Huawei Auto Scaling User Manual

Download

Auto Scaling

User Guide

Issue 01

Date 2020-10-27

HUAWEI TECHNOLOGIES CO., LTD.

No part of this document may be reproduced or transmitted in any form or by any means without prior written consent of Huawei Technologies Co., Ltd.

Trademarks and Permissions

and other Huawei trademarks are trademarks of Huawei Technologies Co., Ltd. All other trademarks and trade names mentioned in this document are the property of their respective holders.

Notice

The purchased products, services and features are stipulated by the contract made between Huawei and the customer. All or part of the products, services and features described in this document may not be within the purchase scope or the usage scope. Unless otherwise specied in the contract, all statements, information, and recommendations in this document are provided "AS IS" without warranties, guarantees or representations of any kind, either express or implied.

The information in this document is subject to change without notice. Every eort has been made in the preparation of this document to ensure accuracy of the contents, but all statements, information, and recommendations in this document do not constitute a warranty of any kind, express or implied.

Auto Scaling User Guide Contents

1 AS Group....................................................................................................................................1

1.1 Creating an AS Group............................................................................................................................................................ 1

1.2 (Optional) Adding a Load Balancer to an AS Group.................................................................................................. 5

1.3 Replacing AS Conguration in an AS Group..................................................................................................................6

1.4 Enabling an AS Group............................................................................................................................................................6

1.5 Disabling an AS Group.......................................................................................................................................................... 7

1.6 Modifying an AS Group.........................................................................................................................................................8

1.7 Deleting an AS Group............................................................................................................................................................ 9

2 AS Conguration................................................................................................................... 10

2.1 Creating an AS Conguration...........................................................................................................................................10

2.2 Using an Existing ECS to Create an AS Conguration............................................................................................. 10

2.3 Using a New Specications Template to Create an AS Conguration............................................................... 14

2.4 Copying an AS Conguration........................................................................................................................................... 20

2.5 Deleting an AS Conguration...........................................................................................................................................20

3 AS Policy.................................................................................................................................. 21

3.1 Overview.................................................................................................................................................................................. 21

3.2 Creating an AS Policy.......................................................................................................................................................... 22

3.3 Managing AS Policies.......................................................................................................................................................... 31

4 Scaling Action........................................................................................................................ 33

4.1 Dynamically Expanding Resources..................................................................................................................................33

4.2 Expanding Resources as Planned.................................................................................................................................... 35

4.3 Manually Expanding Resources....................................................................................................................................... 36

Conguring an Instance Removal Policy...................................................................................................................... 37

4.4

4.5 Viewing a Scaling Action.................................................................................................................................................... 38

4.6 Managing Lifecycle Hooks.................................................................................................................................................39

4.7 Conguring Instance Protection...................................................................................................................................... 45

4.8 Standby Instance................................................................................................................................................................... 46

5 Bandwidth Scaling................................................................................................................ 48

5.1 Creating a Bandwidth Scaling Policy..............................................................................................................................48

5.2 Viewing Details About a Bandwidth Scaling Policy.................................................................................................. 54

5.3 Managing a Bandwidth Scaling Policy.......................................................................................................................... 55

Auto Scaling User Guide Contents

6 AS Group and Instance Monitoring...................................................................................58

6.1 Health Check.......................................................................................................................................................................... 58

6.2 Conguring Notication for an AS Group....................................................................................................................59

6.3 Recording AS Resource Operations................................................................................................................................ 60

6.4 Adding Tags to AS Groups and Instances..................................................................................................................... 63

6.5 Monitoring Metrics............................................................................................................................................................... 64

6.6 Viewing Monitoring Metrics..............................................................................................................................................69

6.7 Setting Monitoring Alarm Rules...................................................................................................................................... 70

Auto Scaling User Guide 1 AS Group

1 AS Group

1.1 Creating an AS Group

Scenarios

Notes

An AS group consists of a collection of instances and AS policies that have similar attributes and apply to the same application scenario. An AS group is the basis for enabling or disabling AS policies and performing scaling actions. The congured AS policy automatically adds or deletes instances to or from an AS group, or maintains a

When creating an AS group, specify an AS one or more AS policies for the AS group.

Creating an AS group involves the conguration of the maximum, minimum, and expected numbers of instances and the associated load balancer.

ECS types supported by must choose a proper AS conguration according to the ECS type supported by the AZs used by the AS group.

● If the ECS type AZs used by the AS group, the following situations will occur:

– If the AS group is disabled, it cannot be enabled.

– If the AS group is enabled, its status will become abnormal when

instances are added to it.

● If the ECS type AZs used by the AS group, the ECSs added by a scaling action are distributed only in the AZs supporting the ECS type. As a result, the instances in the AS group may not be evenly distributed.

xed number of instances in an AS group.

conguration for it. Additionally, add

dierent AZs may vary. When creating an AS group, you

specied in the AS conguration is supported by none of the

specied in the AS conguration is supported only by certain

pre-

Procedure

1. Log in to the management console.

Auto Scaling User Guide 1 AS Group

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click Create AS Group.

4. Set parameters, such as Name, Max. Instances, Min. Instances, and

Expected Instances. Table 1-1 describes the key parameters to be congured.

Table 1-1 AS group parameters

ParameterDescription Example

Value

Region A region is where an AS group resides. N/A

AZ An AZ is a physical region where resources use

independent power supply and networks. AZs are physically isolated but interconnected through an internal network.

Multi-AZ Extensio n Policy

This parameter can be set to Load-balanced or Sequenced.

● Load-balanced: When expanding ECSs in an AS group, the system preferentially distributes ECSs evenly among AZs used by the AS group. If it fails in the target AZ, it automatically selects another AZ based on the sequenced policy.

● Sequenced: When expanding ECSs in an AS group, the system selects the target AZ based on the order in which AZs are selected.

NOTE

This parameter needs to be more AZs are selected.

congured when two or

Name Species the name of the AS group to be created.

The name contains 1 to 64 characters and consists of only letters, digits, underscores (_), and hyphens (-).

N/A

Loadbalanced

N/A

Max. Instance

Species the maximum number of ECS instances in an AS group.

Expected Instance s

Species the expected number of ECS instances in an AS group.

After an AS group is created, you can change this

value, which will trigger a scaling action.

Min. Instance

Species the minimum number of ECS instances in an AS group.

Auto Scaling User Guide 1 AS Group

ParameterDescription Example

Value

congur ation

Species the required AS conguration for the AS group. An AS conguration denes the specications of the ECSs to be added to an AS

group. The specications include the ECS image and system disk size. You need to create the required AS

conguration before creating an AS

group.

VPC Provides a network for your ECSs.

All ECSs in an AS group belong to the same VPC.

Subnet You can select a maximum of ve subnets. The AS

group automatically binds all NICs to the created ECSs. The rst subnet is used by the primary NIC of the ECS by default, and other subnets are used by extension NICs of the ECS.

Load Balancin g

This parameter is optional. A load balancer automatically distributes access trac to all instances in an AS group to balance their service load. It enables higher levels of fault tolerance in your applications and expands application service capabilities.

NOTE

● Up to six load balancers can be added to an AS group.

● After multiple load balancers are added to an AS group, multiple services can be concurrently listened to, thereby improving service scalability. If ELB health check is selected for Health Check Method, when any one of the listeners detects that an instance becomes faulty, AS will replace the faulty instance with a functional one.

If you select Enhanced load balancerIf you select Elastic load balancer,

congure the following

parameters:

● Load Balancer

● Backend ECS Group

● Backend Port:

species the port on which a

backend ECS listens for trac.

● Weight: determines the portion of requests a backend ECS processes compared to other backend ECSs added to the same listener. For more information about load balancing,

Elastic Load Balance User Guide

see

N/A

Auto Scaling User Guide 1 AS Group

ParameterDescription Example

Value

Instance Removal Policy

Species the priority for removing instances from an AS group. If specied conditions are met, scaling actions are triggered to remove instances. AS supports the following instance removal policies:

● Oldest instance created from oldest AS conguration: The oldest instance created based on the oldest conguration is removed from the AS group

rst.

● Newest instance created from oldest AS conguration: The latest instance created based on the oldest conguration is removed from the AS group rst.

● Oldest instance: The oldest instance is removed from the AS group rst.

● Newest instance: The latest instance is removed from the AS group

NOTE

● Removing instances will preferentially ensure that the remaining instances are load balanced in AZs.

● A manually added ECS is removed in the lowest priority. AS does not delete a manually added ECS when removing it. If multiple manually added ECSs must be removed, AS preferentially removes the earliest-added ECS.

rst.

Oldest instance created from oldest AS

congurati on

EIP If EIP has been selected in an AS conguration for

an AS group, an EIP is automatically bound to the ECS added by a scaling action to the AS group. If you select Release, the EIP bound to an ECS is released when the ECS is removed from the AS group. Otherwise, the system unbinds the EIP from the ECS, but does not release it when the ECS is removed from the AS group.

N/A

Auto Scaling User Guide 1 AS Group

ParameterDescription Example

Value

Health Check Method

Advance d Settings

When a health check detects a faulty ECS, AS removes the faulty ECS from the AS group and adds a new one. The health check is implemented using any of the following methods:

● ECS health check: checks ECS running status. If an ECS is stopped or deleted, it is considered as abnormal. This method is selected by default. Using this method, the AS group periodically determines the running status of each ECS based on the health check result. If the health check result shows that an ECS is faulty, AS removes the ECS from the AS group.

● ELB health check: determines ECS running status using a load balancing listener. This health check method is available only when the AS group uses a load balancing listener. When a load balancing listener detects that an ECS is faulty, AS removes the ECS from the AS group.

Congure notications and tags.

You can select Do not congure or Congure now.

N/A

5. Click Next. On the Add AS existing AS

6. Click Create Now.

7. Check the AS group and AS conguration information. Then click Submit.

8. (Optional) Add AS policies. For details, see Creating an AS Policy.

conguration or create one. For details, see .

Conguration page, you can choose to use an

1.2 (Optional) Adding a Load Balancer to an AS Group

Elastic Load Balance (ELB) automatically distributes incoming trac across multiple backend servers based on congured forwarding policies. ELB expands the service capabilities of applications and improves their availability by eliminating single points of failure (SPOFs).

If ELB functions are required, perform the operations provided in this section to add a load balancer to your AS group. The load balancer added to an AS group distributes application added to or deleted from the AS group.

Only a created load balancer can be bound to an AS group, and the AS group and load balancer must be in the same VPC. For details about how to create a load balancer, see group, perform the following operations:

Elastic Load Balance User Guide

trac to all instances in the AS group when an instance is

. To add a load balancer for an AS

● When creating an AS group, load balancer. For details, see Creating an AS Group.

congure parameter Load Balancing to add a

Auto Scaling User Guide 1 AS Group

● If an AS group has no scaling action ongoing, modify parameter Load

Balancing to add a load balancer. For details, see Modifying an AS Group.

1.3 Replacing AS Conguration in an AS Group

Scenarios

If you need to change the ECS specications in an AS group, you need to change the AS conguration.

Eective Time of New AS Conguration

If the AS group has an in-progress scaling action, the new AS conguration will take eect only for instances scaled in later scaling actions.

For example, the current AS conguration of the AS group is as-cong-A, and the new AS conguration is as-cong-B. The instance conguration in the current scaling action is still

The instance conguration will change to as-cong-B in the next scaling action.

as-cong-A.

Figure 1-1 Changing the AS conguration

Procedure

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling. Then click the AS Groups tab.

3. Click the name of the target AS group. On the Basic Information page, click

Change

You can also locate the row containing the target AS group and choose More > Change Conguration in the Operation column.

