Linux Router Project - Initial beta release now available.
After much work I have just made the first beta available. It is dubbed a developers release, but it is very much a usable product for anyone that needs to build a router/terminal server now. (I'm using it in my equivalent to a Portmaster 2e) What has been made is a networking capable minimal root fs that fits on most any small boot medium. (including a 1.44mb floppy!) Once made the boot medium is self contained and upgradable, using a unique work around that allows us to keep the root fs in multpile tgz's instead of the typicial boot disk style of having a single, unmodifiable binary image. The base distribution for this release has been Debian 1.3.1.?, and I have managed to keep the full sysvinit and layout 99% Debian. Now that this base is done we will begin implententing snmp and web based configuration. The root already contains telnetd and tftpd support. (We could sure use more people...join the mailing list!) You can download the current beta via web or ftp: http://www.psychosis.com/linux-router/ ftp.psychosis.com /pub/linux-router/ - http://www.psychosis.com/emc/ Elite MicroComputers 908-541-4214 http://www.psychosis.com/linux-router/ Linux Router Project -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Debian + PC with multi RS... port - n x (text dumb terminal + printer)
Thank you once again, you are so kind to me, but after sending _I_ realized that I may have been cryptic, I'm so sorry... I don't want to use the same output that I send to the display in order to drive the printer, just the opposite thing: I wonder whether or not I can have a Linux box talk SEPARATELY though via ONE ONLY SERIAL LINE with a console AND a printer; or do I need two serial lines and two separate serial ports? I my previous two messages I was meaning: sending output to the screen and getting input from the console - from the point of view of the code of the application running - does not change anything if the session is on a dumb terminal rather than on the main console (*), so now: 1) what about sending output TO A PRINTER which is NOT ON A PARALLEL PORT (lpt1, lpt2, ...) but is instead on a serial line? (Maybe I just need to write software which sends output to its stdout instead of stdprn, and that output is merely redirected to a com port... maybe such multiuser environments have typical and by now traditional solutions to my question... that's why I said I lack the basics.) 2) (already reproposed above) could that serial line be the same that goes to the dumb terminal? If yes, would the OS menage the distinction between the two devices or should _my_application_ (or some wrapper) be aware of the hardware each session runs on and eventually take care of sending special characters to say this goes to the display, this goes to the printer or should I just have two distinct cables run from the Linux box to the place where the dumb terminal and the printer are? ( (*) Nothing changes both if your application just appends output at the bottom of the screen using stdout/stderr and if you use curses to have full screen text output and input (curses uses infos from termcap or terminfo, say to drive a VT100 or much probably also a Wyse like the ones you have or plenty of other types...; if curses can handle the type of terminal you have connected, _your_ software won't need worry about what terminal the session is actually on. ) I suppose your energies are exhausted by now... but I still hope that someone will give me a clue. Nicola Bernardelli [EMAIL PROTECTED] --- Please use [EMAIL PROTECTED] for messages from any kind of robot, such as mailing lists. From that address no autoresponse messages will return even when I'm not at home. --- On Wed, 10 Sep 1997, Nicola Bernardelli wrote: 1) What serial devices allow longer cables than RS232 without use of modems (say 10-100-200 meters)? 2) Are there multi port cards of that kind which run well with Debian? Any brandname + model? 3) What if the text dumb terminals connected to a Linux Box also need a printer each (or almost any of them)? Is it possible that data is sent to the printer and terminal via the same line? Brandname + model? Thanks to anyone willing to give a clue. On Wed, 10 Sep 1997, Joey Hess wrote: Nicola Bernardelli wrote: 3) What if the text dumb terminals connected to a Linux Box also need a printer each (or almost any of them)? Is it possible that data is sent to the printer and terminal via the same line? Brandname + model? I have 2 old wyse 75 terminals, which have a Aux port. I think everything received by the terminal also goes out this port, though I've never tried it. On Wed, 10 Sep 1997, Nicola Bernardelli wrote: Thank you for replying... sure I'm putting one question that involves just _the_basics_ of multi-user environments... I stopped making software for Windog an year ago, I really had enough of Microsoft and Borland, I bring lot of intensive C/C++ days and nights with me, I'm learning quite a lot on Linux which I had been using for about three years before deciding that IT IS greener grass, great tools, great minds searching QUALITY, greener grass indeed... I still lack the basics but now I need them, as I plan to build an application based on PostgreSQL running on a Debain GNU/Linux box with text dumb terminals. So I add one more question: from a software point of view, how would such a terminal+printer couple (if possible) get managed? Just something to say to the OS? On Wed, 10 Sep 1997, Joey Hess wrote: Nicola Bernardelli wrote: So I add one more question: from a software point of view, how would such a terminal+printer couple (if possible) get managed? Just something to say to the OS? The way you set it up is you edit /etc/inittab and add a line similar to: S:23:respawn:/sbin/getty 19200 ttyS3 ZThis has init run a login program on the serial port. From that point on, it's as if you were
Re: Security issues for nfs mount
The traditional unix nfs filesystem is _insecure_ : the i-node generation number, which is part of the file handles, is easy to guess. -- Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: comment about Linux
Hi Is true that I never lose control on my linux box, but some times I lose control in my X-Windows and I must restart the X. Is that usual? I supose that's due to use some aplications. But if this is true, why I lose control on all aplications that I'm using? How do you lose control? I've had X crash on me quite a few times -- but not regularily. It's a pretty big system that is still changing, and there are lots of places for bugs to hide. It's still pretty stable compared to Windows 95 though. Cheers, - Jim pgpF5BWSIjDOK.pgp Description: PGP signature
Re: DHCP Server
Win95 provides no authentication by itself. It needs to belong to an NT Domain or NDS for this. So, without third party support, you're out of luck. Not so. Win95 provides share level authentication of any local filesystems you share. That qualifies as authentication. But only for people accessing the machine via the network. It sounds like the original poster was looking for a utility to monitor usage based on who is using the physical machine itself. The console authentification provided by Win95 is just a routine to unlock a password file for network access - nothing more. You can even bypass it by hitting the 'Cancel' button when it asks for a password. So it's not an effective way of figuring out who is using the machine (or restricting access). There might be a way to configure Win95 (via the registry) to require a valid password when logging in at the console. But I'm not sure how. I would never use a Windows 95 machine as a public access terminal. I've seen somebody build a super-secure public access Win 3.1 machine however (involving lots of bizarre contortions). Windows NT has a real login prompt, so it might be possible to do something more along those lines. Of course, with Linux, it's easy. Cheers, - Jim pgpnzPSwctLbw.pgp Description: PGP signature
Re: Debian ppp server how-to?
