Introduction to cloudstack 4.4

Introduction to cloudstack 4.4

1. Introduction to cloudstack

1. cloudstack is an open-source cloud computing platform with high availability and scalability.

 

2. We will find that cloudstack is more like a commercial product with a very good user interface. By default, each module is well integrated and the installation and deployment process is relatively easy.

 

3. cloudstack supports managing most mainstream hypervisors, such as KVM, xenserver, VMWare, Oracle Vm, and xen.

 

4. cloudstack provides comprehensive user permission management for commercial software, allowing users to build a secure multi-tenant cloud computing environment. It is also compatible with Amazon aws api interfaces and can be used to manage AWS resources.

 

5. cloudstack adopts a typical layered structure: client, core engine, and resource layer. It provides different access methods for various types of customers: Web Console, command shell, and Web Service API. With these features, you can manage and use the underlying computing resources (including hosts, networks, and storage) to complete functions such as allocating virtual machines on hosts and assigning them to Virtual Disks.

 

6. System Architecture diagram:

650) This. width = 650; "src =" http://static.oschina.net/uploads/space/2014/1023/165746_dQpE_727880.jpg "alt =" 165746_dqpe_727880.jpg "/>

If the virtual machine uses xen and KVM, you need to install the cloudstack agent to support interaction between the virtual machine and the management server. The interaction between the management server and the xen server relies on xapi, and the interaction with vcenter and ESX relies on HTTP.

When deploying cloudstack, you need to understand its hierarchy and storage management.

 

7. Deployment diagram:

 

650) This. width = 650; "src =" http://static.oschina.net/uploads/space/2014/1023/165754_oefZ_727880.jpg "alt =" 165754_oefz_727880.jpg "/>

ZONE: Zone corresponds to a data center in reality. It is the largest unit in cloudstack.

Pod: the pod corresponds to a rack. Machines in the same pod are in the same subnet (network segment.

Cluster: a cluster composed of multiple hosts. Hosts in the same cluster have the same hardware, hypervisor, and shared storage. Virtual machines in the same cluster can be migrated from one host to another without interrupting services.

HOST: host is the running VM host.

 

That is, in terms of the Inclusion relationship, a zone contains multiple pods, a pod contains multiple clusters, and a cluster contains multiple hosts.

 

Cloudstack has two types of storage:

 

Primary storage: The primary storage is associated with the cluster. It provides disk volumes for all virtual machines in the cluster. A cluster must have at least one level of storage and be deployed near the host to provide high performance.

 

Secondary storage: the second-level storage is associated with the zone. It stores template files, ISO images, and disk volume snapshots.

 

 

8, Provisioning point (pods ):

 

A provisioning point usually represents a rack. The host in the cabinet is in the same subnet. Each region must contain one or more provisioning points. The provisioning point includes the host and the primary storage server, configure a reserved IP address range for the internal management communication of cloudstack. The reserved IP address range must be unique for each region in the cloud.

 

9, Cluster (clusters ):

 

A cluster is a cluster composed of multiple hosts.

Hosts in the same cluster have the same hardware, hypervisor, and shared storage. Virtual machines in the same cluster can be migrated from one host to another without interrupting services.

 

A cluster consists of one or more hosts and one or more primary storage servers. The cluster size depends on the software of the underlying virtual machine. In most cases, there are basically no suggestions. When VMware is used, each VMware cluster is managed by the vcenter server. The Administrator must register vcenter in this product. Each zone can have multiple vcenter servers. Each vcenter server may manage multiple VMware clusters.

 

10, Host (hosts ):

 

Host is the running VM host.

 

A host is an independent computer. The host machine runs the Guest Virtual Machine and provides corresponding computing resources. Each host has virtual machine software installed to run the guest virtual machine. For example, a server with KVM support enabled, a Citrix xenserver server, or an esxi server can be used as a host.

 

The host machine is the smallest organizational unit in cloudstack deployment. The host is included in the cluster, and the cluster belongs to the provisioning point, while the region contains the provisioning point (that is, in the logic concept, zone> POD> cluster> host ), the new host can be added at any time to provide more resources to the guest virtual machine. cloudstack automatically detects the CPU and memory resources of the host machine. The host is invisible to end users. End users cannot determine the host to which their virtual machines are allocated.

 

11, Storage (primary storage ):

 

Level 1 storage is associated with the cluster, which provides disk volumes for all virtual machines in the cluster. A cluster must have at least one level of storage and be deployed near the host to provide high performance.

 

650) This. width = 650; "src =" http://static.oschina.net/uploads/space/2014/1023/165803_Djtx_727880.png "alt =" 165803_djtx_727880.png "/>

ISCSI and FC-SAN storage are loaded in the xentered LVM format in xenserver. in this format, storage overprovisioning is not supported. If the storage itself supports thin-provisioning of xenserver, cloudstack can also support storage overprovisioning

 

In the global configuration, the storage. overprovisioning. Factor item is used to configure the overprovisioning ratio. The default value is 2, that is, the storage format that supports overconfiguration is n times that of the storage.

