Segmentation violation

[Shibi]

Hi,

I was wondering if I could get some help in finding out why the errors below are occuring. Analysis_Ex is being run on 2 million events, and around 1M events it crashes. Could really use some advice for this problem. Thanks.

Shibi

 *** Break *** segmentation violation



===========================================================
There was a crash.
This is the entire stack trace of all threads:
===========================================================
#0  0x00007f88e2231c3e in waitpid () from /lib/x86_64-linux-gnu/libc.so.6
#1  0x00007f88e21b7f5e in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#2  0x00007f88e598cab7 in TUnixSystem::StackTrace() () from /home/shibi/root/lib/libCore.so
#3  0x00007f88e598f393 in TUnixSystem::DispatchSignals(ESignals) () from /home/shibi/root/lib/libCore.so
#4  <signal handler called>
#5  0x00007f88e2cf17d1 in std::ostream::sentry::sentry(std::ostream&) () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#6  0x00007f88e2cf2fde in std::ostream& std::ostream::_M_insert<double>(double) () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#7  0x0000000000566c8b in Analysis_Ex::Run(ExRootTreeReader*, ExRootTreeReader*, ExRootTreeReader*, ExRootTreeWriter*) ()
#8  0x000000000045ef0e in main ()
===========================================================


The lines below might hint at the cause of the crash.
If they do not help you then please submit a bug report at
http://root.cern.ch/bugs. Please post the ENTIRE stack trace
from above as an attachment in addition to anything else
that might help us fixing this issue.
===========================================================
#5  0x00007f88e2cf17d1 in std::ostream::sentry::sentry(std::ostream&) () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#6  0x00007f88e2cf2fde in std::ostream& std::ostream::_M_insert<double>(double) () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#7  0x0000000000566c8b in Analysis_Ex::Run(ExRootTreeReader*, ExRootTreeReader*, ExRootTreeReader*, ExRootTreeWriter*) ()
#8  0x000000000045ef0e in main ()
===========================================================

[favereau]

Hi,

It looks more like a root-related issue. Could you just check how much of your memory is used by delphes at the time of the crash ?

[anonymous]

Hi,

I used 'top' to find the memory usage at different points during the run. The final line was taken as the crash was occurring. Thanks.

event 312545
29261 shibi 20 0 252m 162m 13m R 74 1.8 4:23.22 Analysis_Ex

event 547167
29261 shibi 20 0 262m 174m 13m R 77 1.9 7:28.72 Analysis_Ex

event 1007945
29261 shibi 20 0 291m 181m 13m D 75 2.0 13:52.60 Analysis_Ex

event 1023218
29261 shibi 20 0 291m 192m 13m R 76 2.1 14:10.73 Analysis_Ex

event 1042101
29261 shibi 20 0 294m 195m 13m R 76 2.2 14:28.95 Analysis_Ex

event 1116861
29261 shibi 20 0 294m 196m 13m S 47 2.2 15:26.91 Analysis_Ex

[anonymous]

Forgot to include the 'top' columns. They are like this:

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND

[anonymous]

Hi,

I'm not sure if the problem is a ROOT problem as said above, but the issue I was having was with my fstream object output to file. I instead used cout as a work around, but it would be better to not have to do that in cases where there are multiple outputs and where the files being analyzed are very large. Has that been an issue otherwise?

Thanks,
Shibi

[favereau]

Hi,

I have no reports of issues like this one, people are often processing big files with delphes. How big is the typical output file ?

[anonymous]

Hi,

My output is around 200 Go.

[anonymous]

That is, 200 Gb is the file size after reconstruction of the showered pythia files. The crash happens around 1.6M events have passed. About 1 out of every 200k events is accepted after cuts, so that the final output file size is of course very small.

Shibi

Replying to anonymous:

Hi,

My output is around 200 Go.

[lmitzka]

Hi,

I seem to have a similar problem. I my case the crash appears already at the first event that Delphes is run on, so memory is not an issue. Here is my error message:

 *** Break *** segmentation violation



===========================================================
There was a crash.
This is the entire stack trace of all threads:
===========================================================
#0  0x00007f389150645e in waitpid () from /lib/x86_64-linux-gnu/libc.so.6
#1  0x00007f389149ba99 in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#2  0x00007f389617ebb8 in TUnixSystem::StackTrace() ()
   from /archive/lmitzka/Doktor/root/lib/libCore.so
#3  0x00007f389617d683 in TUnixSystem::DispatchSignals(ESignals) ()
   from /archive/lmitzka/Doktor/root/lib/libCore.so
#4  <signal handler called>
#5  0x00000000004cc903 in HepMCConverter::~HepMCConverter() ()
#6  0x00000000005a1673 in main ()
===========================================================


The lines below might hint at the cause of the crash.
If they do not help you then please submit a bug report at
http://root.cern.ch/bugs. Please post the ENTIRE stack trace
from above as an attachment in addition to anything else
that might help us fixing this issue.
===========================================================
#5  0x00000000004cc903 in HepMCConverter::~HepMCConverter() ()
#6  0x00000000005a1673 in main ()
===========================================================

Has somebody an idea?
Thanks in advance,
Lukas

[favereau]

The converters/readers have been rewritten as part of the modulardelphes project. The new version should be released very soon, so please check teh website from time to time.

[lmitzka]

Hi,

could you give me a hint what is meant by "very soon"? I sincerely don't want to pressure here, but should I think of it as days, weeks or months?

Thanks again,
Lukas

[favereau]

It will be presented next tuesday in a workshop. Beta version will be public after that.

[favereau]

Delphes 3 beta is on the frontpage ! Could you try this version ?

[lmitzka]

Hello!

Thanks for the fast resolution. All files that failed before do work now.

Lukas