dada wrote: Is there any how-to to setup an ppp server under Debian? I need detailled information becouse I'm newie :-) and I want set up one ppp server at my home. o get/install mgetty package o get/install ppp package o make sure kernel has ppp support, if it doesn't then get kernel source and recompile with ppp support o edit /etc/mgetty/login.config find the line starting with '/AutoPPP/' and modify it so that it looks something like this: /AutoPPP/ - - /usr/sbin/pppd auth +pap login modem :Y.Y.Y.Y where Y.Y.Y.Y is what you want the caller's IP address to be. o add entry in /etc/inittab for like this: S0:23:respawn:/sbin/mgetty -s 38400 /dev/ttyS0 (if the modem is on your first serial port) o as root run 'kill -HUP 1' o enjoy! This should really be added as documentation to the mgetty/ppp packages. Cheers, - Jim pgpRYrVeQlIl7.pgp Description: PGP signature
Re: dpkg --forget-old-unavail do nothing
Why? Running dpkg --help, it says: dpkg --forget-old-unavailforget uninstalled unavailable pkgs Did you have any uninstalled, unavailable packages? If not, then that's why it didn't do anything. If you did, then it's a bug. Cheers, - Jim pgpdzLhEFBqfp.pgp Description: PGP signature
Re: Security issues for nfs mount
Ioannis Tambouras wrote: The traditional unix nfs filesystem is _insecure_ : the i-node generation number, which is part of the file handles, is easy to guess. I'm curious. How would an attack on nfs using this method proceed? Cheers, - Jim pgpYNozTVDntA.pgp Description: PGP signature
Re: Video Cards
We use S3virge cards. They work well at 16bpp 1024*768. They are also cheap. david.. Binary Bar - Australia's first free access internet bar/cafe/gallery. 243 Brunswick Street, Fitzroy, Melbourne, Australia. 12:00pm - 11:00pm http://www.binary.net.au On Wed, 10 Sep 1997, Chad D. Zimmerman wrote: Ok, getting tired of 640x480 and 8bpp color gona do some looking this weekend for a new video card. Anyone have recomendations that i can check into? X Supported of course and able to go 800x600 and 16bpp ... -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Sendmail and domain name
I have had problems from time to time in places like at home where I simply run a script to get my ppp connection going. Like you my sendmail was getting unhappy. One simple solution can be to add your machine to the /etc/hosts file and them sendmail will be able to use the information there. My boot process was ungoddly slow until I did this. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
minicom, modem questions
Hi there, I am trying to get my modem to communicate with the one at work. Using dial-up networking from window$ works, but it is pretty ennoying to reboot to the window$ OS every time I need my modem. I read the Serial and PPP-HOWTO's as well as the PPP-over-minicom mini-HOWTO. I made sure my modem is configured as described in section 9.3 of the PPP-HOWTO (these were default anyway). The point where I get stuck each time is section 9.5, i.e. testing the modem for dial out. What happens is the following: I start minicom, the modem initializes, I dial my phone-number at work, and after some seconds, ... connection is established. As pointed out it is then question of finding out what the ppp-host requires as login procedure. The trouble is that I never receive any prompt, nor any other funny characters as suggested in these HOWTO's. Basically nothing happens. (Remark: I am almost sure that my corporate site uses PAP.) I've already spend too much time on this, hope someone can put me one the right track again, otherwise I will have to abandon the idea. Good help highly appreciated. --Erwin PS. Since a few days (since diving into the above) I keep getting 'lp1 at 0x0378 (polling)' and 'lp1 off-line' in my startup messages (dmesg). Any possible connection or just something else I managed to screw up ? -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
X boot not automatic
Hello, Installed X from debian 1.31 CD. Upon boot the screen comes up with the console login prompt. After I log in the console and do a /etc/init.d/xdm start the X login window comes up and X works fine after logging in. What must be done to get X to come up automatically instead of stopping at the console login prompt? Regards, -- Victor Torrico -- Wildflower Hill, Head Waters, Virginia -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: X boot not automatic
Hello: To automatically start X at boot you need to have a symbolic link from your /etc/rc2.d to your /etc/init.d/xdm. You have to make sure you have a /etc/init.d/xdm. Mine is like this (use the ln command) S99xdm - ../init.d/xdm Also you can use the xf86config facility and answer yes when asked if you want to automatically start X at boot time. regards, From: Victor Torrico [EMAIL PROTECTED] To: Debian User Community debian-user@lists.debian.org Date: Friday, September 12, 1997 5:23 AM Subject: X boot not automatic Hello, Installed X from debian 1.31 CD. Upon boot the screen comes up with the console login prompt. After I log in the console and do a /etc/init.d/xdm start the X login window comes up and X works fine after logging in. What must be done to get X to come up automatically instead of stopping at the console login prompt? Regards, -- Victor Torrico -- Wildflower Hill, Head Waters, Virginia -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] . -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: X boot not automatic SOLVED--
-- Wildflower Hill, Head Waters, Virginia -- On 97/09/12 at 05:22 AM -0400, Victor Torrico wrote: Hello, Installed X from debian 1.31 CD. Upon boot the screen comes up with the console login prompt. After I log in the console and do a /etc/init.d/xdm start the X login window comes up and X works fine after logging in. What must be done to get X to come up automatically instead of stopping at the console login prompt? The /etc/rc2.d/S99xdm symlink was missing. Made symlink in the /etc/rc2.d directory of ln -s /etc/init.d/xdm S99xdm and all was well. Regards, -- Victor Torrico -- Wildflower Hill, Head Waters, Virginia -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: X boot not automatic SOLVED--
Hi, Maybe I am wrong but doesn't the automatic start of X or not depend on /etc/X11/config ? So if you have no-xdm-start-server in /etc/X11/config, even runing the xdm daemon would not start X automatically ? George On Fri, 12 Sep 1997, Victor Torrico wrote: -- Wildflower Hill, Head Waters, Virginia -- On 97/09/12 at 05:22 AM -0400, Victor Torrico wrote: Hello, Installed X from debian 1.31 CD. Upon boot the screen comes up with the console login prompt. After I log in the console and do a /etc/init.d/xdm start the X login window comes up and X works fine after logging in. What must be done to get X to come up automatically instead of stopping at the console login prompt? The /etc/rc2.d/S99xdm symlink was missing. Made symlink in the /etc/rc2.d directory of ln -s /etc/init.d/xdm S99xdm and all was well. Regards, -- Victor Torrico -- Wildflower Hill, Head Waters, Virginia -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] . --- George Kapetanios Churchill College Cambridge, CB3 0DSE-Mail: [EMAIL PROTECTED] U.K. WWW: http://garfield.chu.cam.ac.uk/~gk205/work_info.html --- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
libtermcap
Hello, On Debian there is no libtermcap.so, on Redhat-4.2 there is one. As far as I remember libtermcap is obsolete and should be replaced by libncurses. Am I wrong? Has any body more Information? The problem occurred when installing some foreign binaries build on a Redhat based system. Thanks --- Peter -- -- Peter Weiss, Sonnenstraße 17, D-26123 Oldenburg, Tel: 0441/ 81058 http://www.informatik.uni-oldenburg.de:/~weissp -- -- Slow has got 4 letters so has calm; speed has got 5 letters so has death -- -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: libtermcap
