apt-get update offline???
Buenas... Hace poco he actualizado de slink a potato (durante la party que hubo en santiago), pero como todavía no tengo los CDs de potato, cada vez que quiero instalar un paquete utilizo un script que me crea una lista con los paquetes que necesita bajar y así me los saco en la facultad. El caso es que me gustaría saber si se puede hacer algo parecido con la fase de actualización de la lista de paquetes disponibles, es decir, sacar los ficheros Packages y Releases necesarios para reconstruir la lista de paquetes como si se hiciese un apt-get update, y por supuesto que el apt se crea que lo ha hecho. El objetivo es poder actualizar dicha lista en una máquina que no tiene conexion a internet, de forma que pueda generar la lista de paquetes a instalar con el script anterior, bajarlos en la facultad, copiarlos al /var/cache/apt/archives e instalarlos, todo esto como si la máquina estuviese conectada y lo hubiese hecho directamente. Mirando en la documentación del apt, viene como generar la lista de paquetes a instalar (que es de donde saqué el script), y como mucho como utilizar la configuración del apt de otra máquina en la que se va a conectar. Pero esto no me sirve, ya que la máquina que se conecta a internet no tiene instalado Debian (ni siquiera es Unix, sino que utiliza VMS). Si alguien lo ha hecho y me puede ayudar, por favor, que lo haga. Cualquier otra ayuda (direcciones, documentación, ...) también me vale. Gracias por adelantado. Hasta luego. -- David Muriel. Debian GNU/Linux woody + Emacs 20.5.2 + Gnus v5.8.3 Linux registered user #25632 (http://counter.li.org/) Linux is userfriendly, but is only a bit selective about its friends :-)
default keymap y root
Buenas... Al actualizar hace poco a potato, instalé el paquete console-tools y, según parece, no me configuró el teclado en español. Bueno, hasta ahí no pasa nada, lo configuro con kbdconfig y ya está. El problema es que sólo parece funcionar cuando entro como usuario, y no como root. Si entro como usuario, funcionan todas las teclas, la ñ, acentos, ... Si entro como root, ni la ñ ni los acentos funcionan en la consola (si entro en el emacs si funcionan). He probado cambiando la variable LANG y no pasa nada. No es que necesite utilizarlo ahora mismo, pero me fastidiaría bastante el necesitarlo y no poder usarlo. A ver si alguien me puede echar una mano pare solucionar esto. Gracias. Hasta luego. -- David Muriel. Debian GNU/Linux woody + Emacs 20.5.2 + Gnus v5.8.3 Linux registered user #25632 (http://counter.li.org/) ...El trabajo en equipo es esencial...te permite echarle la culpa a otro
Pregunta sobre libg++2.8.2-dev
Hola. He tenido que instalar el paquete libg++2.8.2-dev para poder disponer de los archivos de cabecera String.h, iostream.h y stl.h (entre otros) que ahora tengo en /usr/include/g++-2. ¿Qué tengo que hacer para que estén disponibles para cualquier programa de C++ con un #include String.h, por ejemplo? He probado a poner /usr/include/g++-2 en el PATH pero no ha funcionado. Hay programas en los que incluyo, por ejemplo, #include mico/naming.h o #include ministl/bstring.h, ficheros que tengo en: - /usr/local/include/mico/naming.h (donde /usr/local/include no está en el PATH.) - /home/emilio/pfc/mico/include/mico/naming.h (donde /home/emilio/pfc/mico/include sí que está en el PATH.) - /usr/local/include/ministl/bstring.h (/usr/local/include no está en el PATH.) - /home/emilio/pfc/mico/include/ministl/bstring.h (/home/emilio/pfc/mico/include sí está en el PATH.) Por tanto, si el ordenador es capaz de encontrarlos perfectamente y uno de los dos directorios en los que se encuentran no está en el PATH (/usr/local/include), yo supuse que los encontraba en el que sí está, pero por lo visto no es así porque, como digo, he incluido en el PATH el directorio en el que están String.h, etc. (/usr/include/g++-2/) y sigue sin encontrarlos. ¿Será entonces que encuentra por defecto todos aquellos archivos que están en /usr/local/include o qué?, ¿qué hay que hacer para que encuentre los que están en /usr/include/g++-2? Gracias y un saludo. Emilio.
Re: Problemas para bajar algunos algunos paquetes(make-pseudo-image)
On Fri, 24 Mar 2000, Miguel Choque wrote: el problema que se me presento es cuando estoy bajando los paquetes de debian por medio de ftp con la utilidad make-pseudo-image para obtener la imagen iso del cd-debian, todo esta perfecto excepto por algunos paquetes que me devolvio error, me indica que no existe, probe con casi todos los servidores de ftp y me devuelve el mismo error, ya revise la lista de Lo que ocurre, al menos con los que he comprobado de la lista de errores que mandas, es que para cada uno que no encuentra lo que hay es ya una versión más moderna. O sea, que no sé bien cómo funciona el make-pseudo-image, pero lo que necesitas son unas listas Packages actualizadas para la Slink que existe estos días (si make-pseudo-image usa apt, con apt-get update resolverás el problema)
Sin sonido por el CD (Solucionado)
Hola Hace no mucho puse un mensaje en esta lista en el que decía que no podia oir ningun tipo de sonido desde el CD, pues bien, averigue la solución, os cuento por si alguien llegara a estar en la misma situación. Tengo una Sound Blaster Live! conectada a unos altavoces Cambridge Soundworks DTTT2500 Desktop Theater 5.1, para la salida de altavoces en la SBLive! hay dos tipos, una, la que todos conocemos, de salida delantera y delantera con el cable típico, y otra con un Din Digital, según que quieras hacer la reproducción de sonido de los altavoces usa una o la otra. Pues bien los DT 5.1 tienen atrás un interruptor que si está en AUTO da preferencia a la entrada del digital DIN e ignora la convencional, si lo pones en ANALOG, usa la entrada convencional. Conclusión, según parece el driver de la SBLive! emu10k1 todavía no está preparado para emitir señal a través del DIGITAL DIN con la música que sale del CD-ROM, pero si con los sonidos de tabla de onda: WAVS, MP3, etc. Weno, después del royo, si alguno tiene el mismo tinglado que yo montado y no oye ningun cd de musica con estos drivers que ponga el interruptor de los DTT2500 a ANALOG cuando esté en Linux. Saludos Daniel
Re: Problemas para bajar algunos algunos paquetes(make-pseudo-image)
Creo que eso a priori no debe importar porque lo que importa es que rsync después hace los arreglos pertinentes sobre los datos que falten, etc... creo :P Al menos a mí me pasó lo mismo con algunos paquetes pero luego con rsync 20 minutos después ya tenía la imagen lista, y hasta ahora el cd grabado no me dio ningun fallo... Saluten Daniel On Sat, Mar 25, 2000 at 10:50:36AM +0100, Fernando Sanchez wrote: On Fri, 24 Mar 2000, Miguel Choque wrote: el problema que se me presento es cuando estoy bajando los paquetes de debian por medio de ftp con la utilidad make-pseudo-image para obtener la imagen iso del cd-debian, todo esta perfecto excepto por algunos paquetes que me devolvio error, me indica que no existe, probe con casi todos los servidores de ftp y me devuelve el mismo error, ya revise la lista de Lo que ocurre, al menos con los que he comprobado de la lista de errores que mandas, es que para cada uno que no encuentra lo que hay es ya una versión más moderna. O sea, que no sé bien cómo funciona el make-pseudo-image, pero lo que necesitas son unas listas Packages actualizadas para la Slink que existe estos días (si make-pseudo-image usa apt, con apt-get update resolverás el problema) -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: GTK (Re: Imposto de Renda no Wine)
Marcio Henrique Leiner wrote: Pessoal, em primeiro lugar o corpo do programa, quer seja em C, C++, java ou LingBR :-? deveria ter que ser projetado para que funcione em qualquer plataforma, apesar do perigo maior de ter sido feito em VB. Quanto ao problema das versoes existe um sistema de compilacao chamado CodeWarior, o qual eu ainda nao testei, que promete, jura de pe junto, garante, etc... que possui VM para Mac, Linux, Win32, BeOS, PlayStation etc... Assim sendo seria apenas o caso de, se este suite funcionar realmente, portar o codigo para algumas das linguagens suportadas, java, C e pascal se nao me engano, e testar em cada uma das VM. Olha, as 3 linguagens mais portáveis que eu conheço: C (e C++), java e perl. Claro que nao ha garantia de que o codigo possa ser portado apenas usando as VM, mas os problemas devem ser reduzidos na fase de beta teste. Este programa e comercial, se era bem que bem mais barato que os Visuais da microsoft ou borland ( nao vejo o preco dele ha tempos ). Eu escolheria uma coisa que fosse padrão no mercado. Alem disso para portabilidade, ainda existe a Qt, que, se nao me engano roda em Win32. A Qt, para uso comercial, custa mais de 2000 dólares. Não curto muito isso... Mas dependendo da aplicação... Se o programa utilizou Delphi, lembrar que ha a possibilidade de termos o Delphi para linux ( alias como ficou isto? ). A Borland tinha dito que dentro de 6 meses saia, isso foi quando eles lançaram o JBuilder pra linux. Assim sendo a portabilidade nao seria um problema tao grande, a menos que o codigo tenha sido *excrito* em VB, ou seja, que seja uma colcha de retalhos. Eu comecei a programar em VB (não conhecia unix). Vivo dizendo e o pessoal acha que é piada, mas é a mais pura verdade: VB NÃO É LINGUAGEM DE PROGRAMAÇÃO!!! É UM PRODUTO DA MICROSOFT, ASSIM COMO O DIRECTOR DA MACROMEDIA!!! Eu não aceito soluções em VB e isso é um fato. Quem sai perdendo é quem compra a aplicação, que está comprando um lixo que, supostamente, funciona. Caso alguem se interesse por dar uma olhada ele tinha uma versao para download e teste e o site, se nao me engano era www.metrowerks.com, mas de qualquer forma tentem www.codewarior.com. Eu ainda recomendo ou GTK+ e Glade, ou wxwindows. É necessário algo rápido, fácil e barato. -- Marcelo Elias Del ValleUIN: 30595143 [EMAIL PROTECTED] http://tilt.8m.com MLinuxer Do you visit GamesNow today?? [EMAIL PROTECTED] http://games_now.netpedia.net Autronix - Tecnologia em Automação Industrial http://members.xoom.com/autronix
libg++
i am install StarOffice-3.1 from a tar because i could not find it in a deb format. it say upon installation, can't load library libg++.so.27 i have the g++ engine installed. what package could i be missing? thankx as always beavis
Re: libg++
nevermind found what i needed! - Original Message - From: Beavis To: debian list Sent: Friday, March 24, 2000 4:03 PM Subject: libg++ i am install StarOffice-3.1 from a tar because i could not find it in a deb format. it say upon installation, can't load library libg++.so.27 i have the g++ engine installed. what package could i be missing? thankx as always beavis
2 monitors on a Rage Mobility
Hi, A friend of mine has a Laptop with a ATI Rage Mobility. When running Windoze he can connect another monitor to his laptop and use both his TFT and the CRT, with a resolution of 1600x600 of the entire desktop. He can't get that to work under Linux. Is there anyway it is possible? I know he had a hard time finding the correct X-server, and now it doesn't support this feature. TIA Bart PS This and getting VMWare to work fine are the only things that keeps him from swapping from MS to Linux. I'm helping the Linux community to get another soul ;-).
Syquest EZFLyer support in Linux
Hey, I have this zip-type drive made by Syquest (the EZFlyer 230MB). I was wondering if their was any support for it in Linux. It's an external drive that plugs into the parallel port. I've heard that you can get Iomega drives similar to this one to work, so I was wondering if this would work to. Thanks, Cameron Matheson
Re: Squid ACLs does not work
OK, I've tried it on my setup and the answer seems to be that you have your http_access statements in the wrong order; try re-arranging this section of squid.conf as follows: http_access allow manager localhost http_access deny manager http_access allow purge localhost http_access deny purge http_access deny !Safe_ports http_access deny CONNECT !SSL_ports http_access deny BanDomains http_access allow localdomain AFAICT, squid uses the first matching ACL that it can find; because you had http_access allow localdomain at the head of the list, squid allows any request from localdomain without reference to subsequent controls. This would also make it important to place these http_access statements after those controlling access to the cachemanager, etc. On Fri, Mar 24, 2000 at 09:09:04PM +0100, [EMAIL PROTECTED] wrote Yes, I ran /etc/init.d/squid restart to reload the config file and the /etc/ban_domains.squid is readable to all, so this should no be a problem. Sven On 24-Mar-2000 John Pearson wrote: On Thu, Mar 23, 2000 at 11:13:42PM +0100, [EMAIL PROTECTED] wrote Hi, I have some problems with squid and its ACLs. I'm using Debian 2.2 with Kernel 2.2.13 and squid 2.2STABLE5. My ACL section in /etc/squid.conf looks like the following. acl all src 0.0.0.0/0.0.0.0 acl manager proto cache_object acl localhost src 127.0.0.1/255.255.255.255 acl SSL_ports port 443 563 acl Safe_ports port 80 21 443 563 70 210 1025-65535 acl purge method PURGE acl CONNECT method CONNECT acl BanDomains dstdomain /etc/ban_domains.squid acl localdomain srcdomain localdomain.own : http_access allow localdomain http_access deny BanDomains http_access allow manager localhost http_access deny manager http_access allow purge localhost http_access deny purge http_access deny !Safe_ports http_access deny CONNECT !SSL_ports And the file /etc/ban_domains.squid looks like... netscape.com microsoft.com msdn.com realnetworks.com But when I try connect to www.microsoft.com the proxy rersolves the hostname and connects. (My browser is configured to use the proxy, of course...). Does anyone have an idea where I made a mistake? HTH, John P. -- [EMAIL PROTECTED] [EMAIL PROTECTED] Oh - I - you know - my job is to fear everything. - Bill Gates in Denmark
Re: apt-get complaining about dependencies, and other upgrade issues
Try downloading the g++ package manually (version 2.91.58 or greater) and installing it using dpkg. Another (brute force) solution would be dpkg -i --force-depends *, again: I don't wish to be held responsible for anything that might go wrong! Ron On Fri, 24 Mar 2000, Jeronimo Pellegrini wrote: :: On Fri, 24 Mar 2000 15:20:59 +0100 (CET), Ron Rademaker [EMAIL PROTECTED] said: What did you mean by go to the main/base and later to the main/lib? If you meant subdirectories, there was none in /var/cache/apt/archives. Anyway, I did what you said a few times, but even though bluefish was upgraded and is working (well, it only crashed when I called weblint)... dpkg still complains about: dpkg: dependency problems prevent configuration of libstdc++2.10-dev: libstdc++2.10-dev depends on g++ (= 2.91.58); however: Package g++ is not configured yet. dpkg: error processing libstdc++2.10-dev (--install): dependency problems - leaving unconfigured Errors were encountered while processing: gcc_1%3a2.95.2-7_i386.deb g++ g77 gobjc libstdc++2.10-dev (?) Maybe upgrading to potato from the state it is now could be a good idea (?) J. apt downloads all your deb packages in /var/cache/apt/archives, so I just went there, to the main/base section and did: dpkg -i --force-depend-versions,conflict * (in other words install everything and don't worry about version depencies or conflicts (conflict are for perl 5.004 to 5.005), it still gives errors but just ignore them. Then go to the main/lib and do the same, then do a dpkg -a --configure, then a apt-get upgrade, if there are still errors repeat the first steps again till it works, if it doesn't allow you to go any further install any packages that apt complains about manually with dpkg -i package (apt has already downloaded it for you). Ron -- Jeronimo Pellegrini Institute of Computing - Unicamp - Brazil http://www.dcc.unicamp.br/~jeronimo mailto:[EMAIL PROTECTED]mailto:[EMAIL PROTECTED]
Re: apt-get complaining about dependencies, and other upgrade issues
I've only once upgraded from hamm to slink and I haven't encountered any problems with that upgrade... i fyou encounter any upgrades just post them on this thread (or start another apt or dpkg thread because I read most of them). Ron On Fri, 24 Mar 2000, Jeronimo Pellegrini wrote: In case I want to upgrade to potato, are the same recommendations for upgrading to slink valid? Like, not upgrading from X, but from text mode... And... Anything else? Thanks a lot! j. -- Jeronimo Pellegrini Institute of Computing - Unicamp - Brazil http://www.dcc.unicamp.br/~jeronimo mailto:[EMAIL PROTECTED]mailto:[EMAIL PROTECTED]
Re: apt-get complaining about dependencies, and other upgrade issues
:: On Sat, 25 Mar 2000 04:43:38 +0100 (CET), Ron Rademaker [EMAIL PROTECTED] said: Try downloading the g++ package manually (version 2.91.58 or greater) and installing it using dpkg. Another (brute force) solution would be dpkg -i --force-depends *, again: I don't wish to be held responsible for anything that might go wrong! Don't worry. I have my /home dir in a separate partition. I'm beginning to think it's easier to re-install slink from my CD-rom and then begin doing upgrades - but not mixing slink with potato... And later I'll upgrade to potato. Anyway - how do I upgrade to potato? Just include frozen in sources.list, apt-get update, and then (not running X), apt-get upgrade? Is that all, os is there something else I have to know? Ron Thanks, J. -- Jeronimo Pellegrini Institute of Computing - Unicamp - Brazil http://www.dcc.unicamp.br/~jeronimo mailto:[EMAIL PROTECTED]mailto:[EMAIL PROTECTED]
Re: Limiting user access in ftp, ssh, samba, etc... 'passwords'
On Fri, Mar 24, 2000 at 09:31:25PM +0100, Antonio Fiol Bonnín wrote: snip I want to have easy freedom in limiting user access. I have killed telnetd, and only sshd. I want to allow some users access through ssh, some through ftpd, and some through samba. How can I turn off user access through ssh, but keep their account, and allow them access through ftp? Can I allow users access to shares through samba, and allow them to ftp in, but not ssh or telnet? Just a note there: I imagine that if you install and make your users use ssh, it's because you do not want the passwords of the users to be visible through the net, isn't it? Other than that, you propose the use of ftp, which uses non-ciphered communication. Well, all the passwords that were not visible with ssh are now visible with ftp. this is a very good point, but as far as i have been able to find, there is no suitable, secure, replacement for ftp (why!?!?!) the only solutions i see are: 1) force users with ssh access to exclusivly use scp. problem with this is some ssh clients on broken platforms either do not have scp, or have very crappy/clumsy implementations. you also have to enter a password over and over again (afait) perhaps this can be delt with by using ssh-agent (i need to get around to looking at that..) but only on *nix platforms. and if you have say a win* or worse macos user, telling them they can't use there purdy drag and drop ftp client will get you shot. and finally if you don't want to give ftp users ssh access this is not an option. (though for those you can put them in a chroot jail and minimize the risks) 2) use ssh to tunnel the ftp connection. I may be doing something wrong but i have never managed to get this to work. this is also a somewhat obscure solution to propose to the clients. it also (again afaict) requires the user to have a interactive ssh account. again this won't work for ftp only users. 3) I think there is SSLized ftp but this would require a SSLized client which are probably not avaiable on macos. (if you have mac users this is a problem) and if the user likes thier more purdy insecure client better you have the problem in 1)... 4) IPsec or some other VPN that encrypts the whole damn thing all the way down to pings. but this is a rather complicated solution, especially with the macos/win* user problem. (or simply lots of users in general) so what can you do? maintain a separate passwd file for ftp? you still have a fairly large breach if someone gets into the ftp account, and the user cannot change this password (i assume using pam_pwdfile) you also get annoyed users having to remember two password and likely just setting the ssh password the same, eliminating any advantage... ftp seems to be an evil that is very hard to eliminate, it has no cross platform secure replacement (yes i have heard of sftp with ssh2, 2 words: no clients and 1(2?) more: non-free) if all your users actually care about your security there are solutions (user gets a *nix OS and uses scp+ssh-agent ;-)) but for most users who don't care about your security if it interferes with there ability to use the purdy insecure macos/win* client software your screwed... I would like nothing better then to be proven wrong on this one! i dare you! show me a solution that solves the problem of convenient file transfer from stubborn users who insist on ease of use over security (purdy GUI drag and drop ftp client over a clumsy scp client on their broken OS) as well as ftp only accounts without promisquous password blabbing across the net. -- Ethan Benson http://www.alaska.net/~erbenson/
yawmppp (not a debian package)
Has anybody managed to get this working under Debian?? I have tried everything in the included faq but still no luck! Roy Pluschke [EMAIL PROTECTED]
Re: esound, lesstif, debian issues
On Thu, Mar 23, 2000 at 05:06:48PM +0100, Lepus wrote: Hi. First of all, it is quite interesting that I didn't get a single reply on my question about the esound dependency problems. Am I he only one who would like to use Gnome and Alsa at the same time under Potato? :) Well, anyway. What was the problem? Right now, I'm doing some research to put together a sound program for linux, and I was under the impression that a program written for OSS/Free would work with Alsa. Since esound is written for OSS as far as I know, I would think that it works under Alsa. Correct me if I'm wrong, though. It would mean that I have to do some more reading. Actually, after I wrote this, I find that I am wrong. Esound is written using the Alsa API, too. So you should definitely be able to use it with Alsa. Reading the source sometimes helps. :-P But correct me if I'm wrong about a program written for OSS being able to run on Alsa. I can't imagine that the Alsa people would break all the other sound apps and still have a user base. Cheers, Chris Gray -- pick, pack, pock, puck: like drops of water in a fountain falling softly in the brimming bowl.
Re: apt-get complaining about dependencies, and other upgrade issues
On Sat, Mar 25, 2000 at 12:54:37AM -0300, Jeronimo Pellegrini wrote: :: On Sat, 25 Mar 2000 04:43:38 +0100 (CET), Ron Rademaker [EMAIL PROTECTED] said: Try downloading the g++ package manually (version 2.91.58 or greater) and installing it using dpkg. Another (brute force) solution would be dpkg -i --force-depends *, again: I don't wish to be held responsible for anything that might go wrong! Don't worry. I have my /home dir in a separate partition. I'm beginning to think it's easier to re-install slink from my CD-rom and then begin doing upgrades - but not mixing slink with potato... And later I'll upgrade to potato. Anyway - how do I upgrade to potato? Just include frozen in sources.list, apt-get update, and then (not running X), apt-get upgrade? Make the last one apt-get dist-upgrade -- Bob Nielsen, N7XY (RN2)[EMAIL PROTECTED] Tucson, AZ DM42nh QRP-L #1985 SOC #77http://www.primenet.com/~nielsen
Re: ...no Masquerade...?
On Fri, Mar 24, 2000 at 02:31:32PM -0500, Andrew Sullivan wrote: No trouble. The other fellow's responses about ipchains c. may also be true. I don't know whether the stock kernel comes with masquerading turned on. Your remarks about what responds to modprobe, though, suggest that you do need to use ipchains. You'll need to remove the ipfwadm module first. Also, get rid of it in modules.conf; you'll need to have a look at the docs for modutils. Once you've taken that out of the kernel (and prevented it from auto-loading), you can use ipchains. Hmm; looks like -nothing's- in the kernel (and no mention of any of these in modules.conf): www2:~# modprobe ipchains modprobe: Can't locate module ipchains www2:~# modprobe ipfwadm modprobe: Can't locate module ipfwadm www2:~# modprobe ipmasq modprobe: Can't locate module ipmasq www2:~# ipmasq IP Masquerade has not been enabled in the kernel. Eh..? -- -- Jeff -- http://www.wellnow.com There's nothing left in the world to prove. All that's worth doing is to love one another, using whatever means are available to serve.
how to stop inetd,nfs,etc. daemons?
hello, how do I stop these daemons without manually deleting them from /etc/init.d? how the heck do I use update-rc.d -f remove -stop something? anyway... I mainly use linux for web surfing,email so what are the other daemons that I do not need and thus can be removed aside from nfs,inetd,lpd?? portmap,ksyslog perhaps? any insights would be appreciated. thanks. __ Get Your Private, Free Email at http://www.hotmail.com
Re: Is it possible for finger to not show .forward?
On Tue, Mar 21, 2000 at 03:49:34PM -, Pollywog wrote: On 21-Mar-2000 Oswald Buddenhagen wrote: Hello, I finally got around to trying out exim's built-in sorting, and it seems to be working great so far. [...] Is there a way to get finger to not show .forward? simply chmod 600 .forward - at least on solaris this works ... I don't think that works on Linux; it did not work for me, anyway. -- Andrew Tried it and it seems to work okay here. (slink) -- David Karlin [EMAIL PROTECTED] Powered by Debian GNU/Linux
Re: how to stop inetd,nfs,etc. daemons?
john == john smith [EMAIL PROTECTED] writes: hello, how do I stop these daemons without manually deleting them from /etc/init.d? how the heck do I use update-rc.d -f remove -stop something? anyway... I mainly use linux for web surfing,email so what are the other daemons that I do not need and thus can be removed aside from nfs,inetd,lpd?? portmap,ksyslog perhaps? Probably not suggested that you stop ksyslog, just in case something happens and you do need info. Actually, can't you just unlink the script from /etc/rc?.d to not have whatever the script started in whatever runlevel you whan. Woo, How can you tell I'm tired...Grammar goes to pot. Marshal any insights would be appreciated. thanks. __ Get Your Private, Free Email at http://www.hotmail.com -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: ...no Masquerade...?
On Sat, Mar 25, 2000 at 01:11:23AM -0500, Jeff Gordon wrote On Fri, Mar 24, 2000 at 02:31:32PM -0500, Andrew Sullivan wrote: No trouble. The other fellow's responses about ipchains c. may also be true. I don't know whether the stock kernel comes with masquerading turned on. Your remarks about what responds to modprobe, though, suggest that you do need to use ipchains. You'll need to remove the ipfwadm module first. Also, get rid of it in modules.conf; you'll need to have a look at the docs for modutils. Once you've taken that out of the kernel (and prevented it from auto-loading), you can use ipchains. Hmm; looks like -nothing's- in the kernel (and no mention of any of these in modules.conf): www2:~# modprobe ipchains modprobe: Can't locate module ipchains www2:~# modprobe ipfwadm modprobe: Can't locate module ipfwadm www2:~# modprobe ipmasq modprobe: Can't locate module ipmasq www2:~# ipmasq IP Masquerade has not been enabled in the kernel. Eh..? Um.. in spite of what Andrew said, they're not modules. ipfwadm is an IP packet firewall/masquerading setup utility that works with kernel 2.0.x; ipchains is similar, but for kernel 2.2.x. To see what masquerading-related modules you have, look in /lib/modules/kernel version/ipv4; with stock kernels, which have IP firewalling masquerading built-in, you should see a bunch of modules for specific protocols, like ip_masq_ftp.o. If you're using a stock Debian kernel you shouldn't need to do anything fancy to use masquerading; try starting with just # ipfwadm -I -l for kernel 2.0.x, or # ipchains -L input for kernel 2.2.x. This should list the default policy and rules for accepting incoming packets, if your kernel supports IP firewalling (which is required for IP masquerading). John P. -- [EMAIL PROTECTED] [EMAIL PROTECTED] Oh - I - you know - my job is to fear everything. - Bill Gates in Denmark
Re: ...no Masquerade...?
Hi, John -- Um.. in spite of what Andrew said, they're not modules. ipfwadm is an IP packet firewall/masquerading setup utility that works with kernel 2.0.x; ipchains is similar, but for kernel 2.2.x. (Okay.) # ipchains -L input for kernel 2.2.x. This should list the default policy and rules for accepting incoming packets, if your kernel supports IP firewalling (which is required for IP masquerading). Okay--the policy at present is one I set up awhile ago, basically allowing bidirectional forwarding on everything. :-) We realized that if we didn't do that, my brother's packets would never make it out the door -- but we're still left with problems of how to get responding packets back to his machine, which (as I understand it, anyway) is where the Masq bits come into play -- and the kernel is saying Masquerading is not enabled -- so are we back at Andrew's original statement, that I need to compile a kernel in which Masquerading -is- enabled as the next order of business...? (I hope so 'cause I'm downloading about 18 megs right now in order to do that. :-) -- -- Jeff -- http://www.wellnow.com There's nothing left in the world to prove. All that's worth doing is to love one another, using whatever means are available to serve.
Re: how to stop inetd,nfs,etc. daemons?
On Sat, 25 Mar 2000, john smith wrote: hello, how do I stop these daemons without manually deleting them from /etc/init.d? how the heck do I use update-rc.d -f remove -stop something? anyway... I mainly use linux for web surfing,email so what are the other daemons that I do not need and thus can be removed aside from nfs,inetd,lpd?? portmap,ksyslog perhaps? Go ahead and manually delete links in the /etc/rcN.d directories. I don't think the update-rc.d command adds enough functionality to warrant learning about it, so why use it (aside from within a package's installation scripts). Keep inetd (InterNETDaemon), aside from browsing the web, some progs use a daemon-client model, this means you also want portmap; lpd is the LinePrinterDaemon, you probably want to keep it (needs inetd to work, IIRC); sysklog is the System and Kernel LOGging daemon, definately keep it running (but you may want to configure logging to be less redundant, see /etc/syslog.conf); NetworkFileSystem can go if you don't mount directories on a network. I suggest you play with one of the rcN.d dirs (not the default runlevel, rc2.d, unless you changed it in /etc/inittab) and see what happens. At worst you will need to reboot. man telinit should get you started. The files in init.d are scripts; you can look at them and see what they are doing, no need to blindly disable stuff unless you are adventurous. Run levels can be a powerful tool, especially if you want or need to get the best performance from your machine. I have set runlevels for: general use; optimized for compiling; optimized for X; one that is untouched; and another that is not much better than single user mode. It makes a big difference on this old and slow box (486-25), but I guess that depends on the system you are running and why you want to disable the daemons and setup scripts. If you are thinking of security issues, you should look at the /etc/services file. any insights would be appreciated. Hope that helps. - Bruce
Expose your business to the Internet
PUT EMAIL MARKETING TO WORK FOR YOU... Call NOW and receive 50,000 FREE emails with your order! WE HAVE OPT-IN LISTS see below for removal. Special Ends Friday March 31, 2000 MLM'ers, We can build your downline. Imagine having a product or idea and selling it for only $10. Now imagine sending an ad for your product or idea to 25 million people! If you only get a 1/10 of 1% response you have just made $250,000!! You hear about people getting rich off the Internet everyday on TV, now is the perfect time for you to jump in on all the action. FACT. With the introduction of the Internet, one primary KEY to conducting your business successfully is creating massive exposure in a cost effective manner. FACT. The experts agree that email marketing is one of the most cost effective forms of promotion in existence today. Electronic mail has overtaken the telephone as the primary means of business communication.(American Management Association) Of online users 41 percent check their email daily. A gold mine for those who can take advantage of bulk email programs- The New York Times Email is an incredible lead generation tool -Crains Magazine Blows away traditional Mailing-Advertising Age It's truly arrived. Email is the killer app so far in the online world-Kate Delhagen, Forrester Research Analyst Why not let a professional company handle your direct email marketing efforts for you? *We will assist you in developing your entire campaign! *We can even create your ad or annoucement for you! *No responses? We resend at no cost! For More Information CALL NOW-702-248-1043 For removal see below. SPECIAL RATES SPECIAL ENDS Friday March 31, 2000 Targeted Rates Upon Request. BONUS!!! Call In and receive 50,000 Extra Emails at No Cost! Call NOW - 702-248-1043 ++ We are terribly sorry if you received this message in error. If you wish to be removed. Please, type REMOVE in the subject line: [EMAIL PROTECTED] ++
Re: Limiting user access in ftp, ssh, samba, etc... 'passwords'
On Fri, 24 Mar 2000, Ethan Benson wrote: On Fri, Mar 24, 2000 at 09:31:25PM +0100, Antonio Fiol Bonnín wrote: snip I want to have easy freedom in limiting user access. I have killed telnetd, and only sshd. I want to allow some users access through ssh, some through ftpd, and some through samba. How can I turn off user access through ssh, but keep their account, and allow them access through ftp? Can I allow users access to shares through samba, and allow them to ftp in, but not ssh or telnet? Just a note there: I imagine that if you install and make your users use ssh, it's because you do not want the passwords of the users to be visible through the net, isn't it? Other than that, you propose the use of ftp, which uses non-ciphered communication. Well, all the passwords that were not visible with ssh are now visible with ftp. this is a very good point, but as far as i have been able to find, there is no suitable, secure, replacement for ftp (why!?!?!) Ask Bill... the only solutions i see are: 1) force users with ssh access to exclusivly use scp. problem with this is some ssh clients on broken platforms either do not have scp, or have very crappy/clumsy implementations. you also have to enter a password over and over again (afait) perhaps this can be delt with by using ssh-agent (i need to get around to looking at that..) but only on *nix platforms. and if you have say a win* or worse macos user, telling them they can't use there purdy drag and drop ftp client will get you shot. and finally if you don't want to give ftp users ssh access this is not an option. (though for those you can put them in a chroot jail and minimize the risks) I believe that a chroot'ed ftp may work well for you, as long as you do not allow ssh users to log in the ftp, nor the ftp users log in the ssh. 2) use ssh to tunnel the ftp connection. I may be doing something wrong but i have never managed to get this to work. this is also a somewhat obscure solution to propose to the clients. it also (again afaict) requires the user to have a interactive ssh account. again this won't work for ftp only users. 