Quick Start Guide

Get IBM GCM 2.0.1 up and running in minutes

For the Impatient

Choose your deployment method and get started immediately.

Single Disk VM (Fastest - 5 minutes)

Recommended for: Quick testing, development environments, or when you don't have a second disk available.
1

Download and Run K3s Installation

bash 
curl -O http://acefs01.ace.ibm.aessatl.arrow.com/downloads/gcm_2_0_1/install_singlenode_k3s_localpath.sh
chmod +x install_singlenode_k3s_localpath.sh
./install_singlenode_k3s_localpath.sh | tee k3s_install.log
2

Set Environment

bash 
export KUBECONFIG=/etc/rancher/k3s/k3s.yaml
3

Download GCM Package

bash 
cd /root
curl -O http://acefs01.ace.ibm.aessatl.arrow.com/downloads/gcm_2_0_1/Grd_Crgrphy_Mngr_2.0.1_Linux_En.tgz
4

Install GCM

bash 
curl -O http://acefs01.ace.ibm.aessatl.arrow.com/downloads/gcm_2_0_1/install_singlenode_gcm.sh
chmod +x install_singlenode_gcm.sh
./install_singlenode_gcm.sh | tee gcm_install.log
Done! Check the installer output for your GCM URL and credentials.

Two Disk VM (With Ceph - 25 minutes)

Recommended for: Production environments requiring high-availability storage with Rook Ceph.
1

Verify Raw Disk

bash 
lsblk

Look for a disk with no FSTYPE (e.g., /dev/sdb)

2

Download and Run K3s with Ceph

bash 
curl -O http://acefs01.ace.ibm.aessatl.arrow.com/downloads/gcm_2_0_1/install_singlenode_k3s_ceph.sh
chmod +x install_singlenode_k3s_ceph.sh
./install_singlenode_k3s_ceph.sh | tee k3s_ceph_install.log
3

Set Environment

bash 
export KUBECONFIG=/etc/rancher/k3s/k3s.yaml
4

Wait for Ceph to be Healthy

bash 
watch kubectl get cephcluster -n rook-ceph

Wait for Phase: Ready and Health: HEALTH_OK

5

Download GCM Package

bash 
cd /root
curl -O http://acefs01.ace.ibm.aessatl.arrow.com/downloads/gcm_2_0_1/Grd_Crgrphy_Mngr_2.0.1_Linux_En.tgz
6

Install GCM

bash 
curl -O http://acefs01.ace.ibm.aessatl.arrow.com/downloads/gcm_2_0_1/install_singlenode_gcm.sh
chmod +x install_singlenode_gcm.sh
./install_singlenode_gcm.sh | tee gcm_install.log
Done! Check the installer output for your GCM URL and credentials.

Troubleshooting One-Liners

Quick commands to check system status and diagnose issues.

Check K3s Status

bash 
kubectl get nodes && kubectl get pods -A

Check Storage

bash 
kubectl get storageclass && kubectl get pvc -A

Check Ceph (if using)

bash 
kubectl get cephcluster -n rook-ceph && kubectl get pods -n rook-ceph

Check GCM

bash 
kubectl get pods -A | grep gcm

View Logs

bash 
journalctl -u k3s -n 100 --no-pager

Reset and Start Over

bash 
/usr/local/bin/k3s-uninstall.sh

Common Issues

"No raw disks found"

Solution:

  • You need a second disk for Ceph
  • Use install_singlenode_k3s_localpath.sh instead

"Ceph cluster stuck in Progressing"

Solution:

  • Check logs: kubectl logs -n rook-ceph -l app=rook-ceph-operator --tail=100
  • May need to wait 5-10 minutes for first-time setup

"Storage validation failed"

Solution:

  • Check: kubectl get storageclass
  • Ensure one is marked (default)

"kubectl: command not found"

Solution:

export PATH="/usr/local/bin:$PATH"
export KUBECONFIG=/etc/rancher/k3s/k3s.yaml

What Gets Installed

K3s (local-path)

  • K3s v1.30.5+k3s1
  • Local-path storage provisioner
  • Flannel CNI
  • CoreDNS
  • Metrics server

K3s (Ceph)

Everything in local-path, plus:

  • Rook Ceph operator v1.14.9
  • Ceph cluster (1 MON, 1 MGR, OSDs)
  • CephFS filesystem
  • CSI drivers

GCM Components

  • GCM application pods
  • Database (PostgreSQL/MongoDB)
  • Web interface
  • API services

Next Steps After Installation

1

Save Credentials

Copy the GCM URL and admin credentials from the installer output

2

Create VM Snapshot

Take a snapshot of your VM for easy backup and recovery

3

Test GCM Access

Log in to the GCM web interface and verify functionality

4

Configure Monitoring

Set up monitoring and alerts for your GCM deployment

5

Integrate with Vault

Connect GCM to HashiCorp Vault for secrets management

Need More Details?

See the Complete Installation Guide for:

  • Detailed workflows and explanations
  • Advanced troubleshooting techniques
  • Architecture decisions and best practices
  • Lessons learned from production deployments