[stress testing] Update prod agent pool. Swap login/pushimages parame…

…ter defaults (#6103)

* Update prod agent pool size

* Swap defaults: -Login and -PushImages to -SkipLogin and -SkipPushImages

eng/common/scripts/stress-testing/deploy-stress-tests.ps1

@@ -1,16 +1,14 @@
-# Set a default parameter set here so we can call this script without requiring -Login and -Subscription,
-# but if it IS called with either of those, then both parameters need to be required. Not defining a
-# default parameter set makes Login/Subscription required all the time.
+# Not defining a default parameter set makes SkipLogin/Subscription required all the time.
 [CmdletBinding(DefaultParameterSetName = 'Default')]
 param(
     [string]$SearchDirectory,
     [hashtable]$Filters,
     [string]$Environment,
     [string]$Repository,
-    [switch]$PushImages,
+    [switch]$SkipPushImages,
     [string]$ClusterGroup,
     [string]$DeployId,
-    [switch]$Login,
+    [switch]$SkipLogin,
     [string]$Subscription,
 
     # Default to true in Azure Pipelines environments

eng/common/scripts/stress-testing/stress-test-deployment-lib.ps1

@@ -43,7 +43,7 @@ function RunOrExitOnFailure()
     }
 }
 
-function Login([string]$subscription, [string]$clusterGroup, [switch]$pushImages)
+function Login([string]$subscription, [string]$clusterGroup, [switch]$skipPushImages)
 {
     Write-Host "Logging in to subscription, cluster and container registry"
     az account show *> $null
@@ -73,7 +73,7 @@ function Login([string]$subscription, [string]$clusterGroup, [switch]$pushImages
         RunOrExitOnFailure kubectl config set-context $clusterName --namespace $defaultNamespace
     }
 
-    if ($pushImages) {
+    if (!$skipPushImages) {
         $registry = RunOrExitOnFailure az acr list -g $clusterGroup --subscription $subscription -o json
         $registryName = ($registry | ConvertFrom-Json).name
         RunOrExitOnFailure az acr login -n $registryName
@@ -86,10 +86,10 @@ function DeployStressTests(
     # Default to playground environment
     [string]$environment = 'pg',
     [string]$repository = '',
-    [switch]$pushImages,
+    [switch]$skipPushImages,
     [string]$clusterGroup = '',
     [string]$deployId = '',
-    [switch]$login,
+    [switch]$skipLogin,
     [string]$subscription = '',
     [switch]$CI,
     [string]$Namespace,
@@ -125,8 +125,8 @@ function DeployStressTests(
         throw "clusterGroup and subscription parameters must be specified when deploying to an environment that is not pg or prod."
     }
 
-    if ($login) {
-        Login -subscription $subscription -clusterGroup $clusterGroup -pushImages:$pushImages
+    if (!$skipLogin) {
+        Login -subscription $subscription -clusterGroup $clusterGroup -skipPushImages:$skipPushImages
     }
 
     $chartRepoName = 'stress-test-charts'
@@ -162,8 +162,8 @@ function DeployStressTests(
             -deployId $deployer `
             -environment $environment `
             -repositoryBase $repository `
-            -pushImages:$pushImages `
-            -login:$login `
+            -skipPushImages:$skipPushImages `
+            -skipLogin:$skipLogin `
             -clusterGroup $clusterGroup `
             -subscription $subscription
     }
@@ -189,8 +189,8 @@ function DeployStressPackage(
     [string]$deployId,
     [string]$environment,
     [string]$repositoryBase,
-    [switch]$pushImages,
-    [switch]$login,
+    [switch]$skipPushImages,
+    [switch]$skipLogin,
     [string]$clusterGroup,
     [string]$subscription
 ) {
@@ -267,7 +267,7 @@ function DeployStressPackage(
         }
         $dockerfileName = ($dockerFilePath -split { $_ -in '\', '/' })[-1].ToLower()
         $imageTag = $imageTagBase + "/${dockerfileName}:${deployId}"
-        if ($pushImages) {
+        if (!$skipPushImages) {
             Write-Host "Building and pushing stress test docker image '$imageTag'"
             $dockerFile = Get-ChildItem $dockerFilePath
 
