On 02/04/2014 01:06 PM, Akemi Yagi wrote:
On Tue, Feb 4, 2014 at 6:08 AM, Pat Riehecky <riehe...@fnal.gov> wrote:
On 02/04/2014 07:24 AM, Elias Persson wrote:
On 2014-02-04 10:40, Matthieu Guionnet wrote:
Le 04/02/2014 04:52, John H. Outlan a écrit :
Several of us over at the forum are having a nautilus problem.
Nautilus is respawning non stop and filling the window list on bottom
panel.

Some with update, some with clean 6.5 x64 install.

Here's the forum link:

http://scientificlinuxforum.org/index.php?act=ST&f=5&t=2634&st=0#entry17238

Hi,
I confirm the reason and the (temporary)  solution.

It's due to the last librsvg2 update
After a yum downgrade librsvg2, Nautilus closes all the listed windows
on bottom panel

Matthieu.
Seemingly caused by: https://bugzilla.redhat.com/show_bug.cgi?id=924414
This issue: https://bugzilla.redhat.com/show_bug.cgi?id=1061085
Thanks to everyone for the detailed information and the bug link!

Pat
Fix coming: https://bugzilla.redhat.com/show_bug.cgi?id=924414#c25

Akemi
One small note. I can confirm the defect -- I had just upgraded a colleague's IA-32 laptop workstation, and the entire desktop lost all icons (the user lives by icons and cannot effectively use a scrolling terminal without written key stroke instructions for each command, including return/enter keystrokes). However, I had installed both librsvg2 and the -devel version, both of which automatically upgraded to the defective release. I had to manually rpm -e of the -devel version before yum would allow the downgrade. After the downgrade, the window manager interface seems to be working. I did attempt to use KDE instead of gnome (this being a login choice from a button on the login GUI screen), and KDE seemed to work for non-gnome applications, but I did not extensively test the proper functioning of KDE. As this "upgrade" automatically was installed on many of our machines, we shall need to manually downgrade a number of machines. Any ETA on the fixed version that should auto-update?

A separate question. Unlike Microsoft, TUV EL is supposed to be a production enterprise system. As this is a major failure of what amounts to a default GUI of EL, does anyone have an idea as to how this defect passed qualification testing prior to production release? I am not expecting SL to test the functionality of TUV production release updates, but rather the correctness of the port from TUV to SL -- this defect clearly is a TUV issue.

Yasha Karant

Reply via email to