Re: [expert] USB Flashcard Reader problem under 9.0
On Wednesday 25 December 2002 18:51, Jason wrote: > Hi All, > > I have a vanilla 9.0 install at work. I plugged an external USB Sony > memory stick reader in and poof, mandrake found it and I could mount it > with Kwikdisk and view the contents via a link ML made on my desktop (I > was duly impressed). Now my problem. I just bought a 6 in 1 (reads > memory stick, CF, SM etc.) external USB card reader for use at home. I > run cooker at home but I have just done a vanilla 9.0 install on a new > HDD. The reader is a Packard Bell and the literature says it is > supported on Linux running 2.4 kernels. I tried plugging this reader in > at home and on my work laptop and...nothing. ML didn't notice it was > plugged in and I cannot get it to work for the life of me. Anyone have > any ideas?? Did you look into the logs? Were there messages from hotplug? Did you look at /dev/sd* ? Was it possibly the case that no such device /dev/sd* existed before you plugged your new reader into the USB port and one or more did appear a couple of seconds after you plugged it in? Does "usbview" show you any device on the USB bus? I know that I can't solve your problem, but maybe these things lead you in the right direction. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] can't compile
On Monday 23 December 2002 05:48, Dan Cox wrote: > I'm trying to compile chkrootkit and am getting the fallowing. > > [root@diablo chkrootkit-0.38]# make sense > gcc -static -o strings strings.c > /usr/bin/ld: cannot find -lc > collect2: ld returned 1 exit status > make: *** [strings] Error 1 > > I don't see any reference to ld or -lc in the make file. Any > suggestions? > > Dan Cox You're trying to link statically to the C-library and the linker fails because it can't find libc.a to link against. "urpmf libc.a" gives a few incorrect hits, but it also shows us the correct one: glibc-static-devel so "urpmi glibc-static-devel" should solve your problem: Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] image editor for postscript?
Quoting Praedor Atrebates <[EMAIL PROTECTED]>: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > I am having a major problem trying to edit a postscript image. I have a > journal article, pdf format, that I've converted to postscript. There is a > > figure in this paper that I need. I cannot extract it with tools like > pdfimages (don't know why this particular figure is immune but it will not > come out with that tool). The only way I can get it is to open up the > postscript page that the figure is on and crop it down to just the figure. > > I have tried and tried to use Gimp but quite honestly, the output/results > from > the Gimp or horrible and unusable. The image, on the page in postscript > form, produced with pdf2ps, is beautiful. The lines are smooth and the text > > is sharp and readable. Run it through the Gimp, however, and the figure is > > rendered into trash. The text becomes stairstepped with gaps, and finer > lines vanish. I don't understand why pdf2ps can create such a beautiful > postscript image of a page (text AND figures) but the Gimp butchers it so > badly. I NEED this figure. Is there another tool available to linux that > can edit postscript images? All I need to do is crop the ps image/file to be > > just the figure. I don't know of any tool that lets you edit postscript files. However, if you only need the figure, and you don't mind going into a pixel based format, why not use gs? gs -sDEVICE=ppmraw -r200 -q -dNOPAUSE -sOutputFile=pic.ppm -- pic.ps converts pic.ps into a ppm file. Change -r200 if you need a higher resolution. If you have a multipage postscript file, you can use pic%d.ppm as the template for the output filenames. Then there is also the 'convert' command from the ImageMagick package, but I didn't see at a glance how to control the resolution of the output images. I hope this helps, Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] mail client
On Thursday 19 December 2002 05:00, Praedor Atrebates wrote: > On Thursday 19 December 2002 02:31 am, Guy Van Sanden wrote: > > That is very strange... > > Although I use Evolution, both my wife and my mother are running Kmail, > > my wife has to see a crash yet. > > My mother did have prolbems with the kmail package build by SuSE (it was > > the KDE 3.0.4 upgrade). The RPM's from the KDE site worked fine > > however. > > Yes. I have tried other mail clients and have always gone back to Kmail. > It has never ever crashed on me or given me any problems. I have also used > pine, evolution (yuck...and no way to order mail by date?), mozilla mail, > and sylpheed (pretty good). > > Something is not right with your system or rpm build, not the software > (Kmail) per se. > > praedor Didn't the original poster say that he had a large number of IMAP accounts? I have only 2 imap accounts and I find it trivial to crash kmail while it is communicating with the imap server. If I make sure to let one imap operations finish before starting the next, it is rather stable, but if I accidentally give 2 imap operations in a row, it is very likely that kmail crashes. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Western digital drives don't work?/maximum capacity
