Re: Severe problem with PyRoot - file segmentation fault

Wim,

> I've never seen it happen for a Linux box, though. What
> is the output of:
>
> $ ldd $ROOTSYS/lib/libHist.so

[james_at_slc4vm PdfErrors]$ ldd /home/james/root/lib/libHist.so

	libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x00111000)
	libm.so.6 => /lib/tls/libm.so.6 (0x00f70000)
	libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x00e1f000)
	libc.so.6 => /lib/tls/libc.so.6 (0x001e5000)
	/lib/ld-linux.so.2 (0x00597000)

> Do you have a stack trace? Thanks.

The best I can do is what I've just reported when reproducing the error in ROOT (Python just swallows it and says 'Segmentation fault'):

root [0] gSystem->Load("libPyROOT")
Warning in <TClassTable::Add>: class TFile already in TClassTable ... Lots of these ...
Warning in <TClassTable::Add>: class TViewer3DPad already in TClassTable (int)0

root [1] TFile *f = new TFile("test.root", "recreate")
root [2] TH1F *h1 = new TH1F("hist", "hist", 10, 0, 10)
root [3] f->Close()
root [4] .q

• Break *** segmentation violation Generating stack trace...
• glibc detected *** corrupted double-linked list: 0x00831858 ***

>> Subsequently inspecting the root file shows it to be empty:
>
> Yes, an f.Write() is needed.

Has this changed recently? I don't recall I used to have to do this, or perhaps I used to be sloppy and call Write() but not Close().

Regards,
James.

>
>
> Best regards,
> Wim
> --
> Wim.Lavrijsen_at_cern.ch -- WLavrijsen_at_lbl.gov -- www.lavrijsen.net
>
> "Your day will be somewhat dictated by authority." --fortune
> cookie
Received on Fri Jun 13 2008 - 18:58:11 CEST