Re: 486 router is very slow
On Sun, 10 Jun 2001 01:52:41 +0200 (CEST), Leen Besselink [EMAIL PROTECTED] wrote: I set up a router with my old 486 computer. I have there potato witch kernel 2.4 installed. try running 2.2 or 2.0 instead, that might help also. In a router, having 2.4 has a _very_ strong point called netfilter. Greetings Marc yes, but then you must know that netfilter eats up much more memory than the previous ipchains. I think maybe this is the problem; having a netfilter firewall with reasonable set of rules requires 32-64 RAM (it depends) I think I saw this discution on the netfilter list stating this... sorry, I didn't follow the disscution here: does it swap a lot? You know, I don't think you can swap something like netfilter rules/information/counters/whatever... I'm not sure how much kernel-'memory' can actually be swapped at all. so this might be the problem.
Re: 486 router is very slow
I set up a router with my old 486 computer. I have there potato witch kernel 2.4 installed. try running 2.2 or 2.0 instead, that might help also.
Re: dbootstrap
On Mon, 21 May 2001, Andrew D Dixon wrote: Leen Besselink wrote: On Fri, 18 May 2001, Andrew D Dixon wrote: Hi All, snip Does anyone know where the source for dbootstrap lives? In bootfloppies maybe ??? Dunno All I can find are the executables in: /debian/dists/stable/main/disks-powerpc/current/prep/images-1.44 on: ftp.us.debian.org There are all the *.bin files there but no source? No, no... I mean the _package_: boot-floppies, try installing that one. Any other clues? Thanks, Andy - New things are always on the horizon.
Re: dbootstrap
On Fri, 18 May 2001, Andrew D Dixon wrote: Hi All, Does anyone know where the source for dbootstrap lives. I wanted to take a look at it and find out what it actually does but I haven't been able to find it. thanks, Andy In bootfloppies maybe ??? Dunno -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] - New things are always on the horizon.
Re: woody and stability
On Fri, 12 Jan 2001, Jerrud wrote: Hey all, I do lots of 3D animations w/ blender (all hail NaN ;-), and while i`m satisfied w/ using potato (solid as a rock for me), there are a few apps that I could use that would make my life a lot more simple. But they are in the unstable tree. My question is this: How would it affect me if I updated to woody? Would my stability be /greatly/ reduced? And (I feel ignorant asking this), how would kernel 2.4 go w/ Woody? If anyone is running this configuration, I would love to hear on how they are doing. Personally I would do the following... collect the few things you want from unstable and compile them from source... (trust me ;) Then install 2.4 kernel... and go ! :) (although I've not been happy about devfs... yet, there are still some things not worked out in that area). Thanks Guys Jerrud -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] - New things are always on the horizon.
Re: woody and stability (Just 1 thing !)
Maybe it's a bad idea to update to 2.4 kernel if your using certain videocards (which perform best with binary-drivers) and XFree 4 ! But trying doesn't harm you ofcourse. :)
Re: sockets
well the problem is with an igmp c program i have, the strace shows:
socket(PF_INET, SOCK_RAW, IPPROTO_IGMP) = -1 EPERM (Operation not permitted)
fstat(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 7), ...}) = 0
old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0)
= 0x40014000
ioctl(1, TCGETS, {B38400 opost isig icanon echo ...}) = 0
write(1, Could not allocate socket.\n, 27Could not allocate socket.
) = 27
munmap(0x40014000, 4096)= 0
_exit(1)= ?
so im assuming a raw socket is a udp socket? and that you need root
You are not allowed to open a raw socket on Linux as non-root, I don't
know if it's different on Solaris (but I would expect it not to be). And
no that's not an UDP socket, it says something about IGMP (Internet Group
Managment Protocol, which is used for Multicast I think).
priviledges for udp sockets? (simalar to why you cant run nmap udp scans
unless you're root).
No, that is also because of Raw sockets in general. There is no reason why
you can't use UDP sockets... just not raw sockets. That's why nmap won't
work for, UDP, TCP and probably not even Ping (it uses rawsockets to
send out very specific packets, well I guess you know).
am i on the right track?
Eventually. :)
Regards,
Marc-Adrian Napoli
Network Admin
Connect Infobahn Australia
+61 2 9212 0387
Re: kernel 2.4, modutils requires libc6 2.1.97
On Sat, 6 Jan 2001, Jameson C. Burt wrote: [..] This flashes a WARNING to me, since libc is very foundational; indeed, a few Debian distributions back, most distribution upgrades failed because of libc changes. Does an upgrade of libc6 from version 2.1.3 to 2.1.97 largely work? [..] Perhaps my best solution would compile modutils version myself. You got it ! :) Besides, I now notice that the kernel source linux-2.4.0.tar.bz2 document .../Documentation/Changes has upped the modutils version requirement from 2.3.15 (in linux-2.4.0-test8) to modutils version 2.4.0 (in linux-2.4.0) which appears nowhere, not even in the .../pool/* directories. Actually, modutils 2.4.0 doesn't even exist (as far as I could find out, I could be mistaken) ! I'm just trying out the one from testing... seems more then ok.
Re: GCC and EGCS
On Sat, 6 Jan 2001, Raffaele Sandrini wrote: Hi all I am really confused over the different compilers wich seems to be together. Does EGCS exist today? or is it included in GCC. Many software recomments to be compiled with egcs (???). I have GCC 2.95.2 installed on my machine If I'm not mistaken this IS egcs. wich seems to be the latest version of GCC. The kernel doc tells me that i have to use GCC 2.72 (???) and only for secondary choice the latest GCC! Depends if you have a new kernel you'll need newer gcc otherwise only older gcc will work (not sure where the 'change point' was, but I think around 2.2.7?). Could someone solute my confusion? Thanks for every help. cheers, Raffaele -- Sent through GMX FreeMail - http://www.gmx.net - New things are always on the horizon.
Re: [2.4.0] migration to devfs
lr-xr-xr-x1 root root 33 Jan 1 1970 /dev/cdroms/cdrom0 - ../ide/host0/bus1/target1/lun0/cd lr-xr-xr-x1 root root 30 Jan 1 1970 /dev/discs/disc0 - ../ide/host0/bus0/target0/lun0/ lr-xr-xr-x1 root root 30 Jan 1 1970 /dev/discs/disc1 - ../ide/host0/bus0/target1/lun0/ lr-xr-xr-x1 root root 30 Jan 1 1970 /dev/discs/disc2 - ../ide/host0/bus1/target0/lun0/ Did those people who whine actaully read the above (and ofcourse the reasons to switch to devfs, instead of that 'hack', although this isn't perfect too, it's getting better !) ?
Re: archive.debian.org
Everybody a happy new year! Thanx. I have a question.. ahhh, guestions..., my speciallity, but I'll try to answer one for a change. I hope I am false but it seems to me that archive.debian.org is down. Are there mirrorsite of this site? Seems it crached: nslookup archive.debian.org 216.66.54.50 nslookup 216.66.54.50 saens.debian.org http://db.debian.org/machines.cgi?host=seans Status: Crashed, hoping to recover. :(( I wish I could remember where I put those CD's ! As I have the same problem probably, Hope this helps, Lennie. - New things are always on the horizon.
Re: adduser automation
On Sat, 30 Dec 2000, :yegon wrote: Is there an easy way to automate the adduser script I'd like to run a script which creates several users listed in a text file (or db) and sets their passwords Specify as many options as needed to: useradd, it's non-interactive if I remember ok. (maybe first you need to create the home directory, and copy /etc/skel) groupadd is also available With all this combined you should be able to do it.
Re: normalizing mp3s under debian/linux
On Sat, 23 Dec 2000, Robert Waldner wrote: On Sat, 23 Dec 2000 10:17:27 EST, Michael P. Soulier writes: I=B4m looking for an app which can normalize my mp3s under linux. All I was able to dig up on myself was normalize[0], but this only can= do with .wav=B4s, and as I=B4m not good (eg I=B4m clueless) at=20 shell-scripting (and perl and almost anything else for that matter) I=20 have no idea how to automatically cope with my 5k+ mp3=B4s... Seems like a nice idea. Is there anything out there for converting to a wav and back? Then you could convert, normalize, and mp3 again. `mpg123 -w` for mp3-wav and gogo to reencode then. But I´ve no clue how to automate thismanually doing this for 5.000 mp3´s would be more than a major PITA, I fear. It's not that hard really, just look at this little scriptie: --- cut hear and buy a new monitor --- #! /bin/sh for FILE in `find . -name '*.mp3'`; do echo $FILE done --- cut hear and buy a new monitor --- TIA, rw -- / Ing. Robert Waldner | Network Engineer | T: +43 1 89933 F: x533 \ \ [EMAIL PROTECTED] |KPNQwest/AT | Diefenbachg. 35, A-1150 / -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] - New things are always on the horizon.
Re: need feedback on PostgreSQL clients
On Wed, 20 Dec 2000, Ian Setford wrote: I am looking for a good GUI client to complement psql. What are good ones to use? I have seen pgAccess, ksql, and kpsql but I would like to know what the Debian community thinks about them. What are their good points and bad points. What would you recommend? As an aside, what would be the best way for non-*nix developers to access Postgres? Personally I prefer to have an Apache webserver running PHP4 access my pqsql server (http://www.phpwizard.net/projects/phpPgAdmin/) There is a simlilair tool for mysql there too (it's what they started with, to bad they didn't use any abstraction layer :/).
Re: 3c509 is being annoying
I have disabled pnp and the card insmods well. I can also give it an address with ifconfig and ping to the card, but I am unable to ping another host (and vice versa). As far as I know I have no firewall running Sounds like routing to me... dunno what though, if you use a 2.2 kernel it should automagically make atleast one route, here is the most simple routing table I could think of: leen:~# route -n Kernel IP routing table Destination Gateway Genmask Flags Metric RefUse Iface 192.168.1.0 0.0.0.0 255.255.255.0 U 0 00 eth0 0.0.0.0 192.168.1.1 0.0.0.0 UG0 00 eth0 What you need is, one route to the 'local network' something like: 192.168.1.0 for example or 192.168.0.0, also (if you want to connect to the outside world, a router/gateway to connect too: 'route add default gw 192.168.1.1' for example) Here is an example /etc/network/interfaces: # An example ethernet card setup: (broadcast and gateway are optional) # iface eth0 inet static address 192.168.0.42 network 192.168.0.0 netmask 255.255.255.0 broadcast 192.168.0.255 gateway 192.168.0.1 (taken from /usr/share/doc/netbase/examples/interfaces) This should set it up correctly. Also what could be the problem is that you have no/wrong nameservers configured in /etc/resolf.conf: nameserver 192.168.1.1 (just an example ofcourse) Because if you have that problem, then that would also be a reason you can't connect to the outside world (or inside even, if you use a namebased system). and no special options in /etc/hosts.deny. hosts.deny, doesn't do anything for you personally, only for people connecting to your computer. Just some ideas, before I go to sleep.. Zz.
Re: Network Monitoring recommendation
On Thu, 14 Dec 2000, sena wrote: On 14/12/2000 at 17:26 +, Craig Coles wrote: I am looking for a tool to monitor the through-put of my connection to the internet. I want to know if my connection is maxing out, or my users just have slow connections to the internet. Is there a debian package to do this, or a recommendation otherwise? Can these tools determine what kind of traffic it is (http,ftp,etc.)? I'm using apache on woody, I know apache does some statistics with server-status but seems to only show averages over uptime. And do you know of any program that also (or just) measures the quantity of data (er.. you know what I mean :)) that passes through an interface? I think ipchains does that, but I think it just counts packets... Is there any way of counting GB/MB/whatever? actually ipchains counts packets and bytes, so no problem there: ipchains -nvL Regards, sena... -- [EMAIL PROTECTED], http://decoy.ath.cx/~sena/ gpg fingerprint: F20B 12A8 A8F6 FD1F 9B1D BA62 C424 8E73 DD2E 47C8 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] - New things are always on the horizon.
Re: remote management
Would this be a dial-in connection to the computer to manage it? What I need is to manage the computer remoteley all the way back to the lilo prompt. Are there Withut special hardware it's impossible to get to the Lilo prompt, after all Linux manages it's own serial ports and it hasn't even started then ! :)
Re: cron/at question.
Three questions: 1) both cron and at seem to run programs in the background. However there is no pid when I do a ps aux, so how do I stop a job from running? Well, they don't get run immediatly ofcourse, only when it's time, you can try to kill them then, but best is to make changed to your cron or at configuration if you don't want something to get run. Maybe you can use something like start-stop-daemon, maybe that can make .pid-files for you (it does say something about it, not sure, never tried it, if someone knows to tell me too). 2) can cron or at jobs be run as one-liners? Eventually I want to write a perl of sh script (not that I know how, but I want to try) incorporating Sure you can do whatever you like (easiest would probably be to make a shell script or something and run that). one of the two. For cron you need to use an editor to write the command, at comes with its own prompt. 3) how can I change the default editor for cron. vi is too cumbersome for me. I think you need to do something like (if you use a C-shell like shell it's different I'm afraid): export EDITOR=emacs or export EDITOR=joe whatever.
RE: cron/at question.
Fine, so where can I find it? If I run radio straight from the command prompt ps and ps aux will list a pid for radio. If I run radio via cron/at there is no pid when I do ps or ps aux. In fact, there is no pid for anything at the time cron/at executes radio. You'd expect at least something, either a pid for radio or a pid for at. That's what I found strange, hence my question. What your saying isn't possible, maybe you're looking in the wrong place. Can you tell us how you're trying to find out what pid it has ? Also, I actually looked at the manpage of start-stop-daemon andit's the '-m' option which will make a pidfile for you.
Re: how to fake from address in an email
We need to send out an email response from a CGI when someone places an order, but we want to set the from address to something other than the user/machine where the CGI is running. How to do this with the various email packages? Currently using exim, but may switch to postfix or sendmail. I think you need something like trusted user/hosts (localhost ?) and or relaying (but don't allow relaying for every1 !). Just look it up (the ways this is achieved in the packages can be very different so).
Re: cron/at question.
Should have given you an example of what I am doing. Here it is # at 23:17 warning: commands will be executed using /bin/sh /usr/local/bin/radio 104 Maybe this command runs sub commands and that's why radio doesn't show up ? Looks like it's something you made, so probably a shell script (normally they wait until subcommands get executed though (unless you use things like: and friends) ? When the clock turns 23:17 I hope this is the computer clock your talking about. :) I'll look into this -m switch for daemons tommorrow. Haven't a clue where to invoke this yet. Bedtime now. man start-stop-daemon or maybe even: info start-stop-daemon
RE: Going Debian: advice request
The way I do it if I want an unstable package is to add unstable to my sources.list, apt-get update apt-get install package -s If all looks OK then I go ahead and do it. Sometimes it will want to upgrade a whole lot of stuff, in which case I don't do it. What I do is download the source of the package and build it, that always seems to work (only thing that can go wrong are the install-scripts).
Re: Unidentified subject!
What ever I put in It say it can't find 'images-1.44/something/rescue.bin. I think there was also an option for it to 'list' possible files that it can use, I think also that's what the manual means. The proper path should be something like: /dists/potato/main/disks-i386/images-1.44/something?/rescue.bin so this is probably what you should enter: /dists/potato/main/disks-i386/
Re: Logging FTP
Does anyone know if FTP-transfers/connections are logged? I'll try to answer this. I tried to find any logging info in the syslog/daemon.log but it seems like it doesn't log anything. sure it does (kinda) the only it logs in the file /var/log/syslog: Dec 12 22:45:00 debian in.ftpd[4509]: connect from 127.0.0.1 It's not much but it does do it ;) That's all the really small and standard ftp (hardly even a) server does. If you want/need logging try installing something like: - wu-ftpd - proftpd - probably more You can do all sorts of things with any of those.
Re: SIOCADDRT: File exists
I use my Debian 2.2r2 as a router, but it always shows the message SIOCADDRT: File exists at boot every time. I use the kernel 2.2.17 that come with the debian 2.2. That's because your using the old /etc/init.d/network file, what you can do is edit that file and comment: #route add -net 127.0.0.0 #route add -net 192.168.0.0 or whatever your network numbers are, the idea is, 2.2 kernels do this them selfs. In Debian 2.2 there is a new file: /etc/network/interfaces which is the new way, just look in the docs of I think netbase (/usr/share/doc/netbase). Hope this helps. - New things are always on the horizon.
Re: Helix-gnome control center problems
On Sat, 9 Dec 2000, Rob VanFleet wrote: For some reason, the gnome control center doesn't want me to configure any sawfish aspects. When I click on any of the sawfish configuration options, the left pane displays nothing, and eventually, the entire app freezes and has to be killed. Another thing I noticed is that the control center often thinks that Enlightenment is the default window manager, even though it is sawfish that is actually running. I don't know if this has anything to with the previous problem, but I thought I would mention it. Well, what I did was, set it to sawfish for a moment and then set it back to enlightenment that seemed to help a little for both problems. But I think we should report it as a bug. -Rob
Re: Need help: Moved /home now lost xmessage use?
Now I did the same from X in root and there's no problem, so it has to be with permissions or my not telling someting I had moved??? Well, did you give your /usr/local and /home the same permissions that they had originally ? You didn't copy anything right ? Just remounted it ??? If you did copy, how did you do that ? cp -a would have been a good way (-a=backup, means so much as keep all permissions and make symlinks and sockets too I think).
Re: port 756/tcp and port 754/udp
netstat -tul showes me that something's listening one port 756/tcp and on port 754/udp - what could that be? even when stopping networking, inetd and all other daemons i'm running i still have these 2 ports open? what is ist? sounds like one of the rpc services, those damn things seem to choose random ports when they startup, install lsof to be sure(the package is lsof-2.2) then run lsof | grep port A good (easier ?) to do this would be to just add the -p switch to netstat: netstat -ptul
Re: IP masquerading
On Sun, 10 Dec 2000, Sebastiaan wrote: Hi, you can make a script and put it in /etc/init.d and make a link to one of the /etc/rcX.d. With the number (like S40firewall) you can set the priority. As an alternative, in Debian you have a /etc/rc.boot where you can put files which must be started at boottime (but not after a init 1; init 2). Actually if you want to make it easier on yourself, Debian has a tool for this: update-rc.d Which you can use to add a script to the run levels you want (or you can choose default).
Re: Moved /home now lost xmessage use? Solved!
On Sun, 10 Dec 2000, Jonathan Gift wrote: Here's the weird thing. It works from root X on another partition. But I created a test user and tried xmessage from there and got the same error. That points to the partition? To fstab? What else? Fstab looks ok, same settings on my new /hda3 as I had on /hda8... I went looking for more info in man but got a can't open temp file. that was the tip off, it was a permision problem but not on /home but /tmp. I gave tmp rwx for all and xmessage works!! BTW, what is the default for /tmp? rwx owner/group/others? drwxrwxrwt6 root root 3072 Dec 10 12:33 tmp The t you get when you do chmod +t if I'm not mistaken. This is what I have here (I don't think I've ever touched it) -- Hey, I think I finally got the hang of i- - New things are always on the horizon.
Re: What's Helixcode?
On Sun, 10 Dec 2000, Bek Oberin wrote: Keep hearing peolpe talk about helix or helixcode or something. I've never heard of this, what on earth is it?? A commercial company trying to help create a better GNOME, it's an upstart of Migual de Incaza (he's the lead developer and fouder of GNOME). http://www.helixcode.com
Automated CPU-optimised-/cross-building of Debian-packages.
