per connection routing
I have a DLS connection to the Internet and a T1 connection to the other office, which in turn is also connected to the Internet. Both connections use IP masquerading with two completely different addresses. I want to put a firewall/router in the middle that would provide: - load balancing for these two connections. Because of NAT, the choice of which way to route has to be done on a per-connection basis. - fault tolerance for the Internet connection: if oen link goes down all new connections shold go through the other. The existing connections will have to timeout. - fault tolerance for the connection to the other office: it T1 goes down, a VPN connection should be established through DSL. Any ideas, siggestions, experinces or pointers ? What would be the most appropriate OS for it ? Sergey.
imp with php4
Has anybody been able to run imp with php4 on potato ? I'm getting error about imap_utf7_encode() being undefined and indeed there is no such function in php4-imap module. I'm running on sparc. Sergey.
Re: Trying to run one process as root, how?
The other solution, setting XAUTHORITY in the root environment to that of the user that owns the X session, also worked for me. I've never seen that before - neat. I'm thinking, shouldn't this be done automatically by /bin/su ? This would make a lot of sense and simplify life significantly. Just check the archives, how often this question pops up. Granted, it will not work in every situation, but still will be useful in most cases. Another way might be to encorporate the setting of XAUTHORITY into the default root's .bashrc Sergey.
Powerchute diffs for deb ?
Has anyone tried to debianize PowerChute Plus ? I intend to create a patch that would put config, log, pid and init files into appropriate places when converting from rpm to deb with alien. But maybe someone already has done that ? Sergey.
Re: Attention: imapd gpoing back to $HOME as mailbox root
On Mon, 24 Jan 2000, Joe Block wrote: Joe Emenaker wrote: Actually, what we *really* need is some sort of consensus. I mean, it would be pretty nice if imapd and other tools (like procmail) all looked in the same default location without any configuration. I know Elm and Pine used $HOME/Mail and $HOME/mail at one time. Surely, I can't be the only one who sees the benefit in having all of the tools look in the same location for the Sent Mail folder, and Drafts, etc. You aren't. Count me as a vote for $HOME/mail $HOME/mail would be OK. Just the mbox driver should be modified then to look in $HOME/mbox instead of $HOME/mail/mbox. Sergey.
Re: Mirror transformation of the Postscript files
On Sun, 26 Dec 1999, Alexey Petrov wrote: Hi All, I have need in mirror transformation of the Postscript files for printing on the transparence. I should be glad for any idea. I'd suggest using CUPS instead of standard lpd. (package cupsys from unstable). This kind of stuff is done by just specifying -o mirror option to the lpr command. There is much more good to it. Unfortunately the best part - the GUI interface and the drivers to several hundreds of printers are commercial (I've got a multiuser package for our department - very cool). But the generic drivers should be enough for most of the typical situations, especially if you have a PostScript printer. I hope at some point the CUPS is going to be THE standard. Sergey.
Re: Can XF86_SVGA do 1280x1024?
On Fri, 17 Dec 1999, Fethi Okyar wrote: I start XF86Setup choose the card in card settings choose the monitor, enter the correct hsync and vsync (i.e. ver_refres = 30-64, hor_refresh=47-104) ^^ In my experience here might be the typical problem. I'm running at [EMAIL PROTECTED] and the limits are 31.5-96 for horisontal and 55-160 for vertical. Anyway, you can manually insert the modeline. Here is mine: Modeline 1280x1024 157.50 1280 1344 1504 1728 1024 1025 1028 1072 +hsync +vsync Sergey. and select 1280x1024 and 1024x768 BUT... when I save and exit , in my XF86Config file there is only one modeline Modeline 1024x768 65.00 1024 1032 1176 1344 768 771 777 806 -hsync -vsync nothing in regard to 1280x1024... So, the question is, how can I get XF86Setup to add a modeline for 1280x1024... -- -- Phil Brutsche [EMAIL PROTECTED] There are two things that are infinite; Human stupidity and the universe. And I'm not sure about the universe. - Albert Einstein --DD1369621D39CDF6CD4F8ED4-- --EA5C5C9880DE2CFB18473039--
Re: pppd died unexpectedly. Where to look?
On Sun, 28 Nov 1999, Hans wrote: biggerSystem: Acer Travelmate 512T running Slink. Generic PCMCIA modem. Goal: setting up ppp connections to two different ISPs. Done: configured pppconfig, kppp, resolv.conf (for DNS) The problem: pon/poff dials in, but refused to connect. kppp the same, and the log from pppd is as follows for both ISPs): Nov 28 19:07:19 mack pppd[407]: pppd 2.3.5 started by root, uid 0 Nov 28 19:07:19 mack pppd[407]: Using interface ppp0 Nov 28 19:07:19 mack pppd[407]: Connect: ppp0 -- /dev/ttyS1 Nov 28 19:07:20 mack pppd[407]: peer refused to authenticate Somehow your ISP uses different approach than your configuration. Try adding -debug option to the pppd options in the kppp, and see if this produces more info. Sergey.
Re: Help with a Dell Latitude CPia
3. My Gold card 56K modem will dial out but will not connect to my ISP. I get the error that the pppd died unexpectedly. Judging from the error message you're using kppp. Try to empty /etc/ppp/config file. If that does not help, add -debug option to the pppd in the kppp options panel, and then look at the logs - may be it'll give you some hint. Sergey.
Re: imap
By the way, if you are using exim, then it supports Maildir directly, no need to use procmail for this. Sergey.
Re: samba: cannot change password from Windws NT
On Tue, 7 Sep 1999, Debian Mail wrote: I setup samba as a PDC for Windows NT. It works. But when I want to change the password from the Win NT machine I get: The User name or old password is incorrect. Letters in... And in the smb logfile I get: smbd/chgpasswd.c:check_oem_password(699) check_oem_password: incorrect password length (445298348). The same when I want to change the password using smbpasswd I didn't change the according lines in smb.conf. They read: unix password sync = true passwd program = /usr/bin/passwd %u passwd chat = *New\spassword:* %n\n *Re-enter\snew\spassword:* %n\n *Password\schanged.* . What could be wrong? I've got similar problem too. It looks like a bug, so I submitted a bug report (bug# 43492). I have not heard form the maintainer yet. You might want to contribute your information to this bug report. Sergey.
Re: full color ghostview ?
On Tue, 17 Aug 1999, Paul Miller wrote: Sergey V Kovalyov wrote: b) is there a sensible replacement for ghostview that is functionally similar the windows' GSview ? I much prefer the 'gv' package over 'ghostview'. It is another front end to 'ghostscript' which has a much better feel and look IMHO. You might give it a shot. I prefere gv too, but it does not eliminate the problem. I spoke about ghostview in the generalized sense (gv, kghostview, ghostview). The all use ghostscript as a back end and produce identically horrible output. Sergey.
Re: full color ghostview ?
On Tue, 17 Aug 1999, Brian Servis wrote: *- On 17 Aug, Sergey V Kovalyov wrote about Re: full color ghostview ? On Tue, 17 Aug 1999, Paul Miller wrote: Sergey V Kovalyov wrote: b) is there a sensible replacement for ghostview that is functionally similar the windows' GSview ? I much prefer the 'gv' package over 'ghostview'. It is another front end to 'ghostscript' which has a much better feel and look IMHO. You might give it a shot. I prefere gv too, but it does not eliminate the problem. I spoke about ghostview in the generalized sense (gv, kghostview, ghostview). The all use ghostscript as a back end and produce identically horrible output. This sounds like a font problem. Do you have the font packages installed for ghostscript? No, this is not the font problem, fonts look ok (please, look at my original post). The problem is that gv (and the family) produce horrible dithered images and do it very slowly. It seems that ghostscript's x11 device is to blame. So I was wondering if there is alternative pastscript viewer with a similar capabilities, which does not suffer from the dithering. Xv looks good, but it does not support multiple pages. If you've seen GSview for windows - this is how it should look like. Sergey.
