XFCE 4.10

2012-05-31 Thread Andrew Z
Hello,
 just curious if you know the time-frame for 4.10 hitting the SL in
packaged form :) ?

Thank you
AZ


Re: [SCIENTIFIC-LINUX-USERS] XFCE 4.10

2012-05-31 Thread Pat Riehecky

On 05/31/2012 10:38 AM, Andrew Z wrote:

Hello,
 just curious if you know the time-frame for 4.10 hitting the SL in 
packaged form :) ?


Thank you
AZ


With xfce making its way into EPEL, we are letting the packaging happen 
up there.


Pat

--
Pat Riehecky
Scientific Linux Developer


One more SL5.7-5.8 gotcha: autofs and special filesystems

2012-05-31 Thread Chris O'Regan
While this is not an SL-specific problem, I thought I'd document it here
to save others from tearing out their hair.

We have a handful of system that run tomcat in a chroot. They
require /proc and so we mount it inside the chroot using autofs. The
entry in the map looked like this before the upgrade to SL5.8:

/path/to/chroot/proc proc-tomcat60

Since there's no real source file system we use 'proc-tomcat60' here to
provide some meaningful information in the output of commands such as
df.

Well, this broke after the update. I'm not sure why, but we now have to
prepend 'proc-tomcat60' with a colon like this:

/path/to/chroot/proc :proc-tomcat60

Seems we were bitten by a number of minor text changes this round of
patching! What a different one character can make. Oy! :-)


-- 
Chris O'Regan ch...@encs.concordia.ca
Senior Unix Systems Administrator, Academic IT Services
Faculty of Engineering and Computer Science
Concordia University, Montreal, Canada


Re: XFCE 4.10

2012-05-31 Thread zxq9

On 06/01/2012 03:16 AM, Glenn Morris wrote:

Andrew Z wrote:


  just curious if you know the time-frame for 4.10 hitting the SL in
packaged form :) ?


Probably no time soon, since it apparently requires newer versions of
Gtk+, Glib etc than exist in SL 6.

http://docs.xfce.org/xfce/building
   Xfce 4.10 requires Gtk+ 2.20 and Glib 2.24

SL6 has 2.18 and 2.22.


This sort of thing is why you exchange cutting edge for stability. 
Its the same issue with moving KDE ahead to  4.3: a lot of the core 
distro would have to go along with it.


The dependency breakages (or, at a minimum, package rebuilds) required 
are a hard enough maintenance problem that its better to just spin off a 
new distro than to deal with rifted library problems. Distros get hard 
to maintain pretty fast, which is why most distros don't make it very far.


Since both XFCE and KDE aren't part of SL or TUV's distro but come from 
EPEL (and probably other repos) its impossible to expect the SL team to 
decide to fork the entire distro for the sake of one window manager, as 
this is completely against their goals.