Hello, Does anyone know how to do cross and/or CPU-optimised building of Debian packages, preferable with the Debian 'standardtools' ? Normally when I build a debian package I do this: dpkg-source -x package-name-version.dsc cd package-name-version dpkg-buildpackage cd .. dpkg -i package-name-version_CPU-type.deb Or something similair. I think there was also a way for one of the tools to do a whole directory of those at ones, btw. But that's about how far I got. tia for any help, Lennie. - New things are always on the horizon.
Re: Automated CPU-optimised-/cross-building of Debian-packages.
On Sat, 9 Dec 2000, Colin Watson wrote:
Leen Besselink [EMAIL PROTECTED] wrote:
Does anyone know how to do cross and/or CPU-optimised building of Debian
packages, preferable with the Debian 'standardtools' ?
For cross-compiling, try the dpkg-cross package.
Yes I just found that one out myself, pretty dim of me. :)
For optimized compiling, try the pentium-builder package (the name
notwithstanding, I don't think it's just for the i386 architecture).
I also found that package, but for some reason, I think dpkg-buildpackage
overwrites my environment settings ??
Normally when I build a debian package I do this:
dpkg-source -x package-name-version.dsc
cd package-name-version
dpkg-buildpackage
'dpkg-buildpackage -rfakeroot' is good, if you install the fakeroot
package, since then you don't need to be root to build packages.
well, as a test I did this:
apt-get install gcc-m68k-linux
(which also wanted to install the appropriate binutils)
so now when I do this:
dpkg-buildpackage -rfakeroot -am68k
I get:
dpkg-buildpackage: source package is tar
dpkg-buildpackage: source version is 1.13.17-2
dpkg-buildpackage: source maintainer is Bdale Garbee [EMAIL PROTECTED]
dpkg-architecture: warning: Specified GNU system type m68k-linux does not
match gcc system type i386-linux.
fakeroot debian/rules clean DEB_BUILD_ARCH=i386 DEB_BUILD_GNU_CPU=i386
DEB_BUILD_GNU_SYSTEM=linux DEB_BUILD_GNU_TYPE=i386-linux
DEB_HOST_ARCH=m68k DEB_HOST_GNU_CPU=m68k DEB_HOST_GNU_SYSTEM=linux
DEB_HOST_GNU_TYPE=m68k-linux
[.. so after a while I get ..]
checking whether the C compiler (m68k-linux-gcc -O2 -g -Wall ) works... no
configure: error: installation or configuration problem: C compiler cannot
create executables.
make: *** [build-stamp] Error 1
make: Leaving directory `/usr/src/leen/tar-1.13.17'
Looking at the config.log file it says:
configure:938: checking whether the C compiler (m68k-linux-gcc -O2 -g
-Wall ) works
configure:954: m68k-linux-gcc -o conftest -O2 -g -Wall conftest.c 15
configure:951: warning: return-type defaults to `int'
/usr/m68k-linux/bin/ld: cannot open crt1.o: No such file or directory
configure: failed program was:
#line 949 configure
#include confdefs.h
main(){return(0);}
When I look up what package(s) contains crt1.o they are these (on the
website):
usr/i486-linuxlibc1/lib/crt1.ooldlibs/libc5-altdev
usr/i486-linuxlibc1/lib/gcrt1.o oldlibs/libc5-altdev
usr/lib/Mcrt1.o devel/libc6-dev
usr/lib/checker/i686-pc-linux-gnu/chkr-crt1.o devel/gccchecker
usr/lib/crt1.odevel/libc6-dev
usr/lib/gcc-lib/i386-gnu/2.95.2/crt1.odevel/gcc-i386-gnu
usr/lib/gcrt1.o devel/libc6-dev
usr/lib/oskit/unix/crt1.o devel/oskit
When I install the oskit (what else can it be ? turns out not to be it)
and an updatedb later (and un umount of a vfat-partition ;)
I find these:
/usr/lib/crt1.o
/usr/lib/gcrt1.o
/usr/lib/Mcrt1.o
/usr/lib/oskit/unix/crt1.o
I'm now have no idea left what this can be. Any ideas, would really help.
:)
Re: CRASHED MACHINE = FB.overflow:1
On Wed, 29 Nov 2000, Eric G . Miller wrote: On Wed, Nov 29, 2000 at 06:27:52PM +, Gordon Dykes wrote: After a few minutes on the net my Debian box crashes and I can not work in ANY terminal. I can change between them but they all have the message FB.overflow:1 I would interpret that to mean Frame Buffer Overflow. Something is doing bad things with your video (it'd have to be something low-level). Also possible, there's a problem with your video hardware (bad memory?). course, I could be interpreting that message incorrectly (never seen it before). To get rid of your problems, I would just Not used FB anymore (check your /etc/lilo.conf (Documentation about framebuffer ? /usr/src/linux/Documentation/)), atleast if you don't need it for XFree4 then that's an easy choice (as long as you have x86 hardware that is).
Re: Peculiar DENY from ipchains?
Anyone ever seen a DENY from IP chains like: Nov 29 19:44:14 john kernel: Packet log: output DENY ppp0 PROTO=1 10.12.1.15:65535 63.81.184.67:65535 L=21 S=0xD0 I=46379 F=0x0042 T=255 (#82) where 10.12.1.15 is my ppp interface address? What's going on with 65535? I'm sending it out, but don't know why. according to /etc/protocols protocol number 1 is ICMP not sure why it specifies the port..for ICMP i think u can just ignore the port .. This is not true, normally it would list the ICMP type, but it seems didn't know what it was or something, because it picked the highest one (that's for example also what it does if it doesn't know the protocol: just look in /usr/share/doc/netbase/ipchains-HOWTO.txt.gz).
Re: nameservers in bind
On Mon, 27 Nov 2000, Sebastiaan wrote: Hello, I am trying to configure my dns so that a www query to my domain will be handled by another nameserver. Yet, I have the following line in /etc/bind/db.domain: www IN NS 202.67.129.130 but when I do a nslookup with q=any, it gives: Non-authoritive answer: www.domain.comnameserver = 202.67.129.130.domain.com If I'm not mistaken it's supposed to be done this way: www IN NS www-dns www-dns IN A 12.12.12.12
Re: hosts for rdate?
On Sat, 25 Nov 2000, Bruno Boettcher wrote: hello, i am searching for some hosts for time updating of my hosts where can i find the listings of hosts offering this service? i am in France BTW, so any host network-near Strasbourg would be fine... Just install or unpack the ntpdata package, if I'm not mistaken this contains (used to anyway) a list of possible servers. Hope this helps, Lennie.
Re: eth0 problem
The card is a Netgear with a Tulip Digital 21040 chip which is what we have in the Redhat and SuSe servers and they do not behave this way. Do they all have the same kernel (read: driver) version ? Maybe Suse/RH use a newer driver ? Because I can remember tulip drivers having problems with this now and then.
Re: dump package list to file?
I want to dump a list of the packages I have installed on a current potatoe installto a file, so I can later build a machine with the same packages? How do I do this? dpkg --get-selections file
re: setting up eth1 interface
###network modules alias eth0 tulip alias eth1 tulip What if you added the right IO, to the lines above ? Did you try that ? Although I'm not sure, as far as I could tell that module didn't have options. #options tulip io=0x400,0x800 This was for both cards right ? The options line is now commented out since neither card would work with that line. modprobe -v eth1 was silent, no errors or messages in /var/log/messages If it's silent that meant it didn't do anything. Maybe you need a different driver then the Tulip one, otherwise I would think he'd find it (especially if you added the IO-parameters). If you run out of ideas I found an other one: compile into the kernel and add this to your lilo: ether=0,0,0,0 eth0 ether=0,0,0,eth1 or something like that. Also did you try looking at the diagnostic program at his webpage ? http://www.scyld.com/network/tulip.html And you could also try downloading the latest driver there. Compile that and try it. BTW, what happends if you haven't inserted the module yet (there is no eth0) and the insert with the right options to find eth1 ? Other ideas, not yet, maybe later. ;) (I'm sure this can work in some way). Now I've gotta go and shower and get to work fast, Lennie.
Re: Setting up eth1 interface
On Thu, 23 Nov 2000, Sean Norris,,, wrote: Hello All, I have been trying to set up a small local network with a linux box as a gateway and my wife's win98 laptop. I am running potato with a fresh 2.2.17 kernel from kernel.org. Currently, a D-link DE-530TX is working well with a tulip driver loaded as a module to connect to my cable modem. The second NIC is a linksys LNE-100TX ver. 4.0, which also uses the tulip driver. However, I cannot get the eth1 interface to initialize. #ifconfig eth1 10.0.0.1 netmask 255.0.0.0 up SIOCSIFADDR: No such device eth1: unknown interface: No such device SIOCSIFNETMASK: No such device eth1: unknown interface: No such device Are you sure the module loaded ? sounds like some kind of driver problem. 'lsmod' will tell you more about that, also you could take a look at the output of 'dmesg'. Maybe that will tell you what went wrong. Hope this helps.
Re: pentium-builder question
So my question is: how to make sure I'm building celeron optimised packages? I've not tested this (as I don't have any such files at hand), but I think one way is to look at the result (in one of two ways): file /usr/src/myprog/myprog or readelf /usr/src/myprog/myprog This will give information about your binary's.
Re: pentium-builder question
Well, thanks to this suggestion, I'm now more or less sure pentium-builder is not doing what it is supposed to do. Now if I only could figure out why not... I just build something with -m486 but I don't see it back in readelf -h or file either. So this wasn't the right way to do it, sorry. :/ So this was not a good way, seems I was wrong.
Re: virtual ips
Hi, I changed from SuSe Linux to debian. How and where do I enter virtual ip as like eth0:1, eth0:1... Dunno what the official place would be to set that up, if there is one, it would probably be /etc/network/interfaces or something. what used to be the way was putting it in: /etc/init.d/network It's just a shell script that sets up all networking stuff: ifconfig eth0 10.0.0.3 netmask 255.255.255.224 broadcast 10.0.0.255 ifconfig eth0:0 10.0.0.2 or something (these numbers are not even thought about, pretty much random !). Thanks Grischa
Re: Debian installation from NT
On Tue, 14 Nov 2000, C. Falconer wrote: At 10:44 AM 11/13/00 -0200, you wrote: I've installed and configured Debian a lot of times in my local network, always from our linux local debian mirror. But I've never installed it from a NT mirror and now I'm having some troubles installing debian via NFS from it. Does the local linux mirror use NFS to serve the files? Does NT use NFS or SMB (a-la samba) AFAIK NFS for NT is a separate product, and not part of the standard package. What I did ones was install an ftp server on the NT machine (you can grab them of the net for free, you could also use IIS or something for it, if you got it). Only problem I ran into was the symbolic links on my CD (where I copied it all from), so what I had to do, was move everything from binary-all to binary-i386 (or other architecture) to get things installed (otherwise you'll have zero length files :( ). But this was just as first install on a system with no CD-rom drive, later I definitly used a different way. :) Maybe this was helpfull.
Re: SysAdmin list and question ..
On Thu, 9 Nov 2000, Alberto Brealey wrote: On Thu, Nov 09, 2000 at 08:53:26AM -0800, Dave wrote: 2. My debian box has several scsi disks on it, and if one is turned off at boot time, the machine mounts the wrong disks on the defined filesystems, even though they are listed correctly in /etc/fstab. Is this is a problem with the way the linux kernel handles device name allocation (is this the way to say it?). the thing is, /dev/sda is going to be the _first_ scsi disc on your system (the first one your kernel recognises), so if you turn on your machine with a only a disc with an id of 5 in the second controller, it will be /dev/sda. but if you hook up another disc with id 2 on the first controller, this new disc will become /dev/sda, nad the previously-know-as-/dev/sda wil become /dev/sdb. so: if you add or remove discs, every hd that has an scsi id higher that the one on the changed device will change its /dev/sdax place. i believe solaris does not have this problem, since it handles discs like /dev/scsi-bus-id/scsi-id or some variation... there a way to force the machine to mount a given disk on a given filesystem? Thanks! not on linux, i don't think so. hope this helps, Alberto. I'm not sure, but maybe linux 2.4 solves this.
Re: Supported hardware for debian
On Wed, 8 Nov 2000, [iso-8859-1] Jostein ?gedal wrote: I am preparing to buy a new machine to run debian linux 2.2. As Im buying it solely for linux, I would like any good advices on what sort of hardware would be recommended for debian Linux. Maybe any of you have some good links or personal experience on this? One way to find out, is look it up here: http://www.debian.org/releases/stable/#new-inst Is has all kinds of information, including information about supported hardware and says, the Debian is like any other Linux and then refers to this document: http://www.linuxdoc.org/HOWTO/Hardware-HOWTO.html Hope this helps, Lennie. PS One tip, buy Good memory, Linux can be a bit more memory critical then for example Windows.
Re: how to load modules at boot time
but if i wanted to be more sophisticated about it, would i add them to /etc/modules? Yes, I think this is the official place (the file also won't be changed, when you (or apt-get upgrade gets) run).
Re: Mozilla M18 on potato (pentium100)
On Sat, 4 Nov 2000, Daniel Borgmann wrote: when i try to run mozilla as root it says: Gtk-WARNING **: cannot open display: :0 - what does that mean?? and yes, i did start it in X. when i do the same as a regular user, i get no error message at all. Sounds like you logged in as a regular user, opened up a terminal program like xterm, Eterm, Konsole, etc., and su'd (with the 'su' command or similair) to root and tryed to start mozilla as root. This does not work in Unix en thus Linux. In Unix de user that 'owns' the 'X' process that is running is the Only user that can modify the X-screen, not even root can do that. This is a (security) feature not a bug. :)
Re: KDE2 -aria
On Tue, 31 Oct 2000, guran remberg wrote: What a fantastic program this apt-get is, I am now writing this in KDE2. I just added deb http://kde.tdyc.com potato.kde2 to my /etc/apt/sources.list and wow. Who ever you are, tdyc, many thanks. That WindowManager I was having was clearly stupid. When I had eight programs running in different desktops, Yes, that is ONE person doing the packaging for KDE2 (ofcourse KDE2 is not a one-man show), he just 'resinged' though, because ONE person out there said bad things to him. But thankfully he did most of the work now, maybe there are only small bugfixes for Potato/KDE2 right now, that need to be done. Also if you are interrested in other non-standard-Debian packages, take a look at: http://www.helixcode.com Maybe that's something for you (You can do that the same way by adding it to /etc/apt/sources.list). APT is really kewl. :) You can even do automatic builds from source (if you need it) ! :) Hope this helps, Lennie.
Re: KDE2 -aria
Hmm - I see the front page of http://kde.tdyc.com/, but on balance the discussion on http://dot.kde.org/971680096/ seems to suggest that he's going to keep doing work on potato for a while longer. I have a message on debian-devel from Ivan Moore less than three hours ago which talks about KDE2 on potato, too. ahh, well, he most have changed is mind, at first it was just I quit, I'll just be doing woody and now he's doing some basic work for potato again (actually, I did notice he did KDE2 for potato, because I did apt-get when KDE2 came out and it was final0 so I should have known). Well I also heared a lot of people (some (all ?) of them new maintainters), did pick up on a lot of packages (what would be kewl is if they got added to 2.2rsomething, but I doubt it). Looking at http://www.debian.org/devel/people, it's scary how many KDE packages he's been maintaining by himself. I know, that's why I was sad about him saying: I quit, he's a good maintainer (haven't had problems)(definitly hard working !).
Re: KDE2 -aria
On 31 Oct 2000, Bob Bernstein wrote: Perhaps it was unintentional, but the sound of this sentence rubs me Everything I saw should be taken lightly first off, I have a Very bad memory and secondly my english 'looks quiet' good, but a lot of the times I don't get the details right. So I'll shut up now, I don't wanna start a bigger thread. *crawling back to where he came from*
Re: Make Menuconfig + curses.h
On Wed, 25 Oct 2000, Hanno [iso-8859-1] B?ttcher wrote: /usr/src/linux. Then I tried to run make menuconfig but I got an error similar to dialog.h:29: curses.h: no such file or directory. I tried I think this is solved by installing the ncurses header files: apt-get install libncurses5-dev sit back have fun ! :) - New things are always on the horizon.
Re: I have apt on RH, now i want dselect (or capt or sth. like that)
On Mon, 16 Oct 2000, Piotr Krukowiecki wrote: I thought it's clear. There are two computers - one in my home (debian, slow modem connection) and one where i study (RH, fast connection). Now, i can take my /var/lib/dpkg/status file, compress it (~100KB) and send it to RH. There i could choose what packages i want (using dselect), download them (apt-get -d dselect-upgrade), take them to my home and install. Why don't you just use masqurading firewalling ?? First you need to have masquerading setup in your RH's kernel, then you add one rule to your ipchains, ipfwadm or iptables (depending on your RH's kernel version) and add it to the startup. Then just point your gateway on your Debian machine at your RH machine. This seems like a lot easier way. And everytime you want the debian machine to use his slow connection, change the Debian's default route over the modem (PPP?) link. PS Here is the HowTo on masq.: http://www.ibiblio.org/pub/Linux/docs/HOWTO/IP-Masquerade-HOWTO - New things are always on the horizon.
Re: DHCP renumbering and NAT
On 30 Sep 2000, David Z Maze wrote: I can take it as a fact of life that I'll occasionally be renumbered; I can deal with this. I guess my question is this: is the ipmasq package clueful enough to recognize when this happens, and tweak the firewall rules appropriately? (Experience this morning suggests not.) Failing this, is there a good way to put some sort of appropriate hook into dhclient to make the right thing happen? TIA... What you can do is just not use the outside _address_ when masq. but the interface name, you can do the same for blocking. Hope this help, Lennie.
Re: Looking for a good Red Hat list also.
On Thu, 28 Sep 2000, Steve Hastings wrote: Hi All, snip Sorry, dunno about that all your other questions. Finally, are there any versions of Linux that runs on the RS6000, in particular the 42T/41T? What you should do is take a look at IBM's pages on this topic: http://oss.software.ibm.com/developerworks/opensource/linux/projects/ppc/ Thanks, Steve
Re: succesor to dpkg
On Wed, 27 Sep 2000, Ben Collins wrote: It was called HPML and it's fairly dead. There is a CVS, but nothing more than a skeleton and a spec document. I think Wichert is working on something else. Yes, and he's gonna propose the non .deb people should use that too (as it's better then deb and rpm and also it would be great to have One packaging systeem): http://www.wi.leidenuniv.nl/~wichert/talks/ (and then choose the talk about package managment obviously). Quiet interresting to read.