Re: full color ghostview ?
On Tue, 17 Aug 1999, Arno wrote: On Tue, Aug 17, 1999 at 11:52:02AM -0400, Sergey V Kovalyov wrote: This sounds like a font problem. Do you have the font packages installed for ghostscript? No, this is not the font problem, fonts look ok (please, look at my original post). The problem is that gv (and the family) produce horrible dithered images and do it very slowly. It seems that ghostscript's x11 device is to blame. So I was wondering if there is alternative pastscript viewer with a similar capabilities, which does not suffer from the dithering. Xv looks good, but it does not support multiple pages. If you've seen GSview for windows - this is how it should look like. From what I've read, you want anti-aliasing. With gs use the command line switch -sDEVICE=x11alpha instead of the normal (implicit) -sDEVICE=x11. In gv you can enable anti-aliasing through a menu option (State - Antialiasing). AFAIK, ghostview has no support for enabling anti-aliasing. Tried that. No difference. It only has effect on things like lines, fonts, but on the representation of color. What would have worked, if a postscript viewer created a bitmap (with appropriate ghostscript device) and then displayed it. I guess that how GSview works. Does such thing exist for Linux ? Sergey.
full color ghostview ?
The problem: ghostview and the family (gv, kghostview) show a horrible dithering and it takes forever to display even simple images. On the other hand, the windows' version of the same thing (GSview) takes just a few seconds and displays in full color (16 or 24 bit, whatever the current depth is). The question: a) is there a way to convince ghostview not to dither, but use full color? or b) is there a sensible replacement for ghostview that is functionally similar the windows' GSview ? Sergey.
Re: Netscape/X key/mouse binding problem
On Thu, 12 Aug 1999, Tim P. Gerla wrote:
(I'm not a subscribed member of this list, so please Cc: any replies.
Thanks!)
Recently I've been having a strange problem with Netscape and X. I'm running
the latest potato stuff. When I start netscape, I get this:
...
Warning: translation table syntax error: Modifier or '' expected
Warning: ... found while parsing 'lt;Btn1Motiongt;:
This is the problem. For some reason the file (most prabaly Netscape.ad)
is modified to display in a browser. So if you replace lt; by and gt;
by everything should work.
Another question is which file is it and why it happened in the first
place.
So first, try locate Netscape.ad and see if this is the file you need.
If not, try to check files in /etc/X11/Xresources and ~/.Xresources.
Sergey.
Re: Debian 2.%$#@*!
Actually, there is another variant. You can install netscape right from debian using apt. Instead of choosing installer, choose the packages netscape-base-x.xx, communicator-smotif-x.xx, and other additions (where x.xx is the version you want). I do it this way and works ok. Sergey. On Sun, 1 Aug 1999 [EMAIL PROTECTED] wrote: Wait! Hold the phone. I tried entering nslookup ftp.uu.net and got an address, so I entered ftp ftp.uu.net and connected! ftp ftp.netscape.com also connects. Cripes. Something so simple! Entering ftp.netscape.com from the ftp prompt gets invalid command. You have to clear the prompt and type in the first ftp. Anyway, I went to /pub and found a large communicator file. Any chance this may be the holy grail? -- Max On Sat, 31 Jul 1999 13:42:16 -0700 (PDT) George Bonser [EMAIL PROTECTED] writes: You probably do not have reverse DNS configured properly. Many sites will not allow you to connect if the in-addr.arpa PTR record is not in place for your system. Can you connect to ftp.uu.net ??? If not, what IP address are you using? Try using an external nameserver and doing an nslookup of that IP address, if it does not return a hostname, that is probably your problem. ___ Get the Internet just the way you want it. Free software, free e-mail, and free Internet access for a month! Try Juno Web: http://dl.www.juno.com/dynoget/tagj.
Re: kppp error - pppd dies unexpectedly
Hi.,. I've removed the entry from /etc/ppp/options, but it still no working... Try making /etc/ppp/options empty, see if this helps. Also add a debug option for ppp in kppp's ppp options tag, and look at /var/log/syslog, see if there is anything interesting. Sergey. when i try to connect to my internet provider with kppp i get a message pppd died unexpectedly, and then it hangs up. Any solution to the problem? [EMAIL PROTECTED] Hi! This is caused by a bug in kppp. It requires that the entry lock in /etc/ppp/options to be removed.
Re: newbie: installing new drivers
Just run modconf. It is the same program that is used during setup. You might have to play with ios and irqs if first attempts fail. Sergey. On Fri, 18 Jun 1999, David Brode wrote: 1. I just added an ethernet card to my system (3Com SC509TP) and I understand that I'm supposed to use the driver 3c509. My question is *how* do I get the system to load that driver? (If I were installing linux again I could use dbootstrap, but I have the feeling there's an easier way.) The boot msg doesn't have any eht0 mention or any other mention of the card. 2. how would your answer change if I told you I had TWO ethernet cards I wanted to install? tia
Re: Window Maker
On Thu, 17 Jun 1999, Brad wrote: On Thu, 17 Jun 1999, Stephen Pitts wrote: The correct way to setup what programs run when you start X is to make an .xsession file, like this: echo exec /usr/X11R6/bin/WindowMaker ~/.xsession chmod +x ~/.xsession If you only want to change the default window manager (and you have root access ;), you could also edit the /etc/X11/window-managers file. All users will use the first wm listed in that file, unless they have a .xsession to change it. IIRC, when you install some of the various window managers the postinst script offers to make this window manager the default by placing it first in that file. i don't recall if WindowMaker is one of them or not. When updating an already installed wm the postinst scripts (usually) don't make the offer again, only when installing new. Actually, the simplest method to start a particular window manager is to name it in a command line, like startx wmaker the default debian /etc/X11/Xsession script checks this argument and, if the file exists and executable, will launch it. Sergey.
Re: ldap/pam problem
If so, then when I tried those modifications, I couldn't figure out how to get reasonable behavior. If you have auth sufficient pam_ldap.so auth required pam_unix_auth.so try_first_pass then if the entry is found in ldap, pam returns and you never execute things like motd, etc. which is not what you want. I used this complex invokation, you'll need an appropriately bug-fixed pam library (Ben, you have my patches..) auth requisite pam_securetty.so auth [success=1 default=ignore] pam_unix_auth.so auth required pam_ldap.so use_first_pass auth optional pam_group.so auth optional pam_mail.so Another solution (that I did) is to just place those optional modules BEFORE pam_ldap.so. Seems to work fine for me. Anyone sees a reason not to do so ? I agree, however, that the complex method got to be fixed - it is too cool not to use it. Sergey.
Re: Using more than one ISP
Is there any easy way of maintaining access to more than one ISP without re-writing the config files for smail and ppp? At present I use a spare computer for this but it's not very convenient. Try just using pppconfig and choose two different provider names (e.g. isp1, isp2). Then to connect you'll use pon isp1. Works for me. Sergey.
Re: Too many posts for my mailbox
On Wed, 16 Jun 1999, Hugo van der Merwe wrote: My mailbox cannot handle the number of posts this list generates, especially not when I'm away for the weekend or even week. What should I do to participate in this mailing list without clogging my mailbox? I suggest you subscribe to debian-user-digest, as I do. You'll recieve few big messages (with 10-15 posts in them). It is very convenint. Sergey.
