Where to find it: in the Hub (your MSP workspace at app.theonestack.com), open RMM, then use the RMM sidebar or page links below:
- RMM → Global → Alert Triage: app.theonestack.com/rmm/alerts
- RMM → Settings → Alerts: app.theonestack.com/rmm/settings/alerts
Overview
Offline alerts notify technicians when a managed device stops checking in. They are one of the most important alerts in RMM because a device going offline can indicate a hardware failure, power outage, ransomware, or a connectivity problem that the client has not yet noticed.
How Offline Detection Works
The RMM agent sends a heartbeat on a short interval. The system monitors these heartbeats continuously. If a device sends no heartbeat for five minutes, it is automatically flipped to Offline, an Agent Offline alert is created, and any automations you have configured for the offline event run.
This five-minute window is built-in and applies to all devices. It accounts for brief connectivity hiccups without generating noise.
Setting Up Offline Notifications
By default, an Agent Offline alert is created in the system when a device goes offline. To make sure the right people are notified:
- Go to Alerts > Alert Rules.
- Find the built-in Agent Offline rule (or create one if it is not present).
- Open the rule and configure notification routing:
- Choose the channels to notify: email, SMS, or PSA ticket.
- Set the severity (typically High or Critical for servers, Medium for workstations).
- Save.
Notifications fire as soon as the alert is created — there is no additional delay beyond the five-minute detection window.
Suppressing Expected Offline Periods
For devices that go offline nightly (sleeping laptops, after-hours shutdowns), use a maintenance window covering that period. Alerts during the window are suppressed and recorded in the rollup but not sent out. See Maintenance Windows.
Escalation
To escalate an offline alert if it is not acknowledged within a certain time, configure the escalation settings in the alert rule or in your PSA integration. For example:
- The alert fires and creates a low-priority ticket in your PSA.
- If the ticket is not updated within 30 minutes, escalate to a senior technician.
Escalation timing and routing are configured in your PSA and notification settings rather than in the RMM alert rule itself.
When a Device Comes Back Online
When the device heartbeats again, it returns to Online status automatically. The offline alert is not automatically resolved — you must resolve it manually. This ensures the event is reviewed even if the device recovers on its own.
Frequently Asked Questions
A device shows offline but the client says it is running fine.
The agent service may have stopped or the device may be blocking outbound port 443. Ask the client to check that the device can reach the internet. If it can, restart the RMM agent service on the device.
Can I get notified only if a server goes offline, not workstations?
Yes. Create separate alert rules for servers and workstations. Scope each rule to the appropriate device group and set different notification routing and severity levels for each.
We have devices that are shut down every night on purpose. Can I stop the offline noise?
Create a recurring maintenance window that covers the overnight hours for those devices. Offline alerts during the window will be suppressed.
How long does offline data stay visible?
Offline devices remain in your device list indefinitely until you delete them. Historical metrics are retained for 14 days at full resolution, 90 days at 15-minute rollups, and 1 year at hourly rollups.