3) I think there is SSLized ftp but this would require a SSLized client which are probably not avaiable on macos. (if you have mac users this is a problem) and if the user likes thier more purdy insecure client better you have the problem in 1)... 4) IPsec or some other VPN that encrypts the whole damn thing all the way down to pings. but this is a rather complicated solution, especially with the macos/win* user problem. (or simply lots of users in general) so what can you do? maintain a separate passwd file for ftp? you still have a fairly large breach if someone gets into the ftp account, and the user cannot change this password (i assume using pam_pwdfile) you also get annoyed users having to remember two password and likely just setting the ssh password the same, eliminating any advantage... separate chroot'ed ftp, and user password changing via chroot'ed telnetd. You setup the user's shell to /bin/passwd and that should be it. (Never tested). ftp seems to be an evil that is very hard to eliminate, it has no cross platform secure replacement (yes i have heard of sftp with ssh2, 2 words: no clients and 1(2?) more: non-free) if all your users actually care about your security there are solutions (user gets a *nix OS and uses scp+ssh-agent ;-)) but for most users who don't care about your security if it interferes with there ability to use the purdy insecure macos/win* client software your screwed... I would like nothing better then to be proven wrong on this one! i dare you! show me a solution that solves the problem of convenient file transfer from stubborn users who insist on ease of use over security (purdy GUI drag and drop ftp client over a clumsy scp client on their broken OS) as well as ftp only accounts without promisquous password blabbing across the net. You won't get security if users do not accept it. However, you can propose them the solution I told you (chroot ftp+telnet/passwd), telling them that they have NO security at all on their files stored there. Install a good log-parser to analyze everything that happens, and if you see something strange... well just analyze them yourself by hand ;) Antonio
Re: Limiting user access in ftp, ssh, samba, etc... 'passwords'
On Sat, Mar 25, 2000 at 11:38:38AM +0100, FIOL BONNIN Antonio wrote: I believe that a chroot'ed ftp may work well for you, as long as you do not allow ssh users to log in the ftp, nor the ftp users log in the ssh. for ftp only yes chroot works quite well. its when you combine shell access and quick file transfer (say for users maintaining a web site or portion of one) separate chroot'ed ftp, and user password changing via chroot'ed telnetd. You setup the user's shell to /bin/passwd and that should be it. (Never tested). hmm, this might work but means blowing a second account on every shell user needing file transfer abilities. and there is no way to keep them from setting both passwords the same making the entire setup useless. You won't get security if users do not accept it. However, you can propose them the solution I told you (chroot ftp+telnet/passwd), telling them that they have NO security at all on their files stored there. if they are maintaining a web site they generally don't care if everyone can access thier files, everyone already can over on port 80... when it comes to compromising the entire system by sending shell passwords over the net in clear with ftp, they don't care if it means giving up ftp altogether... afterall its not thier problem if the system gets cracked and has to be rebuilt... (users are so annoying g) i suppose the only solution until (if) a secure ftp comes along is to be a total BOFH and simply say scp or die, if one can get away with it anyway ;-) Install a good log-parser to analyze everything that happens, and if you see something strange... well just analyze them yourself by hand ;) i need to check out logparsers again, lastime i tried logcheck i ended up dumping it since i got tired of mail delivery being considered suspicious ;-) (didn't have time to rewrite all the rules..) Antonio -- Ethan Benson http://www.alaska.net/~erbenson/
Re: apt-get complaining about dependencies, and other upgrade issues
That's about it, apt-get update, apt-get dist-upgrade And... finished! Ron On Sat, 25 Mar 2000, Jeronimo Pellegrini wrote: :: On Sat, 25 Mar 2000 04:43:38 +0100 (CET), Ron Rademaker [EMAIL PROTECTED] said: Try downloading the g++ package manually (version 2.91.58 or greater) and installing it using dpkg. Another (brute force) solution would be dpkg -i --force-depends *, again: I don't wish to be held responsible for anything that might go wrong! Don't worry. I have my /home dir in a separate partition. I'm beginning to think it's easier to re-install slink from my CD-rom and then begin doing upgrades - but not mixing slink with potato... And later I'll upgrade to potato. Anyway - how do I upgrade to potato? Just include frozen in sources.list, apt-get update, and then (not running X), apt-get upgrade? Is that all, os is there something else I have to know? Ron Thanks, J. -- Jeronimo Pellegrini Institute of Computing - Unicamp - Brazil http://www.dcc.unicamp.br/~jeronimo mailto:[EMAIL PROTECTED]mailto:[EMAIL PROTECTED]
Help modifying a legacy cdrom header file for config and compile
Hello all, I'm setting up a 486DX with a legacy 2x cdrom. (Why? I must be either a masochist or a retrotech.;) The hardware in question is an early SoundBlaster 16 and a Creative Labs CR-563 cdrom. Both are supported in module sbpcd. I've used another eide cdrom to install the system including the sbpcd module. At boot the 2x is detected but not configured and is not available for mounting. /proc/modules lists it as sbpcd 51616 0 (unused). Installation said to configure the sbpcd.h header file (/usr/include/linux/sbpcd.h) and at boot time after probing the default and failing, the system tries again and delivers the following information: SBPCD -0 [x]: Scanning 0x230 (SoundBlaster) [x]: Drive 0 (ID=0): CR-563 (0.75) at 0x230 (type 1) Read linux/Documentation/cdrom/sbpcd and then configure sbpcd.h for your hardware [x]: Data Buffer Size 8 frames. It is at this point that I am stuck - not knowing how to modify the header file sbpcd.h appropriately. The two files above are the ones needed and sbpcd.h is well documented but example free when it comes to syntax. I am not sure whether to use include, define or nothing in adding the following 4 items to the first couple of pages of sbpcd.h. I am pretty sure the information is correct I just don't know how to set this up before a recompile of the kernel so as to be able to use the cdrom normally and remove the eide cdrom for use elsewhere. [from sbpcd.h] 1. Specify interface address and interface type.??? sbpcd=0x230,SoundBlaster#OR its equivalent sbpcd=0x230,1 #SoundBlaster=1 Should I just enter one of these lines as is or use a prefix 2. Define your CDROM port-base address as CDROM_PORT and specify the type of your interface card as SBPRO.?? CDROM_PORT=SBPRO Should I enter one in terms of the other or separately or use a prefix 3. Set SBPRO to 1 for 'true' SoundBlaster cards.? SBPRO 1 As is or with a prefix? 4. Set this to 0 once you have configured your interface definitions right.? #define DISTRIBUTION 1 Should I just remove the '#' sign or the '#device' prefix Many thanks in advance Bill Small is beautiful. Keep it simple.
From field in the 'mail' program
How could I set a mail address (like in this mail) in the 'mail' program? Thanks Attila -- -- - Mail: [EMAIL PROTECTED]; Debian 2.2 Linux / 2.2.13 / exim- - Get my PGP key: gpg --keyserver keys.pgp.com --recv-key 0x2cc33acb -
Re: Squid ACLs does not work
At 11:35 AM 3/24/00 +1200, C. Falconer wrote: [snip] Squid ACLs are messy and not really intended for filtering based on URLs - rather they seem to be for controlling what machines can access your squid cache, and which domains your clients get direct (uncached) access to. I do not agree with you: acl proxyallow url_regex /etc/squid.allow acl proxydeny url_regex /etc/squid.deny and http_access allow proxyallow allowed_hosts http_access deny proxydeny http_access allow allowed_hosts http_access deny all In my squid file do the job just fine! The allow and deny files are all the tools you need. The keywords are flat ASCII and row based and give all the flexibility you need. I don't see the need for any extra software. Regards, Onno
Re: ...no Masquerade...?
