From ec88b6aaa48067fd6a3bd9c0c45d8871a230ca64 Mon Sep 17 00:00:00 2001
From: jgrusewski <jgrusewski@localhost>
Date: Tue, 31 Mar 2026 10:20:06 +0200
Subject: [PATCH] fix: add volumes to all train-* templates for
 compile-and-train compat

templateRef doesn't inherit workflow-level volumes from the referenced
template. All 3 train wrappers (DQN, PPO, supervised) now define
git-ssh-key, training-data, cargo-target-cuda volumes and workspace
volumeClaimTemplate locally.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---
 infra/k8s/argo/train-dqn-template.yaml        | 24 +++++++++++++++++++
 infra/k8s/argo/train-ppo-template.yaml        | 24 +++++++++++++++++++
 infra/k8s/argo/train-supervised-template.yaml | 24 +++++++++++++++++++
 3 files changed, 72 insertions(+)

diff --git a/infra/k8s/argo/train-dqn-template.yaml b/infra/k8s/argo/train-dqn-template.yaml
index 34a2a39ce..840621519 100644
--- a/infra/k8s/argo/train-dqn-template.yaml
+++ b/infra/k8s/argo/train-dqn-template.yaml
@@ -65,6 +65,30 @@ spec:
       - name: cuda-compute-cap
         value: "90"
 
+  # Volumes required by compile-and-train steps
+  volumes:
+    - name: git-ssh-key
+      secret:
+        secretName: argo-git-ssh-key
+        defaultMode: 256
+    - name: training-data
+      persistentVolumeClaim:
+        claimName: training-data-pvc
+        readOnly: true
+    - name: cargo-target-cuda
+      persistentVolumeClaim:
+        claimName: cargo-target-cuda
+
+  volumeClaimTemplates:
+    - metadata:
+        name: workspace
+      spec:
+        accessModes: ["ReadWriteOnce"]
+        storageClassName: scw-bssd
+        resources:
+          requests:
+            storage: 5Gi
+
   templates:
     - name: pipeline
       steps:
diff --git a/infra/k8s/argo/train-ppo-template.yaml b/infra/k8s/argo/train-ppo-template.yaml
index 607fea7c1..f8a5a04bf 100644
--- a/infra/k8s/argo/train-ppo-template.yaml
+++ b/infra/k8s/argo/train-ppo-template.yaml
@@ -69,6 +69,30 @@ spec:
       - name: cuda-compute-cap
         value: "90"
 
+  # Volumes required by compile-and-train steps
+  volumes:
+    - name: git-ssh-key
+      secret:
+        secretName: argo-git-ssh-key
+        defaultMode: 256
+    - name: training-data
+      persistentVolumeClaim:
+        claimName: training-data-pvc
+        readOnly: true
+    - name: cargo-target-cuda
+      persistentVolumeClaim:
+        claimName: cargo-target-cuda
+
+  volumeClaimTemplates:
+    - metadata:
+        name: workspace
+      spec:
+        accessModes: ["ReadWriteOnce"]
+        storageClassName: scw-bssd
+        resources:
+          requests:
+            storage: 5Gi
+
   templates:
     - name: pipeline
       steps:
diff --git a/infra/k8s/argo/train-supervised-template.yaml b/infra/k8s/argo/train-supervised-template.yaml
index 62959d2ea..001020176 100644
--- a/infra/k8s/argo/train-supervised-template.yaml
+++ b/infra/k8s/argo/train-supervised-template.yaml
@@ -73,6 +73,30 @@ spec:
       - name: cuda-compute-cap
         value: "90"
 
+  # Volumes required by compile-and-train steps
+  volumes:
+    - name: git-ssh-key
+      secret:
+        secretName: argo-git-ssh-key
+        defaultMode: 256
+    - name: training-data
+      persistentVolumeClaim:
+        claimName: training-data-pvc
+        readOnly: true
+    - name: cargo-target-cuda
+      persistentVolumeClaim:
+        claimName: cargo-target-cuda
+
+  volumeClaimTemplates:
+    - metadata:
+        name: workspace
+      spec:
+        accessModes: ["ReadWriteOnce"]
+        storageClassName: scw-bssd
+        resources:
+          requests:
+            storage: 5Gi
+
   templates:
     - name: pipeline
       steps: