Resource Monitoring

Use the resources_monitoring block on the coder_agent resource in our Terraform provider to monitor out of memory (OOM) and out of disk (OOD) errors and alert users when they overutilize memory and disk.

This can help prevent agent disconnects due to OOM/OOD issues.

You can specify one or more volumes to monitor for OOD alerts. OOM alerts are reported per-agent.

Prerequisites

Notifications are sent through SMTP. Configure Coder to use an SMTP server.

Example

Add the following example to the template's main.tf. Change the 90, 80, and 95 to a threshold that's more appropriate for your deployment:

resource "coder_agent" "main" {
  arch = data.coder_provisioner.dev.arch
  os   = data.coder_provisioner.dev.os
  resources_monitoring {
    memory {
      enabled   = true
      threshold = 90
    }
    volume {
      path      = "/volume1"
      enabled   = true
      threshold = 80
    }
    volume {
      path      = "/volume2"
      enabled   = true
      threshold = 95
    }
  }
}
See an opportunity to improve our docs? Make an edit.