Reliability
Trust the data. Verify the whole chain.
Reliable operations require visibility into acquisition, stored history, notification delivery, and application health.
A communicating device is a good starting point. For an operator depending on a trend log, the important result is also whether the expected measurements reached storage. For an alarm recipient, the question continues through the delivery of the notification.
Reliable facility operations require visibility across that chain. Nimrod brings acquisition audits, communication events, delivery status, and application health into the operational picture.
Audit the recorded result.
A successful connection and a complete history answer different questions. Data acquisition audits check the final stored trend records against the expected collection result, making missing or incomplete data visible.
The example compares 1,440 expected samples with 1,440 stored samples in a 24-hour window. It is an illustration of an audit result, not a claim that every Point uses that interval. Expected history depends on the configured collection schedule and the audit window.
[ 07 / DATA INTEGRITY & APPLICATION HEALTH ]
Trust the data.
Verify the whole chain.
Monitor communications, recorded trends, notification delivery, and the application itself. When something degrades, bring the warning into the same operational picture as your facility events.
- NetworkCommunicating
- AcquisitionSamples collected
- Trend logRecords verified
Recorded history verified
- Expected samples
- 1,440
- Stored samples
- 1,440
- Missing
- 0
Independent watchdog
Application errors, process health, and available disk space.
Verify the recorded result.
Check that expected trend samples reach storage. Make missing or incomplete history visible.
Catch communication faults.
Detect network events and communication errors, and notify the people responsible.
Follow every recipient.
Track SMS and email delivery outcomes. Bring failures and degraded delivery into the operational interface.
Watch the application itself.
Separate monitoring processes watch cloud and on-premises deployments for application errors and resource pressure.
A system-health example with a completed acquisition audit, recipient delivery outcomes, network errors, and an independent watchdog. All displayed records are illustrative.
Keep communication failures visible.
Network events and communication errors help explain why acquisition may be affected. A timeout on a controller should be visible to the people responsible for that system, alongside the facility conditions they are already monitoring.
This gives operators two related perspectives: whether the communication path is functioning and whether it produced the history the operation expected. Either perspective alone can leave a gap in the picture.
Follow the notification to the recipient.
Sending a notification is another step that deserves its own status. Nimrod tracks delivery information for the expected recipients and brings failures or degraded delivery into the operational interface.
A warning such as “SMS notification delivery is degraded” lets administrators see that the messaging path needs attention. In the example, two deliveries succeed and one SMS fails. That distinction is useful even when the original equipment event has already received a response.
Delivery status reflects the information available from the messaging service. It should be read alongside the incident acknowledgement and operator activity when assessing the response.
Monitor the application itself.
The application is part of the operating chain too. Cloud and on-premises deployments have separate monitoring processes that watch for application errors and health conditions, including available disk space.
An independent watchdog can notify operators when the host is running low on storage or another application-health condition needs attention. Keeping those warnings visible connects the health of the monitoring platform to the people relying on it.
Taken together, these checks make reliability an operating concern from the source measurement through the stored record and the people who need to know.