Bugzilla – Bug 2350
ASSERT in ns3::MacLow::ReceiveError
Last modified: 2016-04-21 03:43:54 UTC
Created attachment 2357 [details] wifi topology i use I'm using ns-3.25 wifi module and having the following assert, is there anything wrong with my configuration or is it a bug in ns3? I have attached the relating files. No problem when running: AddIetfRmcatWifiTestCase4_4(50, 300, 2); AddIetfRmcatWifiTestCase4_4(50, 300, 4); AddIetfRmcatWifiTestCase4_4(50, 300, 8); Happened when running: AddIetfRmcatWifiTestCase4_4(50, 300, 16); AddIetfRmcatWifiTestCase4_4(50, 300, 32); Program received signal SIGABRT, Aborted. 0x00007fff90d65f06 in __pthread_kill () from /usr/lib/system/libsystem_kernel.dylib (gdb) bt #0 0x00007fff90d65f06 in __pthread_kill () from /usr/lib/system/libsystem_kernel.dylib #1 0x00007fff816c04ec in pthread_kill () from /usr/lib/system/libsystem_pthread.dylib #2 0x00007fff908136e7 in abort () from /usr/lib/system/libsystem_c.dylib #3 0x00007fff8c20cf81 in abort_message () from /usr/lib/libc++abi.dylib #4 0x00007fff8c23296a in default_terminate_handler() () from /usr/lib/libc++abi.dylib #5 0x00007fff8c23019e in std::__terminate(void (*)()) () from /usr/lib/libc++abi.dylib #6 0x00007fff8c23022d in std::terminate() () from /usr/lib/libc++abi.dylib #7 0x000000010487aeee in ns3::MacLow::ReceiveError (this=0x106f68c30, packet=..., rxSnr=78992063989.707458, isEndOfFrame=true) at ../src/wifi/model/mac-low.cc:864 #8 0x00000001048b2580 in ?? () from /Users/jeromy/work/cisco/ns3-mari-docker/wx2/ns-allinone-3.25/ns-3.25/build/libns3.25-wifi-debug.dylib #9 0x00007fff5fbfc000 in ?? () #10 0x00007fff5fbfc030 in ?? () #11 0x000000010487a070 in ?? () at ../src/wifi/model/mac-low.cc:2817 from /Users/jeromy/work/cisco/ns3-mari-docker/wx2/ns-allinone-3.25/ns-3.25/build/libns3.25-wifi-debug.dylib #12 0x000000010487a070 in ?? () at ../src/wifi/model/mac-low.cc:2817 from /Users/jeromy/work/cisco/ns3-mari-docker/wx2/ns-allinone-3.25/ns-3.25/build/libns3.25-wifi-debug.dylib #13 0x0000000106f68c30 in ?? () #14 0x000000010487a070 in ?? () at ../src/wifi/model/mac-low.cc:2817 from /Users/jeromy/work/cisco/ns3-mari-docker/wx2/ns-allinone-3.25/ns-3.25/build/libns3.25-wifi-debug.dylib warning: (Internal error: pc 0x0 in read in psymtab, but not in symtab.) #15 0x0000000000000000 in ?? () (gdb) f 7 #7 0x000000010487aeee in ns3::MacLow::ReceiveError (this=0x106f68c30, packet=..., rxSnr=78992063989.707458, isEndOfFrame=true) at ../src/wifi/model/mac-low.cc:864 864 NS_ASSERT (m_lastReceivedHdr.IsQosData ()); (gdb) list 859 if (hdr.GetAddr1 () != m_self) 860 { 861 NS_LOG_DEBUG ("hdr addr1 " << hdr.GetAddr1 () << "not for me (" << m_self << "); returning"); 862 return; 863 } 864 NS_ASSERT (m_lastReceivedHdr.IsQosData ()); 865 NS_LOG_DEBUG ("last a-mpdu subframe detected/sendImmediateBlockAck from=" << m_lastReceivedHdr.GetAddr2 ()); 866 m_sendAckEvent = Simulator::Schedule (GetSifs (), 867 &MacLow::SendBlockAckAfterAmpdu, this, 868 m_lastReceivedHdr.GetQosTid (),
Created attachment 2358 [details] topo header file
Created attachment 2359 [details] test suite i run
It seems possibly related to bug 2313 which was recently closed.
I'm using the least release 3.25. It can easily reproduced here.
It seems you are using MinstrelWifiManager with HT wifi, you should use MinstrelHtWifiManager, which works with both HT and non-HT. It is pending to control this issue as part of bug 1797. It would be better if you upload a diff file so as to easily reproduce the bug. Also, some apps you created are missing and couldn't run your example.
Created attachment 2361 [details] this is a ns3 module ./test.py -s ietf-rmcat-wifi-test-suite -w mari.html -r
Please attach relevant code and try to isolate the issue in a small test case. I actually suspect a wrong configuration as Matias pointed, so please check your configuration first.
I have attached the module that can running above. https://www.nsnam.org/bugzilla/attachment.cgi?id=2361 I also changed to MinstrelHtWifiManager, the issue still happens. There is only a single test there by running ./test.py -s ietf-rmcat-wifi-test-suite -w mari.html -r The wifi configuration is in mari/model/qostool/mari-topo.cc.
if this is this fixed, i can try it out
Does patch for bug 2313 show some improvements?
Hi Jiantao. After dealing with some compilation issues I was able to run your test and reproduce the bug. Be aware that various bugs related to AMPDU support are being discussed right now. I've found that fix for bug 2376 solves your issue. Can you confirm?
It looks like this bug is a duplicate, since it is fixed by proposed patches for currently open issues (bug 2376, bug 2378, bug 2379 and bug 2380). Could you confirm that applying those patches does solve the problem (and especially patch for bug 2376)?
(In reply to Matías Richart from comment #11) > Hi Jiantao. > > After dealing with some compilation issues I was able to run your test and > reproduce the bug. > > Be aware that various bugs related to AMPDU support are being discussed > right now. > I've found that fix for bug 2376 solves your issue. > > Can you confirm? applied the patch fix the crash, thanks!
Patches for other bugs does solve the problem, thus moving to duplicate. *** This bug has been marked as a duplicate of bug 2376 ***