Shulou(Shulou.com)06/01 Report--

At the beginning of this section, we will explain the various sub-services of Nova in detail.

The previous architecture overview section knows that Nova has several sub-services of nova-*, and we will learn the most important ones in turn.

Today we will first discuss nova-api and nova-conductor.

Nova-api

Nova-api is the portal to the entire Nova component, and all requests for Nova are first processed by nova-api. Nova-api exposes several HTTP REST API interfaces to the outside world. In keystone, we can query the endponits of nova-api.

The client can send the request to the address specified by endponits and request an operation from nova-api. Of course, as end users, we don't send Rest AP I requests directly. These API are used by OpenStack CLI,Dashboard and other components that need to be exchanged with Nova.

The HTTP API requests received by Nova-api will be handled as follows: 1. Check whether the parameters passed by the client are legal and valid 2. The client-side HTTP request 3. 0 that invokes other Nova sub-services. Format the results returned by other Nova sub-services and return them to the client

What requests does nova-api receive? In a nutshell, nova-api can respond to any operation related to the life cycle of a virtual machine. Most of the operations can be found on Dashboard.

Open the Instance management interface

Click the drop-down arrow, and the list shows what nova-api can do.

OpenStack uses the term "Instacne" to refer to a virtual machine, which we will use uniformly later.

Nova-conductor

Nova-compute needs to get and update the information of instance in the database. However, nova-compute does not directly access the database, but through nova-conductor to achieve data access.

This has two significant benefits:

Higher system security

Better system scalability

Higher security

In earlier versions of OpenStack, nova-compute could access the database directly, but there were great security risks. Because the nova-compute service is deployed on the compute node, in order to be able to access the database on the control node, you must configure the connection information to access the database in the / etc/nova/nova.conf of the compute node, such as

[database] connection = mysql+pymysql://root:secret@controller/nova?charset=utf8

Imagine that any computing node is *, which will cause the database deployed on the control node to face great risk.

To solve this problem, starting from the G version, Nova introduces a new service nova-conductor, which puts all the operations of nova-compute accessing the database into nova-conductor, and nova-conductor is deployed on the control node. In this way, the direct access to the database by nova-compute is avoided and the security of the system is increased.

Better scalability

Nova-conductor decouples nova-compute from the database and brings another benefit: it improves the scalability of nova.

Nova-compute interacts with conductor through message middleware. This loose architecture allows for the configuration of multiple nova-conductor instances. In a large-scale OpenStack deployment environment, administrators can cope with the growing number of computing nodes accessing databases by increasing the number of nova-conductor.

In the next section, we discuss the computing node scheduling service nova-scheduler.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.