Re: some mutt stuff
On Mon, Feb 12, 2001 at 03:45:51PM -0500, David Merrill wrote: | On Mon, Feb 12, 2001 at 09:30:21PM +0100, Willi Dyck wrote: | | Let me know if you want to see my newsig.sh script also and I will | send it. I'd like to see it. Thanks, -D
Re: gnome-terminal question
On Mon, Feb 12, 2001 at 02:19:20PM -0800, [EMAIL PROTECTED] wrote: | | hi | | | i was curious what needs to be done to get a remote host | (connected through ssh) to display the hostname and directory | in the titlebar of a gnome-terminal window. redhat systems | seem to do this but none of my debian systems do. | I'm not sure what you mean about the RH systems -- I used to run RH and it would normally show the CWD. What I had done was make some launchers (shell scripts work as well) for each of the hosts I would connect to and use the command line to set the title. (I think the opiton is -t) I think there is some (shell) command that can be used as well (from the remote side) because I've seen the title change on me before (usually to a non-useful title like "~"). Sorry, I don't know what the command is. -D PS. My computer is dead right now so I can't try anything with Debian to verify my suggestion.
Re: email-address embedded in homemade kernels
On Mon, Feb 12, 2001 at 11:31:57PM +0100, Diego Biurrun wrote: | Hello! | | I was wondering if it is possible to embed my email address into | homegrown kernels. My current kernel greets me with | | | Linux version 2.2.19pre6 ([EMAIL PROTECTED]) | | | but I would much prefer something like | | | Linux version 2.2.19pre6 ([EMAIL PROTECTED]) | | I don't think it is hard-coded into the kernel. I believe it is specified in a file in /etc (that gets read on boot). I know that if you change the name of localhost in /etc/hosts and reboot your greeting will show the new host name. BTW, Is there a way to "reset" the hostname without rebooting? I had a problem with this a while back -- I called my machine by one name, but the DNS server called it by another. The result was that no one but my own machine could browse my web pages under user's home directories. (apache did some rerouting to the hostname the kernel thought it was, but then other machines couldn't lookup the name) Kind of fun, huh? ;-) -D
Re: Here you have, ;o)
On Tue, Feb 13, 2001 at 11:46:43PM +0800, Shell Hung wrote: | Joris Lambrecht wrote: | > | > Fortunatly this is a linux mailinglist ... | > | But I also think there're someone is using M$ platfrom when receiving | this email, and maybe some of them using OutLook... :-) Mutt compiles and works very nicely with cygwin :-) -D
Re: A debian-based distro for the New
On Tue, Feb 13, 2001 at 03:44:45PM -0500, [EMAIL PROTECTED] wrote: | I got a few friends who are interested in (trying)linux. For TRYING purposes I want them to use a debian based Linux. I love using debian 2.2 | and I am pretty new my-self(1.5 year). I think it might be a little to | overwelming for them. I beleive there are three to chose from Corel, Storm | and Libranet. | | I have tried the first two but Libranet I have not tried yet. Anyone tried Libranet? Are there any reasons not to use the others? I wouldn't want the users to try Linux and despise it. I haven't used any of these 3. I have heard (check the Debian Weekly news from last week, or the week before) that Corel is selling off their dist, and Stormix filed for bankruptcy. I have a friend who uses Corel, and he says it is very easy to configure and is good for people who don't want to learn vi and the FHS to configure the system. I would recommend NOT using RH7.0. No use giving new users (or anyone for that matter) a broken compiler and libc. (This began the end of RH for me :-)) If they want to try upgrading stuff, then apt is the best experience they can have. -D
Re: Wake Up on LAN
On Wed, Feb 14, 2001 at 04:56:10PM +0100, Barthazi Andras wrote: | Hi! | | I should have to write a Backup system at my | company, and I need a program can wake up | the workstations. I know that I have to send | the the "Magic code", do you know a program | can solve it for me? And is it true, that | I can just wake up a machine that suspended | before (not a machine turned off -- with a | motherboard under power)? | I don't know much about WOL, but my ethernet card (LinkSys) has WOL capability. From what I gather, the ethernet card gets plugged into a special plug on the motherboard (there is a cable on the card). When the ethernet card gets some signal, it sends the appropriate signal to the motherboard, which then wakes up the system. I don't know what the signals are or how to config the system but I do know that the motherboard must support it. -D
Re: necesito información sobre Hurd
On Thu, Feb 15, 2001 at 01:24:58AM +0100, wolfman wrote: | |La cuestión es donde puedo encontrar el kernel para bajarmelo??? | I don't understand much Spanish (no hablo mucho espanol), but maybe this will help : http://www.debian.org/ports/hurd/ -D
Re: Fetchmail & Sendmail
On Fri, Feb 16, 2001 at 10:46:10AM -0800, Rob Hudson wrote: | I'm using the fetchmail-procmail-sendmail combo to get mail off of a | mailserver, filter it, and read it locally. I'm using that same mail | server to pass sent mail to. | | In this case, I don't know if sendmail is overkill and if I should be | using something different. For sending, ssmtp works (check the FAQ on www.mutt.org to find a link). ssmtp is like a sendmail replacement, but all it does is forward the mail to a smpt server. No local delivery, etc. | | What I'd like to fix is this... | | When fetchmail gets mail, I think it passes it to sendmail which | checks to make sure the domain exists. Today, my DNS servers listed | in /etc/resolv.conf were down, and all the mail I fetched got bounced | and lost. I'd like for this not to happen and get any and all mail | that is in my box no matter if it's spam and has an unresolvable | domain or not. How can I set this up? | You might be able to tell fetchmail to hand the mail off to procmail directly. If not you can probably have fetchmail just dump the stuff in your inbox, then use the bash script in the procmail man page to filter the mail from the inbox. HTH, -D
Re: Newbie
On Fri, Feb 16, 2001 at 05:51:29PM -0500, Jimmy Kaplowitz wrote: | First change all references to "stable" in your /etc/apt/sources.list to | "woody". Then Shouldn't that be "testing" ? -D
Re: jre problem
On Wed, Feb 19, 2003 at 10:54:10PM +0800, wujf wrote: | | Unable to initialize threads: cannot find class java/lang/Thread ^^ | Could not create Java VM | Here's your problem. You said you were using jdk 1.1.x, so you need to include classes.zip in the CLASSPATH. The problem is the VM (java) can't find the standard library stuff. If you use a newer jdk (1.2.x or 1.3.x) it will find the file automagically (it is called rt.jar in the newer releases). classes.zip should be located somewhere like /usr/share/java/lib/classes.zip. I haven't installed a jdk on Linux yet as I have a Win2k box at work. On it I have jdk1.1.8, jdk1.2.2, and jdk1.3. I installed them in d:\jdk1.1.8, etc. I put "d:\jdk1.1.8\lib\classes.zip" in the CLASSPATH environment variable, or on the commandline as java -classpath "d:\jdk1.1.8\lib\classes.zip" HTH, -D
Re: LILO on my windows partition
On Tue, Feb 20, 2001 at 11:52:44AM -0600, John Hasler wrote: | The Doctor writes: | > Like an idiot, I accidentally wrote a LILO boot into my windows | > partition, so nw booting from my LILO to windows (on /dev/hda3) starts up | > another LILO that croaks with LI is there a way to be able to boot | > windows again? | | Run 'lilo -u /dev/hda3'. This tells lilo to put the mbr back the way it | found it. This is a good lifesaver. I had to install RH6.2 on a machine at work (for testing our product) and it didn't ask, it simply overwrote the mbr. Then I tried grub. Get it from www.gnu.org, download the grub-0.5.96-1-i386.ext2fs file. Then copy it to a floppy using dd or rawrite and reboot. Grub is a piece of cake to get working. It boots both Linux and Win2k out-of-the-box without any trouble. I had never used grub before, but it only took about a minute before I had both systems booting cleanly. The kernel is on the hd _above_ the 1024 cylinder too. You can configure grub as you are booting (unlike lilo which requires you to boot a system first) and then set /boot/grub/menu.lst to however suits you. (just note that grub uses zero-based indexing, so /dev/hda3 is (hd0,2) to grub, you'll understand when you look at the menu.lst file or boot the sample one) I will never use lilo again, grub is just too cool! -D
Re: Debian 2.2 and Linksys LNE100TX - problems
On Fri, Feb 23, 2001 at 11:49:47PM -0700, John Galt wrote: | On Fri, 23 Feb 2001, Jason Price wrote: | | >I will be compiling a new kernel to add SMP support - when I do, is | >there anything special I will need to do to make sure the NIC works? | | Get the newest source possible. Tulip (to wit the LNE) was flaky through | most of the 2.2 series. IIRC they fixed it in the 2.2.18 area. I know | that 2.4 works wonders on it. | Where did you get this information? I have been using a LinkSys LNE100TX for a couple of years now with no problems. Using kernels 2.2.12, 2.2.16, 2.2.17, 2.2.18. The tulip driver that came with 2.2.12 kernel was version 0.90 I think. The one on the floppy in the box was 0.91 (or it was the other way around). I recall reading on Donald Becker's web site(s) very recently that 0.93 would be released soon and that 0.92h was the current latest release. What *revision* of LinkSys card are you using? I had Revision 2.0. I tried to set up my card temporarily in a friend's computer and had trouble installing the driver (Win98). We searched on LinkSys' website and found that there are newer revisions of the card, up to 4.0 I think. I have also read that LinkSys first used the DEC made chip on the cards, then a Lite-On made chip (I think this is what mine uses) and now uses their own chip. All of these use the tulip driver. -D
Re: Network card
On Sat, Feb 24, 2001 at 09:23:58AM -0500, [EMAIL PROTECTED] wrote: | Hi, | | I am doing a new install of Debian 2.2 onto a older box of mine. I am | trying to get past the network modules installation. I started with 2 | Linksys ISA cards and had no luck so pulled them out and put in a PCI | Linksys Etherfast card. I tryed the ne module and ne PCI NE2000 support My PCI LinkSys Etherfast card doesn't use the NE2000 driver, but the Tulip one instead. | module but when it probes I get this error everytime. | | /lib/modules/2.2.18pre21/net/ni5010.o: init_modlues: Device or resource busy | Could it be an IRQ conflict in the settings somewhere? Or maybe it's because the wrong driver is being used. (Though I just read earlier on this list that the ISA cards use the NE2000 driver) -D
Re: D-Link DFE-530TX Probs W. 2.4.2
On Sat, Feb 24, 2001 at 08:37:42PM -0800, David Frey wrote: | | --- "Rick Commo" <[EMAIL PROTECTED]> | | > wrote: | | >I am not using kernel 2.4.2, but the driver that I selected was rtl8139.o. | | | | I think there are 2 versions. Older ones like mine use the via-rhine driver. | Have you solved this yet? Anyways, I bought a DFE-530TX just last week. (It was on sale so I only had to pay tax, and then postage to send in the rebate form) I gave it to a friend, and haven't opened it yet. We checked the web sites, and found that it needs the via-rhine driver. The DE-530TX (no 'F') uses the tulip driver. -D
Re: #! syntax
On Mon, Feb 26, 2001 at 05:50:54PM +0100, Carel Fellinger wrote: | On Mon, Feb 26, 2001 at 08:22:34AM -0800, Michael O'Brien wrote: | > Hola~ | > | > Is there any way to have a #! syntax that will conditionally run a interpreter | > based on a set of fallback locations? | > | > For example, we have a perl install in /dir/bin/perl. However, if you are at a | > non-work machine (ie, at home), you may not have a /dir/bin/perl. So, I'd like | > to have a script that will first try /dir/bin/perl, then if that doesn't | > exist, tries /usr/bin/perl. | | You could consider: | |#!/usr/bin/env perl | | as more installs will have env in the same place. | This is the correct way to begin python scripts (I can't say how the perl community feels) : #!/usr/bin/env python # python code here The idea is that env knows where stuff on the current machine is, so simply ask it for the interperter you want (python, perl, bash, etc). On a tangent, is there a way to manually configure env as to program locations? In particular, how can I have env on my cygwin installation know where python is (Win32 python build)? For now I have my python script in foo.py with a wrapper foo that does : #!/bin/bash python foo.py $* -D