On Sat, Mar 25, 2000 at 03:02:52AM -0500, Jeff Gordon wrote Hi, John -- Um.. in spite of what Andrew said, they're not modules. ipfwadm is an IP packet firewall/masquerading setup utility that works with kernel 2.0.x; ipchains is similar, but for kernel 2.2.x. (Okay.) # ipchains -L input for kernel 2.2.x. This should list the default policy and rules for accepting incoming packets, if your kernel supports IP firewalling (which is required for IP masquerading). Okay--the policy at present is one I set up awhile ago, basically allowing bidirectional forwarding on everything. :-) We realized that if we didn't do that, my brother's packets would never make it out the door -- but we're still left with problems of how to get responding packets back to his machine, which (as I understand it, anyway) is where the Masq bits come into play -- and the kernel is saying Masquerading is not enabled -- so are we back at Andrew's original statement, that I need to compile a kernel in which Masquerading -is- enabled as the next order of business...? (I hope so 'cause I'm downloading about 18 megs right now in order to do that. :-) Are you using a stock Debian kernel, or one which you built yourself? The stock kernels usually include masquerading support. If you're using a Stock 2.2.x kernel you should see masquerading modules (ip_masq_*.o) under /lib/modules/2.2.14/ipv4 (assuming kernel version 2.2.14); if you do then your kernel already has masquerading support built in (if it *is* a stock kernel then you should also have a file like /boot/config-2.2.14 that shows you the kernel configuration used). If you're compiling your own, you need to include support for (assuming kernel 2.2.x) Network Firewalls, IP Firewalls and IP Masquerading. You also need /proc filesystem support and sysctl support (under General Options). If you are using a 2.2.x kernel, also bear in mind that IP forwarding has to be enabled for IP masquerading to work; you can enable forwarding with # echo 1 /proc/sys/net/ipv4/ip_forward and see if it is enabled with # cat /proc/sys/net/ipv4/ip_forward This step is not relevant to 2.0.x kernels; if they have forwarding enabled at compile time then it is enabled. Finally, here are the ipchains rules that perform masquerading on my machine, running kernel 2.2.14: # /sbin/ipchains -L -n Chain input (policy DENY): target prot opt sourcedestination ports ACCEPT all -- 0.0.0.0/0 0.0.0.0/0 n/a ACCEPT all -- 192.168.1.0/240.0.0.0/0 n/a ACCEPT all -- 0.0.0.0/0 203.55.241.211n/a DENY all l- 192.168.1.0/240.0.0.0/0 n/a Chain forward (policy DENY): target prot opt sourcedestination ports MASQ all -- 192.168.1.0/240.0.0.0/0 n/a Chain output (policy DENY): target prot opt sourcedestination ports ACCEPT all -- 0.0.0.0/0 0.0.0.0/0 n/a ACCEPT all -- 0.0.0.0/0 192.168.1.0/24n/a ACCEPT all -- 203.55.241.2110.0.0.0/0 n/a DENY all l- 0.0.0.0/0 192.168.1.0/24n/a # I use the ipmasq package to do this for me; I'm using version 3.2.5, which seems to work here. The only extra tweaking I've done (AFAICR) is to add the line modprobe ip_masq_ftp to the end of /etc/ppp/ip-up.d/00ipmasq; you may want to load the modules (if any) for the protocols you require there, also. Good luck, John P. -- [EMAIL PROTECTED] [EMAIL PROTECTED] Oh - I - you know - my job is to fear everything. - Bill Gates in Denmark
Re: Limiting user access in ftp, ssh, samba, etc... 'passwords'
On Fri, Mar 24, 2000 at 06:59:35PM -0900, Ethan Benson wrote: this is a very good point, but as far as i have been able to find, there is no suitable, secure, replacement for ftp (why!?!?!) 2) use ssh to tunnel the ftp connection. I may be doing something wrong but i have never managed to get this to work. this is also a somewhat obscure solution to propose to the clients. it also (again afaict) requires the user to have a interactive ssh account. again this won't work for ftp only users. See the man pages for ssh-agent. -- Andrew Sullivan Computer Services [EMAIL PROTECTED]Burlington Public Library +1 905 639 3611 x158 2331 New Street Burlington, Ontario, Canada L7R 1J4
Re: ...no Masquerade...?
On Sat, Mar 25, 2000 at 05:37:19PM +1030, John Pearson wrote: Um.. in spite of what Andrew said, they're not modules. I hate it when I write without my brain engaged. Sorry, I wasn't thinking. -- Andrew Sullivan Computer Services [EMAIL PROTECTED]Burlington Public Library +1 905 639 3611 x158 2331 New Street Burlington, Ontario, Canada L7R 1J4
Re: ...no Masquerade...?
On Sun, Mar 26, 2000 at 12:12:42AM +1030, John Pearson wrote: Are you using a stock Debian kernel, or one which you built yourself? The stock kernels usually include masquerading support. Yes--stock 'potato' 2.2.14 is the one telling me IP Masquerading is not enabled in the kernel. If you're using a Stock 2.2.x kernel you should see masquerading modules (ip_masq_*.o) under /lib/modules/2.2.14/ipv4 (assuming kernel version 2.2.14); Strange to say, I know I had those earlier (perhaps under 'slink'?) but I'm aware they went missing at some point, possibly with the first install of 'potato'. ...if you do then your kernel already has masquerading support built in (No, apparently not. You've got me wondering now whether I went through an initial config process with the upgrade that I don't remember, and in which I turned off -something- that was required for Masquerading.) (if it *is* a stock kernel then you should also have a file like /boot/config-2.2.14 that shows you the kernel configuration used). (Yes, that file is present.) If you're compiling your own, you need to include support for (assuming kernel 2.2.x) Network Firewalls, IP Firewalls and IP Masquerading. You also need /proc filesystem support and sysctl support (under General Options). I've now done that, and included the options you mention (though I never came upon the 'sysctl' option during the 'menuconfig' selection process -- I looked for it 2 or 3 times, and finally trusted that it might've been turned 'on' for me along the way; we'll find out. :-) If you are using a 2.2.x kernel, also bear in mind that IP forwarding has to be enabled for IP masquerading to work; you can enable forwarding with # echo 1 /proc/sys/net/ipv4/ip_forward and see if it is enabled with # cat /proc/sys/net/ipv4/ip_forward Okay. This step is not relevant to 2.0.x kernels; if they have forwarding enabled at compile time then it is enabled. Finally, here are the ipchains rules that perform masquerading on my machine, running kernel 2.2.14: [snip] I use the ipmasq package to do this for me; I'm using version 3.2.5, which seems to work here. The only extra tweaking I've done (AFAICR) is to add the line modprobe ip_masq_ftp to the end of /etc/ppp/ip-up.d/00ipmasq; you may want to load the modules (if any) for the protocols you require there, also. Good luck, Your generousness of spirit is appreciated, John; thanks kindly. -- -- Jeff -- http://www.wellnow.com There's nothing left in the world to prove. All that's worth doing is to love one another, using whatever means are available to serve.
Re: Lying to dpkg?
[EMAIL PROTECTED] (Jonas Steverud) wrote: [EMAIL PROTECTED] (Colin Watson) writes: [EMAIL PROTECTED] (Jonas Steverud) wrote: [...] The only problem with this is dselect will tell you the packages made by equivs are obsolete - but I maybe did some wrong when I deleted the .debs? Anyone who knows? Obsolete/local, strictly: it's the local bit that's relevant to you. You can safely ignore this, but if you recreate the .debs and run 'dpkg -A package.deb' then I believe they'll be recorded in the available file and therefore not listed as local. But it will still list the package as obsolete, right? not listed as obsolete/local, I should have said. But ... I mean, the ftpserver will not list it as available and if I'm not misstaken dselect then thinks it's obsolete. The issue isn't that very great, it's just a minor source for irritation - and only sometimes - and I was mostly curious wheather there was something I've overlooked. Ah ... indeed, you're right; now that I've tried it, it works right up to an update. In that case you might want to investigate setting up your own private repository of packages using dpkg-scanpackages; it's not too difficult to do with a bit of trial and error. I'm happy to help you with this if you have problems. -- Colin Watson [EMAIL PROTECTED]
Re: From field in the 'mail' program
Attila Csosz [EMAIL PROTECTED] wrote: How could I set a mail address (like in this mail) in the 'mail' program? I don't think you can. You can fiddle around with header rewriting in your MTA (exim, sendmail, whatever), though, which might solve your problem? For example, when I send mail using 'mail' from here the from address is [EMAIL PROTECTED] right up until it leaves my computer, when it gets rewritten to [EMAIL PROTECTED] -- Colin Watson [EMAIL PROTECTED]
Re: glibc-compat ???
On Thu, 23 Mar 2000, Steve Greenland wrote: On 23-Mar-00, 18:08 (CST), Andor Dirner [EMAIL PROTECTED] wrote: On Thu, 23 Mar 2000, Robert Varga wrote: The other one it breaks is Oracle 8.0, and one needs to convert Redhat compatibility libraries to be able install it, and a patch from Oracle. FWIW, I'm running Oracle 8i (SQL*Plus reports v 8.1.5) with the latest patches (as of a month ago) on a potato box with no obvious problems, I don't have any compatibility libs installed. I said 8.0. I know 8.1.5 works with glibc2.1 since it is explicitly stated in its requirements that it needs it. Of course it should work with it. However I don't really like 8i, since it needs much more (and it should be written as MUCH MORE) resources than 8.0.5. I know there is one aspect of using 8i on linux when compared with 8.0.5, its being free for development purposes. Robert
Re: From field in the 'mail' program
On Sat, Mar 25, 2000 at 01:27:38PM +0100, Attila Csosz wrote: How could I set a mail address (like in this mail) in the 'mail' program? ---end quoted text--- One way is to have your MTA re-write whatever you've got (which I think is the best way), or in your .muttrc: my_hdr From: whatever you want here ...and then you'll probably have to set_hdrs in the rc as well. -- David Kanter [EMAIL PROTECTED]
Re: mutt and Turkish
On 24-Mar-00 Patrick wrote: I've been searching about in www.mutt,org and can't find any reference to what I should ser charset to. Does anyome know where I can find out what the Turkish character set is and do I need to make any other changes to my system for it to work in mutt? On Fri, Mar 24, 2000 at 11:48:03AM +0900, ChangMin Oh wrote: Maybe you need to open your .muttrc, and edit set charset=~~. I don't know about mutt, but the Turkish ISO charset is iso-8859-9 I found good sets of Turkish fonts for X at ftp://ftp.linux.org.tr and ftp://compclup.ceng.metu.edu.tr/pub/linux/turkce and, if you poke around there, you may also find fonts for character consoles (VCs). Good hunting! Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Date: 25-Mar-00 Time: 13:30:53 -- XFMail --
rdate host
Anybody got the name of a good working rdate host handy? All of a sudden I can't connect to my usual, mit.edu. -- Bob Bernstein http://www.ruptured-duck.com
RCS help (possible bug)
hi everyone, i'm trying to learn rcs to assist my uni project, however I'm unsure about the procedure to roll back versions (infact i think i may have found a bug). Say I have node_relations.c at version 1.4, but its stuffed. So i want to go back to version 1.2.1.1 (turned into .1.1 because i just checked it back in after editting 1.2 without specifying a version) and check it in at 1.5 (is that the usual way of doing things?). See the middle of the log to see the 'bug'. i just add comments for you to see what i was doing... so i did a checkout to see the error: $ co node_relations.c RCS/node_relations.c,v -- node_relations.c revision 1.4 done checkout out the good version: $ co -l1.2.1.1 node_relations.c RCS/node_relations.c,v -- node_relations.c revision 1.2.1.1 (locked) done tried to check it back in without locking the latest version (oops): $ ci -r1.5 node_relations.c RCS/node_relations.c,v -- node_relations.c ci: RCS/node_relations.c,v: no lock set by wilfy for revision 1.4 locked it without checking it out: $ rcs -l node_relations.c RCS file: RCS/node_relations.c,v 1.4 locked done checked it in (good, that worked): $ ci -r1.5 node_relations.c RCS/node_relations.c,v -- node_relations.c new revision: 1.5; previous revision: 1.4 enter log message, terminated with single '.' or end of file: . done checked it out again to take a look: $ co node_relations.c RCS/node_relations.c,v -- node_relations.c revision 1.5 done = BUG === wanted to check it out and lock it, but WHY did it get the 1.2.1.1 revision?? $ co -l node_relations.c RCS/node_relations.c,v -- node_relations.c revision 1.2.1.1 (locked) done === BUG proof check it in, unlock it: $ ci node_relations.c RCS/node_relations.c,v -- node_relations.c file is unchanged; reverting to previous revision 1.2.1.1 done check it out (locked) h why is THIS now 1.5?? inconsistant operations! $ co -l node_relations.c RCS/node_relations.c,v -- node_relations.c revision 1.5 (locked) done checked it in so i could do that test again... $ ci node_relations.c RCS/node_relations.c,v -- node_relations.c file is unchanged; reverting to previous revision 1.5 done this happened before $ co node_relations.c RCS/node_relations.c,v -- node_relations.c revision 1.5 done hmm NOW it gets me the 1.5... why did it get out that 1.2.1.1 before?? BUG? $ co -l node_relations.c RCS/node_relations.c,v -- node_relations.c revision 1.5 (locked) done anyway i hope someone can help me, it would be disasterous to misuse rcs and stuff up my project :/ thanks, Paul
Re: C++
Bart, For an excellent online book see 'Thinking in C++' by Bruce Eckel http://www.MindView.net/ For mind-boggling collection of links related to object-oriented programming http://www.cetus-links.org/ Jim Cant On Fri, 24 Mar 2000 12:08:22 -0800, Bart Friederichs wrote: Hi, I am writing a C++ program and I really need info on how to overload operators (especially + and ) and info on streams. Does anybody know a URL where this kind of info can be found? TIA Bart
Re[2]: rdate host
George Bonser [EMAIL PROTECTED] wrote: tick.usno.navy.mil nfg tock.usno.navy.mil Tada!!! Thanks guy. -- Bob Bernstein http://www.ruptured-duck.com
RE: rdate host
On 25-Mar-2000 17:11:25 Bob Bernstein wrote: Anybody got the name of a good working rdate host handy? All of a sudden I can't connect to my usual, mit.edu. I have had trouble with time.nist.gov and for some reason, when my machine cannot connect, it seems to crash, but only if I attempt the update via cron. I don't have this problem when I try an update via command line and that succeeds or fails. -- Andrew
Sound error: Can't allocate DMA buffer
Does anybody know how to get around the sound dma buffer problem? I am pretty sure the problem is that the system can't find any free memory in the bottom 16MB. I think that there is a way to lock in the dma buffer, but I don't remember how, and I haven't managed to find the right documentation. I can get it working by starting a large program to swap out some memory and then kill it, but it is a pain to do that every time I want to use any audio. -- Carl Johnson[EMAIL PROTECTED]
UNIX keyboard on i386
Hello all. I'm trying to get an old UNIX keyboard working on my P100. When I restart the computer with the keyboard I get a errormessage that says something like to many NACK's, check you cables. I have located the .c file that contains that row, but havent been able to do much about it. So my questions are - is it possible to get a UNIX keyboard working on i386 board? - has this already been done, where can i find info? - do i have to make a kernel module that checks all input values it get from the new keyboard, and interpret them and send them away to be transparantly mappad to the old keyboard driver? /nisse
Re: UNIX keyboard on i386
Nils-Erik == Nils-Erik Svangård [EMAIL PROTECTED] writes: Hello all. I'm trying to get an old UNIX keyboard working on my P100. When I restart the computer with the keyboard I get a errormessage that says something like to many NACK's, check you cables. I have located the .c file that contains that row, but havent been able to do much about it. So my questions are - is it possible to get a UNIX keyboard working on i386 board? Careful, I had a friend who fried his keyboard bios trying to use a Alpha keyboard with his PC. I don't know about UNIX keyboards, but there would probably be a lot of hacking to make it work. Just an anecdote. Marshal - has this already been done, where can i find info? - do i have to make a kernel module that checks all input values it get from the new keyboard, and interpret them and send them away to be transparantly mappad to the old keyboard driver? /nisse -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
network setting up (Ethernet,PPP,PLIP,SLIP)
Hi, I am wondering how I set up my personal computer networking at home. Here is what I have. 2 computers: IBM ThinkPad Laptop running window 98 (client) Dell Desktop running Linux debian 2.1 (server) I was thinking the options of setting up the networking between them. 1. PLIP Since I have only one parallel port for each computer respectively, the parallel port on my dell destop is connected on a printer. So PLIP seems not possible to do. 2. PPP/SLIP Interesting, I have a PC card modem on Thinkpad, and another two modems on Dell. The reason I have 2 modems on the linux because one of them is winmodem, It doesn't work under linux. The another modem on the linux is used to connect to ISP through PPP dial up. If this seems to be a good approch, do I need anther modem for Linux to be able to connect to Thinkpad thru SLIP/PPP. Which is better or easier to set up , SLIP or PPP? 3. Ethernet If neither of above two options don't work, I guess I have to buy the Ethernet cards and cable for both compouters. It sounds more complicated to me. Does any of you have the good ideas of what is the best way to set them up. The objective is to allow both computers interconnected and be able to access Internet thru the current PPP dial up account. Thank you and appreciate the response. Daniel
