Cette page n'est pas encore disponible en français, sa traduction est en cours.
Si vous avez des questions ou des retours sur notre projet de traduction actuel,
n'hésitez pas à nous contacter.
Overview
The Monitor Quality feature in Datadog identifies common misconfigurations in monitoring setups, such as alerts that have been muted for more than 60 days or monitors lacking designated recipients. It enables teams to maintain alerting standards and prevent critical alerts from being missed. Use Monitor Quality to identify and resolve misconfigured monitors all on the same page.
From the Manage Monitors page, click the Monitor Quality tab to access a list of monitors that need improvement. Datadog automatically enables this feature after you create your first monitor.
Filter and view the Monitor Quality page based on the teams, creators, services, or environments to organize and manage the list.
Improve monitors and follow best practices
Monitor Quality displays your monitors with the following quality issues:
Muted for over 60 days
Downtimes are useful for silencing alerts during scheduled maintenance, planned outages or system shutdowns, or to stop alerts during weekends and evenings. However, monitors that have been muted, or partially muted, for an extended period (more than 60 days) may indicate an oversight. You can unmute these monitors to resume alerting and ensure comprehensive monitoring coverage.
See which monitors are not alerting due to downtime misconfigurations and unmute them.
Monitors are missing recipients
When a monitor sets off an alert or notification, you want the alerts to go to the team or person that can resolve it and take action. If no recipients are added to the monitor notification, you decrease how proactive you can be with your services. Use the Monitor Quality page to review the monitors that are configured without any recipients.
Missing a delay
Data from cloud integrations (such as AWS, Azure, or Google Cloud) is pulled from the respective APIs with a crawler. These metrics arrive with a delay, which you can account for in your monitor configuration. Monitors with cloud data that do not have an evaluation delay can alert on false positives.
See all the monitors that are crawling cloud data but are missing the recommended delay. For more information on crawled data, see the Cloud Metric Delay page.
@notifications allow you to customize your monitors so that alerts are forwarded to integrations, workflows, or Datadog cases. If @notifications are misconfigured, the expected alerts are not sent to the appropriate channels.
See which monitors have misconfigured notification channels and edit them as necessary.
Composite monitors are missing constituents
Composite monitors evaluate the combined state of multiple sub monitors (constituents) following user-defined logic. Composite monitors that reference deleted constituents do not evaluate or notify. Identify inactive composite monitors and delete them.
Stuck in alert state
Monitors in an ALERT state are indications of an issue in your service that needs your attention. Multiple monitors in a constant ALERT state take the focus away from the issues that might actually need your attention. Investigate why these monitors are stuck in an alert state and edit your configuration accordingly.
Further reading
Documentation, liens et articles supplémentaires utiles:
