Re: Huge font
On Fri, 16 Jun 2000, Glen Sawtell [EMAIL PROTECTED] wrote: I am a new debian user and I have a problem which none of my debian-using friends have come across. It seems that some apps like netscape, GIMP, spruce etc seem to be using a large size helvetica font for the text on buttons and input fields etc. It's quite annoying and I don't think it's normal, is there a font package I am missing or something similar?. Please help. There are 75dpi and 100dpi font packages, you should probably install xfonts-75dpi and rearrange the FontPath entries in /etc/X11/XF86Config to put the path to the 75dpi fonts first. -- Philip Lehman [EMAIL PROTECTED]
Re: Flirting with diaster (a bad installation experience)
On Wed, 24 May 2000, Andy Krietemeyer [EMAIL PROTECTED] wrote: With the DOS tool, fdisk, I deleted a little used DOS partition from my 8.4 gb hard drive to provide ~2 gb freespace and rebooted the Slink rescue floppy into the installation program. As part of the Debian installation, I divided the hard disk free space into 70 mb linux swap (hda4) and 1843 mb linux primary (hda3) partitions. I then installed and configured the base system - uneventfully. I did not select the option 'boot linux directly from the hard drive' - was I under the correct impression that doing so would prevent me from booting Win98? No, not really. For the very first boot after installing lilo, this is true. But all you have to do then is add a record for Windows to /etc/lilo.conf and run lilo (to update the MBR). On the second reboot you will be able to choose either OS (this is what lilo is for, after all). I used dselect to install and configure packages, I configured X. Everything seemed alright. I did not (and don't how I would) install or configure LILO. Is that where I went wrong? I don't recall seeing mention of LILO in the installation dialogs. You probably were almost there. Use lilo, really. It won't mess up your Windows partitions and if you really panic, you can always (like you did) boot from a rescue disk, write a plain MBR and set the bootable flag on your Windows partition. There is no need to actually delete the Linux partitions. I rebooted the machine, thinking it would boot Win98 (and I would have to (temporarily) boot from the Linux boot floppy to get into Linux). No OS booted. An error message along the lines of 'Set up cannot continue, invalid partition' appeared. Can you provide more details of your partitioning sheme? And is your HD set to LBA in the BIOS (should be, but double check)? Perhaps I overreacted, got stupid, or was lucky or unlucky, but I rebooted from a Win98 rescue disk I had created and used DOS fdisk to delete the two linux partions to which I had just installed Linux. I then tried a reboot, and thankfully, my Win98 booted and seems none the worse for the experience. I'd appreciate any insight anyone might have. It may take a while before I get the courage to try reinstalling Debian on a dual OS machine. Try it again right away, good learning experience ;) But first, read the relevant HOWTOs at http://www.linuxdoc.org/HOWTO/; especially: http://www.linuxdoc.org/HOWTO/mini/Linux+Win95.html (obviously) http://www.linuxdoc.org/HOWTO/mini/Multiboot-with-LILO.html (more comprehensive than the first one) http://www.linuxdoc.org/HOWTO/Large-Disk-HOWTO.html (background and why you want to use LBA) There might be even more, just look around. -- Philip Lehman [EMAIL PROTECTED]
Re: ghostscript message
On Thu, 25 May 2000, Agner-Nichols [EMAIL PROTECTED] wrote: forgot to mention that I installed ghostscript as a .deb; went back to the Packages for Slink and see that I should have installed gsfonts -- thanx Philip, that was a nice nudge I was under the impression that you wanted even more additional fonts, well, I do. Didn't mean to sound rude, either. ;) -- Philip Lehman [EMAIL PROTECTED]
Re: Howto find out why a connection is opened....
On Thu, 25 May 2000, Ron Rademaker [EMAIL PROTECTED] wrote: See subject. I'm using a ISDN connection. ...with isdn in autodial mode I guess. Try setting up ipchains and log all output over ippp0. /sbin/ipchains -A input -j ACCEPT -i ippp0 -l Then watch your syslog and try to guess from IP/port/protocol which process might trigger the dialout. See the ipchains HOWTO. Tracking this down is a science. Good luck! -- Philip Lehman [EMAIL PROTECTED]
Re: Howto find out why a connection is opened....
On Thu, 25 May 2000, Philip Lehman [EMAIL PROTECTED] wrote: On Thu, 25 May 2000, Ron Rademaker [EMAIL PROTECTED] wrote: See subject. I'm using a ISDN connection. ...with isdn in autodial mode I guess. Try setting up ipchains and log all output over ippp0. /sbin/ipchains -A input -j ACCEPT -i ippp0 -l ^ This should have read output, sorry. -- Philip Lehman [EMAIL PROTECTED]
Re: Cannot find libz1 and libxpm4
On Wed, 24 May 2000, Kerstin Hoef-Emden [EMAIL PROTECTED] wrote: I guess, my problem is caused by the fact that I use dpkg to investigate dependencies and to install packages and not apt-get (I must admit, that I positioned it low priority on my installing list. Since setting it up didn't work right from the beginning, I skipped it for later experiments. As a last ressort, you can always consult the search engine at http://www.debian.org/distrib/packages (- Search Package Directories) This interface will list all dependencies. Even if you don't download the packages from there (apt is certainly more convenient), it might be helpfull as a means of orientation, especially if you're still exploring the load of packages which are available. -- Philip Lehman [EMAIL PROTECTED]
Re: ghostscript message
On Wed, 24 May 2000, Agner-Nichols [EMAIL PROTECTED] wrote: When running gv (as a replacement for ghostview), I get a Ghostscript message '/invalidfont in findfont' (specifically, Times New Roman). Ad you don't get that with ghostview? Both are just frontends to gs and this is an error message straight from gs itself. Any ideas about where to get additional fonts and how to make them available for ghostscript? (running debian slink, kernel 2.0.36 on a cyrix 486 dx). Put the font files in a directory of your choice, add that to the GS_LIB environment variable, export GS_LIB=/path/to/the/fonts:$GS_LIB and create a 'Fontmap' file in this directory. Format looks like this: % postscript name font file /TimesNewRomanPSMT (times.ttf) ; /TimesNewRomanPS_BoldItalicMT (timesbi.ttf) ; /TimesNewRomanPS_BoldMT (timesb.ttf); /TimesNewRomanPS_ItalicMT (timesi.ttf); See /usr/lib/ghostscript/fonts/Fontmap for examples. How to find out about the ps name depends on the type of font. For type1, there is 'type1inst' (this will create a Fontmap file for you). For truetype, the fastest way I'm aware of is using ttf2afm (in the freetype package) and grep the name from the afm: ttf2afm font.ttf | grep ^FontName -- Philip Lehman [EMAIL PROTECTED]
Testing firewall chains
I have set up firewalling on a gateway which connects my LAN to the net (using ipchains with a 2.2.14 kernel). How could I test my firewall chains from outside? Are there public services out there which would do a port scan on my server, for example? I'm pretty new to advanced networking and after reading the HOWTOs and the ipchains docs I think I understand how it works, but I still feel a bit insecure. Any pointers to more documentation are welcome too. -- Philip Lehman [EMAIL PROTECTED]
Re: Mouse doesn't work in Xwindow, NFS, bug(3.)???
On Mon, 22 May 2000, Justin Megawarne [EMAIL PROTECTED] wrote: And Philip Lehman bestowed upon us all: On Mon, 22 May 2000, Antonio Rodriguez [EMAIL PROTECTED] wrote: prw-r--r--1 root root0 May 22 23:01 /dev/gpmdata If you don't have any gpmdata fifo, create one with `mknod /dev/gpmdata p` and try again. I thought gpm automatically created the FIFO? Or am I wrong? Yes, looking at the time stamp it seems like it even does so everytime it is launched. But it won't do any harm to double check ;) -- Philip Lehman [EMAIL PROTECTED]
Re: Simple Text Editor with Synatx highlighting?
On Tue, 23 May 2000, Armin Wegner [EMAIL PROTECTED] wrote: Use vim. It's hard to learn, but it is worse it. Interesting typo ;) (I'm writing this as a non-native speaker, no offence intended) -- Philip Lehman [EMAIL PROTECTED]
gs, wmaker debs? (was: Re: XFree86 4.0 debs ?)
On Tue, 23 May 2000, James Sleeman [EMAIL PROTECTED] wrote: I would like to install XFree 4.0 on my potato machine, are there debs available yet, or not too far off ? I couldn't even see 4.0 ones in Woody. Take a look at http://www.debian.org/~branden/plans.txt. It seems like the move to the new architecture is non-trivial (haven't tried it myself). A similar question: I've been looking for debs of gs 6.0 and wmaker 0.62. Both are out for quite some time now, but they're not even in woody. Is this because all maintainers are busy getting potato out? I've built wmaker and libproplist myself and it works right out of the box on potato. Merely a matter of half an hour (not flaming or anything, I'm just curious). -- Philip Lehman [EMAIL PROTECTED]
Re: power saving on my desktop...
On 23 May 2000, Matthew Wilson Emmett [EMAIL PROTECTED] wrote: I'd like to have my desktop save power after 30min of inactivity. I've managed to do this before, but I lost my overnight cron jobs, and the clock always got skewed. Do you want that when you're in X or on the console? Three things to consider: 1. Configure and activate APM in your BIOS, compile kernel with APM support (you probably did that already). 2. Install the apmd package (no need to actually run the apm*d*, all you need is apm) and run 'apm -s' or 'apm -S' as needed. Is that what you used a cron job for? Would be unusual. 3. Configure X properly (set timeouts and define power_saver) and run 'xset dpms' from .xinit. -- Philip Lehman [EMAIL PROTECTED]
Re: can't print
On Tue, 23 May 2000, Pollywog [EMAIL PROTECTED] wrote: I have never set up printing in Linux. I just added a printer (Epson Stylus Color 740) via parallel port and I have no idea where to begin. I installed apsfilter and then magicfilter and apsfilter indicates that /dev/lp0 /dev/lp1 etc are not installed, yet I can see they are installed. Forget about aps/magicfilter for the moment. This will come in later. The Printing HOWTO is just confusing me more. What do I need to install? 1. Recompile kernel with parallel port support. The kernel should detect you printer at boot time or when the module (lp) is loaded and log a meaningfull message to syslog. 2. Install lpr 3. Edit /etc/printcap (man printcap). Here's an example: lp:\ :lp=/dev/lp0:\ :sd=/var/spool/lpd/lp:\ :af=/var/log/lp-acct:\ :lf=/var/log/lp-errs:\ :sh:mx#0: 4. See if you can print *something*; e.g. try 'ls -l | lpr'. This should give you some ugly stairs-like output (that's a normal LF/CF issue). You might have to add your user account to the lp group first IIRC. 5. *If* this works, start reading the aps/magicfiter docs. Magicfilters are usually inserted in the print chain as input filters of lpr (I never used apsfilter, but I assume it works more or less like magicfilter). Thus you would add a line like this :if=/usr/local/share/magicfilter/lj5:\ to the printcap entry for your printer. Of course it has to point to a magicfilter sript suitable for your modell. I hope I didn't omit anything essential. Helps? -- Philip Lehman [EMAIL PROTECTED]
Re: Minor Samba issues
On Mon, 22 May 2000, Lindsay Haisley [EMAIL PROTECTED] wrote: The problem turned out to be in /etc/printcap. The default printer pulled up in the share is 'lp', which is basically a config for an old-fashioned dot matrix printer. It has no 'sf' entry. Make sure the printcap entry defining your share contains 'sf', which is 'supress form feed'. Adding 'sf' to the printcap entry fixed the problem for me. Yes, that's it. I didn't consider the server's printcap because printing on the server locally worked fine. Appearantly I have been getting something wrong.
Re: Simple Text Editor with Synatx highlighting?
On Mon, 22 May 2000, Frank Mehnert [EMAIL PROTECTED] wrote: I'm using joe for many purposes. The only feature I miss is syntax highlighting in C. Are there any hints for such a small and fast editor in Debian with that feature built in? jed? -- Philip Lehman [EMAIL PROTECTED]
Re: Mouse doesn't work in Xwindow, NFS, bug(3.)???
On Mon, 22 May 2000, Justin Megawarne [EMAIL PROTECTED] wrote: On Sun, May 21, 2000 at 06:22:23AM -0400, Antonio Rodriguez wrote: A fresh installation using the boot floppies in frozen directory (ftp.debian...), dselect + apt-get used to continue installation. Problems: 1. Mouse (/dev/psaux, ps2 with wheel) doesn't work after setting it up through xf86config; it had always worked before. Mouse works in console, detected by gpm with correct settings. Guessing: Since I had installed imwheel, I thought it could have messed up the settings, removed imwheel, but still doesn't work. Don't know what to do. Yes, I had this too ... it seems you have to `gpm -k` or `killall -9 gpm` for everything to work happily, which is a PITA, to say the least ... the ps2 repeating is reported not to be supported, which I think is a total lie =P If it works after killing gpm, you have a repeating problem. I'm not sure how to fix that side of things. Anyone out there? =) Don't know if I'm getting something wrong here, but all you need to do is have X read from gpm's fifo, i.e. In /etc/X11/XF86Config: Section Pointer ProtocolMouseSystems Device /dev/gpmdata ... EndSection In /etc/gpm.conf: device=/dev/psaux type=ps2 responsiveness= append1=-R# -R for repeat/echo to /dev/gpmdata If you don't start gpm from the stock /etc/init.d/gpm script, just make sure you add -R to the command line. -- Philip Lehman [EMAIL PROTECTED]
Re: Mouse doesn't work in Xwindow, NFS, bug(3.)???
On Mon, 22 May 2000, Antonio Rodriguez [EMAIL PROTECTED] wrote: I tried that and didn't work, so I just dpkg -r gpm. This is not the best solution, but what else??? Check your config step by step (I'm sure it works because I have a similar setup). Does gpm work on the console? If so, make sure you have the -R flag set and check if the fifo is there. What does `ls -l /dev/gpmdata` say? This should be something like prw-r--r--1 root root0 May 22 23:01 /dev/gpmdata If you don't have any gpmdata fifo, create one with `mknod /dev/gpmdata p` and try again. Helps? -- Philip Lehman [EMAIL PROTECTED]
Minor Samba issues
I just installed Samba on a small server to act as a file and print server for a Linux (printing only) and a Windows client (both). Installing and configuring everthing took about ten minutes, SWAT is amazing! So basically everything works like it is supposed to, but there are still some minor issues: (1) Codepages nmbd complains that it can't find the codepage file /etc/samba/codepages/codepage.850. While browsing the docs and the man pages, I found the manpage for make_smbcodepage. As far as I see I have to construct the codepage files required by Samba from some text or binary templates. But where do I get the templates from? It seems like they don't come with the DEB and I didn't find anything on ftp.samba.org either. The manpage for make_smbcodepage talks about source/codepages, does this mean I have to get the source tarball? 2. Printing and extra blank pages When printing from either client, I get an extra blank page after each print job. This is mentioned in the docs but I'm only advised to check the printer config on the server and the settings on the clients, whatever that means. Printing on the server locally with lpr -Plp (the command used by Samba as well) does not produce any extra pages, so I assume the lpd setup on the server side is fine. (a) On the Windows box, the printer driver offers to send an extra separator page, but this option is turned off. Not a Samba issue at this point, but maybe someone has faced this before? (b) On the Linux box, the setup is harder to debug; /etc/printcap uses a magicfilter script as input filter, which basically does the usual PS to PCL conversion (the printer is a non-PS HP LaserJet 5L) and pipes its output to a script which adds print - and pipes on to smbprint. If I pipe a text file straight to the smb print script: no extra page. If I connect the printer to the Linux box and print locally using the same magicfilter file: no extra page. If I even connect the printer to the windows box and use that as a print server for the Linux box (same configuration on the Linux box, all I do is change the printer name): no extra page. Huh? Some configuration details for the Linux client are included below. I would appreciate your thoughts and hints. TIA ### /etc/printcap smb:\ :lp=/dev/lpnull:\ :sd=/var/spool/lpd/smb:\ :if=/usr/local/share/magicfilter/lj5smb:\ :af=/var/log/lp-acct:\ :lf=/var/log/lp-errs:\ :sh:pw#80:pl#72:px#1440:mx#0: ### /usr/local/share/magicfilter/lj5smb (excerpt) [magic] gs [gs options] -sOutputFile=- - | /usr/local/share/samba/smbprint ### /usr/local/share/samba/smbprint ( echo print - cat ) | /usr/bin/smbclient \\server\hp -P -U nobody -N -- Philip Lehman [EMAIL PROTECTED]
kernel: compiler error in function gen_rtx_combine
I'm trying to recompile a 2.2.14 kernel. Compiling fails with the following error msg: gcc -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strict-aliasing -pipe -fno-strength-reduce -m486 -DCPU=486 -c -o namei.o namei.c ../../gcc/combine.c:9305: Internal compiler error in function gen_rtx_combine cpp: output pipe has been closed make[2]: *** [namei.o] Error 1 make[2]: Leaving directory `/usr/src/kernel-source-2.2.14/fs' make[1]: *** [first_rule] Error 2 make[1]: Leaving directory `/usr/src/kernel-source-2.2.14/fs' make: *** [_dir_fs] Error 2 I've also noticed that make segfaults sometimes when I try a make menuconfig. -- Philip Lehman [EMAIL PROTECTED]
weird nfs mount problem
I posted this before but didn't get a reply. Let me give it another try and attempt to elaborate a bit more since I fail to solve this on my own and I have no idea where to even look for the cause. On a potato machine acting as a NFS client, I can't mount any NFS volumes. The mount process just hangs and I have to kill it. While I was looking around for the cause, I noticed some error messages that show up in syslog when lockd is launched form /etc/init.d/nfs-common. This is what happens every time I restart nfs-common: May 8 16:55:30 minerva kernel: lockd_up: no pid, 2 users?? May 8 16:55:30 minerva kernel: SIG: sigpending lied May 8 16:55:39 minerva /sbin/rpc.statd[9045]: Caught signal 15 [...] # restarting May 8 16:56:36 minerva kernel: lockd_up: no pid, 3 users?? May 8 16:56:36 minerva kernel: SIG: sigpending lied ...and so forth. The user count increases by one every time I restart the daemon. When I stop nfs-common and try do to an NFS mount, the mount process still hangs and the volume isn't mounted, but the server claims the mount succeeded and if I run `showmount servername` the client shows up in the list. I browsed the man pages and docs related to NFS and didn't notice anything obvious. I have to bad feeling that the information given here is pretty chaotic, but I don't know how to track this down. On another client running and the very same nfs package everything is fine, so I assume this is not a problem on the server side. This stuff used to just work when I installed nfs-common, put the server volumes in /etc/fstab and added the client to the server's /etc/hosts.allow. But not on this particular client. I will gladly provide more information if neccessary, it's just that I don't know where to search. TIA -- Philip Lehman [EMAIL PROTECTED]
NFS: sigpending lied?
When mounting from a NFS server, I get the following error message on a client running potato whenever I access any NFS mounted volume: kernel: SIG: sigpending lied I can mount, but when I try to unmount the mount process hangs. This happens with one client only, the others are fine. No obvious problems on the server side either with this particular client. All machines (server and clients) are running potato. I don't have the faintest idea what this error msg means. Any hints appreciated. -- Philip Lehman [EMAIL PROTECTED]
lilo mbr on hda with linux on hdc?
I have a machine with two hard disks here, that other OS resides on hda and I added the second drive (hdc) for Debian. The setup installed a MBR on hdc, but my BIOS only supports booting from hda. Is there a way to install lilo in the MBR on hda and have it boot linux from hdc1? I'm familiar with dual-boot setups in general, but I'm a bit clueless facing this particular setup. Any hints appreciated... -- Philip Lehman [EMAIL PROTECTED]
Re: lilo mbr on hda with linux on hdc?
On Thu, 4 May 2000, Oswald Buddenhagen wrote: I have a machine with two hard disks here, that other OS resides on hda and I added the second drive (hdc) for Debian. The setup installed a MBR on hdc, but my BIOS only supports booting from hda. Is there a way to install lilo in the MBR on hda and have it boot linux from hdc1? lilo.conf: change the boot=/dev/hdc into boot=/dev/hda image=/dev/hdc1 should be already there and it is fine. and an option other=/dev/hda1 is needed for booting the other os. I tried that, but LILO hangs with LI-. I did run lilo to update the MBR - several times. In lilo.conf I have: boot=/dev/hda root=/dev/hdc1 ... image=/vmlinuz ... other=/dev/hda1 ... -- Philip Lehman [EMAIL PROTECTED]
Re: lilo mbr on hda with linux on hdc?
On Thu, 4 May 2000, matthschulz wrote: On Thu, 04 May 2000, Philip Lehman wrote: On Thu, 4 May 2000, Oswald Buddenhagen wrote: I have a machine with two hard disks here, that other OS resides on hda and I added the second drive (hdc) for Debian. The setup installed a MBR on hdc, but my BIOS only supports booting from hda. Is there a way to install lilo in the MBR on hda and have it boot linux from hdc1? lilo.conf: change the boot=/dev/hdc into boot=/dev/hda image=/dev/hdc1 should be already there and it is fine. and an option other=/dev/hda1 is needed for booting the other os. I tried that, but LILO hangs with LI-. I did run lilo to update the MBR - several times. In lilo.conf I have: [...] What does lilo say when You run it? Nothing unusual: [EMAIL PROTECTED] ~lilo -v LILO version 21, Copyright 1992-1998 Werner Almesberger Reading boot sector from /dev/hda Merging with /boot/boot.b Boot image: /vmlinuz Added Linux (alias lnx) * Boot other: /dev/hda1, loader /boot/chain.b Added Windows (alias win) /boot/boot.0300 exists - no backup copy made. Writing boot sector. -- Philip Lehman [EMAIL PROTECTED]
Re: how do you set your system clock from a remote time server?
On Tue, 25 Apr 2000, Harry ten Berge [EMAIL PROTECTED] wrote: Chris Mason wrote: Where can I get nptdate? It's in 'xntp3'. ntpdate is a part of the complete 'time server suite'. At least in potato it's broken up into server and client. There is a ntpdate (client only) deb. -- Philip Lehman [EMAIL PROTECTED]
Re: root filesystem corrupted, can't log in
On Tue, 25 Apr 2000, Ethan Benson [EMAIL PROTECTED] wrote: On Mon, Apr 24, 2000 at 02:47:37PM -0700, montefin wrote: It seems the linux kernels from about 2.2.7 thru 2.2.13, and only on IDE boxes, had a recurrent filesystem corruption problem. I was on kernel 2.2.12 when I upgraded to RedHat 6.1 and it slaughtered me. Here's the routine that worked over and over again for me when boot up failed. i can certainly attest to 2.2.13 trashing filesystems, it was always `corrupted blocks' or some such thing too. snip When they are all done, type 'exit' (without the quote marks). Your system will shutdown and you _should_ be able press restart and boot back up clean. Should. Did. Nathan, montefin, aphro, Ethan: thanks to all of for the quick replies. I managed to fix the filesystem some time after midnight (CET) and the box it's up and running again. [...] you will need it. and don't allow any 2.2 kernel earlier then 2.2.14 within a mile of any of your machines :| First thing I did when the machine came up again was upgrading from 2.2.9 to 2.2.14. This is scary. BTW: What's the best way to keep up with glitches like this if you don't follow devel and kernel lists? -- Philip Lehman [EMAIL PROTECTED]
root filesystem corrupted, can't log in
My potato workstation suffered from a power failure and it seems like the partition holding the root filesystem was damaged. When I boot, fsck forces a check and reports an error about duplicate blocks (I'm sorry that I can't provide the precise error messages, but there is no way to catch them). fsck tries to fix the problem but it fails, saying something about an unexpected inconsistency (IIRC) and tries to dump me to a sulogin prompt. At that point I get an error about /sbin/sulogin not beeing found (I guess because the file system is not mounted yet) and the machine simply reboots. If I don't turn it off when the BIOS comes up, it keeps on looping. I'm desperate because the same happens when I boot from a rescue disk. What can I do about that? Any help will definetly be very much appreciated... -- Philip Lehman [EMAIL PROTECTED]
Re: 3 questions. Please, HELP ME!
On Sat, 1 Apr 2000, José María Pongilioni López [EMAIL PROTECTED] wrote: Hi! I'm a Debian Linux 2.1 'slink' user, and I have 3 questions: 1. I have compiled the version 2.2.4-intl of Linux Kernel and, at boot time, my screen shows the following message: SIOCADDRT: Invalid argument I have read the documentation, and says that is a bug in pppd versions 2.2.0 and earlier. I have currently installed pppd v2.3 patch level 5. Why does occur this? How can I patch the kernel? I *think* this occurs when running a 2.2.x kernel with slink, because the /etc/init.d/network script tries to add a route to the loopback interface (lo). This is not neccessary any more with a 2.2.x kernel. Look for a line starting with route add ... in /etc/init.d/network, comment it out and see if that helps. 2. [...] 3. I have the defrag utility. How can I defragment my Linux partition, since a mounted drive cannot be defragmented? There is no need to 'defrag' ext2 filesystems, you don't have to worry about that. -- Philip Lehman [EMAIL PROTECTED]
ip masquerading
I'm trying to set up IP masquerading on a slink/potato box which is supposed to route the traffic on my home LAN over an ISDN dial-up line. I have to admit that I have no experience with advanced networking of this kind. I read the IP masquerading HOWTO. It suggests a sample rc.firewall script to set up masquerading and simple firewalling. It appears to me that this interferes with the /etc/init.d/* scripts used by related Debian packages, and I'd rather do it the Debian way. rc.firewall wants to run: # echo 1 /proc/sys/net/ipv4/ip_forward # echo 1 /proc/sys/net/ipv4/ip_dynaddr I haven't found this in any other script in /etc/init.d/*. What's the default way to do this? Write my own script? And it wants to run: # /sbin/ipfwadm -F -p deny # /sbin/ipfwadm -F -a m -S 192.168.0.0/24 -D 0.0.0.0/0 I guess this is what /etc/init.d/ipmasq is for, but I'm feeling lost as far as the configuration is concerned. The postinstall script asked for the client IPs on the LAN and I entered that, but where is this stored? Do I have to do anything in addition to that, or can I rely on the defaults? I don't need anything fancy, but the setup should be halfway secure. TIA -- Philip Lehman [EMAIL PROTECTED]
Re: Getting xfs-xtt to work
On Sun, 6 Feb 2000, Bart Szyszka [EMAIL PROTECTED] wrote: Try xset +fp unix/:7100 or something similar. xfs-xtt use port 7100, same as xfs. I really have no idea what I'm doing so I'd appreciate it if you guys would be more specific. What's similar What should I do same as xfs? Like I said, I installed xfs-xtt. I don't have xfs (it asks me to remove xfs-xtt if I try to install xfs). Don't I need to do something like this: xset +fp /usr/local/share/fonts/truetype No, you're getting something wrong here. Your X server doesn't need to know the path to the font files, it needs to know were to find the font server who in turn needs info about the path to the actual files. Thus the path definition for the xserver could look like unix/:7100, which basically means font server on localhost, port 7100. You want to put this in the Files section of your /etc/X11/XF86Config file; mine looks like this: Section Files RgbPath/usr/X11R6/lib/X11/rgb FontPath unix/:7100 EndSection The xset stuff is only for testing purposes, since the xserver will refuse to start if something goes wrong with the fonts. When I do I get this: xset: bad font path element (#38), possible causes are: Directory does not exist or has wrong permissions Directory missing fonts.dir Incorrect font server address or syntax This is probably because there is no fonts.dir file. You don't have to worry about this as far as the xserver is concerned as it won't access the files but query the font server instead. But the font server might need this file. xfs and xfsft do, xfstt doesn't, I don't know about xfs-xtt. There should be some instructions that tell you what files you need. -- Philip Lehman [EMAIL PROTECTED]
Re: Fonts too big after getting xfs-xtt to work
On Sun, 6 Feb 2000, Bart Szyszka [EMAIL PROTECTED] wrote: I got xfs-xtt working (thanks!), but I'm having some related problems. Now in some programs, the fonts are too big and I can't change them. My Kpanel's taskbar and menu text fontsize is too big. Changing it to some huge size or some small size or even a completely different font does nothing. It stays at what looks like a 14px Arial/Helvetica no matter what. Any ideas? Is this limited to kde? Then you should launch kfontmanager and make sure that the installed fonts and the fonts used by kde are in sync. Otherwise make sure the path setting used by the font server is sane, order does matter if you have different versions of one font. Put the dirs with the best fonts first. -- Philip Lehman [EMAIL PROTECTED]
Re: [*] about x
i had used redhat for some time, and now i turn to debian(2.1r4).i found my debian run xdm auotmatic, i don't like this, how can i change it? when use redhat,edit the file /etc/initab, and set init 3 to default will be ok, but debian is quite different. If I remeber correctly, there is an option in /etc/X11/config; something like run-xdm. Change that to no-run-xdm or put a '#' on front of it. -- Philip Lehman [EMAIL PROTECTED]
weird scrolling problem
I just upgraded a few packages to the latest potato versions; this included libncurses, libgtk, all the XFree86 stuff (3.3.5 - 3.3.6), and a few others. Now scrolling in almost all X based text editors has become *extremely* slow. PageDown works as usual, but using the arrow keys is painful. This affects all X editors (like gvim, yudit, kwrite) except xemacs. I have no idea what is going on here, the only upgrades that seem relevant to me are those of xlib6g and xserver-svga. I will gladly provide more information if necessary, I just don't know what might be relevant at the moment. Any help appreciated... -- Philip Lehman [EMAIL PROTECTED]
Re: New versions of packages
On Sat, 5 Feb 2000, Joseph A. Martin [EMAIL PROTECTED] wrote: I noticed recently that AbiWord, ssh, and alsa all have newer versions released than what is in potato. Is this normal? Are newer versions of packages not uploaded once a distribution is frozen? That's why they call it frozen ;) Only bug fixes will be included. -- Philip Lehman [EMAIL PROTECTED]
Re: Setting up Debian - II
On Wed, 2 Feb 2000, davidturetsky [EMAIL PROTECTED] wrote: I followed Joe's recommendation and partitioned my 32gb IBM hard drive as follows: c: 6,997.0 mb / 39.2mb swap 258.8mb /usr 5,004.6 mb e: 20,332.2 mb Is this a typo or are you trying to reserve 39.2 megs for the root system? Keep in mind that this partition will contain everything below / except /usr, so this setup would be unusable. -- Philip Lehman [EMAIL PROTECTED]
iso8859-15 input
I would like to be able to type iso8859-15 characters in an X app, to be more specific, I'm after the french oe and OE ligatures which are not in the standard latin1 map. The display part of the whole thing should be ok, I'm running xfsft as font server, some suitable fonts are installed and can be displayed by xfd. But the input part doesn't work. I use xmodmap to map the two characters just like I do for ae and AE, but it doesn't work for characters specific to 8859-15. That is, mapping the characters seems to work (as far as xmodmap is concerned), but I still can't type them. xemacs for example complains about oe being not defined and xev says XLookupString gives 0 characters. I'm don't quite understand wether this is a XF86 only issue or a problem with both X and the specific app. Any clarification and all hints appreciated. Oh, and I'm on Slink/Potato with XF86 3.3.5. TIA -- Philip Lehman [EMAIL PROTECTED]
Re: x11amp/xmms
On Sat, 22 Jan 2000, Wakko Warner wrote: I just have one problem with xmms. I can't find the skin from x11amp which is what's keeping me from upgrading the one copy I have on my machine (I don't like the default skin and I've yet to find one I like other than the x11amp one). Does anyone happen have it or know anyone who has it? I'd love to get a copy. Hint: XMMS supports winamp 1.x and 2.x skins. There are serveral hundred skins available for download at www.winamp.com, I'm sure you'll find something that makes you upgrade ;) -- Philip Lehman [EMAIL PROTECTED]
mounting zip disks and nfs
I'm on a small home lan with a print/file server which has a zip drive connected to it; the server is running slink/potato. Mounting the zip drive and exporting it via nfs works fine but changing disks is rather inconvenient since I have to telnet into the box and unmount/remount the drive on the server. Is there an automount solution out there that would facilitate this? What I have in mind is automatic unmounting of the drive when there is no nfs connect and mounting it as soon someone requests it over nfs. Any recommendations? -- Philip Lehman [EMAIL PROTECTED]
Re: mounting zip disks and nfs
On 21 Jan 2000, Roland Mas wrote: Philip Lehman [EMAIL PROTECTED] writes: Is there an automount solution out there that would facilitate this? What I have in mind is automatic unmounting of the drive when there is no nfs connect and mounting it as soon someone requests it over nfs. Any recommendations? Yeah, like, an auto-mount daemon? Like, something like AMD? Hmm... looks good on first sight, I'll try it out tomorrow. Thanks! -- Philip Lehman [EMAIL PROTECTED]
Re: help
On Fri, 21 Jan 2000, Emmanuel wrote: hello I'm trying to add a gateway on a debian but there's no man on the computer. man route, the option is called -gw. The NET-3-HOWTO decribes the whole procedure in more detail. HTH -- Philip Lehman [EMAIL PROTECTED]
Re: Upgrading pcmcia-* breaks ppp
I recently upgraded my laptop to Linux 2.2.14 (from 2.0.36) with the latest pcmcia-* and modutils, too. I compiled both the kernel and the pcmcia-source myself. I don't know if the kernel upgrade is significant or the upgrade of pcmcia-*. I don't think it's the kernel. I used to run 2.2.9 (and I still do) without problems before the pcmcia-* upgrade. (I can't compile pcmcia-source for Linux 2.0.36 as my compiler is too new. Perhaps I should try and compile pcmcia-* from slink with Linux 2.2.14??? Will that work???) Don't know. Maybe I'll give it a try. There's another odd thing: the dependencies of the pcmcia-modules-2.2.9 package (for my 2.2.9 kernel) indicate that it depends on pcmcia-cs 3.0.14 and *only* 3.0.14. I didn't use the pre-built modules but compiled the pcmcia modules from source. But the source is 3.1.8, could that be the problem? Are there some pcmcia-source packages 3.1.* (but still potato) available? I didn't find any on the debian FTP server. Do you keep them around? I have a xircom combo Ethernet/28.8k Modem card. It works find with Linux 2.0.36 and pcmcia-* from slink. (sidenote: I have noticed recently that this card runs very hot - is this normal?) All I can tell you is that all of my pcmcia cards do that as well and I haven't observed any problems. The Ethernet card part works fine. lehman Jan 15 19:16:30 sirius pppd[352]: pppd 2.3.5 started by lehman, uid 1001 lehman Jan 15 19:16:30 sirius pppd[352]: tcgetattr: Input/output error(5) lehman Jan 15 19:16:31 sirius pppd[352]: Exit. I haven't tried pppd myself, but minicom cannot talk to the modem (no errors, just nothing displayed). lehman ...just as if the modem wasn't inserted at all. But when lehman inserting the modem, the output of cardmgr looks ok: lehman Jan 15 19:17:28 sirius cardmgr[109]: initializing socket 2 lehman Jan 15 19:17:28 sirius cardmgr[109]: socket 2: Serial or Modem lehman Jan 15 19:17:28 sirius cardmgr[109]: executing: 'insmod lehman /lib/modules/2.2.9/pcmcia/serial_cs.o' lehman Jan 15 19:17:28 sirius kernel: tty02 at 0x17f8 (irq = 5) is a 16550A lehman Jan 15 19:17:28 sirius cardmgr[109]: executing: './serial start ttyS2' I get that too. Curiously, with the slink pcmcia-* packages, the modem is assigned tty03, not tty02 - could that be significant? I don't remember that the tty assigned to the card changed, but then again, I didn't upgrade from slink but potato - latest potato. lehman And setserial says: lehman [lehman] ~sudo setserial -a /dev/ttyS2 lehman /dev/ttyS2, Line 2, UART: 16550A, Port: 0x17f8, IRQ: 5 lehman Baud_base: 115200, close_delay: 50, divisor: 0 lehman closing_wait: 3000, closing_wait2: infinte lehman Flags: spd_normal skip_test I get similar results from setserial -a /dev/modem, but setserial -a /dev/ttyS2 just returns an error: /dev/ttyS2: pcmcia controlled device (/var/run/stab) Stale lockfile? lehman I checked /etc/pcmcia.conf and /etc/pcmcia/serial.opts and lehman didn't find anything that looks offensive to me. What's lehman going on here? I don't know how to track this down, any help lehman appreciated... Also, this isn't the worst of my problems... If I try to bootup with the card in, or shutdown, the computer will lock up, and require a hardware reset in order to fix the problem. Inserting/removing the card after bootup seems to be OK though. When I asked about this on the linux-kernel mailing list, somebody suggested I replace the following parameter to my pcmcia.conf file with: PCIC_OPTS='pci_csc=0' however, that doesn't help :-(. -- Sent through Global Message Exchange - http://www.gmx.net
Re: Upgrading pcmcia-* breaks ppp
On Sun, 16 Jan 2000, Christopher S. Swingley wrote: The other poster has suggested removing setserial; is there anything bad that can result from it (i.e. what does setserial do on a laptop)? The reason the poster suggested that is because by default setserial will store and reload serial port configuration information before PCMCIA card services tries to load a serial port associated with the modem card you insert. Often what happens is setserial assigns /dev/ttyS0 - /dev/ttyS3 and irqs 3 and 4. When card services tries to allocate resources and interrupts to the PCMCIA modem card it can't find a free IRQ. I haven't tried this solution yet with 3.1.8, but I'll give it a shot tomorrow and see if I can get my Xircom modem to work with the latest PCMCIA. I (the original poster) went back to pcmcia-source 3.0.14 yesterday and everything is up and running again. pcmcia-cs is 3.1.8-4 and setserial is still installed. -- Philip Lehman [EMAIL PROTECTED]
Re: netscape sharing win98/linux
On Sun, 16 Jan 2000, jason wrote: i think i remember a past post about how to get netscape commicator to share the mail box between win9x/linux partiions.. what's i've done is put the mail box on the windows partition .. mount it from linux and point there.. but it doesn't seem to be working.. i have communicator 4.7 any clues.. or any better suited email programs.. what i really want is the ability to share across my partitions.. and have folders.. If it doesn't have to be an X app, consider pine and pcpine. Never tried that (that is, the pcpine part of it), but it should work. -- Philip Lehman [EMAIL PROTECTED]
anacron jobs for users
Is there a way to set up anacron jobs as a non-root user? There doesn't seem to be an equivalent to the user crontab files and I couldn't find any other obvious solution. I have this small backup script for parts of my home directory and I want it to run as a daily job. The box is a laptop so cron jobs are not quite the way to go, but I'm sure there's a better way than putting sudo -u username myscript in the system-wide anacrontab? -- Philip Lehman [EMAIL PROTECTED]
Re: tex/latex: I can't find the format file `latex.fmt'!
On Tue, 11 Jan 2000 [EMAIL PROTECTED] wrote:
I don't think I've ever seen this before. Having replaced the dying
hard disk, I've done a clean install of slink. The ability to use
latex seems to come and go (mostly go), and I've given the tetex files
a purge/reinstall cycle, but still no luck.
...
I can't find the format file `latex.fmt'!
I've used find on a working machine, and this file doesn't seem to
exist anyway.
It's supposed to be in /var/lib/texmf/web2c/. Does `kpsewhich
latex.fmt` fail as well?
...
I assume that this is something very simple, but I'm stumped.
I'm afraid that's too trivial, but anyway: did you update the ls-R
databases? If it still doesn't work, take a look at
/etc/texmf/texmf.cnf and check if $TEXMFMAIN and $TEXMF are set
correctly. $TEXMF should be something like TEXMF =
{!!$TEXMFMAIN,!!$TEXMFLOCAL;!!$HOMETEXMF}.
--
Philip Lehman [EMAIL PROTECTED]
Postscript, Ghostscript, Unicode, and TT fonts...
This is a somewhat specifc problem, but I didn't get a reply on comp.lang.postscript and maybe somebody on this list can help me out or point me to more specific sources of information. This is the scenario: I'm trying to use unicode truetype fonts with the latest StarOffice on a Linux machine. The problem is that SO supports neither truetype fonts nor unicode. Since I only need about a dozen glyphs 255 (basically all typographical quotation marks, the typographical dashes, the french oe ligatures, the single guillemets, the ellipsis character, the euro sign) I figured I'd just find a way to map them to the code range that is not used in iso8859-1. Since my font server (xfsft) is able to remap fonts on-the-fly, I have him serve the iso10646-1 fonts as microsoft-cp1252 (which supplies all characters I'm interested in). This is the relevant part of the remapping process (in decimal codes): quotesingle 8217- 127 Euro8364- 128 quotesinglbase 8218- 130 florin 402- 131 quotedblbase8222- 132 ellipsis8230- 133 dagger 8224- 134 daggerdbl 8225- 135 circumflex 710- 136 perthousand 8240- 137 Scaron 352- 138 guilsinglleft 8249- 139 OE 338- 140 Zcaron 381- 142 quoteleft 8216- 145 quoteright 8217- 146 quotedblleft8220- 147 quotedblright 8221- 148 bullet 8226- 149 endash 8211- 150 emdash 8212- 151 tilde732- 152 trademark 8482- 153 scaron 353- 154 guilsinglright 8250- 155 oe 339- 156 zcaron 382- 158 Ydieresis939- 159 My next step is to supply SO with hacked font metrics files that correspond to the cp1252 encoding and put all the font names in the appropriate printer definition file. The idea is to make SO believe that the fonts are resident printer fonts and not ttfs. Up to this point, I think I know what I'm doing and everything works fine. The fonts show up on the screen, SO finds the correct metrics for all glyphs and prints them. This is where gs comes in. The print spooler calls gs via an input filter to convert the ps file to pcl and sends gs' output to my LaserJet 5L. At this point, gs gets the glyphs from the original ttf files. Still, this works fine for all characters defined by the iso8859-1 encoding table. What I can't get to work is the unicode part, that is the remapped characters (gs substitutes the '.notdef' character for them). I have a rather fuzzy idea of what postscript is and how it works, so please take the following with a grain of salt. If I redirect the ps output produced by SO to I file, I notice two things, the first being that at several points the encoding is defined as /ISOLatin1Encoding and the second that there is a /changesvec section which seems to provide the mapping information for the characters I'm after. It looks like this: /changesvec [ 16#80 /euro 16#82 /quotesinglbase 16#83 /florin 16#84 /quotedblbase 16#85 /ellipsis 16#86 /dagger 16#87 /daggerdbl 16#88 /circumflex 16#89 /perthousand 16#8a /Scaron 16#8b /guilsinglleft 16#8c /OE 16#8e /zcaron 16#91 /quoteleft 16#92 /quoteright 16#93 /quotedblleft 16#94 /quotedblright 16#95 /bullet 16#96 /endash 16#97 /emdash 16#98 /tilde 16#99 /trademark 16#9a /scaron 16#9b /guilsinglright 16#9c /oe 16#9e /zcaron 16#9f /Ydieresis ] def Obviously, the codes given here correspond to the cp1252 mapping and gs can't find them in the cmap of the ttf file. What I would like to do is to pipe SO's ps output through a perl script that makes some changes and corrects this before sending it to gs and the print spooler. At this point, I'm lost. Should I change the 'ISOLatin1Encoding' to something else or simply fix the codes so that they correspond to the cmap of the ttf file? I tried the latter by hand with a sample file, that is I changed 16#84 /quotedblbase to 16#201e /quotedblbase but to no avail. gs doesn't give any error messages, it simply subsitutes 'notdef' for 'quotedblbase'. Is the whole idea nonsense or is this possible (in theory at least)? And in practice? Please help me out here... TIA -- Philip Lehman [EMAIL PROTECTED]
Re: How to install pine on slink from source
On Wed, 17 Nov 1999, Daniel Mashao wrote: I once did it, but I forgot what I did. Searching the archives I could not find pine installation messages. Anyone remember? I just can't get used to mutt. It smells too much like vi and I find vi hard. Get the source from any Debian mirror, last time I checked they were in debian/project/experimental - Get foo.orig.tar.gz, foo.diff.gz, and foo.dsc (where foo is something like pine plus the version number) - `dpkg-source -x foo.dsc` - `cd foo` - as root: `debian/rules binary` You also need dpkg-dev and the development packages of the libs required by Pine. This will patch the source, compile, and build binary debs ready to install with dpkg/dselect. HTH -- Philip Lehman [EMAIL PROTECTED]
Re: emacs and word processing
Thanks a lot for all the input, this really helps me out. -- Philip Lehman [EMAIL PROTECTED]
emacs and word processing
Recently, I started using XEmacs to edit LaTeX files. The LaTeX mode is nice, but some features related to word processing still pose some problems to me: (1) ...actually not an emacs issue, but anyway: spell checking with ispell under emacs works (sort of), but ispell doesn't recognize 8-bit characters. Same problem with running ispell from the command line, but I can't find anything in the docs (only an option to force 7-bit, which is not quite what I'm looking for). So how do I make ispell process 8-bit input? (2) Emacs' flyspell-mode has support for English, but I need spell checking for English, German, and French. Is this a built-in facility or is it accomplished by ispell as well? And how do I get additional language support? I'm not sure if that matters, but I'm using the mule binary. (3) What about spell checking with multilingual texts, e.g. a German text with quotes in English, French, and Latin? Do I have to do this paragraph by paragraph? (4) I turned on auto-fill-mode and filladapt-mode. Frankly, I have no idea what that really means, but while typing new text word-wrapping works the way I want it now. I guess auto-fill is what I was looking for in the first place, what does fill-adapt do, anyway? ;) (Don't flame me for that, these docs are killing me.) However, when adding a word to a paragraph or deleting some words I miss a convenient way to re-justify the whole paragraph, something like hitting ^J in Pico. (5) Is there some documentation on the X resources used by XEmacs for Joe XEmacs user? The sample Emacs.ad and .xdefaults files are not very comprehensive. I figured out how to change the look of the Athena and Motif style widgets but I still don't know how to set the background color of the text space and the 'shadowWidth' of the minibuffer bar (I have a very specific understanding of the word customization ;). Oh, and if there is any documentation on (X)Emacs which is not systematic but rather task oriented, I'd be interested in some pointers. TIA -- Philip Lehman [EMAIL PROTECTED]
Re: Wp8 and Debian 2.1
On Wed, 27 Oct 1999, NaNoFlaW wrote: I getting a seg fault when I try to run xwp. this is the debug info Program received signal SIGSEGV, Segmentation fault. 0x4017026f in strncmp () and this is the error message 3316 Segmentation fault $Platform/ins/$Exec $Menu $Values $bgcolor No problems with libs ( i have libc5 installed ) Did you install all of libc5, xlib6 (!= xlib6g), and xpm4.7? In addition to that you need a wrapper that rearranges the library path, this is the one I use #!/bin/sh source /etc/profile export LD_LIBRARY_PATH=/lib/libc5-compat:/usr/lib/libc5-compat:$LD_LIBRARY_PATH export LIBRARY_PATH=$LD_LIBRARY_PATH $WPDIR/wpbin/xwp.bin $@ HTH -- Space for hire. Contact Philip Lehman [EMAIL PROTECTED]
Re: Netscape vs M$ IE (Re: Sun goes...)
On Thu, 28 Oct 1999, J Horacio MG wrote: I'd like to try other graphical browsers, but there are not that many around. I tried installing Sun's HotJava... to no success. Meantime, I most times have to resort to good old Lynx. Operasoft is working on a Linux port of Opera 4.0... -- Space for hire. Contact Philip Lehman [EMAIL PROTECTED]
Re: sigh......
On Thu, 28 Oct 1999, Charles Lewis wrote: I have similar problems periodically, although I'm on potato and my sources.list is pointing to http://kde.tdyc.com potato kde contrib rkrusty. I just try again later (sometimes a couple of days) and eventually I get through. I understand they are looking for a stable home for tdyc. There are debs on the mirrors of ftp.kde.org. I never used them so I don't know if they differ from the ones on kde.tdyc.com. You might want to try one of these for example: ftp://ftp.varesearch.com/pub/mirrors/kde/stable/1.1.2/distribution/deb/slink/i386/ ftp://ftp.varesearch.com/pub/mirrors/kde/stable/1.1.2/distribution/deb/potato/i386/ There's a list of all mirrors at http://www.kde.org/mirrors.html HTH -- Space for hire. Contact Philip Lehman [EMAIL PROTECTED]
Re: HELP!! Re: HELP!: problems after POTATO upgrade
On Tue, 26 Oct 1999, J Horacio MG wrote: I updated libncurses4, libncurses4-dev, libreadlineg2_2.1-13.6 (bash_2.02.1-1.8) as well now, but I'm having trouble since libc6-dev conflicts with libstdc++2.9-dev: [...] Fine, so I try to deinstall libstdc++2.9-dev but: h0rus:/tmp# dpkg --purge libstdc++2.9-dev dpkg: dependency problems prevent removal of libstdc++2.9-dev: g++ depends on libstdc++2.9-dev (= 2.91.60). ^^ ^^ [...] install libstdc++2.9-glibc2.1 (2.91.66-1) install libstdc++2.9-glibc2.1-dev (2.91.66-1) install libc6-dev remove the old libstdc++2.9-dev enjoy Also, if I ever get to install libc6-dev (fingers crossed), will I have to reinstall all those libraries which returned all the : ldconfig: warning: can't open /usr/lib/libcrypt.so (No such file or directory), skipping errors? I say this since `dpkg -l' lists all of them as fully installed (I attach a `dpkg -l' log). These errors might be caused by orphaned symlinks. Check if /usr/lib/libcrypt.so actually points to an existing library. If not if you don't have libcrypt installed, remove to link. -- Space for hire. Contact Philip Lehman [EMAIL PROTECTED]
RE: connection cost monitor anyone?
On Tue, 26 Oct 1999, Pollywog wrote: I'm looking for a convenient way to monitor and log telephone costs (using ppp in a country where telephone bills can *really* spoil the day) but I can't seem to find a powerfull utility. At the moment I'm running pppcosts. It does the job, but I would like something more full-featured. Any recommendations? I thought kppp could do this. Perhaps someone who uses kppp will know. It does - if it is running and handles the dial-up. But I rather need a logger that runs either from /etc/ppp/ip-up or as a daemon (monitoring all connections no matter who fires up pppd and how) and a corresponding X interface that displays some info and warning messages (e.g. after a certain online time or depending on the accumulated costs). -- Space for hire. Contact Philip Lehman [EMAIL PROTECTED]
Re: Diskless and Stare Office??
On Mon, 25 Oct 1999, Benak Istvan wrote: Someone please tell me, if I want to use a Diskless clients (p200 with 64 MB ram) and a strong server and 10 Mb ethernet cards, how will run the StarOffice 5.1 on the clients?? Have anyone any experience with this?? Not with loading off the network in particular, but I tried it out on my Notebook, a P233 with 64 MB. SO is useable, but working is painful. This is mostly a RAM issue, I think. It's bearable with 128 MB, but still feels somewhat sluggish. If you ever thought Netscape was a pig... ;) -- Space for hire. Contact Philip Lehman [EMAIL PROTECTED]
Wordperfect 8 woes again
I know that this must have been beaten to death, but after spending a lot of time browsing Corel's site, the list archives, and deja it still segfaults on my slink/potato (running libc6 2.1). I know it's libc5 based, so I installed libc5 and the libc5 versions of xlib6 and xpm4 - to no avail. In detail, that's: libc5 5.4.46-3 xlib6 3.3.5-1 xpm4.7 3.4k-2 That's what was mentioned several times on this list and that's what ldd suggests too. Am I missing something? Any ideas? -- Space for hire. Contact Philip Lehman [EMAIL PROTECTED]
Re: Exim and fetchmail
On Thu, 21 Oct 1999, Richard Clarke wrote: Could someone point me to documentation for exim and fetchmail which is easier to understand than the man files. There is more documentation in /usr/doc or /usr/share/doc. It's more comprehensive, but it's probably not that helpfull if you just want to get started. As I understand, exim is a local mail agent which uses SMTP? to send mail around the localnetwork. Fetchmail delivers remote mail, POP3 in my case, to this local network and the users on it. However, I don't fully understand how it all works yet and my attempt to get either exim or fetchmail working don't seem to bear any success. If you use the deb packages, exim should work almost out of the box. The postinstall script will ask some questions and find a suitable default config for your system. In particular, you probably want to use your ISP's SMTP server as a so-called smart host, pay attention to that part. Try exim first by sending mail to yourself on the local machine, that is, say, [EMAIL PROTECTED]. In case it doesn't work, install the eximon package and run eximon (as root or sudo'ed), that's a graphical interface to exim that will display the log and the contents of the mail queue. Post again if you run into trouble. Try to get that straight before trying out fetchmail, it would be useless without a working MTA like exim. Then, install exim and create a .fetchmailrc file either for root (in /root, for system-wide setups) or for you only (in your home dir). The syntax is pretty straightforward, for example: poll servername # name of the pop server proto pop3 # use pop3 protocoll user username # username for pop server password password # guess what is username # username on *local* host and (optional): fetchall# fetch old and new mail and delete from server And so on for every mail account. .fetchmailrc permissions *have* to be 600 (`chmod 600 .fetchmailrc`). Let's assume you created a system-wide .fetchmailrc in /root and want to poll all servers whenever you connect to your ISP. Put that at the bottom of '/etc/ppp/ip-up': fetchmail -d 600 fetchmail The first line will start fetchmail in daemon mode and tell it to poll all hosts in your .fetchmailrc (/root/.fetchmailrc in this case, since commands in /etc/ppp/ip-up are run as root by default) every 600 seconds. The second line will wake the daemon up and tell it to poll right away (otherwise the first attempt would be made after 600 seconds). Whenever you want to fetch mail in between these intervals, just run `fetchmail` as root, to wake up the daemon process. Don't forget to add this fetchmail --quit to '/etc/ppp/ip.down'. If you don't want/need to run fetchmail as root, the command in /etc/ppp/ip-* should be preceded by a `sudo` command, e.g. sudo my_local_username fetchmail -d 600 Let us know what you come up with. -- Space for hire. Contact Philip Lehman [EMAIL PROTECTED]
Re: Exim and fetchmail
On Thu, 21 Oct 1999, Philip Lehman wrote: root, the command in /etc/ppp/ip-* should be preceded by a `sudo` command, e.g. sudo my_local_username fetchmail -d 600 It's `sudo -u my_local_username...`, forgot the '-u', sorry. -- Space for hire. Contact Philip Lehman [EMAIL PROTECTED]
Re: Exim config question
On 25 Jul 1999, Gary L. Hennigan wrote: I'm in the process of getting a little home network set up. One of the hosts, call it mach2, won't have a connection to the internet, just to other hosts on my home network. The other host, call it mach1, will occasionally connect to the internet via dialup. I have mach1 all set up but can't seem to get mach2 to forward all the mail to mach1. In fact I can't get mach2 to deliver mail directly to mach1 at all. What I want to do is set mach2 so that it uses mach1 as a smarthost. I think I got that right in my exim.conf (configuration 2 from the debian installation) but I can't seem to make it work. Just two ideas: Check if relay_domains is set in mach1's exim.conf and make sure that sender_host_reject_relay corresponds to your situation. I think the default setting is sender_host_reject_relay = *, which would bounce all mail from mach2... -- Philip Lehman [EMAIL PROTECTED]
Re: Gnome runs out of sockets without sound?
On Sat, 24 Jul 1999, Graham Williams wrote: All works great, of course, except Gnome seems to be trying to open some audio device (I haven't compiled any kernels and I don't think my Sound Blaster Pro compatible card is supported in the kernel I have?), If it is a 100% compatible clone, it will work with the SB drivers. Dunno how to deactivate sound in Gnome, but if you'd like some noise anyway, it might be worth a try to set up sound. -- Philip Lehman [EMAIL PROTECTED]
Re: Pascal
On Sat, 24 Jul 1999, peter karlsson wrote: I installed fpc (Free Pascal) before, but now it is listed as obsolete in Dselect. Has it been replaced with something else? No. This means that the original *.deb file is no longer available, which is quite natural if you don't keep Debian CDs in your drive all of the time ;) The word obsolete is somewhat misleading. And, now that I upgraded to the version of libc6 (2.1) from potato (to be able to install some other development packages), all my Pascal programs causes segmentation faults on startup. Recompile them and see if that helps. If not, put the 2.0 version of glibc somewhere in /usr/local/lib, add this location to your environment and to /etc/ld.so.conf, and run ldconfig. And make sure your apps use the right linker. -- Philip Lehman [EMAIL PROTECTED]
Re: Netscape crashing -- a lot.
On Thu, 22 Jul 1999, Adam Shand wrote: i was trying to find a pattern among the info people posted but couldn't really see one. did anyone else figure this out? - p2-266 128mb ram - kernel 2.2.9 - glibc 2.1.1-13 It's the glibc from unstable. -- Philip Lehman [EMAIL PROTECTED]
Re: default font for Xaw - Applications
On Fri, 23 Jul 1999, Bernhard Rieder wrote: Since I installed Xfsft all the text-widgets of Xaw-applications are nearly unreadable (Xaw uses a tiny awful font). Where can I set the default font for _all_ applications ? Technically, in your .xdefaults file. Take a look at editres, it will give you the complete ressource tree for athena based apps and can change ressources on-the-fly. Very useful to find the most convenient settings (you have to put them manually in .xdefaults, though). But if xfsft messed it up, rearranging the paths in xfsft's config file will do the trick. Just copy you old xfs paths and append the paths to your ttf fonts at the end of the list. HTH. -- Philip Lehman [EMAIL PROTECTED]
Re: modem line speed
On Fri, 23 Jul 1999, Shao Zhang wrote: I have 33.6k modem, but windoze reports 56700bps, I have tried to set with the string W2, and it still reports 56700bps. Hopefully, linux will give me a correct value. Invoke chat with the -r option from your /etc/ppp/peers/my_isp file. It will write the connect speed to a (custom) log file. -- Philip Lehman [EMAIL PROTECTED]
Re: shutdown -h now : don't power off
On Fri, 23 Jul 1999, Khalid EZZARAOUI wrote: I set the APM option in the kernel 2.2.10. if I do halt, my box act like the use of reboot : it reboots using the shutdown of windows95 act as i want : power off the computer the command : #ps aux |grep apm root 2169 0.0 0.3 1148 416 pts/0S15:39 0:00 grep apm* to be sure I am recompiling my kernel. Have you tried calling shutdown via /sbin/poweroff? -- Philip Lehman [EMAIL PROTECTED]
gv with neXtaw anyone?
I'm trying to build gv using neXtaw, but it segfaults. Linking to neXtaw works fine with Xaw based apps for me, but I'm not sure about replacing Xaw3d with neXtaw. Is this at all possible or am I getting somthing wrong here? -- Philip Lehman [EMAIL PROTECTED]
Re: Netscape crashing -- a lot.
On Fri, 23 Jul 1999, Adam Shand wrote: i was trying to find a pattern among the info people posted but couldn't really see one. did anyone else figure this out? It's the glibc from unstable. but i thought there were several posts from people claiming that it worked just fine for them and they were using glibc 2.1? I didn't follow this list lately, so I'm probably not up to date. However, this is the last explanation I remember and corresponds to my experience too. Right after moving to glibc 2.1, netscape started giving me those bus errors. I'm not running an all potato system, just slink with some unstable stuff thrown in and glibc seems to be the only relevant part in my case, but I may be overlooking something. This is a *royal* PITA. If you find out more, please post it. -- Philip Lehman [EMAIL PROTECTED]
Re: Netscape crashing -- a lot.
On Fri, 23 Jul 1999, Brad wrote: On Fri, 23 Jul 1999, Matt Kopishke wrote: I had that problem, the fix (for me at least) was deleteing my ~/.netscape Dunno why, but somthing get corupted. Hmmm... this is the first suggestion that has worked for me! i still get bus errors trying to login to dhs.org, but i haven't gotten it to crash on closing a window yet. I just tried that from a new user account and had no problems crashing it ;) I did change some basic settings after launching: refuse all cookies, use a proxy, disable javascript and java - and this was supposed to improve stability. Shouldn't make a difference, though. -- Philip Lehman [EMAIL PROTECTED]
