Merge branch 'w/2.7/improvement/ZENKO-4919' into w/2.8/improvement/ZE…

…NKO-4919

.github/actions/archive-artifacts/action.yaml

@@ -74,26 +74,29 @@ runs:
         set -exu
 
         KAFKA=$(kubectl get pods -n ${NAMESPACE} -lkafka_cr=${ZENKO_NAME}-base-queue -o jsonpath='{.items[0].metadata.name}')
+        KAFKA_PATH="/tmp/artifacts/data/${STAGE}/kafka"
+        
+        mkdir -p ${KAFKA_PATH}
 
         kubectl exec -in ${NAMESPACE} ${KAFKA} -c kafka -- \
             env KAFKA_OPTS= kafka-topics.sh --bootstrap-server :9092 --list \
-            > /tmp/artifacts/data/${STAGE}/kafka-topics.log
+            > ${KAFKA_PATH}/kafka-topics.log
 
         kubectl exec -in ${NAMESPACE} ${KAFKA} -c kafka -- \
             env KAFKA_OPTS= kafka-consumer-groups.sh --bootstrap-server :9092 --list \
-            > /tmp/artifacts/data/${STAGE}/kafka-consumer-groups.log
+            > ${KAFKA_PATH}/kafka-consumer-groups.log
 
         kubectl exec -in ${NAMESPACE} ${KAFKA} -c kafka  -- \
                    env KAFKA_OPTS= kafka-consumer-groups.sh --bootstrap-server :9092 --describe --all-groups \
-                   > /tmp/artifacts/data/${STAGE}/kafka-offsets.log
+                   > ${KAFKA_PATH}/kafka-offsets.log
 
         KAFKA_SERVICE=$(kubectl get services -n ${NAMESPACE} -lkafka_cr=${ZENKO_NAME}-base-queue -o jsonpath='{.items[0].metadata.name}')
         kubectl run -n ${NAMESPACE} kcat --image=edenhill/kcat:1.7.1 --restart=Never --command -- sleep 300        
         kubectl wait -n ${NAMESPACE} pod kcat --for=condition=ready
-        cat /tmp/artifacts/data/${STAGE}/kafka-topics.log | grep -v '^__' | xargs -P 15 -I {} \
+        cat ${KAFKA_PATH}/kafka-topics.log | grep -v '^__' | xargs -P 15 -I {} \
             sh -c "kubectl exec -i -n ${NAMESPACE} kcat -- \
                        kcat -L -b ${KAFKA_SERVICE} -t {}  -C -o beginning -e -q -J \
-                       > /tmp/artifacts/data/${STAGE}/kafka-messages-{}.log"
+                       > ${KAFKA_PATH}/kafka-messages-{}.log"
       env:
         STAGE: ${{ inputs.stage }}
         NAMESPACE: ${{ inputs.zenko-namespace }}

.github/actions/deploy/action.yaml

@@ -64,9 +64,10 @@ runs:
         docker pull ${OPERATOR_IMAGE_NAME}:${OPERATOR_IMAGE_TAG}
         kind load docker-image ${OPERATOR_IMAGE_NAME}:${OPERATOR_IMAGE_TAG}
         cd ./.github/scripts/end2end
-        git clone https://${GIT_ACCESS_TOKEN}@github.com/scality/zenko-operator.git operator
+        git init operator
         cd operator
-        git checkout ${OPERATOR_IMAGE_TAG}
+        git fetch --depth 1 --no-tags https://${GIT_ACCESS_TOKEN}@github.com/scality/zenko-operator.git ${OPERATOR_IMAGE_TAG}
+        git checkout FETCH_HEAD
         tilt ci
       env:
         OPERATOR_IMAGE_TAG: ${{ inputs.zkop_tag }}

.github/scripts/end2end/configs/prometheus.yaml

@@ -36,13 +36,19 @@ spec:
   evaluationInterval: 30s
   logFormat: logfmt
   logLevel: info
+  serviceMonitorNamespaceSelector: {}
+  serviceMonitorSelector:
+    matchLabels:
+      metalk8s.scality.com/monitor: ""
   podMonitorNamespaceSelector: {}
   podMonitorSelector:
     matchLabels:
       metalk8s.scality.com/monitor: ""
+  probeNamespaceSelector: {}
   probeSelector:
     matchLabels:
       metalk8s.scality.com/monitor: ""
+  ruleNamespaceSelector: {}
   ruleSelector:
     matchLabels:
       metalk8s.scality.com/monitor: ""

.github/scripts/end2end/install-kind-dependencies.sh

@@ -58,14 +58,15 @@ kubectl rollout status -n ingress-nginx deployment/ingress-nginx-controller --ti
 