On Sep 12, Peter Weiss wrote On Debian there is no libtermcap.so, on Redhat-4.2 there is one. As far as I remember libtermcap is obsolete and should be replaced by libncurses. Am I wrong? No. Has any body more Information? Yes. The problem occurred when installing some foreign binaries build on a Redhat based system. Install the termcap-compat package to run them. Description: Compatibility package for termcap-based programs. The termcap-compat package provides the libtermcap.so.2 and /etc/termcap files which are required to run non-Debian, binary-only termcap-based programs. . You do not need to install this package to run Debian-packaged programs since Debian GNU/Linux uses terminfo and not termcap. You need this package if a program fails to run with the following error message ...: can't load library 'libtermcap.so.2' or complains about a missing /etc/termcap file. . The termcap-compat package isn't meant to be used to compile programs therefore it doesn't provide all the necesary files for compilation. If you want to compile a program that claims to need termcap, why not try ncurses's termcap emulation instead? It's as simple as linking with ncurses instead of libtermcap (i.e. replace the '-ltermcap' with '-lncurses' in the makefile). Ncurses' termcap emulation routines translate terminfo entries to termcap entries on the fly, so you don't even need an /etc/termcap file. . This package provides: libtermcap.so shared library, version 2.0.8 termcap database, version 9.13.15 HTH, Ray -- LEADERSHIP A form of self-preservation exhibited by people with auto- destructive imaginations in order to ensure that when it comes to the crunch it'll be someone else's bones which go crack and not their own. - The Hipcrime Vocab by Chad C. Mulligan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
RE: X boot not automatic
Hello, Installed X from debian 1.31 CD. Upon boot the screen comes up with the console login prompt. After I log in the console and do a /etc/init.d/xdm start the X login window comes up and X works fine after logging in. What must be done to get X to come up automatically instead of stopping at the console login prompt? Regards, -- Victor Torrico Hi Victor, Edit your /etc/X11/config file so it looks like this: obey-nologin allow-user-resources allow-user-modmap allow-user-xsession start-xfs #the x font server start-xdm #xdm - no-start-xdm means don't start xdm... xdm-start-server #xdm server Martin -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: X boot not automatic SOLVED--
Hello: xdm depends on /etc/X11/config. The config tell indicates how the Xserver is to run. Peter From: G. Kapetanios [EMAIL PROTECTED] To: debian-user@lists.debian.org Date: Friday, September 12, 1997 6:26 AM Subject: Re: X boot not automatic SOLVED-- Hi, Maybe I am wrong but doesn't the automatic start of X or not depend on /etc/X11/config ? So if you have no-xdm-start-server in /etc/X11/config, even runing the xdm daemon would not start X automatically ? George On Fri, 12 Sep 1997, Victor Torrico wrote: -- Wildflower Hill, Head Waters, Virginia -- On 97/09/12 at 05:22 AM -0400, Victor Torrico wrote: Hello, Installed X from debian 1.31 CD. Upon boot the screen comes up with the console login prompt. After I log in the console and do a /etc/init.d/xdm start the X login window comes up and X works fine after logging in. What must be done to get X to come up automatically instead of stopping at the console login prompt? The /etc/rc2.d/S99xdm symlink was missing. Made symlink in the /etc/rc2.d directory of ln -s /etc/init.d/xdm S99xdm and all was well. Regards, -- Victor Torrico -- Wildflower Hill, Head Waters, Virginia -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] . --- George Kapetanios Churchill College Cambridge, CB3 0DSE-Mail: [EMAIL PROTECTED] U.K. WWW: http://garfield.chu.cam.ac.uk/~gk205/work_info.h tml --- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] . -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Debian + PC with multi RS... port - n x (text dumb terminal + printer)
On Fri, 12 Sep 1997, Joey Hess wrote: Nicola Bernardelli wrote: 1) what about sending output TO A PRINTER which is NOT ON A PARALLEL PORT (lpt1, lpt2, ...) but is instead on a serial line? (Maybe I just need to write software which sends output to its stdout instead of stdprn, and that output is merely redirected to a com port... maybe such multiuser environments have typical and by now traditional solutions to my question... that's why I said I lack the basics.) I think you need 2 lines. Assumming you want your software to be able to direct some output to the printer and some output to the serial port, as it wishes without human intervention to flip a switch, you need a separate line for each. Also, while it's possible to have a device that prints out what comes to it on a serial line, a PC's printer port works quite differenlty than it's serial port, (you can't just plug a printer up to it), and so it will be more economical to use a standard printer. But I never heard of PC's with 8 or 16 LPTs, while I hear of multi serial IO cards with that number of ports. If yes, would the OS menage the distinction between the two devices or should _my_application_ (or some wrapper) be aware of the hardware each session runs on and eventually take care of sending special characters to say this goes to the display, this goes to the printer I've never heard of anything that did this. If you actually manage to find a serial terminal + printer combo that is switchable from terminal to printer mode via some escape sequence, then yes, linux could send the sigals. But I think that's unlikly. or should I just have two distinct cables run from the Linux box to the place where the dumb terminal and the printer are? probably. 1) What serial devices allow longer cables than RS232 without use of modems (say 10-100-200 meters)? 2) Are there multi port cards of that kind which run well with Debian? Any brandname + model? 3) What if the text dumb terminals connected to a Linux Box also need a printer each (or almost any of them)? Is it possible that data is sent to the printer and terminal via the same line? Brandname + model? I'll bet you're setting up a point of sale system. Bingo. I don't feel like doing it so much, but I should _hope_ to instead. If I were you, I would set up an ethernet network, (I wouldn't like so much to have that possibly pretty high number of ethernet points, I wouldn't like to be called once in a while and have to go looking for bad connectors and so on... maybe some CPU-intensive task deserves another complete powerful Debian box, maybe even more in future, but using ethernet just to have terminal+printer...) with a linux server, and POS systems that were 286 or 8088 machines with printers attached. Then you would set up software for the POS systems, to let them function as terminals, and/or output what data they receive to their printers. This fixes your cable length problem, you only run one cable, and the price is probably not much larger (unless you get them for free, dumb terminals cost more than you would expect). What is a POS system? What software runs on it? What does such a PC need to boot at startup? What ethernet cards are suitable for a 8088 or 80286? (And - ignoring noise from the fan and need to boot some software - will the people for which I will be building the application find good-looking one-year-warranty such outdated machines? And, last, I will most probably be stuck at 80x25... ok, I shall anyway _not_ be making assumptions on that point...). Nicola Bernardelli [EMAIL PROTECTED] --- Please use [EMAIL PROTECTED] for messages from any kind of robot, such as mailing lists. From that address no autoresponse messages will return even when I'm not at home. --- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
RE: minicom, modem questions
-Hi there, I am trying to get my modem to communicate with -the one at work. Using dial-up networking from window$ -works, but it is pretty ennoying to reboot to -the window$ OS every time I need my modem. -I read the Serial and PPP-HOWTO's as well as the PPP-over-minicom -mini-HOWTO. -I made sure my modem is configured as described in section 9.3 -of the PPP-HOWTO (these were default anyway). -The point where I get stuck each time is section 9.5, i.e. -testing the modem for dial out. -What happens is the following: -I start minicom, the modem initializes, I dial -my phone-number at work, and after some seconds, ... connection -is established. -As pointed out it is then question of finding -out what the ppp-host requires as login procedure. -The trouble is that I never receive any prompt, nor any other funny -characters as suggested in these HOWTO's. Basically nothing happens. -(Remark: I am almost sure that my corporate site uses PAP.) -I've already spend too much time on this, hope someone can put -me one the right track again, otherwise I will have to abandon -the idea. I had the same problem my first time with pppd, use +ua in pppd commad line -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: dpkg --forget-old-unavail do nothing
On Thu, 11 Sep 1997, Jim Pick wrote: Why? Running dpkg --help, it says: dpkg --forget-old-unavailforget uninstalled unavailable pkgs Did you have any uninstalled, unavailable packages? If not, then ^^^ ^^^ Yes. ? that's why it didn't do anything. If you did, then it's a bug. How can I set a package to be unavailable? Andrea Arcangeli FYI, Andrea in Italy is a male name. (I am tired of being mistaken for a female :) -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
dpkg uninstalled package...
