Re: valgrind-snapshot packaged (not officially)
On Wed, 17 Sep 2003, Andrés Roldán wrote: I'm glad to announce that valgrind-snapshot has been packaged. I've made a script that makes the package daily. It is currently not on the official Debian archives so, please, do NOT report bugs to valgrind if you found a bug in valgrind-snapshot. Instead, send me an email with the description of the problem. The source is as follows: deb http://people.fluidsignal.com/~aroldan/debian unstable main Upload this to experimental. That's what's for. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: valgrind-snapshot packaged (not officially)
On Wed, 17 Sep 2003, Andrés Roldán wrote: PS: valgrind-snapshot conflicts with valgrind for obvious reasons. Er, why? I see no reason for it to. Just rename the binaries in the package. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: valgrind-snapshot packaged (not officially)
On Wed, 17 Sep 2003, Andrés Roldán wrote: Adam Heath [EMAIL PROTECTED] writes: On Wed, 17 Sep 2003, Andrés Roldán wrote: PS: valgrind-snapshot conflicts with valgrind for obvious reasons. Er, why? I see no reason for it to. Just rename the binaries in the package. Both packages have the same shared object files (skins) located at /usr/lib/valgrind but may not be compatible each other. So change that directory name. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Anti Open Source Psyops / Mind Tweakers? (they'll never believewe're really aliens dept)
On 31 May 2003, Karl M. Hegbloom wrote: [snip] HAHAHAHAHAHA. I needed that. Thanks for the laugh. Made my day. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian on Cobalt Qube
On Sun, 16 Jun 2002, Russell Coker wrote: The kernel patch package I produced hacks the arch/i386/kernel/Makefile to produce a gzip compressed vmlinux file instead of a regular bzImage. This is because the Qube BIOS is unable to load a bzImage format kernel. bzImage is gzip compressed. bzImage just means the zImage can be larger, hence b for bigger. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Don't install lvm2!
On Sun, 12 May 2002, Andres Salomon wrote:
As for the really-bad-shape part; they work for me (tm); I don't have a
system w/ /usr on a separate partition. I apologize for the bug, but
they happen.
Read -policy and/or the packaging manual. Things in /{{s,}bin,lib} can not
use thinks in /usr.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
SHUTDOWN: Re: Scheduled downtime for murphy.debian.org(lists.debian.org)
On Thu, 10 Jan 2002, Adam Heath wrote: Brainfood is scheduling downtime for murphy.debian.org(which is also lists.debian.org, and runs all the mailing lists), to do a disk upgrade. This is just the addition of a new drive, with no copying of the existing data. We expect downtime to be minimal. The time for this maintenance is scheduled for Friday, January 11, at 9pm UTC. This is 3pm local time, for murphy. Well, it's that time. I'm leaving to go to the colo where murphy is located. I'll be shutting it down from there. This is the warning about it's shutdown.
Re: SHUTDOWN: Re: Scheduled downtime for murphy.debian.org(lists.debian.org)
On Fri, 11 Jan 2002, Adam Heath wrote: Well, it's that time. I'm leaving to go to the colo where murphy is located. I'll be shutting it down from there. This is the warning about it's shutdown. It's been back up for 30m now.
Scheduled downtime for murphy.debian.org(lists.debian.org)
Brainfood is scheduling downtime for murphy.debian.org(which is also lists.debian.org, and runs all the mailing lists), to do a disk upgrade. This is just the addition of a new drive, with no copying of the existing data. We expect downtime to be minimal. The time for this maintenance is scheduled for Friday, January 11, at 9pm UTC. This is 3pm local time, for murphy.
Re: Propossed Project: Odyssey
On Wed, 24 Oct 2001, Timothy H. Keitt wrote: Better yet, lets convince package maintainers not to unnecessarily update all their dependencies to the latest libs in unstable so that packages can be easily backported with 'apt-get -b source ...' My guess is that 60-90% of the packages in unstable do not require the latest lib versions to build, but that maintainers are defaulting their dependencies to be = the latest version in unstable for no reason (of course, package name changes and package reorganization can throw a wrench into things). If maintainers default to only depend on what is in stable whenever possible, many many deb packages would compile just fine on both stable and unstable. This shows a deep misunderstanding of the way shared libraries work. If a library is changed, and uploaded, it may require an update to its /var/lib/dpkg/info/*.shlibs file. When pkgs are then rebuilt against that library, the pkg-version dependency info is taken from this file. That is what causes newer libraries to be depended on. It is not a conscious effect on the maintainer. Additionally, if a new version of a package comes out, that depends on a new library, do you think that the new package should not be allowed into debian, on the fact that backporting to an older version of debian would be problematic? That line of thinking means nothing would ever be upgraded.
Re: X in Woody: you gotta be fscking kidding!
On Fri, 19 Oct 2001, Dimitri Maziuk wrote: 2. Dexconf asks you if you want to overwrite XF86Config-4. If you answer no, it will proceed with questions about your hardware etc. It should FOAD (which part of no does it not understand?). Not quite. It will ask you questions that you didn't answer the last time you ran it. That's the way debconf(which is what dexconf uses as its backend) works. This does not mean it will process those values.
Re: X in Woody: you gotta be fscking kidding!
On Fri, 19 Oct 2001, Dimitri Maziuk wrote: * Adam Heath ([EMAIL PROTECTED]) spake thusly: On Fri, 19 Oct 2001, Dimitri Maziuk wrote: 2. Dexconf asks you if you want to overwrite XF86Config-4. If you answer no, it will proceed with questions about your hardware etc. It should FOAD (which part of no does it not understand?). Not quite. It will ask you questions that you didn't answer the last time you ran it. That's the way debconf(which is what dexconf uses as its backend) works. This does not mean it will process those values. Let me get this staright, with dpkg-reconfigure xserver-xfree86 running on other vc: Um, no. You just asked it to reask you all the questions, by calling reconfigure.
Re: XFree86 news; users of TESTING or UNSTABLE, PLEASE READ
On Wed, 17 Oct 2001, Branden Robinson wrote: doc/xbase-clients/examples/Xmodmap, and the Xsession manual page describes how to write a replacement for /etc/X11/Xsession.d/40xfree86-common_xmodmap. Of course, if you're upgrading, you can just move the -old versions of these files back into place. Be sure to give them different names, or change their contents (say, by adding a comment), so that they aren't moved again on the next upgrade of XFree86. Why give these different names? dpkg passes the previous version to maintainer scripts, so you can check what is being upgraded from. If this version is less that 4.1.0-7, then do the move. Otherwise, not. Don't inflict additional pain on an admin, when this kind of check can be automated.
Re: exploring debian's users and groups
On Tue, 7 Aug 2001, Joey Hess wrote: Debian has always lacked an explanation of what the various users and groups are for. Such a document is useful for sysadmins who must determine the correct way to use various users and groups. It's useful for developers as well, and it might help us find unused users and groups, or find unstated requirements about use of users and groups that could be put in policy. So here's a start. There are a lot of unanswered questions; can you help me answer some of them? -- uucp: HELP: Presumably used for UUCP, which I know nothing of. This is kinda like group mail. People can be added to group uucp, then be able to call the uucp binaries, to interact with the uucp subsystem. uucp also has the ability to copy files(and run programs) between machines, and I think having it as a separate user gives a little more security. HELP: Why is minicom owned by group uucp? Is this a bug? I would think so. majordom: Majordomo has a statically allocated uid on Debian systems for historical reasons. HELP: Do we still even ship that buggy old POS? And can someone remember what the hysterical raisins were? majordomo is non-free. IMHO *NO* static user ids should be given to non-free 'POS'. postgres: HELP: Presumably used by the postgresql database? Security, to keep people from reading the database files. www-data: HELP: Er, I should know this, but this box doesn't run apache and I'm offline. No webfiles should be owned by www-data, as that is what httpd daemons run as. However, some cgi scripts(and the daemons themselves) need to write temp files, so they are given a separate user. dialout: HELP: Is this used for /dev/cua devices or something? cua devices are not used anymore. audio: This group can be used locally to give a set of users access to an audio device. ... video: This group can be used locally to give a set of users access to an video device. It has been discussed in the past, that audio is a poor name for this. We now have video, audio, mixer, joystick, cdrom, dvd(others). All are multi-media devices.
Re: apt-get dselect upgrade problems
On Thu, 12 Jul 2001, Jürgen A. Erhard wrote: As I said, that's bullshit. You're right that dpkg's available file needs to be up to date... but only for dselect and for manual browsing (do that quite often myself). dpkg itself doesn't use it AFAIK. dpkg does. if you do dpkg -i foo.deb, it will update the entry in available.
Re: root file system
On Mon, 30 Oct 2000, Art Edwards wrote: I'm noticing that my root file system has filled up. I'm having a very hard time trying to find out what is growing. Has anybody had the same thing happen? I'm running potato (kernel 2.2.17) on both an athalon and a pentium III machine (both are exhibiting the same engorgement.) BTW, what is the file kcore? It seems to be very big. When was the last time you rebooted? How often do you upgrade? The reason I ask, is that there could still be daemons running, that are using old libraries. These libraries have been deleted from disk, but because a program is still using them, the kernel hasn't actually removed them from disk, but just the reference to them. You'll see this when you run a du, and it'll be smaller than what df tells you. BEGIN GEEK CODE BLOCK Version: 3.12 GCS d- s: a-- c+++ UL P+ L !E W+ M o+ K- W--- !O M- !V PS-- PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z? -END GEEK CODE BLOCK- BEGIN PGP INFO Adam Heath [EMAIL PROTECTED]Finger Print | KeyID 67 01 42 93 CA 37 FB 1E63 C9 80 1D 08 CF 84 0A | DE656B05 PGP AD46 C888 F587 F8A3 A6DA 3261 8A2C 7DC2 8BD4 A489 | 8BD4A489 GPG -END PGP INFO-
Re: netscape 4.75 package has no netscape command?
On Mon, 28 Aug 2000, Peter S Galbraith wrote: Due to the vulnerability of my old netscape 4.08, I decided to upgrade to the woody version (Version: 4.75-1). I installed netscape-base-4, netscape-base-475 and netscape-java-475. The only file under a bin directory is netscape-remote. There is no netscape command! Am I missing something obvious? Yes, you did it wrong. apt-get install communicator-smotif-475(or navigator). This has ALWAYS been the way netscape has been done. And, in any case, I am no longer the maintainer. BEGIN GEEK CODE BLOCK Version: 3.12 GCS d- s: a-- c+++ UL P+ L !E W+ M o+ K- W--- !O M- !V PS-- PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z? -END GEEK CODE BLOCK- BEGIN PGP INFO Adam Heath [EMAIL PROTECTED]Finger Print | KeyID 67 01 42 93 CA 37 FB 1E63 C9 80 1D 08 CF 84 0A | DE656B05 PGP AD46 C888 F587 F8A3 A6DA 3261 8A2C 7DC2 8BD4 A489 | 8BD4A489 GPG -END PGP INFO-
new autofs 3.1.4-7, testers wanted
I am looking for testers for autofs. You can fetch the latest version at: http://incoming.debian.org/autofs_3.1.4-7_i386.deb I am not on debian-user, so please cc me. BEGIN GEEK CODE BLOCK Version: 3.12 GCS d- s: a-- c+++ UL P+ L !E W+ M o+ K- W--- !O M- !V PS-- PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z? -END GEEK CODE BLOCK- BEGIN PGP INFO Adam Heath [EMAIL PROTECTED]Finger Print | KeyID 67 01 42 93 CA 37 FB 1E63 C9 80 1D 08 CF 84 0A | DE656B05 PGP AD46 C888 F587 F8A3 A6DA 3261 8A2C 7DC2 8BD4 A489 | 8BD4A489 GPG -END PGP INFO-
Re: What happend to Netscape?
On Fri, 28 Jan 2000, Colin Watson wrote: [EMAIL PROTECTED] (Christian Dysthe) wrote: [Also to debian-devel] On 27 Jan, paul wrote: | 2) Libc5 version works a little better than libc6. I tried both. I have never been able to see much difference. Well, I was going to try out the libc6 version for a while when it appeared in potato (or am I going mad and it was libc6 all along?), but the netscape-base-4 maintainer appears to have decided that it should be libc5 again. :( I'm reminded of a yo-yo ... Is there any reason why we can't have netscape-libc5 and netscape-libc6 packages? It would seem to solve a lot of problems with people who need one or the other, and eliminate this switching back and forth. Yes, I am working on that. I maintain all of netscape. You will note that netscape-base-4(the wrapper) and plugger both support installations of libc5 and libc6 netscapes. I haven't yet uploaded new netscapes, tho. The scripting to build the debs isn't simple, and I need to test it all as well. BEGIN GEEK CODE BLOCK Version: 3.12 GCS d- s: a-- c+++ UL P+ L !E W+ M o+ K- W--- !O M- !V PS-- PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z? -END GEEK CODE BLOCK- BEGIN PGP INFO Adam Heath [EMAIL PROTECTED]Finger Print | KeyID 67 01 42 93 CA 37 FB 1E63 C9 80 1D 08 CF 84 0A | DE656B05 PGP AD46 C888 F587 F8A3 A6DA 3261 8A2C 7DC2 8BD4 A489 | 8BD4A489 GPG -END PGP INFO-
DO NOT UPGRADE TO POTATO. MENU UPLOAD ON OCT 2 KILLS SYSTEMS
I just did an upgrade. The menu pkg ate memory like no tomorrow. I have a dual-330, 256m ram, 384m swap. Update-menus calls install-menu, and I saw that eating 280m of memory. root 19580 21.6 83.3 282784 215152 pts/8 R15:51 0:13 install-menu /etc/menu-methods//enlightenment-nosound -f --stdin Cease and desist at all costs. I have just been informed on irc that a fixed menu is in incoming. So, it should all be fixed tomorrow. BEGIN GEEK CODE BLOCK Version: 3.12 GCS d- s: a-- c+++ UL P+ L !E W+ M o+ K- W--- !O M- !V PS-- PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z? -END GEEK CODE BLOCK- BEGIN PGP INFO Adam Heath [EMAIL PROTECTED]Finger Print | KeyID 67 01 42 93 CA 37 FB 1E63 C9 80 1D 08 CF 84 0A | DE656B05 PGP AD46 C888 F587 F8A3 A6DA 3261 8A2C 7DC2 8BD4 A489 | 716280FA GPG -END PGP INFO-
RE: apt-get update failing
On Sat, 6 Feb 1999, Jason Gunthorpe wrote: On Fri, 5 Feb 1999 [EMAIL PROTECTED] wrote: Yup. I've removed the Packages.gz files and had mirror reget them several times. I also tried right from the ftp site ... Are you -sure- you have the right version? Hehe. I logged in to irc, and noticed that talon was having this exact same prob. He said he got this error after upgrading to libc6 2.1-1. Downgrading to 2.0.7.19981211-2 fixed it. Adam
vote on slashdot.org: who is your favorite linux distro?
Re: Reinstalling required packages with dselect
First, fix your mailer so it wraps lines at about 75. On Sun, 24 May 1998, Jim Rogers wrote: I accidentally overwrote /bin/more (using instead of |), and dselect won't let me reinstall the package that contains it (util-linux) because it is already installed. I can't uninstall it because it is a required package. How did I get back MORE? Download more.deb yourself. Then dpkg -i it. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Reinstalling required packages with dselect
On Sun, 24 May 1998, Adam Heath wrote: First, fix your mailer so it wraps lines at about 75. On Sun, 24 May 1998, Jim Rogers wrote: I accidentally overwrote /bin/more (using instead of |), and dselect won't let me reinstall the package that contains it (util-linux) because it is already installed. I can't uninstall it because it is a required package. How did I get back MORE? Download more.deb yourself. Then dpkg -i it. Doh. I'm an idiot. I need more sleep. It should be util-linux.deb. Adam(who's face is red) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debianized Netscape for Debian 1.3.1 ?
On Sun, 5 Apr 1998, shaul wrote: Is there any Debianized Netscape for Debian 1.3.1 except for the probably obsolete (and also problematic) Netscape 3.01 (Navigator) ? Yes, I am still working on it. The new version of communicator4.05, and netscape4.05, allow for redistribution. They won't work on hamm, as they are compiled for libc6. There is nothing I can do to change that. Also, they won't be going into 2.0, unless decisions are made by someone other than myself. Please be patient. They will be in 2.1, and will be available on the ftp site in slink. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debianized Netscape for Debian 1.3.1 ?
On Mon, 6 Apr 1998, Tim Sailer wrote: Adam Heath wrote: On Sun, 5 Apr 1998, shaul wrote: Is there any Debianized Netscape for Debian 1.3.1 except for the probably obsolete (and also problematic) Netscape 3.01 (Navigator) ? Yes, I am still working on it. The new version of communicator4.05, and netscape4.05, allow for redistribution. They won't work on hamm, as they are compiled for libc6. There is nothing I can do to change that. Also, they Strange.. I thought hamm was libc6... Yes, you are correct. I meant to say that they won't work on 'bo.' On a related note, the navigator tarball and communicator tarball, are almost identical. Initial inspecation leads me to believe that there will be one -java pkg, and both navigator and communicator will depend on that. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: xdm, X, fvwm2 : newbie questions
On Wed, 25 Mar 1998, Bill Leach wrote: With a ps2 style mouse, it is either use gpm for both consoles and X or kill gpm to run X (for whatever reasons they will not co-exist on a ps2 style mouse). I've never seen the advantages of using /dev/gpmdata as I have never had any problems getting my mouse to work correctly in X11. YMMV though. I think not using the gpmdata method makes it simpler anyhow. Hogwash. I have a ps2 mouse. Gpm and X know nothing about eachother. They both access /dev/psaux fine. I use 2.1.91pre1. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Attn: Newbies Re: Random Broken Pipe
On Sun, 22 Mar 1998, Art Lemasters wrote: Some wonderful developer(s) fixed our manpage readers of the coming release (hamm) so that they will also scroll upwards to the beginning of any file we read. The manpages (accessed by typing man nameofcommand--without the quotes) can also be searched with the / foreslash command in the hamm version. Excellent work... This has nothing to do with man. With bo, the pager that man used was more, and that couldn't go up. Now, the default is less, which does as you describe. This could have been changed manually at any time. You didn't have to wait for hamm. This might not have been done in hamm's man, I am not sure. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: New handling of scripts external to PPPD (ie ip-up, ip-down)
On Mon, 16 Mar 1998, Adam Heath wrote: I have never liked the idea of having to edit either ip-up, or ip-down each time I want to add a script into the ppp chain. This is also hard for separate packages to do automatically, without errors. So here it comes. Sit down for this one. On my system there is *NO* ip-up, or ip-down. Pppd shells out to /etc/ppp/ppp-functions, with the first parameter being a runlevel, and the rest the normal parameters passed to ip-up. Then, ppp-functions sets up some variables, and runs /etc/ppp/rc runlevel. Rc is similiar to the rc that init uses, in that there are kill scripts and start scripts for each level. Also, pppd waits until the external script finishes, so make sure that all programs are either short in duration, or fork into the background. I am thinking of more places to add runlevels. Currently, I am trying to eliminate the connect option from pppd's option files. Ppp-functions would be called like this: /etc/ppp/ppp-functions dial hostname username ip-up parms, with stdin, stdout, and stderr redirected to the serial port. Ppp-functions could then use whatever means nescessary to dial. This would allow for dialing several numbers in round-robin fashion, better redial capabilities, enhanced logging, etc. Hints for the debian developer(don't do this yet, as it is not official) If a package wants to be run when pppd is on, then include a file /etc/ppp/scripts.d/packagename. In the postinst, run the command ppp-update.d. runlevel directory old method (relative to /etc/ppp/) init(just before modem init) init.d cnct(modem finished connectingcnct.d but before ip-layer starts) online(ip-layer is up)online.dip-up offline(ip-layer is down) offline.d ip-down discnct(modem is disconnected)discnct.d idle(pppd is in the holdtime) idle.d ppp-update.d defaults to online.d/S20 and offline.d/K20. Success! I now have no need of a connect option inside of pppd. Currently, pppd calls /etc/ppp-functions like this: /etc/ppp/ppp-functions[8360] /dev/ttyS1 862-4141 michnet [EMAIL PROTECTED] ppp-functions then uses a new runlevel, dial(dial.d) to dial out on the modem. Currently, a script, connector, build a connect script, and executes chat. Multiple phone numbers are supported, but currently they are hard-coded into the script. Redial takes place *outside* of pppd, so more sophisticated redial techniques can be used. I connect to the internet for this session with the connector portion commented out in pppd's source. It is only running /etc/ppp/ppp-functions. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
New handling of scripts external to PPPD (ie ip-up, ip-down)
I have never liked the idea of having to edit either ip-up, or ip-down each time I want to add a script into the ppp chain. This is also hard for separate packages to do automatically, without errors. So here it comes. Sit down for this one. On my system there is *NO* ip-up, or ip-down. Pppd shells out to /etc/ppp/ppp-functions, with the first parameter being a runlevel, and the rest the normal parameters passed to ip-up. Then, ppp-functions sets up some variables, and runs /etc/ppp/rc runlevel. Rc is similiar to the rc that init uses, in that there are kill scripts and start scripts for each level. Also, pppd waits until the external script finishes, so make sure that all programs are either short in duration, or fork into the background. I am thinking of more places to add runlevels. Currently, I am trying to eliminate the connect option from pppd's option files. Ppp-functions would be called like this: /etc/ppp/ppp-functions dial hostname username ip-up parms, with stdin, stdout, and stderr redirected to the serial port. Ppp-functions could then use whatever means nescessary to dial. This would allow for dialing several numbers in round-robin fashion, better redial capabilities, enhanced logging, etc. Hints for the debian developer(don't do this yet, as it is not official) If a package wants to be run when pppd is on, then include a file /etc/ppp/scripts.d/packagename. In the postinst, run the command ppp-update.d. runleveldirectory old method (relative to /etc/ppp/) init(just before modem init)init.d cnct(modem finished connecting cnct.d but before ip-layer starts) online(ip-layer is up) online.dip-up offline(ip-layer is down) offline.d ip-down discnct(modem is disconnected) discnct.d idle(pppd is in the holdtime) idle.d ppp-update.d defaults to online.d/S20 and offline.d/K20. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: New handling of scripts external to PPPD (ie ip-up, ip-down)
On 16 Mar 1998, Ben Pfaff wrote: If a package wants to be run when pppd is on, then include a file /etc/ppp/scripts.d/packagename. In the postinst, run the command ppp-update.d. And in the prerm, presumably? Yes, I assumed the developers would get that bit. I like the proposal, by the way. My only possible concern is that it is perhaps a little too complicated, but sometimes that is needed to buy extra flexibility. I know that. Maybe I could have it still exec ip-up and ip-down at the appropriate times. This is a list of scripts I have in /etc/ppp/scripts.d/ -rwx--x--x 1 root root 288 Mar 15 17:00 chime* -rwx--x--x 1 root root 120 Mar 16 09:53 cleanup* -rwx--x--x 1 root root 247 Mar 14 02:58 distributed-net* -rwxr-xr-x 1 root root 147 Mar 15 18:03 exim* -rwx--x--x 1 root root 105 Mar 16 02:22 fetchmail* -rwx--x--x 1 root root 258 Mar 14 02:59 logger* -rwx--x--x 1 root root 155 Mar 16 02:03 mirror* -rwx--x--x 1 root root 390 Mar 14 02:59 named.cache* -rwx--x--x 1 root root 243 Mar 14 03:00 ntpdate* -rwx--x--x 1 root root 376 Mar 14 03:00 pppstats* -rwx--x--x 1 root root 90 Mar 14 03:00 tcpdump* -rwx--x--x 1 root root 646 Mar 14 03:00 wget* tcpdump start on tty11. pppstats on tty10 mirror is on tty9, getting debian currently. fetchmail runs on tty8, in daemon mode, non-detach Everything else I will let you guess at. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: New handling of scripts external to PPPD (ie ip-up, ip-down)
On 17 Mar 1998, Martin Bialasinski wrote:
Adam Heath [EMAIL PROTECTED] writes:
On my system there is *NO* ip-up, or ip-down. Pppd shells out to
/etc/ppp/ppp-functions, with the first parameter being a runlevel, and the
rest the normal parameters passed to ip-up. Then, ppp-functions sets up
some
variables, and runs /etc/ppp/rc runlevel. Rc is similiar to the rc that
init uses, in that there are kill scripts and start scripts for each level.
There is already something like this, I believe.
$ ls -R /etc/ppp/ip-{up,down}.d
/etc/ppp/ip-down.d:
fetchmail* isdnutils
/etc/ppp/ip-up.d:
fetchmail* isdnutils
Don't know how (or if) your suggestion fits into this.
My suggestion(enhancement) is more refined. The current way doesn't allow
some programs to be run before others. Plus, I have more runlevels. This
would allow for an external dialer, instead of having to have the connect
option in pppd.
Please reread the original post, and think everything through carefully(not
meant as a flame).
Adam
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: lchown(), libc6-2.0.7pre1, and 2.1.8x kernels
On Wed, 11 Feb 1998, Steve Hsieh wrote: Doesn't appear to be good enough on its own... dpkg: error processing ssltelnet_0.11.1-2.deb (--install): error setting ownership of symlink `usr/man/man8/in.telnetd.8.gz': No such file or directory dpkg-deb: subprocess paste killed by signal (Broken pipe) ii libc6 2.0.7pre1-1The GNU C library version 2 (run-time files) The lchown() function shows up in the most recent libc6 package. Is this on its own enough for dpkg to work happily with the 2.1.81+ kernels? -- 2.1.86 swapped syscalls lchown and chown. So 2.1.86 is backward compatible. Please upgrade to 2.1.86 and all will work again. I just checked this on my system with a few packages that hadn't been able to install. They went fine. Adam -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: X Windows
On 11 Feb 1998, Kirk Hilliard wrote: If you have a ps2 mouse, run gpm with the -R repeater switch, and tell X to use ProtocolMouseSystems Device /dev/gpmdata in the Pointer section of your XF86Config so that you will still be able to use gpm when you visit the console for old time's sake. I don't have to do this. I run unstable(does it make a difference?), and both 2.0.33 and 2.1.86. When I switch to text mode, the mouse immediately works. When I go back to X, it sometimes takes a few seconds to start functioning(the mouse). Adam -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Checking mail from several ISPs
On Mon, 9 Feb 1998, Ian Perry wrote: Hi all, I have set up a Debian system to allow several computers to utilise the one internet modem line using IP masquerade, and would like to get linux to check several email accounts at different ISPs, and route them to a mailbox, for the relevant win95 desktop to pick them up automatically, at say 10 minute intervals, while online. There are several email addresses going to several desktops, but there are no repeated email addresses. e.g collected mail for [EMAIL PROTECTED] postbox for machine A collected mail for [EMAIL PROTECTED] postbox for machine A collected mail for [EMAIL PROTECTED] --- postbox for machine B Then the individual machines pick up their mail from the respective postboxes Linux is on a dial up connection so the IP never remains constant, so ppl sending email direct to the machine is not possible. Can anyone give me some ideas of where to start, and which packages may be of use ? I looked for some howto's but nothing really covers this. Somewhere in the ip-up chain, issue this command, fetchmail -d 300 --invisible and in ip-down, fetchmail -q Then in root's home dir, create .fetchmailrc: poll isp1.com proto pop3 user abc pass password is box1 here poll isp2.com proto pop3 user def pass password is box2 here poll isp3.com proto pop3 user ghi pass password is box3 here You should be able to figure out what needs to be changed for your setup. Obviously, you need fetchmail installed. I also recommend running exim, and not smail/sendmail. This way, you could have the masqueraded machines point their smtp mailer to linux, and have it the mail either delivered immediately if online, or have it queued if offline. Smail doesn't work in that instance, and I don't know sendmail. This is my setup here, except that I only have one user(me!). Adam -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Sending email(smail)
On 4 Feb 1998, Ben Pfaff wrote: I would like to have email that I send when offline be queued for later deliver when I connect. How can I do this? When mail is sent when you're offline, the mail daemon queues it, because it can't connect to the remote host. It retries every so often for a few days, then gives up. The easiest thing to do is to just let the daemon queue it, and then, when you're online, run the `runq' command to have it attempt delivery immediately. Of course there are other solutions but this is the simplest. If you use diald it won't work of course. Maybe the diald filters can be configured to not bring the connection up for packets sent to the smtp port? Sorry, but it doesn't. I use Pine, with the smtp host as localhost. Pine comes back as saying it can't be delivered. I believe this is because smail is trying to resolve with dns, can't, and gives up. Any other suggestions? Adam. ps: I had been using windows for email, but just recently converted to fetchmail/procmail and unix/pine. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Sending email(smail)
On 5 Feb 1998, Martin Bialasinski wrote: Adam Heath [EMAIL PROTECTED] writes: Sorry, but it doesn't. I use Pine, with the smtp host as localhost. Pine comes back as saying it can't be delivered. I believe this is because smail is trying to resolve with dns, can't, and gives up. What version of smail do you use? If you use X, make your xterm as wide as possible, try again and tell us the exact errormessage smail gives you. Does mailx -s testmail [EMAIL PROTECTED] testmail .. work? Yes, this does. Smail: 3.2.0.100-4 Pine: 3.96L-2 This is the error message I get: [Mail not sent. Sending error: 450 defer '[EMAIL PROTECTED]', sender] Let me state this again. Sending mail on the cmd line will queue mail for later delivery when not online. However, if I am in Pine, it(Pine) attempts to use the local smtp relay(localhost), and mail delivery fails(doesn't queue). Adam. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Sending email(smail) **FIXED**
On Wed, 4 Feb 1998, Adam Heath wrote: Sorry, but it doesn't. I use Pine, with the smtp host as localhost. Pine comes back as saying it can't be delivered. I believe this is because smail is trying to resolve with dns, can't, and gives up. What version of smail do you use? If you use X, make your xterm as wide as possible, try again and tell us the exact errormessage smail gives you. Does mailx -s testmail [EMAIL PROTECTED] testmail .. work? Yes, this does. Smail: 3.2.0.100-4 Pine: 3.96L-2 This is the error message I get: [Mail not sent. Sending error: 450 defer '[EMAIL PROTECTED]', sender] Let me state this again. Sending mail on the cmd line will queue mail for later delivery when not online. However, if I am in Pine, it(Pine) attempts to use the local smtp relay(localhost), and mail delivery fails(doesn't queue). I installed exim. This can be set to queue all mail destined for non-local hosts, no matter how it entered the system. Smail would queue mail generated with the mail command, but if done through smtp, it would try to resolve dns before queueing. When offline, dns would fail, and pine would error out. Online, however, dns would work and smail would put the mail into the outgoing queue. This is broken, in my mind. For smail to queue outgoing smtp mail received through smtp, you must be online. What is the point in that? Adam. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Backspace in X?
-BEGIN PGP SIGNED MESSAGE- How do I get backspace to work in X? I would like bspc/del to work in xterm/bash like bash on the console. Also, bspc doesn't function AT ALL in staroffice. I have no idea where to look for this. Txs in advance. Adam -BEGIN PGP SIGNATURE- Version: 2.6.3a Charset: noconv iQCVAwUBNNjnyGtDcajeZWsFAQGL3AP/TjzT74G8Ikl8+8rFWhORX1xEBqGk6vIG JDtsCpmG6Dj6Siyu1OlRChEWmyiUOFNkuvMFiVmmz+Y6T6fvUMH1PuIRDTj5+a4A kcIn7GIa65HgBaZhfrt/L9z1MWnCB+zSVEy/siJq07gF43RZDwYNIvwPV2mD9JEh Bnx5UlGJZJg= =9TA9 -END PGP SIGNATURE- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Sending email(smail)
I would like to have email that I send when offline be queued for later deliver when I connect. How can I do this? Adam -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Commands
On Sun, 25 Jan 1998, Matt Thompson wrote: Hi, y'all, Ok, I've been poring over the HOWTO's and man-pages and can't find two little commands THAT I USED TO KNOW! :) A friend told me about these a while back and now I can't remember or find them. One is the command executed by root on a kernel to make it look at a certain partition, allowing it to be used as a boot-floppy image. The other is the command that I use as mattyt so that, as root, I can run make xconfig. It allows root (or any other user, I suppose) access to the X-server, when you're not logged in to XDM as that user. RDEV, first, then either su, sudo, or secure-su. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Debian Pgp question...
| On Tuesday, 13 January 98, at 12:14:19 PM | Smorrill wrote about Debian Pgp question... I have used Pgp for quite awhile with Win95, and thus have already made my secret key id number, etc. I would like to get pgp set up with Linux as I would like to move to using Debian exclusively. Anybody out there familiar with this issue of how I would use my same 1028 bit key (created under Dos, MIT pgp...2.6.2) without having to revoke the existing key a create a new one? Mount the drive that contains your pgp directory in windows. In your home directory, link .pgp to the win95 pgp directory. On my machine, ln -s /dr_c/pgp ~/.pgp 2B OR NOT 2B=FF Adam Heath of Borg-Linux [EMAIL PROTECTED] Join the H.323 effort. Email http://www.debian.org - Get Your Own Linux! [EMAIL PROTECTED] with http://wwp.mirabilis.com/3375265 - Page Me the word subscribe in the body. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: New maintainer request
I am sorry for this. I didn't see the extra TO: field before I sent. I wish I had a life outside Quake. Adam Heath of Borg-Linux [EMAIL PROTECTED] Join the H.323 effort. Email http://www.debian.org - Get Your Own Linux! [EMAIL PROTECTED] with http://wwp.mirabilis.com/3375265 - Page Me the word subscribe in the body. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
New maintainer request
-BEGIN PGP SIGNED MESSAGE- My name is Adam Heath. I am currently attempting to add Linux support for the H.323 video conferencing protocol, both as masquerading, and a proxy. I have also rewritten redir, to support ip-aliasing, and to make it function as a load balancer. However, I have never released that program to the public. You can see from my PGP key that I have already had it signed by a current developer. Please sign me up for debian-private, et al, with [EMAIL PROTECTED] Preferred account name: doogie Type Bits/KeyIDDate User ID pub 1024/DE656B05 1997/12/26 Adam Heath [EMAIL PROTECTED] - -BEGIN PGP PUBLIC KEY BLOCK- Version: 2.6.3i mQCNAzSjA8gAAAEEAMDYM1B3Kkv2FDzxVM8AB1KcBwz08l58dVuk4ZQuV52qbuBD /SWSQIy17xA8NF72yxG2oc16dlmm/9itxKLTcsAI93rHj4eKZdRTEihJoAjS6seY lMJwzQVE4hHa63aRRWKuRQjxwPVbrhcyvW+rdU8iyd31nRywqmtDcajeZWsFAAUR tB9BZGFtIEhlYXRoIDxhZGFtLmhlYXRoQHVzYS5uZXQ+iQEVAwUQNKWVnrf2jhx5 fmQdAQG0vQgAmXVcoo+tgldE4v0pPeVjO5zE8/IOhbEbCVh+SAbl6LV+QFIlQNIL m8ulfG/3kIhr55o4XPJucCBwj9iE3NBYCLu09M3Hh2Oh9Jc86pn72+c3lb9oA2ao 5kfw78b+5JlNDdeB5fvrvfnfdMkwUy1Oncl2tAo8E+sjOeigRq6hysQM1PPgxH5W 3/jg9clL7KqCdOgm5eU/IbuFcDy56mRAErOaJiJSkc1Bvja0PXP1MvTUJ3NSLF7N 8WyYDMW3KkAlz8Mc+nUfa+ZVMa5FK5GNyWa4eolX4UnPEd3H+S5QSNJPIcJ9CoZu AUhNxIsY5a+eU60HY20/lfM67/zAJ9AoOYkAlQMFEDSjA8hrQ3Go3mVrBQEBvx4E AKEt9BZ8vfM8r/b4ugf09X57loSncquWr8VSPUZVB3e3kiPxHclHylMQkt15PbAR 1n6vE4R9ucLI+PuwyO+X2ez2j51HVbBu8dT+BRMvNeyocfH1Ztr6bq9HM6Wf0GvV p/9NoG4F7dyEBLigwJ+hAp3g84eXbjDfWzPN5BEAeL6t =9b+u - -END PGP PUBLIC KEY BLOCK- 2B OR NOT 2B=FF Adam Heath of Borg-Linux [EMAIL PROTECTED] Join the H.323 effort. Email http://www.debian.org - Get Your Own Linux! [EMAIL PROTECTED] with http://wwp.mirabilis.com/3375265 - Page Me the word subscribe in the body. -BEGIN PGP SIGNATURE- Version: 2.6.3i Charset: noconv iQCVAwUBNKYeGmtDcajeZWsFAQFlJwQAwL5h9yh9uycl+DJDENquKHeSXrARQ4Rm gHBjDBMAkQM2uL8vs6hAIy/txzaEVyO9HM9Lo8aiQExIKijkFQGYCKqxDKsHm2AE +uO6RcCT2qp2II8fCcb+rZeoWB7MWXgVWOYXcaRHX7ahzrzPrmA3LwRykXYxD/uW RGlxXHu19As= =4I4h -END PGP SIGNATURE- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Linux w/AWE 64 and Cyrix 6x86 CPU
-Original Message- From: Randy Edwards [EMAIL PROTECTED] To: debian-user@lists.debian.org debian-user@lists.debian.org Date: Monday, December 22, 1997 8:19 PM Subject: Linux w/AWE 64 and Cyrix 6x86 CPU I'm planning on setting up a Soundblaster AWE 64 under Linux. I also plan on using this in a Cyrix system and have heard that the AWE 64 and Cyrix CPUs don't get along well. What I was wondering is if anyone has any words of advice before I start playing with this combination? Thanks in advance. This is only for windows systems. On the AWE64, it only has 32-hardware voices. The upper 32 are done in software. This software checks for a pentium on installation, and that is why cyrix chips don't supposedly work. In linux, that is not a problem, however. I ran a Cyrix-120(ov 133) with an AWE64 with no problems. Adam Heath of Borg-Linux [EMAIL PROTECTED] Join the H323 effort. Email http://www.debian.org - Get Your Own Linux! [EMAIL PROTECTED] with http://wwp.mirabilis.com/3375265 - Page Me the word subscribe in the body. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Skipping to next header file
-Original Message- From: Aaron Walker [EMAIL PROTECTED] To: Debian User List debian-user@lists.debian.org Date: Friday, December 19, 1997 9:44 AM Subject: tar: Skipping to next header file I am trying to unpack the newest kernel source in /usr/src. It goes along fine until it stops and says: tar: Skipping to next header file gzip: linux-2.0.33.tgz: invalid compressed data---crc error tar: Archive - EOF not on block boundary I used the following commands to try to uncompress the kernel source: gzip -dc linux-2.0.33.tgz | tar xvB tar xpvf linux-2.0.33.tgz They both produced the same error. Any ideas? Please help. Thanks. -- ` I use ftp.kernel.org. The cmd zcat filename|tar xv works fine. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: dpkg and deb files
-Original Message- From: Rick [EMAIL PROTECTED] To: debian-user@lists.debian.org debian-user@lists.debian.org Date: Friday, December 19, 1997 6:01 AM Subject: dpkg and deb files HI (Me again) I am still having major problems with dpkg and deb files, as suggested i have re downloaded the deb files several times, both with cute ftp (in binary mode) and msie, yet when ever i try to install them with dpkg i always get the same error messages. unexpected eof, something about the archive not being in gzip format, if fact the same messagaes you get when tring to install anyole file (ie not a deb). I have getting nowwhere at all. I have tried installing from both mounted vfat and msdos partitions, and copying the files to /tmp, but still no joy at all, I have checked the packages file and have not missed any dependcys that i know of (trying to get manpages to work, so i have, man-db, manpages, groff, and a lib required by groff) and nothing will install at all. - *HELP!* Rick I can't remember for sure, but are you mounting with the option binary? Maybe the vfat/msdos filesystem is translating the file on the fly, because it thinks it is a text file. Something like mount -t msdos -o binary,rw /dev/hda2 /dos Adam Heath of Borg-Linux [EMAIL PROTECTED] Join the H.323 effort. Email http://www.debian.org - Get Your Own Linux! [EMAIL PROTECTED] with http://wwp.mirabilis.com/3375265 - Page Me the word subscribe in the body. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Skipping to next header file
-Original Message- From: Aaron Walker [EMAIL PROTECTED] To: Debian User List debian-user@lists.debian.org Date: Friday, December 19, 1997 9:44 AM Subject: tar: Skipping to next header file I am trying to unpack the newest kernel source in /usr/src. It goes along fine until it stops and says: tar: Skipping to next header file gzip: linux-2.0.33.tgz: invalid compressed data---crc error tar: Archive - EOF not on block boundary I used the following commands to try to uncompress the kernel source: gzip -dc linux-2.0.33.tgz | tar xvB tar xpvf linux-2.0.33.tgz They both produced the same error. Any ideas? Please help. Thanks. -- ` I use ftp.kernel.org. The cmd zcat filename|tar xv works fine. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Partitioning
