Sync kubex charts from automation-controller main @ f58951c

charts/kubex-automation-engine/BREAKING.md

@@ -0,0 +1,27 @@
+# Breaking Changes
+
+## 2026-07-20 - GPU enablement defaults and experimental contract
+
+GPU request actions now default to disabled. The following fields changed from `true` to `false`:
+
+- `spec.enablement.gpu.requests.downsize`
+- `spec.enablement.gpu.requests.upsize`
+- `spec.enablement.gpu.requests.setFromUnspecified`
+
+The GPU/KAI experimental contract changed to `v1alpha1-2026-07`. The previous contract is no longer accepted.
+
+Affected resources:
+
+- `AutomationStrategy`
+- `ClusterAutomationStrategy`
+- `GpuRebalancingPolicy`
+- `ClusterGpuRebalancingPolicy`
+- `GpuConsolidationPolicy`
+
+### Migration
+
+After upgrading the CRDs:
+
+1. Replace the existing `spec.experimental.gpuKaiContract` value with `v1alpha1-2026-07` in every affected resource.
+2. For `AutomationStrategy` and `ClusterAutomationStrategy`, explicitly set each desired GPU action under `spec.enablement.gpu.requests` to `true`. The GPU policy kinds only require the contract update in step 1.
+3. Reapply affected resources.

charts/kubex-automation-engine/CHANGELOG.md

@@ -2,6 +2,21 @@
 
 All notable changes to the Kubex Automation Engine Helm chart will be documented in this file.
 
+## [1.8.0] - 2026-07-22
+
+### Breaking
+- **[GPU enablement defaults and experimental contract](./BREAKING.md#2026-07-20---gpu-enablement-defaults-and-experimental-contract)**: GPU request actions now default to disabled. `spec.enablement.gpu.requests.downsize`, `.upsize`, and `.setFromUnspecified` changed from `true` to `false`. The GPU/KAI experimental contract changed to `v1alpha1-2026-07`; the previous contract is no longer accepted. Affects `AutomationStrategy`, `ClusterAutomationStrategy`, `GpuRebalancingPolicy`, `ClusterGpuRebalancingPolicy`, and `GpuConsolidationPolicy`.
+  - After upgrading the CRDs, update `spec.experimental.gpuKaiContract` from `v1alpha1-2026-04` to `v1alpha1-2026-07` in every affected resource.
+  - For `AutomationStrategy`/`ClusterAutomationStrategy`, explicitly set each desired GPU action under `spec.enablement.gpu.requests` to `true` if you rely on GPU request resizing. GPU policy kinds only need the contract update.
+  - Reapply affected resources after making these changes.
+
+### Added
+- Rollback monitoring now reopens automatically when a newer replacement pod appears carrying the same recommendation and matching resources, so a pod recreated by eviction, in-place resize, or a manual restart resumes being tracked instead of starting a brand-new monitoring turn.
+- `RollbackPolicy`/`ClusterRollbackPolicy` gain `spec.enableMonitoringReopen` (default `true`) to opt out of this reopen behavior for a given policy.
+- Manifest-level resource sizing is now reported to the Kubex backend, giving visibility into a workload's declared (manifest) resources alongside its live/adopted values.
+
+---
+
 ## [1.7.0] - 2026-07-14
 
 ### Added

charts/kubex-automation-engine/Chart.yaml

@@ -2,7 +2,7 @@ apiVersion: v2
 name: kubex-automation-engine
 description: A Helm chart for deploying kubex-automation-engine for automated workload rightsizing and resource optimization
 type: application
-version: 1.7.0
+version: 1.8.0
 icon: https://kubex.ai/wp-content/uploads/kubex-logo-landscape.svg
 keywords:
   - kubex

charts/kubex-automation-engine/docs/Automation-Strategies.md

@@ -1,6 +1,6 @@
 # Automation Strategies
 
-> Experimental: GPU/KAI-related fields in this resource are subject to breaking changes. When using them, set `spec.experimental.gpuKaiContract: v1alpha1-2026-04`.
+> Experimental: GPU/KAI-related fields in this resource are subject to breaking changes. When using them, set `spec.experimental.gpuKaiContract: v1alpha1-2026-07`.
 
 `AutomationStrategy` defines how resizing is allowed to happen within a namespace.
 
