Skip to content

SparshSunilNaik/Openstack-Setup-3-nodes

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
configs
configs
 
 
scripts
scripts
 
 
templates
templates
 
 
.gitignore
.gitignore
 
 
README.md
README.md
 
 

Repository files navigation

OpenStack Yoga 3-Node Deployment Scripts

Automated installation scripts for deploying OpenStack Yoga on a 3-node architecture:

  • Controller Node: Identity (Keystone), Image (Glance), Compute API (Nova), Network Server (Neutron), Dashboard (Horizon)
  • Network/Neutron Node: OVS Agent, L3 Agent, DHCP Agent, Metadata Agent
  • Compute Node: Nova Compute, Neutron OVS Agent, KVM/QEMU/Libvirt

📁 Project Structure

openstack-yoga-3node/
├── configs/
│   └── cluster.env          # Main configuration file (EDIT THIS FIRST!)
├── scripts/
│   ├── controller.sh         # Controller node installation
│   ├── neutron.sh            # Network node installation
│   └── compute.sh            # Compute node installation
├── templates/                 # Configuration templates
│   ├── keystone.conf.template
│   ├── glance-api.conf.template
│   ├── nova-controller.conf.template
│   ├── nova-compute.conf.template
│   ├── neutron.conf.template
│   ├── ml2_conf.ini.template
│   ├── openvswitch_agent.ini.template
│   ├── l3_agent.ini.template
│   ├── dhcp_agent.ini.template
│   ├── metadata_agent.ini.template
│   └── ...
├── logs/                      # Installation logs (auto-created)
└── README.md                  # This file

🔧 Prerequisites

Hardware Requirements (per node, minimum)

  • CPU: 2 cores (with VT-x/AMD-V for compute)
  • RAM: 4 GB (8 GB recommended for controller)
  • Storage: 40 GB primary disk
  • Network: Based on your environment (see Networking section)

Operating System

  • Ubuntu 20.04 LTS or Ubuntu 22.04 LTS
  • Fresh installation recommended

Network Connectivity

  • All nodes must be able to reach each other
  • Controller node must be accessible by compute/network nodes
  • Internet access for package installation

🌐 Networking Configuration

Network Model (from manuals)

The installation manuals describe a 2-NIC setup:

  1. Management Network: Internal OpenStack communication (API, RabbitMQ, Database)
  2. Provider/Public Network: External access, floating IPs, tenant networks
┌─────────────────────────────────────────────────────────────────────────────┐
│                           Management Network (100.100.100.0/24)             │
│  ┌─────────────┐          ┌─────────────┐          ┌─────────────┐         │
│  │ Controller  │          │   Neutron   │          │   Compute   │         │
│  │ .1          │          │   .3        │          │   .2        │         │
│  └──────┬──────┘          └──────┬──────┘          └──────┬──────┘         │
│         │                         │                        │                │
└─────────┴─────────────────────────┴────────────────────────┴────────────────┘

┌─────────────────────────────────────────────────────────────────────────────┐
│                         Provider Network (192.168.116.0/24)                  │
│         │                         │                        │                │
│  ┌──────┴──────┐          ┌───────┴──────┐          ┌──────┴──────┐        │
│  │ Controller  │          │   Neutron    │          │   Compute   │        │
│  │ .130        │          │   .134       │          │   .132      │        │
│  │             │          │   (br-ex)    │          │             │        │
│  └─────────────┘          └──────────────┘          └─────────────┘        │
└─────────────────────────────────────────────────────────────────────────────┘

Interface Configuration

Edit configs/cluster.env with your actual interface names. To find them:

# List network interfaces
ip link show
# or
nmcli device status

Common interface naming conventions:

  • Legacy: eth0, eth1
  • Predictable: ens33, ens34, enp0s3, enp0s8

📝 Configuration Steps

Step 1: Clone/Copy Scripts to All Nodes

Copy the entire openstack-yoga-3node folder to each node:

# From your workstation
scp -r openstack-yoga-3node/ root@controller:/root/
scp -r openstack-yoga-3node/ root@neutron:/root/
scp -r openstack-yoga-3node/ root@compute:/root/

