You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Feature Description
I noticed in the documentation for the Greengrass Nucleus Telemetry Agent that the QOS is set to 0. I have noticed my devices rarely send the aggregated daily telemetry data. To improve this, allowing the QOS to be set to 1 through the component configuration could help improve reliability.
The telemetry agent publishes data using the MQTT protocol with a quality of service (QoS) level of 0, which means that it doesn't confirm delivery or retry publishing attempts. Telemetry messages share an MQTT connection with other messages for subscriptions destined for AWS IoT Core.
Use Case
Why do you need this feature?
For my use case, daily aggregated data works fine, but using QOS = 0 makes it unreliable. I am currently using EventBridge to write the data to CloudWatch logs. I have 20 devices, but typically see updated logs every 5-6 days for a single device. I would expect to see all 20 devices publishing telemetry data every day.
Proposed Solution
Add a configuration field "QOS" to the nucleus component:
Feature Description
I noticed in the documentation for the Greengrass Nucleus Telemetry Agent that the QOS is set to 0. I have noticed my devices rarely send the aggregated daily telemetry data. To improve this, allowing the QOS to be set to 1 through the component configuration could help improve reliability.
I am referring to this link: https://docs.aws.amazon.com/greengrass/v2/developerguide/telemetry.html
The telemetry agent publishes data using the MQTT protocol with a quality of service (QoS) level of 0, which means that it doesn't confirm delivery or retry publishing attempts. Telemetry messages share an MQTT connection with other messages for subscriptions destined for AWS IoT Core.
Use Case
Why do you need this feature?
For my use case, daily aggregated data works fine, but using QOS = 0 makes it unreliable. I am currently using EventBridge to write the data to CloudWatch logs. I have 20 devices, but typically see updated logs every 5-6 days for a single device. I would expect to see all 20 devices publishing telemetry data every day.
Proposed Solution
Add a configuration field "QOS" to the nucleus component:
{
"telemetry": {
"enabled": true,
"QOS": 1,
"periodicAggregateMetricsIntervalSeconds": 3600,
"periodicPublishMetricsIntervalSeconds": 86400
}
}
Other
Add detailed explanation, stacktraces, related issues, links for us to have context, etc
https://docs.aws.amazon.com/greengrass/v2/developerguide/telemetry.html
https://docs.aws.amazon.com/greengrass/v2/developerguide/greengrass-nucleus-component.html#greengrass-nucleus-component-configuration-telemetry
The text was updated successfully, but these errors were encountered: