From d16abb294a304cb6866eb2b83eeab1318933ba0e Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:05:15 -0500
Subject: [PATCH 01/39] Inital commit to add GH action to generate report

---
 .github/workflows/report.yaml | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)
 create mode 100644 .github/workflows/report.yaml

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
new file mode 100644
index 0000000..96758af
--- /dev/null
+++ b/.github/workflows/report.yaml
@@ -0,0 +1,31 @@
+name: Generate Data Usage Report
+
+on:
+  pull_request:
+    branches:
+      - main
+
+jobs:
+  deploy:
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v3
+
+    - name: Configure AWS credentials
+      uses: aws-actions/configure-aws-credentials@v3
+      with:
+        aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
+        aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+        # TODO param region
+        aws-region: us-east-2
+
+    - name: Configure kubectl with AWS EKS
+      # TODO param name, region
+      run: |
+        aws eks update-kubeconfig --name eks-dandihub --region us-east-2
+
+    - name: Sanity check
+      run: |
+        kubectl get pods -n jupyterhub

From 713d64cbd12e5aa08b6e4c448f1e04ee15adc6b7 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:13:59 -0500
Subject: [PATCH 02/39] Assume Jupyterhub Provisioning Role

---
 .github/workflows/report.yaml | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 96758af..829dc77 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -21,6 +21,16 @@ jobs:
         # TODO param region
         aws-region: us-east-2
 
+     - name: Assume JupyterhubProvisioningRole
+       # TODO param ProvisioningRoleARN and name ^
+      run: |
+        ROLE_ARN="arn:aws:iam::278212569472:role/JupyterhubProvisioningRole"
+        CREDS=$(aws sts assume-role --role-arn $ROLE_ARN --role-session-name "GitHubActionsSession")
+        export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
+        export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
+        export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
+
+
     - name: Configure kubectl with AWS EKS
       # TODO param name, region
       run: |

From 519360c7b004d9807fde9e524cca853b115facce Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:16:02 -0500
Subject: [PATCH 03/39] Fixup: indent

---
 .github/workflows/report.yaml | 58 +++++++++++++++++------------------
 1 file changed, 29 insertions(+), 29 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 829dc77..162b09f 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -10,32 +10,32 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-    - name: Checkout code
-      uses: actions/checkout@v3
-
-    - name: Configure AWS credentials
-      uses: aws-actions/configure-aws-credentials@v3
-      with:
-        aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
-        aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
-        # TODO param region
-        aws-region: us-east-2
-
-     - name: Assume JupyterhubProvisioningRole
-       # TODO param ProvisioningRoleARN and name ^
-      run: |
-        ROLE_ARN="arn:aws:iam::278212569472:role/JupyterhubProvisioningRole"
-        CREDS=$(aws sts assume-role --role-arn $ROLE_ARN --role-session-name "GitHubActionsSession")
-        export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
-        export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
-        export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
-
-
-    - name: Configure kubectl with AWS EKS
-      # TODO param name, region
-      run: |
-        aws eks update-kubeconfig --name eks-dandihub --region us-east-2
-
-    - name: Sanity check
-      run: |
-        kubectl get pods -n jupyterhub
+      - name: Checkout code
+        uses: actions/checkout@v3
+
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@v3
+        with:
+          aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+          # TODO param region
+          aws-region: us-east-2
+
+      - name: Assume JupyterhubProvisioningRole
+        # TODO param ProvisioningRoleARN and name ^
+        run: |
+          ROLE_ARN="arn:aws:iam::278212569472:role/JupyterhubProvisioningRole"
+          CREDS=$(aws sts assume-role --role-arn $ROLE_ARN --role-session-name "GitHubActionsSession")
+          export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
+          export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
+          export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
+
+
+      - name: Configure kubectl with AWS EKS
+        # TODO param name, region
+        run: |
+          aws eks update-kubeconfig --name eks-dandihub --region us-east-2
+
+      - name: Sanity check
+        run: |
+          kubectl get pods -n jupyterhub

From e6f481441b6512f51f17c22e83a608e66e9551ba Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:17:03 -0500
Subject: [PATCH 04/39] Rename job

---
 .github/workflows/report.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 162b09f..5b1f053 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -6,7 +6,7 @@ on:
       - main
 
 jobs:
-  deploy:
+  generate_data_usage_report:
     runs-on: ubuntu-latest
 
     steps:

From 72496f4a7846ef70856408c154f89651ee8a5d55 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:23:29 -0500
Subject: [PATCH 05/39] Add assumed role to update-kubeconfig

---
 .github/workflows/report.yaml | 4 ++--
 README.md                     | 4 ++++
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 5b1f053..8229875 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -32,9 +32,9 @@ jobs:
 
 
       - name: Configure kubectl with AWS EKS
-        # TODO param name, region
+        # TODO param name, region role-arn
         run: |
-          aws eks update-kubeconfig --name eks-dandihub --region us-east-2
+          aws eks update-kubeconfig --name eks-dandihub --region us-east-2 --role-arn arn:aws:iam::278212569472:role/JupyterhubProvisioningRole
 
       - name: Sanity check
         run: |
diff --git a/README.md b/README.md
index 5af6edf..4337f54 100644
--- a/README.md
+++ b/README.md
@@ -1,5 +1,9 @@
 # Dandihub
 
+TODO
+ - add provisioning role to cluser :q
+ -
+
 This Terraform blueprint creates a Kubernetes environment (EKS) and installs JupyterHub. Based on [AWS Data on EKS JupyterHub](https://github.com/awslabs/data-on-eks/tree/main/ai-ml/jupyterhub).
 
 ## Table of Contents

From 8428d3a2025421775320f39542f4cb9057559499 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:26:10 -0500
Subject: [PATCH 06/39] No need to add ProvisioningRole to masters

---
 README.md | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/README.md b/README.md
index 4337f54..5af6edf 100644
--- a/README.md
+++ b/README.md
@@ -1,9 +1,5 @@
 # Dandihub
 
-TODO
- - add provisioning role to cluser :q
- -
-
 This Terraform blueprint creates a Kubernetes environment (EKS) and installs JupyterHub. Based on [AWS Data on EKS JupyterHub](https://github.com/awslabs/data-on-eks/tree/main/ai-ml/jupyterhub).
 
 ## Table of Contents

From e170b593494033ec743255a6e823dea156d85e60 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:47:01 -0500
Subject: [PATCH 07/39] Deploy a pod to the cluster, and schedule with
 Karpenter

---
 .github/manifests/hello-world.yaml | 20 ++++++++++++++++++++
 .github/workflows/report.yaml      | 20 ++++++++++++++++++++
 2 files changed, 40 insertions(+)
 create mode 100644 .github/manifests/hello-world.yaml

diff --git a/.github/manifests/hello-world.yaml b/.github/manifests/hello-world.yaml
new file mode 100644
index 0000000..1977f33
--- /dev/null
+++ b/.github/manifests/hello-world.yaml
@@ -0,0 +1,20 @@
+# manifests/hello-world-pod.yaml
+apiVersion: v1
+kind: Pod
+metadata:
+  name: hello-world-pod
+spec:
+  containers:
+  - name: hello
+    image: busybox
+    command: ['sh', '-c', 'echo Hello, World! && sleep 30']
+  nodeSelector:
+    NodeGroupType: default
+    NodePool: default
+    hub.jupyter.org/node-purpose: user
+  tolerations:
+  - key: "hub.jupyter.org/dedicated"
+    operator: "Equal"
+    value: "user"
+    effect: "NoSchedule"
+
diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 8229875..d8dff65 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -39,3 +39,23 @@ jobs:
       - name: Sanity check
         run: |
           kubectl get pods -n jupyterhub
+          
+      # Step 4: Deploy Hello World Pod from manifest
+      - name: Deploy Hello World Pod
+        run: |
+          kubectl apply -f manifests/hello-world-pod.yaml
+
+      # Step 5: Wait for Pod to Complete
+      - name: Wait for Hello World Pod to complete
+        run: |
+          kubectl wait --for=condition=Ready pod/hello-world-pod --timeout=60s
+
+      # Step 6: Get Pod Logs to verify it ran successfully
+      - name: Get Hello World Pod logs
+        run: |
+          kubectl logs hello-world-pod
+
+      # Step 7: Cleanup - Delete the Pod
+      - name: Delete Hello World Pod
+        run: |
+          kubectl delete pod hello-world-pod

From bfce04694ef3b4be7fd00a54ae1738f2a02f169e Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:48:20 -0500
Subject: [PATCH 08/39] Fixup: correct path to pod manifest

---
 .github/workflows/report.yaml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index d8dff65..a498d99 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -39,11 +39,11 @@ jobs:
       - name: Sanity check
         run: |
           kubectl get pods -n jupyterhub
-          
+
       # Step 4: Deploy Hello World Pod from manifest
       - name: Deploy Hello World Pod
         run: |
-          kubectl apply -f manifests/hello-world-pod.yaml
+          kubectl apply -f .github/manifests/hello-world-pod.yaml
 
       # Step 5: Wait for Pod to Complete
       - name: Wait for Hello World Pod to complete

From 0993129e3452e35980279e0fa14a99d1b76203fa Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:49:42 -0500
Subject: [PATCH 09/39] Fixup again ugh, rename file

