April 2003 - openSUSE Users - openSUSE Mailing Lists

RE: Upgrade hard-disk using autoyast?
by Richard Parvass 01 Apr '03

01 Apr '03

Whoops. So sorry - didn't see the autoinstall mailing list. I'll try my luck there. -----Original Message----- From: Richard Parvass Sent: Tue 2003-04-01 18:39 To: suse-linux-e(a)suse.com Cc: Subject: Upgrade hard-disk using autoyast? Hi there I thought this would be an easy process, but I'm finding out it's not. I made a clone of my workstation using the autoinstall option of yast, so I obtained a clone.xml, and copied the tar off somewhere else. I had a single partition. I then replaced the drive for a larger one, set out the partitions as I wanted, performed a minimum install, used autoinstall to get a clone2.xml again, then pasted the partition information from clone2 into clone.xml. I put this on a floppy as autoinstall.xml and started the installation process all over again. Now I thought it would use the partition information I gave it and install accordingly, but I was wrong. It took the first partition of 100M which was supposed to be for /boot, and used it as /. I suspect it didn't try to do any partitioning and simply used what was on the disk already, but ignored my custom mount points. hda2 was used as swap (as it should have been), and it then proceded to install, only to fail half way through because there wasn't sufficient space on / !!! Where am I going wrong?

1 0

Upgrade hard-disk using autoyast?
by Richard Parvass 01 Apr '03

01 Apr '03

Hi there I thought this would be an easy process, but I'm finding out it's not. I made a clone of my workstation using the autoinstall option of yast, so I obtained a clone.xml, and copied the tar off somewhere else. I had a single partition. I then replaced the drive for a larger one, set out the partitions as I wanted, performed a minimum install, used autoinstall to get a clone2.xml again, then pasted the partition information from clone2 into clone.xml. I put this on a floppy as autoinstall.xml and started the installation process all over again. Now I thought it would use the partition information I gave it and install accordingly, but I was wrong. It took the first partition of 100M which was supposed to be for /boot, and used it as /. I suspect it didn't try to do any partitioning and simply used what was on the disk already, but ignored my custom mount points. hda2 was used as swap (as it should have been), and it then proceded to install, only to fail half way through because there wasn't sufficient space on / !!! Where am I going wrong?

1 0

Sun Ditches Private-Label Linux(OT)
by Fred A. Miller 01 Apr '03

01 Apr '03

Sun Ditches Private-Label Linux It's planning instead to support as many as four Linux distributors. informationweek.com/story/IWK20030331S0009 -- Fred A. Miller Systems Administrator Cornell Univ. Press Services fm(a)cupserv.org, www.cupserv.org

1 0

KInternet question
by Linux Mail 01 Apr '03

01 Apr '03

Hi, I am using kinternet for dialup. I changed my ISP and so configured a new connection. How can I delete the older ISP from the list in Kinternet? I also get the name of older ISP in the modem configuration (in YAST2) when I check the custom ISP radio button. regards Leo

1 0

How to have 27 lines?
by Blue Zeka 01 Apr '03

01 Apr '03

Hi, Linux terminal mode by default has 25 line. How can I change that? I'm using Linux 7.3 with one program only in text mode. Thanks

1 0

re: [SLE] Re: working with source rpms [WAS [Is there a howto for building a source rpm?]]
by Greg Freemyer 01 Apr '03