 

After the configuration is complete, you can view the storage information in the dashboard of the UI. The primary storage allocated item is the sum of all master storages after the overconfiguration, and the storage item is the sum of the actual sizes of all master storages.

 

12Secondary storage ):

 

The second-level storage VM (ssvm) plays an important role in cloudstack as a system VM. It cannot be implemented without many functions.

 

In simple terms, ssvm is mainly used to manage level-2 storage, that is, operations related to level-2 storage will be completed through it. Each resource domain can have multiple ssvms. When an ssvm is deleted or stopped, it is automatically rebuilt and started.

 

Cloudstack directly accesses ssvm through SSH. an SSH private key has been embedded in the system VM template. At this time, cloudstack only needs to access ssvm through its own public key. The port is 3922, xen and VMWare have different mechanisms. Because VMware prohibits SSH Login by default, VMWare can directly access the public IP of ssvm through the Management Server of cloudstack. after configuring ssvm, cloudstack will re-generate a pair of public key and private key for ssvm, and then use the new key to access ssvm.

 

There is a Java Process in ssvm that is responsible for communicating with port 8250 of cloudstack. This Java Process and related shell scripts are injected by cloudstack after ssvm is created, for example, VMWare has a system. ISO, which stores all the files required by the system VM. After the second-level storage is mounted, the file will be copied to ssvm and extracted from ssvm, and save the relevant files to the specified path.

 

13What are the specific items in the second-level storage?

 

Level 2 storage is associated with the zone, which stores template files, ISO images, and disk volume snapshots. That is to say, the management of these resources is carried out through ssvm.

 

①. Template: the operating system image of the virtual machine can be started, and other configurations such as installed applications are also included.

②. ISO image: A disk image that contains operating system data or boot media.

③ Disk volume snapshot: The saved copy of the VM data, which can be used to recover data or create a new template.

 

14And ssvm to Manage Level 2 storage?

 

Ssvm manages the resources stored in the second-level storage. ssvm first mounts the second-level storage to the local device, and then converts the operations on the second-level storage to the operations on the local folder. In terms of functionality, ssvm is responsible for managing snapshot, volume, ISO, and template. The following describes how to manage these resources from four perspectives: Upload, download, query, and delete.

 

① Upload resources:

 

When the user selects to upload resources, the management server will pass the URL to ssvm, so that ssvm can download resources from the specified URL through wget.

 

② Download resources:

 

When you select to download a specified resource, the Management Server first searches for the corresponding path of the resource from the database and passes the path to ssvm. ssvm uses the built-in Apache server to provide the resource upload service. In this way, users can download the required resources through ssvm.

 

③ Query resources:

 

Ssvm regularly uploads its status information to the management server.

 

④ Delete a resource:

 

Because the second-level storage has been attached to ssvm, it can directly Delete the content in the second-level storage.

 

15, How to verify ssvm?

 

There is a ssvm-check.sh in ssvm, which is primarily responsible for verifying ssvm-related configurations. The verification content includes DNS, Mount second-level storage, connection to Management Server, and related Java processes.

 

16And cloudstack network traffic types:

Network Design Planning is a highlight of cloudstack and a major reason for its wide range of commercial applications. This article discusses in depth the network design methods in cloudstack.

 

Cloudstack designs management, public, customer, and storage networks based on different data traffic types, which can be referred to as PMGs (public, management, guest, storage) networks.

 

① Public network)

 

When a virtual machine needs to access the Internet or external network, it needs to pass through the public network. This means that the customer's virtual machine must be assigned an Internet IP address. you can obtain an IP address on the cloudstack UI for Nat ing or load balancing between guest and public. all hypervisor must share the public VLAN to ensure external access to the virtual machine.

 

 

② Guest Network)

 

When an end user runs a virtual machine instance created by cloudstack, the guest traffic is generated. The communication between virtual machine instances passes through the customer's network.

 

③ Management Network)

 

Communication between resources in cloudstack will generate management traffic, which includes communication between the management server node and the hypervisor cluster, between the system virtual machine, or between other components; when the cluster Scale is small, the management traffic only occupies a small amount of bandwidth.

 

④ Storage Network)

 

Traffic between the primary storage and hypervisor is interconnected, and Stroage traffic is also generated between the primary storage and the secondary storage, such as migration of Virtual Machine templates and snapshots.

 

 

For more information, see: http://54im.com/%E4%BA%91%E8% AE %A1%E7% AE %97/cloudstack/%E7% AE %80%E5%8D%95%E9%83%A8%E7%BD%B2cloudstackcentos6-5%E5% AE %89%E8%A3%85cloudstack-4-3%E4%B9%8B%E6%A6%82%E5%BF%B5%E4%BB%8B%E7%BB%8D.html

This article from the "no blame" blog, please be sure to keep this source http://perofu.blog.51cto.com/6061242/1567338

Introduction to cloudstack 4.4