配置弹性伸缩规则
功能介绍
对弹性伸缩规则进行编辑。
在创建集群并执行作业接口中也可以创建弹性伸缩规则。
接口约束
无
调用方法
请参见如何调用API。
URI
POST /v1.1/{project_id}/autoscaling-policy/{cluster_id}
请求参数
参数 |
是否必选 |
参数类型 |
描述 |
---|---|---|---|
node_group |
是 |
String |
弹性伸缩规则适用的节点类型,当前只支持task节点。 |
auto_scaling_policy |
是 |
AutoScalingPolicy object |
弹性伸缩规则。 |
参数 |
是否必选 |
参数类型 |
描述 |
---|---|---|---|
auto_scaling_enable |
是 |
Boolean |
当前自动伸缩规则是否开启。 |
min_capacity |
是 |
Integer |
指定该节点组的最小保留节点数。 取值范围:[0~500] |
max_capacity |
是 |
Integer |
指定该节点组的最大节点数。 取值范围:[0~500] |
resources_plans |
否 |
Array of ResourcesPlan objects |
资源计划列表。若该参数为空表示不启用资源计划。 当启用弹性伸缩时,资源计划与自动伸缩规则需至少配置其中一种。 |
rules |
否 |
Array of Rule objects |
自动伸缩的规则列表。 当启用弹性伸缩时,资源计划与自动伸缩规则需至少配置其中一种。 |
exec_scripts |
否 |
Array of ScaleScript objects |
弹性伸缩自定义自动化脚本列表。若该参数为空表示不启用自动化脚本。 |
参数 |
是否必选 |
参数类型 |
描述 |
---|---|---|---|
period_type |
是 |
String |
资源计划的周期类型,当前只允许以下类型: daily |
start_time |
是 |
String |
资源计划的起始时间,格式为“hour:minute”,表示时间在0:00-23:59之间。 |
end_time |
是 |
String |
资源计划的结束时间,格式与“start_time”相同,不早于start_time表示的时间,且与start_time间隔不小于30min。 |
min_capacity |
是 |
Integer |
资源计划内该节点组的最小保留节点数。 取值范围:[0~500] |
max_capacity |
是 |
Integer |
资源计划内该节点组的最大保留节点数。 取值范围:[0~500] |
参数 |
是否必选 |
参数类型 |
描述 |
---|---|---|---|
name |
是 |
String |
弹性伸缩规则的名称。 只能由字母、数字、中划线和下划线组成,并且长度为1~64个字符。 在一个节点组范围内,不允许重名。 |
description |
否 |
String |
弹性伸缩规则的说明。 最大长度为1024字符。 |
adjustment_type |
是 |
String |
弹性伸缩规则的调整类型,只允许以下类型: 枚举值:
|
cool_down_minutes |
是 |
Integer |
触发弹性伸缩规则后,该集群处于冷却状态(不再执行弹性伸缩操作)的时长,单位为分钟。 取值范围[0~10080],10080为一周的分钟数。 |
scaling_adjustment |
是 |
Integer |
单次调整集群节点的个数。 取值范围[1~100] |
trigger |
是 |
Trigger object |
描述该规则触发条件。 |
参数 |
是否必选 |
参数类型 |
描述 |
---|---|---|---|
metric_name |
是 |
String |
指标名称。 该触发条件会依据该名称对应指标的值来进行判断。 最大长度为64个字符。 详细指标名称内容请参见"配置MRS集群弹性伸缩" |
metric_value |
是 |
String |
指标阈值。 触发该条件的指标阈值,只允许输入整数或者带两位小数的数。 |
comparison_operator |
否 |
String |
指标判断逻辑运算符,包括:
|
evaluation_periods |
是 |
Integer |
判断连续满足指标阈值的周期数(一个周期为5分钟)。 取值范围[1~288] |
参数 |
是否必选 |
参数类型 |
描述 |
---|---|---|---|
name |
是 |
String |
弹性伸缩自定义自动化脚本的名称,同一个集群的自定义自动化脚本名称不允许相同。 只能由数字、英文字符、空格、中划线和下划线组成,且不能以空格开头。 可输入的字符串长度为1~64个字符。 |
uri |
是 |
String |
自定义自动化脚本的路径。设置为OBS桶的路径或虚拟机本地的路径。
|
parameters |
否 |
String |
自定义自动化脚本参数。 多个参数间用空格隔开。 可以传入以下系统预定义参数:
其他用户自定义参数使用方式与普通shell脚本相同,多个参数中间用空格隔开。 |
nodes |
是 |
Array of strings |
自定义自动化脚本所执行的节点组名称(非自定义集群也可使用节点类型,包含Master、Core和Task三种类型)。 |
active_master |
否 |
Boolean |
自定义自动化脚本是否只运行在主Master节点上。 缺省值为false,表示自定义自动化脚本可运行在所有Master节点上。 |
fail_action |
是 |
String |
自定义自动化脚本执行失败后,是否继续执行后续脚本和创建集群。 说明:
枚举值:
|
action_stage |
是 |
String |
脚本执行时机。 枚举值:
|
响应参数
无
请求示例
配置集群弹性伸缩规则
POST https://{endpoint}/v1.1/{project_id}/autoscaling-policy/{cluster_id} { "node_group" : "task_node_analysis_group", "auto_scaling_policy" : { "auto_scaling_enable" : "true", "min_capacity" : "1", "max_capacity" : "3", "resources_plans" : [ { "period_type" : "daily", "start_time" : "9:50", "end_time" : "10:20", "min_capacity" : "2", "max_capacity" : "3" }, { "period_type" : "daily", "start_time" : "10:20", "end_time" : "12:30", "min_capacity" : "0", "max_capacity" : "2" } ], "exec_scripts" : [ { "name" : "before_scale_out", "uri" : "s3a://XXX/zeppelin_install.sh", "parameters" : "${mrs_scale_node_num} ${mrs_scale_type} xxx", "nodes" : [ "master_node_default_group", "core_node_analysis_group", "task_node_analysis_group" ], "active_master" : "true", "action_stage" : "before_scale_out", "fail_action" : "continue" }, { "name" : "after_scale_out", "uri" : "s3a://XXX/storm_rebalance.sh", "parameters" : "${mrs_scale_node_hostnames} ${mrs_scale_node_ips}", "nodes" : [ "master_node_default_group", "core_node_analysis_group", "task_node_analysis_group" ], "active_master" : "true", "action_stage" : "after_scale_out", "fail_action" : "continue" } ], "rules" : [ { "name" : "default-expand-1", "adjustment_type" : "scale_out", "cool_down_minutes" : "5", "scaling_adjustment" : "1", "trigger" : { "metric_name" : "YARNMemoryAvailablePercentage", "metric_value" : "25", "comparison_operator" : "LT", "evaluation_periods" : "10" } }, { "name" : "default-shrink-1", "adjustment_type" : "scale_in", "cool_down_minutes" : "5", "scaling_adjustment" : "1", "trigger" : { "metric_name" : "YARNMemoryAvailablePercentage", "metric_value" : "70", "comparison_operator" : "GT", "evaluation_periods" : "10" } } ] } }
响应示例
状态码: 200
操作成功。
{ "result" : "succeeded" }
SDK代码示例
SDK代码示例如下。
配置集群弹性伸缩规则
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 |
package com.huaweicloud.sdk.test; import com.huaweicloud.sdk.core.auth.ICredential; import com.huaweicloud.sdk.core.auth.BasicCredentials; import com.huaweicloud.sdk.core.exception.ConnectionException; import com.huaweicloud.sdk.core.exception.RequestTimeoutException; import com.huaweicloud.sdk.core.exception.ServiceResponseException; import com.huaweicloud.sdk.mrs.v1.region.MrsRegion; import com.huaweicloud.sdk.mrs.v1.*; import com.huaweicloud.sdk.mrs.v1.model.*; import java.util.List; import java.util.ArrayList; public class CreateScalingPolicySolution { public static void main(String[] args) { // The AK and SK used for authentication are hard-coded or stored in plaintext, which has great security risks. It is recommended that the AK and SK be stored in ciphertext in configuration files or environment variables and decrypted during use to ensure security. // In this example, AK and SK are stored in environment variables for authentication. Before running this example, set environment variables CLOUD_SDK_AK and CLOUD_SDK_SK in the local environment String ak = System.getenv("CLOUD_SDK_AK"); String sk = System.getenv("CLOUD_SDK_SK"); ICredential auth = new BasicCredentials() .withAk(ak) .withSk(sk); MrsClient client = MrsClient.newBuilder() .withCredential(auth) .withRegion(MrsRegion.valueOf("<YOUR REGION>")) .build(); CreateScalingPolicyRequest request = new CreateScalingPolicyRequest(); AutoScalingPolicyReqV11 body = new AutoScalingPolicyReqV11(); List<String> listExecScriptsNodes = new ArrayList<>(); listExecScriptsNodes.add("master_node_default_group"); listExecScriptsNodes.add("core_node_analysis_group"); listExecScriptsNodes.add("task_node_analysis_group"); List<String> listExecScriptsNodes1 = new ArrayList<>(); listExecScriptsNodes1.add("master_node_default_group"); listExecScriptsNodes1.add("core_node_analysis_group"); listExecScriptsNodes1.add("task_node_analysis_group"); List<ScaleScript> listAutoScalingPolicyExecScripts = new ArrayList<>(); listAutoScalingPolicyExecScripts.add( new ScaleScript() .withName("before_scale_out") .withUri("s3a://XXX/zeppelin_install.sh") .withParameters("${mrs_scale_node_num} ${mrs_scale_type} xxx") .withNodes(listExecScriptsNodes1) .withActiveMaster(true) .withFailAction(ScaleScript.FailActionEnum.fromValue("continue")) .withActionStage(ScaleScript.ActionStageEnum.fromValue("before_scale_out")) ); listAutoScalingPolicyExecScripts.add( new ScaleScript() .withName("after_scale_out") .withUri("s3a://XXX/storm_rebalance.sh") .withParameters("${mrs_scale_node_hostnames} ${mrs_scale_node_ips}") .withNodes(listExecScriptsNodes) .withActiveMaster(true) .withFailAction(ScaleScript.FailActionEnum.fromValue("continue")) .withActionStage(ScaleScript.ActionStageEnum.fromValue("after_scale_out")) ); Trigger triggerRules = new Trigger(); triggerRules.withMetricName("YARNMemoryAvailablePercentage") .withMetricValue("70") .withComparisonOperator("GT") .withEvaluationPeriods(10); Trigger triggerRules1 = new Trigger(); triggerRules1.withMetricName("YARNMemoryAvailablePercentage") .withMetricValue("25") .withComparisonOperator("LT") .withEvaluationPeriods(10); List<Rule> listAutoScalingPolicyRules = new ArrayList<>(); listAutoScalingPolicyRules.add( new Rule() .withName("default-expand-1") .withAdjustmentType(Rule.AdjustmentTypeEnum.fromValue("scale_out")) .withCoolDownMinutes(5) .withScalingAdjustment(1) .withTrigger(triggerRules1) ); listAutoScalingPolicyRules.add( new Rule() .withName("default-shrink-1") .withAdjustmentType(Rule.AdjustmentTypeEnum.fromValue("scale_in")) .withCoolDownMinutes(5) .withScalingAdjustment(1) .withTrigger(triggerRules) ); List<ResourcesPlan> listAutoScalingPolicyResourcesPlans = new ArrayList<>(); listAutoScalingPolicyResourcesPlans.add( new ResourcesPlan() .withPeriodType("daily") .withStartTime("9:50") .withEndTime("10:20") .withMinCapacity(2) .withMaxCapacity(3) ); listAutoScalingPolicyResourcesPlans.add( new ResourcesPlan() .withPeriodType("daily") .withStartTime("10:20") .withEndTime("12:30") .withMinCapacity(0) .withMaxCapacity(2) ); AutoScalingPolicy autoScalingPolicybody = new AutoScalingPolicy(); autoScalingPolicybody.withAutoScalingEnable(true) .withMinCapacity(1) .withMaxCapacity(3) .withResourcesPlans(listAutoScalingPolicyResourcesPlans) .withRules(listAutoScalingPolicyRules) .withExecScripts(listAutoScalingPolicyExecScripts); body.withAutoScalingPolicy(autoScalingPolicybody); body.withNodeGroup(AutoScalingPolicyReqV11.NodeGroupEnum.fromValue("task_node_analysis_group")); request.withBody(body); try { CreateScalingPolicyResponse response = client.createScalingPolicy(request); System.out.println(response.toString()); } catch (ConnectionException e) { e.printStackTrace(); } catch (RequestTimeoutException e) { e.printStackTrace(); } catch (ServiceResponseException e) { e.printStackTrace(); System.out.println(e.getHttpStatusCode()); System.out.println(e.getRequestId()); System.out.println(e.getErrorCode()); System.out.println(e.getErrorMsg()); } } } |
配置集群弹性伸缩规则
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 |