---
 .github/manifests/{hello-world.yaml => hello-world-pod.yaml} | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 rename .github/manifests/{hello-world.yaml => hello-world-pod.yaml} (100%)

diff --git a/.github/manifests/hello-world.yaml b/.github/manifests/hello-world-pod.yaml
similarity index 100%
rename from .github/manifests/hello-world.yaml
rename to .github/manifests/hello-world-pod.yaml

From 87027d220f91ce52f33fc59617ab3d1a96595935 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Wed, 25 Sep 2024 15:59:09 -0500
Subject: [PATCH 10/39] Delete Pod even if previous step times out

(Also increase timeout)
---
 .github/workflows/report.yaml | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index a498d99..bc1d088 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -48,14 +48,17 @@ jobs:
       # Step 5: Wait for Pod to Complete
       - name: Wait for Hello World Pod to complete
         run: |
-          kubectl wait --for=condition=Ready pod/hello-world-pod --timeout=60s
+          kubectl wait --for=condition=Ready pod/hello-world-pod --timeout=300s  # 5 minutes
+        continue-on-error: true  # Allow the workflow to continue even if this step fails
 
-      # Step 6: Get Pod Logs to verify it ran successfully
+      # Step 6: Get Pod Logs to verify it ran successfully, only if Step 5 succeeds
       - name: Get Hello World Pod logs
         run: |
           kubectl logs hello-world-pod
+        if: ${{ success() }}  # Only run this step if the previous step was successful
 
-      # Step 7: Cleanup - Delete the Pod
+      # Step 7: Cleanup - Always run this step, even if previous steps fail
       - name: Delete Hello World Pod
         run: |
           kubectl delete pod hello-world-pod
+        if: ${{ always() }}  # Always run this step, even if other steps fail

From 686f6868eff5b99e866e83a1c1cf4fc32f5c2f82 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 11 Oct 2024 12:52:23 -0500
Subject: [PATCH 11/39] Hack out initial du

---
 .github/manifests/disk-usage-report-job.yaml | 23 ++++++
 .github/scripts/du.py                        | 55 ++++++++++++++
 .github/workflows/report.yaml                | 77 ++++++++++++++------
 NEXTSTEPS                                    | 32 ++++++++
 images/Dockerfile.dandihub_report_generator  | 15 ++++
 5 files changed, 179 insertions(+), 23 deletions(-)
 create mode 100644 .github/manifests/disk-usage-report-job.yaml
 create mode 100755 .github/scripts/du.py
 create mode 100644 NEXTSTEPS
 create mode 100644 images/Dockerfile.dandihub_report_generator

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
new file mode 100644
index 0000000..2c94536
--- /dev/null
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -0,0 +1,23 @@
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: disk-usage-report-job
+spec:
+  template:
+    metadata:
+      labels:
+        app: disk-usage-report
+    spec:
+      containers:
+        - name: disk-usage-report
+          image: IMAGE_PLACEHOLDER
+      restartPolicy: Never
+      nodeSelector:
+        NodeGroupType: default
+        NodePool: default
+        hub.jupyter.org/node-purpose: user
+      tolerations:
+        - key: "hub.jupyter.org/dedicated"
+          operator: "Equal"
+          value: "user"
+          effect: "NoSchedule"
diff --git a/.github/scripts/du.py b/.github/scripts/du.py
new file mode 100755
index 0000000..29bccad
--- /dev/null
+++ b/.github/scripts/du.py
@@ -0,0 +1,55 @@
+#!/usr/bin/env python3
+
+import os
+import subprocess
+import sys
+import json
+
+OUTPUT_FILE = "du_report.json"
+SIZE_THRESHOLD_GB = 1
+SIZE_THRESHOLD_BYTES = SIZE_THRESHOLD_GB * 1024 * 1024 * 1024
+
+# Function to calculate disk usage of a directory in bytes
+def get_disk_usage_bytes(path):
+    result = subprocess.run(['du', '-sb', path], capture_output=True, text=True)
+    size_str = result.stdout.split()[0]  # Get the size in bytes (du -sb gives size in bytes)
+    return int(size_str)
+
+# Function to convert bytes to a human-readable format (e.g., KB, MB, GB)
+def bytes_to_human_readable(size_in_bytes):
+    for unit in ['B', 'KB', 'MB', 'GB', 'TB']:
+        if size_in_bytes < 1024:
+            return f"{size_in_bytes:.2f} {unit}"
+        size_in_bytes /= 1024
+
+def prepare_report(directory):
+    report = {}
+    # List user home dirs in the directory and calculate disk usage
+    for user_dir in os.listdir(directory):
+        user_path = os.path.join(directory, user_dir)
+        if os.path.isdir(user_path):
+            disk_usage_bytes = get_disk_usage_bytes(user_path)
+            report[user_dir] = {
+                "disk_usage_bytes": disk_usage_bytes
+            }
+            if disk_usage_bytes > SIZE_THRESHOLD_BYTES:
+                # TODO: Placeholder for other actions
+                report[user_dir]["action"] = f"Directory size exceeds {SIZE_THRESHOLD_BYTES / (1024**3):.2f}GB, further action taken."
+            else:
+                report[user_dir]["action"] = "No action required."
+
+    for user, data in report.items():
+        data["disk_usage_human_readable"] = bytes_to_human_readable(data["disk_usage_bytes"])
+
+    with open(OUTPUT_FILE, 'w') as f:
+        json.dump(report, f, indent=4)
+
+    print(f"Disk usage report generated at {OUTPUT_FILE}")
+
+
+if __name__ == "__main__":
+    if len(sys.argv) != 2:
+        print("Usage: du.py <directory_to_check>")
+    else:
+        directory = sys.argv[1]
+        prepare_report(directory)
diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index bc1d088..20ea815 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -13,52 +13,83 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@v3
 
+      - name: Log in to DockerHub
+        uses: docker/login-action@v2
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Build and push Docker image
+        uses: docker/build-push-action@v3
+        with:
+          context: .
+          file: images/Dockerfile.dandihub_report_generator
+          push: true
+          tags: ${{ secrets.DOCKERHUB_USERNAME }}/dandihub_report_generator:latest
+
       - name: Configure AWS credentials
         uses: aws-actions/configure-aws-credentials@v3
         with:
           aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
           aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
-          # TODO param region
           aws-region: us-east-2
 
-      - name: Assume JupyterhubProvisioningRole
-        # TODO param ProvisioningRoleARN and name ^
+      - name: Assume ProvisioningRole
         run: |
-          ROLE_ARN="arn:aws:iam::278212569472:role/JupyterhubProvisioningRole"
-          CREDS=$(aws sts assume-role --role-arn $ROLE_ARN --role-session-name "GitHubActionsSession")
+          CREDS=$(aws sts assume-role --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }} --role-session-name "GitHubActionsSession")
           export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
           export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
           export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
 
-
       - name: Configure kubectl with AWS EKS
-        # TODO param name, region role-arn
         run: |
-          aws eks update-kubeconfig --name eks-dandihub --region us-east-2 --role-arn arn:aws:iam::278212569472:role/JupyterhubProvisioningRole
+          aws eks update-kubeconfig --name eks-dandihub --region us-east-2 --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }}
 
+      # TODO remove
       - name: Sanity check
         run: |
           kubectl get pods -n jupyterhub
 
-      # Step 4: Deploy Hello World Pod from manifest
-      - name: Deploy Hello World Pod
+      - name: Replace image placeholder in manifest
+        run: |
+          sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
+
+      - name: Deploy Disk Usage Report Job
+        run: |
+          kubectl apply -f .github/manifests/disk-usage-report-job.yaml
+
+      # TODO should timeout be longer?
+      - name: Wait for Disk Usage Report Job to complete
+        run: |
+          kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=300s
+        continue-on-error: true
+
+      - name: Save Pod logs to file
+        run: |
+          POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
+          kubectl logs $POD_NAME > disk_usage_report.log
+        continue-on-error: true
+
+      # continue-on-error for previous steps so we delete the job
+      - name: Delete Disk Usage Report Job
         run: |
-          kubectl apply -f .github/manifests/hello-world-pod.yaml
+          kubectl delete job disk-usage-report-job
 
-      # Step 5: Wait for Pod to Complete
-      - name: Wait for Hello World Pod to complete
+      - name: Clone dandi-hub-usage-reports repository
         run: |
-          kubectl wait --for=condition=Ready pod/hello-world-pod --timeout=300s  # 5 minutes
-        continue-on-error: true  # Allow the workflow to continue even if this step fails
+          git clone https://github.com/dandi/dandi-hub-usage-reports.git
+          cd dandi-hub-usage-reports
 
-      # Step 6: Get Pod Logs to verify it ran successfully, only if Step 5 succeeds
-      - name: Get Hello World Pod logs
+      - name: Copy log file to repository
         run: |
-          kubectl logs hello-world-pod
-        if: ${{ success() }}  # Only run this step if the previous step was successful
+          DATE=$(date +'%Y-%m-%d')
+          mv ../disk_usage_report.log $DATE_disk_usage_report.log
 
-      # Step 7: Cleanup - Always run this step, even if previous steps fail
-      - name: Delete Hello World Pod
+      # Step 13: Commit and push logs to the repository
+      - name: Commit and push logs
         run: |
-          kubectl delete pod hello-world-pod
-        if: ${{ always() }}  # Always run this step, even if other steps fail
+          git config --global user.name "GitHub Actions"
+          git config --global user.email "actions@github.com"
+          git add disk_usage_report.log
+          git commit -m "Add disk usage report log"
+          git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git
diff --git a/NEXTSTEPS b/NEXTSTEPS
new file mode 100644
index 0000000..44d0177
--- /dev/null
+++ b/NEXTSTEPS
@@ -0,0 +1,32 @@
+DONE
+  - Set AWS_ROLE ARN secret
+  - AWS_ACCESS_KEY_ID
+  - AWS_SECRET_ACCESS_KEY
+
+TODO:
+  - Create Dockerhub Service account
+    - set username & token as secrets
+  - Create Github CI account
+  - Docker Image Tagging:
+    - The Docker image is tagged with latest. For better version control, consider using commit SHA or version numbers.
+  - Log Retrieval:
+    - The logs from the pod are retrieved to help you verify the script's output.
+  - Cleanup:
+    - Deleting the Job ensures that no resources are left running after the workflow completes.
+
+By making these updates, your workflow will now:
+
+    Include your du.py script in a Docker image.
+    Build and push this image to DockerHub.
+    Deploy a Kubernetes Job to your EKS cluster that runs the script.
+    Wait for the Job to complete and retrieve logs.
+    Clean up resources after execution.
+
+Feel free to ask if you need further assistance or clarification on any of these steps!
+
+
+- Get image pushing
+- create private gh repository under dandi org for reports
+
+
+
diff --git a/images/Dockerfile.dandihub_report_generator b/images/Dockerfile.dandihub_report_generator
new file mode 100644
index 0000000..5f46008
--- /dev/null
+++ b/images/Dockerfile.dandihub_report_generator
@@ -0,0 +1,15 @@
+FROM python:3.9-slim
+
+# Set the working directory
+WORKDIR /app
+
+# Copy the du.py script into the container
+COPY .github/scripts/du.py /app/du.py
+
+# Install required packages
+RUN apt-get update \
+    && apt-get install -y coreutils \
+    && rm -rf /var/lib/apt/lists/*
+
+# Set the entrypoint to the script
+ENTRYPOINT ["python3", "/app/du.py"]

From ff52971049feb003996ca3ad328e4733b8c342b3 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 12:38:21 -0600
Subject: [PATCH 12/39] tmp comment out job deployment, test dockerhub build

---
 .github/workflows/report.yaml | 132 +++++++++++++++++-----------------
 1 file changed, 66 insertions(+), 66 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 20ea815..c58da48 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -27,69 +27,69 @@ jobs:
           push: true
           tags: ${{ secrets.DOCKERHUB_USERNAME }}/dandihub_report_generator:latest
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v3
-        with:
-          aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
-          aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
-          aws-region: us-east-2
-
-      - name: Assume ProvisioningRole
-        run: |
-          CREDS=$(aws sts assume-role --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }} --role-session-name "GitHubActionsSession")
-          export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
-          export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
-          export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
-
-      - name: Configure kubectl with AWS EKS
-        run: |
-          aws eks update-kubeconfig --name eks-dandihub --region us-east-2 --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }}
-
-      # TODO remove
-      - name: Sanity check
-        run: |
-          kubectl get pods -n jupyterhub
-
-      - name: Replace image placeholder in manifest
-        run: |
-          sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
-
-      - name: Deploy Disk Usage Report Job
-        run: |
-          kubectl apply -f .github/manifests/disk-usage-report-job.yaml
-
-      # TODO should timeout be longer?
-      - name: Wait for Disk Usage Report Job to complete
-        run: |
-          kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=300s
-        continue-on-error: true
-
-      - name: Save Pod logs to file
-        run: |
-          POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
-          kubectl logs $POD_NAME > disk_usage_report.log
-        continue-on-error: true
-
-      # continue-on-error for previous steps so we delete the job
-      - name: Delete Disk Usage Report Job
-        run: |
-          kubectl delete job disk-usage-report-job
-
-      - name: Clone dandi-hub-usage-reports repository
-        run: |
-          git clone https://github.com/dandi/dandi-hub-usage-reports.git
-          cd dandi-hub-usage-reports
-
-      - name: Copy log file to repository
-        run: |
-          DATE=$(date +'%Y-%m-%d')
-          mv ../disk_usage_report.log $DATE_disk_usage_report.log
-
-      # Step 13: Commit and push logs to the repository
-      - name: Commit and push logs
-        run: |
-          git config --global user.name "GitHub Actions"
-          git config --global user.email "actions@github.com"
-          git add disk_usage_report.log
-          git commit -m "Add disk usage report log"
-          git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git
+      # - name: Configure AWS credentials
+      #   uses: aws-actions/configure-aws-credentials@v3
+      #   with:
+      #     aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
+      #     aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+      #     aws-region: us-east-2
+      #
+      # - name: Assume ProvisioningRole
+      #   run: |
+      #     CREDS=$(aws sts assume-role --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }} --role-session-name "GitHubActionsSession")
+      #     export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
+      #     export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
+      #     export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
+      #
+      # - name: Configure kubectl with AWS EKS
+      #   run: |
+      #     aws eks update-kubeconfig --name eks-dandihub --region us-east-2 --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }}
+      #
+      # # TODO remove
+      # - name: Sanity check
+      #   run: |
+      #     kubectl get pods -n jupyterhub
+      #
+      # - name: Replace image placeholder in manifest
+      #   run: |
+      #     sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
+      #
+      # - name: Deploy Disk Usage Report Job
+      #   run: |
+      #     kubectl apply -f .github/manifests/disk-usage-report-job.yaml
+      #
+      # # TODO should timeout be longer?
+      # - name: Wait for Disk Usage Report Job to complete
+      #   run: |
+      #     kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=300s
+      #   continue-on-error: true
+      #
+      # - name: Save Pod logs to file
+      #   run: |
+      #     POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
+      #     kubectl logs $POD_NAME > disk_usage_report.log
+      #   continue-on-error: true
+      #
+      # # continue-on-error for previous steps so we delete the job
+      # - name: Delete Disk Usage Report Job
+      #   run: |
+      #     kubectl delete job disk-usage-report-job
+      #
+      # - name: Clone dandi-hub-usage-reports repository
+      #   run: |
+      #     git clone https://github.com/dandi/dandi-hub-usage-reports.git
+      #     cd dandi-hub-usage-reports
+      #
+      # - name: Copy log file to repository
+      #   run: |
+      #     DATE=$(date +'%Y-%m-%d')
+      #     mv ../disk_usage_report.log $DATE_disk_usage_report.log
+      #
+      # # Step 13: Commit and push logs to the repository
+      # - name: Commit and push logs
+      #   run: |
+      #     git config --global user.name "GitHub Actions"
+      #     git config --global user.email "actions@github.com"
+      #     git add disk_usage_report.log
+      #     git commit -m "Add disk usage report log"
+      #     git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git

From ca6db8943f7da07f8d158a9af5b1983cebe0a8b9 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 12:42:29 -0600
Subject: [PATCH 13/39] Fixup hyphens for image name

---
 .github/workflows/report.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index c58da48..9bf5526 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -25,7 +25,7 @@ jobs:
           context: .
           file: images/Dockerfile.dandihub_report_generator
           push: true
-          tags: ${{ secrets.DOCKERHUB_USERNAME }}/dandihub_report_generator:latest
+          tags: ${{ secrets.DOCKERHUB_USERNAME }}/dandihub-report-generator:latest
 
       # - name: Configure AWS credentials
       #   uses: aws-actions/configure-aws-credentials@v3

From d228f9daf4a1d4af8a13b1b27b0bb3ceeb4cc02c Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 12:57:24 -0600
Subject: [PATCH 14/39] Write file to output location

---
 .github/scripts/du.py         |   4 +-
 .github/workflows/report.yaml | 132 +++++++++++++++++-----------------
 2 files changed, 68 insertions(+), 68 deletions(-)

diff --git a/.github/scripts/du.py b/.github/scripts/du.py
index 29bccad..260bd07 100755
--- a/.github/scripts/du.py
+++ b/.github/scripts/du.py
@@ -5,7 +5,7 @@
 import sys
 import json
 
-OUTPUT_FILE = "du_report.json"
+OUTPUT_FILE = "/output/du_report.json"
 SIZE_THRESHOLD_GB = 1
 SIZE_THRESHOLD_BYTES = SIZE_THRESHOLD_GB * 1024 * 1024 * 1024
 
@@ -44,7 +44,7 @@ def prepare_report(directory):
     with open(OUTPUT_FILE, 'w') as f:
         json.dump(report, f, indent=4)
 
-    print(f"Disk usage report generated at {OUTPUT_FILE}")
+    print(f"Disk usage report generated at {os.path.abspath(OUTPUT_FILE)}")
 
 
 if __name__ == "__main__":
diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 9bf5526..cd7f8f0 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -27,69 +27,69 @@ jobs:
           push: true
           tags: ${{ secrets.DOCKERHUB_USERNAME }}/dandihub-report-generator:latest
 