I know there are very likely valid technical reasons against WD drives in linux, everything that I have heard about WD drives bothers me deeply. On the other hand, if my Maxtor drive crashes tomorrow, that does not mean that Maxtor is a bad brand, absolutely not. We cannot infer anything about the reliability of the manufacturers from individual drives. Now, if we have large scale reliability data, then we can start talking ... [Actually, this list as a collective may be able to provide such a large scale dataset]. But this is why I encourage everybody to participate in the StorageReview drive reliability survey at http://www.storagereview.com/ Please stop by there and enter information your present and past drives. There is only one trick in this game: You can't view the results of the survey without first registering at StorageReview and then entering information on at least one hard drive. If you decide to register and enter data there, please don't bias the survey by only entering information about drives that have crashed on you or been DOA, enter instead information about all your drives. Right now, the readers of StorageReview have entered data about a total of 9059 drives. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] video card recommendations
Quoting Charles A Edwards <[EMAIL PROTECTED]>: > On Mon, 16 Dec 2002 16:59:24 -0600 > Narfi Stefansson <[EMAIL PROTECTED]> wrote: > > > Leadtek WinFast A170 TH GeForce 4 MX 420 64MB AGP TV-Out > > Asus V7100 Pro Pure MX400 64MB AGP > > Gainward GeForce2 MX400 64MB AGP > > Toss-up, they all use the same chip. > None are fully supported unless you use the closed-source nvidia drivers > but any should give good 2d performance using the nv driver > > > > ATI / Powercolor Radeon 9000 64MB DDR 250MHz TV/DVI > > Not supported so you would only be able to use the vega driver so it is > at this time not worth any extra cost. > > > ATI Radeon 7500 Dual Head 64MB AGP > > Pass, flaky behavior can vary between systems. > Some have no problem others nothing but. > > Would like to add another suggestion that would be in the lower price > range of the cards you listed. > ATI Xpert 2000 32mb using Rage 128 chipset. > Running it on 2 system with full 3d accel and no problems. > There is a problem (freezing when switching back and forth between X and > the virtual consoles) with the Rage 128 Mobility cards but not with > those used in standard desktops. Actually, you may have found the winner ... The ATI Xpert 2000 is not compatible with my motherboard since I need an AGP 2.0 card. However, the ATI Xpert 2000 Pro is an AGP 2.0 card, so it runs at 2x/4x and it is also dirt cheap! I have used the XPert 2000 for a couple of years now and it's been very stable for a while (since approx. 8.2) and unless somebody on this list knows the Pro card to be less stable than the regular Xpert 2000, I'll go for the pro :-) Thanks for the good advice, Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] video card recommendations
Hi everybody. I know this is a frequently asked question on this list, but I have a list of candidates and I would like your recommendations for the best cards, using only free drivers. I have the following cards in mind: Leadtek WinFast A170 TH GeForce 4 MX 420 64MB AGP TV-Out Asus V7100 Pro Pure MX400 64MB AGP Gainward GeForce2 MX400 64MB AGP ATI / Powercolor Radeon 9000 64MB DDR 250MHz TV/DVI ATI Radeon 7500 Dual Head 64MB AGP The radeon 9000 is a budget version, but still more expensive than the others, so that's a minus. My needs are simple, I'm only looking for a card that's good enough and supports AGP 4x, uses free drivers and ... Above all, it musn't ever freeze, including when switching between X and the virtual consoles. I don't do any gaming, so that's really not an issue. It's compatibility with my motherboard, smoothness in 2-D use and stability that count. Thanks, Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] linux on nvidia nforce
Quoting Todd Lyons <[EMAIL PROTECTED]>: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Mark Williamson wrote on Fri, Dec 06, 2002 at 09:08:19PM +1100 : > > > > > On the installation of the Mandrake Linux just install as like any other > > PC, till the part were LILO is to be configured, and make sure that the > > "mem=" gets added to the append line in the lilo config that is > > "mem=(The amount of RAM less how much the Video card uses)" or the > > That's a nice easy workaround. The way I got it to work (with a > customer, I wasn't actually at the machine) was by installing cooker > kernel 2.4.20-1mdk,but that's not a recommended method for just anybody. > Todd and Mark: can you please share with me why you did this? I had a brief opportunity to play with a board with the nforce2 chipset, namely the Asus a7n8x deluxe motherboard, and it booted the 9.0 kernel without any problems. Was the difficulty in getting the 9.0 installed properly or in the bootup with an installed kernel? What else did these 2 workarounds give you? [I.e. the mem= or going to a cooker kernel]. Can people please share what they have working on motherboards with this chipset? Has anyone got usb 2.0 working with this chipset? I presume that would require one to go to a cooker kernel. What about firewire? Or the LAN? I got the NVidia LAN working by using the proprietary driver, but has anyone tried to get the 3Com LAN working or managed not to use the proprietary driver for the NVidia LAN? Does anyone have the sound working without the properietary driver? I had a kernel freeze when I modprobe'd the i810_audio driver. Best, Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] /var/log - apologies, a long one
On Tuesday 10 December 2002 11:39, john wrote: > I tried this on newbie a while ago and got no joy, my /var/log/messages > and /var/log/syslog fill up and I have to delete the entries by hand (if > I try logrotate -f /var/log/messages I get a terminal full of "error: > /etc/logrotate.d contains cron, linuxconf,msec,rpm,sudo,syslog,urpmi,xdm > - text files > It is no surprise that "logrotate -f /var/log/messages" results in a lot of errors since you're telling logrotate to use /var/log/messages as the control file, which is of course very different from telling logrotate to rotate /var/log/messages. The command that is in my /etc/cron.daily/logrotate file is: /usr/sbin/logrotate /etc/logrotate.conf and you may want to add the -f flag there to force the log rotation, so run "/usr/sbin/logrotate -f /etc/logrotate.conf" If this does not rotate your /var/log/messages, have a look at the file that controls that, namely /etc/logrotate.d/syslog. [I presume that you saw the include line in your /etc/logrotate.conf file, it included all the files in the /etc/logrotate.d directory] Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] linux on nvidia nforce
On Thursday 05 December 2002 17:18, Todd Lyons wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Mark Chou wrote on Thu, Dec 05, 2002 at 02:46:15PM -0800 : > > I've been a Mandrake user for quite some time now, and I'm considering > > getting a nVidia nForce(2) motherboard. There were rumors that > > installing Linux (not necessarily MDK) on nForce1/nForce2 systems was > > rather painful, largely due to nVidia's proprietary drivers. (I'm > > well aware of the oxymoronic nature of pairing Linux w/ proprietary > > driver/HW, but in this case the benefits may outweigh the negatives > > for me.) > > I'd like to get some comments from folks who went down this route, and > > what their experiences were. What are some of the gotchas? > > I've got a customer right now who can get neither RedHat nor Mandrake > running on it. Be aware that you'll be up for a challenge. > I received one of those earlier today and will be putting it into my case pretty soon ... please share any solutions that you may find to problems that may arise. I don't care if I can't use any of the peripherals until Mandrake 9.1 comes out, I have a sound card, controller card and a usb 2.0 card. However, if it doesn't even boot, ... then any advice will indeed be gratefully accepted .-) I guess I'll find out this weekend how it goes :-) Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Help needed in copying files from DVD/CD to Harddisk
On Monday 02 December 2002 18:12, Ralph De Witt wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > On Monday 02 December 2002 04:03 pm, FRLinux wrote: > > > When I try to copy files from the CD to the hard drive the system is > > > loosing files and directories from the cd and a refresh of the cd > > > does not show them. > > > > Funily (or not) enough, i got the same problem yesterday after having > > installed my Mandrake 9.0 on a VIA system (don't think it's related > > but who knows). I fixed it by installing a vanilla kernel. I lost > > supermount (which i never use anyway) but all is happy now. > > Hello: > Can anyone else comment on this. Would turning off Supermount in MCC > under Mount points fix this problem. Or do I need to do more? > - -- > I used the following to disable supermount: cp /etc/fstab /etc/fstab.with_supermount supermount -i disable This will give you a new /etc/fstab file and supermount will be completely disabled after the next reboot. Your problems sound typical of the supermount problems and they should go away once you disable it. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] problems with msec daily
On Sunday 01 December 2002 16:10, Angus Beath wrote: > Hi, I've got this error from Cron each day. I'm not sure how to go about > fixing it: > > Subject: Cron nice -n 19 run-parts /etc/cron.daily > Date: Sat, 30 Nov 2002 18:04:19 -0500 (EST) > > bzcat: Can't open input file ./newaliases.1.bz2: No such file or > directory. > bzcat: Can't open input file ./mailq.1.bz2: No such file or directory. > bzcat: Can't open input file ./ctags.1.bz2: No such file or directory. > bzcat: Can't open input file ./aliases.5.bz2: No such file or directory. > > > Any suggestions? > > Angus This error message comes from makewhatis. Most likely the symbolic link /usr/share/man/man1/newaliases.1.bz2 points to a location in your file system that does not exist. The following commands fixed this for me: update-alternatives --auto mta update-alternatives --auto ctags What is going on: In order to allow you to have multiple versions of the same program installed, symbolic links to the man pages and the executables are maintained. Take an example from my system: /usr/bin/gcc is a symbolic link to /etc/alternatives/gcc /etc/alternatives/gcc is a symbolic link to /usr/bin/colorgcc-3.2 Then, if I install another version of gcc in /usr/bin, say /usr/bin/gcc-3.1, and I want to access it as /usr/bin/gcc, I would change the link in /etc/alternatives. I.e. I would let /etc/alternatives/gcc point to /usr/bin/gcc-3.1. Regular users (like you and I) never worry about these symbolic links because they are handled automatically when we upgrade and downgrade our rpms. In this case, something went wrong with the symbolic link for these 2 packages and we had to re-run the script that handles these links. If you are curious, you can look at the files in the directory /var/lib/rpm/alternatives as well as man update-alternatives. I hope this helps, Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] How can i copy 8G to an NT machine and keep.....
On Thursday 28 November 2002 13:34, Ken Walker wrote:
> Is there any way to copy 8Gig of files to a nt machine, keeping all
> ownership/permissions/groups.