@@ -58,6 +58,13 @@ Usage-level `floor` and `ceiling` values apply to all containers by default. Add
 | `spec.enablement.memory.limits.ceiling` | none | Maximum memory limit target. |
 | `spec.enablement.memory.limits.containers.<name>.floor` | none | Container-specific memory limit minimum that overrides the usage-level floor for that container only. |
 | `spec.enablement.memory.limits.containers.<name>.ceiling` | none | Container-specific memory limit maximum that overrides the usage-level ceiling for that container only. |
+| `spec.enablement.gpu.requests.downsize` | `false` | EXPERIMENTAL. Allows reducing GPU requests. |
+| `spec.enablement.gpu.requests.upsize` | `false` | EXPERIMENTAL. Allows increasing GPU requests. |
+| `spec.enablement.gpu.requests.setFromUnspecified` | `false` | EXPERIMENTAL. Allows setting GPU requests when currently unset. |
+| `spec.enablement.gpu.requests.floor` | none | EXPERIMENTAL. Minimum GPU request target. |
+| `spec.enablement.gpu.requests.ceiling` | none | EXPERIMENTAL. Maximum GPU request target. |
+| `spec.enablement.gpu.requests.containers.<name>.floor` | none | EXPERIMENTAL. Container-specific GPU request minimum that overrides the usage-level floor for that container only. |
+| `spec.enablement.gpu.requests.containers.<name>.ceiling` | none | EXPERIMENTAL. Container-specific GPU request maximum that overrides the usage-level ceiling for that container only. |
 | `spec.kai.queue` | `kubex-unlimited-gpu-queue` | EXPERIMENTAL. Default KAI queue label applied for KAI GPU admission mutation. |
 | `spec.kai.setQueueWhenSpecified` | `false` | EXPERIMENTAL. Allows strategy queue value to overwrite an existing `kai.scheduler/queue` label. |
 | `spec.kai.vllm` | none | EXPERIMENTAL. Enables admission-time vLLM tuning for KAI GPU-sharing workloads. Requires `spec.experimental.gpuKaiContract`. |
@@ -180,7 +187,7 @@ Example:
 ```yaml
 spec:
   experimental:
-    gpuKaiContract: v1alpha1-2026-04
+    gpuKaiContract: v1alpha1-2026-07
   kai:
     vllm:
       gpuMemoryUtilizationBufferPercent: 10

charts/kubex-automation-engine/docs/Cluster-Automation-Strategies.md

@@ -1,6 +1,6 @@
 # Cluster Automation Strategies
 
-> Experimental: GPU/KAI-related fields in this resource are subject to breaking changes. When using them, set `spec.experimental.gpuKaiContract: v1alpha1-2026-04`.
+> Experimental: GPU/KAI-related fields in this resource are subject to breaking changes. When using them, set `spec.experimental.gpuKaiContract: v1alpha1-2026-07`.
 
 `ClusterAutomationStrategy` defines how resizing is allowed to happen for cluster-scoped policy flows.
 
@@ -52,6 +52,13 @@ Usage-level `floor` and `ceiling` values apply to all containers by default. Add
 | `spec.enablement.memory.limits.ceiling` | none | Maximum memory limit target. |
 | `spec.enablement.memory.limits.containers.<name>.floor` | none | Container-specific memory limit minimum that overrides the usage-level floor for that container only. |
 | `spec.enablement.memory.limits.containers.<name>.ceiling` | none | Container-specific memory limit maximum that overrides the usage-level ceiling for that container only. |
+| `spec.enablement.gpu.requests.downsize` | `false` | EXPERIMENTAL. Allows reducing GPU requests. |
+| `spec.enablement.gpu.requests.upsize` | `false` | EXPERIMENTAL. Allows increasing GPU requests. |
+| `spec.enablement.gpu.requests.setFromUnspecified` | `false` | EXPERIMENTAL. Allows setting GPU requests when currently unset. |
+| `spec.enablement.gpu.requests.floor` | none | EXPERIMENTAL. Minimum GPU request target. |
+| `spec.enablement.gpu.requests.ceiling` | none | EXPERIMENTAL. Maximum GPU request target. |
+| `spec.enablement.gpu.requests.containers.<name>.floor` | none | EXPERIMENTAL. Container-specific GPU request minimum that overrides the usage-level floor for that container only. |
+| `spec.enablement.gpu.requests.containers.<name>.ceiling` | none | EXPERIMENTAL. Container-specific GPU request maximum that overrides the usage-level ceiling for that container only. |
 | `spec.inPlaceResize.enabled` | `true` | Enables the in-place resize execution path. |
 | `spec.inPlaceResize.containerRestart` | `false` | Allows in-place resize operations that require container restart. |
 | `spec.podEviction.enabled` | `true` | Enables eviction-based fallback. |

