Device Uptime Detection
Identifies devices that haven't restarted within policy and prompts the user to act.
- Runs periodic checks for devices with extended uptime
- Identifies devices that haven't restarted within the configured threshold
- Distinguishes newly flagged devices from those already being tracked
- Notifies device owners through their preferred communication channel
- Consolidates the findings into a structured report
- Opens a ticket with the report attached for IT review
What Device Uptime Detection does
Device Uptime Detection is Harmony's automated monitoring workflow for identifying company-owned devices that haven't been restarted in an extended period. Running on a configurable schedule, Harmony queries for devices whose uptime exceeds a threshold, optionally notifies device owners directly via Slack or Teams to prompt a restart, and creates an IT desk ticket with a CSV report of the most overdue devices. Regular restarts apply security patches, clear memory, and keep systems healthy - this workflow ensures IT and device owners stay on top of devices that are drifting without reboots.
Who Device Uptime Detection Is For
IT operations and security teams responsible for endpoint health and patch compliance.
Devices that run for weeks or months without restarting are silently falling behind on security updates and accumulating memory issues. Without automated monitoring, there's no way to know which devices in the fleet are overdue for a restart until something breaks. This workflow surfaces those devices automatically and can nudge owners directly.
How Device Uptime Detection Works
A cron schedule fires (default: daily at 8 AM UTC). Schedule and timezone are configurable per tenant.
- Query high-uptime devices
The workflow queries device management for all company-owned devices whose days-since-last-restart exceeds a threshold. If owner notifications are enabled, the query uses the (lower) notification threshold to cast a wider net; otherwise it uses the ticket threshold.
- Owner notifications (optional)
If notifications are enabled, Harmony sends a direct message to each device owner whose device exceeds the notification threshold (default: 14 days), asking them to restart. A per-owner frequency gate (default: 7 days) prevents repeat nudges. VIP employee exclusion is configurable.
- Ticket filtering
Devices at or above the higher ticket threshold (default: 30 days) are separated from the broader query, with notification status merged in.
- CSV report
All ticket-threshold devices are serialized into a CSV report (
high_uptime_devices_report), including a "Last Notification" column showing when each owner was last nudged.
A ticket titled "Devices With High Uptime Detected" is created in the IT desk with the CSV attached.
Capabilities
- Two-threshold architecture - Operates on two configurable thresholds: a lower one for owner notifications (default: 14 days) and a higher one for IT ticket creation (default: 30 days), so owners get early warnings before devices appear in IT's queue.
- Direct owner notifications - Sends a Slack or Teams DM to the device owner asking them to restart, reducing IT's workload by prompting self-service action first.
- Notification frequency gate - Tracks when each owner was last notified and suppresses repeat nudges within the configured window (default: 7 days per device).
- VIP employee exclusion - Configurable option to skip notifications for employees marked as VIP.
- CSV report with notification history - The IT ticket CSV includes a "Last Notification" column so IT can see which device owners have already been nudged, avoiding redundant outreach.
- Targeted runs - Supports scoping a run to specific employee IDs for focused audits rather than fleet-wide scans.
- Configurable schedule - Cron schedule and timezone are fully configurable per tenant (default: daily).
- Ticket toggle - Ticket creation can be independently disabled if the tenant only wants owner notifications without IT tickets.
Main use cases
Proactive Security Patch Compliance - IT needs to ensure all devices in the fleet have restarted within the last 30 days to apply pending OS security patches. The daily workflow detects all devices exceeding the threshold, notifies device owners at 14 days to prompt early action, and creates an IT ticket at 30 days for devices that haven't restarted despite the nudge.
Prompting Employees to Restart Before Major Update Windows - IT is rolling out a critical security patch and needs all devices to restart within the week. The notification threshold can be lowered to capture devices that haven't restarted in 7 days, sending owners a targeted message before the IT ticket threshold is reached.
Identifying Specific Employees' Device Health - IT wants to check uptime for a specific team's devices ahead of a compliance audit. The target_employee_ids input allows scoping the run to specific employees, generating a focused report rather than the full fleet.
Integrations
| Integration | Role in the agent flow |
|---|---|
| Asset Management / Device Management API (internal) | Source of device uptime data (days-since-last-restart) per company-owned device |
| Service Desk (internal) | IT ticket created with CSV report attached; tracked for follow-up |
| Notifications system (Slack/Teams) | Sends direct restart reminder to device owners when notification threshold is exceeded |
FAQ
It's a scheduled Harmony workflow, not an interactive chatbot. It runs on a cron schedule (default: daily), queries device management for uptime data, and produces notifications and/or tickets based on what's found.
Meet more Agents
