#28 Change credentials for LocalStack between environments without re…

…plicating all of my environment properties

DRAFT_RELEASE_NOTES.md

@@ -103,8 +103,10 @@ To reduce burden of upgrading aiSSEMBLE, the Baton project is used to automate t
 | upgrade-v2-chart-files-aissemble-version-migration | Updates the helm chart dependencies within your project's deployment resources (<YOUR_PROJECT>-deploy/src/main/resources/apps/) to use the latest version of the aiSSEMBLE                               |
 | upgrade-v1-chart-files-aissemble-version-migration | Updates the docker image tags within your project's deployment resources (<YOUR_PROJECT>-deploy/src/main/resources/apps/) to use the latest version of the aiSSEMBLE                                     |
 | upgrade-mlflow-v2-external-s3-migration            | Update the mlflow V2 deployment (if present) in your project to utilize Localstack for local development and SealedSecrets for remote deployments                                                        |
+| upgrade-spark-application-s3-migration             | Update the pipeline SparkApplication(s) (if present) in your project to utilize Localstack for local development and SealedSecrets for remote deployments |
 | upgrade-foundation-extension-python-package-migration                              | Updates the pyproject.toml files within your projects pipelines folder (<YOUR_PROJECT>-pipelines) to use the updated aiSSEMBLE foundation and extension Python packages with the latest naming convention | 
-|upgrade-helm-chart-files-names-migration       | Updates the Chart.yaml and values*.yaml files within your project's deploy folder (<YOUR_PROJECT>-deploy) to use the new Helm chart naming convention (aissemble-<chart-name>-chart)                     |
+|upgrade-helm-chart-files-names-migration            | Updates the Chart.yaml and values*.yaml files within your project's deploy folder (<YOUR_PROJECT>-deploy) to use the new Helm chart naming convention (aissemble-\<chart-name\>-chart)                     |
+| upgrade-dockerfile-pip-install-migration           | Updates dockerfiles such that python dependency installations fail during the build, rather than at runtime |
 
 To deactivate any of these migrations, add the following configuration to the `baton-maven-plugin` within your root `pom.xml`:
 

build-parent/pom.xml

@@ -69,7 +69,7 @@
         <version.hadoop>3.3.4</version.hadoop>
         <version.neo4j>4.1.5_for_spark_3</version.neo4j>
         <version.aws.sdk.bundle>1.12.262</version.aws.sdk.bundle>
-        <version.baton>0.2.0</version.baton>
+        <version.baton>1.0.0</version.baton>
 
         <!-- ***************** -->
         <!-- Repository Deployment URLs - mostly in aissemble-root -->

extensions/extensions-docker/aissemble-spark/src/main/resources/docker/Dockerfile

@@ -37,6 +37,9 @@ RUN apt-get update -y && apt-get install --assume-yes \
   && apt-get clean \
   && ln -s /usr/bin/python${PYTHON_VERSION} /usr/bin/python
 
+# Pyspark uses `python3` to execute pyspark pipelines. This links out latest python install to that command.
+RUN ln -sf /usr/bin/python3.11 /usr/bin/python3
+
 ## Add spark configurations
 COPY ./src/main/resources/conf/ ${SPARK_HOME}/conf/
 

extensions/extensions-helm/aissemble-localstack-chart/README.md

@@ -25,6 +25,7 @@ https://github.com/localstack/helm-charts/tree/main/charts/localstack
 |----------------------|-----------------------------------------------------------------------------------------------------|-------------------|--------------------------------------------------------------------------|
 | fullnameOverride     | String to fully override common.names.fullname                                                      | No                | s3-local                                                                 |
 | startServices        | Comma-separated list of AWS CLI service names which should be loaded right when starting LocalStack | No                | s3                                                                       |
+| service.type         | Kubernetes Service type                                                                             | No                | LoadBalancer                                                             |
 | enableStartupScripts | Enables execution of startup behaviors                                                              | No                | true                                                                     |
 | startupScriptContent | Base script for triggering creation of localstack resources                                         | No                | Triggers creation of s3 buckets/keys                                     |
 | volumes              | Creates required volumes                                                                            | No                | configMap `localstack-resources` -> `create-s3-resources.sh`             |
@@ -34,9 +35,11 @@ https://github.com/localstack/helm-charts/tree/main/charts/localstack
 
 The following properties are provided by the `aissemble-localstack-chart` chart
 
-| Property | Description                                    | Required Override | Default |
-|----------|------------------------------------------------|-------------------|---------|
-| buckets  | Collection of buckets and keys to create in s3 | No                | []      |
+| Property                 | Description                                         | Required Override | Default            |
+|--------------------------|-----------------------------------------------------|-------------------|--------------------|
+| buckets                  | Collection of buckets and keys to create in s3      | No                | []                 |
+| credentialSecret.enabled | Whether to use a secret to store the S3 credentials | No                | true               |
+| credentialSecret.name    | Name of the credential secret                       | No                | remote-auth-config |
 
