英文:
Pod didn't trigger scale-up because "max node group size reached"
问题
我在Google Cloud Platform中使用GKE，并遇到了以下错误消息。
Pod未触发扩容：12个节点与Pod的节点亲和性/选择器不匹配，达到了3个最大节点组大小。
我正在寻找此问题的原因，但找不到任何答案。
有人知道这个问题的原因吗？另外，如果您能告诉我如何解决这个错误，我将不胜感激。
英文:
I use GKE in Google Cloud Platform, and encountered the following error message.
> pod didn't trigger scale-up: 12 node(s) didn't match Pod's node
> affinity/selector, 3 max node group size reached.
I look for the cause of this issue, but cannot find any answer.
Does anyone know the cause of this issue? Also, I would appreciate if you tell me about the solution for this error.
答案1
得分: 0
Pod正在等待在节点上进行调度，但找不到合适的节点。集群自动缩放器无法在任何可用的节点池及其区域中执行任何操作 - 即使添加新节点也无法解决问题：
- 对于12种（节点池+区域）组合：节点未与Pod的节点亲和性/选择器匹配
- 对于3种（节点池+区域）组合：已达到最大节点组大小 - 已提供的节点计数>= autoscaling.maxNodeCount
此外，GKE具有启用node自动配置的选项。
英文:
It means that Pod is waiting to get scheduled on a node, but suitable node cannot be found. Cluster Autoscaler can't do anything in any of the available node pools and their zones - it still wouldn't fit if a new node is added:
- for 12 combinations of (node pool + zone): node(s) didn't match Pod's node affinity/selector
- for 3 combinations of (node pool + zone): max node group size reached - provisioned node count >= autoscaling.maxNodeCount
Also GKE has an option to enable node auto-provisioning
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论