-      # - name: Configure AWS credentials
-      #   uses: aws-actions/configure-aws-credentials@v3
-      #   with:
-      #     aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
-      #     aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
-      #     aws-region: us-east-2
-      #
-      # - name: Assume ProvisioningRole
-      #   run: |
-      #     CREDS=$(aws sts assume-role --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }} --role-session-name "GitHubActionsSession")
-      #     export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
-      #     export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
-      #     export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
-      #
-      # - name: Configure kubectl with AWS EKS
-      #   run: |
-      #     aws eks update-kubeconfig --name eks-dandihub --region us-east-2 --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }}
-      #
-      # # TODO remove
-      # - name: Sanity check
-      #   run: |
-      #     kubectl get pods -n jupyterhub
-      #
-      # - name: Replace image placeholder in manifest
-      #   run: |
-      #     sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
-      #
-      # - name: Deploy Disk Usage Report Job
-      #   run: |
-      #     kubectl apply -f .github/manifests/disk-usage-report-job.yaml
-      #
-      # # TODO should timeout be longer?
-      # - name: Wait for Disk Usage Report Job to complete
-      #   run: |
-      #     kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=300s
-      #   continue-on-error: true
-      #
-      # - name: Save Pod logs to file
-      #   run: |
-      #     POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
-      #     kubectl logs $POD_NAME > disk_usage_report.log
-      #   continue-on-error: true
-      #
-      # # continue-on-error for previous steps so we delete the job
-      # - name: Delete Disk Usage Report Job
-      #   run: |
-      #     kubectl delete job disk-usage-report-job
-      #
-      # - name: Clone dandi-hub-usage-reports repository
-      #   run: |
-      #     git clone https://github.com/dandi/dandi-hub-usage-reports.git
-      #     cd dandi-hub-usage-reports
-      #
-      # - name: Copy log file to repository
-      #   run: |
-      #     DATE=$(date +'%Y-%m-%d')
-      #     mv ../disk_usage_report.log $DATE_disk_usage_report.log
-      #
-      # # Step 13: Commit and push logs to the repository
-      # - name: Commit and push logs
-      #   run: |
-      #     git config --global user.name "GitHub Actions"
-      #     git config --global user.email "actions@github.com"
-      #     git add disk_usage_report.log
-      #     git commit -m "Add disk usage report log"
-      #     git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@v3
+        with:
+          aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+          aws-region: us-east-2
+
+      - name: Assume ProvisioningRole
+        run: |
+          CREDS=$(aws sts assume-role --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }} --role-session-name "GitHubActionsSession")
+          export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
+          export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
+          export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
+
+      - name: Configure kubectl with AWS EKS
+        run: |
+          aws eks update-kubeconfig --name eks-dandihub --region us-east-2 --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }}
+
+      # TODO remove
+      - name: Sanity check
+        run: |
+          kubectl get pods -n jupyterhub
+
+      - name: Replace image placeholder in manifest
+        run: |
+          sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
+
+      - name: Deploy Disk Usage Report Job
+        run: |
+          kubectl apply -f .github/manifests/disk-usage-report-job.yaml
+
+      # TODO should timeout be longer?
+      - name: Wait for Disk Usage Report Job to complete
+        run: |
+          kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=300s
+        continue-on-error: true
+
+      - name: Save Pod logs to file
+        run: |
+          POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
+          kubectl logs $POD_NAME > disk_usage_report.log
+        continue-on-error: true
+
+      # continue-on-error for previous steps so we delete the job
+      - name: Delete Disk Usage Report Job
+        run: |
+          kubectl delete job disk-usage-report-job
+
+      - name: Clone dandi-hub-usage-reports repository
+        run: |
+          git clone https://github.com/dandi/dandi-hub-usage-reports.git
+          cd dandi-hub-usage-reports
+
+      - name: Copy log file to repository
+        run: |
+          DATE=$(date +'%Y-%m-%d')
+          mv ../disk_usage_report.log $DATE_disk_usage_report.log
+
+      # Step 13: Commit and push logs to the repository
+      - name: Commit and push logs
+        run: |
+          git config --global user.name "GitHub Actions"
+          git config --global user.email "actions@github.com"
+          git add disk_usage_report.log
+          git commit -m "Add disk usage report log"
+          git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git

From 68f707ff011f651eb3cf7c4f69fa4ef48a99c150 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 12:59:49 -0600
Subject: [PATCH 15/39] use kubectl cp to retrieve report

---
 .github/workflows/report.yaml | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index cd7f8f0..968bd5a 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -64,10 +64,10 @@ jobs:
           kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=300s
         continue-on-error: true
 
-      - name: Save Pod logs to file
+      - name: Retrieve generated report file
         run: |
           POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
-          kubectl logs $POD_NAME > disk_usage_report.log
+          kubectl cp $POD_NAME:/output/du_report.json du_report.json -n jupyterhub
         continue-on-error: true
 
       # continue-on-error for previous steps so we delete the job
@@ -80,16 +80,16 @@ jobs:
           git clone https://github.com/dandi/dandi-hub-usage-reports.git
           cd dandi-hub-usage-reports
 
-      - name: Copy log file to repository
+      - name: Copy report file to repository
         run: |
           DATE=$(date +'%Y-%m-%d')
-          mv ../disk_usage_report.log $DATE_disk_usage_report.log
+          mv ../du_report.json $DATE_du_report.json
 
-      # Step 13: Commit and push logs to the repository
-      - name: Commit and push logs
+      # Step 13: Commit and push report to the repository
+      - name: Commit and push report
         run: |
           git config --global user.name "GitHub Actions"
           git config --global user.email "actions@github.com"
-          git add disk_usage_report.log
-          git commit -m "Add disk usage report log"
+          git add $DATE_du_report.json
+          git commit -m "Add disk usage report for $DATE"
           git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git

From ad6b589bfd64720204f9d5f05ea87f833031dd74 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 14:08:57 -0600
Subject: [PATCH 16/39] Combine run blocks to use vars

---
 .github/workflows/report.yaml | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 968bd5a..d6837c3 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -78,16 +78,12 @@ jobs:
       - name: Clone dandi-hub-usage-reports repository
         run: |
           git clone https://github.com/dandi/dandi-hub-usage-reports.git
-          cd dandi-hub-usage-reports
 
-      - name: Copy report file to repository
+      - name: Copy report file to repository, commit and push report
         run: |
+          cd dandi-hub-usage-reports
           DATE=$(date +'%Y-%m-%d')
           mv ../du_report.json $DATE_du_report.json
-
-      # Step 13: Commit and push report to the repository
-      - name: Commit and push report
-        run: |
           git config --global user.name "GitHub Actions"
           git config --global user.email "actions@github.com"
           git add $DATE_du_report.json

From f18e8b7583ddcbada46d32461d6cfa85218fe427 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 14:09:29 -0600
Subject: [PATCH 17/39] Mount efs and pass arg to du script

---
 .github/manifests/disk-usage-report-job.yaml | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index 2c94536..696d8f4 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -11,6 +11,11 @@ spec:
       containers:
         - name: disk-usage-report
           image: IMAGE_PLACEHOLDER
+          args:
+            - "/home/"
+          volumeMounts:
+            - name: persistent-storage
+              mountPath: "/home/"
       restartPolicy: Never
       nodeSelector:
         NodeGroupType: default
@@ -21,3 +26,7 @@ spec:
           operator: "Equal"
           value: "user"
           effect: "NoSchedule"
+      volumes:
+        - name: persistent-storage
+          persistentVolumeClaim:
+            claimName: efs-persist

From 387cfc1240bc67be1fb3564377d9f13de823f857 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 14:15:08 -0600
Subject: [PATCH 18/39] Comment out repo pushing, lets see if the report runs

---
 .github/workflows/report.yaml | 29 +++++++++++++++--------------
 1 file changed, 15 insertions(+), 14 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index d6837c3..a12adbd 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -68,6 +68,7 @@ jobs:
         run: |
           POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
           kubectl cp $POD_NAME:/output/du_report.json du_report.json -n jupyterhub
+          cat du_report.json
         continue-on-error: true
 
       # continue-on-error for previous steps so we delete the job
@@ -75,17 +76,17 @@ jobs:
         run: |
           kubectl delete job disk-usage-report-job
 
-      - name: Clone dandi-hub-usage-reports repository
-        run: |
-          git clone https://github.com/dandi/dandi-hub-usage-reports.git
-
-      - name: Copy report file to repository, commit and push report
-        run: |
-          cd dandi-hub-usage-reports
-          DATE=$(date +'%Y-%m-%d')
-          mv ../du_report.json $DATE_du_report.json
-          git config --global user.name "GitHub Actions"
-          git config --global user.email "actions@github.com"
-          git add $DATE_du_report.json
-          git commit -m "Add disk usage report for $DATE"
-          git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git
+      # - name: Clone dandi-hub-usage-reports repository
+      #   run: |
+      #     git clone https://github.com/dandi/dandi-hub-usage-reports.git
+      #
+      # - name: Copy report file to repository, commit and push report
+      #   run: |
+      #     cd dandi-hub-usage-reports
+      #     DATE=$(date +'%Y-%m-%d')
+      #     mv ../du_report.json $DATE_du_report.json
+      #     git config --global user.name "GitHub Actions"
+      #     git config --global user.email "actions@github.com"
+      #     git add $DATE_du_report.json
+      #     git commit -m "Add disk usage report for $DATE"
+      #     git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git

From 04b4193db3f205038d1d081c53eddcffc545aff8 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 14:22:11 -0600
Subject: [PATCH 19/39] Restrict job to asmacdo for testing