4. In the displayed Change AS Conguration dialog box, select another AS conguration to be used by the AS group.

5. Click OK.

Conguration to the right of Conguration Name.

1.4 Enabling an AS Group

Scenarios

You can enable an AS group to automatically increase or decrease instances.

After an AS group is enabled, its status changes to Enabled. AS monitors the AS policy and triggers a scaling action for AS groups only in Enabled state. After an AS group is enabled, AS triggers a scaling action to automatically add or remove

Auto Scaling User Guide 1 AS Group

instances if the number of instances in the AS group is dierent from the expected number of instances.

● Only AS groups in the Disabled state can be enabled.

● Only AS groups in the Abnormal state can be forcibly enabled. You can choose More > Forcibly Enable to enable an abnormal AS group. Forcibly enabling an AS group does not have adverse consequences.

● After you create an AS group and add an AS conguration to an AS group, the AS group is automatically enabled.

Enabling an AS Group

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling. Then click the AS Groups tab.

3. In the AS group list, locate the row containing the target AS group and click Enable in the Operation column. You can also click the AS group name and then Enable to the right of Status on the Basic Information page to enable the AS group.

4. In the Enable AS Group dialog box, click Yes.

Forcibly Enabling an AS Group

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling. Then click the AS Groups tab.

3. In the AS group list, locate the row containing the target AS group and select Forcibly Enable from the More drop-down list in the Operation column. You can also click the AS group name and then Forcibly Enable to the right of Status on the Basic Information page to enable the AS group.

4. In the Forcibly Enable AS Group dialog box, click Yes.

1.5 Disabling an AS Group

Scenarios

When you are required to stop an instance in an AS group for upgrade, disable the AS group before performing the operation. This prevents the instance from being deleted in a health check. When the instance status restores, enable the AS group again.

If a scaling action, for example, creating an instance or EVS disk, consistently fails (the failure cause can be viewed on the Elastic Cloud Server page) and retries in an AS group, use either of the following methods to stop the retry:

conguration or

● Disable the AS group. Then, the scaling action that is being performed fails and will not retry. Enable the AS group again when the environment recovers or after replacing the AS

● Disable the AS group and change the expected number of instances to the number of existing instances. After the scaling action fails and ends, the scaling action will not retry.

conguration.

NO TE

Auto Scaling User Guide 1 AS Group

After an AS group is disabled, its status changes to Disabled. AS does not automatically trigger any scaling actions for a Disabled AS group. When an AS group has an in-progress scaling action, the scaling action does not stop immediately after the AS group is disabled.

You can disable an AS group when its status is Enabled or Abnormal.

Procedure

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling. Then click the AS Groups tab.

3. In the AS group list, locate the row containing the target AS group and click Disable in the Operation column. You can also click the AS group name and then Disable to the right of Status on the Basic Information page to disable the AS group.

4. In the Disable AS Group dialog box, click Yes.

1.6 Modifying an AS Group

Scenarios

You can modify an AS group as needed. The values of the following parameters can be changed: Name, Max. Instances, Min. Instances, Expected Instances, Health Check Method, Health Check Interval, Instance Removal Policy.

Changing the value of Expected Instances will trigger a scaling action. Then, AS automatically increases or decreases the number of instances to the value of Expected Instances.

If the AS group is not enabled, contains no instance, and has no scaling action ongoing, you can modify Subnet action ongoing, you can modify its AZ and ELB congurations.

Procedure

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling. Then click the AS Groups tab.

congurations. If an AS group has no scaling

3. In the AS group list, locate the row containing the target AS group, click the AS group name to switch to the Basic Information page, and click Modify in the upper right corner.

You can also locate the row containing the target AS group and choose More > Modify in the Operation column.

4. In the Modify AS Group dialog box, modify related data, for example, the expected number of instances.

5. Click OK.

Auto Scaling User Guide 1 AS Group

1.7 Deleting an AS Group

Scenarios

You can delete an AS group when it is no longer required.

Procedure

● If an AS group is not required during a advised to disable it but not delete it.

● For an AS group that has an instance or ongoing scaling action, if you attempt to forcibly delete the AS group and remove and delete the instances in the AS group, the AS group enters the deleting state, rejects new scaling requests, waits until the ongoing scaling action completes, and removes all instances from the AS group. Then, the AS group is automatically deleted. Manually added instances are only removed out of the AS group, while the instances automatically created in a scaling action are removed and deleted. During the preceding process, you are not allowed to perform other operations in the AS group.

● After an AS group is deleted, its AS policies and the alarm rules generated based on the AS policies deleted.

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. In the AS group list, locate the row containing the target AS group and choose More > Delete in the Operation column.

4. In the displayed Delete AS Group dialog box, click Yes.

congured for the AS group will be automatically

specied period of time, you are

Auto Scaling User Guide 2 AS Conguration

2 AS Conguration

2.1 Creating an AS Conguration

Scenarios

Methods

An AS conguration denes the specications of the ECSs to be added to an AS group. The specications include the ECS image and system disk size.

● When you create an AS group, create an AS conguration or use an existing

conguration.

● Create the required AS conguration on the Instance Scaling page.

● Change the AS conguration on the AS group details page.

● Using an existing ECS to create an AS conguration

When you create an AS conguration using an existing ECS, the vCPU, memory, image, disk, and ECS type are the same as those of the selected ECS by default. For details, see Using an Existing ECS to Create an AS

Conguration.

● Using a new specications template to create an AS conguration

If you have special requirements on the ECSs for resource expansion, use a

specications template to create the AS conguration. For details, see

new

Using a New Specications Template to Create an AS Conguration.

2.2 Using an Existing ECS to Create an AS

Conguration

Scenarios

You can use an existing ECS to rapidly create an AS conguration. In such a case, the parameter settings, such as the vCPUs, memory, image, disk, and ECS type in the AS

conguration are the same as those of the selected ECS by default.

Auto Scaling User Guide 2 AS Conguration

Procedure

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click Create AS Conguration.

4. Set the parameters for the AS conguration parameters.

Table 2-1 AS conguration parameters

Parameter Description Example

Region A region is where an AS conguration resides. N/A

Name Species the name of an AS conguration. N/A

conguration. Table 2-1 lists the AS

Value

Conguratio n Template

Select Use specications of an existing ECS and click Select ECS.

The ECS type, vCPUs, memory, image, and disk information in the AS conguration are the same as those of the selected ECS by default.

EIP An EIP is a static public IP address bound to

an ECS in a VPC. Using the EIP, the ECS provides services externally.

The following options are provided:

● Do not use An ECS without an EIP cannot access the Internet. However, it can still be used as a service ECS or deployed in a cluster on a private network.

● Automatically assign An EIP with a dedicated bandwidth is automatically assigned to each ECS. The bandwidth size is

NOTE

If you select Automatically assign, you need to specify Type, Bandwidth Type, Billed By, and Bandwidth.

congurable.

Use

specication s of an

existing ECS

Automaticall y assign

Auto Scaling User Guide 2 AS Conguration

Parameter Description Example

Value

Bandwidth Type

You can select Dedicated or Shared.

● Dedicated: The bandwidth can be used by only one EIP.

● Shared: The bandwidth can be used by multiple EIPs.

NOTE

● This parameter is available only when EIP is set to Automatically assign.

● If you select Dedicated, you can select Bandwidth or

● The shared bandwidth can be billed only by bandwidth. You can select a shared bandwidth to which the EIP is to be added.

Trac for Billed By.

a password.

● Key pair In this mode, keys are used for authenticating the users who attempt to log in to target ECSs. If you select this mode, create or import a key pair on the

Key Pair page.

NOTE

If you use an existing key, make sure that you have saved the key logging in to the ECS will fail.

le locally. Otherwise,

● Password In this mode, the initial password of user root (for Linux) or user Administrator (for Windows) is used for authentication. You can log in to an ECS using the username and its initial password.

Shared

Admin@123

Advanced Settings

This allows you to congure User Data Injection and ECS Group.

N/A

You can select Do not congure or Congure now.

Auto Scaling User Guide 2 AS Conguration

Parameter Description Example

Value

User Data Injection

Enables the ECS to automatically inject user data when the ECS starts for the rst time. This conguration is optional. After this function is enabled, the ECS automatically injects user data upon its rst startup.

For details, see

Guide

Elastic Cloud Server User

The following methods are available:

● As text: allows you to enter the user data in the text box below.

le: allows you to inject script les or

● As other les when you create an ECS. If you select As le, the system automatically injects the

les into a specied directory

when creating an ECS.

– For Linux, specify the path for storing

the injected

le, for example /etc/

foo.txt.

– For Windows, the injected le is

automatically stored in the root directory of disk C. You only need to specify the

le name, such as foo. The

le name can contain only letters and

digits.

NOTE

● For Linux, if you use the password authentication mode, the user data injection function is unavailable.

● If the selected image does not support user data injection, the user data injection function is unavailable.

ECS Group An ECS group allows you to create ECSs on

N/A dierent hosts, thereby improving service reliability.

5. Click Create Now.

6. Click Create Now.

7. If you want to use the newly created AS

conguration, add it to the AS group.

For details, see Replacing AS Conguration in an AS Group.

8. (Optional) Enable the AS group.

If the AS group is in Disabled state, enable it. For details, see Enabling an AS

Group.

Auto Scaling User Guide 2 AS Conguration

2.3 Using a New Specications Template to Create an AS Conguration

Scenarios

If you have special requirements on the ECSs for resource expansion, use a new specications template to create the AS conguration. In such a case, ECSs meeting specications of the template will be added to the AS group in scaling actions.

Procedure

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click Create AS

4. Set the parameters for the AS conguration parameters.

Conguration.

conguration. Table 2-2 lists the AS

Table 2-2 AS

Parameter

Region A region is where an AS conguration

Name Species the name of the AS

Conguratio n Template

CPU Architecture

conguration parameters

Description Example Value

resides.

conguration to be created.

Select Create a new specications template.

If this option is selected, congure parameters, such as the vCPUs, memory, image, disk, and ECS type, to create a new AS

The following two types of CPU architectures are available:

● x86: The x86-based CPU architecture

● Kunpeng: The Kunpeng-based CPU

conguration.

uses Complex Instruction Set Computing (CISC).

architecture uses Reduced Instruction Set Computing (RISC).

N/A

Create a new

specications

template

x86

Auto Scaling User Guide 2 AS Conguration

Parameter Description Example Value

SpecicationsThe public cloud provides various ECS

types for dierent application scenarios.

For more information, see

Server User Guide

Elastic Cloud

Congure the ECS specications, including vCPUs, memory, image type, and disk, according to the ECS type.

Image ● Public image

A public image is a standard, widely used image. It contains an OS and preinstalled public applications and is available to all users. You can congure the applications or software in the public image as needed.

● Private image A private image is an image available only to the user who created it. It contains an OS, preinstalled public applications, and the user's private applications. Using a private image to create ECSs removes the need to congure multiple ECSs repeatedly.

● Shared image A shared image is a private image shared by another public cloud user.

Memoryoptimized ECS

Public image

Auto Scaling User Guide 2 AS Conguration

Parameter Description Example Value

Disk Includes system disks and data disks.

● System Disk Common I/O: uses Serial Advanced Technology Attachment (SATA) drives to store data.

High I/O: uses serial attached SCSI (SAS) drives to store data.

Ultra-high I/O: uses solid state disk (SSD) drives to store data.

If a full-ECS image is used, the system disk is restored using the disk backup. On the console, you can only change the volume type and size. In addition, the volume cannot be smaller than the disk backup.

● Data Disk You can create multiple data disks for an ECS. In addition, you can specify a data disk image for exporting data.