-Original Message- From: Gertjan Klein [EMAIL PROTECTED] To: debian-user@lists.debian.org debian-user@lists.debian.org Date: Friday, December 12, 1997 6:39 PM Subject: Re: Partitioning I doubt if you've spend as much time partitioning as I have developing my boot manager ;-) I have very little experience with NT, but I know for a fact that Windows '95 does _not_ require to be installed to the first primary partition of a harddisk - on my harddisk, it is installed on the fourth. The requirements are: - It must be installed on the first harddisk. This requirement goes for both DOS and W95; for DOS 6.22 and 7.0 (the DOS part of W95) there is a workaround if there are no primary DOS partitions on the first harddisk. - The active flag must be set to the booted partition. If W95 is booted from the fourth partition, but the active flag is set to the first, it will hang. Not true. Look below. All cut and pasted from Win95. Disk Drv Mbytes Free Usage 1 2012100% C: 500 30 --Linux Swap 16 --Linux Root E: 500 F: 500 469 \ 2 1211100% | D: 500 + These are mirrored(raid0) G: 203 | to form everything else. 494 ---+ 3203 --100%-/ Current fixed disk drive: 1 Partition Status TypeVolume Label Mbytes System Usage C: 1 PRI DOS 500 FAT16 25% 2 AEXT DOS 1514 75% Current fixed disk drive: 2 Partition Status TypeVolume Label Mbytes System Usage D: 1 PRI DOS 500 FAT16 41% 2 Non-DOS 161% 3 EXT DOS 697 58% Current fixed disk drive: 3 Partition Status TypeVolume Label Mbytes System Usage 1 Non-DOS 203 100% My setup has the extended partition on the first drive active, which loads LILO, and allows be to boot Win95, one of several Linux kernels, or(and this is special) boot up off floppy. Yes, I type in floppy at LILO: , and it started to read A:. That way, I can turn off booting to disk A in the bios, but still have the capability. BTW, I use Win95 mostly for games. It is also my mail store(75+megs currently), and I don't know of a program that can read Outlook Express mail files. Adam Heath of Borg-Linux [EMAIL PROTECTED] Join the H323 effort. Email http://www.debian.org - Get Your Own Linux! [EMAIL PROTECTED] with http://wwp.mirabilis.com/3375265 - Page Me the word subscribe in the body. Windows 95: -- 32-bit extensions and a graphical Windows v. Linux is shell for a 16-bit patch to an 8-bit operating a no-win situation. originally coded for a 4-bit micropro- cessor written by a 2-bit company that It is nearly impossible to can't stand for 1 bit of competition.look at a penguin and get angry. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: alt-F7 reassignment
I have no idea how you would map alt-esc to switch back to X. However, you can cycle through the tty's by using alt-right_arrow and alt-left_arrow. So, the key sequence alt-f1/alt-left_arrow (or I suppose alt-f12/alt-right_arrow) would get you back to X. This begs the question: why would you need 12 login consoles? =) hope that helps... The description is not alt-Fx, buit Lalt-Fx, where x=1-12, and stands for consoles 1 thru 12. Ralt-Fx, where x=1-12, stands for consoles 13-24. Lalt=Left Alt, Ralt=Right Alt. Once, as a test, I had, upon bootup, ten dosemu sessions running, on consoles 13-22, with each one being a dos bbs acting as a node for incoming telnet. Adam Heath of Borg-Linux [EMAIL PROTECTED] Join the H323 effort. Email http://www.debian.org - Get Your Own Linux! [EMAIL PROTECTED] with http://wwp.mirabilis.com/3375265 - Page Me the word subscribe in the body. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: crash at partition check during boot
-Original Message- From: Joel N. Weber II [EMAIL PROTECTED] To: debian-user@lists.debian.org debian-user@lists.debian.org Date: Friday, November 21, 1997 2:37 PM Subject: crash at partition check during boot I am attempting to get a debian rescue disk to boot. The disk seems to boot fine when tried on another machine. I've also tried writing the same image to a different floppy, and that does not change the error. The machine crashes when after saying `partition check' and then `hda:'. If I specify `hda=noprobe', the machine boots fine but does not see the first hard drive. (It does seem to see hdb.) hda has been repartitioned in DOS fdisk and OS fdisk and then low-level formatted; none of these things seem to prevent linux from crashing. This is the standard boot disk image downloaded a few weeks ago; the kernel version is 2.0.29-7. I suppose I could continue using hda=noprobe, but I'd hate to lose the 500 MB had and have to use only the 300 MB hdb. Do you have any suggestions as to what I could try to fix this problem? Remove hdb. Remove hda and move hdb to hda. Swap hda and hdb. Try hda in another machine. Boot dos(ick!), run FDISK, then run FDISK /MBR. Adam Heath of Borg-Linux [EMAIL PROTECTED] Join the H323 effort. Email http://www.debian.org - Get Your Own Linux! [EMAIL PROTECTED] with http://wwp.mirabilis.com/3375265 - Page Me the word subscribe in the body. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
New Mailing List: Development of H.323 video conferencing
This list is for those who want to develop, and see developed, this protocol, both for masquerading, and as a proxy. Send an email to [EMAIL PROTECTED] with the word subscribe in the body. Adam Heath, of Borg-Linux [EMAIL PROTECTED] http://wwp.mirabilis.com/3375265 -- Page me -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: /etc/init.d/network
-Original Message- From: [EMAIL PROTECTED] [EMAIL PROTECTED] To: debian-user@lists.debian.org debian-user@lists.debian.org Date: Thursday, November 13, 1997 7:19 AM Subject: /etc/init.d/network Howdy all. I've got one machine that's really got me stumped. I've been with Debian since before 0.91 was released - I've installed Debian on several systems, many times. Yet this one's tough... The systems is running on an AMI Titan III M/B w/ 32MB RAM. Dual Pentium Processors (so I guess they both will choke on that opcode...:) two quantum 3.2GB SCSI drives plus the usual complement of Floppy/CD-ROM/ tape drive. I've had two network cards in this machine: first, an HP 100VG PCI and then an Eagle NE2000 ISA. Both exhibit the same problem. Which is... When the system boots /etc/init.d/network is run (I know this is true because I added some echo statements watched the screen while it booted.) Anyway, after I boot the eth0 interface is not setup. If I run /etc/init.d/ network by hand afterwards, everything is fine. I've checked, double- checked, and even had a friend check that there were no hardware conflicts. Interrupts and IO addresses are all fine. I've tried compiling the drivers into the kernel and as loadable modules - nothing seems to help this problem. Does anyone have any suggestions? I can provide several K bytes worth of files if someone wants to take a look at them. So far everything's fairly straight out of the box on this system. Running the Official Debian CD release 1.3.1, though I've recently started migrating to the libc6 tree. Comments and suggestions welcome. Thanks. Chuck I had this problem for a while. What fixed it was removing dhcpc. IE DHCP Client. Eth0 is brought up. Later during bootup, dhcpc tries to find an ip address from a dhcp server, can't find one, and sets the address to 0.0.0.0. At least that is what fixed it on my end. Adam Heath [EMAIL PROTECTED] http://wwp.mirabilis.com/3375265 -- Page me -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: dmsdosfs
-Original Message- From: Luka Pravica [EMAIL PROTECTED] To: debian-user@lists.debian.org debian-user@lists.debian.org Date: Tuesday, November 11, 1997 11:22 PM Subject: dmsdosfs Hi, I tried to use dmsdosfs-0.8.3.0 to access doublespace 3 compressed drive on my widows partition. I followed all instructions, recompiled kernel... But when I try add module to the kernel with insmod dmsdos, I get following error messages: /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_read_super /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_rename /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_unlink /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_put_super /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_rmdir /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_create /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_lookup /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_read_inode /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_mkdir /lib/modules/2.0.30/fs/dmsdos.o: unresolved symbol msdos_unlink_umsdos During compilation of kernel, there were few warnings about dmsdosfs, but no errors. I use kernel 2.0.30. Does anybody knows what I have done wrong? Thanks for any help Luka Compile support for MSDOS fs. And be sure to insmod msdosfs(sp?) after bootup. I have noticed that not all modules that require MSDOS properly contain MSDOS dependency information. Adam Heath [EMAIL PROTECTED] http://wwp.mirabilis.com/3375265 -- Page me -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