01 Apr '03

Anders, Many thanks, I will be trying this out this afternoon. Greg >> On Tuesday 01 April 2003 01:16, you wrote: >> > >> If you want to make any changes to the rpm, the standard procedure >> > >> would be to >> > >> install the src.rpm with rpm -ivh, edit the spec file in >> > >> /usr/src/packages/SPECS and finally build the new rpm with "rpm -bb >> > >> <spec file>" or "rpm -ba <spec file>" if you want a new src.rpm built >> > >> with your new >> > >> edited spec file included. >> > >> > Anders, >> > >> > Thanks, that does seem to work but is not quite what I was looking for. >> > >> > I guess there is not a HowTo that describes using the *.src.rpm and >> > associated spec files. :( >> > >> > I still have a few questions, maybe you can answer them: >> > >> > (for all of the below, I am working with shadow-4.0.2-47.src.rpm) >> > ---- >> > 1) The above ("rpm -ivh; rpm -ba") is apparently missing some dependency >> > checking. >> Yes, there is none. You have to manually make sure the required >> dependencies >> are installed. >> > >> > At least on my machine rpm -ba is dying because I don't have ldap.h >> > installed. >> > >> > The dependency on openldap2-devel is in the specfile under both "# >> > neededforbuild" and "# usedforbuild". >> > >> > How do I get rpm to test for it? >> You don't. Those are tags used by SuSE's internal build system. You can >> use >> them as hints for what you need to install. >> > >> > 2) What is the difference between "# neededforbuild" and "# >> usedforbuild" >> > in the specfile. >> neededforbuild are the direct dependencies that the package uses to >> compile. >> As I understand it, the SuSE build system works in a chroot()ed >> environment. >> Every time a package >> is built, a clean environment is created. usedforbuild describes that >> chrooted environment. As you can see, it includes all the basic packages >> that >> are needed for the compile all the way from aaa_base to the top level >> dependencies. >> > >> > 3) If I want to create a patch file of my own, do I just create a >> > shadow-4.0.2-greg.dif file and reference it in the specfile? >> Yes >> > >> > If so, do I create the diff file with a diff -c like a normal patch? >> That is up to you. In the spec file you apply the patch manually, so you >> can >> give whatever parameters to patch you like. Look at the %patch statements >> in >> the %prep section. >> Incidentally, the standard usage is to use unified patches (-u). But as I >> say >> you can use whatever you prefer. >> > >> > 4) While developing my patches, I want to modify the actual source, >> > compile, and install. >> > >> > The below seems to work, and it installs to /var/tmp/shadow-root, which >> is >> > nice for initial testing. === >> > # rpm -ivh shadow-4.0.2-47.src.rpm (.src.rpm from the CDs) >> > # cd /usr/src/packages/SPECS >> > # rpm -bp shadow.spec # Apply SuSE specific patches >> > # rpm -bc --short-circuit shadow.spec # Run ./configure as SuSE does >> > # cd /usr/src/packages/BUILD >> > # make source code changes >> > # Invoke local makes as necessary >> > # cd /usr/src/packages/SPECS >> > # rpm -bi --short-circuit shadow.spec # Install >> > === >> > For final testing, how do I get rpm to install to /? >> You could change the build root in the spec file. But I always build the >> binary rpm and install it as per normal when I work with them. When I make >> >> changes in rpms, here's my procedure >> rpm -ivh >> rpm -bp >> edit source files and create diffs, edit the spec file to use those diffs >> rpm -bb >> rpm -Uvh on the binary rpm >> I'm sure there are other ways of doing it, with the "interrupted style" >> build >> you describe above, but this is the way I do it and it seems to work. I >> haven't really looked at your method, so I can't discuss it. >> regards >> Anders >> -- >> Check the headers for your unsubscription address >> For additional commands send e-mail to suse-linux-e-help(a)suse.com >> Also check the archives at http://lists.suse.com >> Please read the FAQs: suse-linux-e-faq(a)suse.com