If the image you selected is of the full-ECS image type, you can change the volume type and size and encryption attributes of the data disk restored using the disk backup. Ensure that the volume size is greater than or equal to the disk backup size, and the encryption attributes can be modied only if the disk backup of the full-ECS image locates in the target region.

Common I/O System Disk

for

Security Group

Controls ECS access within or between security groups by dening access rules.

N/A

ECSs added to a security group are protected by the access rules you dene.

Auto Scaling User Guide 2 AS Conguration

Parameter Description Example Value

EIP An EIP is a static public IP address bound

to an ECS in a VPC. Using the EIP, the ECS provides services externally.

The following options are provided:

● Do not use An ECS without an EIP cannot access the Internet. However, it can still be used as a service ECS or deployed in a cluster on a private network.

● Automatically assign An EIP with a dedicated bandwidth is automatically assigned to each ECS. You can set the bandwidth size.

NOTE

If you select Automatically assign, you need to specify Type, Billed By, and Bandwidth.

Bandwidth You can select Dedicated or Shared.

● Dedicated: The bandwidth can be used by only one EIP.

● Shared: The bandwidth can be used by multiple EIPs.

NOTE

● This parameter is available only when EIP is set to Automatically assign.

● If you select Dedicated, you can select Bandwidth or

● The shared bandwidth can be billed only by bandwidth. You can select a shared bandwidth to which the EIP is to be added.

Trac for Billed By.

Automatically assign

Shared

Auto Scaling User Guide 2 AS Conguration

Parameter Description Example Value

pair or a password.

● Key pair In this mode, keys are used for authenticating the users who attempt to log in to target ECSs. If you select this mode, create or import a key pair on the Key Pair page.

NOTE

If you use an existing key, make sure that you have saved the key Otherwise, logging in to the ECS will fail.

le locally.

Advanced Settings

This parameter allows you to congure ECS Group and User Data Injection.

You can select Do not congure or Congure now.

Admin@123

N/A

Auto Scaling User Guide 2 AS Conguration

Parameter Description Example Value

User Data Injection

For details, see

User Guide

Elastic Cloud Server

The following methods are available:

● As text: allows you to enter the user data in the text box below.

le: allows you to inject script les

● As or other les when you create an ECS. If you select As

le, the system automatically injects the les into a specied directory when creating an ECS.

– For Linux, specify the path for

storing the injected

le, for

example /etc/foo.txt.

– For Windows, the injected

le is automatically stored in the root directory of disk C. You only need to specify the foo. The

le name, such as

le name can contain

only letters and digits.

NOTE

● For Linux, if you use the password authentication mode, the user data injection function is unavailable.

● If the selected image does not support user data injection, the user data injection function is unavailable.

ECS Group An ECS group allows you to create ECSs

N/A on dierent hosts, thereby improving service reliability.

5. Click Create Now. The system displays a message indicating that the AS conguration is successfully created.

6. If you want to use the newly created AS conguration, add it to the AS group. For details, see Replacing AS

Conguration in an AS Group.

Auto Scaling User Guide 2 AS Conguration

2.4 Copying an AS Conguration

Scenarios

You can copy an existing AS conguration.

When copying an AS conguration, you can modify parameter settings, such as

conguration name, ECS specications, and image of the existing AS

the conguration to rapidly add a new AS conguration.

Procedure

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click the AS conguration, and click Copy in the Operation column.

4. On the Copy AS Name, Specications, and Image, and congure the ECS login mode based on service requirements.

5. Click OK.

Congurations tab, locate the row containing the target AS

Conguration page, modify parameter settings, such as

2.5 Deleting an AS

Scenarios

When you no longer use an AS conguration, you can delete it. An AS conguration can be deleted only when it is not used by any AS group. You can delete an AS conguration or multiple AS congurations in a batch.

Procedure

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click the AS AS conguration, and click Delete in the Operation column to delete this AS conguration. You can also select multiple AS congurations to be deleted and click Delete in the upper part of the AS in batches.

Conguration

Congurations tab page, locate the row containing the target

conguration list to delete them

NO TE

Auto Scaling User Guide 3 AS Policy

3 AS Policy

3.1 Overview

AS policies can trigger scaling actions to adjust bandwidth or the number of instances in an AS group. An AS policy action and the operation to be performed in a scaling action. When the trigger condition is met, the system automatically triggers a scaling action.

When multiple AS policies are applied to an AS group, a scaling action is triggered as long as one of the AS polices is triggered, provided that the AS policies do not conict with each other.

AS supports the following policies:

● Alarm policy: AS automatically adjusts the number of instances in an AS group or sets the number of instances to the is generated for a congured metric, such as CPU Usage.

The following alarm policy types are available:

Simplied scaling: When a metric hits the congured threshold, an alarm

–

is generated, and AS automatically adjusts the number of instances in an AS group. For example, when the CPU usage is greater than or equal to 30%, an instance is added.

Rened scaling: When a metric falls within a range with both an upper

–

and a lower limit, an alarm is generated, and AS automatically adjusts the number of instances in an AS group. For example, if the CPU usage is greater than or equal to 30% and less than 40%, an instance is added.

denes the condition to trigger a scaling

congured value when an alarm

● Scheduled policy: AS automatically increases or decreases the number of instances in an AS group or sets the number of instances to the value at a specied time.

● Periodic policy: AS automatically increases or decreases the number of instances in an AS group or sets the number of instances to the value at a congured interval, such as daily, weekly, and monthly.

congured

Auto Scaling User Guide 3 AS Policy

Resource Adjustment Modes

● Dynamic

AS adjusts the number of instances or bandwidth when an alarm policy is triggered.

This mode is suitable for scenarios where workloads are unpredictable. Alarm policies are used to trigger scaling actions based on real-time monitoring data (such as CPU usage) to dynamically adjust the number of instances in the AS group.

● Planned

AS adjusts the number of instances or bandwidth when a periodic or scheduled policy is triggered.

This mode is suitable for scenarios where workloads are periodic.

● Manual

AS allows you to adjust resources by manually adding instances to an AS group, removing instances from an AS group, or changing the expected number of instances.

3.2 Creating an AS Policy

Scenarios

You can manage instances in an AS group through AS policies. This section describes how to create an AS policy.

Creating an Alarm Policy

1. Log in to the management console.

1. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

2. Locate the row containing the target AS group and click View AS Policy in the Operation column.

3. On the AS Policies page, click Add AS Policy.

4. Set the parameters listed in Table 3-1.

Table 3-1 AS policy parameters

Parameter

Policy Name Species the name of the AS

Description Example Value

policy to be created.

as-policy-p6g5

Policy Type Select Alarm. Alarm

Auto Scaling User Guide 3 AS Policy

Parameter Description Example Value

Alarm Rule Species whether a new alarm

rule is to be created (Create) or an existing alarm rule will be used (Use existing).

For details about how to use an existing alarm rule, see Setting

Monitoring Alarm Rules.

If you choose to create an alarm, system monitoring and custom monitoring are supported.

● System monitoring requires the parameters in Table 3-2.

● Custom monitoring requires the parameters in Table 3-3.

Alarm Policy Type

This parameter is valid only when Policy Type is set to Alarm. The value can be

Simplied scaling or Rened scaling.

● If you select

Simplied

scaling, you can set an action to be performed when a monitored metric hits the

congured

threshold, for example,

Monitored metric >, ≥, <, or ≤ xx.

● If you select

Rened

scaling, you can set the action to be performed when a monitored metric falls within a range with both an upper and a lower

limit, for example,

≤

Monitored metric < xx.

N/A

Simplied

scaling

Auto Scaling User Guide 3 AS Policy

Parameter Description Example Value

Scaling Action Species 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

specied value.

● Add 1 instance

● Add 10% instances The number of instances to be added is 10% of the current number of instances in the AS group. If the product of the current number of instances and the percentage is not an integer, AS automatically rounds the value up or down:

– Rounds down the

value that is greater than 1. For example, value 12.7 is rounded down to

12.

– Rounds up the

value that is greater than 0 and less than 1 to

1. For example, value 0.67 is rounded up to 1.

For example, there are 10 instances in an AS group, and the scaling action is Add 15% instances. When the AS policy is triggered, AS calculates the number of instances to be added is 1.5 and rounds 1.5 down to 1. After the scaling action is complete, there are 11 instances in the AS group.

Auto Scaling User Guide 3 AS Policy

Parameter Description Example Value

Cooldown Period

To prevent the alarm policy from being frequently triggered, you must set the cooldown period.

Cooldown period species how long any alarm-triggered scaling action will be disallowed after a previous scaling action is complete.

The cooldown period does not apply to scheduled or periodic scaling actions. However, AS recounts the cooling duration after a scheduled or periodic scaling action is complete.

For example, the cooldown period is set to 300 seconds, a scheduled policy is

specied to trigger a scaling action at 10:32, and a previous scaling action triggered by an alarm policy ends at 10:30. Any alarm-triggered scaling action will be denied during the cooldown period from 10:30 to 10:35, but scheduled scaling actions will still be triggered at 10:32. If the scheduled scaling action ends at 10:36, a new cooldown period starts from 10:36 and ends at 10:41.

NOTE

● If a scaling action is triggered by an AS policy, the cooldown period is that which is congured for that AS policy.

● If a scaling action is triggered by manually changing the expected number of instances or by other actions, the cooldown period is that which

congured for the AS group.

is The default cooldown period is 300 seconds.

300

Auto Scaling User Guide 3 AS Policy

Table 3-2 System monitoring parameters

Parameter Description Example Value

Alarm Rule Name

Monitoring Type

Trigger Condition

Monitoring Interval

Consecutive Occurrences

Species the name of the alarm rule. as-alarm-7o1u

Species the type of monitoring

metrics, which can be System

System

monitoring monitoring or Custom Monitoring. Select System monitoring.

Select monitoring metrics supported by AS and set alarm conditions for the

CPU Usage Max.

>70% metrics.

Species the interval at which the

5 minutes alarm status is updated based on the alarm rule.

Species the number of sampling

3 points when an alarm is triggered. If Occurrences is set to n, the sampling points of the alarm rule are the sampling points in n consecutive sampling periods. Only if all the sampling points meet the threshold congured for the alarm rule will the alarm rule status be refreshed as the

Alarm status.

Table 3-3 Custom monitoring parameters

Parameter

Description Example Value

Rule Name Species the name of the alarm rule. as-alarm-7o1u

Monitoring Type

Select Custom monitoring. Custom monitoring meets monitoring

Custom

monitoring

requirements in various scenarios.

Resource Type Species the name of the service for

AGT.ECS

which the alarm rule is congured.

Dimension Species the metric dimension of the

instance_id

alarm rule.

Monitored Object

Trigger Condition

Species the resources to which the alarm rule applies.

Select monitoring metrics supported by AS and set alarm conditions for

N/A

CPU Usage Max.

>70%

the metrics.

Auto Scaling User Guide 3 AS Policy

Parameter Description Example Value

Monitoring Interval

Consecutive Occurrences

5. Click OK.

The newly added AS policy is displayed on the AS Policy tab. In addition, the AS policy is in Enabled state by default.

Species the interval at which the alarm status is updated based on the alarm rule.

Species the number of sampling points when an alarm is triggered. If Occurrences is set to n, the sampling points of the alarm rule are the sampling points in n consecutive sampling periods. Only if all the sampling points meet the threshold congured for the alarm rule will the alarm rule status be refreshed as the

Alarm status.

Creating a Scheduled or Periodic Policy

1. Log in to the management console.

5 minutes

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Locate the row containing the target AS group and click View AS Policy in the Operation column.

4. On the AS Policies page, click Add AS Policy.

Congure the parameters listed in Table 3-4.

