基于Prometheus&Grafana的监控方案[3]-监控样例

流量图

这里写图片描述

node exporter

入口流量

sum by (ip) (rate(node_network_receive_bytes_total{name = 'node_bj_xxx', device != "lo"}[5m]))

出口流量

sum by (ip) (rate(node_network_transmit_bytes_total{name = 'node_bj_xxx', device != "lo"}[5m]))
wmi exporter

入口流量

sum by (ip) (rate(wmi_net_bytes_received_total{name="node_hf_xxx"}[30m]))

出口流量

sum by (ip) (rate(wmi_net_bytes_sent_total{name="node_hf_xxx"}[30m]))

内存&磁盘&CPU等

这里写图片描述
这里写图片描述

node exporter

启动时间

node_boot_time_seconds{ip="xxx.xx.xxx.xx"}*1000

内存大小

node_memory_MemTotal_bytes{ip="xxx.xx.xxx.xx"}

CPU核数

count (sum by (cpu) (node_cpu_seconds_total{ip="xxx.xx.xxx.xx"}))

平均负载

node_load1{ip="xxx.xx.xxx.xx"}
node_load5{ip="xxx.xx.xxx.xx"}
node_load15{ip="xxx.xx.xxx.xx"}

free内存

node_memory_MemFree_bytes{ip="xxx.xx.xxx.xx"}

可用内存

node_memory_MemFree_bytes{ip="xxx.xx.xxx.xx"}+node_memory_Cached_bytes{ip="xxx.xx.xxx.xx"}+node_memory_Buffers_bytes{ip="xxx.xx.xxx.xx"}

内存使用率

100 - ((node_memory_MemFree_bytes{ip="xxx.xx.xxx.xx"}+node_memory_Cached_bytes{ip="xxx.xx.xxx.xx"}+node_memory_Buffers_bytes{ip="xxx.xx.xxx.xx"})/node_memory_MemTotal_bytes{ip="xxx.xx.xxx.xx"}) * 100

磁盘使用率

100 - node_filesystem_free_bytes{ip="xxx.xx.xxx.xx",fstype!~"rootfs|selinuxfs|autofs|rpc_pipefs|tmpfs|udev|none|devpts|sysfs|debugfs|fuse.*"} / node_filesystem_size_bytes{ip="xxx.xx.xxx.xx",fstype!~"rootfs|selinuxfs|autofs|rpc_pipefs|tmpfs|udev|none|devpts|sysfs|debugfs|fuse.*"} * 100

CPU使用率

avg  by (ip)(irate(node_cpu_seconds_total{ip="xxx.xx.xxx.xx", mode != "idle"}[2m])) * 100

磁盘写IO

sum by (ip) (rate(node_disk_written_bytes_total{ip="xxx.xx.xxx.xx"}[30m]))

磁盘读IO

sum by (ip) (rate(node_disk_read_bytes_total{ip="xxx.xx.xxx.xx"}[30m]))
wmi exporter

启动时间

wmi_system_system_up_time {name='node_hf_xxx'} * 1000

C盘大小

wmi_logical_disk_size_bytes{name='node_hf_xxx',volume="C:"}

C盘使用率

(wmi_logical_disk_size_bytes{name='node_hf_xxx',volume='C:'}-wmi_logical_disk_free_bytes{name='node_hf_xxx',volume='C:'})/wmi_logical_disk_size_bytes{name='node_hf_xxx',volume='C:'} 

内存大小

wmi_cs_physical_memory_bytes{name='node_hf_xxx'} 

可用内存

wmi_os_physical_memory_free_bytes{name="node_hf_xxx"} 

内存使用率

1 - (wmi_os_physical_memory_free_bytes{name="node_hf_xxx"} /wmi_os_visible_memory_bytes{name="node_hf_xxx"})

CPU核数

wmi_cs_logical_processors{name='node_hf_xxx'}

CPU使用率

100-(avg by (ip) (irate(wmi_cpu_time_total{ip="xxx.xx.xxx.xx",mode="idle"}[30s]))*100)

磁盘写IO

sum by (ip)(rate(wmi_logical_disk_write_bytes_total{ip="xxx.xx.xxx.xx"}[30m]) )

磁盘读IO

sum by (ip) (rate(wmi_logical_disk_read_bytes_total{ip="xxx.xx.xxx.xx"}[30m]) )

猜你喜欢

转载自blog.csdn.net/zeroctu/article/details/81184109