I have a fleet of ~40 Electrons that are deployed in remote areas for monitoring; many of them will work for several days before ceasing to work and upon failure analysis, it appears they are all in a state of fast blinking cyan. I’ve run the cloud-debug firmware and they all show the same error:
../crypto/mbedtls/library/ssl_tls.c:7075: bad application data message
I can fix them individually by resetting the keys, but that seems like an unsustainable solution. Is there a strategy for either 1) ensuring this doesn’t happen, or 2) fixing the keys remotely?