Step 2: Edit cluster.env

On each node, edit configs/cluster.env:

cd /root/openstack-yoga-3node
vim configs/cluster.env

REQUIRED CHANGES:

Variable Description Example
HOST_CONTROLLER Controller hostname controller01
HOST_COMPUTE Compute hostname compute01
HOST_NEUTRON Network hostname neutron01
IP_CONTROLLER_MGMT Controller management IP 100.100.100.1
IP_CONTROLLER_PUBLIC Controller public IP 192.168.116.130
IP_COMPUTE_MGMT Compute management IP 100.100.100.2
IP_COMPUTE_PUBLIC Compute public IP 192.168.116.132
IP_NEUTRON_MGMT Network management IP 100.100.100.3
IP_NEUTRON_PUBLIC Network public IP 192.168.116.134
IFACE_*_MGMT Management interface per node ens33
IFACE_*_PUBLIC Public interface per node ens34
KEYSTONE_ADMIN_PASS Admin password Change from default!
DB_ROOT_PASS MariaDB root password Change from default!

⚠️ SECURITY WARNING: Change all default passwords before deploying to production!

Step 3: Run Installation Scripts

Order is important! Run scripts in this sequence:

On Controller Node (FIRST):

cd /root/openstack-yoga-3node
chmod +x scripts/*.sh
sudo ./scripts/controller.sh

Wait for completion. Verify with:

source /root/admin-openrc
openstack service list
openstack endpoint list

On Neutron/Network Node (SECOND):

cd /root/openstack-yoga-3node
chmod +x scripts/*.sh
sudo ./scripts/neutron.sh

Wait for completion. Verify from controller:

source /root/admin-openrc
openstack network agent list

On Compute Node (THIRD):

cd /root/openstack-yoga-3node
chmod +x scripts/*.sh
sudo ./scripts/compute.sh

Post-Compute Discovery (ON CONTROLLER):

After compute node installation:

# On controller node
su -s /bin/sh -c "nova-manage cell_v2 discover_hosts --verbose" nova

# Verify
source /root/admin-openrc
openstack compute service list
openstack hypervisor list

✅ Verification Checklist

Basic Services

# On Controller - Source credentials first
source /root/admin-openrc

# Check Keystone
openstack token issue

# List all services
openstack service list

# List all endpoints
openstack endpoint list

# Check compute services
openstack compute service list

# Check network agents
openstack network agent list

# Check hypervisors
openstack hypervisor list

# Check images
openstack image list

Dashboard Access

Open in browser:

http://<CONTROLLER_PUBLIC_IP>/horizon

Login credentials:

  • Domain: default
  • User: admin
  • Password: (your KEYSTONE_ADMIN_PASS from cluster.env)

End-to-End Test (Create Instance)

# Source credentials
source /root/admin-openrc

# Create external network (on network node's br-ex)
openstack network create --external --provider-network-type flat \
  --provider-physical-network physnet1 external

# Create external subnet
openstack subnet create --network external \
  --subnet-range 192.168.116.0/24 \
  --allocation-pool start=192.168.116.200,end=192.168.116.250 \
  --gateway 192.168.116.1 \
  --dns-nameserver 8.8.8.8 \
  external-subnet

# Create internal network
openstack network create internal

# Create internal subnet
openstack subnet create --network internal \
  --subnet-range 10.0.0.0/24 \
  --gateway 10.0.0.1 \
  --dns-nameserver 8.8.8.8 \
  internal-subnet

# Create router
openstack router create router1
openstack router set --external-gateway external router1
openstack router add subnet router1 internal-subnet

# Create security group rules (allow ping and SSH)
openstack security group rule create --proto icmp default
openstack security group rule create --proto tcp --dst-port 22 default

# Create keypair
openstack keypair create mykey > mykey.pem
chmod 600 mykey.pem

# Create a flavor (if not exists)
openstack flavor create --ram 512 --disk 1 --vcpus 1 m1.tiny

# Launch instance
openstack server create --flavor m1.tiny \
  --image cirros \
  --network internal \
  --key-name mykey \
  testvm

# Check instance status
openstack server list

# Create and attach floating IP
openstack floating ip create external
FIP=$(openstack floating ip list -f value -c "Floating IP Address" | head -1)
openstack server add floating ip testvm $FIP

