Red Hat Bugzilla – Bug 237649
[fix available] impress crash
Last modified: 2007-11-30 17:07:43 EST
Description of problem: OOo crashes randomly, often. Version-Release number of selected component (if applicable): openoffice.org-core-2.0.4-5.4.17.1 How reproducible: randomly Steps to Reproduce: 1. Seems to happen when deleting text, but that could be just a guess Additional info: (I) x.org loaded video driver of... (II) Loading /usr/lib/xorg/modules/drivers/i810_drv.so (III) Desktop is: GNOME (IV) libgcj version is: libgcj-4.1.1-52.el5.2-i386 (V) kernel is: Linux 2.6.18-8.1.1.el5.rod #1 SMP Fri Apr 13 13:02:43 EDT 2007 i686 i686 i386 (VI) OpenOffice.org core rpm version is: openoffice.org-core-2.0.4-5.4.17.1-i386 (VII) depth of root window: 24 planes (VIII) accessibility is: false ...start sestatus details ... SELinux status: disabled ...end sestatus details ... ...start stackreport details ... 0x750eee8: /usr/lib/openoffice.org2.0/program/libuno_sal.so.3 + 0x22ee8 0x750fb7b: /usr/lib/openoffice.org2.0/program/libuno_sal.so.3 + 0x23b7b 0x699420: + 0x420 (__kernel_sigreturn + 0x0) 0x58d1356: /usr/lib/openoffice.org2.0/program/libsvt680li.so + 0x2eb356 0x58d0eb6: /usr/lib/openoffice.org2.0/program/libsvt680li.so + 0x2eaeb6 (SfxUndoManager::AddUndoAction(SfxUndoAction*, unsigned char) + 0xf6) 0x70176bc: /usr/lib/openoffice.org2.0/program/libsd680li.so + 0x27d6bc 0x58d12ec: /usr/lib/openoffice.org2.0/program/libsvt680li.so + 0x2eb2ec (SfxUndoManager::EnterListAction(String const&, String const&, unsigned short) + 0x9c) 0x7017747: /usr/lib/openoffice.org2.0/program/libsd680li.so + 0x27d747 0x68ce2f7: /usr/lib/openoffice.org2.0/program/libsvx680li.so + 0x7552f7 (SdrModel::BegUndo(String const&) + 0x57) 0x6ee2e22: /usr/lib/openoffice.org2.0/program/libsd680li.so + 0x148e22 0x6ee2e56: /usr/lib/openoffice.org2.0/program/libsd680li.so + 0x148e56 0x6f68dd4: /usr/lib/openoffice.org2.0/program/libsd680li.so + 0x1cedd4 0x6e8d751: /usr/lib/openoffice.org2.0/program/libsd680li.so + 0xf3751 0x6edc440: /usr/lib/openoffice.org2.0/program/libsd680li.so + 0x142440 0x6e9489a: /usr/lib/openoffice.org2.0/program/libsd680li.so + 0xfa89a 0x3d6c12b: /usr/lib/openoffice.org2.0/program/libvcl680li.so + 0x28012b 0x3d6e199: /usr/lib/openoffice.org2.0/program/libvcl680li.so + 0x282199 0xb0f895: /usr/lib/openoffice.org2.0/program/libvclplug_gtk680li.so + 0x34895 0xb13151: /usr/lib/openoffice.org2.0/program/libvclplug_gtk680li.so + 0x38151 0x379f020: /usr/lib/libgtk-x11-2.0.so.0 + 0x130020 0x109f0b: /lib/libgobject-2.0.so.0 + 0x8f0b (g_closure_invoke + 0x12b) 0x11ae83: /lib/libgobject-2.0.so.0 + 0x19e83 0x11c147: /lib/libgobject-2.0.so.0 + 0x1b147 (g_signal_emit_valist + 0x667) 0x11c539: /lib/libgobject-2.0.so.0 + 0x1b539 (g_signal_emit + 0x29) 0x38b30e8: /usr/lib/libgtk-x11-2.0.so.0 + 0x2440e8 0x37984fa: /usr/lib/libgtk-x11-2.0.so.0 + 0x1294fa (gtk_propagate_event + 0x1ba) 0x37996c7: /usr/lib/libgtk-x11-2.0.so.0 + 0x12a6c7 (gtk_main_do_event + 0x317) 0x35f0ffa: /usr/lib/libgdk-x11-2.0.so.0 + 0x40ffa 0xd5a342: /lib/libglib-2.0.so.0 + 0x2b342 (g_main_context_dispatch + 0x182) 0xd5d31f: /lib/libglib-2.0.so.0 + 0x2e31f 0xd5d885: /lib/libglib-2.0.so.0 + 0x2e885 (g_main_context_iteration + 0x65) 0xaecc61: /usr/lib/openoffice.org2.0/program/libvclplug_gtk680li.so + 0x11c61 0x2257b7: /usr/lib/openoffice.org2.0/program/libvclplug_gen680li.so + 0x517b7 (X11SalInstance::Yield(bool, bool) + 0x37) 0x3b81708: /usr/lib/openoffice.org2.0/program/libvcl680li.so + 0x95708 (Application::Yield(bool) + 0x68) 0x3b817dc: /usr/lib/openoffice.org2.0/program/libvcl680li.so + 0x957dc (Application::Execute() + 0x3c) 0x4410279: /usr/lib/openoffice.org2.0/program/libsoffice.so + 0x26279 (desktop::Desktop::Main() + 0x1779) 0x3b872ac: /usr/lib/openoffice.org2.0/program/libvcl680li.so + 0x9b2ac 0x3b873b5: /usr/lib/openoffice.org2.0/program/libvcl680li.so + 0x9b3b5 (SVMain() + 0x35) 0x4401979: /usr/lib/openoffice.org2.0/program/libsoffice.so + 0x17979 (sal_main + 0x59) 0x4401a04: /usr/lib/openoffice.org2.0/program/libsoffice.so + 0x17a04 (main + 0x44) 0xb9ddec: /lib/libc.so.6 + 0x15dec (__libc_start_main + 0xdc) 0x80484a1: /usr/lib/openoffice.org2.0/program/simpress.bin + 0x4a1 ...end stackreport