Re: [ot] grub with 2 hard disks (was [OT] Grub)
On Tue, Feb 27, 2001 at 11:25:11AM +0800, Joey Kool wrote: | Ok, I was finally able to boot /dev/hdb and sees "GRUB" on the monitor. | But that's all. Seems like the thing hang or something though a cursor | is blinking. | | Another thing I can do is to install grub into the mbr of hd0 but don't | really want to do that as this is a company pc not mine. Don't wanna to | change /dev/hda's mbr held by nt. | | (For purpose of clarity, the previous problem of os not found was due to | the fact that I did not change the bios bootup sequence. I had to specify the 2nd harddisk as the bootup disk instead of the first in bios. | My pc is a IBM 300 GL.) | (I'm going through several days of list mail) When I first read your message, I thought you must be confused. There is only 1 MBR on a PC. That's why it is the MASTER boot record. Since you have changed your BIOS to look for the MBR in a different location (/dev/hdb instead of /dev/hda) that might solve that issue. I haven't tried this yet -- I need to get some Debian CDs to reinstall first -- but according to GRUB's documentation (and assuming BIOS compatibility) you should be able to make /dev/hda the linux disk and /dev/hdb the NT disk (physically switching the disks on the bus). THen use grub's swap functionality to make NT *think* it is on the primary master disk. (But switch the BIOS back to using /dev/hda as the MBR) Another alternative you have is to always boot off the floppy. Not quite ideal, but it isn't so bad either. BTW, GRUB is the coolest thing I have seen yet. I was able to use it to dual boot a Win2k/RH6.2 system I had to install at work the other week with no trouble. The linux kernel is above the 1024 cylinder too. I tried LILO on a floppy first, but after a fair amount of difficulty, it still couldn't boot windows. Being able to configure it /before/ an OS is running is very crucial for inexperienced users. It's shell is very cool too. -D
Re: ESS ES1898 Chipset Sound Card
On Tue, Feb 27, 2001 at 11:58:07PM -0600, Horburapa Mongkol-Q13382 wrote: | Hi, | | I have a No Brand name PCI sound card with ESS ES1898 Allegro Chipset. | Please anyone knows which sound module I have to use for this sound card? | BTW, I think you mean "No-Name Brand" ;-) I have an ESS 18[69][89] sound card at home. (I don't recall exactly which digits it ends in) I use the sound blaster module (sb) and one of the opl3 modules. After I get home I'll check my modules.conf. The sound cards works just fine with the proper DMA configuration. (I had the DMA channel wrong for a long time and sound was horrible -- I disabled it so I wouldn't have to suffer, then I eventually checked all the settings and found I needed to adjust the DMA channel now everything is beautiful) -D
Re: D-Link DFE-530TX Probs W. 2.4.2
On Wed, Feb 28, 2001 at 08:22:08AM -0600, W. Paul Mills wrote: | | The DFE-530TX uses the via-rhine driver, and the DFE-530TX+ | uses the rtl8139 driver. The last I knew, this was not | documented on the DLink web site or on Donald Becker's | web site. Only place you will find out about the rtl8139 | driver is on the disk that comes with the DFE-530TX+ card. | Oh, ok. Thanks. I don't remember if it is the TX+ or TX version. -D | | D-Man <[EMAIL PROTECTED]> wrote: | : On Sat, Feb 24, 2001 at 08:37:42PM -0800, David Frey wrote: | : | | : | --- "Rick Commo" <[EMAIL PROTECTED]> | : | | : | > wrote: | : | | : | >I am not using kernel 2.4.2, but the driver that I selected was rtl8139.o. | : | | : | I think there are 2 versions. Older ones like mine use the via-rhine driver. | : | | | : Have you solved this yet? | | : Anyways, I bought a DFE-530TX just last week. (It was on sale so I | : only had to pay tax, and then postage to send in the rebate form) I | : gave it to a friend, and haven't opened it yet. We checked the web | : sites, and found that it needs the via-rhine driver. The DE-530TX (no | : 'F') uses the tulip driver. | | : -D | | -- | * For God so loved the world that He gave his only begotten Son, * | * that whoever believes in Him should not perish...John 3:16 * | I like this sig :-)
Re: #! syntax
On Wed, Feb 28, 2001 at 06:17:35PM +, David Wright wrote: | Quoting D-Man ([EMAIL PROTECTED]): | > On Mon, Feb 26, 2001 at 05:50:54PM +0100, Carel Fellinger wrote: | | > | You could consider: | > | | > |#!/usr/bin/env perl | > | | > | as more installs will have env in the same place. | > | | > | > This is the correct way to begin python scripts (I can't say how the | > perl community feels) : | > | > | > #!/usr/bin/env python | > | > # python code here | | Well it's difficult to see much difference between these two | suggestions. It's the same, except that I don't know if the perl community takes an opinion regarding the "correctness" of #!/usr/bin/perl vs. #!/usr/bin/env perl The python community prefers the latter version (s/perl/python of course). | | > The idea is that env knows where stuff on the current machine is, so | > simply ask it for the interperter you want (python, perl, bash, etc). | | AFAIK env knows nothing about "where stuff is". It just modifies | the environment (nothing to do here as it happens) and runs perl | or python (which are found through $PATH). | | > On a tangent, is there a way to manually configure env as to program | > locations? In particular, how can I have env on my cygwin | > installation know where python is (Win32 python build)? For now I | > have my python script in foo.py with a wrapper foo that does : | > | > #!/bin/bash | > | > python foo.py $* | | I'm not sure what your difficulty is. Your $PATH presumably contains | python so the env trick should work. The env trick is only necessary | because #! requires the full path: although the path to perl may vary | on different systems, the path to env is usually the same, /usr/bin/env. | So you can use one script on the different systems and $PATH (different | on different systems) will actually find perl, python or whatever. | Hmm, it's working now. I must not have had python in the path before. I have a shell script in the path under my home directory that points to the full path of the python executable. | If you want to run a specific version of python, then just write its | path in your script: | | #!/usr/bin/env /weird/path/python32 That defeats the purpose of using env as the first command ;-) Thanks, -D
Re: permission denied on .bashrc ??
On Thu, Mar 01, 2001 at 06:53:16AM -0800, Tom Schuetz wrote: | I reconfigured both ~./bash_profile and ~/.bashrc to work properly, and they do- ls is automatically aliased as --color, etc. | | But at login, I still get a '~/.bash_profile: permission denied' error. | | When I look at the permissions, I'm listed as the owner. I did these changes as an ordinary user, not as root. What are the permissions? Not the owner/group, but the permissions. Ex : $ ls -l ~/.bashrc -rw-r--r-- 1 derrick None 1615 Feb 28 17:02 /home/derrick/.bashrc ^^ These are the permissions. The owner (me) has read and write permission while the group (None -- must be due to cygwin) has read only permission and the rest of the world has read only permission (probably not a good idea, but it's running on top of winodws, I'm the only user and it's behind a firewall) Maybe you don't have read permission, or maybe it wants to execute it? (I don't think cygwin cares very much about execute permission ;-) HTH, -D
Re: lilo errors
On Tue, Feb 27, 2001 at 07:19:15PM -0800, Mike Egglestone wrote: | Hi all.. | | I started an install of potato 2.2 r2 and installed lilo into | the MBR as normal | The computer is a Pentium 75.. with 24 meg of RAM... | When the computer reboots... I get a ton of 40's across the screen... | forever(lilo doesn't load) Try the archives. There was a discussion on this a while back. I would also highly recommend using grub instead of lilo for the boot loader. 100 times easier to configure, and can be configured from the boot menu (as in you don't need a bootable/running OS to configure the boot loader -- eliminates the chicken vs. egg problem you have) HTH, -D
Re: Lost linux partition
On Fri, Mar 02, 2001 at 12:12:11AM +0800, #KUNDAN KUMAR# wrote: | I boot from boot floppy only, no lilo... and the boot floppy stratightaway | loads the kernel without any lilo prompt.. and then gets stuck at Kernel | Panic: No root file system. | I do have debian rescue disk under the c: and I try to boot from there I | get a consolebut running lilo has no meaning here | I am afraid its lost for ever... does NAV back up the partition table before | repairing? | Kundan | It's still not lost forever, just that lilo needs the partition table to figure out where linux is. (Note to others, NAV overwrote the *partition table*, not the MBR) Boot with your debian rescue disk. At the console, run fdisk and set the partition table the way it was before. Exactly the way it was before. Don't do any sort of partition initialization because that would destroy the data on the disk. (AFAIK fdisk will only change the partition table itself) Then boot again from your boot floppy. LILO/the kernel should then be able to figure out where the root filesystem is. HTH, -D
Re: MS OFFICE PRO 2000 $120
oooh, this will run on my (Debian) computer right? All of Debian's supported architectures too, right? ;-) (just making fun of ridiculous spam) -D On Thu, Mar 01, 2001 at 01:01:31PM -0500, [EMAIL PROTECTED] wrote: | MS office professional 2000 FULL version for $120 !!! NEW !! | | If interested email at [EMAIL PROTECTED] | | Verified Business Paypal owner so purchases FULLY protected | Verified Aol subscriber | Verified Address | | Not burns,oem,accademic,charity,demo or otherwisebut FULLY hologrammed | authentic ms office professional FULL version!! | FULL product cd keys | Word,excel,powerpoint,publisher,access,outlook and small business tools!!! | ***Disclaimer:*** | If this was sent to you by mistake or was unwanted I apologize and will | take you immetiately from my list...thanks again |
Re: using RHL 7's rpm in debian 2.2r2
On Fri, Mar 02, 2001 at 01:10:35AM +0530, Ankit Jain wrote: | If i use src.tgz files to install somethin, say licq, then the 'database of | s/w installed' will not have ne details on 'licq' .. right? Is there any way | i can 'inform' the 'database' of existence of 'licq' or is it fine the way | it is? You can inform the database by installing a .deb package. The system is fine (assuming it compiled properly, and isn't horridly buggy), but the package database won't know it exists and won't know what it depends on, and won't know what files belong to it, etc. It's a maintenance decision. The debian package system makes system maintenace much easier by keeping track of the packages you have. If licq needs, say, Qt (I don't know, I don't use it) and you installed it from the tarball, dpkg won't stop you from happily removing Qt. However, licq won't work any more. If you install from anything other than properly made .deb packages, it becomes your responsibility to keep track of files and dependencies. I would recommend sticking with .deb packages, and finding some way to acquire them. Perhaps you can wait until the next stable release, then get CDs from somewhere? Maybe you can access a computer with a faster/cheaper net connection and use some removable media to transfer the new packages? -D
Re: ESS ES1898 Chipset Sound Card
On Thu, Mar 01, 2001 at 09:59:03AM -0600, Horburapa Mongkol-Q13382 wrote: | Yes, I meant "No-Name Brand". :) | | Could you tell me what you add in modules.conf to make it work? | I have an ESS 1869. As far as RH's sndconfig tool is concerned, it is the same as the 1868. My modules.conf has : alias sound sb alias midi opl3 options opl3 io=0x388 options sb io=0x220 irq=5 dma=1 mpu_io=0x330 As it says in the comments at the beginning of /etc/modules.conf, you shouldn't edit it by hand. Instead, put the lines in the following files : /etc/modutils/aliases : alias sound sb alias midi opl3 /etc/modutils/opl3 : options opl3 io=0x388 /etc/modutils/sb : options sb io=0x220 irq=5 dma=1 mpu_io=0x330 then run /sbin/update-modules as root. The purpose of this is to allow tools to automatically update your modules.conf file as needed without losing any custom changes you may have. Instead you edit separate files that are specific to what you want, and let update-modules combine it appropriately into modules.conf. -D
Re: Getting rid of Vim's startup screen
