Troubleshoot EyesOnIt
Start with what you saw in the EyesOnIt web UI. Record the workspace, camera, time, and time zone. Those details let an administrator check the same event even when their browser displays times in a different zone.
I can’t sign in or see my camera
Section titled “I can’t sign in or see my camera”- At the sign-in screen, check the site address and username. Use Forgot your password? for password recovery or Use a recovery code if you have lost access to your authenticator. If both the authenticator and the saved codes are unavailable, ask the account administrator for MFA recovery. Account recovery
- After sign-in, read the workspace name. In Monitor, clear a folder or narrow camera selection. If the camera is still missing, have an administrator check the organization’s status, the organization’s camera grant, and your own role and camera selection. Verify that you can open the intended camera under your own account. Users and camera access
My live video is blank or has stopped
Section titled “My live video is blank or has stopped”- In Monitor, check whether the source is paused, starting, degraded, or offline. Select it and read its status. For a recorded file, Completed means the run reached its end.
- If a live camera should be running, check the camera’s stream address, credentials, and network reachability from the EyesOnIt server. Compare one failing camera with another on the same node to distinguish a camera problem from a server problem.
- Open the live view again. Confirm that the picture advances and the analyzed frame rate updates. For a cluster, verify the processing owner on System. Camera connection checks
The detection missed an event or fired at the wrong time
Section titled “The detection missed an event or fired at the wrong time”Open the incident or a representative frame. In Image Lab, check the region, object type, size, and confidence against the object you expected. If the frame qualifies but the incident timing differs, check motion settings, analyzed FPS, dwell, reset, and the rule’s schedule and time zone.
Replay a short clip that includes entry and exit. Change one setting at a time and compare the resulting incident time. A still image tests object selection. Moving footage tests dwell, crossing direction, and recovery. Tune motion and timing · Repeat a recorded-video test
The recording or search result is missing
Section titled “The recording or search result is missing”Open Archive at the camera and time of interest. Check the recorded intervals and Jump to next recording across a gap. If you expected an incident clip instead, open the incident and check its evidence status.
In Administration → Recording Storage → Find recordings, widen From and Through to contain the entire segment or clip. Select the intended camera and Recording type. Check recording, retention, and protection settings before deciding the footage is gone.
If the video plays but search misses it, check the search method, camera selection, indexing settings at capture time, and any affected interval under System → Archive integrity. PPE search needs retained assessments. Repeat the search against a known recorded event to confirm coverage. Recover searchable history
An incident occurred but no notification arrived
Section titled “An incident occurred but no notification arrived”- Open Incidents and confirm the camera, rule, and event time. If the incident never opened, start with the detection checks above.
- In the rule’s Actions, verify the selected profile and cooldown. Check that delivery is enabled, the destination is correct, and any customer approval covers the current profile settings and credential version.
- Coordinate a labeled Test delivery with the recipient. Use Check result, then confirm receipt at the destination. If that works, verify a controlled incident from the same rule. Record which of the two deliveries succeeded. Test notification delivery
The system is slow or a node needs attention
Section titled “The system is slow or a node needs attention”Open System → Refresh. Review Compute node health, processing resources, and camera count. Under Recording Storage, check the available space and disk status. Compare current workload and analyzed frame rates with a known healthy interval. For a customer workspace, check its resource allowance with the provider. System health · Cluster failover
For support, record the first affected time, last known good time, workspace and camera, node status, exact error, and steps that reproduce it. Include a screenshot cropped to the affected controls. An administrator can use Administration → Audit to locate related activity and include its support reference. Share the relevant details through your support channel and keep authentication secrets and unrelated customer records private. Audit history and export