IP-MASQ: No longer need to be done as modules
Now, FTP, QUAKE, IRC, CUSEEME, RAUDIO, and VDOLIVE can be compiled directly into the kernel! For routers, this should eliminate the need for insmod, kerneld, and friends. It should be possible to eliminate the need for kerneld support on such setups. Each helper routine can be selectively compiled into the kernel or as a module, or not compiled at all. If compiled into the kernel, then the ports it attaches to are specified on the kernel command line as ip_masq=helper-app,port,port,port, If as a module, then the old way with nsmod helper-app ports=port,port,port,... would work. Attached are patches for 2.0.30 and 2.0.31. Please test this out. I would hope that this could be included into the mainstream kernel. I haven't tried getting it to work with 2.1.x, but will work on that most likely this weekend. Have fun! Adam Heath [EMAIL PROTECTED] http://wwp.mirabilis.com/3375265 -- Page me ps: What is Linus' email address so that I can send this to him? Email me private to keep it off the list with this info. Thanks. ip-masq-patch.gz Description: VRML model
Re: Linux, FAT32 and a mighty odd MBR.
-Original Message- From: Dale Harrison [EMAIL PROTECTED] To: debian-user@lists.debian.org debian-user@lists.debian.org Date: Wednesday, October 29, 1997 9:26 PM Subject: Linux, FAT32 and a mighty odd MBR. Hello. I had a weird problem last night, installing Debian 1.3* on a friends PC. For what it's worth, it's a PentiumII/266, some hideous amount of memory, some hideous amount of HD space. Aside from the SuperMicro motherboard bug [Which has been reported, I'm fairly sure..], you know, the one where the bootable cd sits there at the LDLINUX.SYS line.., everything went fairly smoothly. Until it came to LILO. Installed LILO, booted Linux fine. Added the lines to boot Win95, it took 2 goes for it to install the new record [it never complained the first time about failing]. Problem is, it simply doesn't boot Win95. It says booting Win95, then returns to the LILO prompt. Ok, a bit of a hassle. He needs his Win95 stuff more than Linux, so I'll just remove LILO and figure out something else later. lilo -u, lilo -U both complain that theres no LILO boot signature on /dev/hda. Hence nothing is removed. Hmmm. Boot via Win95 rescue disk and fdisk /mbr. Reboot and LILO's STILL there. Tried this more times than I can recall.. LILO won't die. System Commander reported some wacky things about the MBR, saying things were pretty well screwed with it. The problem remains, I can't seem to overwrite the MBR, not with LILO, not with fdisk. I found out that the Windows partition was FAT32, but surely LILO can handle this? [does it make much of a difference with the MBR?] So, I guess I'm throwing this out to you guys, is this a Windows problem? A Debian problem? A LILO problem? A FAT32 problem? Any suggestions for fixing? Thanks, D. Using this setup /dev/hda1 Win95 /dev/hda2 Linux /dev/hda3 Linux Swap /dev/hda Generic MBR, not Lilo /dev/hda1 Win95 Boot sector /dev/hda2 LILO /dev/hda2 is set as active with FDISK. Bios reads generic MBR off of /dev/hda. MBR reads the first sector off the active partition, ie /dev/hda2. LILO is now active. From here, it either reads the linux kernel, or some other, ie chained, partition. Solution: Use FDISK to change active flag. I don't know why you can't boot win95 from lilo's prompt. Could we see your /etc/lilo.conf file? And a copy of fdisk's output? (echo /etc/lilo.conf;cat /etc/lilo.conf;echo -n;echo -e p\nq\n|fdisk)|email.txt Adam Heath [EMAIL PROTECTED] http://wwp.mirabilis.com/3375265 -- Page me -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: smail POP error
-Original Message- From: Adam Heath [EMAIL PROTECTED] To: Debian User debian-user@lists.debian.org; Luis Francisco Gonzalez [EMAIL PROTECTED] Date: Friday, October 24, 1997 4:11 PM Subject: Re: smail POP error I can't connect using POP to retrieve my mail at a remote server I maintain. Recently, the hard drive filled up, and a new one was installed. Now I can't retrieve my mail. A copy of the log follows. Just a blind guess. What are the /tmp permissions? This is how it is in mine: drwxrwxrwt 15 root root 2048 oct 24 13:26 tmp I seem to recall that there is something special that allows everybody to create files in that directory but only delete those that belong to the current user. Check the policy manual. Luis. drwxrwxrwt 2 root root 1024 Oct 24 06:47 /tmp/ drwxrwxrwt 2 root root 1024 Oct 24 06:42 /var/tmp/ Any more ideas? I am stumped on this one. The command I used to copy the hard drives was: (cd nmnt;tar c *)|(cd mnt;tar xv) Was there anything wrong with that? I solved this by uninstalling and reinstalling QPOPPER. However, this still doesn't explain why it happened in the first place. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
smail POP error
I can't connect using POP to retrieve my mail at a remote server I maintain. Recently, the hard drive filled up, and a new one was installed. Now I can't retrieve my mail. A copy of the log follows. Fri Oct 24 07:31:46 [EMAIL PROTECTED] telnet stjohnscc.com 110 Trying 207.74.100.62... Connected to stjohnscc.com. Escape character is '^]'. +OK QPOP (version 2.2) at stjohnscc.com starting. [EMAIL PROTECTED] user adam +OK Password required for adam. pass -ERR System error, can't open temporary file, do you own it? +OK Pop server at stjohnscc.com signing off. Connection closed by foreign host. Also, when I log in and try to send to my account from root, it never arrives. I also can send from this account. A copy of the /var/spool/smail/msglog/* file follows. Xfail: adam parent: root reason: (ERR133) transport local: failed to open ou tput file: No such file or directory Xsucceed: jason parent: root -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: smail POP error
I can't connect using POP to retrieve my mail at a remote server I maintain. Recently, the hard drive filled up, and a new one was installed. Now I can't retrieve my mail. A copy of the log follows. Just a blind guess. What are the /tmp permissions? This is how it is in mine: drwxrwxrwt 15 root root 2048 oct 24 13:26 tmp I seem to recall that there is something special that allows everybody to create files in that directory but only delete those that belong to the current user. Check the policy manual. Luis. drwxrwxrwt 2 root root 1024 Oct 24 06:47 /tmp/ drwxrwxrwt 2 root root 1024 Oct 24 06:42 /var/tmp/ Any more ideas? I am stumped on this one. The command I used to copy the hard drives was: (cd nmnt;tar c *)|(cd mnt;tar xv) Was there anything wrong with that? -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
DOJ charges Microsoft violations
http://www.news.com/News/Item/0,4,15418,00.html DOJ charges Microsoft violations.url Description: Binary data
Re: Cyrix kernel patch
On Mon, 6 Oct 1997, Shaleh wrote: Is there a patch against 2.0.29 for cyrix chips?? Is anyone using it?? I tried the patch for 2.0.30 kernels (on a 2.0.30 kernel, of course) and it locked up a few times ... so if anyone has had better luck with this patch, please let me know. At this point I'm considering downgrading to kernel 2.0.27, as I'm told a lot of 2.0.30 is broke anyway. Will Where could I get the kernel patch? *** ** Good Idea: Feeding Stray Cats in the Park. Bad Idea: Feeding Stray Cats in the park ... to a bear. *** ** I'm glad someone else likes animaniacs. Adam Heath [EMAIL PROTECTED] Web Pager: http://wwp.mirabilis.com/3375265; Email Pager: mailto:[EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Setting up X
You probably want to try 'gpm -k' before starting X. I understand that gpm interferes with the operation of ps/2 mice under X. I have no idea why this problem should be confined to ps/2 mice. Good luck. I'm new to Linux/Debian. Everything was going well until it came time to set up X windows. The trouble is I can't get X to install my mouse. I know that it's a ps/2 mouse and that there's something called gpm also running. I don't know what this is. gpm says that my mouse is on /dev/ttyS0. Is this correct? In the debian installation, I did choose to install the mouse module. I have no problem running gpm and X with my ps2 mouse. I even switch back and forth. I run 2.0.30. Sometimes, however, when I switch back to X, the mouse doesn't function right away. After about a second delay, it comes back, however. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Starting out...
Overall, I'm very satisfied with the upgrade so far. Note that I upgraded the whole system unit in one shot (CPU/motherboard/memory). Also note that if your motherboard comes with an USB port, make sure you can disable it through the BIOS setup program ; Linux seems to dislike USBs. I have USB on my board. Linux says Unknown PCI device, but it stills boots and runs fine.2.0.30. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Sound Blaster AWE32 and Sound Blaster AWE 64 Gold
Hmm ... I have a SB16 and midi works fine without isapnptools, maybe that's the difference. You probably have a waveblaster extension mounted to i/o 330 using the MPU 401 interface. The SB 32 has a build-in (non MPU 401 compatible) Emu8000 waveblaster that need to be initialized to work correctly. I tried pnp extension for the kernel too and the whole sound card stoped working. Now I have the normal kernel with the awe patches installed and use isapnp to initialize all devices on the sound card. BTW: The normal SB 16 compatible sound works without any special initilization, only the midi part needs extra handling to work correctly. Torsten -- What do you use for the isapnp config file? I tried it, but it could never initialize the MIDI portions. I have to boot dos and use loadlin to get it to work. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Test to see if smail works
Ignore me please. Txs. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: mirror up-to-dateness report
I am experimenting with a script that audits the up-to-dateness of Debian FTP mirror systems. Here's my report so far: These 27 of our 42 listed mirrors are up-to-date. Some may be missing from this list because they are more than 24 hours behind the master copy of the Debian FTP archive, or they don't allow the PASV FTP command (I'll fix my program), or I couldn't contact them this evening, or the root of the Debian archive isn't what is listed in README.mirrors. This report will become more reliable over the next week or so, and will become a regular feature on our web site. Great idea! What would you think about a script that checks the connectivity to a given set of closest mirrors based on the Timezone or so and reports the fastest mirror available? Nice idea, but it shouldn't be something that is run on the debian ftp server. It should be run by all us users trying to find a close, fast mirror. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Kernel assumes 486
Hello, I recently installed Debian GNU/Linux 1.1.13 to a Compaq 386. The base kernel (2.0.6) runs fine, but if I let dselect attempt to install any other kernel, the system hangs on boot with a message like: a 486 is assumed. Giving up. Sounds like a buggy 386. When the kernel starts, it trys to find out what type of cpu it is running on. It starts by assuming a 386, then tests for 486. If that passes, then it tests for a 586, etc. Your 386, at startup, looks like a 486 to the kernel, and so later when it treats it as one it has errors. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
SMAIL and remote domains
I have a BBS that runs in dos but can access POP3 and SMTP mail. I would like this BBS to receive email as a sub-domain of my server. IE. wormhole.hackers-net.com# The BBS. # adam.hackers-net.com# My linux machine However, the BBS is not connected all the time. I need to have 'adam' receive the mail, but hold it in a user's mail box(wormhole) for transfer when the BBS connects. Can anybody help? If this is not the place, then could you tell me what newsgroups might help in this area? Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ ps. The BBS is MS-DOS based, running 2 lines with DOSEMU, and incoming AND outgoing 2 line telnet, with support for RIPTERM, ZMODEM, whatever. The BBS thinks it is talking to a modem the entire time(through a FOSSIL). -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
http://www.sophist.demon.co.uk/ping/
Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ The Ping o' Death Page.url Description: The Ping o' Death Page (Internet Shortcut)
/etc/hosts and BIND autoconversion
A while back I saw mentioned in a message about some utility that would read the '/etc/hosts' file and convert it into the BIND database. I would like to have this program. I have a DOS-based tcp program that can't use a hosts file, but can use BIND. I have BIND running locally, but I need the to have the DOS program be able to access the other systems by name. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
fork()ing and SIGKILL and SIGSTOP
I am writing a program that listens on several ports, using fork() to start a separate thread for each port. I want to be able to kill each of the child processes when the master parent terminates. I do not know how to implement this. I have tried signal(), and wait(), and neither will allow trapping of SIGKILL and SIGSTOP. I have registered a function with atexit(), and that is not run either. I know this can be done, because when PPPD is started, and I issue the command kill PPPD pid, PPPD runs the disconnect script before quitting. Any help would be appreciated(sp?). Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ ps: This is for a port redirection program. It supports multiple desitination addresses, and multiple listening ports, all from one command line and one executable(albeit heavly multithreaded!). -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: fork()ing and SIGKILL and SIGSTOP
On Dec 2, [EMAIL PROTECTED] (Adam Heath) wrote: I am writing a program that listens on several ports, using fork() to start a separate thread for each port. I want to be able to kill each of the child processes when the master parent terminates. I do not know how to implement this. I have tried signal(), and wait(), and neither will allow trapping of SIGKILL and SIGSTOP. I have registered a function with atexit(), and that is not run either. Neither SIGKILL or SIGSTOP can be caught -- otherwise you could write a program that couldn't be stopped without rebooting. I know this can be done, because when PPPD is started, and I issue the command kill PPPD pid, PPPD runs the disconnect script before quitting. kill pid sends SIGTERM, which is probably what you are looking for. The general concept would be for the master to keep track of all of the child pids, and then when receiving SIGTERM, kill the children (kill()), and then wait for them to terminate (wait() or waitpid()). Steve Greenland After I sent this, I got the bright idea to type man kill and discovered this. Everything is now working the way I wanted it to. Now, I am going to make it created a PID file for start-stop-daemon. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Virtual Terminals Greater Than tty8.
I just did mknod to create tty9 through tty12. Everything seems to be working fine, but I thought I would just check to make sure those vt's weren't disabled for a reason. Thanks Paul You can even go to tty24! To switch, hit LEFT ALT-(F1 - F12) for the first 12, RIGHT ALT-(F1-F12) for the upper 12. I have two dos sessions started automatically on tty23 and tty24 for a DOS based BBS telnet. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- This message was delayed because the list mail delivery agent was down.
Re: IPFWADM and telnet
At 00:41 22-11-96 -0500, Adam Heath wrote: |I current connect to the Internet with PPPD, and all works well. I have |compiled my kernel (2.0.24) with IP_MASQ enabled, and have verified that |this works. Here is my question. | |Can it be setup, so that when someone telnets into the Linux box, instead |of the request being handled there, it is masqueraded to one of the private |computers on the LAN? I have a DOS-based BBS that I can setup to accept |incoming telnet, but it won't work through DOSEMU. | It's possible. Have a look around for the 'redir' package, which will do it for you. Not what I want. I have looked at 'redir'. I would like it to be implemented more at the kernel level. Currently, I can run this command: ipfwadm -I -P tcp -a accept -S 0/0 -D 127.0.0.1/32 23 -r 1000 That will take packet coming from anywhere, headed toward the local telnet port, and redirect it toward the local 1000 port. Why can't I also put in a destination system address? I currently setup 'redir' to listen on port 1000, and redirect it across the LAN(actually, across a simulated SLIP link between linux and DOSEMU), but it seems I should be able to do something like it with the kernel. If I need to talk to the kernel guys, then if someone could direct me to the proper mailing list, I will post my question there. Txs. To those who are interested. I have finally got this to work. I hacked the 'redir' source to cycle back and forth between to ip addresses that are hard-coded into the source as each connection comes in. Then I setup a simulated SLIP connection bewteen linux and two dosemu sessions(read the Stacker mini howto). In each dos session, I used 'rlfossil' to translate between the fossil and telnet specs, and ran a dos-based bbs. I then telnetted to my system on port 1000(where 'redir' was set to listen). I got a connect on line 1 from the bbs. I aborted the connect. I then immediately telnetted again to port 1000. This time, it connected on line 2! I about sh*t myself when I saw it! Now I just need to get it to load automatically. I have a script created to run the programs on the linux side. I just need to make a script that can load the dos sessions automatically. If you think this is a lot to do on one machine, but the way I hacked the 'redir' program, it will work with many machines on a private lan. As each telnet connection comes in to the linux machine, it can be redirected to any number of IP address. If the kernel has a route to the address, then it will work. So I could have the 2 bbs channels residing on another computer on the private lan, 1 channel residing in DOSEMU on linux, and even another channel residing on a distant machine on the Internet! I am planning on making a HOWTO out of this, also, to make it easier for others to implement this. (I wonder if it would work with other protocols [FTP, HTTP, whatever]) Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: IPFWADM and telnet
At 00:41 22-11-96 -0500, Adam Heath wrote: |I current connect to the Internet with PPPD, and all works well. I have |compiled my kernel (2.0.24) with IP_MASQ enabled, and have verified that |this works. Here is my question. | |Can it be setup, so that when someone telnets into the Linux box, instead |of the request being handled there, it is masqueraded to one of the private |computers on the LAN? I have a DOS-based BBS that I can setup to accept |incoming telnet, but it won't work through DOSEMU. | It's possible. Have a look around for the 'redir' package, which will do it for you. Not what I want. I have looked at 'redir'. I would like it to be implemented more at the kernel level. Currently, I can run this command: ipfwadm -I -P tcp -a accept -S 0/0 -D 127.0.0.1/32 23 -r 1000 That will take packet coming from anywhere, headed toward the local telnet port, and redirect it toward the local 1000 port. Why can't I also put in a destination system address? I currently setup 'redir' to listen on port 1000, and redirect it across the LAN(actually, across a simulated SLIP link between linux and DOSEMU), but it seems I should be able to do something like it with the kernel. If I need to talk to the kernel guys, then if someone could direct me to the proper mailing list, I will post my question there. Txs. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Fw: IPFWADM and telnet
I current connect to the Internet with PPPD, and all works well. I have compiled my kernel (2.0.24) with IP_MASQ enabled, and have verified that this works. Here is my question. Can it be setup, so that when someone telnets into the Linux box, instead of the request being handled there, it is masqueraded to one of the private computers on the LAN? I have a DOS-based BBS that I can setup to accept incoming telnet, but it won't work through DOSEMU. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
IPFWADM and telnet
I current connect to the Internet with PPPD, and all works well. I have compiled my kernel (2.0.24) with IP_MASQ enabled, and have verified that this works. Here is my question. Can it be setup, so that when someone telnets into the Linux box, instead of the request being handled there, it is masqueraded to one of the private computers on the LAN? I have a DOS-based BBS that I can setup to accept incoming telnet, but it won't work through DOSEMU. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: kerneld and pppd and /sbin/request-route
-- On Sat, 17 Nov 2096, Adam Heath wrote: I have set kerneld to automatically run pppd through /sbin/request-route whenever I need the internet. The only problem is that I still get a timeout(or I just need to re-request the connection) from the kernel. I run FTP, then PPPD dials and connects, but FTP has timed-out. I get the same problem with LYNX and NETSCAPE. Can anyone help? You should use the DIALD package. It will do what you need. Its in Debian form. You will need to modify your setup a bit after you install it though. (Its a bit of a pain to get working, as I found out a few months ago..) but it works really great. I did do this. It still timed out. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
kerneld and pppd and ip masquerading
My friend owns to computers, one running linux, the other with win95. I setup /sbin/request-route to run pppd whenever he needs the internet from w95, and it works ok. Is there a way that he could run some kind of batch file or script on the Win95 machine to comunicate to linux to hang up the connection? He doesn't want to open a telnet connection, and kill PPPD himself. It would be better if he could do it all with one click. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- This message was distributed manually by [EMAIL PROTECTED] after the list initially failed to distribute it.
