We have some Greengrass Core Devices with deployments running on them.
These deployments work successfully - we have the LogManager installed, and we can see their logs showing normal function.
However, the IoT Console on AWS doesn't show the status updating - we have some devices with statuses like:
Status reported
1 month ago
And on the 'components' page, we see:
UpdateSystemPolicyService Dependency 0.0.0
Running
2 months ago 1 month ago <deployment-uuid-from-1-month-ago>
And the deployment-uuid will be an out of date deployment that was running 1 month ago, rather than the actual latest deployment.
The actual current deployment will show like:
<current-deployment-uuid> deployment-name thing-name
Succeeded
15 hours ago
It seems like the update information for this latest 'Succeeded' deployment hasn't updated the overall device information. Is there a fix for this?
Greengrass Nucleus Version: 2.12.1 (same issue persists with 2.12.2)
From subscribing to the device activity via MQTT, it looks all normal.
No 'failed to publish' type messages, or DNS lookup fails in the 'System' log on Cloudwatch (using AWS Cloudwatch to manage log uploads). Shows successful log uploads like:
[INFO] (pool-2-thread-12) com.aws.greengrass.status.FleetStatusService: fss-status-update-published. Status update published to FSS. {trigger=THING_DEPLOYMENT, serviceName=FleetStatusService, currentState=RUNNING}
To help us in debugging this on our side, where does the Core status information come from? A reserved AWS Topic, or an Eventbridge bus? Is it possible to subscribe directly to that information?
Hello,
Did you re-install the Nucleus with the same configuration and core device name used before? If so, please ensure follow those steps before re-installation: https://docs.aws.amazon.com/greengrass/v2/developerguide/uninstall-greengrass-core-v2.html. If not, please open a support case where we can get more details regarding the use case and help troubleshoot.