCKET: ${{ inputs.TESTNET_BUCKET || 'aptos-testnet-backup' }}
-      SUB_DIR: e1
-      HISTORY_START: 862000000
-      # to see historical TXNS_TO_SKIP, check out ce6158ac2764ee9d4c8738a85f3bcdc6bd0cadc1
-      TXNS_TO_SKIP: "0"
-      # 1195000000-1220000000: https://github.com/aptos-labs/aptos-core/pull/13832
-      RANGES_TO_SKIP: "1195000000-1220000000"
-      BACKUP_CONFIG_TEMPLATE_PATH: terraform/helm/fullnode/files/backup/gcs.yaml
-      # workflow config
-      RUNS_ON: "high-perf-docker-with-local-ssd"
-      TIMEOUT_MINUTES: 120 # increase test replay timeout to capture more flaky errors
-      MAX_VERSIONS_PER_RANGE: 2000000
+      NETWORK: "mainnet"
+      IMAGE_TAG: ${{ inputs.IMAGE_TAG }}
+      START_VERSION: ${{ inputs.START_VERSION }}
+      END_VERSION: ${{ inputs.END_VERSION }}