>
> I've tried taring but it stops at 2G.
>
> many thanks
>
> Mr Smiley
I'll take a stab at this.
1) you can have "tar' send to the standard output and "split" break the
standard input into small (< 2GB) files.
tar cf - my_directory | split -b 1000m
will create files xaa, xab, xac, xad, Each of them will be less than
1000 MB in size.
In order to un-tar them, you have to concatenate them and then un-tar:
cat xa* | tar xfp -
Of course, you have to be able to store the tar files of your 8GB of data
before transfering them if this is to work.
By the way, >>don't<< use this to create backups that fit onto CDs since
you lose your data if one CD is damaged.
2) Create a list of all of the files you want to transfer, then create many
tar files, each of them with just a few of your files.
cd my_directory/..
find . -type f > filelist.txt
numfiles=100 # We will put 100 files into each tar-ball/
# Calculate the total number of files:
lines=`wc -l filelist.txt | sed 's/ *\([0-9]*\) .*/\1/'`
numtarballs=`expr $lines / $numfiles`
for (( i = 0; i <= $numtarballs; i++ ));
do
min=`expr 1 + $numfiles \* $i`;
list=`tail +$min filelist.txt | head -${numfiles}`;
echo i = $i;
tar cf ball$i.tar $list;
done
You can use this solution if you don't have enough space to store all the
tar files on your server since you don't have to run the entire for loop
in one run.
Of course you have to adjust the variable numfiles so that your tar-files
don't become too many and so that none of them becomes larger than 2GB.
Hope this helps,
Narfi.
Want to buy your Pack or Services from MandrakeSoft?
Go to http://www.mandrakestore.com
Re: [expert] memory usage
On Thursday 28 November 2002 13:54, Philip Webb wrote: > is there a command or utility which would show > which processes are using how much of which parts of memory? > 'free' gives some numbers, > but just what is using 'buffers' & 'cached' is unknown > nor does it show what is 'used +/- buffers/cache': > the last number seems to vary unpredictably. > 'gkrellm' is pretty & quite good for CPU; > 'top' & 'ksysguard' give a whole table, but lack explanation. Run "top" (without the quotes) on the command line. Then if you hit shift-M, you will sort the processes by memory usage. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] xinitrc problem
Quoting [EMAIL PROTECTED]: > On 20 Nov 2002, Matthew Van Horn wrote: > > > Posted to newbie, but got no response. > > > > Since I upgraded from 8.1 to 9.0, I am having problems using my old > > (8.1) ~/.xinitrc > > > > If the file even exists, then startx will not work. I don't know what > > the cause is, can someone point me in the right direction? > > > > The contents of my old .xinitrc are: > > #!/bin/bash > > export LANG=en_US > > export LC_CTYPE=ja_JP > > export LC_MESSAGES=en_US > > export XMODIFIERS="@im=kinput2" > > kinput2 -wnn -ccdef /home/mattv/myccdef & > > imwheel > > startkde > In addition to what kwan said, I would like to mention that kinput2 is started automatically in 9.0. I would venture to guess that in your case it may be started twice. I have my locale information in ~/.i18n and that's it -- kinput2 is started automatically and that way I achieve the same as your .xinitrc, except for 2 things: 1) You start imwheel. I've never used it so I can't help you there, I don't even know if you even need it anymore. 2) You let kinput2 read your personal ccdef file. If kinput2 is started automatically by the system you can of course not provide command line parameters to it. However, kinput2 also reads the Xresources, so you would have to give the ccdef information that way. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] motherboard recommendations?