# Test connectivity
ping $FIP

🔥 Common Issues and Troubleshooting

1. Database Connection Errors

Symptom: Services fail to start with "Can't connect to MySQL"

Fix:

# On controller, verify MariaDB is listening
ss -tlnp | grep 3306

# Check MariaDB bind address
grep bind-address /etc/mysql/mariadb.conf.d/99-openstack.cnf

# Test connectivity from other nodes
mysql -h <CONTROLLER_MGMT_IP> -u root -e "SELECT 1;"

2. RabbitMQ Connection Failures

Symptom: Services timeout connecting to message queue

Fix:

# On controller, check RabbitMQ status
rabbitmqctl status

# Verify user exists
rabbitmqctl list_users

# Check permissions
rabbitmqctl list_permissions

# Test connection from compute/network
python3 -c "import pika; pika.BlockingConnection(pika.ConnectionParameters('CONTROLLER_IP'))"

3. Neutron Agent Not Appearing

Symptom: openstack network agent list doesn't show the agent

Fix:

# On the affected node, check service status
systemctl status neutron-openvswitch-agent
journalctl -u neutron-openvswitch-agent -f

# Common causes:
# - Wrong tunnel IP in openvswitch_agent.ini
# - RabbitMQ connection issues
# - Wrong Keystone credentials

4. Nova Compute Not Discovered

Symptom: Hypervisor not showing in openstack hypervisor list

Fix:

# On compute node, check nova-compute status
systemctl status nova-compute
journalctl -u nova-compute -f

# On controller, manually discover hosts
su -s /bin/sh -c "nova-manage cell_v2 discover_hosts --verbose" nova

# Check cells
nova-manage cell_v2 list_cells

5. VNC Console Not Working

Symptom: Can't access instance console from Horizon

Fix:

# Verify novncproxy is running on controller
systemctl status nova-novncproxy

# Check VNC settings in nova.conf on compute
grep -A5 "\[vnc\]" /etc/nova/nova.conf

# Verify novncproxy_base_url points to accessible controller IP
# It must be the PUBLIC IP that your browser can reach

6. Instance Can't Get DHCP/Metadata

Symptom: Instance gets no IP or stuck at booting

Fix:

# On neutron node, check DHCP agent
systemctl status neutron-dhcp-agent
journalctl -u neutron-dhcp-agent -f

# Check metadata agent
systemctl status neutron-metadata-agent

# Verify OVS bridges
ovs-vsctl show

# Check if namespace exists
ip netns list

7. SELinux/Firewall Blocking Traffic

Symptom: Various connectivity issues

Fix:

# Check firewall status (Ubuntu uses ufw)
ufw status

# Disable for testing (NOT for production!)
ufw disable

# For production, open required ports:
# Controller: 3306, 5672, 11211, 5000, 8774, 8775, 8778, 9292, 9696, 6080
# Network: 4789 (VXLAN)
# Compute: 4789 (VXLAN)

8. Wrong Interface/IP Configuration

Symptom: Tunnel failures, no connectivity between nodes

Fix:

# Verify local_ip in OVS agent config matches actual interface IP
ip addr show

# Check OVS agent config
cat /etc/neutron/plugins/ml2/openvswitch_agent.ini

# The local_ip must be reachable from other nodes

📋 Service Ports Reference

Service Port Node
MariaDB 3306 Controller
RabbitMQ 5672 Controller
Memcached 11211 Controller
Keystone 5000 Controller
Glance 9292 Controller
Nova API 8774 Controller
Nova Metadata 8775 Controller
Placement 8778 Controller
Neutron 9696 Controller
NoVNC Proxy 6080 Controller
Horizon 80/443 Controller
VXLAN 4789/UDP All

📚 Reference Documentation

📄 License

These scripts are provided as-is for educational and lab purposes.

⚠️ Disclaimer

These scripts are derived from installation manuals and are intended for lab/learning environments. For production deployments, consider:

  • Using official deployment tools (Kolla-Ansible, OpenStack-Ansible, TripleO)
  • Implementing proper security hardening
  • Setting up high availability
  • Configuring proper SSL/TLS
  • Using secrets management solutions

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages