Mutt + Korn
Dear list, I use the Mutt email client together with KDE's Korn new mail notifier, and have the following problem. When new mail arrives, Korn will notice and alert me to the fact. I then go to Mutt and read the new mail. After reading the new mail in Mutt, it is no longer marked as new Mutt-internally. However, Mutt does not automatically resynchronise mailbox state with the mbox file on disk, and because Korn only watches the file on disk, it does not realise that I have read the new mail and keeps telling me I haven't. I have to manually resynchronise the mailbox with the disk file by pressing the $ key in Mutt, and that gets old quickly. So basically my question is: How can I get Mutt to automatically resynchronise in-memory mailbox and mailbox file whenever the newness status of a message in that folder changes? Cheers, Stefan -- Stefan Baums Asian Languages and Literature University of Washington -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: coding system woes / OpenOffice (amended)
#export LC_CTYPE=de_DE.utf8 The official form of the locale name is de_DE.UTF-8. Dont know whether that causes your problem. Stefan -- Stefan Baums Asian Languages and Literature University of Washington -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Configuring APT
Package: * Pin: release a=testing Pin-Priority: 1001 Package: * Pin: release a=unstable Pin-Priority: 400 Package: * Pin: release a=stable Pin-Priority: 300 On further reflection I realised that those numbers will only do what I want (procession from unstable/third-party to testing version) for packages that don't exist in testing to start with. The KDE (3.1) packages do, of course, so P = 1001 for testing will block the installation of any 3.2 version from outside testing. Sigh. Is there any other magic combination of numbers and target-release setting that _will_ do what I want? I'd be willing to do without the slight-downgrade on switching from unstable to testing version effect, and could even forgo stable if need be (though the presence of stable doesn't seem to be the main problem). Stefan -- Stefan Baums Asian Languages and Literature University of Washington -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Configuring APT
Dear list,
I would like to configure APT to do the following:
- install requested and update installed packages from
repository X (unstable or third-party) for as long as those
packages are not available in repository Y (testing or
stable)
- automatically switch to updating those packages from
repository Y instead of X as soon as they become available in
Y, even if that requires a downgrade
Currently, my /etc/apt/preferences file contains
Package: *
Pin: release a=testing
Pin-Priority: 900
Package: *
Pin: release a=stable
Pin-Priority: 800
Package: *
Pin: release a=unstable
Pin-Priority: 700
This does not do what I want because it will keep installing
packages from unstable once a version from unstable is on my
system, even after a testing version of that package becomes
available.
I do not have an /etc/apt/apt.conf file. There is an
/etc/apt/apt.conf.d/70debconf file which contains
// Pre-configure all packages with debconf before they are installed.
// If you don't like it, comment it out.
DPkg::Pre-Install-Pkgs {/usr/sbin/dpkg-preconfigure --apt || true;};
Does anybody know how to do what I want?
Thanks in advance,
Stefan
--
Stefan Baums
Asian Languages and Literature
University of Washington
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Configuring APT
Read the apt_preferences(5) man page, I did, but obviously not closely enough. set the pin for unstable to 100 P =500. Done. However, after reading the man page I believe that I also need to set testing to P 1000 in order to enable downgrading when the unstable or third-party package appears in testing. However, what on earth are you doing? Mixing up all three versions of Debian is NOT supported, and will eventually break in odd ways. Easy: I want to - use testing as the main distribution because it is generally the best compromise between stability and up-to-dateness. - use unstable or third-party repositories for software that isn't in testing yet, but that I really want to have (like KDE 3.2 when it is released in a couple of days). - use stable for packages that have been discontinued in testing and unstable, but for which security updates might still appear in stable. So now my /etc/apt/preferences file contains: Package: * Pin: release a=testing Pin-Priority: 1001 Package: * Pin: release a=unstable Pin-Priority: 400 Package: * Pin: release a=stable Pin-Priority: 300 and I also created an /etc/apt/apt.conf containing APT::Default-Release testing; though strictly speaking that appears to be redundant with the above Pin-Priorities. So if everything works as planned, I would e.g. add one of the third-party KDE 3.2 repositories to /etc/apt/preferences with P = 350 when KDE 3.2 is released. That should allow me to install KDE 3.2 packages from the third-party repository to start with, but later cause APT to automatically switch to the unstable and then testing versions of those packages as they become available, downgrading slightly if necessary. Shouldn't that work? Please tell me why you believe that things will break terribly if one uses more than two distributions, and what you mean by is NOT supported. Thanks for your help, Stefan -- Stefan Baums Asian Languages and Literature University of Washington -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
PHPPGAdmin login problem
