GitHubHelp

rajusem/basic-workload

Create basic workload for OpenShift Clusters

★ 0Forks 0GitHub ↗Compare

README

Offline Workloads for OpenShift

These workloads run entirely within your OpenShift cluster with no external dependencies. They use only the busybox image which is typically cached in most clusters.

Features

  • ✅ 100% Offline - No internet access required
  • ✅ Lightweight - Uses only busybox:latest image
  • ✅ Simple - Easy to deploy and manage
  • ✅ Resource Efficient - Configurable CPU and memory limits
  • ✅ Self-Contained - All files and documentation in this directory

Available Workloads

  1. Simple CPU Workload (workloads/simple-cpu-workload.yaml)

    • CPU-intensive loops
    • Resources: CPU 100m-500m, Memory 64Mi-128Mi
    • Duration: 15 minutes

  2. Simple Memory Workload (workloads/simple-memory-workload.yaml)

    • Memory allocation and usage
    • Resources: CPU 50m-200m, Memory 128Mi-512Mi
    • Duration: 15 minutes

  3. File I/O Workload (workloads/file-io-workload.yaml)

    • File read/write operations
    • Resources: CPU 50m-200m, Memory 64Mi-128Mi
    • Duration: 15 minutes

  4. Network Workload (workloads/network-workload.yaml)

    • Local network connections
    • Resources: CPU 50m-200m, Memory 64Mi-128Mi
    • Duration: 15 minutes

  5. Combined Workload (workloads/combined-workload.yaml)

    • CPU, memory, and I/O simultaneously
    • Resources: CPU 100m-500m, Memory 128Mi-512Mi
    • Duration: 15 minutes

Quick Start

Option 1: Complete Setup Script (Recommended)

chmod +x complete-setup.sh
./complete-setup.sh

Note: Run all commands from the workloads repository directory.

This script will:

  • ✅ Check if you're logged in to OpenShift
  • ✅ Create the namespace (offline-workload)
  • ✅ Deploy all CronJobs
  • ✅ Verify deployment

Option 2: Deploy All Workloads

./deploy-offline-workloads.sh

Option 3: Deploy Individual Workloads

# Create namespace
oc create namespace offline-workload

# Deploy CPU workload
oc apply -f workloads/simple-cpu-workload.yaml

# Deploy Memory workload
oc apply -f workloads/simple-memory-workload.yaml

# Deploy File I/O workload
oc apply -f workloads/file-io-workload.yaml

# Deploy Network workload
oc apply -f workloads/network-workload.yaml

# Deploy Combined workload
oc apply -f workloads/combined-workload.yaml

Run Workloads

Using Scripts

# Run all workloads
./run-workload.sh all

Run specific workload

./run-workload.sh cpu ./run-workload.sh memory ./run-workload.sh io ./run-workload.sh network ./run-workload.sh combined


