kubernetes
resource_error
ai_generated
true
Pod 状态:已驱逐 — 节点资源不足:磁盘压力
Pod status: Evicted — The node was low on resource: disk-pressure
ID: kubernetes/pod-evicted-due-to-disk-pressure
88%修复率
90%置信度
1证据数
2023-09-05首次发现
版本兼容性
| 版本 | 状态 | 引入 | 弃用 | 备注 |
|---|---|---|---|---|
| kubernetes 1.26 | active | — | — | — |
| kubernetes 1.27 | active | — | — | — |
| kubernetes 1.28 | active | — | — | — |
根因分析
节点的磁盘使用率超过阈值(例如 85% 或 90%),触发 kubelet 驱逐 Pod 以释放空间。
English
The node's disk usage exceeded a threshold (e.g., 85% or 90%), triggering the kubelet to evict pods to free space.
官方文档
https://kubernetes.io/docs/concepts/scheduling-eviction/node-pressure-eviction/解决方案
-
SSH 到节点并运行 `df -h` 识别满的分区。通过删除未使用的容器镜像清理:`docker system prune -a` 或 `crictl rmi --prune`。同时检查并删除旧日志:`journalctl --vacuum-size=500M`。
-
使用节点亲和性或污点将工作负载移动到具有足够磁盘的节点:`kubectl taint nodes node1 disk-pressure=true:NoSchedule` 然后将 Pod 重新调度到另一个节点。
无效尝试
常见但无效的做法:
-
95% 失败
The pod will be evicted again immediately if the node's disk pressure persists.
-
60% 失败
Raising thresholds can lead to node instability and data loss; it only delays the problem.
-
90% 失败
Restarting kubelet doesn't free disk space; the underlying disk usage issue remains.