Re: kppp error - pppd dies unexpectedly
Just a guess: did you clean /etc/ppp/options ? You need this file to be empty for the kppp to work. Sergey. On Fri, 11 Jun 1999, Micha Feigin wrote: when i try to connect to my internet provider with kppp i get a message pppd died unexpectedly, and then it hangs up. Any solution to the problem? [EMAIL PROTECTED] _ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com
Re: needed, PAM LDAP help
Debian GNU/Linux potato caesar tty8 caesar login: test Password: Last login: Fri Jun 11 21:40:19 1999 from - on tty8 /bin/login: error in loading shared libraries: /lib/libnss_ldap.so.2: undefined symbol: ldap_set_option I get this problem too after the latest upgrade yesterday. Openldap and its libraries were upgraded; I guess the rest have to be recompiled too with the new libraries. I home the new packages are on the way; if not, we should submit a bug report. Sergey.
Re: needed, PAM LDAP help
I was following the thread on using LDAP to handle authentication of clients, and I am having a lot of trouble getting things to work. First I created a new testing account # adduser testing Second, I downloaded the conversion tools, turned off shadowpasswords, and That was not necessary ^^ updated my LDAP database. Using ldapsearch, I verified that all the accounts had been transfered over. I then downloaded the pam_ldap modules, and changed /etc/pam.d/other to: auth sufficient pam_ldap.so auth required pam_unix_auth.so try_first_pass account required pam_unix_acct.so password required pam_unix_passwd.so shadow session required pam_unix_session.so Note: there is not an /etc/pam.d/login file. That means you havn't installed the login that actually uses pam. The packade is pam-apps Then I deleted the test account from /etc/password, /etc/shadow etc. But I was not able to login with the test account. Well, besides authenticating a user, you also need other information, like uid, home dir, shell. pam_ldap does not provide this. You also need libnss-ldap to get this info. After you install libnss-ldap and pam-apps and try again, you should look into your /var/log/syslog file where slapd will report all requests. This will help determine the problems. Sergey.
Re: We need centralized accounts -- Any docs for ldap passwords?
libpam-ldap will allow password change. The rest have to be done manually (or through some customized software. I am considering Ganymede.) Although there is a nice package pam-mkhomedir that will automatically create homedirs (and copy /etc/skel stuff) if it does not exist. Where can I find that script? It is a package in potato. I think we're going to go with ldap, so I'm going to have to figure out how we want to handle adding/deleting users, etc. I suppose I'll just whip up some scripts, but I wouldn't mind having a good one as a reference. Yea, I am in exactly the same situation as you are. And I am looking at a good way to administer the users etc. Unfortunately, I can't find a working solution (or one requiring minimal changes). What I gathered so far: - Ganymede: java based administration tool, keeps its own database, would require you to buid your own classes to upload to LDAP. I have not been able to make Ganymede work so far; - Webmin: perl based cgi scripts. LDAP stuff has to be done from scratch; - linuxconf has a ldapconf module. Very early in the development and so far mostly deals with configuring slaps instead of administering the users. Did not manage to test it (guess need more time to play with linuxconf) First I need to look in to ldap and see what the tools are for editing the database from the command line (if that's possible). Sure. There are utilities to do everithing (search, change, add, etc.) They are part of openldap-* packages. There is also very convenient gq (also in patato) which allows you to brows you directory and add/modified entries (though not suitable for real administration). Keep me informed of your findings. Sergey.
Re: Problem with RAID
My guess is that you cuold create a custom kernel with the RAID driver compiled in, and put it on the rescue floppy. If the RAID controller BIOS provides the capability to boot from the array, that should be it. However, this is hipotetical, I never dealt with RAID yet. You might ask for advice at VA (and their Linux Hardware Solution part), they are experinced in installing various distros on RAID. Sergey. On Mon, 7 Jun 1999, Brian Schramm wrote: I am sorry, I cannot set them up from the install disks. I need to boot from them or figure out a way to use the cdrom or floppy to boot and get the raid system working. In my job, puting another HD in th system is out of the question. So to sumerize, I need to access from the install disks and need some way to boot it or create a boot cd/floppy for it. Creating any type of a cd-rom (including a custom unit for installing) is NOT out of the question. I have a Debian system with a cd-rom writer in it so I can do amost anything with it. Brian On Tue, 08 Jun 1999, Brian Schramm wrote: I would like to know how I can install Debian on this machine: DEC HX 6000 server 128 Meg ram Adaptec 2940 running the CD-ROM Mylex DAC960 Raid controller with all the hard drives on it DEC 21x4 network card single CPU I have tried Redhat, Caldera and Debian on this system to no avail. Any ideas on how to install this?
Re: PCMCIA Problem
I installed Debian 2.1 (kernel ver. 2.0.36)on my i386 PC. When I try to load the PCMCIA base modules I get the message 'modules/lib/modules/2.0.36/pcmcia/i82365.o init_module:device or resource busy' and in /var/log/messages 'kernel build:2.0.36 unknown Intel PCIC probe: not found' How can I get throught it? My guess - you have a different type of PCMCIA chip. When you configure the PCMCIA (if I remember correctly, it is pcmcia-config program), it tells you, that there are two differend kinds of chips and asks you to select one. I guess yours is not Intel i82365, but the other one. Sergey.
2.2.9 killed my G200 ?
I upgraded my slink, running kernel 2.2.5 to 2.2.9. After that the kernel started complaining: modprobe: can't locate module char-major-4 And wat's worse, my Millenium G200 AGP died completely (probably because of that.) Does anyone has any idea, what might have coused such a problem ? 2.2.5 worked fine. I've tried to use the same config file, tried xconfig, menueconfig - no change. The system is dual PII-400 with 512MB RAM (though, for some reason kernel only reports: Memory: 257604k/262080k available (1132k kernel code, 420k reserved, 2836k data, 88k init) and Adaptec AIC-7890/1 Ultra2 SCSI host adapter. Another my computer running potato has no problem with 2.2.9. (though it's single p5-100 and no SCSI). Could not find a relevant difference. Any ideas ? Dejanew does not seem to produce a reasonable answer. Sergey.
Re: We need centralized accounts -- Any docs for ldap passwords?
On 31 May 1999, Rob Browning wrote: OK, so it sounds like we just need shadow/passwd/group support, and as far as I can tell we should be mostly good to go if we 1) firewall access to the ldap server from outside our subnet. 2) import etc/group and passwd via migrate_foo.pl 3) edit our nssswitch.conf as directed in /usr/doc/libnss-ldap/README 4) cross our fingers. Well, it seems to work well for me (though so far only on a test machine). What I don't really know is how doing this interacts with the normal mechanisms. I would presume that we can just use LDAP for user accounts, and leave the system accounts in /etc/passwd, etc. That's a logical thing to do. You might also want to set mail-clients use this ldap for mail address searching. I'm guessing from the nsswitch entry it'll just fall back to that if LDAP fails on a given lookup, but how does LDAP interact with adduser, userdel, addgroup, /usr/bin/passwd, etc. Does it update the right things, or do we have to do manual synchs? libpam-ldap will allow password change. The rest have to be done manually (or through some customized software. I am considering Ganymede.) Although there is a nice package pam-mkhomedir that will automatically create homedirs (and copy /etc/skel stuff) if it does not exist. If the latter, then it seems like it might be worth us considering not using LDAP at all, and just whipping up some ssh synch thingy for these bits... You'll sure have to weight various pro and cons of both approaches. Ldap will just allow more things to use it for. Sergey.