IPFWADM and telnet
I current connect to the Internet with PPPD, and all works well. I have compiled my kernel (2.0.24) with IP_MASQ enabled, and have verified that this works. Here is my question. Can it be setup, so that when someone telnets into the Linux box, instead of the request being handled there, it is masqueraded to one of the private computers on the LAN? I have a DOS-based BBS that I can setup to accept incoming telnet, but it won't work through DOSEMU. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- This message was distributed manually by [EMAIL PROTECTED] after the list initially failed to distribute it.
kerneld and pppd and /sbin/request-route
I have set kerneld to automatically run pppd through /sbin/request-route whenever I need the internet. The only problem is that I still get a timeout(or I just need to re-request the connection) from the kernel. I run FTP, then PPPD dials and connects, but FTP has timed-out. I get the same problem with LYNX and NETSCAPE. Can anyone help? Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- This message was distributed manually by [EMAIL PROTECTED] after the list initially failed to distribute it.
Re: DOSEmulator for linux
-- From: Paul Seelig [EMAIL PROTECTED] To: [EMAIL PROTECTED] Cc: debian-user@lists.debian.org Subject: Re: DOSEmulator for linux Date: Tuesday, November 12, 1996 7:44 PM On Tue, 12 Nov 1996 [EMAIL PROTECTED] wrote: I was just wondering if any of you out there had installed this on your linux boxes. Yes, works wonderful here! Compile Begins make[1]: *** No rule to make target '/usr/Makefile', needed by 'include/kversion.h'. Stop. make[1]: Leaving directory '/root/dosemu/dosemu-0.64.0/src' Make: *** [default] Error 2 That's it...no matter what I do, I can't get it to compile *shrug* Do you have the kernel sources installed on your system? The compilation of DOSEMU depends upon their header files. Regards, P. *8^) I have this same problem to. And I installed kernel-headers 2.0.6, kernel-source 2.0.6, then I manually untarred the kernel source for 2.0.24. And I still get the problem. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Moving files to a new partition. Please help.
Thank you all for your help, but the command I ended up using was this: cd /mnt find ./ | grep -v ./mnt | grep -v ./proc | cpio -p mkdir proc mkdir mnt # 'find ./' find ALL files, starting at the root directory. # 'grep -v ./mnt' Lets all lines pass through that DON'T contain './mnt', ie the new hard drive. # 'grep -v ./proc' Lets all lines pass through that DON'T contain './proc', to stop a nasty neverending loop. # 'cpio -p' Each line input to this command is open for input as a file, and then written to the current directory. I then shutdown, swapped the harddrives on the cables, and booted off floppy. Reran LILO, and all was well. Setup before: /dev/hda = 420 MB HD mounted on / /dev/hdc = 540 MB HD mounted on /mnt Using LILO to boot off /dev/hda Setup after: /dev/hda = 540 MB HD mounted on / /dev/hdc = 420 MB HD not mounted Using LILO to boot off /dev/hda Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- From: Adam Heath [EMAIL PROTECTED] To: Debian User List debian-user@lists.debian.org Cc: Jason Kusnier [EMAIL PROTECTED] Subject: Moving files to a new partition. Please help. Date: Thursday, November 07, 1996 9:00 PM My friend and I have installed debian, but he has not signed up to this list. He just got a new larger hard drive for debian, and wanted to move his installed setup onto the new one. I have the old one mounted as root, and the new one mounted on '/mnt'. I have run mkfs on the new HD. I need to know how to copy the files from the old to the new, without having and links destroyed, and without entering the neverending loop with regards to '/mnt/mnt/mnt/...'. All attempts that I have done have resulted in files that links pointed being copies, instead of just making a link, and having '/mnt' copied recursively. Thanks ahead of time for your help. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Moving files to a new partition. Please help.
My friend and I have installed debian, but he has not signed up to this list. He just got a new larger hard drive for debian, and wanted to move his installed setup onto the new one. I have the old one mounted as root, and the new one mounted on '/mnt'. I have run mkfs on the new HD. I need to know how to copy the files from the old to the new, without having and links destroyed, and without entering the neverending loop with regards to '/mnt/mnt/mnt/...'. All attempts that I have done have resulted in files that links pointed being copies, instead of just making a link, and having '/mnt' copied recursively. Thanks ahead of time for your help. Adam Heath [EMAIL PROTECTED] http://www.geocities.com/Siliconvalley/Park/6562/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: telnet acces for root ?
-- I've set up a test PC with Debian 1.1 and want to administer it remotely but am refused a telnet connection as root ... is there a way around this ? Do I have to give The way I handle this problem is log in under my account and type su root Then it'll ask for the root password and bada-bing! You're root. Hope this helps! Tim I use a program called op that works like super. Its not quite as quick as bada-bing its more like bada-bing bada-boom but there's much to be said about this mechanism. It provides accountablility, more granular control over rootly power, etc. Bye ... This is all nice and good, but there is a better way. == /etc (2)$ cat secur* # /etc/securetty: list of terminals on which root is allowed to login. # $Id: securetty,v 3.0 1994/09/09 03:38:19 imurdock Exp $ tty1 tty2 tty3 tty4 tty5 tty6 tty7 tty8 ttyp0 ttyp1 ttyp2 ttyp3 ttyp4 ttyp5 ttyp6 ttyp7 ttyp8 ttyp9 Date = Sun Aug 18 11:06:33 EDT 1996 /etc (3)$ who root ttyp0Aug 18 11:06 (adam.hackers-net) Date = Sun Aug 18 11:08:15 EDT 1996 /etc (4)$ == I used telnet in Win95 to access linux, and cut and pasted this output into here. The ttyN are the function key terminals. ttypN are for telnet. Have fun!
Re: netscape
-- i installed netscape_3.0-beta4-1.deb tonight and it said that i needed netscape-v30b4-export.i586-unknown-linux-elf.tar.gz|Z in the /tmp directory. Well ^ this is all v30b5 in the ftp.netscape.com site. What do i have to do to get netscape running? You could try renaming the file. I don't imagine that the differences between the two are very drastic. Well the message i got from dpkg stated do not just rename... guess i could try it but are there any other options that someone may know of. That won't work. I tried. It puked after a while. The maintainer needs to update it.
Re: IP forwarding and/or Masquerading
-- From: N. Salwen [EMAIL PROTECTED] To: [EMAIL PROTECTED] Cc: debian-user@lists.debian.org Subject: Re: IP forwarding and/or Masquerading Date: Monday, July 23, 2096 4:46 AM I'm still struggling to get my PPP connection working properly. Although = the chat script is making the connection, I can;t send any packets to = any machine other than the one I am dailing in on. Did you set the defaultroute option to pppd? Setting the 'defaultroute' in /etc/ppp/options is not enough. When I installed Debian, I selected the defaultroute option for my network card. When PPPD would try to add a defaultroute for the ppp0 interface, it wouldn't do it because a defaultroute already existed. This prevented me from communicating over the ppp0 interface. To fix the problem, I edited /etc/init.d/network and commented out the line that added the defaultroute for the network card. I rebooted, and everything worked fine.
Re: how to debug pppd
I think there is something wrong with my chatscript but don't where is it. Is it possible to make pppd echo all words on screen? something like dip -v. Look at /var/log/messages. Chat writes debugging info to that file. Adam Heath mailto://[EMAIL PROTECTED] --
