KDE 3.1 is in woody has ne1 tried it?

[Christian.Noack]

I managed to install KDE 3.1 with the help of what ironstorm wrote. But after all I must say, that was a pain. I want to be able to compile KDE applications like "kile" and therefore I need kdelibs4-dev. When I try to install them I encounter scaring package dependencies so that my dselect was trying to delete the whole KDE stuff again. I also found out, that kpackage does not work anymore with kde 3.1 The error was: "konsole_grantpty: cannot revoke /dev/ttyp0. Reason: Function not implemented" although konsole_grantpty is installed with setuid. So my personal efforts in installing KDE 3.1 weren't very sucessfull. I have a running KDE 3.1 system by now, but it is not complete and I do not know how to repair it. Any hints? Has anybody got a complete KDE 3.1 whith all development libs included?

Best regards,

Christian

[MichielioZ]

I was wondering :
Maybe it would be easier to create a KDE3.1 version with the, for instance, Modularity-K distro (or Morphix base) ?
Or are there still a lot of dependecies in such a small basic distro ?
Then again : Knoppix Lite is easier to use, as it's allready customized a bit.
How should I go about making a KDE3.1 version out of this ?
(Asked more then once, I know, but there still (to me) isn't a clear answer...)

[Alextreme]

Morphix has a KDE3.1 test module up (in iso form , ready to burn ) at 275megs (full KDE install with the morphix installer and some extra's, but as far as i can remeber no devel libs included, it's a GUI module). Would really like some feedback, but if you want a stable KDE3.1 i would suggest eadz' KnoppixKDE.

Anyway, for more info check www.morphix.org or the mirror. I dont have any specific info on KDE up yet, do have a page on the wiki there reserved for feedback purposes

And no, i won't hack up Morphix base for a single distro, that would be exactly the opposite of what morphix is for

[GREGC]

I tried the morphix kde 3.1 hd-install...not very stable at all. The first attempt didn't recognize the reiserfs partition and errored out.

Then the setup asks for root and morph passwords...changed them to mine, but at login, states "invalid login"...even for root (the one I picked).

Was only able to login as root, and only by pressing enter with no password. Then I was able to change the password through KDE. Weird, huh?

Then APT is completely foobar in Morphix. There was no apt.conf file at all, there was no "available" file (whatever that is) and that caused errors trying to apt-get anything. Entire directories needed for apt-get were missing. I gave up after checking out the Apt-get How-to's at Debian.org. Their troubleshooting section at least assumes that the directories will at least be there (they're missing in Morphix).

I'll wait for the knoppix 3.2 to arrive.

Greg

[RockMumbles]

I have a morphix kde 3.1 hd install and it works OK, users are fine, etc.

As root you have to do a:

touch /var/lib/dpkg/available

for dpkg (apt-get) to function properly.

HTH

rock

[GREGC]

I'd love to give it another go, but I recently discovered Fluxbox...if I can get the desktop icons set up, and some quicklaunch apps, I'd be set.

Would love to see the mem usage comparison between kde3.1 and fluxbox. Anyone have some insight? With 3.0, I only had 3megs free mem (out of 128) at boot.

GregC

[RockMumbles]

Free memory is not really a consideration with linux, as linux will use all of the available ram it can especially when running an X session.

More of a consideration is system overhead (processor and memory requirements) of the running processes. Since fluxbox is way smaller and much simpler it is faster than kde 3.0 (later versions of kde ie: 3.1 and 3.1.1 are both slightly smaller and slightly faster than previous versions of kde).

here is the top output of one of my systems that has been running almost 38 days:

76 processes: 74 sleeping, 1 running, 1 zombie, 0 stopped
CPU states: 4.9% user, 4.1% system, 0.0% nice, 91.0% idle
Mem: 256280K total, 208468K used, 47812K free, 24104K buffers
Swap: 136512K total, 20448K used, 116064K free, 90432K cached
--with openoffice, mozilla the two big user apps running.


here is another top output with openoffice no longer running:

70 processes: 68 sleeping, 1 running, 1 zombie, 0 stopped
CPU states: 1.2% user, 0.8% system, 0.0% nice, 98.0% idle
Mem: 256280K total, 201472K used, 54808K free, 24524K buffers
Swap: 136512K total, 20448K used, 116064K free, 91652K cached

Note I have 7MB more free memory (54808K - 47812K) also note openoffice takes 17.4MB to run with the document I had opened (15.2 running OO.org writer without a document opened)

Also note: the two CPU states cannot really be compared as the values fluctuate with system usage, switching windows etc.

rock

[RockMumbles]

Free memory is not really a consideration with linux, as linux will use all of the available ram it can especially when running an X session.

More of a consideration is system overhead (processor and memory requirements) of the running processes. Since fluxbox is way smaller and much simpler it is faster than kde 3.0 (later versions of kde ie: 3.1 and 3.1.1 are both slightly smaller and slightly faster than previous versions of kde).

here is the top output of one of my systems that has been running almost 38 days:

76 processes: 74 sleeping, 1 running, 1 zombie, 0 stopped
CPU states: 4.9% user, 4.1% system, 0.0% nice, 91.0% idle
Mem: 256280K total, 208468K used, 47812K free, 24104K buffers
Swap: 136512K total, 20448K used, 116064K free, 90432K cached
--with openoffice, mozilla the two big user apps running.


here is another top output with openoffice no longer running:

70 processes: 68 sleeping, 1 running, 1 zombie, 0 stopped
CPU states: 1.2% user, 0.8% system, 0.0% nice, 98.0% idle
Mem: 256280K total, 201472K used, 54808K free, 24524K buffers
Swap: 136512K total, 20448K used, 116064K free, 91652K cached

Note I have 7MB more free memory (54808K - 47812K) also note openoffice takes 17.4MB to run with the document I had opened (15.2 running OO.org writer without a document opened)

Also note: the two CPU states cannot really be compared as the values fluctuate with system usage, switching windows etc.

rock