# coding: utf-8 from huaweicloudsdkcore.auth.credentials import BasicCredentials from huaweicloudsdkmrs.v1.region.mrs_region import MrsRegion from huaweicloudsdkcore.exceptions import exceptions from huaweicloudsdkmrs.v1 import * if __name__ == "__main__": # The AK and SK used for authentication are hard-coded or stored in plaintext, which has great security risks. It is recommended that the AK and SK be stored in ciphertext in configuration files or environment variables and decrypted during use to ensure security. # In this example, AK and SK are stored in environment variables for authentication. Before running this example, set environment variables CLOUD_SDK_AK and CLOUD_SDK_SK in the local environment ak = os.getenv("CLOUD_SDK_AK") sk = os.getenv("CLOUD_SDK_SK") credentials = BasicCredentials(ak, sk) \ client = MrsClient.new_builder() \ .with_credentials(credentials) \ .with_region(MrsRegion.value_of("<YOUR REGION>")) \ .build() try: request = CreateScalingPolicyRequest() listNodesExecScripts = [ "master_node_default_group", "core_node_analysis_group", "task_node_analysis_group" ] listNodesExecScripts1 = [ "master_node_default_group", "core_node_analysis_group", "task_node_analysis_group" ] listExecScriptsAutoScalingPolicy = [ ScaleScript( name="before_scale_out", uri="s3a://XXX/zeppelin_install.sh", parameters="${mrs_scale_node_num} ${mrs_scale_type} xxx", nodes=listNodesExecScripts1, active_master=True, fail_action="continue", action_stage="before_scale_out" ), ScaleScript( name="after_scale_out", uri="s3a://XXX/storm_rebalance.sh", parameters="${mrs_scale_node_hostnames} ${mrs_scale_node_ips}", nodes=listNodesExecScripts, active_master=True, fail_action="continue", action_stage="after_scale_out" ) ] triggerRules = Trigger( metric_name="YARNMemoryAvailablePercentage", metric_value="70", comparison_operator="GT", evaluation_periods=10 ) triggerRules1 = Trigger( metric_name="YARNMemoryAvailablePercentage", metric_value="25", comparison_operator="LT", evaluation_periods=10 ) listRulesAutoScalingPolicy = [ Rule( name="default-expand-1", adjustment_type="scale_out", cool_down_minutes=5, scaling_adjustment=1, trigger=triggerRules1 ), Rule( name="default-shrink-1", adjustment_type="scale_in", cool_down_minutes=5, scaling_adjustment=1, trigger=triggerRules ) ] listResourcesPlansAutoScalingPolicy = [ ResourcesPlan( period_type="daily", start_time="9:50", end_time="10:20", min_capacity=2, max_capacity=3 ), ResourcesPlan( period_type="daily", start_time="10:20", end_time="12:30", min_capacity=0, max_capacity=2 ) ] autoScalingPolicybody = AutoScalingPolicy( auto_scaling_enable=True, min_capacity=1, max_capacity=3, resources_plans=listResourcesPlansAutoScalingPolicy, rules=listRulesAutoScalingPolicy, exec_scripts=listExecScriptsAutoScalingPolicy ) request.body = AutoScalingPolicyReqV11( auto_scaling_policy=autoScalingPolicybody, node_group="task_node_analysis_group" ) response = client.create_scaling_policy(request) print(response) except exceptions.ClientRequestException as e: print(e.status_code) print(e.request_id) print(e.error_code) print(e.error_msg) |
配置集群弹性伸缩规则
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 |
package main import ( "fmt" "github.com/huaweicloud/huaweicloud-sdk-go-v3/core/auth/basic" mrs "github.com/huaweicloud/huaweicloud-sdk-go-v3/services/mrs/v1" "github.com/huaweicloud/huaweicloud-sdk-go-v3/services/mrs/v1/model" region "github.com/huaweicloud/huaweicloud-sdk-go-v3/services/mrs/v1/region" ) func main() { // The AK and SK used for authentication are hard-coded or stored in plaintext, which has great security risks. It is recommended that the AK and SK be stored in ciphertext in configuration files or environment variables and decrypted during use to ensure security. // In this example, AK and SK are stored in environment