Simulation time and ending status.

[hazelPhd]

Hi,

I couldn't find where I can specify the time of the simulation? Also, when the simulation ends, is there a way to be notified that it ended instead of checking manually?

Dose the time calculation in Whitefield run as real-time? Or is it similar to the way Cooja does?

Thank you.

[nyrahul]

The time-calculation in Whitefield is in real-time. Thus for Whitefield, you do not have to explicitly pass the simulation time duration. Whenever, you think the time is up you can externally stop the Whitefield.

For e.g., to stop after 1 hour you can sleep 3600 && ./scripts/wfshell stop_whitefield

Hope it helps.

[nyrahul]

Can you share the wf.cfg and check the whitefield/logs/*.log files for any error?
Also, you can run the regression tests ./regression/regress.sh ./regression/basic to check if anything is failing.

[hazelPhd]

The above image is from airline.log from whitefield/logs/*.log . I've ran the command ./regression/regress.sh ./regression/basic and this is the result:
executing ./regression/basic/TC_two_contiki_nodes.test... waiting for routing table status... waiting 7/20 sec... cmd_node_osname="contiki" cmd_def_route=[NONE] cmd_def_route=fe80::302:304:506:0 cmd_rpl_stats.resets=2 cmd_route_table=fd00::302:304:506:1 128 fe80::302:304:506:1 cmd_ipv6_stats: rcvd=4, sent=2 cmd_icmp_stats: rcvd=5, sent=2 cmd_config_info: rt_maxsz=100, nbr_maxsz:100 [1] [TC_two_contiki_nodes.test] [PASS] [tot_fail:0] tc_msg:reached in 7 sec executing ./regression/basic/TC_nodeexec_invalid_cfg.test... [2] [TC_nodeexec_invalid_cfg.test] [PASS] [tot_fail:0] executing ./regression/basic/TC_no_stackline.test... [3] [TC_no_stackline.test] [PASS] [tot_fail:0] executing ./regression/basic/TC_invalid_cfg.test... [4] [TC_invalid_cfg.test] [PASS] [tot_fail:0] executing ./regression/basic/TC_two_mixed_nodes.test... [5] [TC_two_mixed_nodes.test] [FAIL] [tot_fail:1] tc_msg:expecting return 0 instead got 0 check /whitefield/regression/testcase_report.log for execution log... 1/5 testcases failed!!

[nyrahul]

This looks like your basic system is setup ok (1 failure case seems to be a flaky regression). The problem seems to be with scale i.e., 400 nodes. Can you send me your updated wf.cfg and I can try to check at my end? What is the stackline you experimenting with ? RIOT/Contiki/Contiki-ng?

[hazelPhd]

'#key[start-end]=val ... Description, isMandatory?, supportsRange?, exampleValue

numOfNodes=169

#include=config/prop_model.cfg

#---------[Airline configuration]-------
#randSeed=0xabcdef #If not set every time a random seed is used
fieldX=400 #field space in x direction ... currently 2D model is supp only.
fieldY=400 #field space in y direction
topologyType=grid #grid, randrect (ns3 RandomRectanglePositionAllocator),
gridWidth=5 #Grid topology width if the topologyType=grid
#nodePosition[1]=10,20,0 #Change the node position to the given vector
#nodePromiscuous[2]=1 #Sets promiscuous mode for node
#txPower=0 # Transmit Power (dBm)
#txPower[0]=7 # Transmit Power (dBm)

panID=0xabcd
NS3_captureFile=pcap/pkt
macPktQlen=20 #Maximum number of packets that can be outstanding on mac layer
macMaxRetry=3 #Max number of times the mac packet will be retried

'#---------[Stackline configuration]-------

Format:

nodeExec=path/to/bin $NODEID env1=abc env2=xyz

nodeExec[node-range]=path/to/bin $NODEID env1=abc env2=xyz

where node-range could be 0, 0-10 etc'

nodeExec=thirdparty/contiki-ng/examples/rpl-udp/udp-client.whitefield $NODEID UDPCLI_SEND_INT=30 AUTO_START=1 UDP_PAYLOAD_LE$
#nodeExec="thirdparty/RIOT/tests/whitefield/bin/native/riot-whitefield.elf" -w $NODEID
nodeExec[0]=thirdparty/contiki-ng/examples/rpl-udp/udp-server.whitefield.elf $NODEID
#nodeExec[1]="thirdparty/RIOT/tests/whitefield/bin/native/nodes-ngwhitefield.elf" -w $NODEID'

[hazelPhd]

I'm using Contiki-ng. When I start Whitefield it says that it started OK, but when I execute the monitor script, it shows that Whitefield is stopped.