Coder Social home page Coder Social logo

3-node-ha-openstack-cluster-installation's Introduction

The OpenStack installation package contains codes to install and configure a 3-node OpenStack control cluster based on the architecture detailed out in the SDN POC page. The package will install Keystone, Glance and Nova components along with Neutron configuration on a 3-node cluster based on Havana code release. Additional components such as RabbitMQ, Keepalived, Haproxy and MySQL/Galera will also be installed to support the cluster operation.

openstack_ha

A separated glusterfs cluster is required to serve glance image repository and other shared services. The cluster is essential for nova instance images if to support VM migration across different hypervisor. The glusterfs cluster is recommended to be built on a 2-node cluster with minimum 500G storage space. A installation script gluster_install.sh is included to standup such a cluster.

OpenStack HA Installation Steps

  1. Make sure all control nodes' clock are synced and iptables rules allow needed connection requests

  2. Make sure all Openstack controller nodes have needed entries in /etc/hosts, including VIPs

  3. Make sure appropriate repos are specified in /etc/yum.repos.d4

  4. Yum install openstack-install3 package

  5. Identify an existing glusterfs cluster or configure a new one with following script Example: gluster_install.sh -c glusterfs_vip [-m mount_point] [-b brick_name] [-v volume_name] [-d device_name] ......

  6. Install OpenStack components on the first OpenStack controller node i.e. openstack_install3.sh [-L] -v openstack_controller_vip -c contrail_controller_vip -F first_openstack_controller -S second_openstack_controller -T third_openstack_controller [ -e glusterfs_vip] [-m glusterfs_volume_name]

  7. Repeat step 4 on the second OpenStack controller node

  8. Repeat step 4 on the third OpenStack controller node

  9. If needed, restart following services:

service rabbitmq-server start|stop|restart service mysql start|stop|restart service keepalived start|stop|restart service haproxy start|stop|restart service openstack-keystone start|stop|restart service openstack-glance-api start|stop|restart service openstack-glance-registry start|stop|restart service openstack-nova-api start|stop|restart service openstack-nova-cert start|stop|restart service openstack-nova-conductor start|stop|restart service openstack-nova-consoleauth start|stop|restart service openstack-nova-novncproxy start|stop|restart service openstack-nova-scheduler start|stop|restart service httpd start|stop|restart

That's about it for OpenStack !


The SDN installation package contains codes to install, configure and provision Contrail services on a 3-node cluster based on V1.1 release. The package will install Contrail Database, UI, Control, Analytics and Config components on the 3-code cluster with multi-tenancy enabled. Contrail replaces Neutron by providing SDn solution to the native NaaS in OpenStack. By integrating with OpenStack, Contrail services interact with the already installed OpenStack cluster for credential and message queue services while OpenStack call Contrail API for network services. All interactions are through RESTful API calls.

opencontrail_ha

Contrail HA Installation Steps

  1. Make sure all control nodes' clock are synced with OpenStack control nodes and iptables rules allow needed connection requests

  2. Make sure all Openstack controller nodes have needed entries in /etc/hosts, including VIPs

  3. Make sure appropriate repos are specified in /etc/yum.repos.d

  4. Yum install sdn-install3 package

  5. Install Contrail components on the first Contrail controller node i.e. ./contrail_neutron_install3.sh [-L] -o openstack_controller_vip -F first_contrail_controller -S second_contrail_controller -T third_contrail_controller

  6. Repeat step 4 on the second Contrail controller node

  7. Repeat step 4 on the third Contrail controller node

  8. Restart all services list below all on the 3 Contrail controller nodes

    service keepalived restart service haproxy restart service zookeeper restart service supervisor-analytics restart service supervisor-config restart service supervisor-control restart service supervisor-webui restart service supervisord-contrail-database restart service neutron-server restart

Now you have NaaS with SDN. Almost there......


The compute installation package install, configure and provision Nova compute and vrouter services on a compute node. It fetch credentials from OpenStack controller and insert configuration parameters. This version on compute install supports two network interface allocation model for all-in-one and separation of admin and data services with or without NIC bonding.

Compute Installation Steps

  1. Make sure all compute node's clock is synced with OpenStack/Contrail control nodes

  2. Make sure all Openstack controller nodes have needed entries in /etc/hosts, including VIPs

  3. Make sure appropriate repos and specified in /etc/yum.repos.d

  4. make sure the OS kernel matches the prerequisite

  5. Yum install compute-install3 package

  6. Install Compute components on the designate compute node i.e. compute_install3.sh [-L] -o openstack_controller_vip

  7. Reboot the compute node

The cluster should be up, Have fun!

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.