charts/kubex-automation-engine/docs/Configuration-Reference.md

@@ -290,7 +290,7 @@ Use [Global Configuration Reference](./Global-Configuration.md) for the CR field
 | `globalConfiguration.webhookProbe.resources` | `{}` | Resource requests and limits for the dry-run webhook probe container |
 | `globalConfiguration.webhookProbe.podSecurityContext` | `{}` | Pod security context for the dry-run webhook probe Pod |
 | `globalConfiguration.webhookProbe.securityContext` | `{}` | Container security context for the dry-run webhook probe container |
-| `experimental.gpuKaiContract` | `v1alpha1-2026-04` | Required acknowledgement token for experimental GPU/KAI CR fields rendered by the chart |
+| `experimental.gpuKaiContract` | `v1alpha1-2026-07` | Required acknowledgement token for experimental GPU/KAI CR fields rendered by the chart |
 
 Webhook behavior notes:
 

charts/kubex-automation-engine/docs/GPU-Sharing-with-KAI.md

@@ -5,7 +5,7 @@ This guide shows how to configure GPU sharing with KAI and Kubex Automation Engi
 Tested with KAI `v0.12.16`.
 
 > [!IMPORTANT]
-> GPU/KAI fields and related custom resources are experimental and subject to breaking changes. Set `spec.experimental.gpuKaiContract: v1alpha1-2026-04` on GPU/KAI resources.
+> GPU/KAI fields and related custom resources are experimental and subject to breaking changes. Set `spec.experimental.gpuKaiContract: v1alpha1-2026-07` on GPU/KAI resources.
 
 ## Prerequisites
 
@@ -84,7 +84,7 @@ metadata:
 spec:
   experimental:
     # Required contract version for the current experimental GPU/KAI integration.
-    gpuKaiContract: v1alpha1-2026-04
+    gpuKaiContract: v1alpha1-2026-07
   enablement:
     # Convert matching workloads to the KAI GPU-sharing flow.
     overrideScheduler: "kai"
@@ -137,7 +137,7 @@ metadata:
 spec:
   experimental:
     # Required contract version for the current experimental GPU/KAI integration.
-    gpuKaiContract: v1alpha1-2026-04
+    gpuKaiContract: v1alpha1-2026-07
   scope:
     namespaceSelector:
       operator: In
@@ -229,7 +229,7 @@ Example math:
 
 Behavior:
 
-- requires experimental contract: `spec.experimental.gpuKaiContract: v1alpha1-2026-04`
+- requires experimental contract: `spec.experimental.gpuKaiContract: v1alpha1-2026-07`
 - only runs for KAI GPU request admission mutation
 - only mutates detected vLLM containers
 - updates existing `--gpu-memory-utilization=<value>`
@@ -266,7 +266,7 @@ metadata:
   name: gpu-consolidation-pool-a
 spec:
   experimental:
-    gpuKaiContract: v1alpha1-2026-04
+    gpuKaiContract: v1alpha1-2026-07
   nodeSelector:
     matchLabels:
       cloud.google.com/gke-accelerator: nvidia-tesla-t4

charts/kubex-automation-engine/docs/Global-Configuration.md

@@ -1,6 +1,6 @@
 # Global Configuration
 
-> Experimental: GPU/KAI-related fields in this resource are subject to breaking changes. When using them, set `spec.experimental.gpuKaiContract: v1alpha1-2026-04`.
+> Experimental: GPU/KAI-related fields in this resource are subject to breaking changes. When using them, set `spec.experimental.gpuKaiContract: v1alpha1-2026-07`.
 
 `GlobalConfiguration` defines cluster-wide controller behavior that applies across strategies and policies.
 

charts/kubex-automation-engine/docs/Rollback-Policies.md

@@ -27,6 +27,7 @@ Use them when you want the controller to monitor workloads after a resize is app
 | `spec.backoff.multiplyByTurn` | none (required) | Scales the base duration by the current turn number. Example: with `timePeriod: 1m` and `multiplyByTurn: 2`, turn 1 waits 2m and turn 2 waits 4m. |
 | `spec.backoff.maxAttempts` | none (required) | Maximum number of turns before the controller marks the rollback as permanently failed. |
 | `spec.weight` | `0` | Higher weight wins when multiple rollback policies match. When weights are equal, older policies win. |
+| `spec.enableMonitoringReopen` | `true` | Set to `false` to stop reopening a completed (`monitoringSucceeded`) turn when a newer replacement pod appears carrying the same recommendation and resources. When disabled, only a genuinely new recommendation starts a fresh monitoring turn for an owner that has already succeeded. |
 
 ## `ClusterRollbackPolicy.spec`
 
@@ -43,6 +44,7 @@ Use them when you want the controller to monitor workloads after a resize is app
