Description of problem: When pasting information into nedit, it crahses with the following error message : [gert@apollo tmp]$ nedit nedit: the current locale is utf8 (en_US.UTF-8) nedit: changed locale to non-utf8 (en_US) nedit: xcb_xlib.c:41: xcb_xlib_lock: Assertion `!c->xlib.lock' failed. Aborted [gert@apollo tmp]$ Version-Release number of selected component (if applicable): nedit.x86_64 5.5-12.fc8 installed How reproducible: always Steps to Reproduce: 1. start nedit 2. copy some data to the clipboard with crtl-c (e.g. a URL in firefox) 3. paste the data to nedit with ctrl-v Actual results: nedit crashes with error message "xcb_xlib_lock: Assertion `!c->xlib.lock' failed": [gert@apollo tmp]$ nedit nedit: the current locale is utf8 (en_US.UTF-8) nedit: changed locale to non-utf8 (en_US) nedit: xcb_xlib.c:41: xcb_xlib_lock: Assertion `!c->xlib.lock' failed. Aborted [gert@apollo tmp]$ Expected results: The data from the clipboard bying pasted in the nedit editor window. Additional info:
I get the same error message when I try to run the BOINC manager (http://boinc.berkeley.edu/). Others get the same error in BOINC, and as I read the support case, it's an error in the libxcb package (http://boinc.berkeley.edu/trac/ticket/432). A workaround for openSUSE is provided (http://en.opensuse.org/Xlib.lock), but it doesn't work for me. My system is x86 (Celeron) based.
I'll add a METOO for the Citrix client (hey, it's better than running Windows :). I also had the Java problem as well, but the s/XINERAMA/FAKEEXT/ trick from Bug 254144 seems to have mostly resolved the issue. It appears that the XCB stuff [1] in libX11 causes some software to barf? I see several other bugs in RH Bugzilla against f8 about this same symptom. Perhaps we should have a tracking bug for this issue and mark these as dupes of that? Also, if we could have a temporary workaround similar to SuSe's hack until more packages can converge on fixing their bugs -- that'd be super handy. 1 - http://lists.debian.org/debian-devel-announce/2006/11/msg00010.html
Yes, you are right. It seems like libX11 needs to be fixed first and bug 301691 looks like a good tracker for these issues.
*** This bug has been marked as a duplicate of 301691 ***
No, that's not the bug we were talking about. It should be reopened and bug 390621 is just a tracking bug.
Created attachment 279491 [details] gdb backtrace of nedit assertion failure I think this bug should be reassigned to lesstif since you can get the same assertion failure when pasting into ddd. I've attached the gdb backtrace of the first assertion failure with nedit. NB - The work around does work for me, and there are several more assertion failures after the first if nedit isn't aborted. My hardware is x86 (32 bit).
The only thing that should be enough to fix this is to update libxcb in F8 to 1.1. Tried it out a few days ago and it fixed the assertion problems for me (x86_64).
Bug 301691 reported that libxcb-1.0-4.fc8 was available with a fix to the Java flavor of this problem. I just tested libxcb-1.0-4.fc8 and it fixes the nedit paste crash, also.
Indeed, the update to libxcb fixes the paste bug. [gert@apollo tmp]$ yum list | grep libxcb libxcb.x86_64 1.0-4.fc8 installed libxcb-devel.x86_64 1.0-4.fc8 installed libxcb.i386 1.0-4.fc8 updates libxcb-devel.i386 1.0-4.fc8 updates libxcb-doc.x86_64 1.0-4.fc8 updates [gert@apollo tmp]$
Ok, closing ERRATA then.