Dear list, I am new to databases, but would like to use PostgreSQL with the PHPPGAdmin frontend. So I installed postgresql, apache, and phppgadmin from testing. Then I created a database user test with password test with the following command: sudo -u postgres createuser -P and tried to log in on http://localhost/phppgadmin/ but get the error message Wrong username/password. Access denied. (I put a screenshot of the login screen on http://staff.hum.ku.dk/baums/screenshot.png .) Can anybody help me with this? Many thanks, Stefan -- Stefan Baums Asien-Instituttet Københavns Universitet -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
PCMCIA in 2.4 kernels doesn't work
Hi all, I am having problems getting the PCMCIA code included in the 2.4 kernels to work. If I compile it in as modules and then reboot to the new kernel, I get the error message Starting PCMCIA services: modules/lib/modules/2.4.7/pcmcia/i82365.o: init_module: No such device Hint: insmod errors can be caused by incorrect module parameters, including invalid IO or IRQ parameters /lib/modules/2.4.7/pcmcia/ds.o: init_module: Operation not permitted Hint: insmod errors can be caused by incorrect module parameters, including invalid IO or IRQ parameters cardmgr. If, however, I compile the separate pcmcia-source package for use with the 2.4 kernel, everything works just fine. I'm running testing with kernel-source-2.4.6, and pcmcia-cs and pcmcia-source 3.1.22-0.2potato. Does anybody know what's going on here? TIA, Stefan
Re: PCMCIA in 2.4 kernels doesn't work
On Thursday, 9 August 2001, Michael Heldebrant wrote: For one thing you just said you were running 2.4.6. The /lib/modules/2.4.7 is most likely not there. Sorry, I was trying it out with both 2.4.6 and 2.4.7, and mixed them up in my email. Secondly: Your kernel is probably not set to accept module symbols from other compiled modules. This is actually something that you can solve in two ways I beleive. 1. Compile it yourself (safest, and you already did that). 2. Turn on the set module symbols for compiled modules and hope that the precompiled modules have the same (more of a shot in the dark). I'm compiling the modules myself anyway, but want to use the ones included in 2.4 kernels, not the separate pcmcia-source package. I solved the problem, though, see my other email. Thanks, Stefan
Re: PCMCIA in 2.4 kernels doesn't work
On Thursday, 9 August 2001, Mario Vukelic wrote: I don't think it's worth the trouble to try and get it to work with 2.4.x drivers if they work the other way anyhow. Using the separate package with 2.4 kernels is inconvenient because you have to compile and install two packages where one can be enough. Recently, the separate pcmcia-source package also _stopped_ compiling cleanly with 2.4 kernels, because of changes to the kernels aic7xxx SCSI driver. It takes some fiddling to get it to compile after all. I found the solution to my problem, though. I quote from the documentation on pcmcia-cs.sourceforge.net: The CardBus socket driver in the 2.4 tree is the yenta_socket driver. In your PCMCIA startup options, this driver should be specified in place of the old i82365 driver. You do this in /etc/pcmcia.conf. Easy, really, but the Debian pcmcia-cs documentation does not explain this anywhere. Thanks, Stefan
Re: Trackpoint and serial mouse simultaneously
Sorry to answer my own question, but a bit of research revealed that while Xfree86 3.3 knows only one pointer, gpm can manage multiple ones for it. My /etc/gpm.conf now has: device=/dev/psaux type=ps2 append=-M -t mman -m /dev/ttyS0 and the pointer section in /etc/X11/XF86Config is: Section Pointer ProtocolMouseSystems Device /dev/gpmdata Emulate3Buttons EndSection Cheers, Stefan
Trackpoint and serial mouse simultaneously
Dear list, I'm running an HP Omnibook 5500 laptop here, and would like to use, under X, the internal TrackPoint pointing device and an external serial mouse at the same time (i. e., without restarting X to make the switch). More generally, is it possible to have more than one Pointer device for the same X session under XFree86 3.3? (As long as 4.0 is in unstable I prefer not to switch.) Thanks, Stefan
Re: EMERGENCY: LILO can't load Win95 / Mount fails
I found a Win98 bootdisk, ran fdisk /mbr on C:, but Win95 still did not boot. Then I booted into Linux, reran lilo with /dev/hda1 (the Win95 partition) as the thing to load, and got an error message (something about invalid partition specification). This did not bode well, I went into cfdisk to check, and indeed - the disk is 'empty' now, the MBR wiped clean. Nothing to do then except to give it to a technician at our University (Koebenhavn), who suggested that he might be able to rescue at least the most important files. Thanks for your help to Oswald and Dave, Stefan -- PGP-ID: 4AF752D0
EMERGENCY: LILO can't load Win95 / Mount fails
Hello all! I have a Win95 partition here with important data on it (and no recent backups, of course...) which I can no longer boot from or access from GNU/Linux. The setup is: hda1 contains what cfdisk calls Win95 FAT32 (LBA) (the lost Win95 partition) hda2 is Linux Swap hda3 is unused hda4 contains a Debian 2.0 base system LILO was set up to automatically boot Win95. The only major thing we did was install Ghostscript 6.01 for Windows, then next day when we turned the computer back on, LILO got stuck in booting after printing LI on screen. I then went into the Debian 2.0 system with a rescue disk, wrote a series of lilo.conf's, ran lilo, rebooted. The result was always that LILO got to the point where it told me LILO: Loading win. and then got stuck, retried, and repeated the message over and over again. The lilo.conf's looked like this: boot=/dev/hda vga=normal delay=20 other=/dev/hda1 label=win Then I tried to mount the Win95 partition from GNU/Linux to save the most important data, but the result of mount -t vfat /dev/hda1 /mnt is: [MS-DOS FS Rel. 12,FAT 0,check=n,conv=b,uid=0,gid=0,umask=022,bmap] [me=0x85,cs=0,#f=81,fs=0,fl=32512,ds=12032,de=58188,data=15668,se=38200,ts=74624,ls=9471,rc=0,fc=4294967295] Transaction block size = 512 VFS: Can't find a valid MSDOS filesystem on dev 03:01. mount: wrong fs type, bad option, bad superblock on /dev/hda1, or too many mounted file systems Finally, I tried the crappy Windows QuickRestore CD we got with the computer, which only allows you to completely reformat your harddisk, or to run a virus check. I just did the latter, and the interesting thing is that the virus checker did scan all the files on the Windows partition, so they are still there, and this program could access them. Can anybody provide help with this problem? Many thanks in advance, Stefan and Jenny -- PGP-ID: 4AF752D0
Re: xterm with utf-8
Oops. I found out that UTF-8 support actually _is_ compiled into recent xterms. You just have to enable it with command line options / X resources, and to use am ISO-10646 font. Unifont from the Debian distribution for some reason does not really work, but I got myself Markus Kuhn's ISO-10646 replacements for some standard X fonts, and even though do not yet contain glyphs for the whole range of their encoding, I can now do nifty things like viewing Russian or Greek or Lithuanian script webpages in Lynx! Stefan pgpo1OZMoQA3f.pgp Description: PGP signature
Re: xterm with utf-8
On Sat, Jul 03, 1999 at 10:31:15AM -0400, [EMAIL PROTECTED] wrote: ok (I've really only tested with Markus' font, though Juliuz seems to think it works for his unifont as well). -- there's more work to be done, though (report bugs) The problem with unifont is that the characters themselves are displayed correctly, but are drawn too far apart. Looks like an extra space character is inserted between any two characters. I am not sure which package that bug would belong to: unifont works all right with yudit, but not with xterm xterm works all right with Markus' fonts, but not with unifont Probably unifont, though: There is the same spacing problem with unifont + gnome-terminal. Greetings, Stefan pgpSKHNO6H5Xd.pgp Description: PGP signature
xterm with utf-8
From /usr/doc/xterm/changelog.Debian.gz: xfree86-1 (3.3.3.1-4) unstable; urgency=low * (#011): Debian-specific xterm patches - disable UTF-8 support, upstream author is still working on it How can I reenable the UTF-8 support? I got myself the sources, but do not know _where_ to disable the patch. Stefan pgp9SRuzm9RQY.pgp Description: PGP signature
Smail queue
Hi all. Running recent potato, using fetchmail to get my mail. When fetchmail has done its job, mail is stuck in smail's input queue (the one you see with mailq) and not immediately forwarded to /var/spool/mail/myname, which is the behaviour I would like in combination with fetchmail. How can I get smail to behave that way? TIA, Stefan (Please cc me)
autofs for all users
Hi all, I am using autofs. My /etc/auto.master is: /mnt/amnt /etc/auto.amnt --timeout 1 and my /etc/auto.amnt is: cdrom -fstype=iso9660,ro,user :/dev/hdc floppy -fstype=auto,rw,user:/dev/fd0 If I access /mnt/amnt/floppy for reading, everything is fine. If I try to write something to the disk as ordinary user, I get: cp: cannot create regular file `/mnt/amnt/floppy/test': Permission denied The problem seems to be: If Joe User accesses /mnt/amnt/floppy, the automount daemon, _running as root_, calls mount _as root_ so that root is the owner of /mnt/amnt/floppy and the only one allowed to write there. How can I let ordinary users access the floppy for writing? TIA, Stefan pgppvw40e4FO8.pgp Description: PGP signature
Re: autofs for all users
Thanks, Brian! That worked. pgpwdbDKlpoJw.pgp Description: PGP signature
Title change in rxvt
Hi all, I am running tcsh in rxvt. Is there any way to have the title of the rxvt window change according to the program running in it? (There must be one, as right now as I am typing this I notice that vim has changed the title to VIM - /tmp/mutt-sb51-12020-0.) TIA, Stefan pgp2z8kBWk7Mz.pgp Description: PGP signature
Remove occurences of X from file
Hi all. I use Emacs/MULE and write the following document: Hall� KuryLowicz where (this is the important bit!) the L of this iso-latin-1 email is actually the Polish l (l with a bar through it). Then I save it in the encoding emacs-mule. According to less in an iso-latin-1 terminal the file looks like: Hall82� Kury82�owicz Question: Can anyone clue me how to write a small script to remove the 82 characters from the file? (The point is to then feed it to LaTeX and \inputencoding.) Many thanks in advance, Stefan pgpRRxIqethRu.pgp Description: PGP signature
Re: Online Metafont graphics?
What is your TERM variable set to (I tried rxvt, xterm and xterm-debian), and do you use MFTERM (according to man mf not necessary if TERM is set)? What shell do you use (export sounds like bash; I use tcsh)? Finally, what sort of output can I expect? A large fat dot, or a tiny one I could easily miss (unlikely though)? Thanks for your help, Stefan PS. I tried it from the console, too (TERM = linux, I think): no effect. pgpMqT16cwoj6.pgp Description: PGP signature
Re: Online Metafont graphics?
Well, it's just not working. No dots in sight. Nils, what version of Debian and teTeX are you using? Does the output appear in the same window where you type the commands, or does another xterm pop up? Puzzled, Stefan pgpUmOHv6V910.pgp Description: PGP signature
Re: Online Metafont graphics?