Re: Debian/gnu Linux Install
On Tue, 26 Sep 2000, Jeremy S. Warn wrote: Having tremendous difficulty installing onto harddrive. I completely cleaned my system off, and I'm now stuck with a pc that won't boot off the partitions that the software setup, and, due to loss of drivers, I can't read the CD I bought. I'm VERY desperate for assistance Jeremy S. Warn _ Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com. Share information about yourself, create your own public profile at http://profiles.msn.com. Download an installation (read bootable) floppyset ? And write it to a few floppy's, you should be able to read the CD from that installation set. This is the first thing that comes to mind... and I gotta sleep now... tomorrow my company will get _two_ new perminant internet connections (from zero to two in a day ;)) ! -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: Problem with dpkg
On Sun, 24 Sep 2000, Boris Boutillier wrote: snip Does anyone know what to do when dpkg give such an output when installing a package : dpkg-divert: rename involves overwriting `/usr/sbin/paperconfig.libc5' with different file `/usr/sbin/paperconfig', not allowed snip I don't know how to solve this problem. I just installed the patato and had this problem. I think there should be an option but can't find which one. Well if you mean you want to overwrite you should take a look at: dpkg --force-help and what it says is, you can use this option to overwrite: dpkg --force-overwrite -i file.dpkg snip
Re: OT: SCSI woes?
As some of you know, I've recently converted to a full scsi system. I compiled a 2.4.0-test8 kernel with the following options: snip SCSI device sdb: hdwr sector= 512 bytes. Sectors= 35843671 [17501 MB] [17.5 GB] sdb: sdb1 sdb2 sdb3 Excuse me? sdb? I only have one HD in the system which is broken into three partitions. Windoze, swap, linux. What does the second entry mean? I also had this with test8, but I can't yet remember why, I switched back to the last good test version I was using (2, was the last time I tried a test version, didn't have the time for any others). Not just because of this but also because the whole kernel modules directory structure (in /lib/modules/ you know what I mean (?)) changed and I didn't feel like playing with the /etc/modutils/ again. I would be interrested in hearing why this is though, but I think it's just some kind of test8 specific bug (it's still fun having so many harddisks though ;) you can even access them over those dual devices ! Atleast I could (I tried to startup cfdisk or something I think). (I just looked it up a bit in the kernel archives) seems they also turned some things around in the SCSI initialitation code. There was a whole thread about SCSI scanning so, I'm sure they are working on it. Try using a test9pre kernel, maybe that works better. Or go back to 2.2.17 or something (for now). All I can suggest. About your other issues I don't know.
why traceroute in sbin ?
Hello, Just a quick question, but why is traceroute stationed in a sbin directory ? As normal user I can use it too, so why ? or is this just because of some arcane old tradition ? And some old scripts depend on it ? I think it's so strange. tia for an answer, Lennie.
Re: I'm in trouble
On Fri, 22 Sep 2000, Satyajit Das wrote: snip My problem is now I'm not able to boot Red Hat linux. snip please help me how can I solve this problem, waiting reply urgently because I have so many important files and email. Until we figure out what happend, maybe someone has had the same problem, why don't you just mount that partition, on storm Linux ?: just type this on a console or xterm: mount /dev/hda5 /mnt -v cd /mnt ls -l That should do it, or is there an other obsticle ? And maybe you can 'paste' your /etc/lilo.conf in your email, so we can take a look at it. snip
mtu/mru ISDN dialup.
Hi folks, When I run tcpdump on my ippp0 connection to my ISP, I get a lot of packets like this: tcpdump: listening on ippp0 truncated-ip - 46 bytes missing!0.40.224.225 64.0.127.6: (frag 26486:[EMAIL PROTECTED]) [tos 0x17] [ttl 0] truncated-ip - 28 bytes missing!0.58.208.128 0.0.64.17: (frag 38871:[EMAIL PROTECTED]) [tos 0x17] [ttl 0] 194.165.94.5.53 195.86.113.3.1042: 41* 1/3/3 (159) truncated-ip - 38 bytes missing!0.48.64.158 64.0.127.6: (frag 38871:[EMAIL PROTECTED]) [tos 0x17] [ttl 0] 64.28.67.48.80 195.86.113.3.63822: S 1395078506:1395078506(0) ack 13465379 win 16060 mss 1460,nop,nop,sackOK (DF) truncated-ip - 46 bytes missing!0.40.67.158 64.0.127.6: (frag 38871:[EMAIL PROTECTED]) [tos 0x17] 1.190.68.158 64.0.127.6: (frag 38871:[EMAIL PROTECTED]) [tos 0x17] truncated-ip - 436 bytes missing!64.28.67.48.80 195.86.113.3.63822: P 1:1461(1460) ack 407 win 16060 (DF) This worries me a bit, so I put this in my /etc/ppp/ioptions: mru 1000 mtu 1000 But it doesn't seem to help, what is normal for ISDN connections anyway ? Or does someone think tcpdump just wrong or I got bad cabling (I wouldn't be surprised). tia, Lennie.
[OT?] Linux Volume Manager (was: Re: Disk Partitioning for Newbies (everything you always wanted to know))
On Tue, 19 Sep 2000, Nils Kassube wrote: snip Backup everything. (You do backups, don't you?) Re-partition your hard drive. Have more fun :-) With kernel 2.4 soon, you won't have to really repartition anymore... you can just resize them and create new ones... and so on (on the fly, no reboot, no downtime) (it's called LVM, Linux Volume Manager). Website: http://linux.msede.com/lvm/ There will be some GUI tools out there, that you can use to do all that as well. snip I hope this was helpfull.
Re: Off Topic - putty, RH, debian, title changes upon login
is happening. Under a friends RH box when I ssh into it from PuTTY it will change the titlebar to [EMAIL PROTECTED]: /directory. That's very cool, however, when I ssh into my debian box it just puts up domain.com - PuTTY. Does anyone know why/how this is happening for the RH box or how I could enable this type of behavior for the Debian box? It's rather handy to have a short PS1 and have the full path that your in reflected in the titlebar. It seems like RH sets some kind of environment setting. Although I don't know which one. What you can do is (I think), try and telnet to that RH box... then telnet to localhost and press the escape character (probably control+]), and then type environ list, or some other environ command, this will tell you what environment setting is made. How you can enable such stuff, I don't know (when you do, let me know ;). tia, Lennie.
Re: Starting KDE2
On Sun, 17 Sep 2000, Mike wrote: Hi everyone, Hi. I managed to get kde2 install but it wont start. I used apt-get to get it so it hould be fine, I'm using gdm so i added a menu to that which executed startkde but when it starts it sais that it failed interprocess networking and that dcopserver isnt running. Konqueror works fine if i I've been having the same problems. run it from inside gnome but i would like to have a play with kde2 so has anyone got any ideas?. No such luck yet. I've tried to compile it from source first, this did not work either. Although this is not the exact same problem, they fail at different things, so it seems to me. I think this means there are dependencies not met, but they are not listed in the .deb-files. I've not found out which ones yet. I've been very busy looking at the lists.kde.org and trying things, but it hasn't helped me yet. Thanks I wish. :)
Re: Starting KDE2
On Sun, 17 Sep 2000, Stephan Hachinger wrote: Hi! Don't know if this has got anything to do with your problems, but on my machine, everything runs fine. I've got a potato distro; kde 1.93 and qt 2.2.0beta2 are compiled from source, I've compiled qt with the ./configure-switches mentioned in the kdelibs package, file compiling. The only problem was that startkde was not really started by x, although it was the first line in my window-managers list, but after I created a symlink named /usr/bin/x-window-manager pointing to /usr/local/kde/bin/startkde, everything worked. Has anyone of you experienced the last problem? It semms to me like I'm doing something wrong and my solution is a quick and dirty one!? Kind Regards, Stephan Hachinger Well, I know this one, the official way is through the update-alternatives command (some examples): # /usr/sbin/update-alternatives --display x-window-manager x-window-manager - status is auto. link currently points to /usr/bin/kde2 /usr/bin/gnome-session - priority 400 /usr/bin/sawfish - priority 20 /usr/X11R6/bin/wmaker - priority 7 /usr/X11R6/bin/afterstep - priority 2 /usr/X11R6/bin/9wm - priority 6 /usr/bin/X11/wmaker - priority 400 /usr/bin/kde2 - priority 800 /usr/bin/startkde - priority 800 Current `best' version is /usr/bin/kde2. * /usr/sbin/update-alternatives --install /usr/bin/x-window-manager windower /usr/bin/gnome-session 1100
Re: Problems with the modules
On Sat, 16 Sep 2000, Aaron Brashears wrote: My machine developed the same problem after a kernel recompilation. I was never able to figure it out, and eventually *gasp* re-installed the entire system, which fixed everything. At least these days it's reinstall rather than reformat, reinstall... Your kidding me right ? Ooh your a former windows user ? ;) Just kidding. I personally would never do that, but yes I can understand it. You probably ran update-modules or an other similair tool. Well, I looked around and I found the bad line in /etc/modutils/patchs (first line): path[boot]=/lib/modules/ I changed mine to: path[boot]=/lib/modules/boot/ although maybe I should do boot per kernelversion, but I use the /etc/modules anyway. Also I needed to change more in that file to point to the right modules. I think they changed it because with kernel 2.4 they completly turned the modules directory structure around. What I don't understand is, how come it works, when you reinstall. Maybe there's a bug in the upgrade script (is there such a thing ?). OK, I think it's time to bug report about this. Although I've also seen a year and a half old bugreport, which is about the same as this would be. If anything has the errors writen down or something like that. Maybe it could be helpfull, let me know. Or has other more usefull information, do tell me/the list. Hope this helps and tia, Lennie.