On Fri, Mar 02, 2001 at 03:37:45PM +, Colin Watson wrote: | [EMAIL PROTECTED] (Colin Watson) wrote: | >"Colin Cashman" <[EMAIL PROTECTED]> wrote: | >>Every time I start Vim with a new file, I get an intro screen. How do I | >>surpress that? | > | >Put 'set shortmess=I' (or 'set shm=I') in your ~/.vimrc. | | Bah, turns out the default is 'shortmess=filnxtToO' (intuitive, huh?), | so you need to add to that, not replace it. 'set shortmess=filnxtToOI' better yet, set shortmess+=I (I just wanted to mention the "+=" operator, the intro message doesn't bother me so I have never RTFM'd shortmess) -D
Re: D-Link DFE-530TX Probs W. 2.4.2
On Mon, Mar 05, 2001 at 10:42:23AM +0100, Martin Würtele wrote: | On Sat, Mar 03, 2001 at 06:08:12PM -0500, James Moody wrote: | | i thought that the dlink 530 tx cards have a via rhine chip as cat Just to summarize what I have recently learned regarding the D-Link cards : DE-530TXtulip DFE-530TX via-rhine DFE-530TX+ rtl8139 Apparently D-Link likes that model number, but can't stick with a single chipset for it. ;-) IMO when major changes occur (ie a different chipset) the model number should change significantly and in an obvious way. Too bad marketing doesn't seem to agree. -D
Re: mailing list software
On Mon, Mar 05, 2001 at 05:13:39PM +0100, Peter Lieven wrote: | i'm confronted with setting up a mailing list with about 20-30k | subscribers. i find mailman very useful, | but i want to remove the password option. is there a patch available? I take it you don't want the users to need a password? Might be a bit dangerous allowing people to modify other people's subscriptions. Anyways, I imagine you could simply remove/tweak the section of mailman that verifies a password has been entered and have it simply use the empty string instead. mailman is written in Python so it shouldn't be too hard to read ;-). -D
Re: Debian quit booting
On Fri, Mar 02, 2001 at 09:14:00AM -0500, - wrote: | I have just the one floppy that was made during Debian installation.. I | wanted to make another boot floppy while Debian was working but I didn't | know how. Here is how to make a new boot floppy : Go to www.gnu.org/software/grub/ (the URL might be slightly off). Go to the download section and get the file grub-0.5.96.1-i386_ext2fs.bin (a couple of the separators might be off). Boot one of your other Linux systems. Run dd to install into a new floppy. dd if=grub-... of=/dev/fd0 (I think you need a couple more options, read the docs on the web site) The boot loader you have just installed on a new floppy is called "grub". It works very well, and can even be (re)configured while you are trying to boot (ie from the menu). It is very easy to use. 100x easier than LILO. Edit the file boot/grub/menu.lst to suit. You can have all your Linux (even Windows, *BSD, and others) boot from the same grub setup. When you are happy with your setup you can even put it on your MBR with the grub command setup (hd0) (hd0,1) assuming /boot/grub is on /dev/hda2 and the MBR is on /dev/hda. Alternatively you can just always use the floppy to boot, but you can forget about changing floppies since this one will handle all the systems. Also, make a backup boot floppy, just in case . . . ;-) HTH, -D
Re: Printer recommendations
On Fri, Mar 02, 2001 at 02:17:55PM -0800, kmself@ix.netcom.com wrote: | on Mon, Feb 26, 2001 at 08:26:36PM -0500, Alec Smith ([EMAIL PROTECTED]) wrote: | > I'm faced with recommending a new printer for my parents as their old unit | > is starting to show its age. The goal is to find something for $350 or | > less which is Linux friendly since I'll also be using it. The printer will | > be connected to the LAN through an old HP JetDirect EX. Right now we're | > looking more towards a laser but I'd also be interested in inkjet | > recommendations as well. | > | | My recommendation is for a laserjet. Inkjets are slow, expensive to | run, and tend to be maintenance-happy. | | I just purchased a used HPLJ4m for about $380. The footprint's about | twice what the HPDJ6xxC was, but it prints about ten times faster, | literally, native postscript. If you have the space, an HPLJIIIx is | even less expensive, though I suspect you'll make up the difference on | power bills if it's on all the time. Ideal would be to rig up something | to power on the printer automajikally when needed. More modern LJs have | a built-in powersave option, the IIIs and 4s I looked at did not. | I have an old LJIIIp printer. It works quite well. I don't know how much they cost since I got this as a hand-me-down. It prints 300 dpi, 4ppm. Has PCL5 natively and a PS (Level 1) card with it. A few years back the LJ6l was the "personal" laserjet model. It had PCL builtin, 600 dpi, 6 ppm. I think it ran ~$400 new. I think the 1100 series has replaced the 6l series now. The Laserjet 5/6 M and friends did have a powersave feature with an "instant-on" fuser. Quite nice. 600 dpi, 8 ppm. Toner is quite cheap per-page and gives much better quality than inkjets for black. I would highly recommend getting a laser jet printer. I like them and have seen them give good performance. My particular model was manufactured in '92 and is still chugging along quite well. Models with built-in postscript interpreters are the best since PS is quite a standard format in *nix. Windows drives can deal with it. PCL (version 5 at least) is supported and in my experience usually prints faster. I had it set up to use PCL by default on a RH system using printtool. -D
Re: booting Win95 with LILO
On Sun, Mar 04, 2001 at 04:32:57PM -0800, scud wrote: | Hi everybody, | I decided to boot Win95 once again, for making some changes and install | software for useing with Wine. But the MBR has been rewritten during | Linux installation. I boot Linux with LILO. How exactly shuld look the | lilo.conf file in order to be able to boot Win95. Win95s are at hda3, | which begins at some 1.2GB . I tried this: | | It didn't work, probably because BIOS can't see so far (hda3 begins at | 1.2GB). | If you have any sollution please help. Use grub :-). It will take you ~ 2 seconds to understand how to configure it to boot windows. Here is the section to use : (grub uses zero-based indexing) title Windows root (hd0,2) makeactive chainloader +1 :-) (Umm, where is the Win boot loader? You may need to adjust the offset in that last line. Both the machines I have used it on had windows on hda1.) On one machine, linux starts at ~14GB. Grub boots it quite happily. (can't say the same for lilo). On the other machine, /dev/hda is windows while /dev/hdc is linux. LILO got stuck at 'LI' when trying to boot Linux (never really tried it for windows on that machine). Grub is now on the MBR and is quite happy with both OSes. HTH, -D
CPU architectures (was Re: Win95 won't reboot after Debian install)
On Wed, Mar 07, 2001 at 11:57:44AM +0100, Erdmut Pfeifer wrote: | | This A20-line crap rates as one of the most insane ideas ever put forth | in the whole history of PCs. Anyone who doesn't know already may want to | read up a little on what it's about, for example here | | http://www.phys.uu.nl/~mjanssen/control.php3?chapter=6 | http://www.phys.uu.nl/~mjanssen/control.php3?chapter=9 | (a google search for the exact phrase "A20 line" will turn up a couple | of more links) | | In short, it was born in a desperate attempt at gaining another 64k | (k!) of memory addressable beyond the 1M limit of the so-called "real | mode" of 80x86 processors. It was realised by introducing some weird | mechanics into the PC hardware that allows to switch that special | addressing scheme on or off. | It's one of the prototypical examples of how an ad-hoc solution of no | real benefit can cause headaches for thousands of people. Even years | after its invention it seems to haunt innocent users. Unbelievable! | Oh, interesting. Some more details in how bad the design of the x86 architecture was/is. In my Into to Assembly class we used M68K processors. The prof spent 1 day giving an overview of the layout of the x86, and I was really glad then we weren't using x86 systems! The only advantage I can think of is Pentiums are current and fast while m68k is a dead (as in nothing new happending) product line. Does anyone have any overviews/comparisons of Sparc and Alpha designs? What about ARM? I don't know much (anything realy) about ARM and where it came from or who made it. I'm really more of a software guy, but I find the hardware to be interesting as well. | -- Bugs come in through open windows. Keep Windows shut! -- Nice. -D
Re: debain on windows 2k
On Mon, Mar 05, 2001 at 11:25:42AM +0100, Peter Schnebel wrote: | hi | | On Monday 05 March 2001 09:35, Subramaniam Aiyer (CTS) wrote: | > hi, | > i am trying to install debian linux on my office computer which has | > only one hard disk i.e no partitions | | whats that supposed to mean ?? have you got only one big partition ?? | | > can i go ahead with the insatllation or should i downgrade to windowsNT. | > one of my friend who was installing red hat had to first downgrade to | > NT. Also, the sysadmin gave some reason of file systems. can anybody please | > explain further? | > thanx, | > INDSpeedFreak | | the only problem i was facing in this constellation was the win2k | bootmanager... | don't ( i mean it ) install lilo in the mbr... you definetly will run into | problems with win2k... I agree. But I offer an easier solution -- use grub. It is a piece of cake to dual boot linux and win2k with and even has a Debian package. When I had to install linux on the extra partition of a win2k box at work I couldn't get lilo to chainload windows. Then I tried grub and it took no time at all. Much easier. -D
Re: Printer recommendations
On Thu, Mar 08, 2001 at 05:43:29PM -0900, Ethan Benson wrote: | On Thu, Mar 08, 2001 at 08:52:00AM -0800, Chad Maine wrote: | > http://www.linuxprinting.org has collected stats on this subject. Should | > give you a list of printers to choose from based on how well they work in | > linux, if at all. Of course, any postscript printer will work perfectly. | | beware that linuxprinting.org lists some non-postscript inkjets as | `perfectly' supported, this is not entirely true as it may still be a | absolute *NIGHTMARE* to make one of these miserable things work. I haven't seen what it lists, but the HP DeskJet 6xx my roommate bought worked. (I don't remember the exact model, but it was a new one in the 600 series). I had PCL3 in it. I was on RH at the time, and had it printing just fine using printtool without any major difficulty. This was over samba (which I didn't know how to use). The most time was spent trying to get connected through samba since I had to figure out the linux side, then disabling his cheap win32 firewall. DON'T get a "PPA" printer. Those suck. My friend has one and it hasn't been pretty. (Printing Performance??? Architecture, no performance there!) A PCL printer is just fine (PCL5 and PCL3 from experience) and much faster than PostScript. -D
Re: colors in gvim
On Fri, Mar 09, 2001 at 06:37:04PM -0800, Eric G. Miller wrote: | On Fri, Mar 09, 2001 at 05:47:45PM -0500, Marcelo Chiapparini wrote: | > Dear debianers | > | > I am using the vim-gtk plus vim-rt packages. Although I was able to | > change the font using by gvim, I don't | > know how change the background and foreground colors. I think that it is a | > matter of put | > something in /etc/gvimrc, like I done for the fonts. I looked in the man | > pages and on-line help, but I didn't | > find anything. Any help will be very appreciated. | > Regards, | | You may want something like: | | " set bgcolor to light, so colors are readable | set background=light | | The alternative is "dark". This only sets the tint/shade (I don't know the proper "art" term). If you use a black terminal and run vim, you want "set bg=dark" to make the colors lighter. If you use "set bg=light" the colors will be so dark (against the black) that it will be almost impossible to use. Vice versa for a light background (gvim's default of white). The solution is to change the colors used by the syntax highlighting for "Normal" text. In my .gvimrc I use: highlight Normal guibg=black guifg=grey90 to get a black background and text that is grey90 (not quite a glaring white). I also use "set bg=dark" to make the highlighting readable. See ":help highlight" for more info on how this actually works (or just ask ;-)). Also, there is a cl option "--rv" or "--reverse-video" to reverse the colors from their normal. gvim will actually check to make sure that it is light-on-dark rather than just blindly switching fg & bg. HTH, -D
Re: Where are clear-text passwords stored?
