Nom du modèle
ACS-ESS-LifeCycleBulkyDeleteClusterNodes
Description du modèle
Supprime plusieurs nœuds d'un cluster Container Service for Kubernetes (ACK) simultanément à l'aide d'un hook de cycle de vie. Seuls les événements de réduction de capacité (scale-in) sont pris en charge.
Type de modèle
Automatisé
Propriétaire
Alibaba Cloud
Paramètres d'entrée
|
Paramètre |
Description |
Type |
Obligatoire |
Valeur par défaut |
Limite |
|
clusterId |
ID du cluster. |
String |
Oui |
||
|
drainNode |
Indique si tous les pods doivent être évacués des nœuds avant leur suppression. |
Boolean |
Oui |
||
|
releaseNode |
Indique si les instances Elastic Compute Service (ECS) doivent être libérées. |
Boolean |
Oui |
||
|
pageSize |
Taille de la page. |
String |
Non |
100 |
|
|
essTaskType |
Type de tâche Auto Scaling. |
String |
Non |
${essTaskType} |
|
|
regionId |
ID de la région. |
String |
Non |
${regionId} |
|
|
instanceIds |
IDs des instances ECS. |
List |
Non |
['${instanceIds}'] |
|
|
lifecycleHookId |
ID du hook de cycle de vie. |
String |
Non |
${lifecycleHookId} |
|
|
lifecycleActionToken |
Jeton de l'activité de mise à l'échelle associée aux instances. |
String |
Non |
${lifecycleActionToken} |
|
|
OOSAssumeRole |
Rôle Resource Access Management (RAM) assumé par CloudOps Orchestration Service (OOS). |
String |
Non |
OOSServiceRole |
Paramètres de sortie
Aucun
Stratégie requise pour exécuter le modèle
{
"Version": "1",
"Statement": [
{
"Action": [
"cs:DeleteClusterNodes",
"cs:DescribeClusterNodes",
"cs:DescribeTaskInfo"
],
"Resource": "*",
"Effect": "Allow"
},
{
"Action": [
"ess:CompleteLifecycleAction"
],
"Resource": "*",
"Effect": "Allow"
}
]
}
Références
ACS-ESS-LifeCycleBulkyDeleteClusterNodes
Contenu du modèle
FormatVersion: OOS-2019-06-01
Description:
en: Use lifecyclehook to delete cluster nodes (only supports elastic contraction activities)
zh-cn: 使用生命周期挂钩批量移除CS集群节点（仅支持弹性收缩活动）
name-en: ACS-ESS-LifeCycleBulkyDeleteClusterNodes
name-zh-cn: 使用生命周期挂钩批量移除CS集群节点
categories:
- elastic_manage
Parameters:
clusterId:
Label:
en: ClusterId
zh-cn: 集群ID
Type: String
drainNode:
Label:
en: DrainNode
zh-cn: 是否自动排空节点上的Pod
Type: Boolean
releaseNode:
Label:
en: ReleaseNode
zh-cn: 是否同时移除ECS
Type: Boolean
pageSize:
Label:
en: PageSize
zh-cn: 分页大小
Type: String
Default: '100'
essTaskType:
Label:
en: EssTaskType
zh-cn: 弹性伸缩任务类型
Type: String
Default: '${essTaskType}'
regionId:
Label:
en: RegionId
zh-cn: 地域ID
Description:
en: The ID of region
zh-cn: '地域ID，请使用弹性伸缩系统默认值 ${regionId}'
Type: String
Default: '${regionId}'
instanceIds:
Label:
en: InstanceIds
zh-cn: ECS实例ID列表
Description:
en: The ID list of the ECS instance
zh-cn: 'ECS实例ID列表,请使用弹性伸缩系统默认值 ["${instanceId}"]'
Type: List
Default:
- '${instanceIds}'
lifecycleHookId:
Label:
en: LifecycleHookId
zh-cn: 生命周期挂钩ID
Description:
en: The ID of the lifecycle hook
zh-cn: '生命周期挂钩ID,请使用弹性伸缩系统默认值 ${lifecycleHookId}'
Type: String
Default: '${lifecycleHookId}'
lifecycleActionToken:
Label:
en: LifecycleActionToken
zh-cn: 实例关联的特定伸缩活动的令牌
Description:
en: The token that indicates a specific scaling activity associated with an instance
zh-cn: '实例关联的特定伸缩活动的令牌,请使用弹性伸缩系统默认值 ${lifecycleActionToken}'
Type: String
Default: '${lifecycleActionToken}'
OOSAssumeRole:
Label:
en: OOSAssumeRole
zh-cn: OOS扮演的RAM角色
Type: String
Default: OOSServiceRole
RamRole: '{{ OOSAssumeRole }}'
Tasks:
- Name: describeClusterNodes
Action: 'ACS::ExecuteAPI'
Description:
en: Gets node names
zh-cn: 获取节点名称
OnError: CompleteLifecycleActionForAbandon
Properties:
Service: CS
API: DescribeClusterNodes
Method: GET
Headers:
Content-Type: application/json
URI: '/clusters/{{clusterId}}/nodes'
Parameters:
pageSize: '{{ pageSize }}'
instanceIds:
'Fn::Join':
- ','
- '{{ instanceIds }}'
Outputs:
nodeNames:
Type: List
ValueSelector: '.nodes[].node_name'
- Name: whetherNodeIsReady
Action: 'ACS::Choice'
Description:
en: Choose next task by nodes
zh-cn: 根据节点状态选择下一个任务
Properties:
DefaultTask: deleteClusterNodes
Choices:
- When:
'Fn::Equals':
- []
- '{{ describeClusterNodes.nodeNames }}'
NextTask: 'CompleteLifecycleActionForContinue'
- Name: deleteClusterNodes
Action: 'ACS::ExecuteAPI'
Description:
en: Delete the specified nodes of cluster
zh-cn: 移除集群中指定节点
OnError: CompleteLifecycleActionForAbandon
Properties:
Service: CS
API: DeleteClusterNodes
Method: POST
Headers:
Content-Type: application/json
URI: '/clusters/{{clusterId}}/nodes'
Body:
'Fn::If':
- 'Fn::Equals':
- SPOT_REMEDY
- '{{ essTaskType }}'
- '{"drain_node": {{ drainNode }}, "release_node": true, "nodes": {{ describeClusterNodes.nodeNames }}, "call_back": true, "is_expired": true}'
- '{"drain_node": {{ drainNode }}, "release_node": {{ releaseNode }}, "nodes": {{ describeClusterNodes.nodeNames }}, "call_back": true, "is_expired": false}'
Outputs:
taskId:
Type: String
ValueSelector: '.task_id'
- Name: untilNodesRemovedSuccessfully
Action: ACS::WaitFor
Description:
en: Wait for the node to be removed successfully
zh-cn: 等待节点移除成功
Retries: 26
Delay: 30
OnError: CompleteLifecycleActionForAbandon
OnSuccess: CompleteLifecycleActionForContinue
Properties:
Service: CS
API: DescribeTaskInfo
Method: GET
Headers:
Content-Type: application/json
URI: '/tasks/{{deleteClusterNodes.taskId}}'
Parameters: {}
DesiredValues:
- success
StopRetryValues:
- fail
PropertySelector: .state
- Name: CompleteLifecycleActionForContinue
Action: 'ACS::ExecuteAPI'
Description:
en: Modify lifecycle action for continue
zh-cn: 修改伸缩活动的等待状态为继续完成
OnSuccess: 'ACS::END'
Properties:
Service: ESS
API: CompleteLifecycleAction
Parameters:
RegionId: '{{ regionId }}'
LifecycleHookId: '{{ lifecycleHookId }}'
LifecycleActionToken: '{{ lifecycleActionToken }}'
LifecycleActionResult: CONTINUE
- Name: CompleteLifecycleActionForAbandon
Action: 'ACS::ExecuteAPI'
Description:
en: Complete lifecycle action for Abandon
zh-cn: 修改伸缩活动的等待状态为弃用
Properties:
Service: ESS
API: CompleteLifecycleAction
Parameters:
RegionId: '{{ regionId }}'
LifecycleHookId: '{{ lifecycleHookId }}'
LifecycleActionToken: '{{ lifecycleActionToken }}'
LifecycleActionResult: ABANDON