[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: please help: segmentation fault on large topology



Does it run out of memory? 

- Haobo

On Wed, 26 May 1999, Kamil Sarac wrote:

> Hello all,
> 
> I have a problem with a large topology simulation:
> 
> I have a 400 node transit stub (2 level) topology and I simulate a
> multicast session with one source and 400 receivers. 
> 
> My problem is: simulation fails after a while by exiting with a SIGSEGV.
> gdb gives the below stack state:
> 
> Program received signal SIGSEGV, Segmentation fault.
> 0x13d3bc in NewVar () at gen/ns_tcl.cc:16042
> 16042   EmbeddedTcl et_ns_lib(code);
> (gdb) bt
> #0  0x13d3bc in NewVar () at gen/ns_tcl.cc:16042
> #1  0x13c9ac in MakeUpvar () at gen/ns_tcl.cc:16042
> #2  0x13cb10 in Tcl_UpVar () at gen/ns_tcl.cc:16042
> #3  0xecf70 in OTclOInstVarOne (obj=0xfc96c58, in=0x298c18,
> frameName=0x2346f0 "1", 
>     varName=0xfc89120 "offset_", localName=0xfc89120 "offset_", flags=0)
> at otcl.c:1433
> #4  0xed070 in OTclOInstVarMethod (cd=0x0, in=0x298c18, argc=5,
> argv=0xefff2e10) at otcl.c:1457
> #5  0xeb3d4 in OTclDispatch (cd=0xfc96c58, in=0x298c18, argc=3,
> argv=0xefff2e90) at otcl.c:419
> #6  0xf01d0 in TclInvokeStringCommand () at gen/ns_tcl.cc:16042
> #7  0x10faac in TclExecuteByteCode () at gen/ns_tcl.cc:16042
> #8  0xf0dec in Tcl_EvalObj () at gen/ns_tcl.cc:16042
> #9  0xf0a94 in Tcl_Eval () at gen/ns_tcl.cc:16042
> #10 0xe8818 in TclObject::create_instvar () at gen/ns_tcl.cc:16042
> #11 0xe96e8 in TclObject::bind () at gen/ns_tcl.cc:16042
> etc. (I am attaching the complete stack state to this email).
> 
> When I run it with a topology of 100 nodes, it works fine.
> I use topo-gen.tcl to generate topology and I am using ns version 2.1b4a.
> 
> Is this due to some size limitation, if so is there a way to increase this
> limit or any other ideas ?
> 
> Thanks for your help.
> 
> Kamil.
> 
>