Re: [lwip-users] Randomly delayed frame (STM32070 package including LwIP v1.4.1)

[stevestrong]

I made some further checks.

I activated LWIP_STATS but could not see anything on the serial debug
output, nothing changed, same issue.

Then made some tries with LWIP_DEBUG activated.

First I additionally activated both PBUF_DEBUG and TCP_DEBUG.
I got a lot of debug output, but could not detect any error message.
In this case, from 30 refreshes, I could not reproduce the issue anymore.
I cannot explain why.

Then I activated only TCP_DEBUG, without pbuf_debug.
In this case the issue can be reproduced almost at every refresh, with the
same situation as described in my previous email.

I captured the TCP packets and serial debug output, see attached.
long_delay.pcap <http://lwip.100.n7.nabble.com/file/t2167/long_delay.pcap>  
long_delay_debug_log.txt
<http://lwip.100.n7.nabble.com/file/t2167/long_delay_debug_log.txt>  
The time stamps in the debug log file are ~20 ms later that the wireshark
time stamps, probably due to the delay caused by USB serial.

If my interpretation is correct, from the timestamps can be seen that the
last Tx frame (nr 70) from STM32 with a delay of 2.5 seconds is triggered at
the same time with the 6th "tcp_slowtmr: polling application".

Anyone an idea what can be wrong here?

What should I check next?




--
Sent from: http://lwip.100.n7.nabble.com/lwip-users-f3.html