Table 3-4 Parameter description

Parameter

Policy Name Species the name of the AS

Description Example Value

as-policy-p6g5

policy to be created.

Auto Scaling User Guide 3 AS Policy

Parameter Description Example Value

Policy Type Select Scheduled or Periodic

for expanding resources at a specied time.

If you select Periodic, you are required to congure two more parameters:

● Interval

– One day

– One week

– One month

● Time Range Species a time range during which the AS policy can be triggered.

Time Zone The default value is GMT

+08:00.

GMT+08:00, Beijing, China time, is 8:00 hours ahead Greenwich Mean Time.

Triggered At Species a time at which the AS

policy is triggered.

N/A

GMT+08:00

N/A

Auto Scaling User Guide 3 AS Policy

Parameter Description Example Value

Scaling Action Species an action and the

number 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 specied value.

● Add 1 instance

● Rounds down the value that is greater than 1. For example, value 12.7 is rounded down to

12.

● Rounds up the value that is greater than 0 and less than 1 to

1. For example, value 0.67 is rounded up to 1.

Auto Scaling User Guide 3 AS Policy

Parameter Description Example Value

Cooldown Period

To prevent the alarm policy from being frequently triggered, you must set the cooldown period.

Species how long any alarmtriggered scaling action will be disallowed after a previous scaling action is complete.

The cooldown period does not apply to scheduled or periodic scaling actions. However, AS recounts the cooling duration after a scheduled or periodic scaling action is complete.

For example, the cooldown period is set to 300 seconds, a scheduled policy is

specied to trigger a scaling action at 10:32, and a previous scaling action triggered by an alarm policy ends at 10:30. Any alarmtriggered scaling action will be denied during the cooldown period from 10:30 to 10:35, but scheduled scaling actions will still be triggered at 10:32. If the scheduled scaling action ends at 10:36, a new cooldown period starts from 10:36 and ends at 10:41.

NOTE

● If a scaling action is triggered by an AS policy, the cooldown period is that which is congured for that AS policy.

● If a scaling action is triggered by manually changing the expected number of instances or by other actions, the cooldown period is that which

congured for the AS group.

is The default cooldown period is 300 seconds.

300

6. Click OK.

The newly added AS policy is displayed on the AS Policy tab. In addition, the AS policy is in Enabled state by default.

NO TE

Auto Scaling User Guide 3 AS Policy

If you have created scheduled or periodic AS policies that are triggered at the same time, AS will execute the one created later. This constraint does not apply to alarm-triggered AS policies.

3.3 Managing AS Policies

Scenarios

An AS policy triggered operation. If the conditions are met, a scaling action is triggered to perform the required operation.

This section describes how to manage an AS policy, including modifying, enabling, disabling, executing, and deleting an AS policy.

Modifying an AS Policy

Modify parameter settings of an AS policy if it cannot meet service requirements.

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling. Then click the AS Groups tab.

3. Locate the row containing the target AS group and click View AS Policy in the Operation column. On the displayed page, locate the row containing the target AS policy and choose More > Modify in the Operation column.

4. In the displayed Modify AS Policy dialog box, modify the parameters and click OK.

Enabling an AS Policy

An AS policy can trigger scaling actions only when it and the AS group are both enabled. You can enable one or more AS policies for an AS group as required.

species the conditions for triggering a scaling action as well as the

● Before enabling multiple AS policies, ensure that the AS policies do not conict with one another.

● An AS policy can be enabled only when its status is Disabled.

Locate the row containing the target AS group and click View AS Policy in the Operation column. On the displayed page, locate the row containing the target AS policy and click Enable in the Operation column. To concurrently enable multiple AS policies, select these AS policies and click Enable in the upper part of the AS policy list.

Disabling an AS Policy

Disable a within a specied period of time.

● If all AS policies of an AS group are disabled, no scaling action will be

specied AS policy if you do not want it to trigger any scaling action

triggered for this AS group. However, if you manually change the value of Expected Instances, a scaling action will still be triggered.

NO TE

Auto Scaling User Guide 3 AS Policy

● You can disable an AS policy only when its status is Enabled.

Locate the row containing the target AS group and click View AS Policy in the Operation column. On the displayed page, locate the row containing the target AS policy and click Disable in the Operation column. To concurrently disable multiple AS policies, select these AS policies and click Disable in the upper part of the AS policy list.

Manually Executing an AS Policy

Perform this operation to make the number of instances in an AS group reach the expected number of instances immediately.

● You can manually execute an AS policy if the scaling conditions the AS policy are not met.

● You can manually execute an AS policy only when the AS group and AS policy are both in Enabled state.

Locate the row containing the target AS group and click View AS Policy in the Operation column. On the displayed page, locate the row containing the target AS policy and click Execute Now in the Operation column.

● If Policy Type is set to Alarm and Alarm Policy Type to Rened scaling, the scaling policy cannot be executed immediately.

Deleting an AS Policy

Delete an AS policy that will not be used for triggering scaling actions.

An AS policy can be deleted even when the scaling action triggered based on the AS policy is in progress. Deleting the AS policy does not adversely progress scaling action.

Locate the row containing the target AS group and click View AS Policy in the Operation column. On the displayed page, locate the row containing the target AS policy and choose More > Delete in the Operation column.

congured in

aect the in-

To concurrently delete multiple AS policies, select these AS policies and click Delete in the upper part of the AS policy list.

Auto Scaling User Guide 4 Scaling Action

4 Scaling Action

4.1 Dynamically Expanding Resources

Before using AS to perform scaling actions, you must specify how to perform the scaling actions to dynamically expand resources.

If your demands change frequently, you can also dynamically expanding or reducing resources. When the conditions for triggering an AS policy are met, AS automatically changes the expected number of instances for triggering a scaling action to scale up or down resources. For details about how to create an alarm policy, see Creating an AS Policy.

For example, for a web application that allows users to purchase train tickets, when the CPU usage of the instances that run the application goes up to 90%, an instance needs to be added to ensure that services run properly. When the CPU usage goes down to 30%, an instance needs to be deleted to prevent resource waste. To meet the requirements, you can trigger condition of the rst policy is that the maximum CPU usage becomes greater than 90% and the action is to add one instance. For details, see Figure

4-1. In the second alarm policy, the trigger condition is that the minimum vCPU

usage is less than 30% and the action is to reduce an instance. For details, see

Figure 4-2.

congure two alarm policies. The

congure alarm policies for

Auto Scaling User Guide 4 Scaling Action

Figure 4-1 Alarm policy 01

Auto Scaling User Guide 4 Scaling Action

Figure 4-2 Alarm policy 02

4.2 Expanding Resources as Planned

To satisfy demands that change regularly, you can congure a scheduled or periodic policy to scale resources at how to create a scheduled or periodic policy, see Creating an AS Policy.

Take an online course selection web application as an example. This application is frequently used when a semester starts and seldom used in other periods of the semester. You can semester. The rst policy triggers a scaling action to add an instance when the course selection starts, and the second policy triggers a scaling action to reduce an instance when the course selection ends, meeting students' requirements as well as reducing cost.

congure two scheduled policies at the beginning of each

specied time or periodically. For details about

Auto Scaling User Guide 4 Scaling Action

4.3 Manually Expanding Resources

Scenarios

Expand resources by manually adding instances to an AS group, removing instances from an AS group, or changing the expected number of instances.

Procedure

Adding instances to an AS group

If an AS group is enabled and has no ongoing scaling action, and the current number of instances is less than the maximum, you can manually add instances to the AS group.

Before adding instances to an AS group, ensure that the following conditions are met:

● The instances are not in other AS groups.

● The instances must be in the same VPC as the AS group.

● The AZ to which the instances belong must be within the AZ to which the AS group belongs.

● After instances are added, the total number of instances is less than or equal to the maximum number of instances allowed.

● A batch operation can be performed on a maximum of 10 instances at a time.

To add instances to an AS group, perform the following steps:

1. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

2. Click the AS Groups tab and then the name of the target AS group.

3. On the AS group details page, click the Instances tab and then Add.

4. Select the instances to be added and click OK.

Removing instances from an AS group

You can remove an instance from an AS group, update the instance or rectify the instance fault, and add the instance to the AS group again. The instance removed out of the AS group does not carry application

For example, you can modify the AS However, the AS group does not update instances that are running. In such an event, terminate the instance and replace it in the AS group. Alternatively, remove the instance out of the AS group, update the instance software, and add the instance to the AS group again.

conguration for an AS group at any time.

trac any more.

Restrictions on instance removal are as follows:

● The AS group does not have a scaling action that is being performed, the instances are enabled, and the total number of instances after removal is not less than the minimum number of instances allowed.

Auto Scaling User Guide 4 Scaling Action

● Instances can be removed from an AS group and deleted only if the AS group has no scaling action ongoing, and the instances are automatically created and enabled, and are not used in Storage Disaster Recovery Service (SDRS).

● The instances automatically added to an AS group are billed in Pay-per-use mode by default. When you click Remove for such an instance, the system only removes it out of the AS group but does not delete it. When you click Remove and Delete, the system removes the instance out of the AS group and deletes it.

● If you manually change the billing mode of an instance in an AS group from Pay-per-use to Yearly/Monthly, when you click Remove and Delete for the instance, the system only removes it out of the AS group but does not delete it.

● Instances manually added to an AS group can only be removed, and cannot be removed and deleted.

● A batch operation can be performed on a maximum of 10 instances at a time.

To remove an instance from an AS group, perform the following steps:

1. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

2. Click the AS Groups tab and then the name of the target AS group.

3. Click the Instances tab, locate the row containing the target instance, and click Remove or Remove and Delete in the Operation column.

4. To delete multiple instances from an AS group, select the check boxes in front of them and click Remove or Remove and Delete.

4.4

To delete all instances from an AS group, select the check box on the left of

Instance Name and click Remove or Remove and Delete.

Changing the expected number of instances

Manually change the expected number of instances to add or reduce the number of instances in an AS group for expanding resources.

For details, see Modifying an AS Group.

Conguring an Instance Removal Policy

When instances are automatically removed from your AS group, the instances that are not in the currently used AZs will be removed whether instances are evenly distributed in the currently used AZs. If the load among AZs is unbalanced, AS balances load among AZs when removing instances. If the load among AZs is balanced, AS removes instances following the congured instance removal policy.

AS supports the following instance removal policies:

● Oldest instance: The oldest instance is removed from the AS group this policy if you want to replace old instances by new instances in an AS group.

● Newest instance: The latest instance is removed from the AS group this policy if you want to test a new AS retain it.

rst. Besides, AS will check

conguration and do not want to

pre-

rst. Use

NO TE

Auto Scaling User Guide 4 Scaling Action

● Oldest instance created from oldest AS conguration: The oldest instance created based on the oldest conguration is removed from the AS group rst. Use this policy if you want to update an AS group and delete the instances created based on early AS congurations gradually.

● Newest instance created from oldest AS conguration: The latest instance created based on the oldest

A manually added ECS is removed in the lowest priority. AS does not delete a manually added ECS when removing it. If multiple manually added ECSs must be removed, AS preferentially removes the earliest-added ECS.

conguration is removed from the AS group rst.

4.5 Viewing a Scaling Action

Scenarios

To check whether a scaling action is performed or view scaling action details, perform the operations described in this section.

Viewing Scaling Actions

The following steps illustrate how to view scaling actions of an AS group.

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click the AS Groups tab and then the name of the target AS group.

4. Click the Monitoring tab and view scaling actions. On the Monitoring page, you can view changes in the number of instances and metrics such as CPU Usage.

Viewing Historical Scaling Actions

The following steps illustrate how to view the historical records of scaling actions of an AS group.

1. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

2. Click the AS Groups tab and then the name of the target AS group.

3. Click the Scaling Actions tab. This page displays historical scaling actions of an AS group, including instance scaling and load balancer migration.

Scaling Action ID, Status, Scaling Action Type, Description, Start Time, and

End Time of scaling actions are displayed. Click

ID to view the resource name, status, and failure cause. You can also use the

ltering function in the upper right corner to view scaling actions in a specied period.

before the scaling action

Auto Scaling User Guide 4 Scaling Action

4.6 Managing Lifecycle Hooks

Lifecycle hooks enable you to exibly control creation and removal of ECS instances in AS groups and manage the lifecycle of ECS instances in AS groups.

Figure 4-3 shows the instance lifecycle when no lifecycle hook is added to the AS

group.

Figure 4-3 Instance lifecycle statuses when no lifecycle hook is added to the AS group

Figure 4-4 shows the instance lifecycle when a lifecycle hook is added to the AS

group.

Figure 4-4 Instance lifecycle statuses when a lifecycle hook is added to the AS group

When the AS group performs a scaling action and triggers the lifecycle hook, the scaling action is suspended and the instance that is being added to or removed

Auto Scaling User Guide 4 Scaling Action

from the AS group is set to waiting state, as shown in 2 and 6 in Figure 4-4. During this period, you can perform some custom operations on the instance. For example, you can install or congure software on an instance to be added to the AS group. The suspension of a scaling action will be ended in either of the following scenarios:

● The time when the instance stays in waiting state is longer than the timeout duration.

● A callback operation is performed to stop the instance waiting state.

Application Scenarios

● Instances newly added to an AS group can be bound to a load balancer only after initialization (software installation or on the instances and services start running properly.

● Instances can be removed from an AS group only after they are unbound from the load balancer and have

● Before instances are removed from an AS group, data needs to be backed up and logs need to be downloaded.

● Other scenarios where custom operations need to be performed

nished processing ongoing requests.

conguration) has been performed

Working Rules

After added to an AS group, a lifecycle hook works as follows:

● Adding an ECS instance to an AS group

After an instance is added to an AS group and initialized, a lifecycle hook of the Instance adding type is automatically triggered. The instance enters the Wait (Adding to AS group) state, that is, the instance is suspended by the lifecycle hook. If you have congured a notication object, the system sends a message to the object. After receiving the message, you can perform custom operations, for example, installing software on the instance. After the custom operations, you can perform a callback operation to end the instance waiting state. Alternatively, you can wait until the timeout duration ends, when the system automatically ends the instance waiting state. After the instance waiting state ends, two types of Default Callback Action are available, Continue and Abandon.

– Continue: The instance in waiting state will be added to the AS group.

– Abandon: The instance in waiting state will be deleted and a new

instance will be created.

If you have will be triggered when an instance is added to the AS group. If the Default Callback Action of one lifecycle hook is Abandon, the instance will be deleted and a new instance will be created. If the Default Callback Action of all lifecycle hooks is Continue, the instance is added to the AS group after suspension by the last lifecycle hook is complete.

● Removing an instance from an AS group

When an instance is removed from an AS group, the instance enters the Removing from AS group state. After a lifecycle hook is triggered, the instance enters the Wait (Removing from AS group) state. The system sends messages to the

congured multiple Instance adding lifecycle hooks, all of them

congured notication object. After receiving the message,

nishing

Auto Scaling User Guide 4 Scaling Action

you can perform custom operations, such as uninstalling software and backing up data. After nishing the custom operations, you can perform the default callback operation or wait for the system to end the instance waiting state when the timeout duration expires. After the instance waiting state ends, two types of Default Callback Action are available, Continue and Abandon.

– Continue: The instance is removed from the AS group.

– Abandon: The instance is removed from the AS group.

If there are multiple lifecycle hooks, Continue allows suspension by other lifecycle hooks to time out. The instance will be removed from the AS group only when the status of all lifecycle hooks is Continue. If the Default Callback Action of any lifecycle hook is Abandon, the instance will be directly removed from the AS group.

Use Restrictions

● You can add, modify, or delete a lifecycle hook when the AS group does not perform a scaling action.

● Up to

ve lifecycle hooks can be added to one AS group.

Adding a Lifecycle Hook

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click the name of the AS group to which the lifecycle hook is to be added. On the AS group details page, click the Lifecycle Hooks tab and then Add Lifecycle Hook.

4. In the displayed Add Lifecycle Hook dialog box, set the parameters listed in

Table 4-1.

Table 4-1 Parameter description

Paramete r

Hook Name

Hook Type

Description Example

Value

Species the lifecycle hook name. The name is a string of 1 and 32 characters and can contain letters, digits, underscores (_), and hyphens (-).

Species the lifecycle hook type. The value can be Instance adding or Instance removal. Instance adding sets an instance that is being added to an AS group to Wait (Adding to AS group) state. Instance removal sets an instance that is being removed from an AS group Wait (Removing from AS group) state.

we12_w

Instance adding

Auto Scaling User Guide 4 Scaling Action

ParameterDescription Example

Value

Default Callback Action

Species the default system action when the waiting duration of an instance expires.

When an instance is being added to an AS group, the default callback action can be Continue or Abandon:

● Continue: When only one lifecycle hook is available, the instance will continue to be added to the AS group. When multiple lifecycle hooks are available, the instance will continue to be added to the AS group only when all lifecycle hooks change to Continue state.

● Abandon: The system deletes the instance and creates a new one only when one lifecycle hook is in Abandon state, regardless of whether one or more lifecycle hooks are available.

When an instance is being removed from an AS group, the default callback action can be Continue or Abandon:

● Continue: When only one lifecycle hook is available, the instance will be removed from the AS group. When multiple lifecycle hooks are available, the instance will be removed from the AS group only when all lifecycle hooks change to Continue state.

● Abandon: The system removes the instance from the AS group when one lifecycle hook is in Abandon state, regardless of whether one or more lifecycle hooks are available.

Continue

Timeout Duration (s)

Species the instance waiting duration by default. The value ranges from 300s to 86,400s.

You can prolong the timeout duration or

3600

perform the Continue or Abandon operation before the timeout duration expires. For more information about callback actions, see

Performing a Callback Action.

Auto Scaling User Guide 4 Scaling Action

ParameterDescription Example

Value

Noticatio n Topic

Species a notication object for a lifecycle

hook. For details, see "Creating a Topic" in

Simple Message Notication User Guide

. When an instance is suspended by the lifecycle hook, the system sends a notication to the object. This

notication contains the basic instance information, your customized notication content, and the token for controlling lifecycle operations. An example notication is as follows:

{ "service": "AutoScaling", "tenant_id": "93075aa73f6a4fc0a3209490cc57181a", "lifecycle_hook_type": "INSTANCE_LAUNCHING", "lifecycle_hook_name": "test02", "lifecycle_action_key": "4c76c562-9688-45c6b685-7fd732df310a",

"notication_metadata": "xxxxxxxxxxxxx", "scaling_instance": { "instance_id": "89b421e4-5fa6-4733-bf40-6b07a8657256", "instance_name": "instance_ip": "192.168.0.202" }, "scaling_group": { "scaling_group_id": "fe376277-50a6-4e36bdb0-685da85f1a82", "scaling_group_name": "as-group-wyz01",

"scaling_cong_id": "16ca8027-b6cc-45fc-

af2d-5a79996f685d",

"scaling_cong_name": "as-cong-kxeg" } }

"as-cong-kxeg_RM6OCREY",

N/A

Noticatio n

Message

After a notication object is congured, the system sends your customized notication to the object.

5. Click OK.

The added lifecycle hook is displayed on the Lifecycle Hooks page.

Performing a Callback Action

1. On the AS Groups page, click the name of the target AS group.

2. On the displayed page, click the Instances tab.

3. Locate the instance that has been suspended by the lifecycle hook and click

Wait (Adding to AS group) or Wait (Removing from AS group) in the Lifecycle Status column, as shown in Figure 4-5.

N/A

NO TE

Auto Scaling User Guide 4 Scaling Action

Figure 4-5 Performing a callback operation

Callback operations can only be performed on instances that have been suspended by a lifecycle hook.

4. In the displayed Added Hook dialog box, view the suspended instance and all the lifecycle hooks, and perform callback actions on lifecycle hooks, as shown in Figure 4-6.

Figure 4-6 Added Hook dialog box

Callback actions include:

– Continue

– Abandon

– Extend timeout

If you have performed customized operations before the timeout duration expires, select Continue or Abandon to complete the lifecycle operations. For details about Continue and Abandon, see Table 4-1. If you require more time for customizing operations, select Extend timeout to prolong the timeout

NO TE

Auto Scaling User Guide 4 Scaling Action

duration. Then, the instance waiting duration will be prolonged by 3600 sections each time.

Modifying a Lifecycle Hook

On the Lifecycle Hooks page, locate the target lifecycle hook and click Modify in the Operation column, see Table 4-1 for parameters. You can modify the parameter except Hook Name, such as Hook Type, Default Callback Action, and

Timeout Duration.

Deleting a Lifecycle Hook

On the Lifecycle Hooks page, locate the target lifecycle hook and click Delete in the Operation column.

4.7 Conguring Instance Protection

Scenarios

Congure instance protection if you want to specify one or more ECSs not to be automatically removed from an AS group. After the conguration, when AS automatically reduces the number of ECSs in an AS group, the in-service ECSs with instance protection enabled will not be removed.

Prerequisites

In the following scenarios, ECSs will still be removed from the AS group even if instance protection is enabled:

● The ECS is not healthy according to the health check.

● The ECS is manually removed from the AS group.

● Instances in the abnormal health status cannot provide services. The AS group preferentially ensures that all instances in the group are normal. Therefore, instance protection cannot protect abnormal instances.

● By default, instance protection does not take eect on the ECSs that are newly created in or added to an AS group.

● Instance protection becomes invalid immediately when the target ECS is removed from the AS group.

Enabling Instance Protection

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click the name of the target AS group.

4. Click the Instances tab. Select one or more ECSs and choose Enable Instance

Protection from the More drop-down list. In the displayed Enable Instance Protection dialog box, click Yes.

Auto Scaling User Guide 4 Scaling Action

You can also locate the row containing the target ECS and click Enable

Instance Protection in the Operation column. Then, in the Enable Instance Protection dialog box, click Yes.

Disabling Instance Protection

1. Log in to the management console.

1. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

2. Click the name of the target AS group.

3. Click the Instances tab. Select one or more ECSs and choose Disable Instance

Protection from the More drop-down list. In the displayed Disable Instance Protection dialog box, click Yes.

You can also locate the row containing the target ECS and click Disable Instance Protection in the Operation column. Then, in the Disable Instance Protection dialog box, click Yes.

4.8 Standby Instance

If you want to stop distributing not want to remove them from the AS group, you can set the instances to standby mode. You can set one or more instances in your AS group to standby mode, and stop or restart these instances without removing them from the AS group.

Application Scenarios

You cannot control the lifecycle of ECSs in an AS group. The AS group removes unhealthy ECSs and does not allow you to stop or restart these ECSs. As a result, you cannot use some functions provide by the ECS service, such as resetting the password, reinstalling the OS, and changing the OS.

By setting ECSs to standby mode, you can control their lifecycle and perform operations on the ECSs (such as stopping the ECSs) as you need. This facilitates your management of ECSs in your AS group and applies to various scenarios.

● If you want to change the OS of an ECS added by a scaling action or stop the ECS, you can set the ECS to standby mode. Then you can perform all operations supported by the ECS service. After completing the operations, cancel standby mode for the ECS.