I�m using wmaker-gnome. Just now tried twm: no difference in behaviour. I�m not using xdm. I don�t get any error messages from startx though. No errors in .xsession-errors. I first tried it with tetex-bin 0.9.981113-2 (the slink/stable version), then got tetex-bin 0.9.990406-1 (from potato/unstable): no difference. /home/stefan echo $DISPLAY $TERM $MFTERM :0.0 rxvt rxvt I�m focussing now on trying to make it run in a plain xterm, with the unnecessary MFTERM (and COLORTERM, just to be paranoid) unset: /home/stefan echo $TERM $DISPLAY xterm :0.0 /home/stefan mf This is METAFONT, Version 2.7182 (Web2C 7.3) **\relax *drawdot (20,20); *showit; mf: Window support for X was not compiled into this binary. mf: To do so, rerun configure --with-x, recompile, and reinstall. mf: (Or perhaps you just failed to specify the mode.) *end Transcript written on mfput.log. /home/stefan Well, as I said, according to the source package's debian/rules, mf _is_ compiled with x. My system is configured with texconfig, of course, so IMHO the problem should not be a missing mode specification, but let's do it manually: /home/stefan mf '\mode:=ljfour' This is METAFONT, Version 2.7182 (Web2C 7.3) *drawdot (20,20); ! Extra tokens will be flushed. to be read again addto addto_currentpicture-addto .currentpicture drawdot-...:def_pen_path_.fi.addto_currentpicture .contour.currentpen_path.s... to be read again ; * drawdot (20,20); ? x Transcript written on mfput.log. /home/stefan I don't understand these error messages. At a loss, Stefan On Wed, Apr 14, 1999 at 09:42:28AM +0200, Nils Rennebarth wrote: On Wed, Apr 14, 1999 at 01:08:47AM +0100, Stefan Baums wrote: Nils, what version of Debian and teTeX are you using? Does the output appear in the same window where you type the commands, or does another xterm pop up? Neither of the two. A Giant window titled mf pops up, it's just an X-Window, behaving rather strange with my window manager (can't raise it, can't receive focus) What wm are you using? Are there some messages in /var/log/xdm*? tetex-bin is 0.9.980706-1, but I doubt it depends on that. What are the values of your DISPLAY, TERM, MFTERM? pgpHs1oSoz1Qv.pgp Description: PGP signature
Metafont and X: the solution
Dear Nils, and all you other attentive followers of this thread: I finally got the solution to the MF + X problem. I was already recompiling the sources, but the strange thing was that --with-x actually does appear in debian/rules. So I took a close look at the configure files, and there I found the option --with-mfw, undocumented in the Web2C manual. mfw, I guessed, is mf with window support... /home/stefan locate mfw /usr/bin/mfw /usr/man/man1/mfw.1.gz /usr/src/linux/net/ipv4/ip_masq_mfw.c /home/stefan The man page is the Debian dummy man page. The only problem is that mfw standardly expects a mfw.base base file, which does not exist. So what you do is: use whichever window manager and terminal emulator you want; have your TERM variable set to what you like (in my case rxvt); _but_ set MFTERM to xterm; and finally, when running mfw explicitly demand mf.base as the base file: mfw -base mf Voila, la fenetre. If you want a reasonably sized window, set the X resource Metafont.geometry. And that's it. The one thing I do consider a bug is that mf.base is not symlinked to mfw.base (as it is symlinked to plain.base). And a mention of mfw in /usr/doc/tetex-bin/README.debian would be nice; /usr/man/man1/mfw.1.gz could then point to the mf man page, with maybe the addition of mfw in the NAME section, maybe even one or two sentences in the ONLINE GRAPHICS OUTPUT section pointing the reader to mfw (but then, I don't know if changing the upstream manpage for what is essentially a compile time option is good). I haven�t filed a bug yet; instead I'm sending a copy to the maintainer (hi Christoph) and see what happens next. Have a nice day, Stefan pgpqZijmY2Cof.pgp Description: PGP signature
Online Metafont graphics?
Hi all! I'm trying to teach myself Metafont (using Debian 2.1, tetex from unstable; same problem with stable tetex, though). According to the documentation, mf can do online graphics when compiled --with-x; I checked the debian/rules in the source package, and it is indeed compiled this way. $TERM and $MFTERM are set to rxvt (no difference when using xterm). Now I run mf interactively as follows: /home/stefan mf This is METAFONT, Version 2.7182 (Web2C 7.3beta6) **\relax *drawdot (50,50); showit; end. Transcript written on mfput.log. /home/stefan That should draw a dot at position (50,50) (wherever that is), but I don't get any on-screen output. Any ideas anyone what's going wrong? Thanks in advance, Stefan
Problem with MPEG3s
Hi all! My system is an HP Omnibook 5500 CT running Debian Slink and the Linux 2.2.2 kernel. The following are the sound modules I use (inserted by modprobe cs4232) Module Size Used by cs4232 2408 0 uart401 5904 0 [cs4232] ad1848 14992 0 [cs4232] sound 56396 0 [cs4232 uart401 ad1848] soundcore 2148 5 [sound] and the configuration is options cs4232 io=0x220 irq=5 dma=1 dma2=3 Now, one minor problem is that when my computer comes back from suspend it does not produce any audible output when I play some sound file. When I remove and reinsert the sound modules, it's working all right. (I added the remove / reinsert operations to /etc/apm/.) The real problem is that playing MPEG3s kills my sound system. I experienced this with freeamp and x11amp (slink and potato versions), and several different MPEG files (and all the kernel versions I tried: 2.2.0 - 2.2.2, and even 2.0.36, I think). The file plays all right, but then, when I want to play something else (any format), the only output I get from my speakers is a very soft clicking noise at ca. 1 second intervals. The only way to repair that condition I could find is doing a hard reset (not even a soft reset will do). I have no clue what's happening, and I don't even see in principle how a simple user-space program can screw up my sound system so royally. Any suggestions anyone? Best wishes, Stefan
Re: Problem with MPEG3s
I just rechecked the situation: In fact, simply killing my sound apps, removing the modules, going into suspend, coming back, reinserting the modules, restarting the sound-apps does the trick. But still, you wouldn't want to do that after every MPEG3 title you play... (BTW, even if I start, say, x11amp with a list of more than one title to play, sound stops working when the first title is finished and the second begins. Still no error message, just no sound output anymore...) Stefan
Do you still need screensavers these days?
Hi all, I remember reading somewhere that you don't really need screensavers anymore with modern monitors to prevent damage to them, and that the only reason to run them is to show off to your colleagues or whoever. Now I'm wondering if that is true, and if it applies to (TFT) LCD diplays as well (which is what I am using). Any ideas? Cheers, Stefan pgptYxidiWdLI.pgp Description: PGP signature
dpkg broken?