Re: Floppy Mounting Problems
On Tue, 1 Jun 1999, Brad wrote: On Mon, 31 May 1999, Eric Gillespie, Jr. wrote: I have a stack of floppies that i'm sorting through, and i need to know what's on them. Some of them, however, aren't formatted. I have know way to know which ones aren't until i mount them. Every time i try to mount an unformatted floppy, mount segfaults. Afterwards, if i try to mount another floppy it says mount: /dev/fd0 already mounted or /floppy busy. I can mount a floppy to other directories, so it must be /floppy that's busy, even though /etc/mtab and /proc/mounts both show it as not mounted. Further, i can't remove the floppy module from the kernel. Eventually i run out of directories to mount to and have to reboot. This is unacceptable. Any ideas? Did you try umounting /floppy? I had a similar proble. But with a corrupted floppies. Trying to access a file that happens to be too long produces and error (something like: VFS: trying to access outside the filesystem), and after that there is no way to unmount the floppy. Had to reboot. Does anyone know a way to force unmounting in such a case ? Sergey.
maildirs in pine
I wonder, did anyone manage to make pine support maildir reasnably well ? So that it shows maildirs as just mailboxes and creates new mailboxes by default in maildir format. I've got pine-4.10 from project/experimental. It is capable of reading maildirs, but a) in order to create maildir box foo you have to enter #driver.maildir/foo and it creates them in $HOME instead of $HOME/mail, so you'll have to enter #driver.maildir/mail/foo. b) It lists maildir folders twice, and with a slash in square brackets, like foo[/]. When you press Enter, it correctly brings you into the content. But if you press it instead just cd into that dir. So, can pine be configured correctly, or I'm just waisting my time with maildirs and should just forget about them ? What do you, guys, think ? Sergey.
Re: We need centralized accounts -- Any docs for ldap passwords?
On 13 May 1999, Rob Browning wrote: Jens B. Jorgensen [EMAIL PROTECTED] writes: But if you're not ready for potato then NIS will provide a ready-made solution. It's pretty straightforward. I'd be glad to offer assistance. As for a comparison, well, they're different. NIS has been around a long time, LDAP is newer. Thanks to both of you for the help. We're running all unstable here so getting the ldap packages isn't a problem, but I wasn't completely sure what the tradeoffs are. I'll go look at the web sites Ben mentioned. Actually I had already seen the ldap pacakges, but I wasn't quite sure where to start. I was hoping for a HOWTO or something, but I can always just jump in and figure it out as I poke around. When you install libnss-ldap, there is a short howto in /usr/doc/libnss-ldap I also suggest downloading conversion tools from www.padl.com, which will help populate the LDAP database Sergey.
Re: Cobalt Qube
On Tue, 4 May 1999, G. Crimp wrote: I'm also wondering about mixing architechtures. This guy wants to sit in front of a quiet diskless box at his desk (not entirely solved yet) Why not just keep the existing Sun box and just use it as X-terminal. You can either keep Solaris or install Linux on it. that runs all apps across the net from another box (the Qube, maybe) sitting in another room. You'd probably want something more powerful here. If you have a ix86 diskless box on the desk, is it going to be able to run apps served from a RISC port of Linux ? What you mean served ? You can run applications on RISC and display them on your local worstation. Any architecture. Could one put x86 binaries onto the RISC harddrive and have them served to the diskless box ? Sure. Put them in a separate directory , export it via NFS. Sergey.
Re: KDE and .Xdefaults
On Tue, 27 Apr 1999, Ralf Comtesse wrote: Hi, I just installed KDE from ftp.kde.org. It is the slink version. My box is running potato. The main problem at the moment is, that on startup neither my .Xdefaults nor my .xinitrc is read. I applied the changes to /etc/X11/Xsession as described in the kdm pages. But the defaults or inits are not read. Saying that, I am not quite sure about that, either, since ls -l on my .xsession-errors shows the actual date. But the logs show only the last entries, so I do not know, what commands are executed. So: 1. Am I missing sth? What do I have to do to read .Xdefaults of .xinitrc The best solution I've found: Create an applink in Desktop/Autostart (i.e. New-Application). In the Execute line put xrdb -load $HOME/.Xdefaults and another one for .xinitrc. By the way, xrdb -merge does not work well, xrdb -load is ok. Sergey.
Re: KDE and .Xdefaults
On Thu, 29 Apr 1999 [EMAIL PROTECTED] wrote: In a message dated 4/29/99 12:06:22 AM Central Daylight Time, [EMAIL PROTECTED] writes: The best solution I've found: Create an applink in Desktop/Autostart (i.e. New-Application). In the Execute line put xrdb -load $HOME/.Xdefaults and another one for .xinitrc. By the way, xrdb -merge does not work well, xrdb -load is ok. What exactly does -merge does not work well mean? I've been using it with no problems so far, is there something I should watch out for? It means some settings have no effect - KDE overrides them. Notably fonts and color settings in emacs do not work with -merge but do with -load. Again, this is KDE-specific problem. Sergey.
Re: trouble with LILO
I had similar problem (actually also on 486 33 MHz.) I've solved it by overwriting master boot record. I.e. put in /etc/lilo.conf boot=/dev/hda instead of boot=/dev/hda1 and run lilo again. That should fix it. Sergey. On Fri, 16 Apr 1999, Brian Morgan wrote: I just got done installing Debian 2.1 on a 486 33mhz machine. When rebooting after the install, it hangs on LILO, with the LI appearing on the screen. doesn't ever go anywhere after that. Booting from the boot floppy appears to work fine. Is there any way to correct this? I've installed the 2.0 version on identical machines (Digital DECpc LPv+ 433 sx) with no problems like this. Any suggestions? Brian Morgan http://brian.greenville.edu Computer Support Specialist [EMAIL PROTECTED] IBM Mobile Systems Support618.664.2800 ext. 4241 Greenville College IT Dept. 618.338.4963 - pager __ The end of the world is in late Beta ... no, hang on. Bill Gates just bought it . . . we're safe again.
Re: Hardware advice
I've got a Pentium 75 overclocked to 100MHz, with 64MB RAM and 17 monitor. Works extremely well. Sergey. On Wed, 14 Apr 1999, Brad Stevenson wrote: Hello, I'm just getting started using Linux. Actually, I'm expecting my Slink = CD's to arrive any day now. I'm upgrading an old PC I have to install = Debian on. I'm on a real tight budget, but I don't want to spend my = money on hardware that won't be appropriate. I've found a 1.2 G hard drive for a reasonable price. I know that this = is more than adequate size for the OS, but is it a reasonable size for = a typical system. Basically, do you people who use this OS regularly = recommend a bigger drive?=20 The system I intend to install this on is a 60Mhz Pentium with 16 M ram. = I realize that this would be the minimum for a reasonable Win 95 setup, = but is it reasonable for a Linux setup?=20 Brad Stevenson, CET Product Development Support NCA Microelectronics 506-634-5014 http://www.ncamicro.com
Re: HP Web JetAdmin
On Sun, 11 Apr 1999, Mark Blunier wrote: What kind of problems you have with printing ? I'm not sure about lpr, but lprng works fine by putting :lp=yourprintername.yourdomain%9100 We have hp jetdirect boxes with ethernet connections, but they use print through a novel net interface. I use nprint to print to these. Don't these boxes support TCP/IP as well ? Sergey.