On Sat, Mar 10, 2001 at 12:45:31PM -0800, Robert Cymbala wrote: | | Is this list an exhaustive list of places where my dial-up password is | stored? I'm about to return a laptop after installing Debian for a | friend, and want to make sure all my passwords are deleted. It depends on what you use to dial. On my system it is /etc/wvdial.conf -D
Re: spontaneous partitions type changes (GRUB)
On Mon, Mar 12, 2001 at 03:14:57PM +0100, Carel Fellinger wrote: | Hi Balbir, | | On Sun, Mar 11, 2001 at 02:31:02PM -0500, Balbir Thomas wrote: | > Hi, | > I have installed dos, win98 and debian on my system using GRUB as | > the bootloader. The installation was ok. But every time I boot Good choice of bootloader's if I may say so ;-). | > into windows or dos and try to boot back into linux I get the | > Stage 2 error 17, that the info documentations says is due to an | > unrecogized partitions type. When I check the partition type in | > linux using cfdisk , it is reported as being of type "amoeba". I | > change its type to linux ext2 and write it to disk (Without losing | > any data on the disk) and the error repeats if I boot into win or | > dos. | | I'm not much of a guru, but I suspect that in your GRUB config file ( see | /etc/boot/grub/menu.lst ) you'll find a "hide" command for the linux parti- | tion, but no corresponding "unhide" command in the linux section itself. I saw different symptons when I tried hiding the linux partitions from win2k, then not unhiding them from linux. The partition table wasn't messed up at all, I just couldn't boot because the partition didn't exist. (That's what hiding does, it makes the partition SEEM to not exist) When I unhid the partitions it was fine. Actually, I could still see the linux partions in win2k (though the disks had weird properties and didn't function, as expected). As a result I just removed all the hide/unhide commands. Actually, on this system I was using grub from a floppy disk. I didn't want to overwrite the MBR since it is a company machine. At home I was pleasantly surprised to see grub able to boot linux when I installed it on the MBR (lilo couldn't). I haven't had any trouble with it, but I don't have plain DOS installed. Also, at home it is 2 separate disks. The hide/unhide didn't seem to have any effect. (Better, IMO than simply failing to boot since apparently my BIOS doesn't support it) -D
Re: vi/emacs: Loging shell command and output
On Fri, Mar 09, 2001 at 01:37:42PM -0800, Osamu Aoki wrote: | Sorry, I am using different teminal this time. (windoze) | | > I also found that "export term=xterm" stops autocolor-ls to producing | > escape sequences. Recorded text after this is printable. | | This may be bogus statement but setting term to proper mode shall stop | autocolor-ls to producing color. I think I think it is "set term to a terminal not supporting color and autocolor-ls won't output color escapes". Also, if stdout is not a terminal, color codes are omitted (eg "ls --color=auto | less"). On a couple of Solaris 7/8 systems I have access to, 'xterm' doesn't support color (in termcap), but does have underlining and bolding of text. It is very annoying since my terminals/emulators support color just fine (when logging in to a Linux box). I couldn't find any terms in /etc/termcap that supported color except for a few special ones that didn't work at all with my emulator. -D
Yamaha sound card (FW: pls help me)
I got this message in private e-mail today. Melvin needs help configuring a Yamaha sound card. Can anyone help? (cc replies to him since his subscription was rejected (see below)) -D - Forwarded message from Melvin Sebastian <[EMAIL PROTECTED]> - From: Melvin Sebastian <[EMAIL PROTECTED]> Date: Mon, 12 Mar 2001 20:41:33 + (GMT) To: [EMAIL PROTECTED] Hi, My name is melvin and i'm a fresher into the linux environment. I'm not much exposed to the unix platform. but after installing linux i found it extremely good and easy to work with. But since i don know anybody in this field i get stuck up while configuring something or the other.Like for example right now i'm stuck up with configuring the sound card in debian.and i'm using a pci yamaha sound card. and like in RedHat it is not able to configure with snconfig tool. Can u pls help me out of this any way or can u tell me whom to escalate these issues to. i've heard that the mailing list helps a lot but even after sbscribing there, my subscription was rejected. Can u pls help me get over this somehow.. regards Melvin - End forwarded message -
Re: wraping lines in mutt
On Tue, Mar 13, 2001 at 11:35:42AM -0800, Steven E. Harris wrote: | [EMAIL PROTECTED] writes: | | > In my case, it is vi. I have a .exrc file in my home directory that | > has the following contents: | > | > set wl=76 | | I tried creating such a file, and (n)vi does note that the "wraplen" | variable is set to 76, but I can still type way past that 76 column | point. I thought that maybe vi would chop the lines up upon exit, but I don't know about nvi, but in vim the "formatoptions" must also be set properly to wrap the lines. (Also, in vim the variable "textwidth" must be set, not "wraplen") Someone else said to set it to "t", mine is set to "tcq". Also, gqip or gqap will rewrap the current paragraph. gqG will reqwrap from here to the end of the buffer, gq} will rewrap from here to the end of the current paragraph. -D
Re: spontaneous partitions type changes (GRUB)
On Wed, Mar 14, 2001 at 12:50:31AM +0100, Carel Fellinger wrote: | On Mon, Mar 12, 2001 at 06:52:42PM -0500, D-Man wrote: | > I saw different symptons when I tried hiding the linux partitions from | > win2k, then not unhiding them from linux. The partition table wasn't | > messed up at all, I just couldn't boot because the partition didn't | > exist. (That's what hiding does, it makes the partition SEEM to not | > exist) | | No it doesn't, it simply changes the partition type by toggling one | bit, 0b (windows) becomes 1b (hidden windows), 83 (Linux ext) becomes | 93 (amoebe), and reverse. So indeed the partition table is not messed | up, only the type fields in the partition table are mangled, but in a Oh, ok. I was under the impression that the hide/unhide was a function of the BIOS. Thanks for the explanation. | > When I unhid the partitions it was fine. Actually, I could still see | > the linux partions in win2k (though the disks had weird properties and | > didn't function, as expected). As a result I just removed all the | > hide/unhide commands. | | hide/unhide is only working for osses/loaders that pay attention to | the partition type. Linux/lilo for one, doesn't (as far as I know), | and I don't know of win2k. I'm surprised it did recognize linux | partitions though. Or did it only recognize the linux partitions if | they were hidden? In that case it thought it saw amoebe partitions It recognized the partitions in the same way whether they were hidden or not. They showed up as E: and F:, but other than showing up in "My Computer", they did nothing else. I think the properties listed it as 0 bytes or something. It wasn't quite correct, but it made sense because windows doesn't understand ext2. This is different than my Win98 machine at home which doesn't even acknowledge the existence of the ext2 partitions. -D
Re: linux card
On Wed, Mar 14, 2001 at 08:54:24AM -0600, Matt Fair wrote: | | I'm not sure if this had gone through last time I submitted this. | I need to create a card that when scanned in a credit card scanner it | connects to my server via the phone line. | | I would like to have the web server receive information like what | someone purchased, who they purchased from and record that on the | server. | | This card isn't a credit card where it completes a transaction, but it | tracks what people purchase and gives them points based on what they | purchase to give prizes away. | | Does anyone know where I could find more information, like a HOWTO, or | something that could explain this process in-depth? I don't know of any how-tos, but I think the basic steps would be : 1) get a scanner. IMO barcode cards are better (more reliable, less susceptible to accidental demagnetization, etc), but choose what you (or your boss) likes best. 2) get the docs for the scanner, a device driver for it is nice too 3) figure out from the docs/driver how to receive input from the scanner -- hopefully it has an interrupt to alert you that it has input for you, the barcode scanners I use at one job connects through the keyboard port. the terminal doesn't know whether it is getting the input from the scanner, or if it was typed in manually 4) write a program that receives the data from the scanner and does what you want with it -- in this case connect to a server and log it in a database There are a number of stores around my area that do this sort of thing. The software the records the data must have access to both the customer's id (from that scanner) and the transaction from the sales register (to know what was purchased). I think that this question isn't exactly on-topic for debian-user since it will involve more than simply configuring debian system. It will involve writing some code to bring the various components together and process the data so that it is useful to you. If you don't have programming experience, I recommend the python language. The language is very nice, high-level, and easy to grasp. The python community is also very friendly and helpful and has a nice tutor list, [EMAIL PROTECTED] HTH, -D
Re: linux card
On Wed, Mar 14, 2001 at 12:15:26PM -0500, Mike Dresser wrote: | brian moore wrote: | | > And... then you may as well go to Radio Schlock and get one of those | > silly CueCat scanners. Throw away the software without looking at it, | > so you don't have to accept their silly terms Then you have a free | > barcode reader. See .sig :) | I always wondered what that weird looking sig was supposed to mean ;-). | Aye, except that he might want a _working_ scanner. I've got one of | them at home, and the recognition... well... it'd be faster to type | the numbers in, than repeatedly scan the barcode, praying it'll read | this time. I've worked with proper scanners, there's a _big_ | difference. Yeah. My part-time job is in the video department of a fairly large grocery store. The scanners aren't CueCat, but they are aging and often don't work quite right. It's really a pain when it gets a bit or two off and give a different number. Or just ignores the first couple of decimal digits in the code. The "real" scanners are on the cash registers at the front end. I haven't worked front-end, but they don't have nearly as much trouble as video has. -D
Re: FW: D-Link DFE-530TX+ via-rhine.o module
On Mon, Mar 19, 2001 at 01:26:31AM -0800, David Carlile wrote: | Thanks for the quick reply. | | I tried modconf and selected rtl8193. I didn't include any parameters. I got | an error message: "Device or resource busy. Hint: this error can be caused | by incorrect module parameters, including invalid IO or irq parameters." | | Any ideas? I think that browsing /proc/pci can give some information, but I haven't tried that yet. Have you used the card in windows? What does it report as the IRQ and IO addresses? | BTW, I checked the card and the link light is on, which kind of suprises me | because I didn't think I had a driver loaded... Is it possible a driver was I would expect the link light to be on if you have the card plugged into the service (or cable modem) because a link exists. If you unplug the card from the network the link light will go off. This is hardware independent of the software drivers. It is a convenient way to get information about which part of the card isn't working. My card (LinkSys) has link, 100, duplex, and tx lights to indicate if a link (network) exists, the network is 100 Mbps, the network is full duplex, the card is transfering data right not. -D
Re: executable compatible with Debian and Redhat?