For example, you can change the AS time. This conguration will be used by any instance that is created in the AS group. However, the AS group does not update instances that are running. You can stop these instances, and the AS group will replace them. Alternatively, you can set the instances to standby mode, update software on them, and then cancel standby mode for them.

● If an instance in your AS group associated with a load balancer becomes faulty, you can set the instance to standby mode, after which the load balancer will no longer distribute access log in to the instance, locate and rectify the fault, and restart the instance. After the instance recovers, cancel standby mode for the instance to receive

trac again.

trac to some instances in your AS group but do

conguration for your AS group at any

trac to the instance. Then you can

NO TE

Auto Scaling User Guide 4 Scaling Action

Working Rules

● Set instances to standby mode.

After you set an instance to standby mode, the instance will be automatically unbound from the load balancer associated with the AS group. The instance is still in the AS group, but no health check will be performed on the instance. In this case, load on other instances will increase. To reduce load on other instances and ensure proper service running, you can select Add the same number of new instances to the AS group when setting the instance to standby mode.

● An instance can be set to standby mode only when the instance is enabled and the AS group has no ongoing scaling action.

● Scaling actions will not remove standby instances from the AS group.

● You can manually remove standby instances from the AS group.

● Cancel standby mode for instances.

After you cancel standby mode for an instance, it will be in running state and receive instance will be bound to the load balancer. After the instance starts running properly, health check will be performed on it again.

trac again. If a load balancer is associated with the AS group, the

Standby mode can be canceled for an instance only when the instance is in standby mode and the AS group has no ongoing scaling action.

Setting Instances to Standby Mode

1. Log in to the management console.

2. Click Service List.

3. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

4. Click the name of the target AS group.

5. Click the Instances tab. Select one or more instances, click More, and select

Set to Standby in the drop-down list. In the displayed dialog box, select Add the same number of new instances to the AS group as you need and click Yes.

Canceling Standby Mode for Instances

1. Log in to the management console.

1. Click Service List.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click the name of the target AS group.

4. Click the Instances tab. Select one or more instances, click More, and select Cancel Standby in the drop-down list. In the displayed dialog box, click Yes.

Auto Scaling User Guide 5 Bandwidth Scaling

5 Bandwidth Scaling

5.1 Creating a Bandwidth Scaling Policy

Scenarios

You can automatically adjust the your purchased EIP bandwidth and shared bandwidth using a bandwidth scaling policy. This section describes how to create an bandwidth scaling policy.

When creating a bandwidth scaling policy, you need to information. The system supports three types of bandwidth scaling policies: alarmbased, scheduled, and periodic.

The basic information for creating a bandwidth scaling policy includes the policy name, resource type, policy type, and trigger condition.

Creating an Alarm-based Bandwidth Scaling Policy

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Bandwidth Scaling.

3. Click Create Bandwidth Scaling Policy.

4. Set parameters, such as the policy name, policy type, and trigger condition. For details, see Creating a Bandwidth Scaling Policy.

Table 5-1 Alarm policy parameters

congure basic

Parameter

Region Species the region where the AS group resides. N/A

Policy Name

Description Example

Value

Species the name of the bandwidth scaling policy.

The name consists of only letters, digits, underscores (_), and hyphens (-).

N/A

Auto Scaling User Guide 5 Bandwidth Scaling

Parameter Description Example

Value

Resource Type

Species the type of the bandwidth to be managed. You can select EIP or Shared

EIP

bandwidth. Shared bandwidths do not support alarm policies. Set Resource Type to EIP.

EIP Species the public network IP address whose

N/A

bandwidth needs to be scaled.

NOTE

Currently, only pay-per-use EIPs can be scaled. Yearly/ monthly EIPs cannot be used to create AS groups.

Policy Type Select Alarm. Alarm

Alarm Rule You can use an existing alarm rule or create a

N/A new one. Alternatively, click Create Alarm Rule on the right side of the Alarm parameter and create an alarm rule on the Alarm Rules page. For details, see Creating an Alarm Rule.

To create an alarm rule,

congure the following

parameters:

● Rule Name

Species the name of the new alarm rule, for example, as-alarm-7o1u.

● Trigger Condition

Select a monitoring metric and trigger condition based on the metric. Table 5-2 lists the supported monitoring metrics. An example value is Outbound

Trac Avg. >

100 bit/s.

● Monitoring Interval

Species the period for the metric, for example, 5 minutes.

● Consecutive Occurrences

Species the number of consecutive times, for example, one time, for triggering a scaling action during a monitoring period.

Alarm Policy Type

This parameter is valid only when Policy Type is set to Alarm. The value can be Simplied

Simplied

scaling scaling or Rened scaling.

● If you select Simplied scaling, you can set

an action to be performed when a monitored metric hits the

congured threshold, for

example, Monitored metric >, ≥, <, or ≤ xx.

● If you select Rened scaling, you can set the

action to be performed when a monitored metric falls within a range with both an

upper and a lower limit, for example,

≤

Monitored metric < xx.

Auto Scaling User Guide 5 Bandwidth Scaling

Parameter Description Example

Value

Scaling Action

Species the execution action in the AS policy.

The following scaling action options are available:

● Add

When a scaling action is triggered, the bandwidth is increased.

● Reduce

When a scaling action is triggered, the bandwidth is decreased.

● Set to

The bandwidth is set to a

NOTE

The step (minimum unit for bandwidth adjustment) varies depending on the bandwidth value range. The bandwidth will be automatically adjusted to the nearest value according to the actual step.

● If the bandwidth is less than or equal to 300 Mbit/s, the default step is 1 Mbit/s.

● If the bandwidth ranges from 300 Mbit/s to 1000 Mbit/s, the default step is 50 Mbit/s.

● If the bandwidth is greater than 1000 Mbit/s, the default step is 500 Mbit/s.

xed value.

N/A

Cooldown Period

Species a period of time in the unit of second after each scaling action is complete. During the

300

cooldown period, scaling actions triggered by alarms will be denied. Scheduled and periodic scaling actions are not restricted.

Table 5-2 Monitoring metrics supported by the alarm policy

Metric

Description

Inbound Bandwidth Indicates the network rate of inbound trac.

Inbound Trac Indicates the network trac going into the cloud

platform.

Outbound Bandwidth Indicates the network rate of outbound trac.

Outbound Trac Indicates the network trac going out of the cloud

platform.

Outbound Bandwidth Usage

Indicates the usage of network rate of outbound trac Unit: Percent.

5. After setting the parameters, click Create Now.

Auto Scaling User Guide 5 Bandwidth Scaling

The newly created bandwidth scaling policy is displayed on the Bandwidth

Scaling page and is in Enabled state by default.

Creating an Alarm Rule

When creating an alarm-based bandwidth scaling policy, you can click Create Alarm Rule on the right of Rule Name to create an alarm rule. To do so, perform

the following operations:

1. Click Create Alarm Rule on the right of Rule Name to switch to the Alarm Rules page of Cloud Eye.

2. On the Alarm Rules page, click Create Alarm Rule in the upper right corner.

3. Set parameters based on Figure 5-1 and Table 5-3. For more information about how to set alarm rules, see

Figure 5-1 Creating an alarm rule

Cloud Eye User Guide

Auto Scaling User Guide 5 Bandwidth Scaling

Table 5-3 Key parameters for creating an alarm rule

ParameterDescription Example Value

Name Species the name of the alarm rule. alarm-

bandwidth

Resource Type

Dimension Species the item of the monitored service.

Monitorin g Scope

Method Select the method of creating an alarm. Set

Alarm Policy

4. After setting the parameters, click Create.

Species the name of the service to which the alarm rule applies. Set this parameter to Elastic IP and Bandwidth.

Bandwidth scaling adjusts the bandwidth. Therefore, set this parameter to Bandwidths.

Species the resources to which the alarm rule applies. Set this parameter to Specic resources. Search for resources by bandwidth name or ID, which can be obtained on the page providing details about the target EIP.

this parameter to Create manually.

Species the alarm policy for triggering the alarm rule. Set this parameter as required. For details about the monitoring metrics, see Table 5-2.

Elastic IP and Bandwidth

Bandwidths

Specic

resources

Create manually

N/A

5. On the Create Bandwidth Scaling Policy page, click on the right of Rule Name, and select the created alarm rule.

Alternatively, create your desired alarm rule on the Cloud Eye page before creating a bandwidth scaling policy. Ensure that the during alarm rule creation are the bandwidth of the EIP selected for the bandwidth scaling policy to be created. After the alarm rule is created, you can select the rule when creating a bandwidth scaling policy.

specic resources selected

Creating a Scheduled or Periodic Bandwidth Scaling Policy

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Bandwidth Scaling.

3. Click Create Bandwidth Scaling Policy.

4. Set parameters, such as the policy name, policy type, and trigger condition. For details, see Table 5-4.

Auto Scaling User Guide 5 Bandwidth Scaling

Table 5-4 Scheduled or periodic policy parameters

ParameterDescription Example

Value

Region Species the region where the AS group resides. N/A

Policy Name

Species the name of the bandwidth scaling policy.

The name consists of only letters, digits, underscores (_), and hyphens (-).

Resource Type

Species the type of the bandwidth to be managed. You can select EIP or Shared bandwidth.

EIP Species the public network IP address whose

bandwidth needs to be scaled. This parameter is mandatory when Resource Type is set to EIP.

NOTE

Currently, only pay-per-use EIPs can be scaled. Yearly/ monthly EIPs cannot be used to create AS groups.

Shared Bandwidt h

Policy Type

Species the shared bandwidth to be scaled. This parameter is mandatory when Resource Type is set to Shared bandwidth.

Species the policy type. You can select a scheduled or periodic policy.

If you select Periodic, you are required to congure two more parameters:

● Time Range Species a time range during which the AS policy can be triggered.

● Interval

– One day

– One week

– One month

as-policyp6g5

EIP

N/A

TriggeredAtSpecies a time at which the AS policy is

N/A

triggered.

Auto Scaling User Guide 5 Bandwidth Scaling

ParameterDescription Example

Value

Scaling Action

Cooldow n Period

Species the action to be performed.

The following scaling action options are available:

● Add When a scaling action is triggered, the bandwidth is increased.

● Reduce When a scaling action is triggered, the bandwidth is decreased.

● Set to The bandwidth is set to a

NOTE

The step (minimum unit for bandwidth adjustment) varies depending on the bandwidth value range. The bandwidth will be automatically adjusted to the nearest value according to the actual step.

● If the bandwidth is less than or equal to 300 Mbit/s, the default step is 1 Mbit/s.

● If the bandwidth ranges from 300 Mbit/s to 1000 Mbit/s, the default step is 50 Mbit/s.

● If the bandwidth is greater than 1000 Mbit/s, the default step is 500 Mbit/s.

xed value.

Species a period of time in the unit of second

after each scaling action is complete. During the cooldown period, scaling actions triggered by alarms will be denied. Scheduled and periodic scaling actions are not restricted.

N/A

300

5. After setting the parameters, click Create Now.

5.2 Viewing Details About a Bandwidth Scaling Policy

Scenarios

You can view details about a bandwidth scaling policy, including its basic information and execution logs. Policy execution logs record details about policy execution. This section describes how to create an bandwidth scaling policy.

Procedure

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Bandwidth Scaling.

3. On the Bandwidth Scaling page, click the name of a bandwidth scaling policy to go to the page showing its basic information and view its details. You can view basic information about the scaling policy, including Policy Type, Trigger Condition, and Scaling Action.

NO TE

Auto Scaling User Guide 5 Bandwidth Scaling

Viewing Execution Logs of a Bandwidth Scaling Policy

In the Policy Execution Logs area on the bandwidth scaling policy details page, you can view the policy execution logs. Policy execution logs record the execution status, execution time, original value, and target value of a bandwidth scaling policy.