Re: HP Web JetAdmin
On Sun, 11 Apr 1999, John Stevenson wrote: If anyone gets this going I would like to hear about it, also if they can actually print successfuly using linux to a HP printer with a direct ethernet connection. What kind of problems you have with printing ? I'm not sure about lpr, but lprng works fine by putting :lp=yourprintername.yourdomain%9100 I also found that the best filer for HP printers is ifhp, not the CTI-ifhp provided by debian, but newer ifhp from lprng home site www.astart.com. It can act as a magicfilter, supports duplex-simplex selection and trays, gets page count etc. Wish someone had packaged it (don't have time myself) for debian. Sergey.
Re: Netscape and *.png files
I've got a similar problem but with *.pdf files. The solution was found looking at this list archives. Uncomment the following line in the file /etc/netscape4/config NO_LD_PRELOAD=yes Sergey. On Mon, 5 Apr 1999, Christopher Swingley wrote: Hello! I'm having trouble getting Netscape to read portable network graphic (*.png) files. I've tried Navigator 4.08 and 4.51 via the netscape4 package and in both cases netscape simply quits with a Bus error when I try to open a *.png file in the browser. I've got libpng0g and libpng2 installed, and *.png files work fine with the Gimp 1.0.3 and ImageMagick. Thanks! Chris -- ++--+ | Christopher S. Swingley| 108 O'Neill Building --x7092 | | Programmer / Analyst | 940 Sheep Creek Rd -- 479-9729 | | IARC - Frontier| [EMAIL PROTECTED] (email) | | University of Alaska Fairbanks | www.frontier.iarc.uaf.edu:8080/ | | Fairbanks, AK 99775 |~cswingle (web) | ++--+
Re: What DO you lose with Linux ???
I'll tell you what I'm loosing in linux. More precisely, what I think is the main problem with linux. It is lack of convenient and integrated support for fonts and printing. Fonts: if you can see a font, does not mean you can print it; if you can print with font, does not mean you can see it on the screen. Especially international. Too many formats, and each application has to provide its own fonts: X, tetex, staroffice, wordperfect, abiword, ghostscript - all have separate fonts. TrueType fons are accessible through xfstt, but you still can't print them and xfstt reports incorrect character set. Printing: It works acceptable when you have a simple dumb printer, but when you've got a $3000 beast there is no simple way to use its capabilities: no way to select 1-or-2-sided printing, a paper tray, etc. With Windows, Mac and SGI - no problem, just choose the appropriate option. The only way I see in linux - is to create a buch of queues for each combination of options. It would be cool to have a SGI's glp-like utility, but it seem to need a SysV type printing. Sergey.
Re: Debian on IBM thinkpad 380XD
Well, a month ago I've installd slink on ThinkPad 385D. The tecra disks did not help. I ended up compiling a custom kernel as zImage on another workstation and replacing the kernel image and modules on installation disks. If you need, I can try to send you these floppy images, if they are still alive. Sergey. On Mon, 15 Mar 1999, Brian Morgan wrote: Anyone have info on installing Debian on an IBM thinkpad 380XD? I'm installing the latest 2.1 version, and not having much luck with the rescue disk. It comes to the boot: screen. I press enter, and then it reboots the system. Repeats with different rescue disks on different thinkpads (all 380's). Thanks, Brian Brian Morgan http://brian.greenville.edu Computer Support Specialist [EMAIL PROTECTED] IBM Mobile Systems Support618.664.2800 ext. 4241 Greenville College IT Dept. 618.338.4963 - pager __ Never criticize anybody until you have walked a mile in their shoes, because by that time you will be a mile away and have their shoes.
SGI's glp analog ?
I wonder if there is a program for linux analogous to SGI's glp ? I.e. i need an easy and convenient method to select such parameters as whether to print one- or two-sided, flip horizontally or vertically, the paper size (the input tray), whether to print a banner, etc. Sergey.
Re: kdm
On Tue, 9 Mar 1999, Noah L. Meyerhans wrote: -BEGIN PGP SIGNED MESSAGE- On Tue, 9 Mar 1999, Alec Smith wrote: Difference is kdm is for starting the KDE environment only. xdm starts a more traditional setup using fvwm2, windowmaker, or whichever you prefer. Actually, that is quite untrue. Kdm provides a menu from which you can select the session that you wish to run (KDE, Window Maker, FVWM, etc). Kdm also provides the ability to shutdown or reboot your system from the login screen. It can also provide icons for the available usernames on the system (like the SGI login). I actually prefer wdm, the WINGS Desktop Manager to either. It has about the same functionality as KDM, but it is built with WINGS, the Window Maker interface, so it looks quite nice. It's included with slink. There is one very nice feature in kdm in 1.1 that is unfortunately absent in wdm: kdm remembers your last session, when you type your username it switches the menue to the last used session, so that you don't have to click every time you login (if you prefere non-default one). Of course, it is not importain for a single user. Sergey.
Re: xdm kdm
Hi, I have got kde installed, but I never configured my systems to use kdm. In my /etc/X11/config, I still have the line start-xdm. and even I ^^^ That's the reason. If you want to start kdm instead, change it to start-kdm. do a ps, I can only see xdm running, and no kdm. But when I shutdown my linux box, I got the message: Shut down the kdm display manager It just says it executes the /etc/init.d/kdm script (which actually checks for the start-kdm line on startup but does not on shutdown) checking the local xservers Shut down the xdm display manager xdm is not running Here the kdm script already has stopped the xdm. I have to give you a little warning though. If you upgrade to slink, it does not have /etc/X11/config file. Instead, xdm is started as soon as you install the xdm package. In order to switch to kdm you'll need to: 1) remove original /etc/init.d/kdm (it still uses that config file) and copy it from /etc/init.d/xdm. 2) edit the new /etc/init.d/kdm to replace xdm with kdm. 3) disable original /etc/init.d/xdm either by inserting exit 0 somewhere in the beginning of the script, or removing links to xdm from /etc/rc.*/ Sergey.
Re: StarOffice 5
On Tue, 2 Mar 1999, Robert V. MacQuarrie wrote: I've installed each version of SO and have never had any problems using the provided 'setup' utility. I really only have 1 minor complaint with SO5 and thats the fact that it takes up so much disk spack in the users account. I remember one of the SO developers had hacked a multiuser patch for the in SO4 called AutoMultiUserSetupForSO40Linux.tar.gz I think. This patched the /net option if i remember correctly and worked beautifully. Here are the diffs in size of my SO5 and SO4 installs: [timberwolf:adren:~$] du -hs /usr/local/Office40/ 113MOffice40 [timberwolf:adren:~$] du -hs ~/StarOffice (this is 4.0) 12M StarOffice [timberwolf:adren:~$] du -hs ~/Office50 140MOffice50 [timberwolf:adren:~$] This is a HUGE difference. The multiuser patch was great! 12M for 4.0 to 140M with 5.0. Does anyone know if there was a multiuser patch for SO5 aswell? I like having a 12M user dir as opposed to a 140M dir. You don't need patch, /net option is already there, just use it. In the home directory it uses only 4 MB. Sergey.
Maxtor dynamic drive overlay
I wonder if anyone knows what is Maxtor Dynamic Drive Overlay and how to get rid of it. It seems to be a part of the hard drive. And it causes my boot to take about 3 minutes (to get to lilo). I don't have such problems with other hard drives and I did not have such a problem with this one before I've installed Linux. Any ideas ? Sergey.
Re: KDE 1.1 don´t install under hamm
Actually, they've fixed the packages today (version 2 now). Just installed. Seems to be OK, except that the kdm startup script is incompatible with slink (it wants the start-kdm line /etc/X11/options file, which is absent in slink). So I've copied from xdm and modified it. Sergey. On Wed, 10 Feb 1999, Stephen Pitts wrote: On Wed, Feb 10, 1999 at 12:33:47PM -0400, Pablo Lorenzzoni wrote: Hi friends! This is the first time I write here. Sorry for it is a problem that bring me to write this one, but... Debian life is Debian life... I?ve installed Debian 2.0 about a month ago, and I love it. It is very much better than Slackware that I ran before, or RH that I tried before Debian. I?ve installed Gnome (0.30) which I get from ftp.gnome.org in .deb format. It worked great, but it was not KDE (which I love), so I decided to install KDE. I downloaded KDE 1.1 from ftp.varesearch.com which I get in .deb format too. When trying to install (dpkg -i kdelibs2g_1..) it returns a message that sounds like subprocess: gzip -dc: it isn?t in gzip format or something like that, and quit. Can someone help me? []s, TIA Pablo Lorenzzoni Get free e-mail and a permanent address at http://www.amexmail.com/?A=1 -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null Same problem happened to me. I downloaded it from ftp.varesearch.com and ftp.kde.org. The MD5 sums computed OK too. Searching through the kde mailing list archives, I saw one other person with the same error message. Perhaps someone who is 'in the know' about KDE in .deb format could reply? -- Stephen Pitts [EMAIL PROTECTED] webmaster - http://www.mschess.org
Re: Slink, xbase-common and xdm...
How about just pressing Ctrl-Alt-F1 and login through the text console ? AFAIK, you problem is a known bug - missing ;; in the /etc/X11/Xsession. Hopefully it will get fixed soon. Sergey. On Tue, 9 Feb 1999, Hogland, Thomas E. wrote: Interesting problem - I've upgraded to slink as part of kernel 2.2.1 on two different PC's, and I think I goofed on one. When the new X stuff went in and asked if I should use my current config files, the PC at work was told No - leave mine alone; the PC at home was told Yes - overwrite with new ones. PC at work is fine, but PC at home now immediately boots into xdm and refuses to let me log in... Won't take root, me, nothing. Tried booting from the boot disk I made - slowly boots, then starts xdm :-(...Anyone know how to kill this? Even Ctl-Alt-Backspace just restarts it...
README files for XF86Setup in slink
Hey guys, where are those /usr/X11R6/lib/X11/doc/README.* files required for XF86Setup in slink ? Are they in a separate package ? Which ? I can't find them. Or is this just a bug ? Anyone's aware and working on it ? Sergey.
Re: xfstt - Confusion?
Well, I bet you have xfstt package from hamm. It was old and broken. Get xfstt from slink - it should work fine. The actual xfstt executable is indeeed located in /usr/X11R6/bin. In /etc/init.d thereis a script that is used to start and stop the daemon. Another thing: the slink version of xfstt keeps fonts in /usr/share/fonts/truetype Good luck. Sergey. On Sat, 26 Dec 1998, John wrote: Curt wrote: Try this: execute (as root) /etc/init.d/xfstt start. This works for me. I've been too lazy to seek a fix for this since I so seldom reboot. Curt Daugaard [EMAIL PROTECTED] I tried Curt's suggestion (thank you Curt), but for some reason I don't have xfstt in the directory /etc/init.d. On my computer xfstt is located in /usr/X11R6/bin. Is this correct? Or, do I not have it installed correctly? I then tried executing the file with /usr/X11R6/bin/xfstt start - that didn't work. The only options I can find by using --help are -- sync, -- port portno and --unstrap That having failed, I tried every options I could think of. I know I tried some I shouldn't have cause fsck sure had fun straightening out my drive after I hung it up. :) Anyone know what I'm doing wrong?? Thanks John On Fri, Dec 25, 1998 at 03:25:17PM -0600, John wrote: Merry Christmas All! After reading several recent postings, I decided to try installing some true type fonts using xfstt. I did the following: 1. copied some .ttf fonts from my windows95 disk to /var/ttfonts 2. ran xfstt --sync 3. edited XF86Config by adding FontPath unix/:7101 as the next line in the area of XF86Config that has all the FontPath directories listed. 4. quit xwindows (fvwm95) 5. ran startx - it failed with the following error _FontTransSocketUNIXConnect: Can't connect: errno = 2 failed to set default font path ' listed all the font paths including unix/:7101 fatal server error: could not open default font fixed etc,etc I then tried the command xset +fp unix/:7101 - it failed with the following: 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. My main confusion is about the line unix/:7101. Every other entry for fonts is a path and starts with / this one doesn't. Is it a path or somthing I'm not familiar with? also which fonts.dir is it searching for? I have several on my system, but none called either unix or 7101 Have I missed something obvious? I'm running hamm if that makes a difference. Thanks for any hints. John
Re: kde can't find the library...
Where did you get the kde ? Is it the pre1.1 form ftp.timsnet.com ? If it is, you'd need to get these several files from somewhere else - these packages are broken. E.g. get an RPMs from ftp.kde.org and take required files from there. Sergey. On Mon, 21 Dec 1998, Daniel Elenius wrote: Shao Zhang writes: Hi, for some reason, kde can't find the lib /usr/X11R6/lib... It complains something about: kaudioserver: error in loading shared libraries libmediatool.so.0: cannot open shared object file: No such file or directory kfm: error in loading shared libraries libkhtmlw.so.0: cannot open shared object file: No such file or directory kcontrol: error in loading shared libraries libkdeui.so.0: cannot open shared object file: No such file or directory if I do an export of LD_LIBRARY_PATH wiht /usr/X11R6/lib, then it is fine... but if I put in kde in my .xsession, it won't run again... complains the same problem... my ld.so.conf has the following: /usr/X11R6/lib /usr/X11R6/lib/Xaw3d /usr/lib/libc5-compat /lib/libc5-compat Could someone tell me where I got wrong?? You should put 'exec startkde' as the last line in your .xsession. You may also need to run ldconfig. Then start x with 'startx'. With xdm you probably have to put that line in .xinitrc, but I don't run xmd, so I don't know. But I think your problem comes from the fact that you try to run 'kde' or 'startkde' from the console. Am I right. 'kde/startkde' are X programs, and therefore have to be put in .xsession/.xinitrc. Let me know if the problem persists, 'cause I had the same problem, and I fixed in on my machine.
