Managing VMs with OpenStack
To start log in to OpenStack deployed at BITP using credentials given to you.
There are different authentication providers as your cloud could be a part of larger system having users who come from other services. We are going to use Keystone, OpenStack service that provides API client authentication, service discovery, and distributed multi-tenant authorization.
Other login options might include EGI (European Grid Infrastructure) or Elixir, distributed infrastructure for life-science information.
One can be a part of [VO](https://en.wikipedia.org/wiki/Virtual_organization_(grid_computing)) (virtual organisation), which has access to different OpenStack instances as well as Grid infrastructures using alternative authentication methods. For example, EGI provides a manual on how to set up cloud infrastructure over grid.
Dashboard overview
After the login you will see OpenStack dashboard with the right menu featuring following sections:
API access. Links to the OpenStack services dealing with different different part of the cloud such as network or authentication. We will use this information later when creating virtual machines (VMs) from the linux command line.
Compute. Computing resources management.
Overview. Summary of available resources and running instances.
Instances. Virtuale machines in use (including suspended ones).
Images. Operating systems (OS) images, that were uploaded into the cloud. Many wide spread OS provide ready to use cloud images, for example Ubuntu. The images might come in different formats such as QCOW2 or VMDK, and depending on the specific OpenStack instance it is important to know who to convert between formats. OS must be specifically preconfigure, so OpenCloud could interact with it, however, it is not difficult to create image yourself. For example, one can convert VirtualBox image into raw format using its VBoxManage command line tool or create image manually using OpenStack manual.
Key Pairs. Encryption key (public key) and decryption key (private key) pair for secure data transmission. It's an alternative way how to access VM via SSH (secure shell protocol) instead of password authentication method, which is prohibited by defaults in many OS as less reliable. User can either create new key pair withing OpenStack or import existing one.
Server Group. It provides a mechanism to group servers according to certain policy.
Volumes. Storage management.
Volumes. Overview of created volumes.
Snapshots. Typically snapshot is an exact capture of what a volume looked like at a particular moment in time, including all it's data. The OpenStack snapshot mechanism allows you to create new images from running instances. This is very convenient for upgrading base images or for taking a published image and customizing it for local use.
Groups. Volume groups.
Group Snapshots. Snapshot groups.
Network. Network management.
Network Topology. Visual representation of network topology. Each OpenStack project can have its own private (internal) network(s), which must be connected to public network, so one can access the machine via SSH. Typically "public" network doesn't necessarily mean direct access via public IP and many OpenStack operators provide SSH access to a "jump host" from where one can login to the machine. Later we will use BITP jump host to connect to newly created VMs.
Networks. The list of available networks.
Routers. The list of available routers, which redirects traffic from one network to another.
Secutiry Groups. A set of rules one can assign to the particular VM instance to allow or restrict usage of specific network protocols and ports.
Floating IPs. Floating IPs are just publicly routable IPs that one typically buys from an ISP (Internet Service Provider).
Orchestration. Orchestration is typically a service for managing the entire lifecycle of infrastructure and applications within OpenStack clouds. More you can read in the OpenStack manual.
Stacks. These flexible template languages enable application developers to describe and automate the deployment of infrastructure, services, and applications. The templates enable creation of most OpenStack resource types, such as instances, floating IP addresses, volumes, security groups, and users. The resources, once created, are referred to as stacks.
Resource Types. There are many resources types such as server or volume, the full list can be found here.
Template Versions. OpenStack template versions.
Template Generator. An editor for templates creation.
Identity. Projects and users management.
Projects. Projects list in the particular OpenStack instance.
Users. Users list in the particular OpenStack instance.
Application credentials. Users can create application credentials to allow their applications to authenticate to keystone.
Create simple VM with vnc login
Now when we are familiar with OpenStack dashboard it is time to create our first VM. Follow the instructions:
Navigate to Project -> Compute -> Instances and press Launch Instance.
Under Details menu input instance name.
Under Source menu select boot source Image, set volume size as 1GB, select Yes to delete volume on instance delete, press arrow near the CirrOS image.
Under Flavor menu select m1.tiny instance type.
Press Launch Instance.
After that wait till the power state of the VM is Running and press VM name. You will be able to see VM overview, its network interfaces, boot log, console and action log. Check boot log to ensure that VM is running indeed.
Next step is accessing VM via [Virtual Network Computing (VNC)] (https://en.wikipedia.org/wiki/Virtual_Network_Computing) interface, graphical desktop-sharing system that uses the Remote Frame Buffer protocol (RFB) to remotely control another computer. Go to console tab and click on the dark area to activate mouse capture. CirrOS has password enabled login by default, the credentials are listed above the login prompt. After you login, check if you can access internet by executing followin command:
VM with SSH login
Using VNC is less convenient then working through SSH, while it does not allow to use key authentication. In order to create VM and login via SSH follow the instructions:
Navigate to Project -> Compute -> Instances and press Launch Instance.
Under Details menu input instance name.
Under Source menu select boot source Image, set volume size as 1GB, select Yes to delete volume on instance delete, press arrow near the Ubuntu-cosmic image.
Under Flavor menu select m1.tiny instance type.
Under Key pair menu press Create Key Pair, enter kay pair name and press create. Copy pricate key to clipboard, open any text editor, paste the key and save the file. You won't be able to access VM without private key.
Press Launch Instance.
After that wait till the power state of the VM is Running and press VM name. You will be able to see VM overview, its network interfaces, boot log, console (for web access, although you won't be ablle to login as we use key authentication) and action log. Verify if console shows login prompt to be sure that VM is running indeed.
Next step is accessing VM via SSH. Unfortunatelly, we do not have many public IPs, so we are going to access VM through the jump host. Log in to cloud-11 host using provided credentials:
Launching VM with Vagrant
Using graphical interface (GUI) is nice, but takes time, especially if one needs to launch and destroy VMs regularly. There are many ways how to do it via command line using config files with all required information. We are going to use Vagrant, open-source software product for building and maintaining portable virtual software development environments, e.g. for VirtualBox, KVM, Hyper-V, Docker containers, VMware, and AWS. It tries to simplify software configuration management of virtualizations in order to increase development productivity. There are many user provided plugins for Vagrant including OpenStack provider. In particular, we will use OpenStack plugin from this repository. Before proceeding examine the repository and read about implemeted functionality.
You should see following message:
Now clone the repository with Vagrant configuration
Last updated