Source connector performance harness (airbytehq#23786)

* remove arguments * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * test * Source Performance harness (airbytehq#23922) * test * initial commit * Working dev * src-platform-dst * Integrate secrets * Update .dockerignore * Update build.gradle * Update base.sh * Build harness module and use it to measure * Build harness module and use it to measure * Build harness module and use it to measure * Automated Change * Integrate secrets * Integrate secrets * Integrate secrets * Integrate secrets * test * test * test * test * test * test * test * test * test * test * test * test * test * test * Automated Change * test * Automated Change * better error handling * temp * Automated Change * use connector image name * use connector image name * use connector image name * use connector image name * use connector image name * use connector image name * use connector image name * use connector image name * add test name argument * add test name argument * add test name argument * add test name argument * add test name argument * add test name argument * Automated Change * add bottleneck_stream1 dataset * Automated Change * fix dependency * fix dependency * test downgrade of commons to 41.1 * Automated Change * test downgrade of commons to 41.0 * test * test * test * test * test * exit condition * exit condition * exit condition * test * test * test * test * test * test * sanity * readme * change command name * update gradle settings * sanity * architecture aware docker build * sanity * put run harness step in a script * sanity * sanity --------- Co-authored-by: rodireich <rodireich@users.noreply.github.com>
samos123 · Mar 30, 2023 · 0a38d4e · 0a38d4e
1 parent 836e4f4
commit 0a38d4e
Show file tree

Hide file tree

Showing 18 changed files with 1,185 additions and 8 deletions.
diff --git a/.github/workflows/connector-performance-command.yml b/.github/workflows/connector-performance-command.yml
@@ -19,15 +19,171 @@ on:
       uuid:
         description: "Custom UUID of workflow run. Used because GitHub dispatches endpoint does not return workflow run id."
         required: false
-      connector-acceptance-test-version:
-        description: "Set a specific connector acceptance test version to use. Enter 'dev' to test, build and use a local version of Connector Acceptance Test."
-        required: false
-        default: "latest"
-      local_cdk:
-        description: "Run Connector Acceptance Tests against the CDK version on the current branch."
+      dataset:
+        description: "Name of dataset to use for performance measurement. Currently supports 1m, 10m, 20m."
         required: false
+        default: "1m"
 jobs:
-  Stub:
+  uuid:
+    name: "Custom UUID of workflow run"
+    timeout-minutes: 10
+    runs-on: ubuntu-latest
+    steps:
+      - name: UUID ${{ github.event.inputs.uuid }}
+        run: true
+  start-test-runner:
+    name: Start Build EC2 Runner
+    needs: uuid
+    timeout-minutes: 10
+    runs-on: ubuntu-latest
+    outputs:
+      label: ${{ steps.start-ec2-runner.outputs.label }}
+      ec2-instance-id: ${{ steps.start-ec2-runner.outputs.ec2-instance-id }}
+    steps:
+      - name: Checkout Airbyte
+        uses: actions/checkout@v3
+        with:
+          repository: ${{ github.event.inputs.repo }}
+          ref: ${{ github.event.inputs.gitref }}
+      - name: Check PAT rate limits
+        run: |
+          ./tools/bin/find_non_rate_limited_PAT \
+            ${{ secrets.GH_PAT_BUILD_RUNNER_OSS }} \
+            ${{ secrets.GH_PAT_BUILD_RUNNER_BACKUP }}
+      - name: Start AWS Runner
+        id: start-ec2-runner
+        uses: ./.github/actions/start-aws-runner
+        with:
+          aws-access-key-id: ${{ secrets.SELF_RUNNER_AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ secrets.SELF_RUNNER_AWS_SECRET_ACCESS_KEY }}
+          github-token: ${{ env.PAT }}
+  performance-test:
+    timeout-minutes: 240
+    needs: start-test-runner
+    runs-on: ${{ needs.start-test-runner.outputs.label }}
+    steps:
+      - name: Link comment to workflow run
+        if: github.event.inputs.comment-id
+        uses: peter-evans/create-or-update-comment@v1
+        with:
+          comment-id: ${{ github.event.inputs.comment-id }}
+          body: |
