ping 监控插件,探测远端目标地址能否 ping 通,如果机器没有禁 ping,这就是一个很好用的探测机器存活的手段
要探测的机器配置到 targets 中,targets 是个数组,可以配置多个,当然也可以拆成多个 [[instances]]
配置段,比如:
[[instances]]
targets = [ "10.4.5.6" ]
labels = { region="cloud", product="n9e" }
[[instances]]
targets = [ "10.4.5.7" ]
labels = { region="cloud", product="zbx" }
上例中是 ping 两个地址,为了信息更丰富,附加了 region 和 product 标签
systemctl edit categraf
Increase the number of open files:
[Service]
LimitNOFILE=8192
Restart Categraf:
systemctl restart categraf
On most systems, ping requires CAP_NET_RAW
capabilities or for Categraf to be run as root.
With systemd:
systemctl edit categraf
[Service]
CapabilityBoundingSet=CAP_NET_RAW
AmbientCapabilities=CAP_NET_RAW
systemctl restart categraf
Without systemd:
setcap cap_net_raw=eip /usr/bin/categraf
Reference man 7 capabilities
for more information about
setting capabilities.
When using method = "native"
, you will need permissions similar to the executable ping program for your OS.
该 README 的同级目录下,提供了 dashboard.json 就是监控大盘的配置,alerts.json 是告警规则,可以导入夜莺使用。