Re: Off Topic: SCSI Recommendations?
snip I haven't run into any problems with my onboard U2W 7xxx chip, so... So have I, or haven't if you like. :) snip Anyhow what you want to look for is AHA/AIC 7xxx devices. Linux supports most (all?) of them. If you look at the 2.4 TODO list it says non-PCI cards are not yet supported or something along that line, but I'm sure you are gonna be using PCI anyway so. snip HardDrive 2 seagate ST39175LW barricudda's which are 9.1Gb ultra2 wide scsi 7200 rpm drives. I was figuring one for linux and one for windows (gotta game fellas). Good, fairly fast, not exactly cheap, but I'm planning on getting a couple of 18GB 'cudas. I dunno what your gonna do, but I think it would be a good idea to split those drives in 2 parts, first part for windows second for Linux. That would give you more speed I would think. But then again, some guru may come and tell me otherwise. And I don't wanna scare you or anything, but 2.4 used to have a small problem with Barricudda's I think, they trashed the fs. But it's been resolved months ago. I personally use IBM drives, but I guess the Barricudda's are cheaper (?). I haven't had any problems with them (I actually heared of one fellow who dropped his IBM HD down the stairs by acident they still worked, although this is hearsay). I've run all kinds of pre releases of 2.4, although not the last ones (because they have a different modules directory layout, I think I need to update my modutils first, I'm still running an unstable now). Hope all this helps, Lennie.
[OT] Switching kernels without reboot (was: Re: Debian vs. Red Hat)
btw: Is it possible to switch kernels without rebooting. I cannot believe that's possible Actually, I think someone was working on that. Well, first he wants to make it so you can build an other kernel in userspace or something (this is already possible with special kernels). I think Solaris can actually do that, update a running kernel.
[OT] Switching kernels without reboot (continued) (was: Re: Debian vs. Red Hat)
I've not been at the OLSymposium but I did remember reading this about Werner Almesberger: http://www.ottawalinuxsymposium.org/2000/booting.html: Booting Linux: The History and the Future The IA-32 Linux boot process has since 1991 evolved from using boot floppy to Shoelace and now LILO. We will discuss the history of the boot process and some of the features intended for the future. We will cover new features such as the ability to launch a new Linux kernel within Linux without resetting the system as well as numerous other new features to enhance the IA-32 boot process. We will also cover some commonly misunderstood current features including initrd and big image support. Werner Almesberger, Ph.D. Werner received his master's in Computer Science from ETH Zurich in 1992. From 1993 to 1994 he developed ATM switch control software at the IBM Zurich Research Lab. He completed his Ph.D. in Communication Systems at EPFL (Ecole Polytechnique Federale de Lausanne) in 1999. He became addicted to Linux with kernel version 0.12 in 1992, and quickly became involved in very interesting kernel work including; file systems, device drivers, and the LILO bootloader. His ongoing involvement includes; the LILO bootloader, the psmisc tools, ATM and Differentiated Services on Linux, as well as a bit of hacking on the Linux-7110 project bringing Linux to PDA such as the Psion palmtop. There is also a link to his (Institute ?) website: http://icawww.epfl.ch/ and his own page is at: http://ica1www.epfl.ch/~almesber/ There he has his ToDo list (http://icawww1.epfl.ch/~almesber/todo.html), of which one of the items is: write code to boot Linux kernel from within Linux (2-4 days) Well, this was an optimistic estimate. It's now more like 1-2 weeks, after getting the hairy struff done. This seems pretty fast to me, anyway. I hope he'll get around to do it soon. ;) See ya, Leen Besselink.
Re: [OT] Switching kernels without reboot (was: Re: Debian vs. Red Hat)
On Thu, 14 Sep 2000, Jason Quigley wrote: This may be of interest: http://www.gnu.org/software/hurd/hurd.html Thank you but I already knew of it's existance, personally I think a little like Linus Torvalds, monolitic kernels are always gonna be faster then Microkernels. Although a mix of things put together would probably the better approach (I also personally think, splitting up more parts of the Linux kernel would do that). just my thoughts AND no let's not discous this _on the list_. ;) Lennie.
Re: X, GNOME, and Debian
use GNOME on top of X, the screen disappears into a bunch of little stripes, and the keyboard locks up (even alt-ctrl-backspace doesn't work). This was with the GNOME 1.0.53 from Debian. I installed Helix (GNOME 1.2), same problem. I've done XF86Setup, both putting the Happens regardless of windowmanager - sawfish, afterstep, windowmaker. did you try using a windowmanager without using GNOME ? Did this solve the problem ? (just trying to clear things up here, I don't think I could help you with this problem to be honest). Did you try using a different resolution, perferable a lower one ? Hope this helps make things clearer, Lennie
Re: fsplit
On Wed, 13 Sep 2000, Jeff Woodford wrote: Hi all, Would anyone happen to know in which package I can find the fsplit (Fortran split) utility? Thanks, -Jeff Woodford if you already have it installed, you can find it this way: dpkg -S /usr/bin/filename or you can try looking it up on the website (although this does put some load on the cgi.debian.org). At the end of http://www.debian.org/distrib/packages (or a mirror there of) it says: Search the content of the Latest Release. There you can search for a file by name(part) (with/without path). I just did this for you and this is what it came up with (it doesn't look to helpfull to me, I hope it's more usefull to you): FILE PACKAGE __ usr/bin/tiffsplit graphics/libtiff-tools usr/share/doc/scansort/fsplit.txt graphics/scansort usr/share/man/man1/tiffsplit.1.gz graphics/libtiff-tools
Re: Internet Cafe
On Thu, 14 Sep 2000, ChrisHellberg wrote: I want to set up an internet cafe at a hostel and am investigating various ways of going about things. I think windows 2000 would be the snip down the linux path, what would be the best way to log a workstation out when credit expires? I rekon it's a matter of creating a simple TCP/IP connection to a server and listening to logout requests :/ Or keep a local program running to do that, it would seem more usefull to me. Or if you buy a (few) really fast server machines, you could run X on that and have cheapo clients, I would expect it would be more easy to centrally administer those (could be cheaper too). But then again, I've never tried that. just a few thoughts.
Re: Sparc IPX mouse issue
On Wed, 21 Jul 1999, Mike Lieberman wrote: We are having problems getting Debian to see/find/use the Sun mouse connected to our type 5 keyboard. gpmcongif wants to find /dev/mouse which doesn't exist. Xwindows won't run for the same reason. We know the mouse is working as we have Solaris on this box and the mouse was just fine. Anyone using a Sparc IPX have any ideas? well, I can tell you... /dev/mouse is one most systems... maybe all, I don't know... just a like to the real device... and IPX is ps/2 right ? maybe it's: /dev/ps1 ? although I'm not much of a Sun user... hope this helps somehow. /* Mike Lieberman[EMAIL PROTECTED] */ /* President*/ /* Net Wright LLC */ /* http://www.netwright.net */ /* Voice and Fax: 307-857-1053 */ -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: debian
http://www.toms.net/rb/ is what you are looking for. it's the best in town. IMHO. On Sun, 18 Jul 1999, tyler spivey wrote: as most of you know from irc, i am a blind user using a terminal. thats why i need a linux boot disk with module support 2.0.36 so i can test a driver on my old 386 without overwriteing dos. [EMAIL PROTECTED] SDF Public Access UNIX System - http://sdf.lonestar.org -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: Back Again - CDRom
On Sat, 17 Jul 1999, Artur Correia wrote: OK, As expected, ran into some more problems: #1. How the hell do i see the contents of a cdrom? I've tried in the shell mount /dev/hdb and mount /dev/cdrom, but i got a message saying that that file was not in the file fstab or mtab... and, in fact, it isn't. The starnge part is tha if i run dselect with the cd rom with the debian distribution the program accesses the cd and install all the packages from it? So, what do i do? Desperately waiting for a solution, AC Tks mount [-fnrsvw] [-t vfstype] [-o options] device dir -- from the mount manpage. did you see this: ^^^ you need to have an entry for that in the /etc/fstab file or you need to specify the mountpoint, which is just an empthy dir. did I make sense to you ? also, try reading the whole man page, I also think there is a manpage about fstab. Hope this helps... it should :)
Re: Linux LAN probs - a shaggy dog story?
yeah... maybe something like half dublex - full duplex... but then again... mostly this causes no connection at all... I dunno... maybe it's a bad cable. It very much sounds like hardware to me. Although a bad driver could be the cause. What kind of card is it ? On Fri, 16 Jul 1999, Jens B. Jorgensen wrote: This sounds more like a hardware problem than a config problem because _some_ of your pings are making it through. If it were a config problem I would expect you to get _no_ ping responses. Patrick Kirk wrote: Hi all, I have two working Linux boxes now. cat has IP 10.0.0.1 and dog has 10.0.0.2. When I reboot cat it can ping dog but reports over 50% losses. Likewise dog and ping cat with close on 50% losses. If I tru to telnet from dog to cat, I get the cat login prompt and then the system freezes as I enter a username. From that point neither machine can ping the other. I also have a NT box. If I put it on instead of dog, cat can ping NTbox. But NTbox cannot ping cat. Request times out. cat is my default gateway to the Internet, tho obviously is as useful as a chocolate teapot right now. Can anyone see what I'm doing wrong? IDEA: is it an ipmasq problem? Patrick Wise Chinese Proverb: If tired of computer winning at chess, try it at kick-boxing instead -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Jens B. Jorgensen [EMAIL PROTECTED] -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: redirect logs to tty?
try this, you will need to have syslogd reload it's config, after changing this (kill -1 `cat /var/run/syslogd.pid`). In /etc/syslogd: # send all logging information to /dev/tty8: *.* /dev/tty8 it's too bad apache has it's own logging mechanism, other wise, the logging would also show up there. But pretty much everything else goes there. Hope this helps you, Leen. On Wed, 14 Jul 1999, andreas [iso-8859-1] pålsson wrote: Hello. My question is, if it is possible to get the logging-information (syslog,messages) to one of the virtual consoles instead? There's 6 of them and I only need 3 at most, and it would be nice to see log-information by just pressing Alt+F6.. Any solution or reference to documentation are welcome. Regards... -- == andreas pålsson == [EMAIL PROTECTED] == -- -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: multiple IPs on one netcard
On Sat, 10 Jul 1999, Paul Miller wrote: Is it possible to a network card configured to use multiple IP address (ie, 4 static + 4 dynamic = 8 IPs) at once? All of the IPs are on the same network and netmask. If this is possible, where can I find more information on how to do this? Thanks -Paul This is called IP-aliasing. Try reading this document: http://metalab.unc.edu/pub/Linux/docs/HOWTO/mini/IP-Alias btw.. it's always good to look at the documentation at: http://metalab.unc.edu/LDP/ (there are also mirrors) rather before asking on a list, but before. :) hope this helps a lot, Leen. PS I think you can also look at: www.linux-howto.com (it's a lot easier to remember and it take a load of the other documentationservers).