 # cert-manager
 kubectl apply --validate=false -f https://github.com/jetstack/cert-manager/releases/download/${CERT_MANAGER_VERSION}/cert-manager.yaml --wait
-# kubectl apply --validate=false -f - <<EOF
-# apiVersion: cert-manager.io/v1
-# kind: ClusterIssuer
-# metadata:
-#   name: artesca-root-ca-issuer
-# spec:
-#   selfSigned: {}
-# EOF
+kubectl rollout status -n cert-manager deployment/cert-manager-webhook --timeout=10m
+kubectl apply -f - <<EOF
+apiVersion: cert-manager.io/v1
+kind: ClusterIssuer
+metadata:
+  name: artesca-root-ca-issuer
+spec:
+  selfSigned: {}
+EOF
 
 # prometheus
 # last-applied-configuration can end up larger than 256kB  which is too large for an annotation

.github/scripts/end2end/patch-coredns.sh

@@ -33,6 +33,7 @@ corefile="
     rewrite name exact s3.dr.zenko.local ingress-nginx-controller.ingress-nginx.svc.cluster.local
     rewrite name exact sts.dr.zenko.local ingress-nginx-controller.ingress-nginx.svc.cluster.local
     rewrite name exact iam.dr.zenko.local ingress-nginx-controller.ingress-nginx.svc.cluster.local
+    rewrite name exact prom.dr.zenko.local ingress-nginx-controller.ingress-nginx.svc.cluster.local
     rewrite name exact shell-ui.dr.zenko.local ingress-nginx-controller.ingress-nginx.svc.cluster.local
     rewrite name exact website.mywebsite.com ingress-nginx-controller.ingress-nginx.svc.cluster.local
     kubernetes cluster.local in-addr.arpa ip6.arpa {

.github/workflows/alerts.yaml

@@ -1,10 +1,11 @@
 name: Test alerts
 
 on:
-  push:
-    branches-ignore:
-      - 'development/**'
-      - 'q/*'
+  workflow_call:
+    secrets:
+      GIT_ACCESS_TOKEN:
+        description: 'GitHub token'
+        required: true
 
 jobs:
   run-alert-tests:

.github/workflows/end2end.yaml

@@ -126,6 +126,10 @@ jobs:
       - name: Verify monitoring dashboard versions
         run: bash ./.github/scripts/check_versions.sh
 
+  check-alerts:
+    uses: ./.github/workflows/alerts.yaml
+    secrets: inherit
+
   check-workflows:
     runs-on: ubuntu-22.04
     steps:
@@ -388,7 +392,7 @@ jobs:
           cache-to: type=gha,mode=max,scope=end2end-ctst
 
   end2end-http:
-    needs: [build-kafka, build-test-image, check-dashboard-versions]
+    needs: [build-kafka, build-test-image]
     runs-on:
       - ubuntu
       - focal
@@ -437,7 +441,7 @@ jobs:
         run: kind delete cluster
 
   end2end-pra:
-    needs: [build-kafka, check-dashboard-versions, lint-and-build-ctst]
+    needs: [build-kafka, lint-and-build-ctst]
     runs-on: ubuntu-22.04-16core
     env:
       GIT_ACCESS_TOKEN: ${{ secrets.GIT_ACCESS_TOKEN }}
@@ -497,7 +501,7 @@ jobs:
         run: kind delete cluster
 
   end2end-https:
-    needs: [build-kafka, build-test-image, check-dashboard-versions]
+    needs: [build-kafka, build-test-image]
     runs-on:
       - ubuntu
       - focal
@@ -549,7 +553,7 @@ jobs:
         run: kind delete cluster
 
   end2end-sharded:
-    needs: [build-kafka, build-test-image, check-dashboard-versions]
+    needs: [build-kafka, build-test-image]
     runs-on:
       - ubuntu-22.04-8core
       # Enable this for Ring-based tests
@@ -589,7 +593,7 @@ jobs:
         run: kind delete cluster
 
   ctst-end2end-sharded:
-    needs: [build-kafka, lint-and-build-ctst, check-dashboard-versions]
+    needs: [build-kafka, lint-and-build-ctst]
     runs-on:
       - ubuntu-22.04-8core
     steps:
@@ -638,6 +642,7 @@ jobs:
   write-final-status:
     runs-on: ubuntu-latest
     needs:
+      - check-alerts
       - check-dashboard-versions
       - check-workflows
       - build-doc