+            #### Note: The following `dataset=` values are supported: `1m`<sub>(default)</sub>, `10m`, `20m`, `bottleneck_stream1`
+            > :runner: ${{github.event.inputs.connector}} https://github.com/${{github.repository}}/actions/runs/${{github.run_id}}
+      - name: Search for valid connector name format
+        id: regex
+        uses: AsasInnab/regex-action@v1
+        with:
+          regex_pattern: "^((connectors|bases)/)?[a-zA-Z0-9-_]+$"
+          regex_flags: "i" # required to be set for this plugin
+          search_string: ${{ github.event.inputs.connector }}
+      - name: Validate input workflow format
+        if: steps.regex.outputs.first_match != github.event.inputs.connector
+        run: echo "The connector provided has an invalid format!" && exit 1
+      - name: Filter supported connectors
+        if: "${{ github.event.inputs.connector != 'connectors/source-postgres' }}"
+        run: echo "Only connectors/source-postgres currently supported by harness" && exit 1
+      - name: Checkout Airbyte
+        uses: actions/checkout@v3
+        with:
+          repository: ${{ github.event.inputs.repo }}
+          ref: ${{ github.event.inputs.gitref }}
+      - name: Install Java
+        uses: actions/setup-java@v3
+        with:
+          distribution: "zulu"
+          java-version: "17"
+      - name: Install Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: "3.9"
+      - name: Install CI scripts
+        # all CI python packages have the prefix "ci_"
+        run: |
+          pip install --quiet -e ./tools/ci_*
+      - name: Write source-harness credentials
+        run: |
+          ci_credentials connectors-performance/source-harness write-to-storage
+        env:
+          GCP_GSM_CREDENTIALS: ${{ secrets.GCP_GSM_CREDENTIALS }}
+      - name: build harness
+        shell: bash
+        run: |
+          echo "Building... source-harness" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY
+          ./gradlew :airbyte-integrations:connectors-performance:source-harness:build -x check
+      - name: build connector
+        shell: bash
+        run: |
+          echo "Building... ${{github.event.inputs.connector}}" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY # this is a blank line
+          echo "Running ./gradlew :airbyte-integrations:connectors:source-postgres:build -x check" 
+          ./gradlew :airbyte-integrations:connectors:source-postgres:build -x check
+      - name: KIND Kubernetes Cluster Setup
+        uses: helm/kind-action@v1.4.0
+        with:
+          config: ./tools/bin/source-harness-kind-cluster-config.yaml
+      - name: Run harness
+        id: run-harness
+        shell: bash
+        env:
+          CONN: ${{ github.event.inputs.connector }}
+          DS: ${{ github.event.inputs.dataset }}
+          prefix: '{"type":"LOG","log":{"level":"INFO","message":"INFO i.a.i.p.PerformanceTest(runTest):165'
+          suffix: '"}}'
+        run: |
+          kubectl apply -f ./tools/bin/admin-service-account.yaml
+          kind load docker-image airbyte/source-postgres:dev --name chart-testing
+          kind load docker-image airbyte/source-harness:dev --name chart-testing
+          export CONNECTOR_IMAGE_NAME=${CONN/connectors/airbyte}:dev
+          export DATASET=$DS
+          envsubst < ./tools/bin/source-harness-process.yaml | kubectl create -f -
+          POD=$(kubectl get pod -l app=source-harness -o jsonpath="{.items[0].metadata.name}")
+          kubectl wait --for=condition=Ready --timeout=20s "pod/$POD"
+          kubectl logs --follow $POD
+          EOF=$(dd if=/dev/urandom bs=15 count=1 status=none | base64)
+          echo "RUN_RESULT<<$EOF" >> $GITHUB_OUTPUT
+          kubectl logs --tail=1 $POD | while read line ; do line=${line#"$prefix"}; line=${line%"$suffix"}; echo $line >> $GITHUB_OUTPUT ; done 
+          echo "$EOF" >> $GITHUB_OUTPUT
+      - name: Link comment to workflow run
+        uses: peter-evans/create-or-update-comment@v2
+        with:
+          reactions: '+1'
+          comment-id: ${{ github.event.inputs.comment-id }}
+          body: |
+            ## Performance test Result: 
+            ```
+            ${{ steps.run-harness.outputs.RUN_RESULT }}
+            ```
+      # need to add credentials here
+  # In case of self-hosted EC2 errors, remove this block.
