Created attachment 2947 backtrace of the crash forwarded from https://bugzilla.redhat.com/show_bug.cgi?id=584455 architecture: i686 Attached file: backtrace cmdline: /usr/libexec/tvp-svn-helper --log /home/bdhameli/workspace2/SIPRe component: thunar-vcs-plugin executable: /usr/libexec/tvp-svn-helper kernel: 2.6.32.11-99.fc12.i686 package: thunar-vcs-plugin-0.1.2-2.fc12 rating: 4 reason: Process /usr/libexec/tvp-svn-helper was killed by signal 11 (SIGSEGV) release: Fedora release 12 (Constantine) How to reproduce ----- 1. Use SubVersion Thunar Plug-in 2. Checkout new working copy 3. Right on directory -> SVN -> Log 4. Click on the first entry (and the only one for me) of revision list
There's Thunar support in RabbitVCS. http://blog.rabbitvcs.org/archives/178 Unfortunately, there's some strange dependency to Nautilus, too. It doesn't make much sense to offer support for Thunar if you have to install also Nautilus to get it working. First, we would have to sort this out. https://bugzilla.redhat.com/show_bug.cgi?id=717485
(In reply to comment #1) > There's Thunar support in RabbitVCS. That won't fix this crash. > http://blog.rabbitvcs.org/archives/178 > > Unfortunately, there's some strange dependency to Nautilus, too. It doesn't > make much sense to offer support for Thunar if you have to install also > Nautilus to get it working. First, we would have to sort this out. > > https://bugzilla.redhat.com/show_bug.cgi?id=717485 Again, this is completely unrelated to this bug report. Please only us the downstream bugzilla for tracking this.
Created attachment 3778 abrt.log with thunar-vcs-plugin 0.1.4 This abrt.log contains all info (backtrace, memory map, xsession errors etc) from a crash with Thunar 1.2.2 and thunar-vcs-plugin 0.1.4 on Fedora 15.
Created attachment 3779 backtrace of the crash Sorry, actually the interesting parts were not included in abrt.log
(In reply to comment #2) > (In reply to comment #1) > > There's Thunar support in RabbitVCS. > > That won't fix this crash. Sorry for my badly thought comment. I did not want to rant against Xfce with that or do other spamming. But maybe, could it help to fix the bug by doing things similiar to the other source part? > Again, this is completely unrelated to this bug report. Please only us the > downstream bugzilla for tracking this. Yeah, this should have gone to downstream bugzilla. Sorry again. :(
We now have a new bug report downstream at https://bugzilla.redhat.com/show_bug.cgi?id=722679 and even a patch that - according to the bug reporter - fixes the issue. it can be found at https://bugzilla.xfce.org/show_bug.cgi?id=6404 I will issue an update in Fedora and keep you informed.
New version of the patch: https://bugzilla.redhat.com/attachment.cgi?id=513486
(In reply to comment #7) > New version of the patch: > https://bugzilla.redhat.com/attachment.cgi?id=513486 It looks like that quark function is indeed the problem. With this patch I now understand the back trace. I can't use the static quark since that can't be used in a dynamic library which gets unloaded. Like thunar plugins. I don't see the need on the extra if g_quark_try_string, since the g_quark_string already searches for an existance.
I think there is a mix up in the bugs here. 1. This bug 6404 is about a crash in the tvp-svn-helper. This is what the first backtrace shows. 2. https://bugzilla.xfce.org/show_bug.cgi?id=6322 is about the plugin crashing Thunar. The abrt.log and the backtrace on this 6404 bug report show the 6322 bug. The solution mentioned in the above few comments is from bug 6322. Status update for this bug 6404: For this bug 6404 the original cause of the segfault in tvp-svn-helper is not found. A bit technical detail of the 6404 bug: There is a strcmp which fails (as it seems in a NULL pointer). At a location in the code where there shouldn't be a NULL pointer. The string is retreived from the svn library and it then g_strdup and stored for later use. And now it seems that when it is used it is a NULL pointer. The code from would suggest that either the g_strdup allocation failed or that g_strdup got a NULL pointer supplied. If g_strdup got a NULL pointer as input, that NULL pointer would be the key value is a apr_hash_t. which should be NULL as far as I know. Further investigation on bug 6404 is required.
-- GitLab Migration Automatic Message -- This bug has been migrated to xfce.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.xfce.org/thunar-plugins/thunar-vcs-plugin/-/issues/4. Please create an account or use an existing account on one of our supported OAuth providers. If you want to fork to submit patches and merge requests please continue reading here: https://docs.xfce.org/contribute/dev/git/start#gitlab_forks_and_merge_requests Also feel free to reach out to us on the mailing list https://mail.xfce.org/mailman/listinfo/xfce4-dev