Crash and assert failure during script with ToS and CCA mechanism

[Łukasz M]

Hi,

I work on a script concerning ToS, CCA and generally QoS.

I have tried two releases of 3.26 version: ns-3.26 and ns-3.26-dev and I got segmentation fault in one case and assert failure in another. I have no ideas how I can avoid it.

Can you take a look at it? It is very important, because my master thesis based on this script.

I attached two files with described problems and the script.

Best Regards,

Łukasz

[Łukasz M]

Can anyone help?

[Tommaso Pecorella]

Hi,

the Simulator::Destroy (); cal should be at the very end of the cycle. Otherwise the applications will be deleted and accessing their objects will be fatal.

Beside this, you did hit a bug. Please check Bugzilla if there's some bug open about this, and (in case there's nothing similar) contact the Wi-Fi module maintainer.

Thanks,

T.

On Sunday, January 15, 2017 at 11:21:45 AM UTC+1, Łukasz M wrote:

Can anyone help?

[Sebastien Deronne]

Which version are you using?

I hit a different issue when I run your script:

msg="Ipv4AddressGeneratorImpl::Add(): Address Collision: 192.168.1.1", file=../src/internet/model/ipv4-address-generator.cc, line=367

libc++abi.dylib: terminating

I suspect the bug was fixed in ns-3-dev and the new issue is no longer pointing to wifi.

[Łukasz M]

@Sebastien - names of pictures include used version.

@Tommaso - I did as you advised concerning "Simulator::Destroy()" and it works correct, but unfortunatelly, after some time, there is another bug:

msg="Packet in blockAck manager retry queue is not Qos Data", file=../src/wifi/model/block-ack-manager.cc, line=531

terminate called without an active exception

Command ['/home/bieniu/workspace/ns-allinone-3.26/ns-3.26/build/scratch/ht-wifi-network_tos'] terminated with signal SIGIOT. Run it under a debugger to get more information (./waf --run <program> --command-template="gdb --args %s <args>").

I used 3.26 version. Also, I attached the script.

Could you help with it?

Best Regards 

[Łukasz M]

I forgot about used script. 

[Tommaso Pecorella]

Good news everyone! -- Prof. Farnsworth

The bug is there BUT I'm not able to reproduce it consistently.

On my MacOS it shows.

On my Linux box it doesn't.

The hunt begins (and I have little or no time to hunt, sorry).

T

[Tommaso Pecorella]

Correction: it happens on Linux too, only much, much later in the simulation.

It seems a memory corruption issue, I'll try to run it through Valgrind.

T.

[Łukasz M]

Yes, it happens on Linux and, as you wrote, it occurs later, not at the beginning of the simulation.

Thanks in advance, Tommaso!

I will be waiting for your response and more details.

Best Regards,

Łukasz

[Sebastien Deronne]

Sorry but I get a different error if I run your new script:

msg="Packet has no Traffic ID", file=../src/wifi/model/qos-utils.cc, line=163

libc++abi.dylib: terminating

[Łukasz M]

Tommaso, did you check this issue already?

[Łukasz M]

@Sebastien - maybe you used another version of ns-3 than me. I have used ns-3.26 version.

As you can see, Tommaso has the same errors like me. 

[Gmail]

I use the latest ns-3-dev

--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to the Google Groups "ns-3-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ns-3-users+...@googlegroups.com.
To post to this group, send email to ns-3-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ns-3-users.
For more options, visit https://groups.google.com/d/optout.

[Łukasz M]

Tommaso? Do you have any solutions?

It is very important for me, because my master thesis depens on this. 

[Tommaso Pecorella]

Hi,

sorry but I'm extremely busy these days. As I said, it seems a memory issue, but I don't know which one.

My suggestion is to try with valgrind and an extremely short run. If there's a memory failure it should find it.

Cheers,

T.