@@ -290,8 +290,8 @@ function DeployStressPackage(
 
             Run docker push $imageTag
             if ($LASTEXITCODE) {
-                if ($login) {
-                    Write-Warning "If docker push is failing due to authentication issues, try calling this script with '-Login'"
+                if (!$skipLogin) {
+                    Write-Warning "If docker push is failing due to authentication issues, try calling this script without '-SkipLogin'"
                 }
             }
         }

eng/pipelines/templates/jobs/stress-test-release.yml

@@ -74,7 +74,5 @@ jobs:
           -Filters $(Filters)
           -Environment '${{ parameters.Environment }}'
           -Repository '$(Agent.JobName)'
-          -PushImages
-          -Login
           -DeployId '$(Build.BuildNumber)'
           -CI

tools/stress-cluster/chaos/README.md

@@ -52,17 +52,12 @@ package dependencies, and building and pushing docker images. The script must be
 
 If using bash or another linux terminal, a [powershell core](https://docs.microsoft.com/powershell/scripting/install/installing-powershell-core-on-linux?view=powershell-7.1) shell can be invoked via `pwsh`.
 
-The first invocation of the script must be run with the `-Login` flag to set up cluster and container registry access.
-
 ```
 cd <stress test search directory>
-
-<repo root>/eng/common/scripts/stress-testing/deploy-stress-tests.ps1 `
-    -Login `
-    -PushImages
+<repo root>/eng/common/scripts/stress-testing/deploy-stress-tests.ps1
 ```
 
-To re-deploy more quickly, the script can be run without `-Login` and/or without `-PushImages` (if no code changes were
+To re-deploy more quickly, the script can be run with `-SkipLogin` and/or with `-SkipPushImages` (if no code changes were
 made).
 
 ```
@@ -622,7 +617,7 @@ Follow the below commands to execute a sample test.
 
 ```
 cd ./examples/network_stress_example
-pwsh ../../../../../eng/common/scripts/stress-testing/deploy-stress-tests.ps1 -Login -PushImages
+pwsh ../../../../../eng/common/scripts/stress-testing/deploy-stress-tests.ps1
 ```
 
 Verify the pods in the job have booted and are running ok (with chaos network failures):

tools/stress-cluster/cluster/README.md

@@ -78,8 +78,7 @@ resource values from the newly provisioned dev environment that are required by
 Avoid checking in the updated dev values, they are for local use only.
 
 ```
-# -Login only needs to be run once or if the azure container registry credentials have expired (~24 hours)
-<tools repo>/eng/common/scripts/stress-testing/deploy-stress-tests.ps1 -Login -Environment dev
+<tools repo>/eng/common/scripts/stress-testing/deploy-stress-tests.ps1 -Environment dev
 ```
 
 ## Playground Cluster
@@ -122,7 +121,7 @@ Steps for deploying the stress test addons helm chart:
 1. Increment the version number in stress test addons' [Chart.yaml](https://github.com/Azure/azure-sdk-tools/blob/main/tools/stress-cluster/cluster/kubernetes/stress-test-addons/Chart.yaml) (e.g. 0.1.0 -> 0.1.1).
 1. Run [deploy.ps1](https://github.com/Azure/azure-sdk-tools/blob/main/tools/stress-cluster/cluster/kubernetes/stress-test-addons/deploy.ps1).
 1. Update all the helm chart versions for stress-test-addons dependency references in `azure-sdk-tools/tools/stress-cluster/chaos/examples/**/Chart.yaml`.
-1. Run azure-sdk-tools\eng\common\scripts\stress-testing\deploy-stress-tests.ps1 script in the [examples](https://github.com/Azure/azure-sdk-tools/tree/main/tools/stress-cluster/chaos/examples) directory, this will update all the nested helm charts (-login tag is needed for the first run).
+1. Run azure-sdk-tools\eng\common\scripts\stress-testing\deploy-stress-tests.ps1 script in the [examples](https://github.com/Azure/azure-sdk-tools/tree/main/tools/stress-cluster/chaos/examples) directory, this will update all the nested helm charts (the -SkipLogin parameter can be used to speed up the script or if interactive login isn't supported by the shell).
    1. Run `kubectl get pods -n examples -w` to monitor the status of each pod and look for Running/Completed and make sure there are no errors.
 1. Update all the stress tests' Chart.yaml files across the other repos in the same manner.
 
@@ -166,8 +165,7 @@ may still be an invalid kubernetes manifest, so the example stress test should a
 the full set of changes:
 