RE:KDE and kppp
I am not 100% sure this is the reason, can you just run pon as a normal user ? If not, add yourself to the group dip (add you username to the line dip:x:30:yourusernamehere ) When you can run pon and poff as a nirmal user, then see if kppp works as well. That worked in my case. Sergey. On Mon, 7 Dec 1998, John Greer wrote: I have kde installed on my machine and seem to have everything installed and working OK. The only thing that is a hang-up right now is that I am unable to run the kppp program as a normal user. It gives an error message that the setuid script is unable to run and this I can not start pppd. Does anyone have any clues on this??? John John Greer Information Systems Specialist Mansfield Library The University of Montana (406)243-6063 Knowledge is the only fountain both of the love and the principles of human liberty. Completion of Bunker Hill Monument, June 17, 1843. P. 93. -Daniel Webster
Re: installing linux debian on Toshiba Satellite
I've installed hamm on Toshiba Tecra 700. tecra disk did not help - it seems to still use bzImage while zImage is needed. What I did (after seeing advice on this list) was to put kernel on a dos partition and use loadlin.exe (and the debian install.bat). And after I install everything, I still had to compile my own kernel (to make zImage). I don't know what the status of boot disks for slink, I never managed to boot them (it was a week ago). For the explanations of why zImage is needed see Manoj's message several days ago (about ThinkPad). Sergey. On Mon, 7 Dec 1998, Joey Hess wrote: Use the tecra boot disks. I have debian installed happily on my toshiba sattellite 445cdx, btw. There are even some debian packages (toshiba-hotkey, toshiba-fan) specifically for these laptops. Vazquez Jr, E A wrote: Don't expect much help here, I've been trying for DAYS to get a response to my questions but apparently no-one likes laptop installs. - Ed I'm trying to install debian on a Toshiba Satellite. I get as far as booting linux.. but then the Toshiba logo comes up and the cycle restarts. I'm sure I read a post recently about this but I can't remember what it was. There must be a simple way of getting round this -- any ideas? Anthony -- Anthony Campbell - running Linux Debian 2.0 [EMAIL PROTECTED] http://www.achc.demon.co.uk The Moving Finger writes; and, having writ, Moves on... - Edward Fitzgerald -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- see shy jo
Re: xfstt font database should be in /var
On Tue, 3 Nov 1998, Stephen J. Carpenter wrote: well this si a FIXME of sorts) :) xfstt currently requires write acess to a font database it stores (which by all rights should be in /var) as it stands this is in the same directory as the fonts... Why exactly xfstt requires write access when it is invoked with --daemon option and does not when without ? I don't see any reason to write unless we do --sync. Moreover, it should not quit when the write fails (again, unless ussued --sync option). May be you'd better correct that and leave database in place (/usr/share). Sergey.
Re: xfstt + xset
Use ps aux to check that xfstt is actually running. I've encountered a wierd problem today that xfstt by itself works, but xfstt --daemon does not. Very strange - it worked yesterday. I don't think I've changed anything important. Anyone got a similar problem ? Sergey. On 3 Nov 1998, Gossamer wrote: I've just installed xfstt and a bunch of truetype fonts. When I run 'xfstt --sync' it works fine, and startind the daemon from the /etc/init.d/xfstt script works. But I can't get the server to talk to it: [EMAIL PROTECTED] ~/cs3/ai ps aux | grep xfstt root 227 7.4 1.3 1332 520 ? S12:53 0:50 /usr/X11R6/bin/xfstt [EMAIL PROTECTED] ~/cs3/ai xset fp+ unix:/7101 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 Any ideas? bekj -- : --Neophilic-Hacker-Grrl-Geek-Eclectic-Gay-Disabled-Boychick-- : [EMAIL PROTECTED] http://www.tertius.net.au/~gossamer/ : The POP3 server service depends on the SMTP server service, : which failed to start because of the following error: The : operation completed successfully. -- Windows NT Server v3.51
Re: xfstt + xset
On Tue, 3 Nov 1998, Stephen J. Carpenter wrote: On Tue, Nov 03, 1998 at 12:07:05AM -0500, Sergey V Kovalyov wrote: Use ps aux to check that xfstt is actually running. I've encountered a wierd problem today that xfstt by itself works, but xfstt --daemon does not. Very strange - it worked yesterday. I don't think I've changed anything important. Anyone got a similar problem ? Wow someone already using -daemon option (I added it and rushed it in to the upstream author very shortly before releace :) ) anyway...it is -daemon NOT --daemon yes this is counter-intuitive to the way MANY programs work but... ALL o fth eother xfstt options use only 1 '-' and ALL are long options. I just followed the convention the author used. Try it an let me know. I am using xfstt 0.9.10-1 and the --daemon option is actually part of /etc/init.d/xfstt script, I don't use it manually. On the attempt to issue only one dash -daemon it complains of the wrong syntax. The problem is that starting xfstt manually works fine, but starting it from init.d does not. Neither does xfstt --daemon. And I'm not sure what's wrong. Sergey.
Re: xfstt + xset
Even more wierd: I have a server and a client, the client mounts /usr via nfs from server; /var is local, the rest is syncronized via cfengine. /etc/init.d/xfstt start works on the server, does not on the client. Just xfstt works everywhere. Any idea what the so subtle difference between these machines might be? Sergey. On Tue, 3 Nov 1998, Stephen J. Carpenter wrote: On Tue, Nov 03, 1998 at 12:07:05AM -0500, Sergey V Kovalyov wrote: Use ps aux to check that xfstt is actually running. I've encountered a wierd problem today that xfstt by itself works, but xfstt --daemon does not. Very strange - it worked yesterday. I don't think I've changed anything important. Anyone got a similar problem ? Wow someone already using -daemon option (I added it and rushed it in to the upstream author very shortly before releace :) ) anyway...it is -daemon NOT --daemon yes this is counter-intuitive to the way MANY programs work but... ALL o fth eother xfstt options use only 1 '-' and ALL are long options. I just followed the convention the author used. Try it an let me know. -Steve Sergey. On 3 Nov 1998, Gossamer wrote: I've just installed xfstt and a bunch of truetype fonts. When I run 'xfstt --sync' it works fine, and startind the daemon from the /etc/init.d/xfstt script works. But I can't get the server to talk to it: [EMAIL PROTECTED] ~/cs3/ai ps aux | grep xfstt root 227 7.4 1.3 1332 520 ? S12:53 0:50 /usr/X11R6/bin/xfstt [EMAIL PROTECTED] ~/cs3/ai xset fp+ unix:/7101 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 Any ideas? bekj -- : --Neophilic-Hacker-Grrl-Geek-Eclectic-Gay-Disabled-Boychick-- : [EMAIL PROTECTED] http://www.tertius.net.au/~gossamer/ : The POP3 server service depends on the SMTP server service, : which failed to start because of the following error: The : operation completed successfully. -- Windows NT Server v3.51 -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- /* -- Stephen Carpenter [EMAIL PROTECTED] --- [EMAIL PROTECTED] */ We do everything by custom, even belive by it; our very axioms, let us boast of free-thinking as we may, are oftenest simply such beliefs as we have never questioned --Thomas Carlyle
Re: xfstt + xset
On Tue, 3 Nov 1998, Stephen J. Carpenter wrote: On Tue, Nov 03, 1998 at 12:11:20PM -0500, Sergey V Kovalyov wrote: Even more wierd: I have a server and a client, the client mounts /usr via nfs from server; /var is local, the rest is syncronized via cfengine. /etc/init.d/xfstt start works on the server, does not on the client. Just xfstt works everywhere. Any idea what the so subtle difference between these machines might be? Ahhh well this si a FIXME of sorts) :) xfstt currently requires write acess to a font database it stores (which by all rights should be in /var) as it stands this is in the same directory as the fonts... Yep, I've just come to the same conclusion. And you can guess I mount /usr ro. this same thing happens when you run xfstt as a normal user. With NFS on you probably have root_squash on (which is the default) so root on the client can not write tot he database! sorry for the inconvinece...I should work on this! any ideas on where the font database should be? How about /var/cache/xfstt ? Sergey.
/var/lib
I have trouble understanding /var/lib. I'm trying to set up a network of debian workstations. They will have /usr nfs-mounted from a server (ro), local /var and the / copied over using cfengine (excludung hardware-dependent files). The question is what should I do with /var/lib ? Is it something that can be generated locally ? Does not seem to be: tex requires quite a lot of files in there in order to work. Can I mount it ro via nfs ? But then why is it /var ? Presumably I need a write access there. Do I have to distribute the contents of /var/lib to all clients ? But that's quite a lot (I see 25Mb), and then I can accidentally overwrite local files - so this is no better than nfs. So, what is /var/lib ? What should I do with it ? Sergey.
Re: OffTopic: Synchronization of /etc or .dot-files... CVS?