This is correct. I have a RH7.0 system: $ gcc --version 2.96 $ rpm -q glib glib-1.2.8-4 (I don't know if the glib is stable or not) Check on gcc.gnu.org and you will see that 2.96 was the label for the devel tree and is binary incompatible with all but version 2.96. If you want to "upgrage" (don't!) your system to use gcc 2.96 it will be binary compatible with other executables made with 2.96. Of course, then things made with the stable 2.95.2 won't work. (also, the c++ optimizer has a bug that gives an internal compiler error and quits when compiling certain code) -D PS. When I get some free time I want to install Debian on my other disk and see how I like it. Anybody want to offer a comparison of Debian and RedHat with both pros and cons? (on a diff thread of course) On Thu, 30 Nov 2000 20:56:35 Jonathan D. Proulx wrote: | Hi, | | WARNING! | this is pure hearsay, which I have no documentation for | | I heard about this problem (from what I consider knowlegable people), | red hat seems to be using a development branch of either gcc or glibc | (I forget which, sorry), the result RedHat 7.0 is frequently if not | always binary incompatable with any other GNU/Linux implementation. | | As I said this is hear say, I'll try and find references one way or | the other. | | Corrections welcome flames > /dev/null | | -Jon | | | -- | To UNSUBSCRIBE, email to [EMAIL PROTECTED] | with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] |
[OT] Apple IIe help please
Hi all. I apologize for the off topic-ness of the message and the cross-post. I have a friend who has a program she likes (written in BASIC) on an Apple IIe. I have a way to get into the code and list it on the screen. What I am looking for is someone who has had some experience using an Apple IIe who can tell me how I can get the listing to be redirected to the printer port. Also if it is possible to get a copy of the program file onto a disk for an IBM compatible comptuer (Windows or Linux) that would be great. Thanks in advance for all the gurus out there who can help me. -D
Re: Does unsubscribing from this list really work?
A solution (though not ideal) : configure procmail to dump all messages for debian-user to /dev/null :-) -D On Mon, 04 Dec 2000 16:45:20 Ringo De Smet wrote: | On Monday 04 December 2000 22:20, you wrote: | > I can't understand all this dump stuff about failing to unsubscribe. | > under every mail it is described clearly and exactly. just follow the | > instruction. | > it does work, _defenetly_ | | Well yes, the instructions are clear, but for some reason, I never get | a confirmation message. I don't even get a confirmation message if I | try to unsubscribe through the web frontend. I tried to | subscribe/unsubscribe with different email addresses. IT JUST DOESN'T | WORK! *sigh* | | Ringo | -- | ** Ringo De Smet Ringo.DeSmet AT bigfoot.com **
Re: Mail deleted locally -> deleted on the server?
If I understand what you are saying, I don't think it will work like that. I use fetchmail, but not mutt. Fetchmail simply gets the mail from the server and stores it locally. AFAIK mutt (or any other mailer) can't communicate with fetchmail since fetchmail doesn't take direction from other apps. If you aren't running fetchmail in daemon mode then fetchmail won't even be running while you read your mail. I think you need some other sort of setup, like maybe using a mailer that can access the IMAP folder directly instead of using fetchmail. (maybe fetchmail can accept requests to delete particular messages form the server, another more experienced person will have to comment on that) -D On Mon, 04 Dec 2000 22:54:14 Ignasi Tura wrote: | But, as an example for my situation: | | I download the mail. The mail is kept with the 'keep' option. I read it with | mutt and I decide to delete one of that emails and keep the others. If I was | using Netscape Messenger or Eudora Light, for example, the next time I | downloaded mail that letter would be also deleted in the server, and the | others would be kept. | | I would like this behaviour, as I also access Yahoo with the browser. | | If it's RTFM again, please forgive me. I can't figure how the mutt and | fetchmail can communicatein this way. | | | Best, | | | Ignasi |
Re: LI- Lilo just stops working
I had the problem of LILO stopping at "LI" when booting from the MBR on my machine. Upon further research I found that my BIOS is too cheap to boot from /dev/hdc. (Using LILO to boot from floppy worked, but couldn't find /dev/hda). My solution is to use loadlin.exe (from AUTOEXEC.BAT). -D
Re: OT: Perl or PHP
I would recommend not using Perl. Perl can let you do lots of complex text manipulation really easily, but it is a nightmare to debug, let alone maintain. It does too much magic with a lack of data types and automagic variables. I would recommend Python if you want to do CGI scripting. But as someone else said, you might want PHP instead of CGI scripting. -D On Tue, 05 Dec 2000 04:27:02 [EMAIL PROTECTED] wrote: | hi. sorry for the off-topic post. i'm embarking on a journey to be a web | developer and would like to know if i should use mod.perl or mod.php for cgi? | i've searched and read alot of faqs and so far they're all just howtos. the | only conclusion i can come up with is with php u can use it directly with | html | (plus also call it from html code) whereas with perl the only way u can use | it | is to call it from your html code. | | am i right with that assumption? | |
Re: erase my adress on your list!!!!
You've got a PhD. If you can't figure out how to get off the list, make a filter that dumps it to /dev/null. ( hint: $ man procmail ) On Fri, 08 Dec 2000 22:17:09 Jim Kroger wrote: | | UNSUBSCRIBE UNSUBSCRIBE UNSUBSCRIBE | | _ | James K. Kroger, Ph.D. | Center for the Study of Brain, Mind, and Behavior | Department of Psychology | 3-N-4D Green Hall | Princeton University | Princeton, NJ 08544-1010, USA | Tel: (609) 258-1291 | Fax: (609) 258-1113 | [EMAIL PROTECTED] | http://www.princeton.edu/~kroger/home/ |
Re: Too many processes for Tomcat
I don't know what Tomcat is, but I did notice the references to threads in the ps output. AFAIK using threads causes ps (and top) to report more processes than really exist : each thread is reported as a process. -D On Sat, 09 Dec 2000 05:50:38 Daniel de los Reyes wrote: | I just installed Tomcat and realiced it runs on 33 processes . Is this | normal? | Attached is the result of a ps ax | | -- | __ | Daniel de los Reyes | S2-Selling Soluciones | Valencia Spain | e-mail: [EMAIL PROTECTED] | Powered by Debian GNU-Linux 2.2r2 | __ | | PID TTY STAT TIME COMMAND | 318 ?S 0:01 /usr/lib/j2sdk1.3/bin/i386/native_threads/java | -Dtomcat.home=/usr/local/tomcat org.apache.tomcat.startup.Tomcat | 445 ?S 0:00 /usr/lib/j2sdk1.3/bin/i386/native_threads/java | -Dtomcat.home=/usr/local/tomcat org.apache.tomcat.startup.Tomcat | 446 ?S 0:05 /usr/lib/j2sdk1.3/bin/i386/native_threads/java | -Dtomcat.home=/usr/local/tomcat org.apache.tomcat.startup.Tomcat | 450 ?S 0:00 /usr/lib/j2sdk1.3/bin/i386/native_threads/java | -Dtomcat.home=/usr/local/tomcat org.apache.tomcat.startup.Tomcat | 457 ?S 0:00 /usr/lib/j2sdk1.3/bin/i386/native_threads/java | -Dtomcat.home=/usr/local/tomcat org.apache.tomcat.startup.Tomcat | 458 ?S 0:00 /usr/lib/j2sdk1.3/bin/i386/native_threads/java | -Dtomcat.home=/usr/local/tomcat org.apache.tomcat.startup.Tomcat | 459 ?S 0:00 /usr/lib/j2sdk1.3/bin/i386/native_threads/java | -Dtomcat.home=/usr/local/tomcat org.apache.tomcat.startup.Tomcat | 460 ?S 0:00 /usr/lib/j2sdk1.3/bin/i386/native_threads/java | -Dtomcat.home=/usr/local/tomcat org.apache.tomcat.startup.Tomca |
Re: Sound card interrupt problem (ESS ES1878)
Playing CD is different than playing other sounds. The cd drive sends the audio directly to the sound card with no software intervention. (I had a cd playing after the system had shutdown). I have an ESS1869 that works just fine; that is once I fixed the IRQ (or was it the DMA channel?). I had a horrible echo effect before I fixed it. My modules.conf looks like this: = # sound card options (works!!) # -- run 'sndconfig', select "ESS 1868" # -- IO base 0x220, IRQ 5, DMA 1, MPU IO 0x330 # Sound modules dependencies: # sound : none # sb : uart401.0 # uart401 : sound.0 # opl3 :sound.0 # deps are listed in /lib/modules/2.2.16/modules.dep, # deps are loaded automagically by insmod # use the 'sb' sound driver alias sound sb options sb io=0x220 irq=5 dma=1 mpu_io=0x330 # after loading the sound (sb) module, load the midi (opl3) module post-install sound /sbin/insmod opl3 # ?? post-install awe_wave /bin/sfxload /etc/midi/GU11-ROM.SF2 # the midi (opl3) module alias midi opl3 options opl3 io=0x388 === BTW, this is for RH7 (copied from my RH6.1 setup). I don't think the distro would matter here since it is at the kernel level HTH, -D On Tue, 12 Dec 2000 17:18:54 Tiarnan O Corrain wrote: | Hello... | | I recently installed Debian on a Compaq Armada 7770DMT laptop | with an ES 1878 sound card. I've compiled support for the Soundblaster | into the kernel (2.2.17), which allows me to play audio cds, but gives a | weird | echo effect when I try to play wave files using 'play'. | | I've compiled the kernel twice, once using IRQ 7 and once IRQ | 8, with the same effects. Here is the relevant section from | dmesg... | | Sound initialization started | ESS chip ES1878 detected | ESS1688: Invalid IRQ 8 | sb: Interrupt test on IRQ8 failed - Probable IRQ conflict | at 0x220 irq 8 dma 1,5 | Sound initialization complete | | Does anyone have any ways of changing the IRQ and DMA settings | without recompiling the kernel, or suggestions for the correct | settings, or how to come by then? | | Thanks | | Tiarnán Ó Corráin <[EMAIL PROTECTED]> | "Parturient montes, nascetur ridiculus mus."
Re: Relation(exim,fetchmail,mutt)=?