---
 .github/manifests/disk-usage-report-job.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index 696d8f4..c7f60a4 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -12,7 +12,7 @@ spec:
         - name: disk-usage-report
           image: IMAGE_PLACEHOLDER
           args:
-            - "/home/"
+            - "/home/asmacdo"
           volumeMounts:
             - name: persistent-storage
               mountPath: "/home/"

From a443081ce5a1ae8bbc60532af88c4ccdeb474858 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 14:30:46 -0600
Subject: [PATCH 20/39] Sanity check. Just list the directories

---
 .github/scripts/du.py | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/.github/scripts/du.py b/.github/scripts/du.py
index 260bd07..12e0c0f 100755
--- a/.github/scripts/du.py
+++ b/.github/scripts/du.py
@@ -41,6 +41,7 @@ def prepare_report(directory):
     for user, data in report.items():
         data["disk_usage_human_readable"] = bytes_to_human_readable(data["disk_usage_bytes"])
 
+
     with open(OUTPUT_FILE, 'w') as f:
         json.dump(report, f, indent=4)
 
@@ -51,5 +52,9 @@ def prepare_report(directory):
     if len(sys.argv) != 2:
         print("Usage: du.py <directory_to_check>")
     else:
-        directory = sys.argv[1]
-        prepare_report(directory)
+        path = sys.argv[1]
+        directories = [d for d in os.listdir(path) if os.path.isdir(os.path.join(path, d))]
+
+        with open(OUTPUT_FILE, 'w') as f:
+            f.write("\n".join(directories))
+        # prepare_report(directory)

From 99ac2646b0779dc2ee6178d7f20091ccc8dc745f Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 14:43:17 -0600
Subject: [PATCH 21/39] Job was deployed, but never assigned to node, back to
 sanity check

---
 .github/workflows/report.yaml | 57 +++++++++++++++++++++++------------
 1 file changed, 38 insertions(+), 19 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index a12adbd..5889c90 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -50,32 +50,51 @@ jobs:
         run: |
           kubectl get pods -n jupyterhub
 
-      - name: Replace image placeholder in manifest
+      - name: Deploy Hello World Pod
         run: |
-          sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
+          kubectl apply -f .github/manifests/hello-world-pod.yaml
 
-      - name: Deploy Disk Usage Report Job
+      - name: Wait for Hello World Pod to complete
         run: |
-          kubectl apply -f .github/manifests/disk-usage-report-job.yaml
+          kubectl wait --for=condition=Ready pod/hello-world-pod --timeout=300s  # 5 minutes
+        continue-on-error: true  # Allow the workflow to continue even if this step fails
 
-      # TODO should timeout be longer?
-      - name: Wait for Disk Usage Report Job to complete
+      - name: Get Hello World Pod logs
         run: |
-          kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=300s
-        continue-on-error: true
+          kubectl logs hello-world-pod
+        if: ${{ success() }}  # Only run this step if the previous step was successful
 
-      - name: Retrieve generated report file
+      - name: Delete Hello World Pod
         run: |
-          POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
-          kubectl cp $POD_NAME:/output/du_report.json du_report.json -n jupyterhub
-          cat du_report.json
-        continue-on-error: true
-
-      # continue-on-error for previous steps so we delete the job
-      - name: Delete Disk Usage Report Job
-        run: |
-          kubectl delete job disk-usage-report-job
-
+          kubectl delete pod hello-world-pod
+        if: ${{ always() }}  # Always run this step, even if other steps fail
+      #
+      # - name: Replace image placeholder in manifest
+      #   run: |
+      #     sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
+      #
+      # - name: Deploy Disk Usage Report Job
+      #   run: |
+      #     kubectl apply -f .github/manifests/disk-usage-report-job.yaml
+      #
+      # # TODO should timeout be longer?
+      # - name: Wait for Disk Usage Report Job to complete
+      #   run: |
+      #     kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=300s
+      #   continue-on-error: true
+      #
+      # - name: Retrieve generated report file
+      #   run: |
+      #     POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
+      #     kubectl cp $POD_NAME:/output/du_report.json du_report.json -n jupyterhub
+      #     cat du_report.json
+      #   continue-on-error: true
+      #
+      # # continue-on-error for previous steps so we delete the job
+      # - name: Delete Disk Usage Report Job
+      #   run: |
+      #     kubectl delete job disk-usage-report-job
+      #
       # - name: Clone dandi-hub-usage-reports repository
       #   run: |
       #     git clone https://github.com/dandi/dandi-hub-usage-reports.git

From 6ee89b2eeebb6732115abc2a0c47d5856f7930b2 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 14:55:31 -0600
Subject: [PATCH 22/39] change from job to pod

---
 .github/manifests/disk-usage-report-job.yaml | 51 ++++++--------
 .github/workflows/report.yaml                | 73 ++++++++++----------
 2 files changed, 59 insertions(+), 65 deletions(-)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index c7f60a4..36ded0f 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -1,32 +1,27 @@
-apiVersion: batch/v1
-kind: Job
+apiVersion: v1
+kind: Pod
 metadata:
   name: disk-usage-report-job
 spec:
-  template:
-    metadata:
-      labels:
-        app: disk-usage-report
-    spec:
-      containers:
-        - name: disk-usage-report
-          image: IMAGE_PLACEHOLDER
-          args:
-            - "/home/asmacdo"
-          volumeMounts:
-            - name: persistent-storage
-              mountPath: "/home/"
-      restartPolicy: Never
-      nodeSelector:
-        NodeGroupType: default
-        NodePool: default
-        hub.jupyter.org/node-purpose: user
-      tolerations:
-        - key: "hub.jupyter.org/dedicated"
-          operator: "Equal"
-          value: "user"
-          effect: "NoSchedule"
-      volumes:
+  containers:
+    - name: disk-usage-report
+      image: IMAGE_PLACEHOLDER
+      args:
+        - "/home/asmacdo"
+      volumeMounts:
         - name: persistent-storage
-          persistentVolumeClaim:
-            claimName: efs-persist
+          mountPath: "/home/"
+  restartPolicy: Never
+  nodeSelector:
+    NodeGroupType: default
+    NodePool: default
+    hub.jupyter.org/node-purpose: user
+  tolerations:
+    - key: "hub.jupyter.org/dedicated"
+      operator: "Equal"
+      value: "user"
+      effect: "NoSchedule"
+  volumes:
+    - name: persistent-storage
+      persistentVolumeClaim:
+        claimName: efs-persist
diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 5889c90..50168ff 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -50,51 +50,50 @@ jobs:
         run: |
           kubectl get pods -n jupyterhub
 
-      - name: Deploy Hello World Pod
-        run: |
-          kubectl apply -f .github/manifests/hello-world-pod.yaml
-
-      - name: Wait for Hello World Pod to complete
-        run: |
-          kubectl wait --for=condition=Ready pod/hello-world-pod --timeout=300s  # 5 minutes
-        continue-on-error: true  # Allow the workflow to continue even if this step fails
-
-      - name: Get Hello World Pod logs
-        run: |
-          kubectl logs hello-world-pod
-        if: ${{ success() }}  # Only run this step if the previous step was successful
-
-      - name: Delete Hello World Pod
-        run: |
-          kubectl delete pod hello-world-pod
-        if: ${{ always() }}  # Always run this step, even if other steps fail
-      #
-      # - name: Replace image placeholder in manifest
-      #   run: |
-      #     sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
-      #
-      # - name: Deploy Disk Usage Report Job
+      # - name: Deploy Hello World Pod
       #   run: |
-      #     kubectl apply -f .github/manifests/disk-usage-report-job.yaml
+      #     kubectl apply -f .github/manifests/hello-world-pod.yaml
       #
-      # # TODO should timeout be longer?
-      # - name: Wait for Disk Usage Report Job to complete
+      # - name: Wait for Hello World Pod to complete
       #   run: |
-      #     kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=300s
-      #   continue-on-error: true
+      #     kubectl wait --for=condition=Ready pod/hello-world-pod --timeout=300s  # 5 minutes
+      #   continue-on-error: true  # Allow the workflow to continue even if this step fails
       #
-      # - name: Retrieve generated report file
+      # - name: Get Hello World Pod logs
       #   run: |
-      #     POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
-      #     kubectl cp $POD_NAME:/output/du_report.json du_report.json -n jupyterhub
-      #     cat du_report.json
-      #   continue-on-error: true
+      #     kubectl logs hello-world-pod
+      #   if: ${{ success() }}  # Only run this step if the previous step was successful
       #
-      # # continue-on-error for previous steps so we delete the job
-      # - name: Delete Disk Usage Report Job
+      # - name: Delete Hello World Pod
       #   run: |
-      #     kubectl delete job disk-usage-report-job
+      #     kubectl delete pod hello-world-pod
+      #   if: ${{ always() }}  # Always run this step, even if other steps fail
       #
+      - name: Replace image placeholder in manifest
+        run: |
+          sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
+
+      - name: Deploy Disk Usage Report Job Pod
+        run: |
+          kubectl apply -f .github/manifests/disk-usage-report-job.yaml
+
+      # TODO should timeout be longer?
+      - name: Wait for Disk Usage Report Job to complete
+        run: |
+          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=300s
+        continue-on-error: true
+
+      - name: Retrieve generated report file
+        run: |
+          kubectl cp disk-usage-report-job:/output/du_report.json du_report.json
+          cat du_report.json
+        continue-on-error: true
+
+      # continue-on-error for previous steps so we delete the job
+      - name: Delete Disk Usage Report Job
+        run: |
+          kubectl delete pod disk-usage-report-job
+
       # - name: Clone dandi-hub-usage-reports repository
       #   run: |
       #     git clone https://github.com/dandi/dandi-hub-usage-reports.git

