Closed Sarvesh-Thiruppathi closed 2 months ago
This occurs because of the way the alarm detection works, if our alarm is already triggered it just doesn't go to sleep but because the sampleFlag is no longer true, and as such it just gets stuck in sleep. We could just push the interrupt time x amount from when we realized the alarm has been already triggered
This was fixed by simply reattaching to the RTC interrupt and sampling again
Describe the bug Running WeatherChimesTippingBucket code w/ DO probe (ChimeDO.txt). The interrupt duration was set to one minute. During one of the cycles, the MQTT broker refused to connect and it continued trying to upload until the limit was exceeded after four attempts. By this point, the one-minute alarm interrupt got triggered before Hypnos went to sleep resulting in the following message being spammed onto the serial monitor: "Alarm triggered during sample, specified sample duration was too short! Resampling..."
Log file from the Hypnos Board: output_4.log (Scroll to the bottom)
Hardware in Use
To Reproduce Steps to reproduce the behavior:
(If possible, test it out on a WeatherChime board)
Expected behavior Resample data and log to SD/upload to MongoDB
Code
Output
Additional context N/A