I’m using a 865 module.
This mdm have a little library that logs any error or exception that can occurs.
I’m using too the watchdog library.
Both libraries works great logging the error and reboot the module when an error occurs or when the GPRS connection fails several times. This reboot command is sent to the module using the MDM2 to do not get any issue about socket status and others issues.
But even all the checks and cares the module hangs and do not reboot properly it stay in an unknown status. I must perform a turn off/on to get control again.
It’s very clear that the internal watchdog is based on software and not in hardware, so I think if this is a power source problem or other issue the module never detect them and reboot.
What I want to do is easy, the module must be alive always do not care about exceptions, errors, power source noises or failures or low voltages. the module must take care about all these issues and stay all the time online.
So, what steps i must perform to reach a all time alive system ?
The Python watchdog is software, the way to go to protect from hardware problems is to add a very small microcontroller to supervise for example a periodic module GPIO toogle. Is cheap and strong.
Hi Cosmin, thank you for response.
My app is using a external device with a uP so i will implement the WDT on it.
What i’m suspecting is that is not a problem with errors, it’s a problem with the reboot command.
I test this software in a 862 device, this is a RS S3 board kit and in the X700 I/O connector I build a SPI device to control some outputs with a shift register.
The issue is that this device have a ten led bar, and the first one is connected directly to the VCC.
Don’t ask why but when the module goes to an unknown status this led switch off too.
Looks like a power-on overcurrent protection. The leds at the S3 are all switched on.
I did not test if the module reboot from this situation using the RESET pin.
I keep testing.
Hit enter to search or ESC to close
Knowledge Base & Download Zone