 | `spec.backoff.multiplyByTurn` | none (required) | Scales the base duration by the current turn number. Example: with `timePeriod: 1m` and `multiplyByTurn: 2`, turn 1 waits 2m and turn 2 waits 4m. |
 | `spec.backoff.maxAttempts` | none (required) | Maximum number of turns before the controller marks the rollback as permanently failed. |
 | `spec.weight` | `0` | Higher weight wins when multiple rollback policies match. When weights are equal, older policies win. |
+| `spec.enableMonitoringReopen` | `true` | Set to `false` to stop reopening a completed (`monitoringSucceeded`) turn when a newer replacement pod appears carrying the same recommendation and resources. When disabled, only a genuinely new recommendation starts a fresh monitoring turn for an owner that has already succeeded. |
 
 ## Example: Namespaced Rollback Policy
 

charts/kubex-automation-engine/docs/gpu-consolidation-policy.md

@@ -1,6 +1,6 @@
 # GPU Consolidation Policy
 
-> Experimental: GPU/KAI fields and related custom resources are subject to breaking changes. Set `spec.experimental.gpuKaiContract: v1alpha1-2026-04`.
+> Experimental: GPU/KAI fields and related custom resources are subject to breaking changes. Set `spec.experimental.gpuKaiContract: v1alpha1-2026-07`.
 
 `GpuConsolidationPolicy` is a cluster-scoped controller that looks at scheduled pods carrying the `gpu-fraction` annotation and tries to consolidate them off an underutilized node.
 
@@ -30,7 +30,7 @@ metadata:
   name: gpu-consolidation-pool-a
 spec:
   experimental:
-    gpuKaiContract: v1alpha1-2026-04
+    gpuKaiContract: v1alpha1-2026-07
   nodeSelector:
     matchLabels:
       kubex.ai/gpu-pool: pool-a
@@ -47,7 +47,7 @@ metadata:
   name: gpu-consolidation-l40s
 spec:
   experimental:
-    gpuKaiContract: v1alpha1-2026-04
+    gpuKaiContract: v1alpha1-2026-07
   nodeSelector:
     matchExpressions:
     - key: kubex.ai/gpu-pool
@@ -67,7 +67,7 @@ metadata:
   name: gpu-consolidation-h100
 spec:
   experimental:
-    gpuKaiContract: v1alpha1-2026-04
+    gpuKaiContract: v1alpha1-2026-07
   nodeSelector:
     matchLabels:
       kubex.ai/gpu-pool: training-h100

charts/kubex-automation-engine/docs/gpu-rebalancing-policies.md

@@ -1,6 +1,6 @@
 # GPU Rebalancing Policies
 
-> Experimental: GPU/KAI fields and related custom resources are subject to breaking changes. Set `spec.experimental.gpuKaiContract: v1alpha1-2026-04`.
+> Experimental: GPU/KAI fields and related custom resources are subject to breaking changes. Set `spec.experimental.gpuKaiContract: v1alpha1-2026-07`.
 
 `GpuRebalancingPolicy` and `ClusterGpuRebalancingPolicy` emit GPU rebalancing recommendations (upsize and downsize) from Prometheus utilization.
 
@@ -37,7 +37,7 @@ metadata:
   namespace: default
 spec:
   experimental:
-    gpuKaiContract: v1alpha1-2026-04
+    gpuKaiContract: v1alpha1-2026-07
   scope:
     labelSelector:
       matchLabels:
@@ -89,7 +89,7 @@ metadata:
   name: gpu-rebalance-cluster
 spec:
   experimental:
-    gpuKaiContract: v1alpha1-2026-04
+    gpuKaiContract: v1alpha1-2026-07
   scope:
     namespaceSelector:
       operator: In

charts/kubex-automation-engine/values.yaml

@@ -4,7 +4,7 @@ crdCheck:
   skip: false
   # Minimum version of the kubex-crds chart required before this chart can be installed/upgraded.
   # Bump this whenever a new CRD field or schema change is required.
-  minCRDsHelmVersion: "1.7.0"
+  minCRDsHelmVersion: "1.8.0"
 # -- Whether to create the kubex-gateway-config Secret automatically
 createSecrets: true
 
@@ -54,14 +54,14 @@ image:
   # -- Image pull policy
   pullPolicy: IfNotPresent
   # -- Image tag
-  tag: "1.7.0"
+  tag: "1.8.0"
 
 cleanup:
   image:
     # -- Image repository for the pre-delete cleanup job.
     repository: "densify/kubex-automation-cleanup"
     # -- Image tag for the pre-delete cleanup job. Defaults to the controller image tag.
-    tag: "1.7.0"
+    tag: "1.8.0"
     # -- Image pull policy for the pre-delete cleanup job.
     pullPolicy: "IfNotPresent"
   # -- Optional pod security context for the pre-delete cleanup job
@@ -323,7 +323,7 @@ crds: {}
 # Experimental feature acknowledgements for unstable APIs.
 experimental:
   # -- Required acknowledgement token for GPU/KAI experimental fields.
-  gpuKaiContract: "v1alpha1-2026-04"
+  gpuKaiContract: "v1alpha1-2026-07"
 
 # Global configuration defaults
 # These will be used to create the default GlobalConfiguration CR

charts/kubex-crds/Chart.yaml