Re: network setting up (Ethernet,PPP,PLIP,SLIP)
PLIP doesn't sound like the best solution for you, as you would need to get some kind of parallel switch - and those are notorious for causing problems rather than fixing them. SLIP/PPP sounds like a possible solution. You do not have to use the modems at all. If you have a serial port on your server (you do) and a serial port on your client (you do) then you can connect a 'null-modem' cable between the two computers. You can then set the laptop (client) as a terminal on the server (desktop), or run PPP or SLIP (PPP seems to be more common and supported) to give you a more traditional protocol for networking (TCP/IP or NetBIOS, or whatever for resource sharing). Ethernet. Although PPP will work for you, ethernet is really the only way to go. It is much faster, and although it way seem as though ethernet setup will be complicated, it is not, as linux is a network operating system, it really has the required protocols and hooks for ethernet running all the time already. My suggestions: buy a PCI ethernet card for the desktop (I like the Netgear FA-103 myself - it uses the DEC Tulip chip and there is WIDE support all over the place for it) and then find a pc-card ethernet card for the laptop. That will be the expensive part, but you should be able to find one for $50USD. There are two other options. Parallel to ethernet adapters and USB to ethernet adapters. If your laptop has USB ports (maybe) then you can buy an external adapter that converts from USB to ethernet, and almost all adapters will have Win98 drivers. The other option is very similar - a parallel to ethernet adapter that has Win98 drivers. This is a good list for setup questions - when you decide which way to go - let us help you with the configuration - but also read the relevent howtos - Ethernet, or PPP, or Serial, or PLIP, or ... . -Percival On Sat, Mar 25, 2000 at 03:55:08PM -0600, Daniel Yang wrote: Hi, I am wondering how I set up my personal computer networking at home. Here is what I have. 2 computers: IBM ThinkPad Laptop running window 98 (client) Dell Desktop running Linux debian 2.1 (server) I was thinking the options of setting up the networking between them. 1. PLIP Since I have only one parallel port for each computer respectively, the parallel port on my dell destop is connected on a printer. So PLIP seems not possible to do. 2. PPP/SLIP Interesting, I have a PC card modem on Thinkpad, and another two modems on Dell. The reason I have 2 modems on the linux because one of them is winmodem, It doesn't work under linux. The another modem on the linux is used to connect to ISP through PPP dial up. If this seems to be a good approch, do I need anther modem for Linux to be able to connect to Thinkpad thru SLIP/PPP. Which is better or easier to set up , SLIP or PPP? 3. Ethernet If neither of above two options don't work, I guess I have to buy the Ethernet cards and cable for both compouters. It sounds more complicated to me. Does any of you have the good ideas of what is the best way to set them up. The objective is to allow both computers interconnected and be able to access Internet thru the current PPP dial up account. Thank you and appreciate the response. Daniel
Can't extract base2_1.tgz no matter what
I am trying to install debian. I have previously installed RedHat on other machines and found it much easier to install. (I'm trying to work out something for public libraries to show how to set up a proxy-server.) I have a Dell Dimension XPS 466, with 64M of RAM and two old SCSI disks. I have tried downloading CD-images from ftp.eecs.umich.edu and using the cd-image making program. After that failed I decided that there was a problem with the images or my CD writer so I purchase the current CDs for Slink from LinuxMall. In ALL cases, regardless of the source the outcome was the same. I would created the floppy image from resc1440.bin with raw_write2. I would start the process at the boot prompt with linux aha152x=0x340,11,7. I would go through installing and everything would work until I got to the step Install the base system. In all cases this would end with the error message: There was a problem extracting the Base System from /instmnt/debian/dists/slink/main/disk_i386/2.1.8-1999-02-22/base2_1.tgz I have no idea what is going on. The install program appears to have no trouble with the CD. I have managed to install an old version previously (but I couldn't find dhcpcd on the old hamm cd and I figured I'd be better off with the current CD set anyhow.) I have absolutely no ideas. If it couldn't access the CD or extract files I should have had a problem long before I got to that step. All the earlier initializations, installing the kernel and system modules worked. (Is there a way to tell the install program to mount two CDs - I have two CD drives on the SCSI chain - but that is another issue.) -- Josh Kuperman [EMAIL PROTECTED]
Re: Can't extract base2_1.tgz no matter what
On Sat, Mar 25, 2000 at 06:02:35PM -0500, Josh Kuperman wrote: I have a Dell Dimension XPS 466, with 64M of RAM and two old SCSI disks. I have tried downloading CD-images from ftp.eecs.umich.edu and using the cd-image making program. After that failed I decided that there was a problem with the images or my CD writer so I purchase the current CDs for Slink from LinuxMall. In ALL cases, regardless of the source the outcome was the same. I would created the floppy image from resc1440.bin with raw_write2. I would start the process at the boot prompt with linux aha152x=0x340,11,7. I would go through installing and everything would work until I got to the step Install the base system. In all cases this would end with the error message: There was a problem extracting the Base System from /instmnt/debian/dists/slink/main/disk_i386/2.1.8-1999-02-22/base2_1.tgz I have no idea what is going on. I ran in to a similar problem installing Potato from floppies. It turned out I hadn't made my / partition large enough. Since you have two disks, make the root partition fairly large on one of the disks. It'll be easy enough to move the partitions around with cp, tar or cpio after you get the base system installed. The install program appears to have no trouble with the CD. I have managed to install an old version previously (but I couldn't find dhcpcd on the old hamm cd and I figured I'd be better off with the current CD set anyhow.) I have absolutely no ideas. If it couldn't access the CD or extract files I should have had a problem long before I got to that step. All the earlier initializations, installing the kernel and system modules worked. (Is there a way to tell the install program to mount two CDs - I have two CD drives on the SCSI chain - but that is another issue.) -- ++ | Eric G. Milleregm2@jps.net | | GnuPG public key: http://www.jps.net/egm2/gpg.asc | ++
question please help
I have Virtual PC version 3.0.(Macintosh Platform) I cannot install Debian 2.1 on the new hard drive I just created. Does anyone know how to install Debian onto Virtual PC ? please help, thanks robert reply to: [EMAIL PROTECTED]
DVD in 2.2 Debian w/2.2.14
I am trying to get DVDs to play in Linux. I read the DVD-HowTo on OpenDVD.org, but when i try to compile the CSS utilities, get get a bunch of errors dealing with the dvd stuff. Namely that dvd_struct and DVD_ consts can't be found. It seems like I'm missing a header file or something, but I can't figure out what it is. Does anyone know? I could post the error messages if needed, but if anyone's run across this, they probably know what I'm talking about. Anyone get it to work on Debian 2.2 (Which is what i'm running)? Rob -- [EMAIL PROTECTED] 1999-00 | Aka Khyron the Backstabber | LI NN N U U X X O ICQ# 2325055| LI N NN U U X | LLL I N N UUU X X O Shackles cannot keep me bound | Those who can, do. forever. I'm outta here. |
modules not working please help
I have just installed Debian potato (2.2.14). The installer gave an error when I tried to install the modules for my network cards. I know for certain that they work, I have tried them previously. They are an smc-ultra and a lance card. When I modprobe them, I get a message like: smc-ultra.c: No SMC Ultra card found (i/o = 0x280).[that is the right io value, it says it on the card] /lib/modules/2.2.14/net/smc-ultra.o: init_module: Device or resource busy /lib/modules/2.2.14/net/smc-ultra.o: insmod /lib/modules/2.2.14/net/smc-ultra.o failed /lib/modules/2.2.14/net/smc-ultra.o: insmod smc-ultra failed Please help me install them properly. Thanx Adrian
Re: Sound error: Can't allocate DMA buffer
Hi Carl, Does anybody know how to get around the sound dma buffer problem? I am pretty sure the problem is that the system can't find any free memory in the bottom 16MB. I think that there is a way to lock in the dma buffer, but I don't remember how, and I haven't managed to find the right documentation. I can get it working by starting a large program to swap out some memory and then kill it, but it is a pain to do that every time I want to use any audio. I had the same problem with some kernels, but with 2.2.14 it works fine. I saw that there is a patch at www.alsa-project.org , maybe thats what you need. /Jonas