I think the problem with your statement is the word *have*. I don't know anything about exim and very little about mutt. I use fetchmail, procmail, and elm/balsa/mahogany (depending on the situation). fetchmail : fetches the mail and hands it off to a local MDA procmail : a local MDA that can filter mail to various folders, even other addresses elm : a local MUA , reads mail in mbox folders balsa, mahogany : MUA , can read local folders as well as remote IMAP/POP Some mailers such as balsa, mahogany, netscape provide the ability to retrieve mail from the remote server since it is very common to do these days, however you are not required to configure them for that. fetchmail is a very mature program that only fetches mail and is independent of the MUA. Some acronyms: MTA : Mail Transfer Agent (ie sendmail) : used to transfer mail between servers MDA : Mail Delivery Agent : used to deliver mail to local users MUA : Mail User Agent : the program you use to read the mail HTH, -D On Tue, 12 Dec 2000 19:38:38 Debian User wrote: | Can someone explain what's the relation between exim, fetchmail | and mutt or any other reader? If fetchmail fetches it, then why | the servers have to be defined again for exim or mutt? | Thanks, | A.R. |
Re: HT configure an IDE/ATAPI ZIP drive?
I had this problem too. Not sure what changed on my system to correct it. Actually, since I rarely use it I'm not even sure if it still works. Zip drives are kind of buggy with respect to ejecting. Sorry I can't be of help. -D On Tue, 12 Dec 2000 23:28:08 Dan Griswold wrote: | Hi all, | | I have an internal IDE/ATAPI ZIP 100 in my box. I can mount, | read/write, and umount it. But I cannot eject it. :-( | | Now, I know I've done it before. But that was under SuSE 6.1 (some | time ago). And of course, I could boot into some unmentionable OS and | eject the disk. Ugh. | | So, I figure that I need to configure something properly in the | kernel. I've recompiled the kernel 3 times tonight, and modules too, | and still no luck. | | Anybody know what I'm overlooking? | | Thanks in advance, | | Dan | | -- | -- | Dan Griswold | [EMAIL PROTECTED] | -- |
Re: [OT] Recent increases in d-u posts?
You may be on to something here. I started looking at Debian parly because of all the good things Debian users had to say about it in the other mailing lists (ie gnome-list). Also I upgraded to RH7 =p. (What vendor would be so stupid as to use an incomplete compiler that is totally incompatible with everything?) I haven't yet installed Debian, just looking for the time. :-) -D PS. I am in school too, so that may have had an effect On Wed, 13 Dec 2000 03:43:07 Jaye Inabnit ke6sls wrote: | | Damon, | | I think it's the redhat upgrade to 7.0! heeheehee. | | Also, it could be that so many books are hitting the shelves about linux | and linux+pure=debian. | | Linux in school is also exposing this distro to the next generation. | | Microsoft STILL wants money for that thingy they sell. | | Gosh, this is without actually committing any real thought to the matter :) | | Tatah | | On Wednesday 13 December 2000 00:13, Damon Muller wrote: | | > > Hi folks, | > | > I was just wondering if anyone else has noticed a signigicant increase | > in posts in the last few weeks. It seems that, not that long ago, I'd be | > lucky to get over 100 in a single day. These days it seems to be | > regularly over 200. | > | > It's a bit weird, because it's not as if there are any really long | > threads or flame wars. There doesn't seem to have been a noticable | > change in the sorts of things that have been posted (with the exception, | > maybe, that there seem to be a lot of XFree 4-related posts (and also | > the recent Dr Kroger thing, but that was relatively minor)). | > | > I'm not suggesting splitting the lists, or that newbies should go away, | > or anything like that. I'm just wondering if anyone else has any | > thoughts on the phenomenon. Maybe it's just beacue people are æxpecting | > to have time off over christmas where they can work out this linux | > thing. | > | > cheers, | > | > damon |
Re: Debian is not for me
I think you should browse the archives now. I recall seeing a subject similar to "Re es1371 PROBLEM SOLVED" yesterday. -D On Tue, 12 Dec 2000 22:26:03 Clayton Stapleton wrote: | It is a Sound Blaster 16 PCI sound card. The system BIOS is set to recognize | PnP cards and is used by Win 98 and Suse 6.4 distro's so I do not want to | change it. The sound chip is a Ensoniq 1371, IRQ 11 and i/o=6800-683F. I | have read the PnP Howto and Alsa mini-Howto with no clue on how to get | the sound card recognized. Any help would be appricated. | Clay Stapleton
Re: [OT] Recent increases in d-u posts?
> > >Also I upgraded to RH7 =p. (What vendor would be so stupid > >as to use an incomplete compiler that is totally incompatible with > everything?) > > The type of vendor sellling to people who don't compile stuff. > Maybe they did it as an experiment, perhaps one day they will strip gcc from > their distro completely. > > tka > Maybe that's why I joined this list. :-) Not a smart move on RH's part. Remember the marketing slogan : "Satisfied customers tell others, and that makes for more business for us." and it's counterpart : "Dissatisfied customers tell others, thus hurting our business." -D
Re: serving user's homes with apache
For one thing, the user's home directory must be excecutable (otherwise apache can't cd to it). Also, apache must be configured to know which directory in $HOME is the "magic" one. I believe that public_html is the default. Also, public_html must be both world readable and world executable. (When discussing directories, executable means that you can cd to it, if it is readable, but not executable you can ls it but not cd, if it is executable but not readable you can cd but not ls -- this is what you want ( rwx-x $HOME ) and ( rwx---r-x $HOME/public_html ) HTH, -D On Thu, 14 Dec 2000 07:45:27 Daniel de los Reyes wrote: | Apache is enabled to server users home dirs by default. | However if I place a file in public_html and point a browser to it I get a | permission denied error | | Why is this? | -- | __ | Daniel de los Reyes | S2-Selling Soluciones | Valencia Spain | e-mail: [EMAIL PROTECTED] | Powered by Debian GNU-Linux 2.2r2 | __ |
Re: Porting applications to Debian
It would be best to have a system set up to test on, but if it runs on Unix and doesn't do anything that is system/implementation dependent it should work. What they probably really mean is they would like you to create a Debian specific package they can install. This simply means making the appropriate config file and then building the package (which would require a Debian system). HTH, -D On Fri, Dec 15, 2000 at 01:22:17PM +, Simon Broad wrote: > I'm pretty new to Linux, but have been asked by a customer to port a Unix > product to Debian. > > This may sound daft to a lot of you, but would I need to port the product on > a Debian box or, as I already have SuSE and RH available will a port on one > of these be OK? > > I'm sure there are many other issues than just which distribution is in use. >
Re: Java2
You could try kaffe or gcj. Kaffe is a free JVM implementation and gcj is a Java compiler that can output .class files or native object code. -D On Sat, 16 Dec 2000 19:38:35 Dale Morris wrote: | I downloaded the j2sdk1.3 package from Blackdown, it installed fine, but | doesn't work when I try to setup staroffice. Any suggestions? I'm using | debian 2.2 (libranet 1.8.2 that's been apt-get upgraded). I suppose | there's some permission somewhere that has to be changed, but I'm not | sure where it is. | | thanks in advance | | -- | "If you only have a hammer, you tend to see every problem as a nail." | -- Maslow |
Re: Need advice: Compaq Presarios and Debian
Inform Compaq that M$ isn't the only OS vendor. :-) I have a Compaq Presario (don't get one, it's better to build your own comp from parts) that came with Windows 98 preinstalled. I added RedHat to it (first 5.2 - didn't like my video card, then 6.1 , now 7.0 ). I am planning on installing Debian Potato early next week. I wonder why your BIOS got flashed. You probably should have installed a better BIOS ;-). My BIOS can't boot from the second IDE bus, thus I can't use LILO and must first boot DOS and run loadlin. -D On Fri, 15 Dec 2000 13:51:09 Tom Schuetz wrote: | I had Red Hat installed on my Presario as a dual boot system. It had its own | hard drive, and worked somewhat, but eventually it flashed the BIOS and the | whole motherboard had to be replaced. | | I was told (tersely) by Compaq Support that Presarios are NOT GOOD with | Linux. | | But, I WANT LINUX. Namely, I want Debian. Anybody know about Presarios and | Debian? |
Re: OT - web browsers
On Sat, 16 Dec 2000 03:52:15 Ken Weingold wrote: | | Yeah, I think you are. I am talking about copying a URL from a | terminal window or even the web browser itself. Open a new browser | window and you still have text in the Location bar. | I don't have this problem : I set it to open a new window with "blank" page. I do have to hit the "Enter" key to make it go to the new URL though. Now I recall having a button, but I think it was called "Go", on older versions of Netsape. (this was before I had my own computer and before I knew about Unix) -D
Re: Need advice: Compaq Presarios and Debian
On Sat, 16 Dec 2000 03:04:09 Peter Jay Salzman wrote: | d-man, you might want to set your wrapmargin a bit shorter. What is the preferred size? I had it at 80, this time it's 70. :-) | as for lilo, you can have two devices on each ide chain. i assume | you have | windows on one hd and linux on another hd. why not just make them | a | mast/slave pair on the first IDE chain? Nice try. Here's what happened when I got my HD and added it to install Linux : Original setup, out-of-the-box from Compaq : Quantum Bigfoot, 8GB, /dev/hda (I'll use the Linux naming since it is nice and concise) DVD drive, /dev/hdc Zip drive, /dev/hdd floppy , /dev/fd0 (my floppy came on the floppy connector) My Compaq is a Presario 5035 mini-Tower (very mini!) I bought a 10GB Maxtor drive (3.5"). My tower only has 4 drive bays: HD, DVD, Zip, floppy so I intended to leave the disk somewhere but use it anyways. As it turns out there is just enough room on top of the power supply and behind the floppy drive that the cover fits on and you can't tell from the outside. As for connecting it, the BIOS has no way to manually specify what the disks are (like older BIOS's on my dad's 286 and 486). Since /dev/hdb was open I figured I would put the disk there. After a lot of frustration and rebooting I realized that my BIOS could only recognize my Maxtor drive if it was in "Cable Select" mode. Of course, I don't have the special cable that allows 2 disks on it in "Cable Select" mode. The BIOS wouldn't see it in Master or Slave mode, and then the BIOS couldn't find the other disk on the bus either. I tried all combinations of disks together with the Master/Slave configs. Since the Maxtor drive had to be in Cable Select mode it had to be by itself on a bus. The BIOS doesn't care about Zip drives so putting it with the zip drive was equivalent to being by itself. I moved the DVD drive to /dev/hdb and made it a slave while the 'doze disk is still /dev/hda. My current setup: orig hd : /dev/hda dvd : /dev/hdb Linux hd : /dev/hdc zip : /dev/hdd | | boy, everytime i open her computer, i always get the feeling that | compaq | didn't just put any thought into working on your computer. they | specifically | designed it to discourage you to! Yes definitely! Not only that, but they put everything except the modem on the mother board. You can't upgrade or even just replace the video or sound without needing a new mother board. As for booting, someone else suggested using a floppy. Yes, Linux can boot fine with LILO on a floppy, but the floppy had no /dev/hda file so I couldn't ever boot that other OS. That's why I settled for loadlin. Fortunately I don't have the need to boot the other OS very often these days. -D
Re: mailagent question
I don't know anything about mailagent, but I use procmail (in a very
simple way). Here's a much shorter introduction than the man pages
give:
my procmailrc file looks like:
:0:
*
ex:
# debian-user
:0:
* [EMAIL PROTECTED]
lists/debian-user
You can also replace the with an e-mail address to send the
message to. (BTW, '#' starts a comment like python and bash)
HTH,
-D
On Sat, Dec 16, 2000 at 06:41:26PM +0100, Robert Waldner wrote:
>
> Hi!