From a8f6ed30157424af1b90192a32c04ccb7ed4bf92 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 15:00:07 -0600
Subject: [PATCH 23/39] deploy pod to same namespace as pvc

---
 .github/manifests/disk-usage-report-job.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index 36ded0f..7a5e424 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -2,6 +2,7 @@ apiVersion: v1
 kind: Pod
 metadata:
   name: disk-usage-report-job
+  namespace: jupyterhub
 spec:
   containers:
     - name: disk-usage-report
@@ -25,3 +26,4 @@ spec:
     - name: persistent-storage
       persistentVolumeClaim:
         claimName: efs-persist
+

From 664853b13abef65e910667d3306fc6554cb0df13 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 15:04:29 -0600
Subject: [PATCH 24/39] Use ns in action

---
 .github/workflows/report.yaml | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 50168ff..de0ad87 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -80,19 +80,19 @@ jobs:
       # TODO should timeout be longer?
       - name: Wait for Disk Usage Report Job to complete
         run: |
-          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=300s
+          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=30s -n jupyterhub
         continue-on-error: true
 
       - name: Retrieve generated report file
         run: |
-          kubectl cp disk-usage-report-job:/output/du_report.json du_report.json
+          kubectl cp disk-usage-report-job:/output/du_report.json du_report.json -n jupyterhub
           cat du_report.json
         continue-on-error: true
 
       # continue-on-error for previous steps so we delete the job
       - name: Delete Disk Usage Report Job
         run: |
-          kubectl delete pod disk-usage-report-job
+          kubectl delete pod disk-usage-report-job -n jupyterhub
 
       # - name: Clone dandi-hub-usage-reports repository
       #   run: |

From e35c97439adc430f9fb1a113ed1db5ab2f8fc2ec Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 15:09:33 -0600
Subject: [PATCH 25/39] increase timeout to 60s

job shouldnt take that long, but this is wall time, includes docker pull, etc
---
 .github/workflows/report.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index de0ad87..4bcf1b4 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -80,7 +80,7 @@ jobs:
       # TODO should timeout be longer?
       - name: Wait for Disk Usage Report Job to complete
         run: |
-          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=30s -n jupyterhub
+          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=60s -n jupyterhub
         continue-on-error: true
 
       - name: Retrieve generated report file

From a8af5f278231575cac52a5575a4ada4fc89ab1a4 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 15:15:26 -0600
Subject: [PATCH 26/39] fixup: image name in manifest

---
 .github/workflows/report.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 4bcf1b4..170841f 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -71,7 +71,7 @@ jobs:
       #
       - name: Replace image placeholder in manifest
         run: |
-          sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/disk_usage_report:latest"'|' .github/manifests/disk-usage-report-job.yaml
+          sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/dandihub-report-generator:latest"'|' .github/manifests/disk-usage-report-job.yaml
 
       - name: Deploy Disk Usage Report Job Pod
         run: |

From 024cf6e08a55cdffde89435e181dd5b939d6dabd Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 15:18:53 -0600
Subject: [PATCH 27/39] increase timeout to 150

took almost 60 sec to start up
---
 .github/workflows/report.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 170841f..04b264c 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -80,7 +80,7 @@ jobs:
       # TODO should timeout be longer?
       - name: Wait for Disk Usage Report Job to complete
         run: |
-          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=60s -n jupyterhub
+          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=150s -n jupyterhub
         continue-on-error: true
 
       - name: Retrieve generated report file

From 49c346ea840b1606612482ae4e88c5f004aa8236 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 15:29:59 -0600
Subject: [PATCH 28/39] override entrypoint so i can debug with exec

---
 .github/manifests/disk-usage-report-job.yaml | 5 +++--
 .github/workflows/report.yaml                | 2 +-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index 7a5e424..e2e81a3 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -7,8 +7,9 @@ spec:
   containers:
     - name: disk-usage-report
       image: IMAGE_PLACEHOLDER
-      args:
-        - "/home/asmacdo"
+      command: ["/bin/sh", "-c", "sleep 300"]
+      # args:
+      #   - "/home/asmacdo"
       volumeMounts:
         - name: persistent-storage
           mountPath: "/home/"
diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 04b264c..34241ee 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -80,7 +80,7 @@ jobs:
       # TODO should timeout be longer?
       - name: Wait for Disk Usage Report Job to complete
         run: |
-          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=150s -n jupyterhub
+          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=360s -n jupyterhub
         continue-on-error: true
 
       - name: Retrieve generated report file

From 0191c858bd006e4608f08dea5ad0d90a5c1ba180 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 15:31:45 -0600
Subject: [PATCH 29/39] bound /home actually meant path was /home/home/asmacdo

---
 .github/manifests/disk-usage-report-job.yaml | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index e2e81a3..81ff08d 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -7,12 +7,12 @@ spec:
   containers:
     - name: disk-usage-report
       image: IMAGE_PLACEHOLDER
-      command: ["/bin/sh", "-c", "sleep 300"]
-      # args:
-      #   - "/home/asmacdo"
+      args:
+        - "/home/asmacdo"
       volumeMounts:
         - name: persistent-storage
-          mountPath: "/home/"
+          mountPath: "/home"
+          subPath: "home"
   restartPolicy: Never
   nodeSelector:
     NodeGroupType: default

From 3eb9157f6f8d15e4cb3615fdb9cdef112c88084b Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Fri, 8 Nov 2024 15:43:35 -0600
Subject: [PATCH 30/39] Create output dir prior to writing report

---
 .github/scripts/du.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/.github/scripts/du.py b/.github/scripts/du.py
index 12e0c0f..0b2ceb7 100755
--- a/.github/scripts/du.py
+++ b/.github/scripts/du.py
@@ -41,7 +41,7 @@ def prepare_report(directory):
     for user, data in report.items():
         data["disk_usage_human_readable"] = bytes_to_human_readable(data["disk_usage_bytes"])
 
-
+    os.makedirs(os.path.dirname(OUTPUT_FILE), exist_ok=True)
     with open(OUTPUT_FILE, 'w') as f:
         json.dump(report, f, indent=4)
 
@@ -55,6 +55,7 @@ def prepare_report(directory):
         path = sys.argv[1]
         directories = [d for d in os.listdir(path) if os.path.isdir(os.path.join(path, d))]
 
+        os.makedirs(os.path.dirname(OUTPUT_FILE), exist_ok=True)
         with open(OUTPUT_FILE, 'w') as f:
             f.write("\n".join(directories))
         # prepare_report(directory)

From 676a00e2e1d18cc0548443a3124af482669c14b7 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Mon, 11 Nov 2024 08:54:57 -0600
Subject: [PATCH 31/39] pod back to job

---
 .github/manifests/disk-usage-report-job.yaml | 54 +++++++++++---------
 .github/workflows/report.yaml                |  9 ++--
 2 files changed, 34 insertions(+), 29 deletions(-)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index 81ff08d..387aead 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -1,30 +1,34 @@
-apiVersion: v1
-kind: Pod
+apiVersion: v1/batch
+kind: Job
 metadata:
   name: disk-usage-report-job
   namespace: jupyterhub
 spec:
-  containers:
-    - name: disk-usage-report
-      image: IMAGE_PLACEHOLDER
-      args:
-        - "/home/asmacdo"
-      volumeMounts:
+  template:
+    metadata:
+      labels:
+        app: disk-usage-report
+    spec:
+      containers:
+        - name: disk-usage-report
+          image: IMAGE_PLACEHOLDER
+          args:
+            - "/home/asmacdo"
+          volumeMounts:
+            - name: persistent-storage
+              mountPath: "/home"
+              subPath: "home"
+      restartPolicy: Never
+      nodeSelector:
+        NodeGroupType: default
+        NodePool: default
+        hub.jupyter.org/node-purpose: user
+      tolerations:
+        - key: "hub.jupyter.org/dedicated"
+          operator: "Equal"
+          value: "user"
+          effect: "NoSchedule"
+      volumes:
         - name: persistent-storage
-          mountPath: "/home"
-          subPath: "home"
-  restartPolicy: Never
-  nodeSelector:
-    NodeGroupType: default
-    NodePool: default
-    hub.jupyter.org/node-purpose: user
-  tolerations:
-    - key: "hub.jupyter.org/dedicated"
-      operator: "Equal"
-      value: "user"
-      effect: "NoSchedule"
-  volumes:
-    - name: persistent-storage
-      persistentVolumeClaim:
-        claimName: efs-persist
-
+          persistentVolumeClaim:
+            claimName: efs-persist
diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 34241ee..1c2692a 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -73,26 +73,27 @@ jobs:
         run: |
           sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/dandihub-report-generator:latest"'|' .github/manifests/disk-usage-report-job.yaml
 
-      - name: Deploy Disk Usage Report Job Pod
+      - name: Deploy Disk Usage Report Job
         run: |
           kubectl apply -f .github/manifests/disk-usage-report-job.yaml
 