+  stop-test-runner:
+    name: Stop Build EC2 Runner
+    timeout-minutes: 10
+    needs:
+      - start-test-runner # required to get output from the start-runner job
+      - performance-test # required to wait when the main job is done
+      - uuid
     runs-on: ubuntu-latest
+    if: ${{ always() }} # required to stop the runner even if the error happened in the previous jobs
     steps:
-      - run: echo "Connector Performance harness stub"
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@v1
+        with:
+          aws-access-key-id: ${{ secrets.SELF_RUNNER_AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ secrets.SELF_RUNNER_AWS_SECRET_ACCESS_KEY }}
+          aws-region: us-east-2
+      - name: Checkout Airbyte
+        uses: actions/checkout@v3
+      - name: Check PAT rate limits
+        run: |
+          ./tools/bin/find_non_rate_limited_PAT \
+            ${{ secrets.GH_PAT_BUILD_RUNNER_OSS }} \
+            ${{ secrets.GH_PAT_BUILD_RUNNER_BACKUP }}
+      - name: Stop EC2 runner
+        uses: supertopher/ec2-github-runner@base64v1.0.10
+        with:
+          mode: stop
+          github-token: ${{ env.PAT }}
+          label: ${{ needs.start-test-runner.outputs.label }}
+          ec2-instance-id: ${{ needs.start-test-runner.outputs.ec2-instance-id }}
diff --git a/airbyte-commons/src/main/java/io/airbyte/commons/features/EnvVariableFeatureFlags.java b/airbyte-commons/src/main/java/io/airbyte/commons/features/EnvVariableFeatureFlags.java
@@ -64,6 +64,16 @@ public String fieldSelectionWorkspaces() {
     return getEnvOrDefault(FIELD_SELECTION_WORKSPACES, "", (arg) -> arg);
   }
 
+  @Override
+  public String strictComparisonNormalizationWorkspaces() {
+    return "";
+  }
+
+  @Override
+  public String strictComparisonNormalizationTag() {
+    return "";
+  }
+
   // TODO: refactor in order to use the same method than the ones in EnvConfigs.java
   public <T> T getEnvOrDefault(final String key, final T defaultValue, final Function<String, T> parser) {
     final String value = System.getenv(key);

diff --git a/airbyte-commons/src/main/java/io/airbyte/commons/features/FeatureFlags.java b/airbyte-commons/src/main/java/io/airbyte/commons/features/FeatureFlags.java
@@ -37,4 +37,20 @@ public interface FeatureFlags {
    */
   String fieldSelectionWorkspaces();
 
+  /**
+   * Get the workspaces allow-listed for strict incremental comparison in normalization. This takes
+   * precedence over the normalization version in destination_definitions.yaml.
+   *
+   * @return a comma-separated list of workspace ids where strict incremental comparison should be
+   *         enabled in normalization.
+   */
+  String strictComparisonNormalizationWorkspaces();
+
+  /**
+   * Get the Docker image tag representing the normalization version with strict-comparison.