>
> I can´t get mailagent to forward or bounce any mail. My latest try
> looks like:
>
> Content-Type: /application.*octet.*stream.*/i {BOUNCE [EMAIL PROTECTED];
>
> Out of the man-page this should work, but mailagent simply ignores it...
>
> (If anyone has written a script converting mailagent-rules into
> procmail ones I´d be grateful also cause I´ll switch sometime next year,
> but in the meantime I´m stuck with mailagent...).
>
> TIA,
> &rw
> --
> / Ing. Robert Waldner | Network Engineer | T: +43 1 89933 F: x533 \
> \ <[EMAIL PROTECTED]> |KPNQwest/AT | Diefenbachg. 35, A-1150 /
>
>
Re: New IDE drive w/o rebooting?
I'm assuming the machine is a PC. Usually you can't add IDE disks without rebooting, also I'm not sure it's a good idea to mess with the cables while the power is on. IDE disks are identified by the BIOS at boot time. Zip IDE disks are an exception : the BIOS does nothing to support them. As far as the BIOS is concerned, ZIP drives don't exist. (See my post in the archives relating how I got my Linux HD to work). You may be able to get away with it with a zip drive, but if the power destroys it, I am not liable. ;-) It shouldn't be so bad to reboot the machine, it should only take a couple of minutes. HTH, -D On Mon, Dec 18, 2000 at 11:52:32AM -0800, [EMAIL PROTECTED] wrote: > Hi, all, > > Is this possible? I need to install an IDE Zip 250 drive in the server here > at > work to make available on the LAN. I seem to recall that you can 'refresh' > the > SCSI bus to add drives without rebooting. > > Is this possible (or even recommended) with an internal IDE drive? > > TIA :) > > Matthew Thompson http://mattyt.net > [EMAIL PROTECTED] http://www.oz.net/~mattyt > "They that can give up essential liberty to obtain a little temporary safety > deserve neither liberty nor safety." - Benjamin Franklin > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Mutt and group reply
Even better than that, mutt understands lists. Put this in your .muttrc: subscribe debian-user@lists.debian.org Then mutt will know that that address is a list. Also, you will want to change the "index_format" string otherwise all messages will have "debian-user" as the name. This is easy to do: go to the documentation for index_format. Put 'set index_format=""' in the .muttrc. For put the default string except change the "L" to "F". When you want to reply to a list message use "L" for list-reply. Then mutt will only mail the list (so people don't get multiple copies of each reply) and set a header option so that you don't get put in the "To" or "CC" of replies (since you are on the list). HTH, -D On Mon, Dec 18, 2000 at 11:47:07AM -0800, Rob Hudson wrote: > Is there a command for .muttrc that tells mutt not to include me in > the group reply (pressing g on a message)? > > Thanks. > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Stupid question
I have an ethernet connection at school. At boot time root does the "ifup eth0" to DHCP boot the card. Services such as sendmail, apache, ftpd, etc. run as they should and work fine. I believe each service has a user/group associated with it that has limited permissions. As long as the users/groups are allowed to use the network interface it shouldn't matter which user actually created that interface. I recall an option when configuring the network to allow users other than root to bring up/down the interface. I chose 'no' so that no one else could boot me off the network. As a result I had to su to root one time when the card couldn't DHCP boot (some external problems) to bring the interface back up when the network was fixed. HTH, -D On Mon, Dec 18, 2000 at 09:21:02PM +0100, Gary Jones wrote: > Okay, stupid question time. > > What is the best way of connecting to the 'net? I don't mean the > mechanicals, which connection type to use, that sort of thing, but > rather which account(s) should do so. Preferably I don't want to > connect as root, but some things (e.g. collecting mail or news) might > be better done as root or might /need/ to be done as root, or at > least some specific user with the right permissions which might be > different for the different tasks. What's the best thing to do? I've > never really seen a decent discussion about this, since I started > fiddling about with Linux (on and off, about 2 years). > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Removal from list
An alternate solution, since I assume you want the mail once you get back from your trip, is to use a filter program to send all mail to this list to /dev/null. procmail, mailagent, and filter can do the job. HTH, -D On Wed, Dec 20, 2000 at 10:28:50AM -0500, Eileen Orbell wrote: > Hi all, > > I read a few weeks back the many attempts the Professor had made to be > removed from this list and through his final frustrations he GOT a little > mad. Anyway, I am leaving the country for 3 weeks tomorrow and felt it was > better to unsubscribe rather than come back to 3 thousand plus emails. So > following the instructions provided at the bottom of the debian email: > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". > > I did just that. But oh dear here is the reply: > > You have not been removed, I couldn't find your name on the list. > What I did find were the following approximate matches: > 1121 [EMAIL PROTECTED] 32752 [EMAIL PROTECTED] > 1121 [EMAIL PROTECTED] 25797 [EMAIL PROTECTED] > > But wait those are my email address's h Confused ? Yes I am... > > > > > Eileen Orbell > Software & Internet Applications > Capitol College > mailto:[EMAIL PROTECTED] > mailto:[EMAIL PROTECTED] > This is Linux Country. On a quiet night you can hear Windows 98 reboot!" > > > > > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Install/Config q's
Last night I installed Potato using the 'network' method. It was very nice! (especially with my 100Mb/s connection). Debian has more packages in the distro than RH! Very cool! I found some debian packages that I hadn't been able to find rpms for previously. The only problem is that many of the packages are old (ie gnome-*, sawmill, python, etc). When was Potato released? (that may be the problem and woody might have the newer packages) How unstable is woody? When I looked at the inittab file, it had a comment saying that runlevels 2-5 are mutlti-user. Ok, but not enough information. I have been using RH for 2 years, and it has runlevel 5 for X and 3 for full multi-user. Does Debian use the same runlevels? I know some distros use different nubers than RH. (The inittab should explicitly list each runlevel) I'll probably be back later with more configuration questions, but that's all that's on my mind for now. -D
Re: how to reply to messages
On Thu, Dec 21, 2000 at 03:47:56PM -0200, Christoph Simon wrote: > This seems to be for people having a reply-group or reply-all > button. So they can choose to reply only to the sender or also to the > list. > I really like this feature of the list. I used to wonder myself, but I heave learned the usefulness of it. Another list I am on has the list as the Reply-To. If I select "group-reply" then the list gets 2 copies, a To and Cc. Now I'm using mutt and I use its list feature so only the list gets a message sent to it and I don't get extra copies. Sometimes, however, people ask a question, but aren't on the list. In that case they need the extra cc. -D
Re: Quick tip on transforming a tarball into a .deb
Actually the options are: -taBuild binary and source rpm from tarball -tbBuild binary rpm only from tarball -baBuild binary and source rpm from given spec file (source must be where the spec wants it) -bbBuild the binary rpm only from the given spec When a working spec file is provided by the maintainer, building an rpm is very simple. Are the debian rules significantly different than those given in a spec file? Would it be feasible to make a packaging tool that would make a proper .deb if given the source and a spec file? I just think it would be a very good idea for all involved if a uniform description file was to be used for all (major) packaging systems. This would allow developers to maintain only 1 file and allow users/packagers to build any sort of package (debian or rpm) from the source. Just my $0.02. -D On Fri, Dec 22, 2000 at 06:15:40AM +0800, csj wrote: > I need to compile some stuff in Debian (available in Debian but three version > numbers stale). I have already successfully compiled them in Mandrake. > > The problem is that the application's various bits and pieces will be > splattered across my system. While they'll probably wind up in /usr/local, I > want a more manageable method of installing/deinstalling these files. I'm > thinking of creating a .deb to "trap" the compilation. > > Can someone give me a quick tip on how to do this, convert a tarball into a > deb in one fell swoop? I'm a bit lazy to RTFM. In Mandrake I simply type > something like rpm -bb (or is it rpm -tb?) to produce a binary from a > .src.rpm (I know, not quite a tarball). What's the quick and dirty Debian > equivalent?
Re: Install Debian 2.2 from one floppy and one harddrive folder
I'm not sure what the difficulty is. I recently installed Debian using a network install. You can't do an install with just 1 floppy, but will need 2. The kernel resides on 1 floppy, the root directory on another. I haven't actually gotten into the root floppy, but from what I understand it just contains the basics like a shell, apt, file utilities, and an http client (for the network access). So make 2 disks from Rescue.bin (to boot from) and Root.bin (the root directory for the install system). Then boot with the rescue disk. It will ask for the root disk once the kernel is loaded. Put the root disk in and the rest of the install system will load. Then it will walk you through the installation. You said you are familiar with a disk install so I assume you know how to work out the details of it. I would imagine (though haven't tried it) that it would be the same as a network install except that instead of specifying a network interface and a host to get packages from you would specify a disk. HTH, -D On Thu, Dec 21, 2000 at 05:24:15PM -0500, Bart Szyszka wrote: > Hi, > > I'm having trouble figuring out how to install Debian 2.2 with just a Rescue > disk and with the rest of the files needed > for a base system to be put in one folder on a DOS partition. This is how > I've always installed Debian slink, but would > like to start out with Debian potato on a different computer. I'm reading > through the install guide at debian.org, but > all I'm finding is multi-floppy installations or installations that make you > retain a multi-folder directory structure. > I'd appreciate some help with this. I tried getting most of the folder that > has 'images-1.44' in it and I tried making > a Rescue fisk from the rescue.bin in images-1.44, but when I try the disk and > have it boot the installation, it asks > me for a Root disk afterwords. I just want a single floppy to kickstart the > install and have it grab everything else > from my harddrive. Is that possible with Debian potato? Or what about woody? > > (please CC: me at [EMAIL PROTECTED] since I'm subbed to the digest... I'm > visiting home for three weeks from college and > need to use web-based e-mail) > > -- > Bart Szyszka ICQ:4982727 [EMAIL PROTECTED] >
Re: swap space and memory
I have 64MB RAM and 256MB swap. When I run free it tells me I have 61MB RAM and 244MB swap. Interesting. Also, I had read in a How-To that Linux wouldn't use more than 128MB of swap in a single partition. Before I relayed that info, I though I would test it (I had only recently expanded my swap beyond the 128MB size). I got it to use 241MB of swap (with free telling me I had 2MB free). Must be outdated info in the how-to. However, I think that the system is really using all the memory available. It must have something to do with the way free gets its information. (but I have no proof) -D On Thu, Dec 21, 2000 at 05:46:48PM -0800, Denzil Kelly wrote: > When I partitioned my hard drive, I noticed that I was > unable to make a swap partition of 128 MB. For > whatever reason, the largest partition I was able to > make was 122 MB. Why is this? Also I've noticed that > all of my memory isn't detected either. Why is this? I > have 128 MB installed, but this is what is reported by > free > > total used free sharedbuffers > cached > Mem:1241203 45 17 42 > Swap: 122 0 122 > Total: 246120 125 >
Re: swap space and memory
I think you are right nate. When I started with Linux a couple years ago, the 2.0 series was current (like 2.0.36 or something). It was then that I read in a how-to that even if you make a really big swap partition, only 128MB would be used. Last night I tried to confirm this with my current 2.2.16 kernel but was able to use all 244MB available. -D On Fri, Dec 22, 2000 at 12:41:33AM -0800, Nate Amsden wrote: > did you install this on debian 2.1? or anything using the 2.0 kernel? > from what i remember this was a kernel 2.0 limitation. 124MB > reported by the system is normal. 124*1024=126,976kB , add > some more for the space the kernel takes up when it loads into > memory and any memory used by modules when the system loads > and you have 128MB. also bios settings such as shadow can > effect how much memory is available. > > nate
Re: I need a filter for Epson Stylus Color 400
What is the mechanism for setting up a printer on Debian? I installed Debian recently but haven't had the time to configure it yet (so I'm still using RH). In RH I run printtool as root and it handles the details of printer config. It lists filters for Epson printers "Epson Sylus 800 & ESC/P 2 printers" and "Epson Stylus Color (UP)". If either of these are close enough to your printer I could set it up on my system and send you the config files. (it might help you figure what you need on your system for it to work) -D On Fri, Dec 22, 2000 at 01:23:34PM +0100, Santiago Fernandez wrote: > I need printcap entries and a filter (maybe from magicfilter) for a > Epson Sylus Color 400. Has anyone some experience configuring this > printer? I've tried different possibilities without success. > > I've read README-StylusColor from magicfilter but it doesn't offer any > help: I know I should buy another printer, but I've spend all my money > on CD Technology, and a friend of mine has given me this ridiculous > printer. So, that's all I've got to print my documents at home. > > Thanks in advance, > > -- > Santi > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to track *part* of unstable?