Hi, everybody. I managed to fry my motherboard yesterday (bad memory module) and I'm therefore looking for a new one. And I was hoping you could help me choose one :-) I'm looking for a motherboard for an Athlon, most likely based on the KT333 or the KT400 chipset and I'm very interested in hearing stories of successes/failures with using 9.0 on such boards, esp: MSI KT4 Ultra SR (Mandrake reports it working) MSI KT3 Ultra (Mandrake tested) Gigabyte 7VAXP (Mandrake certified) Soyo KT400 Dragon Ultra (??) I personally find the Gigabyte motherboard very tempting since it's certified to have everything working: USB 2.0, firewire, ...and it costs around $110. If you have any shutdown problems, problems with setting up peripherals, etc. I'd greatly appreciate to hear about it. Thanks, Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] floppy timeout and I/O error
On Tuesday 29 October 2002 10:53, Larry Sword wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > H. Narfi Stefansson wrote: > |I wanted to mount /dev/fd0 on /mnt/floppy and that was completely > |unsuccessful. My /var/log/messages got filled with the messages listed > |below. > |This is with supermount disabled. > |Does anybody know how I can get around this? > | > |Thanks, > | > |Narfi. > Never mind, this turned out to be a hardware problem -- most likely the connectors were not seated properly. Sorry for that, Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] floppy timeout and I/O error
I wanted to mount /dev/fd0 on /mnt/floppy and that was completely unsuccessful. My /var/log/messages got filled with the messages listed below. This is with supermount disabled. Does anybody know how I can get around this? Thanks, Narfi. Below are my /etc/fstab and a short segment of /var/log/messages /etc/fstab: /dev/hda1 / ext3 defaults 1 1 none /dev/pts devpts mode=0620 0 0 /dev/hda7 /home reiserfs notail 1 2 /dev/hdc/mnt/cdrom auto codepage=850,iocharset=iso8859-1,noauto,nosuid,ro,umask=0,user,nodev 0 0 /dev/fd0/mnt/floppy auto iocharset=iso8859-1,codepage=850,sync,unhide,noauto,nosuid,umask=0,user,nodev 0 0 /dev/hda10 /archive xfs defaults 1 2 /dev/hda9 /opt xfs defaults 1 2 none /proc proc defaults 0 0 /dev/hda8 /projects reiserfs notail 1 2 /dev/hda11 /tmp xfs defaults 1 2 /dev/hda6 /usr reiserfs notail 1 2 /dev/hda5 swap swap defaults 0 0 /dev/sda1 /mnt/removable auto user,iocharset=iso8859-1,kudzu,codepage=850,noauto,umask=0,exec 0 0 /dev/sdb1 /mnt/memory_card auto user,iocharset=iso8859-1,kudzu,codepage=850,noauto,umask=0,exec 0 0 Oct 28 23:30:48 localhost kernel: Oct 28 23:30:48 localhost kernel: floppy driver state Oct 28 23:30:48 localhost kernel: --- Oct 28 23:30:48 localhost kernel: now=131363 last interrupt=131063 diff=300 last called handler=e28231a0 Oct 28 23:30:48 localhost kernel: timeout_message=floppy start Oct 28 23:30:48 localhost kernel: last output bytes: Oct 28 23:30:48 localhost kernel: 0 90 131063 Oct 28 23:30:48 localhost kernel: 13 90 131063 Oct 28 23:30:48 localhost kernel: 0 90 131063 Oct 28 23:30:48 localhost kernel: 1a 90 131063 Oct 28 23:30:48 localhost kernel: 0 90 131063 Oct 28 23:30:48 localhost kernel: 3 90 131063 Oct 28 23:30:48 localhost kernel: c1 90 131063 Oct 28 23:30:48 localhost kernel: 10 90 131063 Oct 28 23:30:48 localhost kernel: 7 80 131063 Oct 28 23:30:48 localhost kernel: 0 90 131063 Oct 28 23:30:48 localhost kernel: 8 81 131063 Oct 28 23:30:48 localhost kernel: e6 80 131063 Oct 28 23:30:48 localhost kernel: 0 90 131063 Oct 28 23:30:48 localhost last message repeated 2 times Oct 28 23:30:48 localhost kernel: 1 90 131063 Oct 28 23:30:48 localhost kernel: 2 90 131063 Oct 28 23:30:48 localhost kernel: 12 90 131063 Oct 28 23:30:48 localhost kernel: 1b 90 131063 Oct 28 23:30:48 localhost kernel: ff 90 131063 Oct 28 23:30:48 localhost kernel: last result at 131063 Oct 28 23:30:48 localhost kernel: last redo_fd_request at 131063 Oct 28 23:30:48 localhost kernel: 20 0 Oct 28 23:30:48 localhost kernel: status=50 Oct 28 23:30:48 localhost kernel: fdc_busy=1 Oct 28 23:30:48 localhost kernel: DEVICE_INTR=e2822060 Oct 28 23:30:48 localhost kernel: fd_timer.function=e2821fb0 Oct 28 23:30:48 localhost kernel: cont=e282bb50 Oct 28 23:30:48 localhost kernel: CURRENT=d798e920 Oct 28 23:30:48 localhost kernel: command_status=-1 Oct 28 23:30:48 localhost kernel: Oct 28 23:30:48 localhost kernel: floppy0: floppy timeout called Oct 28 23:30:48 localhost kernel: end_request: I/O error, dev 02:00 (floppy), sector 0 Oct 28 23:30:48 localhost kernel: FAT: unable to read boot sector Oct 28 23:30:48 localhost kernel: FAT: freeing iocharset=iso8859-1 Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] mdk9.0 with japanese user
On Sunday 13 October 2002 06:02 pm, you wrote: > Hi Narfi, > > I created a new acount, but it does not have an option to create it as > japanese as you did. I set it in the KDE control centre, menus where > japanese, but still no input. > > using my .i18n file in my english KDE i can input in japanese with > kinput&wnn fine in mozilla. But not in any KDE applications. > > Have you got working input in KDE? > I'm sorry, I gave you the wrong info for how to set up a new user ... the programs had changed since RC3 ... I created a new account, logged into KDE as that user and ran /usr/sbin/localedrake and selected Japanese there. I logged out and in again and then everything was in Japanese, kinput2 was started automatically and KDE accepted its input. I immediately ran /usr/lib/qt3/bin/qtconfig, went to the Interface tab and changed "XIM input style" to "Over the spot" since I find "On the spot" terribly confusing. I then went to the K-menu->Configuration->KDE->Personalization->Country and Language [of course this was in Japanese] and changed the country into "C", logged out and in again. This time, kinput2 was also started automatically and KDE accepted its input but all the menus and dialogs were in English. Narfi. ps. I've had emacs crash if it's running in either one of the environments I've described, so I had to write a wrapper shell-script that unsets the XIM-variables, sets the language to English and then runs /usr/bin/emacs. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] error in a kernel-header file?
Hi,
can anybody here explain to me the rationale behind having
/usr/include/linux/videodev.h include fs.h?
This is done in kernel-headers in 9.0, but it's not done in stock 2.4.18,
2.4.19 or any of the pre-20 patches.
I've emailed cooker to ask about this, but nobody has responded [it's a 9.0
question, after all]. I was hoping that somebody here has both the time
and the knowledge...
There has been quite a discussion about this on the mjpeg-users list
because mjpegtools fails to compile due to this bug/feature(!). In fact,
the following program fails to compile on 9.0
#include
#include
int main()
{
return 0;
}
but compiles on the other linux systems that I have access to.
Narfi.
Want to buy your Pack or Services from MandrakeSoft?
Go to http://www.mandrakestore.com
Re: [expert] mdk9.0 with japanese user
Here's what I did in 9.0 final: 1) Selected Japanese as a second language during the installation 2) After I was running the 9.0 final, I dropped into a console and did # mv ~/.kde ~/.kde-old and kept my ~/.i18n file intact. 3) I logged into KDE and had it recreate my settings. That's it, I was instantly able to type in Japanese. Your ~/.i18n looks fine to me. If this does not work for you, you should try to create a new account on your machine, log into KDE, switch to Japanese in the KDE control panel and then log out. If you can't log into this account after the switch, you've most likely got some system problems [or something went wrong between RC3 and 9.0 final, I have do admit that I haven't done this in the final] Narfi. One hint for future use: It took me a long time to figure out that with these settings, KDE applications refuse to run over ssh if one does not have kinput2 running, this is a change from 8.2 with KDE 3.03 Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] msec, kernel panic and zip drive
Summary: When msec is run (at 4:03 am) it seems to trigger a kernel panic. I have also got the same kernel panic when shutting down, right after Alsa has been stopped. In either case, when this happens, I have to resort to the SysRq sequence. This seems to be related to the zip drive: # uname -a Linux xxx.new.rr.com 2.4.18-6mdk #1 Thu Mar 28 22:32:10 EST 2002 i686 unknown I've looked for this on Cooker and have seen some vague talk about ide-cd possibly being at fault, but nothing about how to fix this in 8.2. Does anybody here know what to do? Attached: /etc/fstab and a part of /etc/lilo.conf. Below: Output from /var/log/messages. Thanks, Narfi. CROND[12083]: (root) CMD (nice -n 19 run-parts /etc/cron.daily) anacron[12086]: Updated timestamp for job `cron.daily' to 2002-05-23 kernel: Device not ready. Make sure there is a disc in the drive. kernel: sda : READ CAPACITY failed. kernel: sda : status = 0, message = 00, host = 0, driver = 28 kernel: Current sd00:00: sense key Not Ready kernel: Additional sense indicates Medium not present kernel: sda : block size assumed to be 512 bytes, disk size 1GB. kernel: /dev/scsi/host0/bus0/target0/lun0: I/O error: dev 08:00, sector 0 kernel: I/O error: dev 08:00, sector 2 kernel: I/O error: dev 08:00, sector 0 kernel: Unable to handle kernel paging request at virtual address 204f2f8d kernel: printing eip: kernel: c0161873 kernel: *pde = kernel: Oops: kernel: CPU:0 kernel: EIP:1010:[devfs_read_super+51/208]Tainted: P kernel: EIP:1010:[]Tainted: P kernel: EFLAGS: 00010202 kernel: eax: cd53ccc0 ebx: 204f2f49 ecx: edx: cd53ccc0 kernel: esi: cdf38380 edi: df7bdbc0 ebp: c9c504a0 esp: ddbfbf28 kernel: ds: 1018 es: 1018 ss: 1018 kernel: Process msec_find (pid: 12182, stackpage=ddbfb000) kernel: Stack: cdf38380 c0161d06 df7bdbc0 c02676a0 cdf38380 cdf38400 cdf383ec kernel:c9c504a0 c0142780 c9c504a0 ddbfbfa0 c0142c80 c9c504a0 fff7 000d kernel:bfffecd8 c0142e2f c9c504a0 c0142c80 ddbfbfa0 cdf392e0 c01349e7 cdf392e0 kernel: Call Trace: [devfsd_ioctl+54/304] [do_poll+96/240] [fifo_open+160/601] [fifo_open+591/601] [fifo_open+160/601] kernel: Call Trace: [] [] [] [] [] kernel:[do_readv_writev+663/704] [do_simd_coprocessor_error+35/128] kernel:[] [] kernel: kernel: Code: 66 8b 43 44 25 00 f0 00 00 66 3d 00 60 75 0d f6 43 10 04 74 /dev/hde1 / ext2 defaults 1 1 none /dev/pts devpts mode=0620 0 0 none /dev/shm tmpfs defaults 0 0 /dev/hde7 /home ext2 defaults 1 2 /dev/sdb1 /mnt/camera auto noauto,ro /mnt/cdrom /mnt/cdrom supermount dev=/dev/scd0,fs=iso9660,ro,--,iocharset=iso8859-1 0 0 /mnt/cdrom2 /mnt/cdrom2 supermount dev=/dev/hdg,fs=iso9660,ro,--,iocharset=iso8859-1 0 0 /mnt/floppy /mnt/floppy supermount dev=/dev/fd0,fs=vfat,--,iocharset=iso8859-1,sync,codepage=850 