Re: segfault at 0 ... error 4 in curl-loader

[bnoble <bn...@gm...>]

You bet, thanks for the quick reply:

Core was generated by `curl-loader -t 4 -f
/home/support/curl-loader/curl-loader-0.50/conf-examples/ht'.
Program terminated with signal 11, Segmentation fault.
#0  0x0000000000437de0 in Curl_done ()
(gdb) bt
#0  0x0000000000437de0 in Curl_done ()
#1  0x0000000000426a32 in multi_runsingle ()
#2  0x0000000000426e5f in curl_multi_perform ()
#3  0x0000000000428629 in curl_multi_socket_all ()
#4  0x0000000000411bfa in mperform_hyper (bctx=0x7fffff4aa410,
still_running=0x7ff32f33de94) at loader_hyper.c:683
#5  0x0000000000411520 in next_load_cb_hyper (fd=-1, kind=1,
userp=0x7fffff4aa410) at loader_hyper.c:398
#6  0x00000000004648f8 in event_base_loop ()
#7  0x0000000000411995 in mget_url_hyper (bctx=0x7fffff4aa410) at
loader_hyper.c:567
#8  0x00000000004118b0 in user_activity_hyper
(cctx_array=0x7ff330b42010) at loader_hyper.c:506
#9  0x000000000040aadf in batch_function (batch_data=0x7fffff4aa410)
at loader.c:345
#10 0x00007ff331d52070 in start_thread () from /lib64/libpthread.so.0
#11 0x00007ff3310db11d in clone () from /lib64/libc.so.6
#12 0x0000000000000000 in ?? ()
(gdb) up
#9  0x000000000040aadf in batch_function (batch_data=0x7fffff4aa410)
at loader.c:345
345       rval = ua_array[loading_mode] (bctx->cctx_array);
(gdb) up
#10 0x00007ff331d52070 in start_thread () from /lib64/libpthread.so.0
(gdb) up
#11 0x00007ff3310db11d in clone () from /lib64/libc.so.6
(gdb) up
#12 0x0000000000000000 in ?? ()
(gdb) up

On Tue, Dec 1, 2009 at 1:07 PM, Robert Iakobashvili <cor...@gm...> wrote:
> Hi bnoble,
>
> On Tue, Dec 1, 2009 at 10:42 PM, bnoble <bn...@gm...> wrote:
>>
>> Hello,
>>
>> I receive the following error when attempting to run the 10k
>> configuration example, or any configuration which exceeds 1500-3000
>> CLIENTS_NUM_MAX:
>>
>> curl-loader[3975]: segfault at 0 ip 0000000000440094 sp
>> 00007f96b17c2bc0 error 4 in curl-loader[400000+8a000]
>>
>> Sometimes it crashes immediately, sometimes it runs until ~3000
>> clients are loaded.  I have played with every conceivable sysctl
>> parameter with no luck.  I have tried running the application with
>> four threads, I have adhered to every recommendation I could find.  Am
>> I missing something obvious?  I encountered similar problems with an
>> old Ubuntu VM previously; this, however, is not a VM instance.
>>
>> OS: OpenSuSE 11.1
>> Linux 2.6.27.37-0.1-default #1 SMP 2009-10-15 14:56:58 +0200 x86_64
>> x86_64 x86_64 GNU/Linux
>> Quad-core 2Ghz Xeon (Tylersburg) 4M cache with 8GB RAM
>>
>> Thanks.
>
>
> Are you sure, that your sockets/open descriptors limit is sufficient?
>
>
> Please:
>
> 1. Compile it with debugging information
> $make clean; make debug=1 optimize=0
>
> 2. keep core allowed in your shell
> #ulimit -c unlimited
>
> 3. Run it till crash and get the core file
>
> 4. open the core and the curl-loader application by gdb debugger
>    and explore the backtrace by bt command
>
> You might wish to send the trace to this list.
>
>
> --
> Truly,
> Robert Iakobashvili, Ph.D.
> ......................................................................
> www.ghotit.com
> Assistive technology that understands you
> ......................................................................
>
> ------------------------------------------------------------------------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing.
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> curl-loader-devel mailing list
> cur...@li...
> https://lists.sourceforge.net/lists/listinfo/curl-loader-devel
>
>