Take a look at cfengine package in slink. Sergey. On Thu, 29 Oct 1998, Alexander Kushnirenko wrote: Hi, It's a little bit off-topic. We have a set of Debian computers, which are not quite the same. It would be useful to synchronize certain files on all nodes like /etc/profile or /etc/X11/Xresources, or /etc/menu and so on. But of course not all the files. Reflection of this problem is synchronization of user's .dot-files on different nodes. Would CVS be an appropriate tool for that purpose? Or is there better one tool? Thanks, Sasha.
kvt and xterm-color
The KDE's kvt keeps setting TERM to xterm-color, which is no good since other systems do not like it. I'd like to make it just xterm on a system-wise basis (so that every user does not have to do it manually). Do you know which file is to be modified ? What puzzles me though is that another box that I've setup a month ago (also slink) does not have the problem and gives plain xterm, and I don't recall changing anything. I could not figure not what is different. Any ideas ? Sergey.
exim on a smarthost configuration
A server exports /var/spool/mail via NFS and acts as a smarthost for the satellites to relay their messages to outside. Everything works except for relaying. Server refuses to relay with the message: 550 relaying to whatever address prohibited by aministrator How should I setup /etc/exim.conf to enable relaying from the satellites (and only from satellites or at least only from within local domain) to anywhere. Thanks, Sergey.
Re: xterm configuration
Install packages: xaw-wrappers, xaw3dg and/or xaw95g - you'll get a nice 3D scrollbar in xterm, ghostview etc. Sergey. On Fri, 30 Oct 1998, tony mollica wrote: Hi. Can't seem to find a way to change the configuration of my xterms in fvwm2. I would like to change the default scrollbars from the hatched display to a more solid type. I've been through a lot of documentation and I've either missed it or haven't found the right info yet. Can someone suggest a way to do this? (debian2.0, 2.0.34) thanks, tony mollica [EMAIL PROTECTED]
Re: Catch-22 - Help!
On Wed, 28 Oct 1998, Jerry E. McGoveran wrote: At 06:37 PM 10/28/98 -0600, Andy Kennedy wrote: On Wed, 28 Oct 1998, Jerry E. McGoveran wrote: I have a Debian installation in which the ethernet card driver is apparently not installed. There were some errors at this stage of the installation process, but the screen drew and reset to the inst. menu too fast to read anything. I finished the installation, and now I need to update the kernel and the drivers. I can't compile a kernel (downloaded 2.0.35 src) because I don't have gcc. I can't install the gcc package because I don't have network access in Linux - only under Win95.A Gcc is available at sunsite.unc.edu and most of the mirrors. Download it w/ win95 and then use tar to unpack it. Oh. I had assumed that it had to be installed via the package route. Silly me. The package route is still better, but you don't have to use dselect. Download packages using Win95, then mount vfat partition from within linux and use dpkg -i pakcagename.deb to install it. Sergey.
Re: Printing to a Novell network printer
One option is to install Novell-nfs on the Netware server. It includes lpd. At least that what we do at NYU to enable Windows, Mac and Unix to print. Sergey. On Wed, 28 Oct 1998, Clovis Sena wrote: Hi all, so is that possible? How can I print to Novell network printer, where i put the queues? NOTE: I work in a environment that includes Win95+dos6.22 as clients, Win NT and Novell 3.12 as servers ( running a mix of database+ intranet nad internet access + mail(most uses outlooks/messenger) ) ; so I would like to introduce Linux as a option as server ( files + web services) as workstation ( that needs access all of these resources. Its driving me crasy! So I need helP :).
more on ( mount NFS during boot problem )
Even more wierd, I've reinstalled kernel-image (2.0.35 right from slink) the NFS started working, but dhcpcd stopped - just gives me IP 0.0.0.0 Reinstallation of dhcpcd did not help. Any ideas ? Sergey. On Fri, 23 Oct 1998, Sergey V Kovalyov wrote: Hi, I have a wierd problem: during boot process mount fails with the following message: Starting portmapper... Mounting remote filesystems... mount: RPC: Program not registered But after the boot mount -a works perfectly. This runs latest slink from unstable. I did not have such a problem on a similar machine a month before. Any idea of what might cause the problem and how to work around it ? Something, I think, relevant: later in the process netbase starts its own portmapper. Sergey.
mount NFS during boot problem
Hi, I have a wierd problem: during boot process mount fails with the following message: Starting portmapper... Mounting remote filesystems... mount: RPC: Program not registered But after the boot mount -a works perfectly. This runs latest slink from unstable. I did not have such a problem on a similar machine a month before. Any idea of what might cause the problem and how to work around it ? Something, I think, relevant: later in the process netbase starts its own portmapper. Sergey.
Re: Hi/True Color mode for X ! please help
On Mon, 19 Oct 1998, Linh Dang wrote: I have just installed ham from SLS. Everything is fine except X in Hi (16bits) or True (24bits) color. I have a Sony CPD 200sf and a PCI ATI 3D Rage II+ with 8Mb on board. I have an excellent [EMAIL PROTECTED] in 8bits depth but when I switch to a 16/24/32 bits mode, I get an 1-inch-wide column of noise at 1/4 of the screen on the left side. All the other area of the screen is OK except for that column. - I configured X with XF86Setup and tuned it with xvidtune. Try launch xvidtune from an xterm and see whether small changes can solve the problem. Usually it helps in such an intermittent case. You can also try to look at /etc/X11/XF86Config to see if there is anything suspisious. - Windog95 has no problem in [EMAIL PROTECTED] in 16bits. Try to dig through Win95 files. Some video drivers actually have a configuration file with timings (I have such a file for my Matrox Millenium II, so I've made timings in Linux exactly the same as in Win95). Another suggestion: rename your XF86Config and run xf86config program. It usually produces somewhat different configuration. Use the best video timings of these two methods. Sergey.
have anyone got dhcpcd working ?
Hi, I wonder, have anyone manage to get a working dhcpcd ? I've tried to install it once, but the package seemd to be broken. Installation script complained about absence of /etc/dhcp directory but refused to create one, so I had to do it by hand and reinstall. Then the daemon itself could not create a socked - it requested type 17, which even did not have a name and hence the kernel tried to load non-existing module, something like net-11 Finally, I gave up. So I wonder - did I miss something, or it is indeed broken? Thanks, Sergey.
Install roblem (SCSI ?)
Here is a tough one I'm afraid (have not found any answer on dejanews). I was trying to install hamm onto an old Dell 486SX/33 server, and upon boot the kernel gives the following error (I tried bo installation disks - the same result): scsi0: Encountered spurious interrupt scsi0: BRKADRINT error (0x1): Illegal Host Access Unable to handle kernel NULL pointer dereference at virtual address c000 current-tss.cr3=00101000, %cr3=00101000 *pde=00102067 *pte= Oops= CPU:0 EIP: 0010:[001bda4f] EFLAG: 00010206 blablabla... STACK .. After that the hamm version just freezes completely, while the bo still allows fo scrolling with Shift-PgUp. According to Win95, which BTW works on this machine fine, it has Adaptec AIC-777x EISA SCSI Host Adapter io: 8c00-8cff, irq 11, mem: D-d3fff SMC EtherCard PLUS Elite 16T (WD/8073W) io: 0300-031f, irq 10, mem: cc000-ccfff Tseng Labs ET4000 display adapter well, it also complained about lack of BIOS32 extension but I think it's irrelevant. So, any idea on how to get around ? Can I just specify something suitable on the boot command line, or do I have to create a custom install disk (what to watch for in this case ?) Please, cc me directly to [EMAIL PROTECTED] because I subscribed to digest only. Thanks, Sergey. -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