I installed some custom packages in /usr/local. Now, when i install other packages which depends on that, dselect says that x depends on y; hower y is not installed (but y is in /usr/local, manually instaled. How can i tell to dpkg that i have the package `y' ? vang -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
libtermcap, tputs, and clisp
I'm trying to recompile the clisp package, which installs and compiles it's own gnu readline library. The display routines are having problems locating functions such as tputs, which the comments in the source code indicate should come from the locate termcap library. I'm using the most current version of the ncurses/termcap libs in unstable .. is there a problem with them? Secondly, do we have a standard GNU readline library someplace that I could just link clisp to, rather than having it compile it's own lib? Will --- [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] http://www.cis.udel.edu/~lowe/ For PGP Public Key, visit my website. --- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Best Ethernet Card
Hi! Can anyone tell me which 100MB Ethernet Card is best for Debian (driver quality, stability, support and also performance)? I'm planning to switch to 100MB Ethernet Network and want to be shure that my new network card will work without any problems. Thanks in advance! __ Leszek Gerwatowski [EMAIL PROTECTED] __ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Best Ethernet Card
On Fri, 12 Sep 1997, Leszek Gerwatowski wrote: Can anyone tell me which 100MB Ethernet Card is best for Debian (driver quality, stability, support and also performance)? I'm planning to switch to 100MB Ethernet Network and want to be shure that my new network card will work without any problems. I've been using 3COM 3C905-TX without any problems. No idea how it compares with others. -- Jean Pierre -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: libtermcap, tputs, and clisp
Will Lowe [EMAIL PROTECTED] writes: I'm trying to recompile the clisp package, which installs and compiles it's own gnu readline library. The display routines are having problems locating functions such as tputs, which the comments in the source code indicate should come from the locate termcap library. I'm using the most current version of the ncurses/termcap libs in unstable .. is there a problem with them? Don't use termcap, it's evil (and against policy). If you have ncurses3.4 and ncurses3.4-dev from hamm installed you should have no problem linking with something that use tputs. Secondly, do we have a standard GNU readline library someplace that I could just link clisp to, rather than having it compile it's own lib? Yes, libreadlineg2 and libreadlineg2-dev. -- James -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re[2]: Debian + PC with multi RS... port - n x (text dumb t
Printing to the aux port of a terminal has been done before. SCO Xenix used to do it (about 10 years ago) with Wyse terminals. I have heard about printing through vt type terminals off dec vax machines. I once wrote a program to do it through hp terminals (with rte-a, not unix). It probably doesn't work the way you want. Usually the terminal keyboard is locked until the print is finished. Making the terminal useful for input at the same time is generally not possible. It would require a very clever terminal and an extremely clever driver. jim __ Reply Separator _ Subject: Re: Debian + PC with multi RS... port - n x (text dumb ter Author: [EMAIL PROTECTED] at ~AMSCCSSW Date:9/12/97 8:35 AM On Fri, 12 Sep 1997, Joey Hess wrote: Nicola Bernardelli wrote: 1) what about sending output TO A PRINTER which is NOT ON A PARALLEL PORT (lpt1, lpt2, ...) but is instead on a serial line? (Maybe I just need to write software which sends output to its stdout instead of stdprn, and that output is merely redirected to a com port... maybe such multiuser environments have typical and by now traditional solutions to my question... that's why I said I lack the basics.) I think you need 2 lines. Assumming you want your software to be able to direct some output to the printer and some output to the serial port, as it wishes without human intervention to flip a switch, you need a separate line for each. Also, while it's possible to have a device that prints out what comes to it on a serial line, a PC's printer port works quite differenlty than it's serial port, (you can't just plug a printer up to it), and so it will be more economical to use a standard printer. But I never heard of PC's with 8 or 16 LPTs, while I hear of multi serial IO cards with that number of ports. If yes, would the OS menage the distinction between the two devices or should _my_application_ (or some wrapper) be aware of the hardware each session runs on and eventually take care of sending special characters to say this goes to the display, this goes to the printer I've never heard of anything that did this. If you actually manage to find a serial terminal + printer combo that is switchable from terminal to printer mode via some escape sequence, then yes, linux could send the sigals. But I think that's unlikly. or should I just have two distinct cables run from the Linux box to the place where the dumb terminal and the printer are? probably. 1) What serial devices allow longer cables than RS232 without use of modems (say 10-100-200 meters)? 2) Are there multi port cards of that kind which run well with Debian? Any brandname + model? 3) What if the text dumb terminals connected to a Linux Box also need a printer each (or almost any of them)? Is it possible that data is sent to the printer and terminal via the same line? Brandname + model? I'll bet you're setting up a point of sale system. Bingo. I don't feel like doing it so much, but I should _hope_ to instead. If I were you, I would set up an ethernet network, (I wouldn't like so much to have that possibly pretty high number of ethernet points, I wouldn't like to be called once in a while and have to go looking for bad connectors and so on... maybe some CPU-intensive task deserves another complete powerful Debian box, maybe even more in future, but using ethernet just to have terminal+printer...) with a linux server, and POS systems that were 286 or 8088 machines with printers attached. Then you would set up software for the POS systems, to let them function as terminals, and/or output what data they receive to their printers. This fixes your cable length problem, you only run one cable, and the price is probably not much larger (unless you get them for free, dumb terminals cost more than you would expect). What is a POS system? What software runs on it? What does such a PC need to boot at startup? What ethernet cards are suitable for a 8088 or 80286? (And - ignoring noise from the fan and need to boot some software - will the people for which I will be building the application find good-looking one-year-warranty such outdated machines? And, last, I will most probably be stuck at 80x25... ok, I shall anyway _not_ be making assumptions on that point...). Nicola Bernardelli [EMAIL PROTECTED] --- Please use [EMAIL PROTECTED] for messages from any kind of robot, such as mailing lists. From that address no autoresponse messages will return even when I'm not at home. --- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] .
Re[2]: Debian + PC with multi RS... port - n x (text dumb t
Printing to the aux port of a terminal has been done before. SCO Xenix used to do it (about 10 years ago) with Wyse terminals. I have heard about printing through vt type terminals off dec vax machines. I once wrote a program to do it through hp terminals (with rte-a, not unix). It probably doesn't work the way you want. Usually the terminal keyboard is locked until the print is finished. Making the terminal useful for input at the same time is generally not possible. It would require a very clever terminal and an extremely clever driver. jim __ Reply Separator _ Subject: Re: Debian + PC with multi RS... port - n x (text dumb ter Author: [EMAIL PROTECTED] at ~AMSCCSSW Date:9/12/97 8:35 AM On Fri, 12 Sep 1997, Joey Hess wrote: Nicola Bernardelli wrote: 1) what about sending output TO A PRINTER which is NOT ON A PARALLEL PORT (lpt1, lpt2, ...) but is instead on a serial line? (Maybe I just need to write software which sends output to its stdout instead of stdprn, and that output is merely redirected to a com port... maybe such multiuser environments have typical and by now traditional solutions to my question... that's why I said I lack the basics.) I think you need 2 lines. Assumming you want your software to be able to direct some output to the printer and some output to the serial port, as it wishes without human intervention to flip a switch, you need a separate line for each. Also, while it's possible to have a device that prints out what comes to it on a serial line, a PC's printer port works quite differenlty than it's serial port, (you can't just plug a printer up to it), and so it will be more economical to use a standard printer. But I never heard of PC's with 8 or 16 LPTs, while I hear of multi serial IO cards with that number of ports. If yes, would the OS menage the distinction between the two devices or should _my_application_ (or some wrapper) be aware of the hardware each session runs on and eventually take care of sending special characters to say this goes to the display, this goes to the printer I've never heard of anything that did this. If you actually manage to find a serial terminal + printer combo that is switchable from terminal to printer mode via some escape sequence, then yes, linux could send the sigals. But I think that's unlikly. or should I just have two distinct cables run from the Linux box to the place where the dumb terminal and the printer are? probably. 1) What serial devices allow longer cables than RS232 without use of modems (say 10-100-200 meters)? 2) Are there multi port cards of that kind which run well with Debian? Any brandname + model? 3) What if the text dumb terminals connected to a Linux Box also need a printer each (or almost any of them)? Is it possible that data is sent to the printer and terminal via the same line? Brandname + model? I'll bet you're setting up a point of sale system. Bingo. I don't feel like doing it so much, but I should _hope_ to instead. If I were you, I would set up an ethernet network, (I wouldn't like so much to have that possibly pretty high number of ethernet points, I wouldn't like to be called once in a while and have to go looking for bad connectors and so on... maybe some CPU-intensive task deserves another complete powerful Debian box, maybe even more in future, but using ethernet just to have terminal+printer...) with a linux server, and POS systems that were 286 or 8088 machines with printers attached. Then you would set up software for the POS systems, to let them function as terminals, and/or output what data they receive to their printers. This fixes your cable length problem, you only run one cable, and the price is probably not much larger (unless you get them for free, dumb terminals cost more than you would expect). What is a POS system? What software runs on it? What does such a PC need to boot at startup? What ethernet cards are suitable for a 8088 or 80286? (And - ignoring noise from the fan and need to boot some software - will the people for which I will be building the application find good-looking one-year-warranty such outdated machines? And, last, I will most probably be stuck at 80x25... ok, I shall anyway _not_ be making assumptions on that point...). Nicola Bernardelli [EMAIL PROTECTED] --- Please use [EMAIL PROTECTED] for messages from any kind of robot, such as mailing lists. From that address no autoresponse messages will return even when I'm not at home. --- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] .
Re: Best Ethernet Card
On Sep 12, Jean Pierre LeJacq wrote On Fri, 12 Sep 1997, Leszek Gerwatowski wrote: Can anyone tell me which 100MB Ethernet Card is best for Debian (driver quality, stability, support and also performance)? I'm planning to switch to 100MB Ethernet Network and want to be shure that my new network card will work without any problems. I've been using 3COM 3C905-TX without any problems. No idea how it compares with others. However, there are known problems with the 3c905 as well. Under some situations/configurations, the 905 has some fairly severe problems. I have one, and unfortunately suffer from those problems (dropped packets, the driver/card freezes, and only resumes after dropping out of busmastering mode, etc.). I will say this, though, the turnaround time for driver releases is very fast. Donald Becker does a great job with driver releases. The problems I've been having are known, and are also supposedly rare. It seems like my problems are more the exception than the rule. I'm not trying to steer anyone away from the card, far from it, but the driver is not 100% perfect yet, either. What is, though. My $0.15 worth. (inflation's an evil thing) Paul Kuykendall -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: X boot not automatic SOLVED--
On Fri, 12 Sep 1997, Victor Torrico wrote: -- Wildflower Hill, Head Waters, Virginia -- On 97/09/12 at 05:22 AM -0400, Victor Torrico wrote: Hello, Installed X from debian 1.31 CD. Upon boot the screen comes up with the console login prompt. After I log in the console and do a /etc/init.d/xdm start the X login window comes up and X works fine after logging in. What must be done to get X to come up automatically instead of stopping at the console login prompt? The /etc/rc2.d/S99xdm symlink was missing. Made symlink in the /etc/rc2.d directory of ln -s /etc/init.d/xdm S99xdm and all was well. The preferred method of creating these links is to use update-rc.d (included in the dpkg package). Bob Bob Nielsen Internet: [EMAIL PROTECTED] Tucson, AZ AMPRnet: [EMAIL PROTECTED] http://www.primenet.com/~nielsen -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: DHCP Server
Win95 provides no authentication by itself. It needs to belong to an NT Domain or NDS for this. So, without third party support, you're out of luck. Not so. Win95 provides share level authentication of any local filesystems you share. That qualifies as authentication. sigh grin I *hate* to waste space on Winbloze here grin However. Without a network server to authenticate you (i.e. user level authentication), you can just bypass the login screen by hitting Esc. Once Windows is loaded up, you can still access the whole network and even shares on other Win95 computers - no authentication there at all. (Of course, there *is* a level of security - *if* you password protect the shares.) Windows sucks! Long live Linux! Unfortunately, my employer - and most of the rest of the commercial world - doesn't feel the same way grin I still spend much more time administrating Win95/WinNT every day than I do Linux :-( Anyway, let's get back to discussing a real OS, eh? later, Kevin Traas Baan Business Systems Systems Analyst Langley, BC, Canada [EMAIL PROTECTED] (604) 882-8169 http://www.baan-bbs.ca --- Linux is not user-friendly. It _is_ user-friendly. It's just not ignorant-friendly or idiot-friendly. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Debian + PC with multi RS... port - n x (text dumb terminal + printer)
Nicola Bernardelli wrote: But I never heard of PC's with 8 or 16 LPTs, while I hear of multi serial IO cards with that number of ports. That's right, neither have I. I'll bet you're setting up a point of sale system. Bingo. I don't feel like doing it so much, but I should _hope_ to instead. with a linux server, and POS systems that were 286 or 8088 machines with printers attached. Then you would set up software for the POS systems, to let them function as terminals, and/or output what data they receive to their printers. This fixes your cable length problem, you only run one cable, and the price is probably not much larger (unless you get them for free, dumb terminals cost more than you would expect). What is a POS system? POS = point of sale. Hm, there are probably commercial solutions here as well. Of course then you don't get to run debian on it. :-/ What software runs on it? What does such a PC need to boot at startup? If you were willing to go with 386 class PC's, you could strip them down to a special ethernet card with the kernel in ROM (or, a normal ethernet card and a floppy disk drive), a video card, a printer port, and 2 to 4 MB of memory (plus keyboard, screen, and printer, of course). That can boot linux over the network, and then you could write the POS app in linux. What ethernet cards are suitable for a 8088 or 80286? Will the people for which I will be building the application find good-looking one-year-warranty such outdated machines? Well, it's possible thay you will be able to find all the parts you need new. I'm not sure if unused 286 or 386 chips are still being sold - everything else can be bought new, though. -- see shy jo -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
How to NFS install.