       # TODO should timeout be longer?
       - name: Wait for Disk Usage Report Job to complete
         run: |
-          kubectl wait --for=condition=complete pod/disk-usage-report-job --timeout=360s -n jupyterhub
+          kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=360s -n jupyterhub
         continue-on-error: true
 
       - name: Retrieve generated report file
         run: |
-          kubectl cp disk-usage-report-job:/output/du_report.json du_report.json -n jupyterhub
+          POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
+          kubectl cp $POD_NAME:/output/du_report.json du_report.json -n jupyterhub
           cat du_report.json
         continue-on-error: true
 
       # continue-on-error for previous steps so we delete the job
       - name: Delete Disk Usage Report Job
         run: |
-          kubectl delete pod disk-usage-report-job -n jupyterhub
+          kubectl delete job disk-usage-report-job -n jupyterhub
 
       # - name: Clone dandi-hub-usage-reports repository
       #   run: |

From c085751ef1440b4af9ce6c61af48c266a1db27e5 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Mon, 11 Nov 2024 09:18:16 -0600
Subject: [PATCH 32/39] Fixup use the correct job api

---
 .github/manifests/disk-usage-report-job.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index 387aead..ed2bcde 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -1,4 +1,4 @@
-apiVersion: v1/batch
+apiVersion: batch/v1
 kind: Job
 metadata:
   name: disk-usage-report-job

From 3e18a379b1f13835913c7726e48dee35af92a429 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Mon, 11 Nov 2024 09:27:49 -0600
Subject: [PATCH 33/39] Add namespace to pod retrieval

---
 .github/workflows/report.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 1c2692a..0cbe18d 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -85,7 +85,7 @@ jobs:
 
       - name: Retrieve generated report file
         run: |
-          POD_NAME=$(kubectl get pods --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
+          POD_NAME=$(kubectl get pods -n jupyterhub --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
           kubectl cp $POD_NAME:/output/du_report.json du_report.json -n jupyterhub
           cat du_report.json
         continue-on-error: true

From 0fa5ece95edf5a7e3b82fb5cf0b5ce2fe90835e0 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Mon, 11 Nov 2024 10:38:43 -0600
Subject: [PATCH 34/39] write directly to pv to test job

---
 .github/scripts/du.py         | 19 +++++++++++--------
 .github/workflows/report.yaml |  7 -------
 2 files changed, 11 insertions(+), 15 deletions(-)

diff --git a/.github/scripts/du.py b/.github/scripts/du.py
index 0b2ceb7..1c3a403 100755
--- a/.github/scripts/du.py
+++ b/.github/scripts/du.py
@@ -4,8 +4,9 @@
 import subprocess
 import sys
 import json
+from datetime import date
 
-OUTPUT_FILE = "/output/du_report.json"
+OUTPUT_DIR = "/home/asmacdo/du_reports/"
 SIZE_THRESHOLD_GB = 1
 SIZE_THRESHOLD_BYTES = SIZE_THRESHOLD_GB * 1024 * 1024 * 1024
 
@@ -41,11 +42,12 @@ def prepare_report(directory):
     for user, data in report.items():
         data["disk_usage_human_readable"] = bytes_to_human_readable(data["disk_usage_bytes"])
 
-    os.makedirs(os.path.dirname(OUTPUT_FILE), exist_ok=True)
-    with open(OUTPUT_FILE, 'w') as f:
-        json.dump(report, f, indent=4)
-
-    print(f"Disk usage report generated at {os.path.abspath(OUTPUT_FILE)}")
+    # os.makedirs(os.path.dirname(OUTPUT_FILE), exist_ok=True)
+    # output_file =
+    # with open(OUTPUT_FILE, 'w') as f:
+    #     json.dump(report, f, indent=4)
+    #
+    # print(f"Disk usage report generated at {os.path.abspath(OUTPUT_FILE)}")
 
 
 if __name__ == "__main__":
@@ -55,7 +57,8 @@ def prepare_report(directory):
         path = sys.argv[1]
         directories = [d for d in os.listdir(path) if os.path.isdir(os.path.join(path, d))]
 
-        os.makedirs(os.path.dirname(OUTPUT_FILE), exist_ok=True)
-        with open(OUTPUT_FILE, 'w') as f:
+        os.makedirs(os.path.dirname(OUTPUT_DIR), exist_ok=True)
+        current_date = date.today().strftime('%Y-%m-%d')
+        with open(f"OUTPUT_DIR/{current_date}.json", "w") as f:
             f.write("\n".join(directories))
         # prepare_report(directory)
diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index 0cbe18d..acb9dc1 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -83,13 +83,6 @@ jobs:
           kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=360s -n jupyterhub
         continue-on-error: true
 
-      - name: Retrieve generated report file
-        run: |
-          POD_NAME=$(kubectl get pods -n jupyterhub --selector=job-name=disk-usage-report-job -o jsonpath='{.items[0].metadata.name}')
-          kubectl cp $POD_NAME:/output/du_report.json du_report.json -n jupyterhub
-          cat du_report.json
-        continue-on-error: true
-
       # continue-on-error for previous steps so we delete the job
       - name: Delete Disk Usage Report Job
         run: |

From e1ecbc33b8b820e68d3fc3d3242cebd4137a4b35 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Mon, 11 Nov 2024 10:47:13 -0600
Subject: [PATCH 35/39] fixup script fstring

---
 .github/scripts/du.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/scripts/du.py b/.github/scripts/du.py
index 1c3a403..05043ef 100755
--- a/.github/scripts/du.py
+++ b/.github/scripts/du.py
@@ -59,6 +59,6 @@ def prepare_report(directory):
 
         os.makedirs(os.path.dirname(OUTPUT_DIR), exist_ok=True)
         current_date = date.today().strftime('%Y-%m-%d')
-        with open(f"OUTPUT_DIR/{current_date}.json", "w") as f:
+        with open(f"{OUTPUT_DIR}/{current_date}.json", "w") as f:
             f.write("\n".join(directories))
         # prepare_report(directory)

From 082d3ccfe348885d7adfaeec4bbfb2682c7af4c3 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Mon, 11 Nov 2024 10:47:37 -0600
Subject: [PATCH 36/39] no retry on failure, we were spinning up 5 pods, lets
 just fail 1 time

---
 .github/manifests/disk-usage-report-job.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index ed2bcde..b487280 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -9,6 +9,7 @@ spec:
       labels:
         app: disk-usage-report
     spec:
+      backoffLimit: 0  # No retry on failure
       containers:
         - name: disk-usage-report
           image: IMAGE_PLACEHOLDER

From d46ea446adaffc63c6cb14b4ea1f33ba660467dc Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Mon, 11 Nov 2024 10:49:22 -0600
Subject: [PATCH 37/39] Fixup backup limit job not template

---
 .github/manifests/disk-usage-report-job.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index b487280..adf966e 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -4,12 +4,12 @@ metadata:
   name: disk-usage-report-job
   namespace: jupyterhub
 spec:
+  backoffLimit: 0  # No retry on failure
   template:
     metadata:
       labels:
         app: disk-usage-report
     spec:
-      backoffLimit: 0  # No retry on failure
       containers:
         - name: disk-usage-report
           image: IMAGE_PLACEHOLDER

From 965a81eb6374077f8657066761f8808c7e7d08f3 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Mon, 11 Nov 2024 10:52:54 -0600
Subject: [PATCH 38/39] Initial report

---
 .github/manifests/disk-usage-report-job.yaml |  2 +-
 .github/scripts/du.py                        | 19 ++++++-------------
 2 files changed, 7 insertions(+), 14 deletions(-)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index adf966e..488735e 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -14,7 +14,7 @@ spec:
         - name: disk-usage-report
           image: IMAGE_PLACEHOLDER
           args:
-            - "/home/asmacdo"
+            - "/home/"
           volumeMounts:
             - name: persistent-storage
               mountPath: "/home"
diff --git a/.github/scripts/du.py b/.github/scripts/du.py
index 05043ef..e626769 100755
--- a/.github/scripts/du.py
+++ b/.github/scripts/du.py
@@ -42,12 +42,11 @@ def prepare_report(directory):
     for user, data in report.items():
         data["disk_usage_human_readable"] = bytes_to_human_readable(data["disk_usage_bytes"])
 
-    # os.makedirs(os.path.dirname(OUTPUT_FILE), exist_ok=True)
-    # output_file =
-    # with open(OUTPUT_FILE, 'w') as f:
-    #     json.dump(report, f, indent=4)
-    #
-    # print(f"Disk usage report generated at {os.path.abspath(OUTPUT_FILE)}")
+    os.makedirs(os.path.dirname(OUTPUT_DIR), exist_ok=True)
+    current_date = date.today().strftime('%Y-%m-%d')
+    with open(f"{OUTPUT_DIR}/{current_date}.json", "w") as f:
+        json.dump(report, f, indent=4)
+    print(f"Disk usage report generated at {os.path.abspath(OUTPUT_FILE)}")
 
 
 if __name__ == "__main__":
@@ -55,10 +54,4 @@ def prepare_report(directory):
         print("Usage: du.py <directory_to_check>")
     else:
         path = sys.argv[1]
-        directories = [d for d in os.listdir(path) if os.path.isdir(os.path.join(path, d))]
-
-        os.makedirs(os.path.dirname(OUTPUT_DIR), exist_ok=True)
-        current_date = date.today().strftime('%Y-%m-%d')
-        with open(f"{OUTPUT_DIR}/{current_date}.json", "w") as f:
-            f.write("\n".join(directories))
-        # prepare_report(directory)
+        prepare_report(path)

