3.2.2. 创建调度程序策略文件
您可以通过使用所需的 predicates 和 priorities 创建 JSON 文件来更改默认调度行为。然后,您可以通过 JSON 文件生成配置映射,并将 cluster
调度程序对象指定为使用该配置映射。
流程
配置调度程序策略:
使用所需的 predicates 和 priorities,创建一个名为
policy.cfg
的 JSON 文件。调度程序 JSON 文件示例
{ "kind" : "Policy", "apiVersion" : "v1", "predicates" : [ 1 {"name" : "MaxGCEPDVolumeCount"}, {"name" : "GeneralPredicates"}, {"name" : "MaxAzureDiskVolumeCount"}, {"name" : "MaxCSIVolumeCountPred"}, {"name" : "CheckVolumeBinding"}, {"name" : "MaxEBSVolumeCount"}, {"name" : "MatchInterPodAffinity"}, {"name" : "CheckNodeUnschedulable"}, {"name" : "NoDiskConflict"}, {"name" : "NoVolumeZoneConflict"}, {"name" : "PodToleratesNodeTaints"} ], "priorities" : [ 2 {"name" : "LeastRequestedPriority", "weight" : 1}, {"name" : "BalancedResourceAllocation", "weight" : 1}, {"name" : "ServiceSpreadingPriority", "weight" : 1}, {"name" : "NodePreferAvoidPodsPriority", "weight" : 1}, {"name" : "NodeAffinityPriority", "weight" : 1}, {"name" : "TaintTolerationPriority", "weight" : 1}, {"name" : "ImageLocalityPriority", "weight" : 1}, {"name" : "SelectorSpreadPriority", "weight" : 1}, {"name" : "InterPodAffinityPriority", "weight" : 1}, {"name" : "EqualPriority", "weight" : 1} ] }
根据调度程序 JSON 文件创建配置映射:
$ oc create configmap -n openshift-config --from-file=policy.cfg <configmap-name> 1
- 1
- 输入配置映射的名称。
例如:
$ oc create configmap -n openshift-config --from-file=policy.cfg scheduler-policy
输出示例
configmap/scheduler-policy created
编辑调度程序 Operator 自定义资源以添加配置映射:
$ oc patch Scheduler cluster --type='merge' -p '{"spec":{"policy":{"name":"<configmap-name>"}}}' --type=merge 1
- 1
- 指定配置映射的名称。
例如:
$ oc patch Scheduler cluster --type='merge' -p '{"spec":{"policy":{"name":"scheduler-policy"}}}' --type=merge
在修改了
Scheduler
配置资源后,等待openshift-kube-apiserver
pod 重新部署。这可能需要几分钟。只有重新部署 pod 后,新的调度程序才会生效。通过查看
openshift-kube-scheduler
命名空间中调度程序 pod 的日志,来验证已配置调度程序策略。以下命令检查由调度程序注册的 predicates 和 priorities:$ oc logs <scheduler-pod> | grep predicates
例如:
$ oc logs openshift-kube-scheduler-ip-10-0-141-29.ec2.internal | grep predicates
输出示例
Creating scheduler with fit predicates 'map[MaxGCEPDVolumeCount:{} MaxAzureDiskVolumeCount:{} CheckNodeUnschedulable:{} NoDiskConflict:{} NoVolumeZoneConflict:{} GeneralPredicates:{} MaxCSIVolumeCountPred:{} CheckVolumeBinding:{} MaxEBSVolumeCount:{} MatchInterPodAffinity:{} PodToleratesNodeTaints:{}]' and priority functions 'map[InterPodAffinityPriority:{} LeastRequestedPriority:{} ServiceSpreadingPriority:{} ImageLocalityPriority:{} SelectorSpreadPriority:{} EqualPriority:{} BalancedResourceAllocation:{} NodePreferAvoidPodsPriority:{} NodeAffinityPriority:{} TaintTolerationPriority:{}]'