0 0 /dev/hda1 /mnt/win_c vfat iocharset=iso8859-1,codepage=850,umask=002,gid=100,setgid=100 0 0 /dev/hda5 /mnt/win_d vfat iocharset=iso8859-1,codepage=850,umask=002,gid=100,setgid=100 0 0 /mnt/zip /mnt/zip supermount dev=/dev/sda4,fs=vfat,--,iocharset=iso8859-1,sync,codepage=850 0 0 /dev/hde9 /opt ext3 defaults 1 2 none /proc proc defaults 0 0 /dev/hde8 /projects ext3 defaults 1 2 /dev/hde6 /usr ext2 defaults 1 2 /dev/hde10 /var ext2 defaults 1 2 /dev/hde5 swap swap defaults 0 0 # added by NeTraverse - DO NOT REMOVE THIS LINE, it's used for uninstall default=win4lin # end added by NeTraverse - DO NOT REMOVE THIS LINE, it's used for uninstall boot=/dev/hde map=/boot/map install=/boot/boot.b vga=normal keytable=/boot/us.klt lba32 prompt nowarn timeout=100 message=/boot/message menu-scheme=wb:bw:wb:bw ignore-table disk=/dev/hda bios=0x81 disk=/dev/hde bios=0x80 image=/boot/vmlinuz label=linux root=/dev/hde1 initrd=/boot/initrd.img append="nobiospnp devfs=mount hdc=ide-scsi hdb=ide-scsi" vga=788 read-only Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Help! i18n and KDE problem on Mandrake 8.2
On Thursday 18 April 2002 19:09, you wrote: > The problem: > KDE doesn't respect locale setting (LANG, LANGUAGE, LC_ALL...) > Description: >1. I installed Mandrake 8.2 using its CD1 and CD2. During its > installation, I chose Simplified Chinese (zh_CN.GB2312) as installation > language, and also chose other languages (includeing en_US) to be > available after installation. >2. After installation, as expected, KDE uses Chinese fonts for > display. Then I decided to switch back to English fonts without much > success: I changed /etc/sysconfig/i18n to use en_US, and I don't have > local .18n; I also changed /etc/menu-methods/lang.h to make lang() and > languages() return en_US only. Now, running "locale" shows the > following: > You must modify the language for KDE in 2 places: The ~/.i18n [well, you chose /etc/sysconconf/i18n] as well as in the KDE control center. You can now go to the KDE control center->Personalization->Country and language and change Country, Language and Charset to your liking. You may also want to change your fonts because right now they are best suited for Chinese. You can also use localedrake: Run /usr/sbin/localedrake as a regular user and select the language that you want to use. You can run it either in a console or in X. The localedrake modifies your ~/.i18n file and the KDE control center settings, all in one sweep. It's actually very nice for people that want to switch from one language environment to another. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] mdk8.2 with japanese user
[From the way you wrote your question, I'm assuming that you wanted it to go to the entire list and not just to me personally, therefore I am replying to the entire list. If that was not the case, then I apologize] On Friday 31 May 2002 17:58, you wrote: > On Fri, 31 May 2002 12:39:41 -0500 > > "H. Narfi Stefansson" <[EMAIL PROTECTED]> wrote: > > On Friday 31 May 2002 04:00, you wrote: > > > Has anyone got 8.2 working sucessfully with Japanese input/display? > > Yes, > > http://www.math.wisc.edu/~stefanss/japanese > Anyone seen similar docs for Korean? I can get everything but input > working right for my wife. I need to be able to allow her to input > English and Korean. (Limited Chinese would be nice but not really > needed,) > > James I don't know anything about Korean so I may very well be wrong. Bear with me in my guessing game. By looking at http://www.mandrakelinux.com/l10n/advance.php3 which is admittedly a somewhat obsolete document, I walk away with the following information: The ~/.i18n file for a completely Korean environment should look something like this: LC_ALL=ko ENC=kr XIM=Ami XMODIFIERS="@im=Ami" and I'd guess that the ami input method editor will be started automatically in 8.2 when you start KDE. Try to look at the webpage I wrote about Japanese in 8.2 and see if you can't figure out how to use the information about Japanese to set up Korean properly. But remember that you should have selected Korean as a second language during install, this is very important -- I cannot tell you which packages you would otherwise be missing. Good luck, Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] mdk8.2 with japanese user
