Skip to main content



      Home
Home » Eclipse Projects » Eclipse Titan » Wrong decoding of matched / unmatched messages paths in log files(log files)
Wrong decoding of matched / unmatched messages paths in log files [message #1833350] Fri, 09 October 2020 18:58 Go to next message
Eclipse UserFriend
Hi,

I have here some strange decoding of the message paths within the log files. See attached log file and minimal extract from the SocketCAN test port. (SocketCAN is not needed in operation)

Typical examples are:
00:54:25.987967 4 MATCHING Isobus_test_decoder_debug.ttcn:457(function:f_ptc_J1939VTApplication) Matching on port pt_j1939PDU .phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq.pdu.pdu.ecu2vt{
    getNumberOfSoftKeysReq := {
        vtfunction := 194
    }
} with {
    getHardwareReq := {
        vtfunction := 199
    }



Search for:
".phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq.pdu.pdu."


00:54:25.987784 4 PORTEVENT Isobus_test_decoder_debug.ttcn:223(altstep:alt_awaitPhaseStartReq) Receive operation on port pt_sync succeeded, message from mtc: @Isobus_test_decoder_debug.Command_MTC2PTC : {
    phaseStartReq := {
        phase := e_testbodyEnd (8),
        phase_int := 8
    }
} id 9


Search for:
" .phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq.phaseStartReq"


Here is how to compile the code on a linux kernel with SocketCAN-j1939 kernel module installed. E .g. Ubuntu 20.05. ( Linux kernel >= 5.4)

Unzip the attached SockectCAN test port with J1939 support.
cd demo
make


Run the ttcn test cases:
ttcn3_start SocketCAN J1939.cfg J1939_Isobus_test_decoder_debug.tc_can_j1939_IsobusVtEcuSimulationDebug


ttcn3_logmerge -o log_merged.log *.log;ttcn3_logformat -o log_merged_format.log log_merged.log;
<editor> log_merged_format.log


Re: Wrong decoding of matched / unmatched messages paths in log files [message #1833362 is a reply to message #1833350] Sun, 11 October 2020 06:50 Go to previous messageGo to next message
Eclipse UserFriend
Hi Michael,

I tried to compile on an Ubuntu 18.04 machine, but the compilation fails with

:
g++ -c -DLINUX -I. -I/home/james00/titan.core/Install/include -I. -I../src -Wall -O2  -o J1939_Isobus_test_decoder_debug.o J1939_Isobus_test_decoder_debug.cc
g++ -c -DLINUX -I. -I/home/james00/titan.core/Install/include -I. -I../src -Wall -O2  -o Isobustest_test_encoder.o Isobustest_test_encoder.cc
g++ -c -DLINUX -I. -I/home/james00/titan.core/Install/include -I. -I../src -Wall -O2  -o SocketCAN_PT.o ../src/SocketCAN_PT.cc
../src/SocketCAN_PT.cc:19:10: fatal error: linux/can/j1939.h: No such file or directory
 #include <linux/can/j1939.h>
          ^~~~~~~~~~~~~~~~~~~
compilation terminated.
Makefile:177: recipe for target 'SocketCAN_PT.o' failed
make: *** [SocketCAN_PT.o] Error 1



Are there maybe some missing requirements possibly not listed in README.md?

BR

Elemer
Re: Wrong decoding of matched / unmatched messages paths in log files [message #1833364 is a reply to message #1833362] Sun, 11 October 2020 10:17 Go to previous messageGo to next message
Eclipse UserFriend
Hi Elemer,

your kernel is too old. Your SocketCAN implementation does not yet provide the j1939 interface file "linux/can/j1939.h".

But in this case this should not be a problem, as the example used in does not use the j1939 SockectCAN test port to demonstrate the bug.

ttcn3_start SocketCAN J1939.cfg J1939_Isobus_test_decoder_debug.tc_can_j1939_IsobusVtEcuSimulationDebug



To compile the SocketCAN test port without J1939 interface the file "SocketCAN_PT.cc" provides a define:

In file /src/SocketCAN_PT.cc line 17 please uncomment:
#define CANJ1939_SUPPORT
to:
//#define CANJ1939_SUPPORT

However, there are 2 newer sections in the file, which I have forgotten to wrap with
#ifdef CANJ1939_SUPPORT
...
#endif //CANJ1939_SUPPORT

This needs to be fixed in future releases.

Therefore it is easiest, if you replace the "SocketCAN_PT.cc" with the attached file. Then it should compile through.


[Updated on: Sun, 11 October 2020 10:20] by Moderator

Re: Wrong decoding of matched / unmatched messages paths in log files [message #1833454 is a reply to message #1833364] Tue, 13 October 2020 12:50 Go to previous message
Eclipse UserFriend
Hi Michael,

A trouble report has been created:

https://bugs.eclipse.org/bugs/show_bug.cgi?id=567844

BR

Elemer
Previous Topic:OpenSSL 1.1.1.g not supported
Next Topic:Compilation on Windows with Cygwin
Goto Forum:
  


Current Time: Thu Apr 24 14:04:05 EDT 2025

Powered by FUDForum. Page generated in 0.04286 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software

Back to the top