 # Migration From v1 Structure
 

extensions/extensions-helm/aissemble-localstack-chart/templates/credential-secret.yaml

@@ -0,0 +1,9 @@
+{{- if .Values.credentialSecret.enabled }}
+apiVersion: v1
+kind: Secret
+metadata:
+  name: {{ .Values.credentialSecret.name }}
+stringData:
+  AWS_ACCESS_KEY_ID: "123"
+  AWS_SECRET_ACCESS_KEY: "456"
+{{- end }}

extensions/extensions-helm/aissemble-localstack-chart/tests/credential_secret_test.yaml

@@ -0,0 +1,34 @@
+suite: localstack
+templates:
+  - credential-secret.yaml
+tests:
+  - it: Should contain correct default values when enabled
+    asserts:
+      - isKind:
+          of: Secret
+      - equal:
+          path: metadata.name
+          value: remote-auth-config
+      - equal:
+          path: stringData.AWS_ACCESS_KEY_ID
+          value: "123"
+      - equal:
+          path: stringData.AWS_SECRET_ACCESS_KEY
+          value: "456"
+  - it: Should be able to set the secret name 
+    set:
+      credentialSecret:
+        name: test-name
+    asserts:
+      - isKind:
+          of: Secret
+      - equal:
+          path: metadata.name
+          value: test-name
+  - it: Should not exist when disabled
+    set:
+      credentialSecret:
+        enabled: false
+    asserts:
+      - hasDocuments:
+          count: 0

extensions/extensions-helm/aissemble-localstack-chart/values.yaml

@@ -1,6 +1,8 @@
 localstack:
   fullnameOverride: s3-local
   startServices: s3
+  service:
+    type: LoadBalancer
   enableStartupScripts: true
   startupScriptContent: |
     #!/bin/sh
@@ -17,4 +19,7 @@ localstack:
           - key: create-s3-resources.sh
             path: create-s3-resources.sh
 
-buckets: []
+buckets: []
+credentialSecret:
+  enabled: true
+  name: remote-auth-config

foundation/foundation-mda/src/main/resources/templates/deployment/argocd/v2/sealed.secret.yaml.vm

@@ -9,4 +9,4 @@ metadata:
   name: remote-auth-config
 stringData:
   YOUR_SECRET_KEY: YOUR_SECRET_VALUE
-  YOUR_SECRET_KEY: YOUR_SECRET_VALUE
+  YOUR_SECRET_KEY: YOUR_SECRET_VALUE

foundation/foundation-mda/src/main/resources/templates/deployment/mlflow/v2/mlflow.chart.yaml.vm

@@ -31,4 +31,4 @@ appVersion: "1.0.0"
 dependencies:
   - name: aissemble-mlflow-chart
     version: ${versionTag}
-    repository: oci://ghcr.io/boozallen
+    repository: oci://ghcr.io/boozallen

foundation/foundation-mda/src/main/resources/templates/deployment/mlflow/v2/mlflow.values-dev.yaml.vm

@@ -9,9 +9,8 @@ aissemble-mlflow-chart:
       service:
         type: LoadBalancer
     externalS3:
-      existingSecret: ""
       host: "s3-local"
       port: 4566
-      accessKeyID: "123"
-      accessKeySecret: "456"
-      protocol: http
+      protocol: http
+      existingSecretAccessKeyIDKey: "AWS_ACCESS_KEY_ID"
+      existingSecretKeySecretKey: "AWS_SECRET_ACCESS_KEY"

foundation/foundation-mda/src/main/resources/templates/deployment/spark-operator/spark-application-base-values.yaml.vm

@@ -70,42 +70,39 @@ sparkApp:
       cores: 1
       coreLimit: "1200m"
       memory: "2048m"
+      #if (${useS3Local})
+      # Setup these secret key references within your SealedSecret 
+##      # See our guide for using SealedSecret's in your project to learn more
+##      # TODO: LINK-TO-GUIDE-HERE
+      envFrom:
+        - secretRef:
+            name: remote-auth-config
+      #end
       env:
         - name: KRAUSENING_BASE
           value: /opt/spark/krausening/base
         #if (${enableDataLineageSupport})
         - name: ENABLE_LINEAGE
           value: "true"
         #end
-        #if (${useS3Local})
-        - name: AWS_ACCESS_KEY_ID
-          value: "123"
-        - name: AWS_SECRET_ACCESS_KEY
-          value: "456"
-        - name: STORAGE_ENDPOINT
-          value: "http://s3-local:4566"
-        #end
       #if (${isJavaPipeline})
       javaOptions: "-DKRAUSENING_BASE=/opt/spark/krausening/base"
       #end
     executor:
       cores: 1
       memory: "4096m"
+      #if (${useS3Local})
+      envFrom:
+        - secretRef:
+            name: remote-auth-config
+      #end
       env:
         - name: KRAUSENING_BASE
           value: /opt/spark/krausening/base
         #if (${enableDataLineageSupport})
         - name: ENABLE_LINEAGE
           value: "true"
         #end
-        #if (${useS3Local})
-        - name: AWS_ACCESS_KEY_ID
-          value: "123"
-        - name: AWS_SECRET_ACCESS_KEY
-          value: "456"
-        - name: STORAGE_ENDPOINT
-          value: "http://s3-local:4566"
-        #end
       #if (${isJavaPipeline})
       javaOptions: "-DKRAUSENING_BASE=/opt/spark/krausening/base"
       #end

