cisco-system-traffic-generator/trex-core

TRex crashed all time when running ASTF test cases

ciscohack opened this issue · 2 comments

Hi Team,

TRex ASTF profile test cases not running and causing a traceback on version V3.04. Below is the traceback. it would be good if someone addressed this issue. all astf script closed with below traceback

Here are the someone of scripts detail I used and found issue.

https://github.com/cisco-system-traffic-generator/trex-core/blob/master/scripts/astf/emix2.py

https://github.com/cisco-system-traffic-generator/trex-core/blob/master/scripts/astf/http_eflow2.py

https://github.com/cisco-system-traffic-generator/trex-core/blob/master/scripts/astf/http_advanced.py

Command I ran is :--

./t-rex-64 --cfg /etc/trex_cfg_n29.yaml --astf -f astf/emix2.py -m 1 -l 1000 -t traffic_per=0.9,s_delay=100000

Error: signal 11:

*** traceback follows ***

1 0x5cb066 ./_t-rex-64() [0x5cb066]
2 0x7f6d4f17b630 /lib64/libpthread.so.0(+0xf630) [0x7f6d4f17b630]
3 0x55250c CFlowTable::rx_handle_packet(CTcpPerThreadCtx*, rte_mbuf*, bool, unsigned char) + 492
4 0x58d273 unsigned short CFlowGenListPerThread::handle_rx_pkts(bool) + 259
5 0x58aea6 CFlowGenListPerThread::handle_rx_flush(CGenNode*, bool) + 358
6 0x5bcc3f CNodeGenerator::handle_slow_messages(unsigned char, CGenNode*, CFlowGenListPerThread*, bool) + 335
7 0x555582 int CNodeGenerator::flush_file_realtime<24, false>(double, double, CFlowGenListPerThread*, double&) + 1394
8 0x821492 TrexDpCoreAstfBatch::start_astf() + 1586
9 0x8217b9 TrexDpCoreAstfBatch::start_scheduler() + 9
10 0x728f19 TrexDpCore::start() + 89
11 0x5b257d CFlowGenListPerThread::start(std::string&, CPreviewMode&) + 93
12 0x534457 CGlobalTRex::run_in_core(unsigned char) + 407
13 0x55a1a9 ./_t-rex-64() [0x55a1a9]
14 0xa07072 eal_thread_loop + 434
15 0x7f6d4f173ea5 /lib64/libpthread.so.0(+0x7ea5) [0x7f6d4f173ea5]
16 0x7f6d4e262b0d clone + 109

*** addr2line information follows ***

basic_string.h:3295
??:0
flow_table.cpp:666
bp_sim_tcp.cpp:375
bp_sim_tcp.cpp:163
bp_sim.cpp:3996
bp_sim.cpp:3640
trex_astf_batch.cpp:134
trex_astf_batch.h:57
trex_dp_core.cpp:142
bp_sim.cpp:4496
trex_watchdog.h:109
main_dpdk.cpp:6360
eal_thread.c:141
??:0
??:0

./t-rex-64: line 106: 5113 Segmentation fault ./_$(basename $0) $INPUT_ARGS $EXTRA_INPUT_ARGS

hhaim commented

@ciscohack is it an issue only with v3.04?

@hhaim I have seen the same crash on v3.04 as well but then I rebooted the TRex server and after that no crash observed on either version