variables for authentication. Before running this example, set environment variables CLOUD_SDK_AK and CLOUD_SDK_SK in the local environment ak := os.Getenv("CLOUD_SDK_AK") sk := os.Getenv("CLOUD_SDK_SK") auth := basic.NewCredentialsBuilder(). WithAk(ak). WithSk(sk). Build() client := mrs.NewMrsClient( mrs.MrsClientBuilder(). WithRegion(region.ValueOf("<YOUR REGION>")). WithCredential(auth). Build()) request := &model.CreateScalingPolicyRequest{} var listNodesExecScripts = []string{ "master_node_default_group", "core_node_analysis_group", "task_node_analysis_group", } var listNodesExecScripts1 = []string{ "master_node_default_group", "core_node_analysis_group", "task_node_analysis_group", } parametersExecScripts:= "${mrs_scale_node_num} ${mrs_scale_type} xxx" activeMasterExecScripts:= true parametersExecScripts1:= "${mrs_scale_node_hostnames} ${mrs_scale_node_ips}" activeMasterExecScripts1:= true var listExecScriptsAutoScalingPolicy = []model.ScaleScript{ { Name: "before_scale_out", Uri: "s3a://XXX/zeppelin_install.sh", Parameters: ¶metersExecScripts, Nodes: listNodesExecScripts1, ActiveMaster: &activeMasterExecScripts, FailAction: model.GetScaleScriptFailActionEnum().CONTINUE, ActionStage: model.GetScaleScriptActionStageEnum().BEFORE_SCALE_OUT, }, { Name: "after_scale_out", Uri: "s3a://XXX/storm_rebalance.sh", Parameters: ¶metersExecScripts1, Nodes: listNodesExecScripts, ActiveMaster: &activeMasterExecScripts1, FailAction: model.GetScaleScriptFailActionEnum().CONTINUE, ActionStage: model.GetScaleScriptActionStageEnum().AFTER_SCALE_OUT, }, } comparisonOperatorTrigger:= "GT" triggerRules := &model.Trigger{ MetricName: "YARNMemoryAvailablePercentage", MetricValue: "70", ComparisonOperator: &comparisonOperatorTrigger, EvaluationPeriods: int32(10), } comparisonOperatorTrigger1:= "LT" triggerRules1 := &model.Trigger{ MetricName: "YARNMemoryAvailablePercentage", MetricValue: "25", ComparisonOperator: &comparisonOperatorTrigger1, EvaluationPeriods: int32(10), } var listRulesAutoScalingPolicy = []model.Rule{ { Name: "default-expand-1", AdjustmentType: model.GetRuleAdjustmentTypeEnum().SCALE_OUT, CoolDownMinutes: int32(5), ScalingAdjustment: int32(1), Trigger: triggerRules1, }, { Name: "default-shrink-1", AdjustmentType: model.GetRuleAdjustmentTypeEnum().SCALE_IN, CoolDownMinutes: int32(5), ScalingAdjustment: int32(1), Trigger: triggerRules, }, } var listResourcesPlansAutoScalingPolicy = []model.ResourcesPlan{ { PeriodType: "daily", StartTime: "9:50", EndTime: "10:20", MinCapacity: int32(2), MaxCapacity: int32(3), }, { PeriodType: "daily", StartTime: "10:20", EndTime: "12:30", MinCapacity: int32(0), MaxCapacity: int32(2), }, } autoScalingPolicybody := &model.AutoScalingPolicy{ AutoScalingEnable: true, MinCapacity: int32(1), MaxCapacity: int32(3), ResourcesPlans: &listResourcesPlansAutoScalingPolicy, Rules: &listRulesAutoScalingPolicy, ExecScripts: &listExecScriptsAutoScalingPolicy, } request.Body = &model.AutoScalingPolicyReqV11{ AutoScalingPolicy: autoScalingPolicybody, NodeGroup: model.GetAutoScalingPolicyReqV11NodeGroupEnum().TASK_NODE_ANALYSIS_GROUP, } response, err := client.CreateScalingPolicy(request) if err == nil { fmt.Printf("%+v\n", response) } else { fmt.Println(err) } } |
更多编程语言的SDK代码示例,请参见API Explorer的代码示例页签,可生成自动对应的SDK代码示例。
状态码
状态码 |
描述 |
---|---|
200 |
操作成功。 |
错误码
请参见错误码。