solution/deps.yaml

@@ -6,7 +6,7 @@ backbeat:
   dashboard: backbeat/backbeat-dashboards
   image: backbeat
   policy: backbeat/backbeat-policies
-  tag: 8.6.49
+  tag: 8.6.51
   envsubst: BACKBEAT_TAG
 busybox:
   image: busybox
@@ -16,7 +16,7 @@ cloudserver:
   sourceRegistry: ghcr.io/scality
   dashboard: cloudserver/cloudserver-dashboards
   image: cloudserver
-  tag: 8.8.35
+  tag: 8.8.36
   envsubst: CLOUDSERVER_TAG
 drctl:
   sourceRegistry: ghcr.io/scality
@@ -113,7 +113,7 @@ sorbet:
   policy: sorbet/sorbet-policies
   dashboard: sorbet/sorbet-dashboards
   image: sorbet
-  tag: v1.1.12
+  tag: v1.1.13
   envsubst: SORBET_TAG
 stern: # tail any pod logs with pattern matchin
   tag: 1.30.0
@@ -136,7 +136,7 @@ vault:
 zenko-operator:
   sourceRegistry: ghcr.io/scality
   image: zenko-operator
-  tag: v1.6.3
+  tag: v1.6.5
   envsubst: ZENKO_OPERATOR_TAG
 zenko-ui:
   sourceRegistry: ghcr.io/scality

tests/ctst/common/common.ts

@@ -7,6 +7,8 @@ import assert from 'assert';
 import { Admin, Kafka } from 'kafkajs';
 import {
     createBucketWithConfiguration,
+    putMpuObject,
+    copyObject,
     putObject,
     runActionAgainstBucket,
     getObjectNameWithBackendFlakiness,
@@ -62,7 +64,7 @@ export async function cleanS3Bucket(
 }
 
 async function addMultipleObjects(this: Zenko, numberObjects: number,
-    objectName: string, sizeBytes: number, userMD?: string) {
+    objectName: string, sizeBytes: number, userMD?: string, parts?: number) {
     let lastResult = null;
     for (let i = 1; i <= numberObjects; i++) {
         this.resetCommand();
@@ -74,7 +76,9 @@ async function addMultipleObjects(this: Zenko, numberObjects: number,
         if (userMD) {
             this.addToSaved('userMetadata', userMD);
         }
-        lastResult = await putObject(this, objectNameFinal);
+        lastResult = parts === undefined
+            ? await putObject(this, objectNameFinal)
+            : await putMpuObject(this, parts, objectNameFinal);
     }
     return lastResult;
 }
@@ -144,7 +148,20 @@ Given('an existing bucket {string} {string} versioning, {string} ObjectLock {str
 
 Given('{int} objects {string} of size {int} bytes',
     async function (this: Zenko, numberObjects: number, objectName: string, sizeBytes: number) {
-        await addMultipleObjects.call(this, numberObjects, objectName, sizeBytes);
+        const result = await addMultipleObjects.call(this, numberObjects, objectName, sizeBytes);
+        assert.ifError(result?.stderr || result?.err);
+    });
+
+Given('{int} mpu objects {string} of size {int} bytes',
+    async function (this: Zenko, numberObjects: number, objectName: string, sizeBytes: number) {
+        const result = await addMultipleObjects.call(this, numberObjects, objectName, sizeBytes, undefined, 1);
+        assert.ifError(result?.stderr || result?.err);
+    });
+
+Given('{string} is copied to {string}',
+    async function (this: Zenko, sourceObject: string, destinationObject: string) {
+        const result = await copyObject(this, sourceObject, destinationObject);
+        assert.ifError(result?.stderr || result?.err);
     });
 
 Given('{int} objects {string} of size {int} bytes on {string} site',
@@ -156,12 +173,20 @@ Given('{int} objects {string} of size {int} bytes on {string} site',
         } else {
             Identity.useIdentity(IdentityEnum.ACCOUNT, Zenko.sites['source'].accountName);
         }
-        await addMultipleObjects.call(this, numberObjects, objectName, sizeBytes);
+        const result = await addMultipleObjects.call(this, numberObjects, objectName, sizeBytes);
+        assert.ifError(result?.stderr || result?.err);
     });
 
 Given('{int} objects {string} of size {int} bytes with user metadata {string}',
     async function (this: Zenko, numberObjects: number, objectName: string, sizeBytes: number, userMD: string) {
-        await addMultipleObjects.call(this, numberObjects, objectName, sizeBytes, userMD);
+        const result = await addMultipleObjects.call(this, numberObjects, objectName, sizeBytes, userMD);
+        assert.ifError(result?.stderr || result?.err);
+    });
+
+Given('{int} mpu objects {string} of size {int} bytes with user metadata {string}',
+    async function (this: Zenko, numberObjects: number, objectName: string, sizeBytes: number, userMD: string) {
+        const result = await addMultipleObjects.call(this, numberObjects, objectName, sizeBytes, userMD);
+        assert.ifError(result?.stderr || result?.err);
     });
 