RE: multiple IPs on one netcard
On Sat, 10 Jul 1999, Paul Miller wrote: Is it possible to have Linux split up the bandwidth automatically on aliased IPs? hmm... I think the question is, do you really want to ? I think it would be better to have the networkcard do as much as possible, and not leave bandwidth unused. The HOWTO only covered static IPs, it is possible to use dynamics as well? Well, I'm not sure... I could try looking in to it for you. I'm gonna do it today. So at late, you will have an answer from me tomorrow. If I couldn't do it. That doesn't mean it can't be done, though. Because a lot of people are working on all sorts of things. Thanks -Paul -Original Message- From: Leen Besselink [mailto:[EMAIL PROTECTED] Sent: Saturday, July 10, 1999 9:41 AM To: Paul Miller Cc: Debian User Subject: Re: multiple IPs on one netcard On Sat, 10 Jul 1999, Paul Miller wrote: Is it possible to a network card configured to use multiple IP address (ie, 4 static + 4 dynamic = 8 IPs) at once? All of the IPs are on the same network and netmask. If this is possible, where can I find more information on how to do this? Thanks -Paul This is called IP-aliasing. Try reading this document: http://metalab.unc.edu/pub/Linux/docs/HOWTO/mini/IP-Alias btw.. it's always good to look at the documentation at: http://metalab.unc.edu/LDP/ (there are also mirrors) rather before asking on a list, but before. :) hope this helps a lot, Leen. PS I think you can also look at: www.linux-howto.com (it's a lot easier to remember and it take a load of the other documentationservers). -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
RE: multiple IPs on one netcard
Is it possible to have Linux split up the bandwidth automatically on aliased IPs? hmm... I think the question is, do you really want to ? I think it would be better to have the networkcard do as much as possible, and not leave bandwidth unused. The HOWTO only covered static IPs, it is possible to use dynamics as well? Well, I'm not sure... I could try looking in to it for you. I'm gonna do it today. So at late, you will have an answer from me tomorrow. If I couldn't do it. That doesn't mean it can't be done, though. Because a lot of people are working on all sorts of things. I tried it out, I found a moment to do so. As I have a cable modem (with dhcp). I tried to do so it doesn't work. I think it's because the dhcpcd that I use, tries to get hardware control to the card and not the logical device. Maybe this is wrong, but maybe it has to, maybe it's possible, I don't know. If you really need it to be done, you could ask any good programmer to add it to Linux (if you got the money to pay him or her). But, I would first ask some extra people first. :) Because, I too hate wasting money. Hope this helps. PS the maintaner of dhcpcd for Debian is: [EMAIL PROTECTED], he should know more about it. But, don't ask him, just wait a day or so, for people to answer this email, he's probably a busy man.
Re:
On Tue, 6 Jul 1999, Jack Wilkinson wrote: I'm having the same problem... just figured it was a limitation of the standard IP masquerading setup... I can recieve dcc's, but can't send... is anyone actually able to dcc send from behind a masq? - Original Message - From: The Buht Man To: debian-user@lists.debian.org Sent: Tuesday, July 06, 1999 3:34 AM i cannot get my DCC to work behind 2.1 debian , linux 2.2.10 .. i have the irc masq module loaded , ip masqing stuff is loaded ... and i cant get a dcc to work going throught he linux box to my windows machine :( I have not tried this, but there is an ident, that supports masq. (Because, that is your problem) If an dcc request is done, the 'server' will do an ident request, if it doesn't get an answer or a wrong anwser it just waits or stops. There is an ident that does support masq. I think it's called midentd. (try looking it up at http://freshmeat.net) I have not had the time to try it out, but it should work. If some1 can try it out, it would be great. Because, I would like to know. tia, Leen.
Re: Using gnome-apt as non-root
On Sun, 27 Jun 1999, Havoc Pennington wrote: On Sat, 26 Jun 1999, Stuart Ballard wrote: 1) Is there a way to run programs as root from X, respecting such things as the current GTK theme; OR No, the theme is a per-user setting, so it won't apply to you while you're root. You could set the theme for the root user though. 2) Is there a way to enable a regular user to run apt/dpkg/gnome-apt? No, installing packages requires root permissions. gnome-apt might let you do package selections as a regular user eventually, but you'll always have to be root to actually install. Maybe it's a reason to change it's behavour, maybe we could have it ask for a password and do sudo in the background or something, when installing. Just my 2 cents. Havoc -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: Problem with route and kernel 2.2.10
On Sun, 27 Jun 1999, Ralf Comtesse wrote: Hello, I compiled kernel 2.2.10 today and almost everything went fine. The only problem is when booting. In /etc/rcS.d S40network is called and there you find the line route add -net 127.0.0.0 wich gives me SIOCADDRT: Invalid argument This is now done autimagically, in 2.2.x kernels. also in some 2.1.x kernels ofcourse. Was there change that I did not get? Sth. else I missed? The network seems alrigt. How do I get rid of the error message (without deleting the line ;-)) ) well, uncomment it :) Regards Ralf Hope this answers your questions. Ralf Comtessee-Mail: [EMAIL PROTECTED] Baerwaldstr. 63 Tel: +49-30-28599230 10961 Berlin Fax: +49-30-28599231 Member of the Cetus Team. For OO links visit: http://www.cetus-links.org -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: How to set default resolution in XF86?
On Sat, 26 Jun 1999, Tracy Nelson wrote: Whenever I start X, it comes up with a default resolution of (I think) 640x480. I normally run at 1280x1024, which means I have to hit CTRLALTKP+ two or three times to select the correct resolution. Is there any way I can specify that as the default resolution? Is it a matter of just resequencing some lines in my XF86Config file? I'd also like to use a resolution of 1152x870, but that resolution wasn't provided under xf86config. Is there some way I can calculate the vertical and horizontal frequencies for this resolution and add them to my XF86Config file? 1280x1024 is a little much for my .31DP monitor... Thanks! -- Tracy Nelson Well there is a file /etc/X11/XF86Config the easy way to do this is just outcomment the lines you don't need in that file with a '#'. The ones you need to comment are: Subsection Display Depth 16 Modes 640x480 800x600 1024x768 ViewPort0 0 EndSubsection or just remove the one you don't need. Like: 640x480 or put them on the end of the line. You need to do this in section Device and in Screen. you probably have multiple sections screeen, you can do them all or do the one you think is the right one. Hope this helps. PS this is not the proper way, I think the proper way is using xf86config or XF86Setup, but, I don't use them for 'fine tuning'.
Re: Debian Router???
On Thu, 10 Jun 1999, J.H.M. Dassen wrote: On Wed, Jun 09, 1999 at 19:22:33 +0200, Leen Besselink wrote: it's probably easier to use the Linux Router Project http://www.linuxrouter.org/ . It's Debian-based if I'm not mistaken. Actually... I'm sure.
RE: where can i get the source for debian in a format i can read
On Thu, 10 Jun 1999, Shaleh wrote: On 10-Jun-99 matthew lamb wrote: where can i get debian source code for the kernal so i can port it to another platform?? There is no debian source code. Debian is based on the Linux kernel. try ftp.us.kernel.org, look in /pub/linux/kernel. What hardware are you porting to? Try looking here for ports that are already being/have been done: http://www.ctv.es/USERS/xose/linux/linux_ports.html
Fw: Re: Hope you don't mind this (friends!) (fwd)
-- Forwarded message -- Date: Thu, 10 Jun 1999 11:59:03 +0200 From: Maik Bakker [EMAIL PROTECTED] To: Tim Stahlberg [EMAIL PROTECTED], Sheri Hudacin [EMAIL PROTECTED], Mette Poulsen [EMAIL PROTECTED], Melinda Fetcko [EMAIL PROTECTED], Leen Besselink [EMAIL PROTECTED], Gundel Hacker [EMAIL PROTECTED], Dave Kathy Hinshaw [EMAIL PROTECTED], Berit Hansen [EMAIL PROTECTED], Annett Beinling [EMAIL PROTECTED] Subject: Fw: Re: Hope you don't mind this (friends!) - Original Message - From: Jorge V.Sánchez Compte [EMAIL PROTECTED] To: [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Tuesday, June 08, 1999 11:31 PM Subject: Fwd: Re: Hope you don't mind this (friends!) From: Pilar Salcedo [EMAIL PROTECTED] To: [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED] Subject: Re: Hope you don't mind this (friends!) Date: Tue, 08 Jun 1999 17:26:25 PDT MIME-Version: 1.0 Received: from 209.14.120.3 by wy1lg.hotmail.com with HTTP;Tue, 08 Jun 1999 17:26:25 PDT X-Originating-IP: [209.14.120.3] A ball is a circle, No beginning, no end. It keeps us together Like our Circle of Friends But the treasure inside for you to see is the treasure of friendship You've granted to me. Today I pass the friendship ball to you. Pass it on to someone who is a friend to you. Ok, this is a test to see how many friends you have !!! Don't be hurt by this if no one sends it back, it's just a test. Ok, here goes, instantly when you receive this letter, you're requested to send it to at least 10 people, including the person who sent it to you. __ Get Your Private, Free Email at http://www.hotmail.com __ Get Your Private, Free Email at http://www.hotmail.com __ Get Your Private, Free Email at http://www.hotmail.com __ Get Your Private, Free Email at http://www.hotmail.com __ Get Your Private, Free Email at http://www.hotmail.com __ Get Your Private, Free Email at http://www.hotmail.com
Re: Unwanted Graphical Login and other woes...
