Auto Scaling Group

Available in Classic

The following describes how to create an Auto Scaling Group based on launch configurations.

Auto Scaling Group screen

The basic description for the use of an Auto Scaling Group is as follows:

Check Auto Scaling Group list

You can check information on each Auto Scaling group. The following describes how to view the information.

1. Click the environment you are using in the Region menu and Platform menu of NAVER Cloud Platform console.
2. Click the Services> Compute > Auto Scaling menus in that order.
3. Click the Auto Scaling Group menu.
4. When the Auto Scaling Group list appears, check the summary information or click an Auto Scaling group to check the details.
   • Auto Scaling Group name: the name entered when creating the Auto Scaling group
   • Launch Configuration name: the name of the launch configuration specified for the Auto Scaling group
   • Minimum capacity: the minimum number of servers of the Auto Scaling group
   • Maximum capacity: the maximum number of servers of the Auto Scaling group
   • Expected capacity: the expected number of servers of the Auto Scaling group. If not entered, servers are created according to the minimum capacity.
   • Number of servers: the current number of servers
   • Health check type: health check standard of the Auto Scaling group
   • Server name prefix: the prefix attached to the Auto Scaling server
   • Zone(s): the zone where the Auto Scaling server is assigned
   • Default cooldown value (second): the cooldown time value to apply when performing scaling
   • Server configuration: detailed specifications of the Auto Scaling server
   • My Server image: the name of the My Server image used. A blank is displayed if the default image is used.
   • Load balancer: the name of the connected Load Balancer when the health check type is load balancer
   • Health check grace period (second): health check grace period value
   • Return policy: the server return policy set when creating Auto Scaling groups
   • NAT Gateway: whether to use NAT Gateway
   • Peer Host: the name of the peer host when the NAT Gateway is used

Create Auto Scaling Group

The following describes how to create an Auto Scaling group on NAVER Cloud Platform's console.

1. Click the environment you are using in the Region menu and Platform menu of NAVER Cloud Platform console.
2. Click the Services> Compute > Auto Scaling menus in that order.
3. Click the Auto Scaling Group menu.
4. Click the [Create Auto Scaling Group] button.
5. When the Auto Scaling Group creation page appears, take the following steps in order.
   • 1. Select launch configuration
   • 2. Set group
   • 3. Set up policy/schedule
   • 4. Set notification
   • 5. Final confirmation

1. Select launch configuration

Select the configuration template of the server to be included in the Auto Scaling group, i.e. select the launch configuration, and click [Next].

2. Set group

Specify the Auto Scaling Group settings information and click the [Next] button.

• Auto Scaling group name: must be 1 - 255 characters long. You may enter lowercase letters, numbers, and/or hyphens
• Zone: select the zone where the Auto Scaling group will be deployed
• Server name prefix: enter the prefix to be attached to the Auto Scaling server
• Minimum capacity: enter the minimum number of servers of the Auto Scaling group
• Maximum capacity: enter the maximum number of servers of the Auto Scaling group
• Expected capacity: enter the expected number of servers of the Auto Scaling group. If not entered, the minimum number of servers are created after the group is created.
• Default cooldown value (second): enter the cooldown time value to wait without reacting to other alarms until the server is fully ready when performing scaling
• Health check grace period: considering the fact that a health check cannot be properly responded to due to an update immediately after server creation, enter a grace period during which there is no problem even if the health check fails
  • If the health check standard is Load Balancer, you must enter the health check grace period.
• Health check type: health check standard of the Auto Scaling group. By default, a server is selected, and Load Balancer can be selected.
• Load Balancer: select whether to connect Load Balancer
  • If you click the Connect to Load Balancer checkbox to select it, the health check type is set to Load Balancer, and you can select the load balancer to connect (up to 10 can be connected). If Load Balancer is selected, the Auto Scaling server is automatically bound to the load balancer.
• Return policy: set the server return policy
  • Default setting: return the oldest server first among the servers
  • The customer directly specifies APIs: the user specifies return targets through APIs and returns them
    • API URL to be called: API URL to request the server list to return to the user when server return is required
    • Response wait time: user response wait time. It is returned to the default setting if it is exceeded
    • API authentication (optional): additional means of authentication for users' API
• For NAT Gateway: select whether to use NAT Gateway

HTTP query API, LIKE REST API, but not REST
HTTP only, not yet HTTPS
HTTP method: POST
POST body data type: json
{autoScalingGroupName:"asg-1", terminationCandidateInstances:[{instanceName:"s-1", instanceNo:"10342", instanceIpAddress:"10.7.31.4"}], adjustmentMagnitude:2}
Result data type: json
{autoScalingGroupName:"asg-1", selectedInstanceNoList:[{"10342", "20024", "89103"}]}
Size of list must be >= adjustmentMagnitude:2 and <= # of terminationCandidateInstances

3. Set up policy/schedule

Set the scaling policy or schedule, and click the [Next] button.

• Set later: set it later, not now
• Set policy: set the policy for increasing/decreasing the number of servers (up to 10 can be created)
  • Policy name: enter the policy name
  • Set scaling: change increase/decrease, change ratio, and set the scaling policy from among fixed values
    • Increase/decrease change: enter the number of servers to increase or decrease
    • Ratio change: enter a certain ratio at which the server will increase or decrease in comparison to the current group size
    • Fixed value: enter the fixed value of the group size
  • Minimum adjustment width: you can enter it only when you selected Ratio change, and it specifies the minimum adjustment width
  • Cooldown (second): enter the waiting time value without reacting to other alarms when scaling is performed
• Set schedule: set the schedule for increasing/decreasing the number of servers (up to 100 can be created)
  • Schedule name: enter the name of the schedule
  • Set scaling: enter the minimum capacity, maximum capacity, or expected capacity for performing scaling
    • You must enter at least one of the three items.
  • Time Zone: select a time zone
  • Start time: enter the date and time to start the schedule. If there is no input, the start date is designated as the earliest iteration schedule after saving the settings.
  • Set repetition: if repeated execution is required, enter the repetition cycle, but enter the same rules as Linux Cronjob
  • End time: when setting repetition, enter the date and time to end the repetition. If not, it is repeated infinitely.

