(1)node exporter 标准性能指标
1)监控项
cpu使用率: (100 - (avg by (instance)(irate(node_cpu_seconds_total{mode="idle"}[5m])) 100))
内存使用率:(100 - (((node_memory_MemFree_bytes+node_memory_Buffers_bytes+node_memory_Cached_bytes)/node_memory_MemTotal_bytes) 100))
磁盘使用率:(1- (node_filesystem_free_bytes{fstype=~"ext3|ext4|xfs"} / node_filesystem_size_bytes{fstype=~"ext3|ext4|xfs"}) ) * 100
2)prometheus规则
groups:
- name: alert-rule
rules:
- alert: NodeFilesystemUsage-high
expr: (1- (node_filesystem_free_bytes{fstype=~"ext3|ext4|xfs"} / node_filesystem_size_bytes{fstype=~"ext3|ext4|xfs"}) ) * 100 > 80
for: 2m
labels:
severity: warning
annotations:
summary: "{ {$labels.instance}}: High Node Filesystem usage detected"
description: "{ {$labels.instance}}: Node Filesystem usage is above 80% ,(current value is: { { $value }})"
- alert: NodeMemoryUsage
expr: (100 - (((node_memory_MemFree_bytes+node_memory_Buffers_bytes+node_memory_Cached_bytes)/node_memory_MemTotal_bytes) * 100)) > 80