A thin later around virsh to allow one to setup a host for running vms
On the host where vms will run you can use the vminstall.sh script to create a base vm from which you can create independent vm instances. By default it creates vm's installed with fedora release on to a bootable disk.
On the host where you have created your base vm, with vminstall.sh
, you can save the base vm image with vmsave.sh
. This script will
create a tar file that you can store to allow others to directly use the vmload.sh
script rather than having to install their own.
On the host where you plan to run vm's you can use the vmload script to create the base vm from a saved image of it hosted on a webserver.
These scripts can be run anywhere. They will ssh their commands to the VM host for you to get access to them.
You can use this on any host to create a new vm, of the name specified, on the vmhost. This will clone a new independent version of the basevm for you. Eg. newvm dev
You can then use the vm
command to access it.
This is the main daily command you will use. With no arguments it will list your vm's that you have created with vmnew
on the vm host. If you pass it the name of one of these vm's it will start it, if needed, and attach to the console.
This coammand will ssh to the specified vm on the vmhost.
This attempts to determine the ip address of the specified vm assuming it has an interfaced attached to virbr0 (default network)