 Given('a tag on object {string} with key {string} and value {string}',

tests/ctst/features/dmf.feature

@@ -93,6 +93,87 @@ Feature: DMF
     |           Non versioned |           1 |        100 |
     |               Suspended |           1 |        100 |
 
+    @2.7.0
+    @PreMerge
+    @Dmf
+    @ColdStorage
+    Scenario Outline: Overwriting of a cold object with mpu
+    Given a "<versioningConfiguration>" bucket
+    And a transition workflow to "e2e-cold" location
+    And <objectCount> objects "obj" of size <objectSize> bytes
+    Then object "obj-1" should be "transitioned" and have the storage class "e2e-cold"
+    And dmf volume should contain <objectCount> objects
+    Given <objectCount> mpu objects "obj" of size <objectSize> bytes
+    Then object "obj-1" should be "transitioned" and have the storage class "e2e-cold"
+    And dmf volume should contain 1 objects
+
+    Examples:
+    | versioningConfiguration | objectCount | objectSize |
+    |           Non versioned |           1 |        100 |
+    |               Suspended |           1 |        100 |
+
+    @2.7.0
+    @PreMerge
+    @Dmf
+    @ColdStorage
+    Scenario Outline: Overwriting of a cold object with copyObject
+    Given a "<versioningConfiguration>" bucket
+    And a transition workflow to "e2e-cold" location
+    And 2 objects "obj" of size <objectSize> bytes
+    Then object "obj-1" should be "transitioned" and have the storage class "e2e-cold"
+    And object "obj-2" should be "transitioned" and have the storage class "e2e-cold"
+    And dmf volume should contain 2 objects
+    When i restore object "obj-1" for 5 days
+    Then object "obj-1" should be "restored" and have the storage class "e2e-cold"
+    Given "obj-1" is copied to "obj-2"
+    Then object "obj-2" should be "transitioned" and have the storage class "e2e-cold"
+    And dmf volume should contain 2 objects
+
+    Examples:
+    | versioningConfiguration | objectSize |
+    |           Non versioned |        100 |
+    |               Suspended |        100 |
+
+    @2.7.0
+    @PreMerge
+    @Dmf
+    @ColdStorage
+    Scenario Outline: Overwriting of a cold object with mpu
+    Given a "<versioningConfiguration>" bucket
+    And a transition workflow to "e2e-cold" location
+    And <objectCount> objects "obj" of size <objectSize> bytes
+    Then object "obj-1" should be "transitioned" and have the storage class "e2e-cold"
+    And dmf volume should contain <objectCount> objects
+    Given <objectCount> mpu objects "obj" of size <objectSize> bytes
+    Then object "obj-1" should be "transitioned" and have the storage class "e2e-cold"
+    And dmf volume should contain 1 objects
+
+    Examples:
+    | versioningConfiguration | objectCount | objectSize |
+    |           Non versioned |           1 |        100 |
+    |               Suspended |           1 |        100 |
+
+    @2.7.0
+    @PreMerge
+    @Dmf
+    @ColdStorage
+    Scenario Outline: Overwriting of a cold object with copyObject
+    Given a "<versioningConfiguration>" bucket
+    And a transition workflow to "e2e-cold" location
+    And 2 objects "obj" of size <objectSize> bytes
+    Then object "obj-1" should be "transitioned" and have the storage class "e2e-cold"
+    And object "obj-2" should be "transitioned" and have the storage class "e2e-cold"
+    And dmf volume should contain 2 objects
+    When i restore object "obj-1" for 5 days
+    Then object "obj-1" should be "restored" and have the storage class "e2e-cold"
+    Given "obj-1" is copied to "obj-2"
+    Then object "obj-2" should be "transitioned" and have the storage class "e2e-cold"
+    And dmf volume should contain 2 objects
+
+    Examples:
+    | versioningConfiguration | objectSize |
+    |           Non versioned |        100 |
+    |               Suspended |        100 |
 
     @2.7.0
     @PreMerge
@@ -125,4 +206,4 @@ Feature: DMF
         | versioningConfiguration | objectCount | objectSize | restoreDays |
         |           Non versioned |           2 |        100 |          15 |
         |               Versioned |           2 |        100 |          15 |
-        |               Suspended |           2 |        100 |          15 |
+        |               Suspended |           2 |        100 |          15 |

tests/ctst/features/pra.feature

@@ -18,7 +18,8 @@ Feature: PRA operations
     Given a DR installed
     Then the DR source should be in phase "Running"
     And the DR sink should be in phase "Running"
-    Then the kafka DR volume exists
+    And the kafka DR volume exists
+    And prometheus should scrap federated metrics from DR sink
 
     # Check that objects are transitioned in the DR site
     Given access keys for the replicated account