在 aqua-sim 中运行基于 vectore 的转发协议后出现分段错误 [英] Segmentation fault after running vectore based forwarding protocol in aqua-sim

问题描述

有人可以帮我追踪这个错误吗，我在 400 个节点上使用基于向量的转发协议 tcl 示例，但我收到了这个错误

can someone please help me trace this error, i'm using vector based forwarding protocol tcl example on 400 nodes and i got this error

UWSink (id:0): I get the packet data no.0 from 378 
SK 0: Num_Recv 412, InterArrival 0.298128
UWSink (id:0): I get the packet data no.11 from 262 
SK 0: Num_Recv 413, InterArrival 0.057902
UWSink (id:0): I get the packet data no.4 from 79 
SK 0: Num_Recv 414, InterArrival 0.368108
UWSink (id:0): I get the packet data no.10 from 79 
SK 0: Num_Recv 415, InterArrival 0.741583
UWSink (id:0): I get the packet data no.4 from 262 
SK 0: Num_Recv 416, InterArrival 0.216878

Program received signal SIGSEGV, Segmentation fault.
0x000000000060871c in TclRegisterLiteral ()

推荐答案

另请参考.您的另一个 AquaSim VBF 线程，其中包含大部分模拟文件:VBFP 可能不起作用在水上模拟

Also ref. your other AquaSim VBF thread, with the most of a simulation file : VBFP doesnt work probably in aqua-sim

... 节点 4 - 398 丢失.我通过一个随机的外部文件填写了一些节点设置，4 - 85:source 82-nodePositions.我建议您将完整的模拟文件.tcl 上传到例如Dropbox，然后发布一个共享链接.

... node 4 - 398 are missing. I filled in some node settings, 4 - 85 by a random external file : source 82-nodePositions . I will suggest that you upload the complete simulation file.tcl to e.g. Dropbox, and post a shared link.

您的错误:像 AquaSim 这样的旧 ns2 代码可能会因迟到的 glibc/libc6 而崩溃，例如Ubuntu 17.xx .

Your errors : Old ns2 code like AquaSim can crash with a late glibc / libc6, e.g. Ubuntu 17.xx .

我使用三个Linux 操作系统"进行了测试:PCLinuxOS 2017 - 64、Ubuntu 16.04 - amd64、CentOS 6.9 - i386.

I made the tests with three 'Linux OS' : PCLinuxOS 2017 - 64, Ubuntu 16.04 - amd64, CentOS 6.9 - i386.

示例结果，CentOS 6.9 - 32 :

Example result, CentOS 6.9 - 32 :

$ ns-aqua-el6 A.Moh-VBF-n400.tcl

the file name is test.data
the sending interval is 10.0
num_nodes is set 401
Width=100
INITIALIZE THE LIST xListHead
Node 400 is sending first!!
starting Simulation...
GOD: the old file name istest.data
GOD: the new file name istest.data
SINK 0 : terminates (send 0, recv 0, cum_delay 0.000000)
god: the energy consumped is 16681.600000
NS EXISTING...

跟踪文件 test.tr 没问题.92.4 ... 96.8 MB.nam trace test.nam 没有"trace lines" :表示模拟文件.tcl 中缺少某些 nam 设置.或者是我的随机节点设置造成的.

The trace file test.tr is OK. 92.4 .. 96.8 MB. The nam trace test.nam has no "trace lines" : Indicates that some nam settings are missing in the simulation file.tcl . Or caused by my random node settings.

完整结果https://www.dropbox.com/s/fov2r82s88rfhaq/A.Moh-Nov12-result2.tar.gz?dl=0

参考.您的评论 #2:得到了您的新示例VBF 401 节点".在 PCLinuxOS 2017、CentOS 7.4、Ubuntu 16.04、CentOS 6.9 - 64 和 32 位 CentOS 6.9 上测试 → 两个可用结果可以从 https://www.dropbox.com/s/nl6pnyz3cdklu5g/VBF_example_5-n401-result.tar.gz?dl=0

EDIT : Ref. your comment #2 : Got your new example, "VBF 401 nodes". Tested with PCLinuxOS 2017, CentOS 7.4, Ubuntu 16.04, CentOS 6.9 - 64, and the 32bits CentOS 6.9 → The two usable results can be downloaded from https://www.dropbox.com/s/nl6pnyz3cdklu5g/VBF_example_5-n401-result.tar.gz?dl=0

这篇关于在 aqua-sim 中运行基于 vectore 的转发协议后出现分段错误的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！