### Manual CLI Commands

```bash
# CPU workload
oc create job --from=cronjob/simple-cpu-workload cpu-test-$(date +%s) -n offline-workload

# Memory workload
oc create job --from=cronjob/simple-memory-workload memory-test-$(date +%s) -n offline-workload

# File I/O workload
oc create job --from=cronjob/file-io-workload io-test-$(date +%s) -n offline-workload

# Network workload
oc create job --from=cronjob/network-workload network-test-$(date +%s) -n offline-workload

# Combined workload
oc create job --from=cronjob/combined-workload combined-test-$(date +%s) -n offline-workload

Monitor Workloads

Using Scripts

./status.sh

Manual CLI Commands

# View CronJobs
oc get cronjobs -n offline-workload

# View Jobs
oc get jobs -n offline-workload

# View Pods
oc get pods -n offline-workload

# View resource usage
oc top pods -n offline-workload

# View logs
oc logs -f -n offline-workload -l app=simple-cpu

Modify CPU and Memory Values

Method 1: Edit YAML Files (Recommended)

  1. Open the workload YAML file (e.g., workloads/simple-cpu-workload.yaml)
  2. Locate the resources section:
resources:
  requests:
    cpu: "100m"      # Change this value
    memory: "64Mi"  # Change this value
  limits:
    cpu: "500m"      # Change this value
    memory: "128Mi"  # Change this value

  1. Modify the values as needed:

    • CPU values: Use m for millicores (e.g., "100m" = 0.1 cores, "500m" = 0.5 cores)
    • Memory values: Use Mi for megabytes or Gi for gigabytes (e.g., "64Mi", "128Mi", "1Gi")

  2. Apply the changes:

oc apply -f workloads/simple-cpu-workload.yaml

Method 2: Use oc patch Command

# Update CPU request
oc patch cronjob simple-cpu-workload -n offline-workload --type='json' -p='[
  {"op": "replace", "path": "/spec/jobTemplate/spec/template/spec/containers/0/resources/requests/cpu", "value": "200m"}
]'

# Update CPU limit
oc patch cronjob simple-cpu-workload -n offline-workload --type='json' -p='[
  {"op": "replace", "path": "/spec/jobTemplate/spec/template/spec/containers/0/resources/limits/cpu", "value": "1000m"}
]'

# Update memory request
oc patch cronjob simple-cpu-workload -n offline-workload --type='json' -p='[
  {"op": "replace", "path": "/spec/jobTemplate/spec/template/spec/containers/0/resources/requests/memory", "value": "128Mi"}
]'

# Update memory limit
oc patch cronjob simple-cpu-workload -n offline-workload --type='json' -p='[
  {"op": "replace", "path": "/spec/jobTemplate/spec/template/spec/containers/0/resources/limits/memory", "value": "256Mi"}
]'

Method 3: Use oc edit Command

oc edit cronjob simple-cpu-workload -n offline-workload

This opens the CronJob in your default editor. Find the resources section and modify the values, then save and exit.

Resource Value Examples

Low Resource Configuration:

resources:
  requests:
    cpu: "50m"
    memory: "32Mi"
  limits:
    cpu: "200m"
    memory: "64Mi"

Medium Resource Configuration (Default):

resources:
  requests:
    cpu: "100m"
    memory: "64Mi"
  limits:
    cpu: "500m"
    memory: "128Mi"

High Resource Configuration:

resources:
  requests:
    cpu: "200m"
    memory: "128Mi"
  limits:
    cpu: "1000m"
    memory: "256Mi"

Stop Workloads

Suspend CronJobs (Stop Scheduling New Jobs)

# Suspend all CronJobs
oc patch cronjob simple-cpu-workload -n offline-workload -p '{"spec":{"suspend":true}}'
oc patch cronjob simple-memory-workload -n offline-workload -p '{"spec":{"suspend":true}}'
oc patch cronjob file-io-workload -n offline-workload -p '{"spec":{"suspend":true}}'
oc patch cronjob network-workload -n offline-workload -p '{"spec":{"suspend":true}}'
oc patch cronjob combined-workload -n offline-workload -p '{"spec":{"suspend":true}}'

# Resume CronJobs
oc patch cronjob simple-cpu-workload -n offline-workload -p '{"spec":{"suspend":false}}'
oc patch cronjob simple-memory-workload -n offline-workload -p '{"spec":{"suspend":false}}'
oc patch cronjob file-io-workload -n offline-workload -p '{"spec":{"suspend":false}}'
oc patch cronjob network-workload -n offline-workload -p '{"spec":{"suspend":false}}'
oc patch cronjob combined-workload -n offline-workload -p '{"spec":{"suspend":false}}'

Delete Running Jobs

# Delete all jobs
oc delete jobs --all -n offline-workload

# Or delete specific job
oc delete job <job-name> -n offline-workload

Clean Up Everything

./cleanup.sh

# Or manually
oc delete cronjobs --all -n offline-workload
oc delete jobs --all -n offline-workload
oc delete pods --all -n offline-workload
oc delete namespace offline-workload

Schedule

All workloads are scheduled to run every 2 hours and each job runs for 15 minutes.

To change the schedule, edit the schedule field in the YAML files:

  • "0 */2 * * *" - Every 2 hours (current)
  • "0 * * * *" - Every hour
  • "*/30 * * * *" - Every 30 minutes
  • "0 9 * * 1-5" - Every weekday at 9 AM

Customization

Change Resource Limits

Edit the resources section in any YAML file:

resources:
  requests:
    cpu: "100m"
    memory: "64Mi"
  limits:
    cpu: "500m"
    memory: "128Mi"

Change Duration

Edit the timeout value in the command section:

timeout 900  # 900 seconds = 15 minutes
timeout 600  # 600 seconds = 10 minutes

Change CPU Intensity

For CPU workloads, edit the loop count in the command section:

while [ $i -lt 100000 ]; do  # Increase 100000 for more CPU stress

Change Memory Allocation Size

For memory workloads, edit the MEM_SIZE variable:

MEM_SIZE=256  # Change this value (in MB)

Monitoring

These workloads generate metrics that can be observed in your observability stack:

  • CPU Metrics: CPU usage, throttling
  • Memory Metrics: Memory usage, RSS, working set
  • I/O Metrics: Read/write operations, disk usage
  • Network Metrics: Connection counts, bandwidth

Documentation Files

This directory contains:

  • README.md (this file) - Overview and quick start
  • START-HERE.md - Quick start guide from login
  • CLI-COMMANDS.md - Complete CLI commands reference
  • QUICK-REFERENCE.md - Quick command reference
  • complete-setup.sh - Complete setup script
  • deploy-offline-workloads.sh - Deploy workloads script
  • run-workload.sh - Run workloads script
  • status.sh - Check status script
  • cleanup.sh - Clean up script
  • workloads/*.yaml - Workload CronJob definitions

Troubleshooting

Jobs not starting

# Check CronJob status
oc describe cronjob <cronjob-name> -n offline-workload

# Check for resource quotas
oc describe quota -n offline-workload

# Check if CronJob is suspended
oc get cronjob <cronjob-name> -n offline-workload -o jsonpath='{.spec.suspend}'

Pods failing

# Check pod events
oc describe pod <pod-name> -n offline-workload

# Check pod logs
oc logs <pod-name> -n offline-workload

Image pull errors

The workloads use busybox:latest which should be available in most clusters. If you get image pull errors:

# Check if image is available
oc run test-pod --image=busybox:latest --rm -it --restart=Never -n offline-workload -- /bin/sh

Insufficient resources

# Check available resources
oc describe nodes
oc top nodes

# Reduce resource requests/limits in the YAML files

Notes

  • All workloads use the busybox:latest image
  • Workloads run entirely within the cluster
  • No external network access required
  • Suitable for air-gapped environments
  • Minimal resource footprint
  • All files and documentation are self-contained in this directory

Contributors

dvandra

Issues