From 7366d2d318a2b38496d552fcb050be8c86914155 Mon Sep 17 00:00:00 2001
From: Austin Macdonald <austin@dartmouth.edu>
Date: Mon, 11 Nov 2024 12:35:01 -0600
Subject: [PATCH 39/39] disable report

see PR for comment
---
 .github/manifests/disk-usage-report-job.yaml |   2 +-
 .github/workflows/report.yaml                | 208 +++++++++----------
 2 files changed, 105 insertions(+), 105 deletions(-)

diff --git a/.github/manifests/disk-usage-report-job.yaml b/.github/manifests/disk-usage-report-job.yaml
index 488735e..161f177 100644
--- a/.github/manifests/disk-usage-report-job.yaml
+++ b/.github/manifests/disk-usage-report-job.yaml
@@ -12,7 +12,7 @@ spec:
     spec:
       containers:
         - name: disk-usage-report
-          image: IMAGE_PLACEHOLDER
+          image: dandiarchive/dandihub-report-generator:latest
           args:
             - "/home/"
           volumeMounts:
diff --git a/.github/workflows/report.yaml b/.github/workflows/report.yaml
index acb9dc1..8424fd6 100644
--- a/.github/workflows/report.yaml
+++ b/.github/workflows/report.yaml
@@ -1,104 +1,104 @@
-name: Generate Data Usage Report
-
-on:
-  pull_request:
-    branches:
-      - main
-
-jobs:
-  generate_data_usage_report:
-    runs-on: ubuntu-latest
-
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v3
-
-      - name: Log in to DockerHub
-        uses: docker/login-action@v2
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_TOKEN }}
-
-      - name: Build and push Docker image
-        uses: docker/build-push-action@v3
-        with:
-          context: .
-          file: images/Dockerfile.dandihub_report_generator
-          push: true
-          tags: ${{ secrets.DOCKERHUB_USERNAME }}/dandihub-report-generator:latest
-
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v3
-        with:
-          aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
-          aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
-          aws-region: us-east-2
-
-      - name: Assume ProvisioningRole
-        run: |
-          CREDS=$(aws sts assume-role --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }} --role-session-name "GitHubActionsSession")
-          export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
-          export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
-          export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
-
-      - name: Configure kubectl with AWS EKS
-        run: |
-          aws eks update-kubeconfig --name eks-dandihub --region us-east-2 --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }}
-
-      # TODO remove
-      - name: Sanity check
-        run: |
-          kubectl get pods -n jupyterhub
-
-      # - name: Deploy Hello World Pod
-      #   run: |
-      #     kubectl apply -f .github/manifests/hello-world-pod.yaml
-      #
-      # - name: Wait for Hello World Pod to complete
-      #   run: |
-      #     kubectl wait --for=condition=Ready pod/hello-world-pod --timeout=300s  # 5 minutes
-      #   continue-on-error: true  # Allow the workflow to continue even if this step fails
-      #
-      # - name: Get Hello World Pod logs
-      #   run: |
-      #     kubectl logs hello-world-pod
-      #   if: ${{ success() }}  # Only run this step if the previous step was successful
-      #
-      # - name: Delete Hello World Pod
-      #   run: |
-      #     kubectl delete pod hello-world-pod
-      #   if: ${{ always() }}  # Always run this step, even if other steps fail
-      #
-      - name: Replace image placeholder in manifest
-        run: |
-          sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/dandihub-report-generator:latest"'|' .github/manifests/disk-usage-report-job.yaml
-
-      - name: Deploy Disk Usage Report Job
-        run: |
-          kubectl apply -f .github/manifests/disk-usage-report-job.yaml
-
-      # TODO should timeout be longer?
-      - name: Wait for Disk Usage Report Job to complete
-        run: |
-          kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=360s -n jupyterhub
-        continue-on-error: true
-
-      # continue-on-error for previous steps so we delete the job
-      - name: Delete Disk Usage Report Job
-        run: |
-          kubectl delete job disk-usage-report-job -n jupyterhub
-
-      # - name: Clone dandi-hub-usage-reports repository
-      #   run: |
-      #     git clone https://github.com/dandi/dandi-hub-usage-reports.git
-      #
-      # - name: Copy report file to repository, commit and push report
-      #   run: |
-      #     cd dandi-hub-usage-reports
-      #     DATE=$(date +'%Y-%m-%d')
-      #     mv ../du_report.json $DATE_du_report.json
-      #     git config --global user.name "GitHub Actions"
-      #     git config --global user.email "actions@github.com"
-      #     git add $DATE_du_report.json
-      #     git commit -m "Add disk usage report for $DATE"
-      #     git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git
+# name: Generate Data Usage Report
+#
+# on:
+#   pull_request:
+#     branches:
+#       - main
+#
+# jobs:
+#   generate_data_usage_report:
+#     runs-on: ubuntu-latest
+#
+#     steps:
+#       - name: Checkout code
+#         uses: actions/checkout@v3
+#
+#       - name: Log in to DockerHub
+#         uses: docker/login-action@v2
+#         with:
+#           username: ${{ secrets.DOCKERHUB_USERNAME }}
+#           password: ${{ secrets.DOCKERHUB_TOKEN }}
+#
+#       - name: Build and push Docker image
+#         uses: docker/build-push-action@v3
+#         with:
+#           context: .
+#           file: images/Dockerfile.dandihub_report_generator
+#           push: true
+#           tags: ${{ secrets.DOCKERHUB_USERNAME }}/dandihub-report-generator:latest
+#
+#       - name: Configure AWS credentials
+#         uses: aws-actions/configure-aws-credentials@v3
+#         with:
+#           aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
+#           aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+#           aws-region: us-east-2
+#
+#       - name: Assume ProvisioningRole
+#         run: |
+#           CREDS=$(aws sts assume-role --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }} --role-session-name "GitHubActionsSession")
+#           export AWS_ACCESS_KEY_ID=$(echo $CREDS | jq -r '.Credentials.AccessKeyId')
+#           export AWS_SECRET_ACCESS_KEY=$(echo $CREDS | jq -r '.Credentials.SecretAccessKey')
+#           export AWS_SESSION_TOKEN=$(echo $CREDS | jq -r '.Credentials.SessionToken')
+#
+#       - name: Configure kubectl with AWS EKS
+#         run: |
+#           aws eks update-kubeconfig --name eks-dandihub --region us-east-2 --role-arn ${{ secrets.AWS_PROVISIONING_ROLE_ARN }}
+#
+#       # TODO remove
+#       - name: Sanity check
+#         run: |
+#           kubectl get pods -n jupyterhub
+#
+#       # - name: Deploy Hello World Pod
+#       #   run: |
+#       #     kubectl apply -f .github/manifests/hello-world-pod.yaml
+#       #
+#       # - name: Wait for Hello World Pod to complete
+#       #   run: |
+#       #     kubectl wait --for=condition=Ready pod/hello-world-pod --timeout=300s  # 5 minutes
+#       #   continue-on-error: true  # Allow the workflow to continue even if this step fails
+#       #
+#       # - name: Get Hello World Pod logs
+#       #   run: |
+#       #     kubectl logs hello-world-pod
+#       #   if: ${{ success() }}  # Only run this step if the previous step was successful
+#       #
+#       # - name: Delete Hello World Pod
+#       #   run: |
+#       #     kubectl delete pod hello-world-pod
+#       #   if: ${{ always() }}  # Always run this step, even if other steps fail
+#       #
+#       - name: Replace image placeholder in manifest
+#         run: |
+#           sed -i 's|IMAGE_PLACEHOLDER|'"${{ secrets.DOCKERHUB_USERNAME }}/dandihub-report-generator:latest"'|' .github/manifests/disk-usage-report-job.yaml
+#
+#       - name: Deploy Disk Usage Report Job
+#         run: |
+#           kubectl apply -f .github/manifests/disk-usage-report-job.yaml
+#
+#       # TODO should timeout be longer?
+#       - name: Wait for Disk Usage Report Job to complete
+#         run: |
+#           kubectl wait --for=condition=complete job/disk-usage-report-job --timeout=360s -n jupyterhub
+#         continue-on-error: true
+#
+#       # continue-on-error for previous steps so we delete the job
+#       - name: Delete Disk Usage Report Job
+#         run: |
+#           kubectl delete job disk-usage-report-job -n jupyterhub
+#
+#       # - name: Clone dandi-hub-usage-reports repository
+#       #   run: |
+#       #     git clone https://github.com/dandi/dandi-hub-usage-reports.git
+#       #
+#       # - name: Copy report file to repository, commit and push report
+#       #   run: |
+#       #     cd dandi-hub-usage-reports
+#       #     DATE=$(date +'%Y-%m-%d')
+#       #     mv ../du_report.json $DATE_du_report.json
+#       #     git config --global user.name "GitHub Actions"
+#       #     git config --global user.email "actions@github.com"
+#       #     git add $DATE_du_report.json
+#       #     git commit -m "Add disk usage report for $DATE"
+#       #     git push https://${{ secrets.GITHUB_TOKEN }}@github.com/dandi/dandi-hub-usage-reports.git