+   *
+   * @return The Docker image tag representing the normalization version with strict-comparison
+   */
+  String strictComparisonNormalizationTag();
+
 }
diff --git a/airbyte-integrations/connectors-performance/source-harness/.dockerignore b/airbyte-integrations/connectors-performance/source-harness/.dockerignore
@@ -0,0 +1,4 @@
+*
+!Dockerfile
+!build
+!base.sh
diff --git a/airbyte-integrations/connectors-performance/source-harness/Dockerfile b/airbyte-integrations/connectors-performance/source-harness/Dockerfile
@@ -0,0 +1,26 @@
+FROM airbyte/integration-base-java:dev AS build
+WORKDIR /airbyte
+
+ENV APPLICATION source-harness
+
+COPY build/distributions/${APPLICATION}*.tar ${APPLICATION}.tar
+
+RUN tar xf ${APPLICATION}.tar --strip-components=1 && rm -rf ${APPLICATION}.tar
+
+FROM airbyte/integration-base-java:dev
+ARG TARGETARCH
+WORKDIR /airbyte
+
+ENV APPLICATION source-harness
+# Kubectl and socat are needed in order to create a pod similar to cloud orchestrator
+# It brings up in cluster the source pod to measure, and communicates over socat
+RUN curl -O https://s3.us-west-2.amazonaws.com/amazon-eks/1.25.6/2023-01-30/bin/linux/${TARGETARCH}/kubectl
+RUN chmod +x ./kubectl
+RUN mv ./kubectl /bin/
+RUN yum install -y socat && yum clean all
+
+COPY --from=build /airbyte /airbyte
+COPY base.sh .
+
+LABEL io.airbyte.version=0.1
+LABEL io.airbyte.name=airbyte/source-harness
diff --git a/airbyte-integrations/connectors-performance/source-harness/README.md b/airbyte-integrations/connectors-performance/source-harness/README.md
@@ -0,0 +1,6 @@
+# source-harness
+
+Performance harness for source connectors
+
+This component is used by the `/connector-performance` github action and is used in order to test throughput of 
+source connectors on a number of datasets.
diff --git a/airbyte-integrations/connectors-performance/source-harness/base.sh b/airbyte-integrations/connectors-performance/source-harness/base.sh
@@ -0,0 +1,20 @@
+#!/usr/bin/env bash
+
+set -e
+
+function echo2() {
+  echo >&2 "$@"
+}
+
+function error() {
+  echo2 "$@"
+  exit 1
+}
+
+# todo: make it easy to select source or destination and validate based on selection by adding an integration type env variable.
+function main() {
+  nohup bash -c "socat tcp-listen:9000,reuseaddr,fork \"exec:printf \'HTTP/1.0 200 OK\r\n\r\n\'\" &";
+  cat <&0 | /airbyte/bin/"$APPLICATION" "$@"
+}
+
+main "$@"
diff --git a/airbyte-integrations/connectors-performance/source-harness/build.gradle b/airbyte-integrations/connectors-performance/source-harness/build.gradle
@@ -0,0 +1,25 @@
+plugins {
+    id 'application'
+    id 'airbyte-docker'
+}
+
+repositories {
+    maven {
+        url 'https://airbyte.mycloudrepo.io/public/repositories/airbyte-public-jars/'
+    }
+}
+
+application {
+    mainClass = 'io.airbyte.integrations.performance.Main'
+    applicationDefaultJvmArgs = ['-XX:+ExitOnOutOfMemoryError', '-XX:MaxRAMPercentage=75.0']
+}
+dependencies {
+    implementation project(':airbyte-db:db-lib')
+    implementation project(':airbyte-integrations:bases:base-java')
+    implementation libs.airbyte.protocol
+    implementation 'io.fabric8:kubernetes-client:5.12.2'
+    implementation 'org.apache.commons:commons-lang3:3.11'
+    implementation files(project(':airbyte-integrations:bases:base-java').airbyteDocker.outputs)
+    implementation 'io.airbyte:airbyte-commons-worker:0.42.0'
+    implementation 'io.airbyte.airbyte-config:config-models:0.42.0'
+}