ONOS_autocluster

This script is intended to automatically deploy an ONOS cluster, a fork from original work by ederollora/ONOS_autocluster

Prerequisite (you may skip it if already installed)

• Docker engine installed on your Ubuntu server with a non-user privileges to access all docker commands, use this link
• Mininet
• sudo apt install jq python-is-python2

Steps to do

• Run git clone https://github.com/abazh/ONOS_autocluster.git

You may encounter problem when running the script if jq and python2 is not install on your server Just install the dependencies sudo apt install jq python-is-python2

• Go the directory cd ONOS_autocluster
• Create ONOS cluster (by default 2 instances) ./create_cluster.sh
• Set sshkey and env parameters ./set_env_sshkey.sh

Testing (with an example mytopo.py)

mytopo.py consists of 6 switches and 6 hosts

• Go to directory cd ONOS_autocluster
• Run the onos-netfcg onos-netcfg $OC1 mytopo.json

setting the label name of the switches and hosts in ONOS

• Run the example sudo python3 mytopo.py $ONOS_INSTANCES
• In the mininet shell pingall or h1 ping h5

run the topology and connected all switches to the two ONOS Instances in the pingall and ping command does not can activate apps in ONOS or do some steps with host-to-host intent from ONOS GUI

• On the other terminal, run onos

it will direct you to ONOS CLI, you can just press TAB to see other command completion

• You may activate the necessary ONOS app, app activate fwd or
• checking paths between two switches paths --disjoint of:0000000000000001 of:0000000000000003
• Go to ONOS GUI with your web browser to open http://[your_ip_address or localhost]:8181/onos/ui