Scalyr Agent collects CPU metrics from the hosts. For users who'd like to build a dashboard to monitor the overall CPU usage across all hosts from the account, they can simply apply the following JSON config.
{
graphs: [
{
"graphStyle": "pie",
"query": "metric ='proc.stat.cpu_rate' ($type == 'system' or type = 'user')| group cpu_time = avg(value) by serverHost | sort -cpu_time | limit 5",
"title": "Top 5 Hosts by Most CPU Time"
},
{
"graphStyle": "pie",
"query": "metric = 'sys.cpu.count' | group cpu_count = avg(value) by serverHost | sort -cpu_count | limit 5",
"title": "Top 5 Hosts by CPU Count"
},
{
"graphStyle": "report",
"query": "metric ='proc.stat.cpu_rate' ($type == 'system' or type = 'user') | group cpu_time = avg(value) by serverHost, serverIP | sort -cpu_time",
"title": "Hosts by Most CPU Time"
}
]
}
It shows a report of the real CPU time (the unit is 1/100 seconds) of the hosts in descending order, plus a pie graph that displays the hosts that have the most CPUs.
Comments
0 comments
Please sign in to leave a comment.