On Tue, 8 Jun 1999, Jean Pierre LeJacq wrote: On Wed, 9 Jun 1999, Andrew J Fortune wrote: I don't know what I have done, but Linux (using slink) is now booting up to a graphical login. This is not what I want at the moment, and I was wondering if anyone knew what the problem might be ? this is not normally a problem, but I am trying to resolve other problems associated with startx and file resolutionand this is not allowing me to test some things. You probably have xdm installed. You can either remove the package with dpkg or disable the automatic startup by removing the symbolic links in /etc/rc[0-6].d/S??xdm. To do this with the login screen, first switch to another virtual console using the Alt-F? where ? in 1, 2, 3, ... actually it's CTRL-ALT-F?...where? in 2, 3, 4, ...
Re: Debian Router???
On Wed, 9 Jun 1999, Anthony Landreneau wrote: Greetings, I am building a debian router, however I prefer to use a FLASH ROM card instead of a hard drive. The largest I can afford is a 60 meg card. Can I set up a debian router with all the drivers (three cards two DS3 and one Gigabit), VI as an editor (don't hate me because I like VI ok) and keep it running under 60 megs? Anyway have something like this running? Share the info? Would be appriciated! Thanks. Anthony Landreneau DoD Network Security Administrator Infinity Data Systems New Orleans Louisiana (504)455-8973 You can but, it's probably easier to use the Linux Router Project, for this, with some extra's or something. Whatever you want.
Re: getting around a firewall?
others -- ssh is most certainly blocked. I want to be able to securely connect to a remote server (Debian based) in shell. I'm told one way to do this is to tunnel ssh proxy through other port like http. What are other ways to achieve the end result? I think I heard tell that there is something in java that allows secure connections from httpd (right from a web page). well... let's see.. I think Java would use the standard port for ssh too... couldn't you just put sshd behind a different port..? just set it up as from inetd... and change the portnumber... how hard can it be ? (like http... or ftp... or something else... that isn't blocked) hope this helps.
Re: OFFTOPIC: SDRAM PC100
On Wed, 2 Jun 1999 [EMAIL PROTECTED] wrote: The question is: if I buy a SDRAM PC100 chip, will it work on a computer with bus speed of 66Mhz ? Nope. Although there should be still some of this older stuff around. try: www.pricewatch.com hope this helps.
Re: [s2192888@cse.unsw.edu.au: Re: Can you help me with scsi in debian please]
I think the module name is general scsi support. Hope this helps you. Lennie. On Sun, 9 May 1999, Shao Zhang wrote: Hi, My anwser to this question is to compile the scanner driver as the module. But I am not sure. Any other suggestions?? Thanks. - Forwarded message from CyrusP [EMAIL PROTECTED] - From: CyrusP [EMAIL PROTECTED] To: Shao Zhang [EMAIL PROTECTED] Date: Sun, 9 May 1999 14:50:48 +1000 (EST) X-Sender: [EMAIL PROTECTED] Subject: Re: Can you help me with scsi in debian please In-Reply-To: [EMAIL PROTECTED] New problem, (i) If I boot up kernel with scanner power on, then kernel can find it and if I check /proc/scsi/scsi, it shows that it detects the scanner and when I try scan everything is cool. (ii) If I boot up kernel with scanner turned off, it doesn't detect. If I check /proc/scsi/scsi, it doesn't show scanner. My question is how can I turn my scanner on after I boot up and have it be detected. Thanks Cyrus - End forwarded message - -- Shao Zhang - Running Debian 2.1 ___ _ _ Department of Communications/ __| |_ __ _ ___ |_ / |_ __ _ _ _ __ _ University of New South Wales \__ \ ' \/ _` / _ \ / /| ' \/ _` | ' \/ _` | Sydney, Australia |___/_||_\__,_\___/ /___|_||_\__,_|_||_\__, | Email: [EMAIL PROTECTED] |___/ _ -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: X over network
On Thu, 29 Apr 1999, Paul Nathan Puri wrote: What client software do I need to run X over the network? What are the hardware requirements? Is there a good HOWTO on running X over networks? yes, there is a mini-howto, it's called: Remote-X-Apps Thanks. NatePuri Certified Law Student Debian GNU/Linux Monk McGeorge School of Law Sacramento, CA [EMAIL PROTECTED] -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: samba/network neighborhood question
I think I know... I think you havd samba running from inetd, so when win95/98 issues a request... the samba server doesn't answer right away (it needs to be started first). That would be like starting apache from inetd, it takes a little time to get the first response... so what you need to do is...: 1. you can make samba active... bye, typing the smbname or ipaddress in the startmenu/search/computer... so samba get's started... or: 2. what would be better... make samba a daemon and not start from inetd. just my 2 cents, Leen. On Fri, 23 Apr 1999 [EMAIL PROTECTED] wrote: Funny..Just my experiences on the samba subject. I've recently moved my network to a new apartment. Worked perfectly for a while, then my single win98 machine began having troubles connecting to the samba server. Sometimes the server wouldn't show up, sometimes the 'network' would be down, sometimes it couldn't browse the network at all. And through it all I could still access the internet from the machine. (Samba box also the network gateway) I actually had to threaten to reinstall windows on the win98 box before it started working again. No settings tweaking or rebooting helped. I actually had to go look for the installation CD to make it work again. I still have no idea what the 'problem' was or how it was 'fixed'. On Thu, 22 Apr 1999, Ben Frame wrote: I just got Samba installed and it seems to be working fine. But it doesn't always show up in my Network Neighborhood under Win95. Both the Debian machine and Win95 machine are on the same subnet and both are in a workgroup called linux. I've made several changes to my smb.conf file, and consequently restarted the debian machine a few times. Sometimes it shows up in Network Neighborhood, and sometimes it doesn't. Sometimes it will show up later, but not immediately. So my question is, what makes it show up (or not) in Network Neighborhood? I can see the Debian machine if I do a search for it with Win95 (start/find/computer/name). And for now I've just placed a shortcut to it on my desktop to keep from having to search every time. If anyone can shed any light, I would certainly appreciate it. Ben Frame [EMAIL PROTECTED] -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: Is ssh 2 incompatible with ssh 1?
Well, let me say what I did to get it working: I installed ssh1 first... with deamons and all... then I installed ssh2... over that. then edited /etc/ssh2/ssh2_config set this to yes, when I found it in the manpages: Ssh1AgentCompatibility yes so, now when a ssh1 client tries to connect to my system... the ssh2d(aemon) will start the ssh1d(aemon) with the options he himself got. I hope this helps you too. Leen. (Well actually, I already had ssh2 installed... but just installed ssh1 over it and ssh2 over that again... and had to make some small changes in /usr/local/sbin/, something about a link... I think that's also in the manpages). On Fri, 23 Apr 1999, Lazarus Long wrote: On Friday, April 23, 1999 at 10:06:19 -0400, Arcady Genkin wrote: From: Arcady Genkin [EMAIL PROTECTED] Message-ID: [EMAIL PROTECTED] X-Mailer: Gnus v5.6.45/XEmacs 20.4 - Emerald Sender: Arcady Genkin [EMAIL PROTECTED] I've installed ssh 2.10 and tried to connect to my school via ssh (I know that they run ssh protocol) and got the following: bash-2.01$ ssh cdf.utoronto.ca Disconnected; protocol version not supported. They are using vershio 1, AFAIK. Does this mean that v.2 is incompatible with it or do I have smth misconfigured? May I suggest bash-2.01$ ssh -v cdf.utoronto.ca for more complete information on which ssh/sshd and protocol versions are being used? BTW, it *is* possible to connect to a ssh2 system using ssh1, if using recent versions. I do this with my ssh2-using-ISP regularly, using ssh1 here (because I dislike ssh2's licensing.) (And I connect in the other direction as well.) -- PGP Public Key available on request: Type Bits/KeyIDDate User ID pub 1024/CFED2D11 1998/03/05 Lazarus Long [EMAIL PROTECTED] Key fingerprint = 98 2A 56 34 16 76 D5 21 39 93 99 EA 89 D4 B5 A2 -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: W95 defrag
On Thu, 22 Apr 1999, H C Pumphrey wrote: Greetings, fellow Debian fans, This is only a proto-debian question, I'm afraid, but I have tried to RTFM, honest. I'm trying to defrag the disc on a W95 laptop prior to using FIPS to re-partition it so I can put Debian on it as well[1]. W95 defrag will move a lot of stuff (which it colours turquoise) to the beginning of the disc, but (a) it wants to put some things at the end and (b) there are some things which it colours red/white and won't move at all. well, I don't know if your using the latest version... but, if you are... (maybe earlier versions do the same)... fips will let you make a backup of your fat(32) AND very importantly it will let you know till what point the disk is used... (do you understand what I mean ?, hope so for you) so, you can try running it... and it will tell you if you can make use of as mcuh diskspace as you want to. also, mostly the files that defrag leaves lying around... are the mirror and image files... you would be able to spot those in the C:\ dir. Hope this helps... --- snip ---
Re: Whoops
On Sun, 18 Apr 1999, Richard Harran wrote: I did something really stupid: #mv /usr/bash somewhere_else (don't ask). Then I exited root, and (of course) I can't log in as root to fix it. I'm still logged in as a normal user, but anything using a script with /bin/bash or /bin/sh doesn't work. Does anyone have any ideas on how to fix this. PS I don't think I have a rescue floppy Thanks Slightly Sheepish: Rich Do you have something like sudo or su installed ? other wise... you can maybe make a tomsrtbt disk... and restore it from there... (you can make those from DOS too... so) I'm out of ideas now... -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
