HTTP POST with NCS v1.0.0-rc4

[Hoang]

Dear Bjørn Kvaale, 

I just tried to run LTE-M/HTTP POST with NCS v1.0.0-rc4 (with nrf git fetch origin pull/943/head:branch1)

but, I COULD NOT POST any more:

**** Booting Zephyr OS build v1.14.99-ncs2-rc4-495-ga8933a2ead01 *****
LTETESTING is starting 
 LTE is connecting........... 
Checking LTE_M signal.............

+COPS: 0,2,"24201",7

OK

OK

+CEREG: 2,1,"765D","0110C303",7

OK


Original String is: +CEREG: 2,1,"765D","0110C303",7

OK

Found the LTE_M signal 
Closing socket

  find LTE_M signal........... 
***** USAGE FAULT *****
  Illegal load of EXC_RETURN into PC
***** Hardware exception *****
Current thread ID = 0x200203a8
Faulting instruction address = 0x0
Fatal fault in thread 0x200203a8! Aborting.
Exception occurred in Secure State
***** HARD FAULT *****
  Fault escalation (see below)
***** BUS FAULT *****
  Precise data bus error
  BFAR Address: 0x50008120
***** Hardware exception *****
Current thread ID = 0x20021ae4
Faulting instruction address = 0x21992
Fatal fault in ISR! Spinning...

It worked normal with the old version v0.4.0

1. when i used the same setup, i got error: 

1> zephyr/subsys/net/ip/libsubsys__net__ip.a(net_if.c.obj): In function `net_if_ipv6_calc_reachable_time':
1> D:/hng/projects/CyCollector/ncs_v1.0.0_rc4/zephyr/subsys/net/ip/net_if.c:2067: undefined reference to `sys_rand32_get'
1> collect2.exe: error: ld returned 1 exit status

2. I added the option: 

# Disable native network stack to save some memory
CONFIG_NET_IPV4=n
CONFIG_NET_IPV6=n

then, Build Solution: Build completed

but, I got the error 

My code:

LTE_M_NB-IoT.7z

Thanks so much

Regards, 

hng

[Simon]

I was able to post to a webserver, with ncs v1.0.0 and modem v1.0.0. Could you test my attached example?

http_post.zip

Best regards,

Simon

[Hoang]

Hi @Simon

Thanks so much, I just backed from vocation, I will try your suggestion and contact with you soon

Mvh, 

Hoang Nguyen 

[Hoang]

Dear Simon, 

Ja, it works, but I got another "funny" error ( because I dont know why!!!) 

I removed: 

printk("Hello from start of main");
     char tmp_str[MAXLINE + 1];
     snprintf(tmp_str,500,
		"$GPGGA,%02d%02d%02d.200,%8.3f,N,%09.3f,W,1," 
		"12,1.0,0.0,M,0.0,M,,*", 2019, 07, 23, 59.3033, 10.454);
     //printf("tmp_string is: %s \n",tmp_str); 
     

and just ran:

int main(void){

     printk("Hello from start of main \n");
     char send_buf[MAXLINE + 1];
     int send_data_len_post = snprintk(send_buf,
                                         500,
				         POST_TEMPLATE, HTTP_PATH,
				         HTTP_HOST, strlen(TEST_STRING),
				         TEST_STRING);

     printf("send_buf is: %s \n",send_buf);
     printk("send err: %d\n\r", send_data_len_post);
    
     //snprintk_post("OMG whyyyyyyyyyyyy \n"); 
     while (1) {
          //5 seconds delay
          k_sleep(5000);
          
          }
   }

it didnt work:

SPM: NS image at 0xc000
SPM: NS MSP at 0x20024cf8
SPM: NS reset vector at 0xec1d
SPM: prepare to jump to Non-Secure image.
***** Booting Zephyr OS v1.14.99-ncs2 *****
Hello from start of main %
***** USAGE FAULT *****
  Illegal use of the EPSR
***** Hardware exception *****
Current thread ID = 0x20021e9c
Faulting instruction address = 0xec1c
Fatal fault in essential thread! Spinning...

then, I tried to add an extra variable (but i will not use it !!!!)

char tmp[MAXLINE + 1];

and, it works:

SPM: NS image at 0xc000
SPM: NS MSP at 0x20024cf8
SPM: NS reset vector at 0xec1d
SPM: prepare to jump to Non-Secure image.
***** Booting Zephyr OS v1.14.99-ncs2 *****
Hello from start of main 
send_buf is: POST /? HTTP/1.1

Host: 142.93.37.192

Connection: keep-alive

Content-type: application/x-www-form-urlencoded

Content-length: 19



111 from nRF9160 DK 
send err: 153

I dont know why!!!????

Thanks,

mvh

Hoang 

[Simon]

I am sorry for the delay. The support staff is reduced during the summer holidays, and you may experience delayed answers. Have you been able to resolve this issue?

Best regards,

Simon

[Hoang]

Hi Simon, 

I understand it, not yet

Mvh

Hoang

[Simon]

Hmm.. That is really strange. I was able to recreate the error message. I tried to minimize the project to try to locate the problem, and I eventually ended up with the following project, which was giving the same error:

#include <zephyr.h>
#include <net/socket.h>
#include <stdio.h>

int main(void)
{
     //char tmp_str2[2];
     char send_buf[4096];

     while (1) {
          
          }
	
}

However, if I set CONFIG_SIZE_OPTIMIZATION=y (and removed the optimize nothing config), then it worked fine. 

I will report the problem internally. In the mean time, I guess you have to declare tmp_str as a temporary solution.

Best regards,

Simon

[Simon]

The solution to this is to set CONFIG_NO_OPTIMIZATIONS to y, and increase the main stack size through CONFIG_MAIN_STACK_SIZE or place it in the data section instead of on the stack by declaring and initializing it outside of the main function.

Best regards,

Simon