4. Set notification

After setting the event notification function, click the [Next] button.

• Set later: set it later, not now
• Notification details: enable the function to notify when an event occurs
  • Notification details: select an event to be notified
    • Administrative Suspension refers to a situation in which the use of a launch configuration is suspended because the launch configuration is determined to be defective.
  • Set notification: select the person to be notified
    • You can select only from those registered in the notification target management group
    • To add a new notice recipient, click [Manage notification targets] to add the recipient. (See Notification target addition guide)

5. Final confirmation

Check the set details and then click the [Create Auto Scaling Group] button.

• A group is created and added to the Auto Scaling Group list.

Integrate monitoring event

If you set the Auto Scaling execution rule as a policy, you can make it so that the Auto Scaling policy is executed based on monitoring events.
Set the event policy of the Auto Scaling group in the Services > Management & Governance > Monitoring > Group event setting menu. For more information on the setting method, see Auto Scaling group event setting guide.

• An Auto Scaling group must be created, and there must be a scaling policy within the group.
• Items related to CPU usage are mainly used as event items.

Set and check Auto Scaling Group

You can change the policy, schedule, launch configuration, event notification, and detailed process of a created Auto Scaling group. Also, you can check the scaling execution history of the Auto Scaling group or the servers in the group. The following describes how to execute it.

1. Click the environment you are using in the Region menu and Platform menu of NAVER Cloud Platform console.
2. Click the Services> Compute > Auto Scaling menus in that order.
3. Click the Auto Scaling Group menu.
4. Select an Auto Scaling group and click the [Set and manage] button.
5. Click the setting tab to be changed and change the contents.

   • [Launch Configuration] tab: change the launch configuration connected to the group

   • [Policy] tab: change the scaling policy

     • If you click the [Create] button, a new scaling policy is added. (See Set policy)
     • If you click the [Edit] or [delete] button, you can edit or delete the selected policy, and if you click the [Execute] button, the selected policy is executed immediately

   • [Schedule] tab: set the scaling schedule

     • If you click the [Create] button, a new scaling schedule is added. (See Set schedule)
     • When the [Edit] or [delete] button is clicked, the selected schedule is edited or deleted

   • [History] tab: check the scaling execution history

     • Check the detailed information on the history items selected when the [Detailed information] button is clicked

   • [Set notification] tab: change notification events and notification targets

     • To add a new notice recipient, click [Manage notification targets] to add the recipient. (See Notification target addition guide)

   • [Server] tab: check the list and status of the servers in the group

   • [Manage process] tab: check the process status of the group and pause/restart it

     Process	Description
     LAUNCH (LANCH)	When a scale-out (server expansion) condition occurs, check the setting value to create and insert a server. When paused, server creation stops, and when restarted, scale-out is performed according to the expected quantity, if necessary
     TERMINATE (TERMT)	When a scale-in (server reduction) condition occurs, check the setting value and return the server. When paused, server return is stopped, and when restarted, scale-in is performed according to the expected quantity
     HEALTH CHECK (HTHCK)	Check server health every 10 seconds. When paused, health check ends, and when restarted, health check resumes from the start point
     REPLACE UNHEALTHY (RPUNH)	If the health check of a specific server fails, the failure history is recorded in the database, periodic monitoring is performed, and the server is replaced with another server. When paused, even if health check fails, the server is not returned, and scaling is performed according to the expected quantity. When restarted, the remaining servers with health check failures are returned again.
     SCHEDULED ACTIONS (SCACT)	Scaling is performed according to the registered schedule. When paused, even if it is schedule start time, scaling is not performed, and when restarted, scaling is performed according to the schedule. Scaling due to health check failure or change in the expected quantity is normally performed regardless of whether the process is stopped or not
     ADD TO LOAD BALANCER (ADTLB)	If the health check type is Load Balancer, the additional created servers are bound to the Load Balancer. When paused, binding is not performed, but when restarted, binding is performed only on servers added later (Binding is not requested for servers added during the pause)
     ALARM NOTIFICATION (ALMNO)	Scaling execution events according to Cloud Insight alarms are notified to the users. When paused, the Cloud Insight alarm is ignored, scaling is not performed, and transmission of notices is stopped. When restarted, Cloud Insight alarms that occur later are delivered and scaling is performed and notified.

Edit Auto Scaling Group information

The following describes how to edit the detailed information of a created Auto Scaling group.

1. Click the environment you are using in the Region menu and Platform menu of NAVER Cloud Platform console.
2. Click the Services> Compute > Auto Scaling menus in that order.
3. Click the Auto Scaling Group menu.
4. Select an Auto Scaling group and click the [Edit] button.
5. Edit the detailed information and click the [OK] button.
   • For more information on each item, see 2. Set group.

Delete Auto Scaling Group

The following shows how to delete a created Auto Scaling group.

1. Click the environment you are using in the Region menu and Platform menu of NAVER Cloud Platform console.
2. Click the Services> Compute > Auto Scaling menus in that order.
3. Click the Auto Scaling Group menu.
4. Click the Auto Scaling Group to delete and click the [Delete] button.
5. When the Delete pop-up window appears, click the [OK] button.
   • The Auto Scaling group disappears from the Auto Scaling Group list.