Hi all. OK, so yesterday, in the process of updating my system to the newest slink, I broke dpkg. I cannot install packages anymore. The error messages are as follows: Preparing to replace modconf 0.2.23 (using modconf_0.2.24.deb) ... Unpacking replacement modconf ... gzip: stdout: Broken pipe dpkg-deb: subprocess gzip -dc returned error exit status 1 dpkg: error processing modconf_0.2.24.deb (--unpack): subprocess dpkg-deb --fsys-tarfile returned error exit status 2 Errors were encountered while processing: cron_3.0pl1-50.deb freetype2-dev_1.2-3.deb freetype2_1.2-3.deb kernel-doc-2.0.36_2.0.36-2.deb libg++2.8.2_2.91.60-4.deb modconf_0.2.24.deb E: Sub-process returned an error code Exit 100 Actually, the problem might be not dpkg as such: to remedy the situation, I grabbed dpkg and gzip from stable and manually (ar -x ; tar xvzf ; cp -R) installed them, but that has not changed the situation. Some other package must have broken dpkg. Any ideas? Stefan
DPI
Hi, I just read your dpi request on the Debian mailing list archives. In case you don't have an answer yet: I change my X dpi setting with the command-line option -dpi (startx -dpi 96); if you have learned of any more elegant way to do this, please tell me. Another thing seems to be the default resolution for which xfs serves fonts. You set this in /etc/X11/xfs/config (the factory setting in my case was 75 and 100 dpi). See man xfs. If you set the more general X resolution with -dpi, I do not think that the xfs setting has much further effect. Regards, Stefan
Unpacking Mac HQX files in GNU/Linux
Hi all! OK, so I download this true type font pack which turns out to be Mac compressed. The filename is STEDTpack50.sea.Hqx Does anyone know how to get at its contents under GNU/Linux? Is there even a Debian package for this? Or do I have to mail the site maintainers and beg them to make their stuff available in something other than incompatible-with-the-rest-of-the-world Mac garb? Thanks in advance, Stefan pgpwNammEgUHW.pgp Description: PGP signature
Re: Unpacking Mac HQX files in GNU/Linux
On Mon, Dec 14, 1998 at 09:46:43AM +, David Stern wrote: On Mon, 14 Dec 1998 15:17:17 GMT, Stefan Baums wrote: OK, so I download this true type font pack which turns out to be Mac compressed. The filename is STEDTpack50.sea.Hqx Does anyone know how to get at its contents under GNU/Linux? ftp://ftp.debian.org/debian/dists/stable/main/binary-i386/otherosfs/macu tils_2.0b3-6.deb Sorry, I could have thought of checking otherosfs myself... Anyway, I got macutils, did hexbin STEDTpack50.sea.Hqx and then macunpack STEDTpack50.sea.bin and ended up with a directory containing Read_Me_v5.0.bin, STEDT.bin, and STEDT_Font_5.0_Reference.bin. My guess is that STEDT.bin is the font file. So I tried all the converters in macutils on that file, but I couldn't get the .ttf's out of it. Since I have no clue how Macs work, I'm at a loss what to do now. Can anyone help? Thanks in advance, Stefan pgpvXHXUkE4Tk.pgp Description: PGP signature
What's using my harddisk? or, how to silence my computer
Hi all! I'm using Debian (the most recent packages) on a laptop with a pretty noisy harddisk (a Toshiba harddisk inside an HP Omnibook), and I'm sick to death of hearing it all the time I'm using my computer (and, of course, it's draining the battery as well). So I set my harddisk timeout to 30 seconds. But _something_ keeps accessing the disk. I tried strace on the most obvious candidates (xemacs, gnome-panel), but could not find the guilty program. Any suggestions anyone on how to monitor harddisk access on my computer? There must be a way to at least write a text on it without constant disk access (autosave concerns aside). Thanks in advance, Stefan pgpdxPphF5Yfw.pgp Description: PGP signature
libstdc++2.9 2.91.58-5 broken??
Hi all! I just tried to upgrade my system to the packages in slink, and now my libstdc++2.9 (2.91.58-5) seems to be broken. Whenever I try to do things like man any_program or apt-get install some_package or dselect I get the error message /usr/lib/libstdc++-libc6.0-1.so.2: undefined symbol: __register_frame_info Is that really the fault of libstdc++2.9? And anyway, does anyone know how to repair this? Best wishes, Stefan
Toshiba and HP's GNU/Linux compatibility
Hi all! I'll buy a notebook computer shortly, and one of my prime concerns is that GNU/Linux will not only run on it somehow, but actually use the available hardware features. At the moment, I'm considering the Toshiba Satellite 300 CDT and the HP Omnibook 2000 (+ internal CDROM for its accessory bay). To judge from the Linux on Laptops Page (http://www.cs.utexas.edu/users/kharker/linux-laptop/), Toshiba in general seems to be quite compatible with Linux, and HP far less so (though, or maybe because, it has the more interesting hardware). But the exact models I'm interested in have not been reviewed there. So, is anyone out there running GNU/Linux on one of these machines and can tell me about his experiences? Many thanks in advance, Stefan pgpOxf8ff6nT9.pgp Description: PGP signature
Re: Toshiba and HP's GNU/Linux compatibility
On Sun, May 17, 1998 at 11:47:24AM -0500, robert havoc pennington wrote: On Sun, 17 May 1998, Stefan Baums wrote: I'll buy a notebook computer shortly, and one of my prime concerns is that GNU/Linux will not only run on it somehow, but actually use the available hardware features. At the moment, I'm considering the If you really want to be sure your best bet is to get it with Linux preinstalled from one of the Linux vendors, like VA Research. They can install Windows too, I think. The if the hardware isn't Linux-friendly you have someone to hold responsible. Havoc Pennington http://pobox.com/~hp I'm living in Germany, and ordering a notebook from VA Research (in the USA) seems like going too far. Besides, their web page gives the impression that they do not just bundle any notebook with Linux, and the one they advertise (the NEC 6260 MX) a) is at $5560 a bit above my budget, and b) uses a proprietary NeoMagic graphics chipset which is something I try to avoid on principle. Come to think of it, I do not know about any German vendor offering Linux preinstalled. Thanks for your suggestion, Stefan pgpz1uWkAPQkD.pgp Description: PGP signature
HP Laserjet 6L
Hi all, is anyone running the HP Laserjet 6L with Debian GNU/Linux? I'm using magicfilter 1.2-20, and the 6L is not in magicfilterconfig's list of supported printers. Can I use another filter (say, for the 4L) and will I be able to get 600x600 dpi with that? Thanks for your help, Stefan pgpzwnJE5OEJp.pgp Description: PGP signature
SCWM 0.5-3 broken?