@@ -2,7 +2,7 @@ apiVersion: v2
 name: kubex-crds
 description: CRDs for Kubex Automation Engine
 icon: https://kubex.ai/wp-content/uploads/kubex-logo-landscape.svg
-version: 1.7.0
+version: 1.8.0
 keywords:
   - crd
   - kubex

charts/kubex-crds/templates/rightsizing.kubex.ai_automationstrategies.yaml

@@ -180,22 +180,22 @@ spec:
                               Any bound omitted for a specific container falls back to the usage-level floor/ceiling.
                             type: object
                           downsize:
-                            default: true
-                            description: downsize controls whether resource downsizing
+                            default: false
+                            description: downsize controls whether GPU request downsizing
                               actions are permitted.
                             type: boolean
                           floor:
                             description: floor is the minimum allowed resource quantity
                               for this usage.
                             type: string
                           setFromUnspecified:
-                            default: true
+                            default: false
                             description: setFromUnspecified controls whether an unspecified
-                              resource value may be set.
+                              GPU request may be set.
                             type: boolean
                           upsize:
-                            default: true
-                            description: upsize controls whether resource upsizing
+                            default: false
+                            description: upsize controls whether GPU request upsizing
                               actions are permitted.
                             type: boolean
                         type: object
@@ -330,7 +330,7 @@ spec:
                       EXPERIMENTAL: this acknowledgement value is required for unstable GPU/KAI behavior.
                     enum:
                     - none
-                    - v1alpha1-2026-04
+                    - v1alpha1-2026-07
                     type: string
                 required:
                 - gpuKaiContract

charts/kubex-crds/templates/rightsizing.kubex.ai_clusterautomationstrategies.yaml

@@ -180,22 +180,22 @@ spec:
                               Any bound omitted for a specific container falls back to the usage-level floor/ceiling.
                             type: object
                           downsize:
-                            default: true
-                            description: downsize controls whether resource downsizing
+                            default: false
+                            description: downsize controls whether GPU request downsizing
                               actions are permitted.
                             type: boolean
                           floor:
                             description: floor is the minimum allowed resource quantity
                               for this usage.
                             type: string
                           setFromUnspecified:
-                            default: true
+                            default: false
                             description: setFromUnspecified controls whether an unspecified
-                              resource value may be set.
+                              GPU request may be set.
                             type: boolean
                           upsize:
-                            default: true
-                            description: upsize controls whether resource upsizing
+                            default: false
+                            description: upsize controls whether GPU request upsizing
                               actions are permitted.
                             type: boolean
                         type: object
@@ -330,7 +330,7 @@ spec:
                       EXPERIMENTAL: this acknowledgement value is required for unstable GPU/KAI behavior.
                     enum:
                     - none
-                    - v1alpha1-2026-04
+                    - v1alpha1-2026-07
                     type: string
                 required:
                 - gpuKaiContract