 ```
-# -Login only needs to be run once or if the azure container registry credentials have expired (~24 hours)
-<tools repo>/eng/common/scripts/stress-testing/deploy-stress-tests.ps1 -Login
+<tools repo>/eng/common/scripts/stress-testing/deploy-stress-tests.ps1
 ```
 
 For more helm debugging info, see [here](https://helm.sh/docs/chart_template_guide/debugging/).

tools/stress-cluster/cluster/azure/cluster/cluster.bicep

@@ -4,6 +4,8 @@ param groupSuffix string
 param dnsPrefix string = 's1'
 param clusterName string
 param location string = resourceGroup().location
+param defaultAgentPoolMinNodes int = 6
+param defaultAgentPoolMaxNodes int = 20
 // AKS does not allow agentPool updates via existing managed cluster resources
 param updateNodes bool = false
 
@@ -31,9 +33,9 @@ var systemAgentPool = {
 
 var defaultAgentPool = {
   name: 'default'
-  count: 6
-  minCount: 6
-  maxCount: 20
+  count: defaultAgentPoolMinNodes
+  minCount: defaultAgentPoolMinNodes
+  maxCount: defaultAgentPoolMaxNodes
   mode: 'User'
   vmSize: 'Standard_D8a_v4'
   type: 'VirtualMachineScaleSets'

tools/stress-cluster/cluster/azure/main.bicep

@@ -7,6 +7,8 @@ param clusterLocation string = 'westus3'
 param staticTestKeyvaultName string
 param staticTestKeyvaultGroup string
 param monitoringLocation string = 'centralus'
+param defaultAgentPoolMinNodes int = 6
+param defaultAgentPoolMaxNodes int = 20
 param tags object
 // AKS does not allow agentPool updates via existing managed cluster resources
 param updateNodes bool = false
@@ -75,6 +77,8 @@ module cluster 'cluster/cluster.bicep' = {
         updateNodes: updateNodes
         location: clusterLocation
         clusterName: clusterName
+        defaultAgentPoolMinNodes: defaultAgentPoolMinNodes
+        defaultAgentPoolMaxNodes: defaultAgentPoolMaxNodes
         tags: tags
         groupSuffix: groupSuffix
         workspaceId: logWorkspace.outputs.id

tools/stress-cluster/cluster/azure/parameters/pg.json

@@ -20,6 +20,12 @@
     "staticTestKeyvaultGroup": {
       "value": "rg-stress-secrets-pg"
     },
+    "defaultAgentPoolMinNodes": {
+      "value": 2
+    },
+    "defaultAgentPoolMaxNodes": {
+      "value": 10
+    },
     "tags": {
       "value": {
         "environment": "pg",

tools/stress-cluster/cluster/azure/parameters/prod.json

@@ -23,6 +23,12 @@
     "staticTestKeyvaultGroup": {
       "value": "rg-stress-secrets-prod"
     },
+    "defaultAgentPoolMinNodes": {
+      "value": 2
+    },
+    "defaultAgentPoolMaxNodes": {
+      "value": 10
+    },
     "tags": {
       "value": {
         "environment": "Prod",

tools/stress-cluster/cluster/kubernetes/generator/Stress.Generator/Program.cs

@@ -44,7 +44,7 @@ public static void Main(string[] args)
                 Console.WriteLine($"2. Add test code to ./src/");
                 Console.WriteLine($"3. Update 'Dockerfile' (see contents for help).");
                 Console.WriteLine($"4. Run the following command from within your language repository to deploy the package:");
-                Console.WriteLine($"   pwsh -c $(git rev-parse --show-toplevel)/eng/common/scripts/stress-testing/deploy-stress-tests.ps1 -Login -PushImages");
+                Console.WriteLine($"   pwsh -c $(git rev-parse --show-toplevel)/eng/common/scripts/stress-testing/deploy-stress-tests.ps1");
                 Console.WriteLine($"********************************************************************************");
             });
         }