Hi all, my system is Debian 1.3 (manually updated to libc6 and using many other 2.0 packages as well). I have installed SCWM 0.5-3 and tried to start it using the Afterstep (my current WM) menu and $HOME/.xsession. In either case, initially it seemed to start up (though I didn't get as far as seeing any SCWM elements onscreen), but then it crashed and I was back at xdm. Any suggestions what might be going wrong or how I could log this to check out? Thanks in advance, Stefan pgpaMZn9U5Cxm.pgp Description: PGP signature
Emacs Menu/Scrollbar Color
Hi all, I'd like to use GNU Emacs under X with a black background and white foreground. The problem is, this should be true of the menubar as well, but Xresources background foreground don't do the job. (It's possible, though: I saw a screenshot.) And can I set the scrollbar color separately? (This seems to be an unrelated problem: the aforementioned Xresources _do_ affect the scrollbar.) I couldn't find the relevant information in the man pages or info documents. Any ideas? Thanks in advance, Stefan pgpWJ3exRgNZ2.pgp Description: PGP signature
scwm 0.5-3 / menu 1.5-7 incompatible? (was: SCWM 0.5-3 broken?)
On 08-Feb-98 Stefan Baums wrote: my system is Debian 1.3 (manually updated to libc6 and using many other 20 packages as well). I have installed SCWM 0.5-3 and tried to start it, but then it crashed and I was back at xdm. I could pin this down to a problem with /etc/X11/scwm/scwm_menus (generated by the menu package, version 1.5-7). If I place an empty scwm_menus file in ~/.scwm, thus preventing the reading of the system scwm_menus file, scwm starts smoothly if slooowly. So scwm 0.5-3 and menu 1.5-7 (most recent hamm versions both) seem to be incompatible. Can anyone replicate this? All the best, Stefan pgpltKMZr8OiB.pgp Description: PGP signature
xfmail 1.2p0-1 seg-faults
Hi all, I'm running xfmail 1.2p0-1 under hamm (i.e., I think all the other packages which I think _might_ be relevant to this problem are from hamm). When xfmail has to deal with new mail (in this new version marked in pink) on startup (or even folder change (?)) it crashes with a Segmentation fault. When I restart it, the new mail is no longer considered new (and not pink anymore) and xfmail works fine (until the next new mail arrives, that is). Could this even be an upstream problem? Have a nice day, Stefan pgprK5TITXhfJ.pgp Description: PGP signature
xlib6 3.3.1-2 + emacs 19.34-11 problem (?)
Hi all, my Debian system is basically bo. Some time ago I upgraded to libc6 2.0.6-2 and libc5 5.4.38-0.1 and didn't experience any problems with my old libc5 programs. But then I upgraded xlib6 to 3.3.1-2 and installed xlib6g 3.3.1-2 in order to use the new libc6 xfmail 1.2, and things took a bad turn: When I try to start emacs now I get a segmentation fault (this happens under X _and_ tty), and gv doesn't start either but produces the error message gv: No message text for missing-resources. Is this really the fault of my new xlib6? If so, what can I do about it? You know, I _need_ my emacs :-) Greetings, Stefan pgp4e2OJbn9Aj.pgp Description: PGP signature
Re: xlib6 3.3.1-2 + emacs 19.34-11 problem (?)
On 18-Jan-98 James Troup wrote: Stefan Baums [EMAIL PROTECTED] writes: When I try to start emacs now I get a segmentation fault (this happens under X _and_ tty), and gv doesn't start either but produces the error message gv: No message text for missing-resources. Upgrade your emacs and gv; earlier versions had hardcoded paths to the X libraries and so can't find the libc5 ones in /usr/lib/libc5-compat. That did it. Thanks, James. pgp2S3vPvF5mb.pgp Description: PGP signature
Re: xlib6 3.3.1-2 + emacs 19.34-11 problem (?)
Hi again, I did what James recommended for gv and it worked (thanks). -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Unidentified subject!
Hi all, would it be feasible to provide a WWW gateway to debian-user? Take your mailing archive: One would have to add the option to only read the messages of, say, the last three days, or even, if that's possible, to let the user decide which selection of messages to read. For me, the main problem with reading debian-user from the mailing archive (apart from the server's periodically being down...) is that I'd have to load the whole list of links for, say, July (that's 363 KB) even if it's the 29th and I only want to have a look at last day's new messages. I'm no html or Java buff, but wouldn't it even be possible to include forms for posting/replying? At the very least, you could use mailto:debian-user@lists.debian.org; links which invoke your favorite mailer. If this is practical (as an alternative, not necessarily a replacement for the mailing list) you'd have the convenience of a news group with the accessibility of a web site. Stability (in case of server failure) could be achieved by mirroring. And as for off-line reading, am I wrong in assuming there's some Debian tool for getting files/directories from WWW servers? I realize that all this would need some getting used to, but then again, I suggest it as an alternative to not a replacement for the mailing list. And it would obviate the necessity to split debian-user (we don't seem to be able to agree on _how_ or even _if_ to split it :-)). Greetings, Stefan -- mailto:[EMAIL PROTECTED] pgp1cds4N0eSA.pgp Description: PGP signature
Splitting up Debian-User
Ooops, my Unidentified Subject mail is about the above... -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: mail encrypting/pgp
Paul Miller wrote I noticed that almost everyone on this list uses pgp.. I installed the debian package and I'm wondering how do auto-encrypt/decrypt all my mail. I use pine (pinepgp package) and netscape communicator 4.01b6 For pine, set display-filters=_BEGINNING(-BEGIN PGP)_ /usr/lib/pinepgp/check _TMPFILE_ _RESULTFILE_ sending-filters=/usr/lib/pinepgp/crypt _TMPFILE_ _RECIPIENTS_, /usr/lib/pinepgp/sign _TMPFILE_, /usr/lib/pinepgp/auto _TMPFILE_ _RECIPIENTS_ in your ~/.pinerc and activate compose-send-offers-first-filter in pine's [S]etup-[C]onfig menu. This will automatically encrypt outgoing messages if you have the recipient's public key, decrypt incoming messages if they have been encrypted using your public key, display the status of pgp signatures, and add pgp public keys contained in incoming messages to your keyring. Greetings, Stefan PS. Just to test your setup, this message is pgp signed and my public key is included. 0.insertkeys.1591.exmh Description: keys of Stefan Baums sbaums@stud.uni-goettingen.de Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] pgpa9IBBVPtbb.pgp Description: PGP signature
Re: bo and hamm-packages
I have installed bo and would like to add some hamm-packages (which not exist in bo) without upgrading to hamm. Is this possible -- provided that there are no dependency problems? Yes. -- Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
From: Field in Mutt and (Ex)mh
Hi all, I'm connected to the Internet by a PPP modem line, and depend on a POP3 host stud.uni-goettingen.de for getting my mail. So far, I handled my mail with Netscape, which directly contacted this host for sending and receiving my mail, so Smail configuration was really of no concern. Now I would like to switch to another mailer with more features, such as PGP support, filters for sorting my mail, you name it. So I told Smailconfig I'm an Internet site deepthought (my computer's name) and would like to use the aforementioned stud.uni-goettingen.de as smarthost. I set up fetchmail to get my mail once on startup. My problem lies in getting my mail system to put From: Stefan Baums [EMAIL PROTECTED] at the top of my outgoing mail. (sbaums is the name of my POP3 account.) Now, Netscape (internal movemail for getting mail, localhost as SMTP host for outgoing mail), Mail(to), XMailTool, and Pine work fine. (This message was sent by Pine.) But Mutt and Exmh have From: Stefan Baums [EMAIL PROTECTED]. (stefan is my user name at deepthought.) A perusal of Mutt and Exmh's documentation was unilluminating (to me). As for Smail, I tried a line from_field=From: Stefan Baums [EMAIL PROTECTED] in /etc/smail/config, but it had no perceptible effect at all. Any suggestions short of renaming my computer stud.uni-goettingen.de (wouldn't that be problematic because this name exists already, anyway?) and changing my username from stefan to sbaums? Greetings, Stefan -- Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: From: Field in Mutt and (Ex)mh
On Jul 28, Alexander Koch wrote Quoting Stefan Baums ([EMAIL PROTECTED]): You surely want to put something like my_hdr From: [EMAIL PROTECTED] (Stefan Baums) in your ~/.muttrc - internally Mutt will do the Name change. Nope. No effect. (I tried localsite stud.uni-goettingen.de: no effect either.) I had to admit I can't really believe you... Sorry. Besides, localsites only affects set localsignature=file. Only. What're you doing there, it works like a charm here? Did you quit Mutt to reload the changes? :- Maybe you want to do a set edit_hdrs and edit the From: manually, see if it goes through you mta unchanged. Ahm... sorry. Now it works. (This message is sent by Mutt; nice prog.) I fiddled around a bit with .muttrc, tried to start Mutt from Rxvt and from Afterstep's Wharf, and now it's ok (from both Rxvt and Wharf). No idea why it didn't work at once. Anyway, thanks a lot. Greetings, Stefan -- Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] pgpADKR0mnWeW.pgp Description: PGP signature
Re: From: Field in Mutt and (Ex)mh
Alexander Koch wrote: You surely want to put something like my_hdr From: [EMAIL PROTECTED] (Stefan Baums) in your ~/.muttrc - internally Mutt will do the Name change. Nope. No effect. (I tried localsite stud.uni-goettingen.de: no effect either.) -- Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: why kernel 2.0.29 in Debian 1.3.1?
SCO emulation. Perhaps there are other stories of Ibcs2 or other 2.0.30 failures the Debian users would like to share? # insmod msdos /lib/modules/2.0.30/fs/msdos.o: unresolved symbol fat_get_entry_R1dbb54c9 ... [plus about twenty other unresolved symbol errors] Greetings, Stefan -- Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] pgpxs9uKuwY1f.pgp Description: PGP signature
Re: XF86 3.3-3
Christopher Jason Morrone wrote: I upgraded from 3.2 using the 3.3-3 packages, and now when I try to startx, the screen flickers and goes black, and the keyboard locks up. I'm using the S3 server(#9 motion 771 card), and he's using standard SVGA for his Trident I believe. In any case, the XF86 3.3 packages are definitely broken. I'd fill out a bug report, but unfortunately, I can't be any more specific than that right now. I had the same problem with an S3 Trio 64 V+ graphics card. My flicker (or more nearly pulsing (of xdm repeatedly trying to start?)) was quite nasty, so I quickly rebooted with the rescue disk, deleted /usr/X11R6/bin/xdm, rebooted from the harddisk, and downgraded. Is 3.3 choking on my old configs? Or could it have been the S3 server? Greetings, Stefan -- Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Watchdog Trouble
Hello, I'm running Debian 1.3.0, kernel 2.0.30, and my software watchdog timer is having trouble starting getting started at boot time. The .deb installed all right, my kernel is configured CONFIG_WATCHDOG=y CONFIG_WATCHDOG_NOWAYOUT=y # CONFIG_WDT is not set CONFIG_SOFT_WATCHDOG=m # CONFIG_PCWATCHDOG is not set and ls -l /dev/watchdog says crw-rw 1 root sys 10, 130 May 28 02:49 /dev/watchdog Nevertheless, after startup my system log tells me watchdog[106]: starting daemon (2.1): 10s mla=12 watchdog[106]: cannot open /dev/watchdog watchdog[106]: stopping daemon (2.1) Please note that I _can_ start it manually: watchdog[251]: starting daemon (2.1): 10s mla=12 kernel: Software Watchdog Timer: 0.04, timer margin: 60 sec Any ideas? Greetings, Stefan -- Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Watchdog Trouble
Bruce Perens wrote: It sounds as if the watchdog daemon tries to start before the software watchdog kernel module is loaded. Is watchdog in /etc/modules? Yup, you're right. In /etc/modules, I activated auto (letting kerneld take care of things) and deactivated the rest, rebooted, et voila, it works. 5 minutes from submitting a problem to receiving the answer that solves it is what I call support. Thanks a lot. This is one great mailing list. Stefan -- Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
nethack lock file
Hi all, After installing Debian 1.3.0 (from scratch): $ nethack NetHack, Copyright 1985-1996 By Stichting Mathematisch Centrum and M. Stephenson. See license for details. No write permission to lock /var/lib/games/nethack/perm! Hit space to continue: hitting space, program aborts $ ls -l /var/lib/games/nethack/perm -rw-rw-r-- 1 root games 0 May 27 09:02 /var/lib/games/nethack/perm So what am I to do (chmod o+w perm as root has no effect; deleting perm results in couldn't find lock file perm plus abort)? Greetings, Stefan -- Stefan Baums Universitaet Goettingen mailto:[EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: [tex] Re: babel and 8-bit files
Frederic Dumont wrote:
On 11-Jun-97 Stephan Tassart wrote:
But, when I run latex2e on a 8-bit file (with characters such as é,
ö instead of \'e, \o), I got nothing (for the 8-bit characters). Is
it a misconfiguration from the Debian distribution or a problem which
comes from initex (and in that case, why having compiled a 7-bit
initex instead of a 8-bit?)
simply add
/usepackage[T1]{fontenc}
^---SHOULD BE \
and it will be solved.
Not quite. Unless you add
\usepackage[latin1]{inputenc}
as well, you'll get `SS' instead of the German ß (sz).
--
-
Stefan Baums
Universitaet Goettingen
[EMAIL PROTECTED]
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] .
Trouble? e-mail to [EMAIL PROTECTED] .
Re: Problem getting German Umlaute
Paul Seelig wrote: In article [EMAIL PROTECTED], [EMAIL PROTECTED] (Stefan Baums) writes: It works all right for bash (thanks so far), but not for emacs, or tcsh (which I'd like to use), where the effect is nil. Make sure to load de-latin1.map using the program 'kbdconfig' and reboot (must this really be necessary?). It happened to me twice during some Debian installations that i accidently loaded de-latin1-nodeadkeys.map and therefore the emacs on the target machines showed the same symptoms. Umlaut-problem number 2 (emacs from bash) has been successfully solved: the German-HOWTO solution actually works once you have configured your shell. Remains umlaut-problem number 3: tcsh. The man pages for tcsh do have something to say on the subject, only it (i.e. setting LANG / LC_CTYPE, using setfont / loadkeys) doesn't work. Could anyone mail me the relevant part of his/her tcsh-under-Debian configuration? Greetings, Stefan -- - Stefan Baums Universitaet Goettingen [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Problem getting German Umlaute
Christian Meder wrote: On Jun 8, Stefan Baums wrote I recently installed Debian ... I seem to fail in the task of getting it to display German Umlaute (a, o, u, sz (if your mailer can handle them, they should look like this: ä, ö, ü, ß)) ... Hi Stefan, I haven't checked out yet why my setup does function while the one from the German HOWTO doesn't. Here are the relevant parts of my setup. In /etc/profile or ~/.bash_profile LC_MESSAGES=de_DE LC_CTYPE=iso_8859_1 export LC_MESSAGES LC_CTYPE Note that I didn't set LANG In ~/.emacs (standard-display-european t) Please tell me if this does function for you too ! It works all right for bash (thanks so far), but not for emacs, or tcsh (which I'd like to use), where the effect is nil. Greetings, Stefan --- - Stefan Baums Universitaet Goettingen [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Problem getting German Umlaute
Hello, I recently installed Debian 1.2 on my notebook and everything went fine. Almost everything. While I am duly impressed by Debian's complex package management system and wide range of packages, I seem to fail in the simple task of getting it to display German Umlaute (a, o, u, sz (if your mailer can handle them, they should look like this: ä, ö, ü, ß)), or, for that matter, any accented letters (´e, etc.). At install time I _did_ indicate, when prompted, a German keyboard. I read the German-HOWTO and tried setfont lat1u-16.psf and loadkeys de-latin1, with the result that pressing the Umlaut keys produced beeps instead of funny characters, or vice versa. All the above is true of the text console; I haven't installed X. Funny thing is, when supposed to enter my user name at the login prompt, I _do_ get correctly displayed Umlaute on pressing their keys; then, of course, login tells me it didn't recognize my spelling exercise as a valid user; and on second trial at the login prompt it doesn't work anymore; neither does it at the console (I tried tcsh and bash); and emacs, too, doesn't recognize Umlaute (and yes, I did add the German-HOWTOs lines to my .emacs-file). Updating to Debian-1.3-development (from March) didn't do any good, either. Any help out there? PS. My Umlaute are rendered as follows: pressing a results in d, o in v, u in |, sz in _, ´e in i, etc. At a loss, Stefan -- - Stefan Baums Universitaet Goettingen [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
