Setting Up an Automatically Scalable Discuz! Forum

Application Scenarios

Discuz! is one of the most mature and widely used forum software programs. Website traffic reaches its top in peak hours. If a forum website is deployed on multiple servers to cope with peak hours, some servers must be sitting idle at other times. This increases the costs and leads to a waste of resources.

AS helps you address the preceding problems. After you use AS on the servers for deploying a forum website, AS maintains the proper running of the website and reduces costs by automatically adjusting the number of servers based on the predefined scaling policy. This section describes how to use AS to automatically increase or reduce the number of ECSs for a web service, for example, Discuz!.

Solution Introduction

Table 1 lists the steps required to set up an automatically scalable Discuz! forum website. This section focuses on how to implement AS. The increase or decrease of website traffic causes the rise or drop of the ECS CPU usage. You can configure two CPU usage alarm policies, one for increasing one ECS when the CPU usage exceeds 70% and the other for reducing one ECS when the CPU usage is less than 30%. In this way, there will remain a proper number of ECSs for the forum website.

Creating an AS Configuration

An AS configuration specifies the specifications of ECS instances to be added. To enable the instances to automatically carry web services, use the image discuz_centos6.5 (40 GB) and ensure the parameter settings in the AS configuration the same as those of discuz02.

1. Log in to the management console. Under Compute, click Auto Scaling.
2. On the Instance Scaling page, click Create AS Configuration.
   Configure parameters listed in Table 2. Retain default settings for other parameters.

   Table 2 Key AS configuration parameters

   Parameter	Description	Example Value
   Configuration Template	Select Create new template and configure the parameters, such as ECS type, vCPUs, memory, image, and disk, to create an AS configuration.	Create new template
   Specifications	You can select multiple flavors to minimize the probability of capacity expansion failures due to insufficient resources of a flavor. Set Flavor selection policy to Sequenced or Cost-centric as required.	s3.medium.2 s3.large.2
   Image	Specifies the software and system configuration template for the instances in an AS group.	discuz_centos6.5 (40 GB)
   Disk	Stores data and manages the stored data for the instances in an AS group.	System disk: high I/O, 40 GB Data disk: high I/O, 100 GB
   Security Group	Controls ECS access within or between security groups by defining access rules.	sg-DISCUZ
   EIP	Not required if you have enabled a load balancer in an AS group. The system automatically associates instances in the AS group to the load balancer. These instances will provide services via the EIP bound to the load balancer.	Do not use

3. After setting the parameters, click Create Now.

Creating an AS Group

1. Click Create AS Group.
   Configure parameters listed in Table 3. Retain default settings for other parameters.

   Table 3 Key AS group parameters

   Parameter	Description	Example Value
   Max. Instances	Specifies the maximum number of instances in an AS group.	50
   Expected Instances	Specifies the expected number of instances in an AS group. The ECSs in this practice are manually added to the AS group. To prevent scaling actions before the manually adding, set Expected Instances to 0.	0
   Min. Instances	Specifies the minimum number of instances in an AS group.	0
   VPC	Provides the network used by instances in an AS group. Ensure that the parameter value is the same as the VPC in which discuz02 is deployed.	VPC-DISCUZ
   Subnet	Manages networks in the VPC. Select the subnet created when you apply for the VPC.	vpc-test
   Load Balancing	Evenly distributes traffic to instances in an AS group. A backend port is a service port on which a backend ECS listens for traffic, for example, set Backend Port to 80 and Weight to 1.	Enhanced load balancer
   Health Check Method	ELB health check is recommended, in which heartbeat messages are sent to backend ECSs for check.	ELB health check

2. After setting the parameters, click Create Now.
3. Back to the AS group list. The AS group is successfully created if its status changes to Enabled.

Creating an AS Policy

To automatically scale ECSs, configure two alarm policies to monitor CPU usage. One (as-policy-discuz01) is used to increase the number of ECSs when the website traffic rises, and the other (as-policy-discuz02) is used to reduce the number of ECSs when the website traffic drops.

1. Locate the row containing the created AS group as-group-discuz and click View AS Policy in the Operation column.
2. On the displayed page, click Add AS Policy.
   Configure parameters listed in Table 4 for as-policy-discuz01. When the system detects that the CPU usage exceeds 70% for three consecutive times, as-policy-discuz01 is triggered and an ECS is added to the AS group.

   Table 4 Key parameters for creating AS policy as-policy-discuz01

   Parameter	Description	Example Value
   Policy Name	Specifies the name of the AS policy to be created.	as-policy-discuz01
   Policy Type	Select Alarm.	Alarm
   Alarm Rule	Specifies whether a new alarm rule is to be created (Create) or an existing alarm rule will be used (Use existing).	Create
   Rule Name	Specifies the name of the alarm rule.	as-alarm-cpu-01
   Monitoring Type	Specifies the type of monitoring metrics, which can be System monitoring or Custom monitoring. Select System monitoring.	System monitoring
   Trigger Condition	Select monitoring metrics supported by AS and set alarm conditions for the metrics.	CPU Usage Max. >70%
   Monitoring Interval	Specifies the interval at which the alarm status is updated based on the alarm rule.	5 minutes
   Consecutive Occurrences	Specifies the number of sampling points when an alarm is triggered.	3
   Scaling Action	Specifies an action and the number or percentage of instances. The following scaling action options are available: Add Adds instances to an AS group when the scaling action is performed. Reduce Removes instances from an AS group when the scaling action is performed. Set to Sets the expected number of instances in an AS group to a specified value.	Add 1 instance
   Cooldown Period	To prevent an alarm-based policy from being triggered repeatedly by the same event, configure a cooldown period.	900

3. Click OK.
4. Click Add AS Policy again and create AS policy as-policy-discuz02. When the system detects that the CPU usage is lower than 30% for three consecutive times, as-policy-discuz02 is triggered and an ECS is removed from the AS group.
5. Click OK.
6. Back to the AS policy list. The AS policies are successfully created if their statuses change to Enabled.

Manually Adding an ECS to an AS Group

Perform the following operations to manually add ECS discuz02 to the AS group:

1. Click the name of the AS group as-group-discuz to switch to the page providing details about the AS group.
2. Click the Instances tab and manually add discuz02 to the AS group.

Changing the Minimum Number of Instances

To ensure that discuz02 is not removed from the AS group, perform the following operations to change the minimum number of instances:

1. Click the name of the AS group as-group-discuz to switch to the page providing details about the AS group.
2. Click Modify in the upper right corner of the page. Set Min. Instances to 1.
3. Click OK.

Verifying Configurations

Check whether the forum website can be used. If the CPU usage of ECSs in the AS group remains higher than 70%, as shown on the Monitoring tab of the page providing details about the AS group, an ECS will be automatically added to the AS group (shown on the Scaling Actions tab). If the CPU usage remains lower than 30% and the AS group contains at least two ECSs, an ECS will be automatically removed from the AS group. If not, contact technical support to locate the fault.

Practice Extensions

• To deploy new applications on ECSs, use AS lifecycle hooks to perform customized operations on the instances to be added to or removed from an AS group. For details, see Managing Lifecycle Hooks.
• To modify the specifications of ECSs in an AS group, create a new AS configuration first. For details, see Creating an AS Configuration from a New Specifications Template. Then, replace the AS configuration used by the AS group with the one you created. For details, see Changing the AS Configuration for an AS Group.