F5 Server Load balancer service platform configuration case

Source: Internet
Author: User

F5 is one of our most popular Server Load balancer products, so here we will introduce its configuration on the actual business platform. Through this case, I hope you will have a clear understanding of the specific use and configuration of this product. For more information, please refer to the following section.
I. Network Topology
650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314V210-0.jpg "/>

2. Plan internal and external IP addresses
Plan internal IP addresses
The server Load balancer and each node in the application server cluster form an internal network. The following uses two application nodes in the cluster as an example to plan the IP address of the internal network.
Big-IP 2000 internal vlan address 192.168.20.129
In addition, the management address of Big-IP 2000 is 192.168.1.245, which is fixed in the hardware and cannot be changed.
The following table lists the IP addresses of the two application nodes.
Application node IP Address Mask broadcast address Default Gateway
Portal1 192.168.255.131 255.255.255.255.128 192.168.255.255 192.168.255.129
Portal 192.168.255.132 255.255.255.255.128 192.168.255.255 192.168.255.129

Note: Because the inherent management address of F5 Server Load balancer is 192.168.1.245, the broadcast address of the node cannot use the network segment 192.168.1 to avoid conflicts.

Plan external IP addresses
Big-IP 2000 external vlan management address 192.168.20.30
You can access https: // 192.168.20.30 in a browser to manage the configuration of Big-IP 2000.
Big-IP 2000 external application address (virtual IP) 192.168.20.31
You can access the Portal Server through http: // 192.168.20.31 in a browser.
Application node external IP address planning
Internal IP address of the application node external IP address of the node
Portal1 192.168.255.131 192.168.255.34
Portal 192.168.255.132 192.168.255.35

3. Configure Big-IP 2000
Access Big-IP 2000 through the external management address of Big-IP 2000, enter the operation interface, in the operation interface select "Configure your BIG-IP (R) Controller using the Configuration Utility" hyperlink, go to the configuration page. The "Big-IP 2000 configuration page" is displayed.
650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314R510-1.jpg "/>

Set Network
Click the "NetWork" hyperlink in the left-side menu of 01, select the "VLANs" page, and click the "ADD" button. In the displayed interface, allocate ports of Big-IP 2000, use the "untagged" button to select the port, as shown in example 2 "set network-set VLANs.
650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314V528-2.jpg "/>

 

Select the "Self IP Address" Page and click the "ADD" button. On the displayed page, set the IP Address. The result is "set network-Set Self IP Address. Click the IP Address in the Address column to view and modify the IP Address Configuration. 3:

650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314QT8-3.jpg "/>

Select NATs from the menu on the left, and click add to add the ing between the internal IP address and the external IP address, as shown in Figure 4:
650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314W229-4.jpg "/>

Set Pools
The server Pool that receives requests according to a specific policy.
The Portal Server requires two node pools: pool1 and pool2. Pool2 only serves various administrators. pool1 serves both common users and various administrators. Because the Administrator has the permission to modify the business tree, therefore, only the Administrator's operation on the same node will not cause errors caused by simultaneous modification ).
Add node pool
Select the "pools" hyperlink in the left-side menu to go to the pools page, as shown in 5 "set Pools page.

650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314V2Z-5.jpg "/>

Go to the "Pools" Page and click the "ADD" button to go to the page for adding a pool, as shown in 6 "adding a pool page.

650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314T535-6.jpg "/>

In the Load Balancing Method section, select "Least Connecttion (Menber )".
In the Resourse section, enter the "Member Address" parameter in sequence the internal IP Address of the Portal Server application node 192.168.20.131, and enter "Service" parameter 80 (this is the port used for portal installation.
The "Member Ratio" parameter is set to 1.
After entering the first parameter in the Resourse section, click the button to select the data in the "Current Members" box and enter the next data.
In this way, all the four nodes 192.168.255.131, 192.168.255.132, 192.168.255.255.255.255.141 are input to the pool.
After the operation is complete, click the "DONE" button on the page to save the data and return to the previous interface.

Note: The Load Balancing Method can be divided into the following types:
Round Robin: This is the default configuration method. User connection requests are evenly distributed to all nodes used as Server Load balancer instances. If the processing speed of the device that acts as a server node is the same as that of the memory, this processing method is the most suitable.
Radio: server nodes determine the proportion of their respective processing requests based on the processing capability, and then send the requests to the server node for Processing Based on the proportion, this can solve the problem of different server models in the pool.
Dynamic Radio: This is similar to Radio, but their request processing proportion is dynamically allocated based on the processing capability of each node. Node processing capability monitoring can be determined in real time by obtaining the current number of connections of each node, the fastest response time, and other dynamic performance.
Fastest: The device obtains the node with the Fastest processing speed and sends the Processing request to the Fastest node.
Least Connections: When all nodes have similar processing capabilities, the current Processing request is sent to the service node with the Least number of Connections.
Observed: This is a combination of Fastest and Least Connections modes. All nodes are sorted based on the total number of connections and response speed. A larger proportion of subsequent user requests will be processed by nodes with fewer connections and faster processing speed. This method is applicable when the processing performance of nodes is inconsistent.
Predictive: This method uses the Observed method to sort nodes, however, the device also considers the changing trend of the processing capabilities of these nodes to determine whether the processing capabilities of a node are increasing or decreasing, allows nodes with increasing processing capabilities to process more requests.

Modify the persistence of the node pool
Before and after a user accesses the WEB, the request may have a certain degree of continuity. For example, the data sent by the client in the previous request is used in the request processing process, in this way, these requests must be sent to the same node for processing, that is, to maintain request consistency, is a feature of the node pool.
There are multiple machines in pool1. You need to set the "persistence" attribute of pool1 so that the access of the same user is concentrated on the same Portal Server application node.
Click the pool1 link in the lower-left corner of Figure 5 and select the Persistence page to configure 7 "Modify the persistence of the node pool:

650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314U347-7.jpg "/>

In this way, the Server Load balancer can differentiate users by adding cookies to the user's IP address information, and fix the requests of the same user on the same Portal Server application node.

Configure Virtual Server
Through a rule, the client can access the virtual IP address of the virtual server and forward the request to a suitable node in the node pool for processing. You must configure the node pool before configuring the Virtual Server. The virtual IP address of the virtual Server in iIMUSE is the IP address when the user accesses the Portal Server.
Select the "Virtual Servers" hyperlink in the left-side menu and select the "Virtual Servers" page.
Click "ADD" and enter 192.168.20.31 as the "Address" parameter. Enter 80 as the "Service" parameter, as shown in.
Click "Next" to go to the "Configure Basic Properties" page. You do not need to modify the parameters here.
Click "Next" to go to the "Select Physical Resources" Page and configure the Role. Select PortalCluster ".
Figure 9 shows the configuration result.
650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314TA2-8.jpg "/>

 

650) this. width = 650; "border =" 0 "alt =" "src =" http://www.bkjia.com/uploads/allimg/131227/02314W461-9.jpg "/>

 

By now, the configuration is complete, and the rest is the test goes online.
 

Summary: The F5 Server Load balancer WEB interface configuration is relatively intuitive and easy to use. The key points of the configuration are to pay attention to the IP planning of the entire network, and the POOLS node pool also has the configuration of Virtual Server. In short, it is necessary to make reasonable configuration and Optimization Based on business applications.

This article is from the "dripping water and stone" blog and will not be reposted!

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.