Message Bridge stops working after a while

[coderbyheart]

I had to restart the task to launch a new message bridge, because not message were forwarded from nRF Cloud.

We need better health monitoring of the bridge that checks, that message are coming through. This should be solved using a lambda that publishes a device message to nRF Cloud every minute and waits for the message to arrive the websocket within 10 seconds, if it arrives, health is good, if not health is bad. Record outcome in Cloudwatch Metrics, so we can set up alerts.

• device ID: health-check
• fingerprint: 29a.ch3ckr

Additional metrics to add:

• time between message creation and when it arrives at Message bridge #49

[pudkrong]

We have added metrics

• checkMessageFromWebsocket is number of health check run
• success is number of success
• fail is number is fail
• receivingMessageDuration is duration in seconds from sending message from simulator device until it arrives through the websocket

[pudkrong]

@coderbyheart For the metric time between message creation and when it arrives at Message bridge, just want to confirm you want to track a time between message creation to Message bridge.

Is it ok to capture this metric via IoT rule?

[coderbyheart]

@coderbyheart For the metric time between message creation and when it arrives at Message bridge, just want to confirm you want to track a time between message creation to Message bridge.

time between message creation and when it arrives at Message bridge = time of message retrieved by message bridge - time of creation

where

• time of message retrieved by message bridge = when the message bridge receives the message on the Account Device connection
• time of creation = the timestamp in the lambda that starts the health check

Is it ok to capture this metric via IoT rule?

As long as that does not change the time, yes.