1 0

problems installing qt_3.1.1
by Fabio De Francesco 01 Apr '03

01 Apr '03

I am trying to install qt_3.1.1 on SuSE 8.1 from sources. I put all the stuff in /usr/local/qt and then: ./configure -thread, gmake, gmake install. It doesn't work and I obtain: g++ -shared -Wl,-soname,libqui.so.1 -Wl,-rpath,/usr/local/qt/lib -o libqui.so.1.0.0 .obj/release-shared-mt/qwidgetfactory.o .obj/release-shared-mt/domtool.o .obj/release-shared-mt/uib.o .obj/release-shared-mt/database.o .obj/release-shared-mt/moc_database2.o -Wl,-rpath,/usr/local/qt/lib -L/usr/local/qt/lib -L/usr/X11R6/lib -L/usr/X11R6/lib -lqt-mt -lGLU -lGL -lXmu -lXrender -lXinerama -lXft -lfreetype -lXext -lX11 -lm -lSM -lICE -ldl -lpthread /usr/lib/gcc-lib/i486-suse-linux/3.2/../../../../i486-suse-linux/bin/ld: cannot find -lqt-mt collect2: ld returned 1 exit status gmake[4]: *** [../../../lib/libqui.so.1.0.0] Error 1 gmake[4]: Leaving directory `/usr/local/qt/tools/designer/uilib' gmake[3]: *** [sub-uilib] Error 2 gmake[3]: Leaving directory `/usr/local/qt/tools/designer' gmake[2]: *** [sub-designer] Error 2 gmake[2]: Leaving directory `/usr/local/qt/tools' gmake[1]: *** [sub-tools] Error 2 gmake[1]: Leaving directory `/usr/local/qt' gmake: *** [install] Error 2 What can I do to solve the problem? Thank you, Fabio M De Francesco

1 0

Problem with HP Officejet USB PTAL CUPS and runlevel editor
by Roger James 01 Apr '03

01 Apr '03

Help, Has anyone got any ideas on this one. I have a USB connected HP Officejet printer. To talk to it I need to the HP specific ptal (hpoj) package. The latest version of this package integrates with the runlevel editor system (as it should do). However with the default settings ptal-init is run after hotplug has started but before hwscan has run. This usually means that ptal-init cannot find a connected printer and often causes a panic in printer.o. To solve this I added hwscan as a Required-Start: dependency in the ptal init.d script. This all works OK ptal now runs after hwscan, the printer is always found and the kernel doesn't panic. However this now means that ptal-init is being run after cupsd in the startup order. For some reason this means that cupsd fails to start. Ok. I thought this would be easy to fix. So I added a ptal dependency to the cups init.d script. However, whenever I do this the runlevel editor silently (no error messages) fails to save the updated configuration and sets the cups service to disabled. If I take "ptal" off the end of the Required-Start: line everything is back to where it was. Does anyone know what is happening here. If there another way to fix this. Help!!! Roger

1 0

Printer full-duplex issue using SuSE 8.1
by Nicholas Parsons 01 Apr '03

01 Apr '03

Hi Everyone, I have a network printer here at work which I use quite frequently. It is a hp color lazer jet 4550dn. Recently we upgraded the printer so that it supports full-duplex printing (print to both sides of a page). I updated the settings on some of my colleagues windows 2000K boxes without a problem to support this feature. I dual boot SuSE 8.1 and windows 2K on my machine here. I looked around YaST in the printer configuration and could find no settings where I could enable this full-duplex mode. Does anyone know how to enable this option? Thanks in advance! ~~Nick _______________________________________________ Join Excite! - http://www.excite.com The most personalized portal on the Web!

1 0

Is anybody else using KDevelop 2.5.1, SuSE 8.1 & KDE 3.1.[12]?
by Steven T. Hatton 01 Apr '03

01 Apr '03

I've been having problems building with KDevelop's framework using KDE 3.1.1 and KDE 3.1.2 from SuSE. My cvs KDE builds seem to work most of the time, but somethimes they don't provide the proper support for KDevelop. The problem I'm having is that ./configure pukes on: ../kdevelop/configure ... ... checking for KDE... libraries /opt/kde3/lib, headers /opt/kde3/include checking if UIC has KDE plugins available... configure: error: not found - you need to install kdelibs first. Anybody else seen this? STH

1 0