Get Listed Request: N3uron SparkPipe v0.0.5

daniel-paesa commented 3 months ago

Organization Name and Organization URL

N3uron Connectivity Systems: https://n3uron.com/

Sparkplug Conformance Profile(s) covered by this product

Sparkplug Host Application

TCK Version Used, digital SHA-256 Fingerprint, and Download URL

Sparkplug TCK v3.0.0: https://download.eclipse.org/sparkplug/3.0.0/Eclipse-Sparkplug-TCK-3.0.0.zip SHA-256: A70B2C2F00D67AC714EADD5AC50F6241E0EFA26036D95CA8EC667D491021B86B

Public URL of TCK Results Summary

https://dl.n3uron.com/SparkplugTCKResults.log

Summary of the Environment Used for Certification - Devices, Software Components, Operating Systems, Cloud Services, etc

SparkPipe v0.0.5 (AWS EC2) HiveMQ Community Edition 2024.6

Accept EFTL Terms

[X] I agree to the terms of EFTL

All TCK Requirements are met

[X] I attest that all TCK requirements have been met

bryce-nakatani commented 3 months ago

I post processed your results with the TCK's Python postprocessor (attached). The results from the SendCommand test are incomplete. Specifically it looks like NCMD and DCMD messages were not evaluated. summary.zip

daniel-paesa commented 3 months ago

I had issues using the Python script to generate the HTML report, so I uploaded the raw logs.

Regarding the SendCommand test, we are unable to properly evaluate this test as our software acts as a host application but does not support writing to arbitrary metrics, it acts as a bridge between the MQTT broker with Sparkplug traffic and publishes the metrics it receives to Kafka.

Sending commands is implemented and it requests rebirths if necessary (out-of-order timeout, etc..) but is not interactive and cannot be requested by the user on-demand.

bryce-nakatani commented 3 months ago

The logs are not an issue. I processed them to make the appearance of the results more consistent. In the 3.0.0 TCK, if you copied the TCK's "src" directory tree to the same directory as the python script, the postprocessor will work.

The implementation makes sense and thank you for the explanation. Would your application issue a "Node Control/Rebirth" NCMD to an unknown or an offline edge node?

daniel-paesa commented 3 months ago

It issues a "Node Control/Rebirth" command in the following cases:

Receives data from an unknown (never received a birth) node
Receives an unknown metric not specified in the previous birth
Receives a metric with an unknown alias
Receives more than one metric from the same node using the same alias (duplicate)
Receives a device birth without a previous node birth
Receives device metrics without a previous device birth
Receives out-of-order messages and the configured timeout (defaults to 5s) has passed.
Receives a metric after a node or device death