Crash PPPOS with mQTT error AWS cloud disconnected

If you are asking a question, please follow this template:

  1. My goal is: Publish the massage on AWS IoT. Massage is being published every time it gets connected with PPPOS.
  2. My actions are: Every time After 4 callback function this 3 massages gets printed on console and the cloud gets disconnected and retry again to connected.
  3. The result I see is:
  4. My expectation & question is: Why these 3 massages are printed console? why This is generated?

#define MBEDTLS_ERR_NET_SEND_FAILED -0x004E /**< Sending information through the socket failed. */
Looks like you are trying to send data while pppos is not yet connected.
See Command failed: ATE0, Error 9 (phase 0), reconnect

It publishes data for 6 to 7 times at start. and then this problem occurs and It again goes from reconnection. And again gets connected.

I too am having this issue on some of my devices (also connecting to an AWS IoT ATS endpoint. I find that many devices are fine, but several have intermittent issues. Let me know if you managed to resolve the issue?

@rojer - seems it’s not just me still having issues with this…

@Shivam_Kavathe did your issues occur on a Wrover ESP32 board with PSRAM enabled?

Modules is same,Wrover ESP32 but, I dont know about PSRAM enabled. How can I check if PSRAM enabled?

Do you have this in your mos.yml build_vars:
CONFIG_SPIRAM_SUPPORT=y

You can see if it is enabled in the boot messages

[May 11 17:54:32.603] I (479) spiram: Found 64MBit SPI RAM device
[May 11 17:54:32.608] I (484) spiram: SPI RAM mode: flash 40m sram 40m
[May 11 17:54:32.614] I (489) spiram: PSRAM initialized, cache is in low/high (2-core) mode.
...
[May 11 17:54:33.522] I (1407) spiram: SPI SRAM memory test OK
...
[May 11 17:54:33.569] I (1445) spiram: Adding pool of 4096K of external SPI memory to heap allocator
...
May 11 17:54:33.595] I (129) spiram: Reserving pool of 32K of internal memory for DMA/internal allocations