Sporadic resets using X-CUBE-AWS-H7S

[nirnittpl]

I am using X-CUBE-AWS-H7S (v1.0.0) on a STM32H7S78-DK.

The bootloader and app were built directly from the package with no source code changes at all, yet the UART-based CLI shows that the board is reset every now and then (sometimes every ~5 minutes, sometimes after more than an hour).

Ruled out power supply issues and ST-Link issues.

Resets happening both with an Ethernet cable connected and without it, both with a valid AWS IoT configuration and without it, both when using a serial console (with an ST-Link based virtual COM port) and without it.

Stack/heap utilization seems normal (and stable) up to the moment where the board is reset.

Built with no errors using STM32CubeIDE with no configuration changes (OEMiROT, overwrite mode)

 

Is this a known issue with X-CUBE-AWS-H7S ?

 

[alister]

Possibly the CMSIS v2's tcpip_thread stack issue described at https://community.st.com/t5/stm32cubemx-mcus/bugs-caveats-misc-fixes-stm32h7s-cube/td-p/710992

 

[nirnittpl]

Thank you @alister 

Unfortunately increasing the tcpip_thread stack size did not help. Sporadic reboots are continuing.

(same behavior even when I kill the lwIP task through the CLI soon after wakeup, along with EthData and EthNet, as well as without killing those)

Is there a public git repo for X-CUBE-AWS-H7S in which I can track changes and possible fixes for this issue moving forward ?