5.3 Managing a Bandwidth Scaling Policy

Scenarios

You can adjust the bandwidth through a bandwidth scaling policy.

This section describes how to manage bandwidth scaling policies, including enabling, disabling, modifying, deleting, and immediately executing a bandwidth scaling policy.

Enabling a Bandwidth Scaling Policy

A bandwidth scaling policy can be enabled only when its status is Disabled.

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Bandwidth Scaling.

3. In the bandwidth scaling policy list, locate the row containing the target policy and click Enable in the Operation column.

4. In the displayed Enable Bandwidth Scaling Policy dialog box, click Yes.

Disabling a Bandwidth Scaling Policy

A bandwidth scaling policy can be disabled only when its status is Enabled.

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Bandwidth Scaling.

3. In the bandwidth scaling policy list, locate the row containing the target policy and click Disable in the Operation column.

4. In the displayed Disable Bandwidth Scaling Policy dialog box, click Yes.

After a bandwidth scaling policy is disabled, its status changes to Disabled. AS does not automatically trigger any scaling action based on a Disabled bandwidth scaling policy.

Modifying a Bandwidth Scaling Policy

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Bandwidth Scaling.

3. In the bandwidth scaling policy list, locate the row containing the target policy and click the policy name to switch to its details page.

NO TE

Auto Scaling User Guide 5 Bandwidth Scaling

Click Modify in the upper right corner of the page.

You can also locate the row containing the target policy, click More in the Operation column, and select Modify.

4. Modify parameters. You can modify the following parameters of a bandwidth scaling policy: Policy Name, EIP, Policy Type, Scaling Action, and Cooldown Period.

5. Click OK.

A bandwidth scaling policy which is being executed cannot be modied.

Deleting a Bandwidth Scaling Policy

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Bandwidth Scaling.

3. In the bandwidth scaling policy list, locate the row containing the target policy, click More in the Operation column, and select Delete.

4. In the displayed Delete Bandwidth Scaling Policy dialog box, click Yes.

You can also select one or more scaling policies and click Delete above the list to delete one or more scaling policies.

● You can delete a bandwidth scaling policy when you no longer need it. If you do not need it only during a specied period of time, you are advised to disable rather than delete it.

● A bandwidth scaling policy can be deleted only when it is not being executed.

Executing a Bandwidth Scaling Policy

By executing a bandwidth scaling policy, you can immediately adjust the bandwidth to that wait until the trigger condition is met.

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Bandwidth Scaling.

3. In the bandwidth scaling policy list, locate the row that contains the target policy and click Execute Now in the Operation column.

4. In the displayed Execute Bandwidth Scaling Policy dialog box, click Yes.

You can also go to the bandwidth scaling policy details page and click Execute Now in the upper right corner.

congured in the bandwidth scaling policy, instead of having to

NO TE

Auto Scaling User Guide 5 Bandwidth Scaling

● A bandwidth scaling policy can be executed only when the policy is enabled and no other bandwidth scaling policy is being executed.

● Executing a bandwidth scaling policy does not aect automatic adjustment of the bandwidth when the trigger condition of the policy is met.

● If Policy Type is set to Alarm and Alarm Policy Type to bandwidth scaling policy cannot be executed immediately.

Rened scaling, tjhe

Auto Scaling User Guide 6 AS Group and Instance Monitoring

6 AS Group and Instance Monitoring

6.1 Health Check

A health check removes abnormal instances from an AS group. Then, AS adds new instances to the AS group so that the number of instances is the same as the expected number. There are two types of AS group health check.

● ECS health check: checks ECS running status. If an ECS is stopped or deleted, it is considered as abnormal. ECS health check is the default health check mode for an AS group. The AS group periodically uses the check result to determine the running status of every instance in the AS group. If the health check result shows that an ECS is faulty, AS removes the ECS from the AS group.

● ELB health check: determines ECS running status using a load balancing listener. If the AS group uses load balancers, the health check method can also be ELB health check. If you add multiple load balancers to an AS group, the ECS is considered normal only when all load balancers detect that the ECS status is normal. If any load balancer detects that the ECS is abnormal, the ECS will be removed from the AS group.

In both ECS health check and ELB health check modes, AS removes abnormal instances from AS groups. However, the removed instances are processed dierently in the two modes:

For instances automatically added to an AS group during scaling actions, AS removes and deletes them. For instances manually added to an AS group, AS only removes them from the AS group.

● Automatically added instances are billed in Pay-per-use mode by default. In such an event, AS removes and deletes such an instance if it becomes abnormal. For an instance that is manually added to the AS group, AS only removes the instance out of the AS group.

● If you manually change the billing mode of an instance from Pay-per-use to Yearly/Monthly, AS only removes the instance out of the AS group.

When an AS group is disabled, checking instance health status continues. However, AS will not remove instances.

Auto Scaling User Guide 6 AS Group and Instance Monitoring

6.2 Conguring Notication for an AS Group

Scenarios

After the SMN service is provisioned, you can promptly send AS group information, such as successful instance increasing, failed instance increasing, successful instance decreasing, failed instance decreasing, or AS group exception to the user using the SMN service. This helps the user learn the AS group status.

congure notication for an AS group, you need to specify a notication event

To and topic. You can notication topic is pre-congured on the SMN console. When the live network complies with the group sends a notication to the user.

Procedure

1. Log in to the management console.

congure up to ve notications for a AS group. The

notication scenario that matches the notication topic, the AS

1. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling. Then click the AS Groups tab.

2. Click the name of the target AS group. On the AS group details page, click the Notication tab and then click Add Notication.

3. Set the parameters listed in Table 6-1.

Table 6-1 Parameter description

Parameter

Send NoticationToSelect a created topic. For how to

Notication

Conditions

Description Example Value

f123

create a topic, see

Notication User Guide

When at least one of the following conditions is met, SMN sends a notication to the user:

● Instance creation succeeds

● Instance creation fails

● Instance removal succeeds

● Instance removal fails

● Errors occur in an AS group

Simple Message

N/A

4. Click Save.

Auto Scaling User Guide 6 AS Group and Instance Monitoring

6.3 Recording AS Resource Operations

Scenarios

AS can use the Cloud Trace Service (CTS) to record resource operations. CTS can record operations performed on the management console, operations performed by calling APIs, and operations triggered within the cloud system.

If you have enabled CTS, when a call is made to the AS API, the operation will be reported to CTS which will then deliver the operation record to a bucket for storage. With CTS, you can record operations associated with AS for later query, audit, and backtrack operations.

Obtaining AS Information in CTS

After you enable CTS in the application system, the system logs the API calling operations performed on AS resources. On the Cloud Trace Service console, you can view operation records for the last 7 days. To obtain more operation records, you can enable the Object Storage Service (OBS) and synchronize operation records to the OBS in real time.

specied OBS

Table 6-2 list the AS operations that can be recorded by CTS.

Table 6-2 AS operations that can be recorded by CTS

Operation

Creating an AS group

Modifying an AS group

Deleting an AS group

Enabling an AS group

Disabling an AS group

Creating an AS

conguration

Deleting an AS

conguration

Resource Type Trace Name

scaling_group createScalingGroup

scaling_group modifyScalingGroup

scaling_group deleteScalingGroup

scaling_group enableScalingGroup

scaling_group disableScalingGroup

scaling_conguration createScalingConguration

scaling_conguration deleteScalingConguration

Deleting AS

congurations

in batches

scaling_conguration batchDeleteScalingConguration

Auto Scaling User Guide 6 AS Group and Instance Monitoring

Operation Resource Type Trace Name

Creating an AS policy

Modifying an AS policy

Deleting an AS policy

Enabling an AS policy

Disabling an AS policy

Executing an AS policy

Removing an instance

Removing instances in batches

Adding instances in batches

scaling_policy createScalingPolicy

scaling_policy modifyScalingPolicy

scaling_policy deleteScalingPolicy

scaling_policy enableScalingPolicy

scaling_policy disableScalingPolicy

scaling_policy executeScalingPolicy

scaling_instance removeInstance

scaling_instance batchRemoveInstances

scaling_instance batchAddInstances

Enabling instance protection in a batch

Disabling instance protection in a batch

Conguring a notication

Deleting a

notication

Creating a lifecycle hook

Modifying a lifecycle hook

Deleting a lifecycle hook

scaling_instance batchProtectInstances

scaling_instance batchUnprotectInstances

scaling_notication putScalingNotication

scaling_notication deleteScalingNotication

scaling_lifecycle_hook createLifecycleHook

scaling_lifecycle_hook modifyLifecycleHook

scaling_lifecycle_hook deleteLifecycleHook

Auto Scaling User Guide 6 AS Group and Instance Monitoring

Viewing Audit Logs

1. Log in to the management console.

2. Click

in the upper left corner to select a region and a project.

3. Click Trace List in the navigation pane on the left.

4. You can use

lters to query traces. The following lters are available:

– Trace Source, Resource Type, and Search By

Select a lter criterion from the drop-down list.

When you select Trace name for Search By, you need to select a specic trace name.

When you select Resource ID for Search By, you need to select or enter a specic resource ID.

When you select Resource name for Search By, you need to select or enter a specic resource name.

– Operator: Select a specic operator (at user level rather than tenant

level).

– Trace Status: Available options include All trace statuses, normal,

warning, and incident. You can only select one of them.

– Start time and end time: You can specify the time period to query traces.

5. Click on the left of the required trace to expand its details.

6. Locate the required trace and click View Trace in the Operation column. A dialog box is displayed, showing the trace content.

CTS Log Entries

Each log entry consists of a trace in JSON format. A log entry indicates an AS API request, including the requested operation, the operation date and time, operation parameters, and information about the user who sends the request. The user information is obtained from the Identity and Access Management (IAM) service.

The following example shows CTS log entries for the CreateScalingPolicy action:

{ "time": "2016-12-15 15:27:40 GMT+08:00", "user": { "name": "xxxx", "id": "domain": { "name": "xxx", "id": "30274282b09749adbe7d9cabeebcbe8b" } }, "request": { "scaling_policy_name": "as-policy-oonb", "scaling_policy_action": { "operation": "ADD", "instance_number": 1 }, "cool_down_time": "", "scheduled_policy": { "launch_time": "2016-12-16T07:27Z" }, "scaling_policy_type": "SCHEDULED", "scaling_group_id": "ec4051a7-6fbd-42d2-840f-2ad8cdabee34" }, "response": {

"6283d2920e4d3d917e6fa5e31ddebe",

Auto Scaling User Guide 6 AS Group and Instance Monitoring

"scaling_policy_id": "6a38d488-664b-437a-ade2-dc45f96f7f4c" }, "code": 200, "service_type": "AS", "resource_type": "scaling_policy", "resource_name": "as-policy-oonb", "resource_id": "6a38d488-664b-437a-ade2-dc45f96f7f4c", "source_ip": "10.190.205.233", "trace_name": "createScalingPolicy", "trace_rating": "normal", "trace_type": "ConsoleAction", "api_version": "1.0", "record_time": "2016-12-15 15:27:40 GMT+08:00", "trace_id": "f627062b-c297-11e6-a606-eb2c0f48bec5" }

6.4 Adding Tags to AS Groups and Instances

Scenarios

If you have many resources of the same type, you can use a tag to manage resources allocated to them.

exibly. You can identify specied resources quickly using the tags

Using a tag, you can assign custom data to each AS group. You can organize and manage AS groups, for example, classify AS group resources by usage, owner, or environment.

Each tag contains a key and a value. You can specify the key and value for each tag. A key can be a category associated with certain values, such as usage, owner, and environment.

For example, if you want to distinguish the test environment and production environment, you can allocate a tag with the key environment to each AS group. For the test environment, the key value is test and for the production environment, the key value is production. You are advised to use one or more groups of consistent tags to manage your AS group resources.

After you allocate a tag to an AS group, the system will automatically add the tag to the instances automatically created in the AS group. If you add a tag to an AS group or modify the tag, the new tag will be added to the ECSs automatically created in the AS group. Creating, deleting, or modifying the tag of an AS group will have no impact on the ECSs in the AS group.

Restrictions of Using Tags

You must observe the following rules when using tags:

● Each AS group can have a maximum of 10 tags added to it.

● Each tag contains a key and a value.

● You can set the tag value to an empty character string.

● If you delete an AS group, all tags of it will also be deleted.

Adding a Tag to an AS Group

1. Log in to the management console.

Auto Scaling User Guide 6 AS Group and Instance Monitoring

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. Click the AS group name. On the AS group details page, click the Tags tab and then click Add Tag.

4. Set the parameters listed in Table 6-3.

Table 6-3 Tag naming rules

Parameter Requirement Example Value

Tag Key ● The value cannot be empty.

● An AS group has a unique key.

● A key can contain a maximum of 36 characters, including digits, letters, underscores (_), hyphens (-), and Unicode characters from \u4e00 to

\u9f.

Tag Value ● The value can be an empty

character string.

● A key can have only one value.

● A tag value can contain a maximum of 43 characters, including digits, letters, underscores (_), periods (.), hyphens (-), and Unicode characters from \u4e00 to

5. Click OK.

Modifying or Deleting Tags of an AS Group

Organization

Apache

\u9f.

1. Log in to the management console.

1. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling. Then click the AS Groups tab.

2. Click the AS group name. On the Basic Information page, click the Tags tab.

3. Locate the row that contains the tag and click Edit or Delete in the Operation column.

After clicking Edit,

After you click Delete, the added tag will be deleted.

congure required parameters. For details, see Table 6-3.

6.5 Monitoring Metrics

Description

This section describes the monitoring metrics reported by AS to Cloud Eye and denes the namespace for the metrics. You can use Cloud Eye to query monitoring metrics and alarms generated by AS.

Auto Scaling User Guide 6 AS Group and Instance Monitoring

Namespace

SYS.AS

Monitoring metrics

Table 6-4 lists the AS metrics supported by Cloud Eye.

Table 6-4 AS metrics

Metric

Metric Description Value

cpu_util CPU

Usage

mem_utilMemory

Usage

CPU usage of an AS group

Formula: Total CPU usage of all ECSs in an AS group/Number of ECSs in the AS group

Unit: Percent

Memory usage of an AS group

Formula: Total memory usage of all ECSs in an AS group/Number of ECSs in the AS group

Unit: Percent

NOTE

This metric is unavailable if the image has no VM Tools installed.

Monitore

Range

d Object & Dimensio n

≥0% Object:

AS group

Dimensio n:

AutoScali ngGroup

≥0% Object:

AS group

Dimensio n:

AutoScali ngGroup

Monito ring Interval (Raw Data)

5 minutes

network_ outgoing _bytes_ra te_inban d

Inband Incoming Rate

Number of incoming bytes per second on an ECS in an AS group

Formula: Total inband incoming rates of all ECSs in an AS group/

≥0 Byte/s

Object: AS group

Dimensio n:

AutoScali ngGroup

5 minutes

Number of ECSs in the AS group

Unit: Byte/s

Auto Scaling User Guide 6 AS Group and Instance Monitoring

Metric ID

instance_ num

disk_rea d_bytes_ rate

Metric Description Value

Range

Inband Outgoing Rate

Number of outgoing bytes per second on an ECS in an AS group

≥0 Object:

Formula: Total inband outgoing rates of all ECSs in an AS group/ Number of ECSs in the AS group

Unit: Byte/s

Disks Read Rate

Number of bytes read from an AS group per second

≥0 Byte/s

Formula: Total disks read rates of all ECSs in an AS group/Number of ECSs in the AS group

Unit: Byte/s

Monitore d Object & Dimensio n

AS group

Dimensio n:

AutoScali ngGroup

Object: AS group

Dimensio n:

AutoScali ngGroup

Monito ring Interval (Raw Data)

5 minutes

disk_writ e_bytes_r ate

disk_rea d_reques ts_rate

Disks Write Rate

Disks Read Requests

Number of bytes written to an AS group per second

Formula: Total disks write rates of all ECSs in an AS group/Number of ECSs in the AS group

Unit: Byte/s

Number of read requests per second sent to an ECS disk in an AS group

Formula: Total disks read rates of all ECSs in an AS group/Number of ECSs in the AS group

Unit: Request/s

≥0 Byte/s

≥0 request /s

Object: AS group

Dimensio n:

AutoScali ngGroup

Object: AS group

Dimensio n:

AutoScali ngGroup

5 minutes

Auto Scaling User Guide 6 AS Group and Instance Monitoring

Metric

Metric Description Value

disk_writ e_reques ts_rate

Disks Write Requests

cpu_usage(Agent)

CPU Usage

Number of write requests per second sent to an ECS disk in an AS group

Formula: Total disks write rates of all ECSs in an AS group/Number of ECSs in the AS group

Unit: Request/s

Agent CPU usage of an AS group

Formula: Total Agent CPU usage of all ECSs in an AS group/Number of ECSs in the AS group

Unit: Percent

Monitore

Range

d Object & Dimensio n

≥0 request /s

Object: AS group

Dimensio n:

AutoScali ngGroup

0-100%Object:

AS group

Dimensio n:

AutoScali ngGroup

Monito ring Interval (Raw Data)

5 minutes

1 minute

mem_us edPercen t

load_ave rage1

load_ave rage5

(Agent) Memory Usage

(Agent) 1-Minute Load Average

(Agent) 5-Minute Load Average

Agent memory usage of an AS group

Formula: Total Agent memory usage of all ECSs in an AS group/ Number of ECSs in the AS group

Unit: Percent

Average CPU load of all ECSs in an AS group in the last 1 minute Unit: none

Average CPU load of all ECSs in an AS group in the last 5 minutes Unit: none

0-100%Object:

AS group

Dimensio n:

AutoScali ngGroup

≥0 Object:

AS group

Dimensio n:

AutoScali ngGroup

≥0 Object:

AS group

Dimensio n:

AutoScali ngGroup

1 minute

NO TE

Auto Scaling User Guide 6 AS Group and Instance Monitoring

Metric ID

load_ave rage15

gpu_usa ge_gpu

gpu_usa ge_mem

Metric Description Value

Range

(Agent) 15Minute Load

Average CPU load of all ECSs in an AS group in the last 15 minutes Unit: none

≥0 Object:

Average

(Agent) GPU Usage

Agent GPU usage of an AS group

Formula: Total Agent

0-100%Object:

GPU usage of all ECSs in an AS group/Number of ECSs in the AS group

Unit: Percent

(Agent) Video Memory Usage

Agent GPU memory usage of an AS group

Formula: Total Agent GPU memory usage of

0-100%Object:

all ECSs in an AS group/ Number of ECSs in the AS group

Unit: Percent

Monitore d Object & Dimensio n

AS group

Dimensio n:

AutoScali ngGroup

AS group

Dimensio n:

AutoScali ngGroup

AS group

Dimensio n:

AutoScali ngGroup

Monito ring Interval (Raw Data)

1 minute

Dimension

Monitoring metrics are classied into metrics with Agent and without Agent. For some OSs, you need to install the Agent to obtain the corresponding monitoring metrics. In this case, select the monitoring metrics with Agent, for example, (Agent) Memory Usage.

OSs determine whether the Memory Usage, Inband Outgoing Rate, and Inband Incoming Rate metrics are supported. For details, see

Before using Agent monitoring metrics, make sure that the Agent plug-in has been installed on the instances in the AS group. For details, see How Can I Install the Agent Plug-in on

the Instances in an AS Group to Use Agent Monitoring Metrics?

Key

Value

Elastic Cloud Server User Guide

AutoScalingGroup AS group ID

NO TE

Auto Scaling User Guide 6 AS Group and Instance Monitoring

6.6 Viewing Monitoring Metrics

Scenarios

The cloud platform provides Cloud Eye to help you obtain the running status of your ECSs. This section describes how to view details of AS group metrics to obtain information about the status of the ECSs in the AS group.

Prerequisites

The ECS is running properly.

● Monitoring metrics such as CPU Usage and Disks Read Rate are available only when there is at least one instance in an AS group. If not, only the Number of Instances metric is available.

● Monitoring data is not displayed for a stopped, faulty, or deleted ECS. After such an ECS restarts or recovers, the monitoring data is available.

Viewing Monitoring Metrics on Auto Scaling

1. Log in to the management console.

2. Under Computing, click Auto Scaling. In the navigation pane on the left, choose Instance Scaling.

3. On the AS Groups page, its name.

4. Click the Monitoring tab to view the monitoring data.

You can view data of the last 1, 3, and 12 hours. If you want to view data for a longer time range, click View details to go to the Cloud Eye page, hover

your mouse over a graph, and click

nd the AS group to view monitoring data and click

Viewing Monitoring Metrics on Cloud Eye

1. Log in to the management console.

2. Click in the upper left corner to select a region and a project.

3. Under Management & Deployment, click Cloud Eye.

4. In the navigation pane on the left, choose Cloud Service Monitoring > Auto Scaling.

5. Locate the row that contains the target AS group and click View Metric in the Operation column to view monitoring data.

You can view data of the last 1, 3, and 12 hours. Hover your mouse over a

graph and click

It can take a period of time to obtain and transfer the monitoring data. Therefore, wait for a while and then check the data.

to view data for a longer time range.

NO TE

Auto Scaling User Guide 6 AS Group and Instance Monitoring

6.7 Setting Monitoring Alarm Rules

Scenarios

Setting ECS alarm rules allows you to customize the monitored objects and notication policies and determine the running status of your ECSs at any time.

Procedure

1. Log in to the management console.

2. Click

3. Under Management & Deployment, click Cloud Eye.

4. In the navigation pane, choose Alarm Management > Alarm Rules.

5. On the Alarm Rules page, click Create Alarm Rule to create an alarm rule for the AS service or modify an existing alarm rule of the AS service.

6. After setting the parameters, click Finish.

in the upper left corner and select the desired region and project.

● For more information about how to set alarm rules, see

● You can create alarm rules on the Cloud Eye console to dynamically expand resources.

Cloud Eye User Guide

Huawei Auto Scaling User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual

Contents

1 AS Group

1.1 Creating an AS Group

1.2 (Optional) Adding a Load Balancer to an AS Group

1.4 Enabling an AS Group

1.5 Disabling an AS Group

1.6 Modifying an AS Group

1.7 Deleting an AS Group

3 AS Policy

3.1 Overview

3.2 Creating an AS Policy

3.3 Managing AS Policies

4 Scaling Action

4.1 Dynamically Expanding Resources

4.2 Expanding Resources as Planned

4.3 Manually Expanding Resources

4.5 Viewing a Scaling Action

4.6 Managing Lifecycle Hooks

4.8 Standby Instance

5 Bandwidth Scaling

5.1 Creating a Bandwidth Scaling Policy

5.2 Viewing Details About a Bandwidth Scaling Policy

5.3 Managing a Bandwidth Scaling Policy

6 AS Group and Instance Monitoring

6.1 Health Check

6.3 Recording AS Resource Operations

6.4 Adding Tags to AS Groups and Instances

6.5 Monitoring Metrics

6.6 Viewing Monitoring Metrics

6.7 Setting Monitoring Alarm Rules