On Friday 31 May 2002 04:00, you wrote: > Hi, > Similar to how i setup my mdk8.1 machine to work with Japanese in KDE > etc I have now been trying to do the same on 8.2. However X either > crashes or the whole system crashes while starting X. If it does start > kinput does not work. > > I used the same technical info as last time, modifying the .i18n file > and the KDE fonts etc. > > Has anyone got 8.2 working sucessfully with Japanese input/display? > > JG [Sorry for the possible repost -- I believe the first attempt failed] Yes, http://www.math.wisc.edu/~stefanss/japanese The main difference that I found between 8.1 and 8.2 is that if the fonts are not set properly and Japanese is enabled, KDE will crash on startup. For that reason, I used localedrake to set the locale to Japanese [this sets the locale and the fonts for KDE] and then I modified the KDE control center and the ~/.i18n file to my liking. The webpage has detailed directions. Please keep in mind that I did test these instructions by creating a new user account and following the relevant sections in detail. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] OpenOffice and libstdc++.so.4
On Wednesday 22 May 2002 09:02, you wrote: > On Tuesday 21 May 2002 07:06 pm, Brian Schroeder wrote: > > I recently downloaded OpenOffice.org-1.0-3mdk.i586.rpm from cooker, > > but when trying to install it, it complains that it needs > > libstdc++.so.4. I have had a good look around, but haven't been > > able to find it. > > > > If the cooker version of OpenOffice was built with it, it must be > > around somewhere. Can anyone give me a pointer?Also, if I > > install this library, is it likely to break anything else? > > Probly, cooker has become very incompatible with 8.2 over the > last several weeks. Libs, gcc3.1, etc. Try this one, it also has a > freetype2 fix the cooker version doesn't (for legal reasons). > http://ranger.dnsalias.com/mandrake/mandrake8.2/OpenOffice.org-1.0-3mdk. >i586.rpm Actually it doesn't have "the freetype fix" [unless he changed that going from 1.0-2 to 1.0-3, but I'm unaware of that], but it allows you to install a better freetype2 and thus improve the look of the fonts in OpenOffice.org. Here's the announcement Ranger posted on pclinuxonline: http://www.pclinuxonline.com/modules.php?name=News&file=article&sid=2139 "... I have compiled OpenOffice.org-1.0-2mdk from cooker on 8.2. This release has changed so that OpenOffice.org uses the system freetype2 rather than it's own copy. See below what the difference is The difference is that if you install a bytecode-interpreter-enabled freetype2, that you can get OpenOffice.org to look something like this, as opposed to what it looks like if you use the build from OpenOffice.org. ..." You'll find a link to the bytecode-interpreter-enabled freetype2 hosted by the penguin liberation front in the actual announcement. Beware of the legal restrictions of its use, though. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] ps2pdf sorcery
On Sunday 21 April 2002 12:48, you wrote: > On Sat, 20 Apr 2002 17:26:02 -0500 > > Jason Guidry <[EMAIL PROTECTED]> wrote: > > Is there any way to input multiple postscript files through ps2pdf and > > have them output to one pdf? the man page mentions nothing of this. > > > > Is there any other program that will do this? > > > > thanks > > > > -- > > Jason Guidry > > http://www.gmaestro.org > You can try using psmerge to merge your postscript files before sending them to ps2pdf. From "man ps2pdf": Psmerge merges PostScript documents into a single document. It only works in the specific case the the files were created using the same application, with the same device setup and resources (fonts, procsets, patterns, files, etc) loaded. Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] experiences with K7T266 Pro2/K7T266 Pro2 RU
Hi everybody. I am thinking about upgrading my computer sometime shortly after 8.2 comes out. Right now I'm looking at MSI K7T266 pro2 or the pro2 RU. The RU version has on-board RAID which is a minus in my mind, but it also has USB 2.0 and I'm hoping that is a plus. Does anybody on the list have either of the 2 boards? I'm specifically wondering about: Does the integrated sound card work in Mandrake w/o a lot of hassle? Does the integraded network card work in Mandrake w/o a lot of hassle? And if you have the RU: Can you disable the raid controllers in the BIOS so that they don't occupy any IRQs? What about the USB 2.0 ports? Is anyone aware of any issues/difficulties regarding those ports? I'd like to be able to use USB 2.0 when the time comes that it's properly supported by the kernel and I have some gimmicks to play with :-) Supposedly 8.2 beta4 has USB 2.0 in the kernel so it's all coming along... Thanks, Narfi. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
