Host Monitor

문서 > 모니터링 > Monitor Types > Host Monitor

이 페이지는 아직 한국어로 제공되지 않으며 번역 작업 중입니다. 번역에 관한 질문이나 의견이 있으시면 언제든지 저희에게 연락해 주십시오.

Overview

Infrastructure monitoring provides visibility into your entire IT environment, including cloud-hosted and on-prem servers, through many integrations. Use the Host monitor to stay informed on which hosts are or are not submitting data to ensure continuous visibility.

Every Datadog Agent reports a service check called datadog.agent.up with the status OK. You can monitor this check across one or more hosts by using a host monitor.

AIX Agents do not report the datadog.agent.up service check. You can use the metric datadog.agent.running to monitor the uptime of an AIX Agent. The metric emits a value of 1 if the Agent is reporting to Datadog.

Monitor creation

To create a host monitor in Datadog, use the main navigation: Monitors –> New Monitor –> Host.

Pick hosts by name or tag

Select the hosts to monitor by choosing host names, tags, or choose All Monitored Hosts. If you need to exclude certain hosts, use the second field to list names or tags.

The include field uses AND logic. All listed names and tags must be present on a host for it to be included.
The exclude field uses OR logic. Any host with a listed name or tag is excluded.

Examples

Monitor	Include	Exclude
Include all hosts with the tag `env:prod`	`env:prod`	leave empty
Include all hosts except hosts with the tag `env:test`	`All Monitored Hosts`	`env:test`

Set alert conditions

In this section, choose between a Check Alert or Cluster Alert:

A check alert tracks if a host stops reporting for a given amount of time. Too much time following a check run can be a sign of problems with data submission from the host.

Enter the number of minutes to check for missing data. The default value is 2 minutes.

If datadog.agent.up stops reporting an OK status for more than the minutes specified, an alert is triggered.

A cluster alert tracks if some percentage of hosts have stopped reporting for a given amount of time.

To set up a cluster alert:

Decide whether or not to group your hosts according to a tag. Ungrouped calculates the status percentage across all included hosts. Grouped calculates the status percentage on a per group basis.
Select the percentage for alert and warn thresholds. Only one setting (alert or warn) is required.
Enter the number of minutes to check for missing data. The default value is 2 minutes.

If datadog.agent.up stops reporting an OK status for more than the minutes specified and the percentage threshold is reached, an alert is triggered.