foundation/foundation-upgrade/src/main/java/com/boozallen/aissemble/upgrade/migration/v1_7_0/MlflowV2ExternalS3Migration.java

@@ -12,6 +12,7 @@
 import com.boozallen.aissemble.upgrade.migration.AbstractAissembleMigration;
 import com.boozallen.aissemble.upgrade.pojo.Chart;
 import com.boozallen.aissemble.upgrade.pojo.MlflowValues;
+import com.boozallen.aissemble.upgrade.pojo.MlflowValues.Mlflow;
 import com.boozallen.aissemble.upgrade.pojo.Chart.Dependency;
 import com.boozallen.aissemble.upgrade.util.FileUtils;
 import com.boozallen.aissemble.upgrade.util.YamlUtils;
@@ -122,12 +123,11 @@ protected boolean performMigration(File chartFile) {
 
         LinkedList<String> linesToAddValuesDev = new LinkedList<String>(Arrays.asList(
             "externalS3:",
-            "existingSecret: \"\"",
             "host: \"s3-local\"",
             "port: 4566",
-            "accessKeyID: \"123\"",
-            "accessKeySecret: \"456\"",
-            "protocol: http"
+            "protocol: http",
+            "existingSecretAccessKeyIDKey: \"AWS_ACCESS_KEY_ID\"",
+            "existingSecretKeySecretKey: \"AWS_SECRET_ACCESS_KEY\""
         ));
 
         return migrateValuesFile(this.valuesFile, this.valuesObject, linesToAddValues) 
@@ -146,8 +146,8 @@ private boolean migrateValuesFile(File file, MlflowValues helmValuesObject, Link
         logger.info("Migrating file: {}", file.getAbsolutePath());
         List<String> newFileContents = new ArrayList<>();
         List<String> originalFile;
-		try {
-			originalFile = FileUtils.readAllFileLines(file);
+        try {
+            originalFile = FileUtils.readAllFileLines(file);
             logger.debug("Read in {} lines", originalFile.size());
 
             int tabSize;
@@ -159,7 +159,8 @@ private boolean migrateValuesFile(File file, MlflowValues helmValuesObject, Link
                 newFileContents.add("aissemble-mlflow:");
                 newFileContents.add(SPACE.repeat(tabSize) + "mlflow:");
 
-                indentValues(linesToAdd, tabSize);
+                FileUtils.indentValues(linesToAdd.subList(0, 1), tabSize * 2);
+                FileUtils.indentValues(linesToAdd.subList(1, linesToAdd.size()), tabSize * 3);
                 newFileContents.addAll(linesToAdd);
             }
             else {
@@ -184,7 +185,8 @@ private boolean migrateValuesFile(File file, MlflowValues helmValuesObject, Link
                         // (leading space on current line) - (leading space on previous line)
                         tabSize = (line.length() - line.stripLeading().length()) - 
                                     (lineBeingAppended.length() - lineBeingAppended.stripLeading().length());
-                        indentValues(linesToAdd, tabSize);
+                        FileUtils.indentValues(linesToAdd.subList(0, 1), tabSize * 2);
+                        FileUtils.indentValues(linesToAdd.subList(1, linesToAdd.size()), tabSize * 3);
 
                         if (addMlFlowHeader) {
                             linesToAdd.addFirst(SPACE.repeat(tabSize) + "mlflow:");
@@ -211,15 +213,4 @@ private boolean migrateValuesFile(File file, MlflowValues helmValuesObject, Link
 		}
         return false;
     }
-
-    private void indentValues(LinkedList<String> values, int tabSize) {
-        for (int i = 0; i < values.size(); i++) {
-            // indent the first header values 2 tabs and the nested elements 3 tabs
-            if (i < 1) {
-                values.set(i, SPACE.repeat(tabSize * 2) + values.get(i));
-            } else {
-                values.set(i, SPACE.repeat(tabSize * 3) + values.get(i));
-            }
-        }
-    }
 }