Linux systems that are configured to autoinstall packages may not be configured to autoreboot (it may be desirable to time this manually). This check enables alerts to be fired in the case where reboots are not performed in a timely manner.
If you are using Agent v6.8+ follow the instructions below to install the Reboot Required check on your host. See our dedicated Agent guide for installing community integrations to install checks with the Agent prior to version 6.8 or the Docker Agent:
Install the developer toolkit.
Clone the integrations-extras repository:
git clone https://github.com/DataDog/integrations-extras.git.
ddevconfig with the
ddev config set extras ./integrations-extras
To build the
ddev -e release build reboot_required
Run the following command to install the integrations wheel with the Agent:
datadog-agent integration install -w <PATH_OF_REBOOT_REQUIRED_ARTIFACT_>/<REBOOT_REQUIRED_ARTIFACT_NAME>.whl
Configure your integration like any other packaged integration.
reboot_required.d/conf.yamlfile in the
conf.d/folder at the root of your Agent's configuration directory. See the sample reboot_required.d/conf.yaml for all available configuration options.
Make sure you create a dd-agent (user that runs the Datadog agent) writable directory for the agent, and used by this check. The default of
/var/run/dd-agentis ideal. The snippet below should suffice.
sudo mkdir /var/run/dd-agent sudo chown dd-agent:dd-agent /var/run/dd-agent
Run the Agent's
status subcommand and look for
reboot_required under the Checks section.
No metrics are collected.
The reboot_required check does not include any events.
To create alert conditions on these service checks in Datadog, select 'Custom Check' on the Create Monitor page, not 'Integration'.
The check returns:
OKif the system does not require a reboot or for less than
WARNINGif the system has required a reboot for longer than
CRITICALif the system has required a reboot for longer than
Need help? Contact Datadog support.