I am going to set up a new machine at home this weekend to replace one that is currently running Caldera Standard 1.1. While reading the list mail I had seen mention that the system could be installed via NFS. After reading the install doc on my mirror here at the shop I couldn't find mention of it. The machines that are running right now were installed as Debian 1.2 and upgraded so I haven't seen the 1.3 install disks. All of the installs of Debian that I have done going back to .93R5 were done on floppy. Is there a separate document that has other info on NFS install? It probably isn't all that hard to figure out but a point in the right direction would be nice. Thank you. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Debian + PC with multi RS... port - n x (text dumb terminal + printer)
If you were willing to go with 386 class PC's, you could strip them down to a special ethernet card with the kernel in ROM (or, a normal ethernet card and a floppy disk drive), a video card, a printer port, and 2 to 4 MB of memory (plus keyboard, screen, and printer, of course). That can boot linux over the network, and then you could write the POS app in linux. IGEL, LLC makes a line of systems called the Etherminal. These are Intel (or compatible) based systems that have Linux and X Windows on EEPROM. They also have a root filesystem on NVRAM (CMOS). These are nice systems. The current models are the 5x and the 5j. The 5x has built-in Ethernet (3 media types) 2 serial (one for the mouse) and a parallel port. They also have builtin Cirrus SVGA. The 5j has the same plus additional RAM, and Netscape navigator. A 5js is going to be available RSN - it's the 5j w/ sound. All of the Etherminal 5's use a 586 processor. If those systems are too much, the Etherminal 3x's may still be available. These systems have the same svga/2s/1p/Ethernet ports on-board, but use a 386sx-40 processor and have 4-8 MB RAM. Those systems may sound puny but they are amazingly fast!! I've sold a few of them to my wife's veterinary clinic, and am amazed with their performance and their reliability. We set the 3x's up in Feb. 1996 and have not had to do *anything* to them since. Oh, yeah. It took longer to get all of the components out of the box and connected than it did to configure the machines. For the record - yes, I am associated with IGEL. I'm an IGEL VAR. My opinion of their systems is not based on that tho - I'm *truly* impressed with the ease of configuration/setup and the reliability of the Etherminals. Chuck -- Chuck Stickelman, Owner E-Mail: [EMAIL PROTECTED] Practical Network DesignVoice: (419) 529-3841 9 Chambers Road FAX:(419) 529-3625 Mansfield, OH 44906-1302 USA -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
printing in netscape
hello, all, when i try to print from netscape, i get this message in my xconsole: Sep 12 09:53:40 mattyt lpd[1143]: lp: Daemon filter 'f' terminated (13) ??? Matt Thompson MZI, Inc. v-206.430.3726 707 S. Grady Wayf-206.430.3420 Renton, WA 98055 [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Best Ethernet Card
Leszek Gerwatowski [EMAIL PROTECTED] writes: Hi! Can anyone tell me which 100MB Ethernet Card is best for Debian (driver quality, stability, support and also performance)? I'm planning to switch to 100MB Ethernet Network and want to be shure that my new network card will work without any problems. Thanks in advance! The current wisdom says to avoid the 3com 3c905, and go with a DEC Tulip based card. I do know from personal experience that the 3c905 has problems - haven't tried a Tulip card yet, but will be using one very soon. The other card I see mentioned a lot, but never see much as far as recommendations, is the Intel EtherExpress 100. Anyone have any thoughts on those? Later, Dale -- + finger for pgp public key -+ | Dale E. Martin | University of Cincinnati Savant Research Laboratory | | [EMAIL PROTECTED]| http://www.ececs.uc.edu/~dmartin | +--+ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Best Ethernet Card
On Fri, Sep 12, 1997 at 01:09:24PM -0400, Dale Martin wrote: The current wisdom says to avoid the 3com 3c905, and go with a DEC Tulip based card. I do know from personal experience that the 3c905 has problems - haven't tried a Tulip card yet, but will be using one very soon. The Tulip cards seem to suffer from minimal braindamage. They are cheap and fast. Some people have driver problems which seem to be related to the way different models implement the SROM. I've 10-T cards from DEC, Kingston, and Linksys without any trouble at all. I would expect similar results from the 100Mb stuff. Jeff -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: dpkg uninstalled package...
Viorel ANGHEL [EMAIL PROTECTED] wrote: I installed some custom packages in /usr/local. Now, when i install other packages which depends on that, dselect says that x depends on y; hower y is not installed (but y is in /usr/local, manually instaled. How can i tell to dpkg that i have the package `y' ? Get my package equivs-1.0.3-1.deb-unoff.src.tar.gz from ftp://ietpd1.sowi.uni-mainz.de/pub/debian/unofficial/source/; and customize it according to your needs. It is a dummy package useful for exactly this purpose. It's about time to contribute it officially... Cheers, P. *8^) -- Paul Seelig [EMAIL PROTECTED] African Music Archive - Institute for Ethnology and Africa Studies Johannes Gutenberg-University - Forum 6 - 55099 Mainz/Germany My Homepage in the WWW at the URL http://www.uni-mainz.de/~pseelig -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: dpkg --forget-old-unavail do nothing
dpkg --forget-old-unavailforget uninstalled unavailable pkgs Did you have any uninstalled, unavailable packages? If not, then ^^^ ^^^ Yes. ? that's why it didn't do anything. If you did, then it's a bug. How can I set a package to be unavailable? You don't - it usually happens when the package was previously available on an FTP site (or on a CD-ROM, or wherever you get your Packages files from) and no longer is. I assume that this command might be useful in the following case: You had selected a package (ie. xtetris) using dselect and the ftp method when it was available from ftp.debian.org, but never installed it. The flags in the dpkg database (/var/lib/dpkg/status) would be set to install it. Perhaps you waited a few days, and the package was deleted from ftp.debian.org (ie. xtetris violated a trademark). When dselect is run again, it would clear the available list (/var/lib/dpkg/available), and re-load it using the Packages file from ftp.debian.org. Now you've got a case where you have something _selected_ to be installed, but it is still _uninstalled_, and it is also _unavailable_. That's where that command might be useful (especially to tools like dselect which internally call dpkg). [ Disclaimer: I'm just guessing about how dpkg might work. It might not actually work this way. ] Cheers, - Jim pgpnbLnry9hUZ.pgp Description: PGP signature
DSELECT/DPKG Problems with Apache 1.1.3-6
Hello, I've been having problem with Apache 1.1.3-6 as supplied with Debian 1.3.1. Basically I've been using DSELECT, but reverted to DPKG in an effort to get more info on the problem. Alas, even the highest level of --debug failed to provide me with enough. A brief history: Apache 1.1.3-6 installed successfully Apache 1.1.3-6 uninstalled successfully Apache 1.2 installed successfully (Manually into /usr/local using the source from the Apache website). Apache 1.2 removed successfully (Had some problems I couldn't resolve) Apache 1.1.3-6 now won't install, despite successful a 'purge'. The interesting thing is that the version number is still visible in DSELECT even after 'purge'. On inspection, this is reflected in /var/lib/dpkg/status. Whether that is important or not I don't know. Has anyone encountered problems similar to this, or is there another switch that will point me in the right direction? . Or am I missing something really obvious? :-) Any help would be appreciated. I really don't fancy reinstalling the whole system :) Thanks in advance Nigel Price PS. Here's the most recent log with the debug options on. Start Log PC2154# dpkg --debug=3773 --install /cdrom/stable/binary/web/apache_1.1.3-6.deb D10: ensure_pathname_nonexisting `/var/lib/dpkg/tmp.ci' (Reading database ... 22370 files and directories currently installed.) Unpacking apache (from .../binary/web/apache_1.1.3-6.deb) ... D20: process_archive conffile `/etc/cron.daily/apache' no package, no hash D20: process_archive conffile `/etc/init.d/apache' no package, no hash D01: process_archive oldversionstatus=not installed conflictor=none D02: fork/exec /var/lib/dpkg/tmp.ci/preinst ( preinst install ) dpkg: error processing /cdrom/stable/binary/web/apache_1.1.3-6.deb (--install): subprocess pre-installation script returned error exit status 1 D02: fork/exec /var/lib/dpkg/tmp.ci/postrm ( postrm abort-install ) D10: ensure_pathname_nonexisting `/var/lib/dpkg/tmp.ci' D10: ensure_pathname_nonexisting running rm -rf D10: ensure_pathname_nonexisting `/var/lib/dpkg/reassemble.deb' Errors were encountered while processing: /cdrom/stable/binary/web/apache_1.1.3-6.deb PC2154# End Log -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Where is Project.tmpl
I've just installed Debian from the Official 1.3.1 CD. I'm now running into problems when developing applications for the X Window System. It appears to me that some development files are missing, namely Project.tmpl and others. And even though I know where it should be if its in the original XFree86 Distribution archive, I'm unable to locate them in the .deb packaging system. I will really appreciate it if someone can tell me where Project.tmpl and other X Development files should be installed from. Thanks Jason -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Where is Project.tmpl
I think it's in the xdevel package. Bruce -- Can you get your operating system fixed when you need it? Linux - the supportable operating system. http://www.debian.org/support.html Bruce Perens K6BP [EMAIL PROTECTED] 510-215-3502 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: DSELECT/DPKG Problems with Apache 1.1.3-6
The pre-install script in that package is failing - it says returning status 1, and it should return 0. You can use dpkg to extract the package archive, and then try to run DEBIAN/preinst by hand. That should let you figure out what is wrong with the script. Sometimes it is something simple like the interpreter does not exist (the file mentioned after the #! on the first line) or there's no exit statement at the end of the script. Sometimes the script runs another program that fails or doesn't exist. Bruce -- Can you get your operating system fixed when you need it? Linux - the supportable operating system. http://www.debian.org/support.html Bruce Perens K6BP [EMAIL PROTECTED] 510-215-3502 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: DHCP Server
On Thu, 11 Sep 1997, Jim Pick wrote: There might be a way to configure Win95 (via the registry) to require a valid password when logging in at the console. But I'm not sure how. I would never use a Windows 95 machine as a public access terminal. I've seen somebody build a super-secure public access Win 3.1 machine however (involving lots of bizarre contortions). This is possible with a Windows 95 Policy, but the only good way to enforce it is to use an NT server domain and place the newly created policy file config.pol in the NETLOGON directory. You'll need to use user-level security on the W95 machines, and you'll need to require (in the policy) a valid user logon (and you can dictate which NT domain shall be used, too). It's quite a neat feature, although it too can be slightly subverted. I used this in a computer store earlier this year. We set up a policy which took away a lot of functionality for the demo logon, gave the demo logon a blank password, and required a validated logon to use the computer. This restricted what demo users could do, especially browse our network and play/add DOS games. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: printing in netscape
On Fri, 12 Sep 1997, Matthew C. Thompson wrote: hello, all, when i try to print from netscape, i get this message in my xconsole: Sep 12 09:53:40 mattyt lpd[1143]: lp: Daemon filter 'f' terminated (13) What do you have as print command in the window which pops up in Netscape when you try to print? Just a simple lpr works fine for me. Bob Bob Nielsen Internet: [EMAIL PROTECTED] Tucson, AZ AMPRnet: [EMAIL PROTECTED] http://www.primenet.com/~nielsen -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Project.tmpl
% grep Project.tmpl Contents-i386 usr/X11R6/lib/X11/config/Project.tmplx11/xlib6-dev I keep a copy of the Contents file locally in case I can't find a file that should be in a package somewhere. If 'dpkg -S file' can't find it then I do the above. My. $.02. Brian -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Security issues for nfs mount
I could resist to your request, Jim, and appear before you with further clarifications, for you are an active contributor in the Debian project and we are quite fortunate to have you here among us; moreover, there in an ancient saying, that hard is the knowledge of the good. And the knowledge of nfs is a great part of knowledge. If I had not been busy, I might have reviewed all the relevant rcf's, which is a complete education in nfs mechanics and then I should have been at once able to answer your question about the IO mechanics. But, indeed, I have only a limited time, and therefore, I do not know the truth about such matters; I will, however, gladly assist you in the investigation of them. We say that an nfs server running nfs-ver_1, or nfs-ver_2, can be tricked to perform unauthorized IO. But, there is a good deal of difficulty in describing this sort of knowledge in a public list, therefore we better investigate the description of it, instead on the specific steps that we are now supposing. Lets reflect on the nature of the io request. The file handle is a simple structure of 32 bytes for version 2 (although this was increased with version 3 to 64 bytes) and is created by the server, who passes it back to the client, and then the client uses the file handle to access the file. All the server does does is validate the file handle (the ticket) when the io request is made. Supply the right handle and you are in. Guessing the fields of the structure is not extremely difficult. Most of the information is already known: major and minor device numbers, i-node number, etc. Everything is known except the i-node generation number whose purpose is to provide security and is chosen when the open request is made. The generation number is implementation dependent; it may, even, be a sequential(!) number, but allow, for the sake of this discussion, this to be a truly random number. My distinct recollection from the relevant rfc [1] is that the width of this number is small, either 16 or 32 bits, we could, of course, if we must, go back to the specification and verify the truth of this, yet this weakness of (traditional) nfs is so well known and is as old as I dare remember. It will not take a giant to place many requests until one succeeds, and it does not prevent an 12-year old to run an exploit that was written very many years ago. [1] The opaque structure, the file handle definition, is a simple typedef in rfc1094 . The actual details should then be in the XDR rfc1014 , I must assume, for I do not have it here locally to examine. The traditional unix nfs filesystem is _insecure_ : the i-node generation number, which is part of the file handles, is easy to guess. I'm curious. How would an attack on nfs using this method proceed? -- Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Security issues for nfs mount
Hi, Although I am not familiar at all with the inner workings of nfs the description below indicates a risk that an unauthorised client may read files on the specific directory which is being exported by nfs read only. However my worry is not whether somebody else will read the files which in my cased is only a piece of software. My worry is whether anybody can write to the directory being exported or any other directory of my computer , given that the stuff on the exported directory are mathematical software unrelated with the workings of the operating system . I would be grateful if someone could clarify this point Thanks George On Fri, 12 Sep 1997 [EMAIL PROTECTED] wrote: I could resist to your request, Jim, and appear before you with further clarifications, for you are an active contributor in the Debian project and we are quite fortunate to have you here among us; moreover, there in an ancient saying, that hard is the knowledge of the good. And the knowledge of nfs is a great part of knowledge. If I had not been busy, I might have reviewed all the relevant rcf's, which is a complete education in nfs mechanics and then I should have been at once able to answer your question about the IO mechanics. But, indeed, I have only a limited time, and therefore, I do not know the truth about such matters; I will, however, gladly assist you in the investigation of them. We say that an nfs server running nfs-ver_1, or nfs-ver_2, can be tricked to perform unauthorized IO. But, there is a good deal of difficulty in describing this sort of knowledge in a public list, therefore we better investigate the description of it, instead on the specific steps that we are now supposing. Lets reflect on the nature of the io request. The file handle is a simple structure of 32 bytes for version 2 (although this was increased with version 3 to 64 bytes) and is created by the server, who passes it back to the client, and then the client uses the file handle to access the file. All the server does does is validate the file handle (the ticket) when the io request is made. Supply the right handle and you are in. Guessing the fields of the structure is not extremely difficult. Most of the information is already known: major and minor device numbers, i-node number, etc. Everything is known except the i-node generation number whose purpose is to provide security and is chosen when the open request is made. The generation number is implementation dependent; it may, even, be a sequential(!) number, but allow, for the sake of this discussion, this to be a truly random number. My distinct recollection from the relevant rfc [1] is that the width of this number is small, either 16 or 32 bits, we could, of course, if we must, go back to the specification and verify the truth of this, yet this weakness of (traditional) nfs is so well known and is as old as I dare remember. It will not take a giant to place many requests until one succeeds, and it does not prevent an 12-year old to run an exploit that was written very many years ago. [1] The opaque structure, the file handle definition, is a simple typedef in rfc1094 . The actual details should then be in the XDR rfc1014 , I must assume, for I do not have it here locally to examine. The traditional unix nfs filesystem is _insecure_ : the i-node generation number, which is part of the file handles, is easy to guess. I'm curious. How would an attack on nfs using this method proceed? -- Ioannis Tambouras [EMAIL PROTECTED], West Palm Beach, Florida Signed pgp-key on key server. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] . --- George Kapetanios Churchill College Cambridge, CB3 0DSE-Mail: [EMAIL PROTECTED] U.K. WWW: http://garfield.chu.cam.ac.uk/~gk205/work_info.html --- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Debian ppp server how-to?
On Fri, 12 Sep 1997, dada wrote: Is there any how-to to setup an ppp server under Debian? I need detailled information becouse I'm newie :-) and I want set up one ppp server at my home. Regards. Depending how many users you want to server try checking out Portslave at http://homepage.cistron.nl/~miquels/radius/ I found it to be an excellent program and fairly self-documenting. We're using it as a temp solution until PRI prices drop here in BC Canada and we order our PM3's. mike... Micro$oft, what do you want to spend today? -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Help with Bibtex
I am for the first time trying to use bibtex.
When I am trying to run bibtex on the following file (I deleted
nonrelevant lines)
---
Moenie dieselfde gebed oor en oor bid nie. \cite[]{Mostert:1992}
\bibliographystyle{harvard}
\bibliography{spies}
---
The following .bbl-file was produced
--
\begin{thebibliography}{1}
\input{babelbst.tex}
\newcommand{\Capitalize}[1]{\uppercase{#1}}
\newcommand{\capitalize}[1]{\expandafter\Capitalize#1}
\harvarditem{Mostert}{1992}{Mostert:1992}
Mostert, B., \harvardyearleft 1992\harvardyearright{}.
\newblock \emph{Geestelike Oorlogvoering}.
\newblock Christelike Uitgewersmaatskappy, Vereeniging.
\end{thebibliography}
-
resulting in the error message:
--
ERROR: LaTeX Error: File `babelbst.tex' not found.
--- TeX said ---
---
There is no `babelbst.tex' file on my system.
How can I solve this problem?
Johann.
Johann Spies
[EMAIL PROTECTED]
Windsorlaan 19
Pietermaritzburg
3201
Suid Afrika (South Africa)
Tel/Faks Nr. +27 331-46-1310
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] .
Trouble? e-mail to [EMAIL PROTECTED] .