details ... ...start sample ldd details ... linux-gate.so.1 => (0x00da8000) libuno_sal.so.3 => /usr/lib/openoffice.org2.0/program/libuno_sal.so.3 (0x006b1000) libuno_salhelpergcc3.so.3 => /usr/lib/openoffice.org2.0/program/libuno_salhelpergcc3.so.3 (0x002fd000) libstore.so.3 => /usr/lib/openoffice.org2.0/program/libstore.so.3 (0x0097a000) libdl.so.2 => /lib/libdl.so.2 (0x001ec000) libpthread.so.0 => /lib/libpthread.so.0 (0x003e7000) libstlport_gcc.so => /usr/lib/openoffice.org2.0/program/libstlport_gcc.so (0x00f0c000) libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x001f0000) libm.so.6 => /lib/libm.so.6 (0x005de000) libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x00110000) libc.so.6 => /lib/libc.so.6 (0x00a4d000) libcrypt.so.1 => /lib/libcrypt.so.1 (0x00d39000) /lib/ld-linux.so.2 (0x001b9000) ...end sample ldd details ...
From the stack it's adding something to the undo which causes this, Is this from impress, or from draw ? It might help to know what view you are using if this is impress, e.g. notes view, outline view or normal.
Created attachment 153357 [details] likely patch for this This is the likely patch for this.
5.1.0 planning and bug selection is over, so proposing for 5.2.0
Yes, it is impress (the presentation tool) and I am in outline mode. Does this only happen in outline mode?
This request was evaluated by Red Hat Product Management for inclusion in a Red Hat Enterprise Linux maintenance release. Product Management has requested further review of this request by Red Hat Engineering, for potential inclusion in a Red Hat Enterprise Linux Update release for currently deployed products. This request is not yet committed for inclusion in an Update release.
Yes I will test it myself. Additionally, you can try to reproduce this by: 1) opening an existing relatively big presentation 2) switching to outline mode 3) making rapid changes to the outline, specifically deleting text and changing outline levels of items, etc. I don't quite grok the patch cause I don't know ooo well, but I would think that it's a combination of adding to the undo list that does this, with outline mode specifically
build underway
Packages are in /mnt/redhat/brewroot/packages/openoffice.org/2.0.4/5.4.22 if this works, please set to "verified"
Hi Nayfield I tested it for OOo version openoffice.org-core-2.0.4-5.4.17.1 with en_US, bn_IN locales. But Could not reproduce it still. :( May I request you to provide some more information how it was reproduced, and any special action that performed to reproduce it. I performed it with a huge document in en_US and bn_IN, in outline mode and tried to delete it both with Delete and Backspace key, and did repeated Undo. No how it reproduced. Please provide informations or test steps to follow.
Created attachment 154734 [details] OOffice crash on Undo
You have accessibility enabled. Can you repeat your tests without accessibility enabled. There are very likely accessibility bugs, and I'm more than willing to fix these given the various acks, but the original report is of undo in a non accessible environment, so testing with a11y enabled is more a test of a11y than a test of this problem and might be showing a different problem.
sure, no prob.
without a11y, i could not reproduce it either.
oky doky, well I suggest that we'll have to take this on trust then :-) The patch looks correct to me. mbabej: If you want to repeat the *a11y* tests on FC-7/rawhide and see if they still exist and file bugs if so that'd be great, then I'll take a look at that separate but also important problem.
Created attachment 155596 [details] traceback Reproduced twice the crash with openoffice.org-2.0.4-5.4.17 on x86_64 and once on i386 when I executed the steps in Comment #9 with the slides got from Comment #15 five times. After upgrade to openoffice.org-2.0.4-5.4.22 I did not find the crash in five times execution.
Reproduced twice the crash with openoffice.org-2.0.4-5.4.17 -> After upgrade to openoffice.org-2.0.4-5.4.22 I did not find the crash -> setting to verified if no-one disagrees
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on the solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHBA-2007-0849.html