I don't think that can be done. Woody (unstable) is based on glibc 2.2 while potato is based on glibc 2.1. Thus any binaries for woody that use the C library at all won't run (even if you get them to install) on potato. You could probably upgrade the libc independent of the rest of the system though (since it is binary backwards compatible). The other thing you will want to be careful of is dependencies. Gnucash depends on guile and maybe a couple of other libs. Other apps/libs may also depend on those libs. If those needed libs aren't backwards compatible you may inadvertantly break other parts of your system. I think it would be best for you to try and find a source package and build it on your system. If gnucash itself (not the prebuilt package) doesn't require newer libs this is the best route. -D On Fri, Dec 22, 2000 at 01:35:28PM -0600, Richard Cobbe wrote: > Greetings, all. > > I'm still fairly new to debian, so I'm not all that familiar with dpkg, > dselect, and apt. > > Is there a way to track the unstable branch for only certain packages? I'd > like to install the unstable version of gnucash. I downloaded the .deb and > tried an apt-get install, but it failed due to a number of unsatisfied > dependencies. Most of these dependencies simply required later versions of > packages that I've already installed---from potato. > > Ideally, I'd like to be able to use dselect to grab this version of > gnucash and satisfy its dependencies, but I want to leave the rest of my > system at stable, where it is now. Is there a good way to do this? > > The only thing I can think of is keeping two sources.list files, one for > stable and the other for unstable, and switching them out as necessary. Is > there a better way? > > (Also, and forgive me for asking this, but please CC me directly on any > responses. I've subscribed to this list and received confirmation of my > subscription, but I'm not receiving any of the list traffic, and I'm not > sure why.) > > Thanks much, > > Richard >
Re: Starup script - how can I make it start later? (Firestarter)
I haven't done this yet, but I have a feeling that when I finish installing Debian I will be playing with the rc?.d scripts. The scripts are actually symlinks to the real scripts. The symlinks have a name like [KS][0-9][0-9][a-zA-Z]+ The symlinks that start with S are for startup and the ones with K are for shutdown. The 2-digit number tells the system which order to run the scripts. The text after the number is the name of the service. If you increase the number, it will start later. It would be a good idea to read the docs first. There is a program called sysconf or some similar name that is used to keep the init system in sync. Check the archives or wait for the other gurus to explain it better than me. HTH, -D On Fri, Dec 22, 2000 at 07:34:39PM +, Phillip Deackes wrote: > I need to start a script containing ipchains settings derived freom > Firestarter. The docs suggest I start the script in /etc/rc.x. If I do > this, on boot up I get error messages which do not occur if I run the > script after the system has rebooted. This leads me to summise that the > script is starting too early. How can I make sure it is the last thing to > run on boot up? > > Anyone else using Firstarter? The script is firewall.sh and contains > ipchains references. > > I can try it again and copy over the error messages I receive on boot, but > I don't know how to get these messages again to copy - they appear during > boot up but they are not in /var/log/messages or /var/log/system. > > -- > Phillip Deackes > Using Storm Linux 2000 > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: What is the file name of the ps2 mouse driver?
Maybe there isn't a separate mouse driver file but it is in some other
file instead?
-D
On Fri, Dec 22, 2000 at 03:31:45PM -0600, John Foster wrote:
> Thanks guys for all of the pointers. I am really not a newbie and have
> covered all of the areas responded to. I do not find a mouse driver in
> my /lib/modules/kernel-2.2.13/misc or any of the other dirs here. There
> should be a driver ps2mouse.o or the like created when the kernel is
> compiled and installed. I used the make-kpkg buildkernel command and
> then installed the resulting kernel-**.deb files. This has worked for
> several debian generations. I checked the kernel distributions from
> kernel.org & did not find a ps2mouse driver in the tarball at all. Is
> there one? Am I missing something.
> My current gpmconfig file:
>
> device=/dev/psaux
> responsiveness=
> repeat_type=raw
> type=ps2
> append=""
>
> My current gpm-root config file:
> button 1 {
> name "ttys"
>
> ""f.nop
> "login on a new tty" f.mktty
> ""f.nop
> "tty 1" f.jptty "1"
> "tty 2" f.jptty "2"
> "tty 3" f.jptty "3"
> "tty 4" f.jptty "4"
> "" f.nop
> "tty 5" f.jptty "5"
> "tty 6" f.jptty "6"
> "tty 7" f.jptty "7"
> "tty 8" f.jptty "8"
> ""f.nop
> "more of them..." {
>
> "tty 9" f.jptty "9"
> "tty 10" f.jptty "10"
> "tty 11" f.jptty "11"
> "tty 12" f.jptty "12"
> "" f.nop
> "tty 13" f.jptty "13"
> "tty 14" f.jptty "14"
> "tty 15" f.jptty "15"
> "tty 16" f.jptty "16"
> ""f.nop
> "more of them..." {
> "tty 17" f.jptty "17"
> }
> }
>
> }
>
> button 2 {
> name "system status"
> foreground red
> background black
> border yellow
> head bright yellow
>
> "" f.nop
> "%b %d %Y" f.time
> "%H:%M" f.time
> "" f.nop
> "load: "f.load
> "free:" f.free
> "" f.nop
> "report disk usage to ~/du" f.bgcmd "du ~ | sort -rn > ~/du"
> "print mail headers to tty" f.bgcmd
>"grep '^From ' /var/spool/mail/$USER | tail"
> }
>
> button 3 {
> name "far ttys"
>
> foreground black
> background red
> border bright yellow
> head bright yellow
>
> "tty 9" f.jptty "9"
> "tty 10" f.jptty "10"
> "tty 11" f.jptty "11"
> "tty 12" f.jptty "12"
> "" f.nop
>
> "tty 13" f.jptty "13"
> "tty 14" f.jptty "14"
> "tty 15" f.jptty "15"
> "tty 16" f.jptty "16"
> "" f.nop
>
> Excerpt from my current Xf86config file:
>
> Section "Pointer"
>Protocol"PS/2"
>Device "/dev/mouse"
>Emulate3Timeout 50
>Resolution 100
>Buttons 3
>Emulate3Buttons
> EndSection
>
> Keep in mind that the mouse in X; works, but Xterm & console does not.
> Also all of this hardware was working on a previous system that had been
> upgraded from hamm, bo, slink, to potato. I installed 2 new hard drives
> and decided to do a clean install of Debian 2.2r2 recently downloaded
> from the debian.org website.
> ANY other ideas please
>
> Thanks!
> John
>
>
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: backingup /home/foo/.* files
If you really like plain vi features, vim has a compatibility mode: :set compatible to set all options at values that are (approximately) equivalent to vi's features. -D (vim has a lot of nice features that vi doesn't have, and it's cool to take advantage of them)
Re: 'sendmail' option -commented- in Muttrc but working??
Hmm, not sure about what you meant in your messages, but I'll try to help anyways. I use mutt on my linux box with sendmail and at work with cygwin using ssmtp as the MTA. On my linux box I don't have the "set sendmail" option in my mutt rc file. It uses the default value given in the manual and works fine (since I have sendmail). At work I have the sendmail option set to "ssmtp" (with some ssmtp options in the quotes). If you have the option commented (usually said "commented out") that is equivalent to not having it at all. ex: # set sendmail="foobar" is commented out and will have no effect. It seemed like you said sending mail worked fine like this. Try this command in a shell: ls -l `which sendmail` It may be that the exim package installs a symlink called "sendmail" that links to exim. If so, then when mutt tries to exec sendmail it will succeed, but actually exec exim instead. Hope this helps. :-) -D PS. I would guess that english isn't your native language, and I don't want to you feel insulted after reading my message, I just don't want to assume anything since that results in misunderstandings (and even more confusion on your part when it doesn't work the way you want it to)
Re: Confusion over library names
When libraries are named, the developer picks some sort of version number for it. That version number goes after the .so part of the filename. My guess for that libstdc++ you have is that it is for g++ 2.7.2 (old!). You probably have another one for 2.95.2. If you need to build a program yourself, you should have a symlink to the correct version of the shared library file. If you change what libraries are installed on the system or the path for them I think it is good to run ldconfig (packages should do this automatically). ldconfig looks for libraries and tells ld where they are. Then ld just uses it's config to find the library when it needs to. I'm not really too clear on the exact details here, but the gist of it should be correct. (If not please let me know so that I can correct my understanding!) HTH, -D :-)
Re: newbie: 3c905C eth0 card not detected
Are you sure about that? The question was about a 905 series card, not a 590 series card. The 905 series are PCI cards, and I believe the 590 series (or was it 509 series?) are ISA. Both 3Com and RedHat say that the 3c905b card does NOT work with Linux. I don't know about the 3c905c. I bought a much cheaper LinkSys card instead and have been very happy with it. -D On Fri, Dec 29, 2000 at 09:46:03AM -0500, Carl Fink wrote: > On Fri, Dec 29, 2000 at 07:12:36PM +0530, M K Saravanan wrote: > > > > I have been using RH linux. after reading several success stories of debian, > > decided to switch over to deb. I installed Debian2.1. It is not detecting > > my > > ethernet card [3Com 3c905c 10/100]. how to activate it? > > Personally, I did > > insmod 3c59x > > as root. > -- > Carl Fink [EMAIL PROTECTED] > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: newbie: 3c905C eth0 card not detected
On Fri, Dec 29, 2000 at 08:55:04PM +0100, Robert-Jan Kuijvenhoven wrote: > At 13:49 29-12-00 -0500, you wrote: > >Both 3Com and RedHat say that the 3c905b card does NOT work with > >Linux. > > 3c905b does not work, they say? Take a look at this: I'm glad it works for you. It was a few years ago that I recall reading on the web sites that it didn't work. In any case I am glad that I thought it didn't since the Linksys card was many times cheaper! (the Linksys was ~$20 while the 3Com was about ~90) :-) -D
Re: Java???
On Sat, Dec 30, 2000 at 06:18:16PM -0500, Michael P. Soulier wrote: > I thought that the JRE was depricated. That's what I read on the Java > homepage anyway. I highly doubt that. The JRE is just the VM without the compiler. You can't deprecate the VM for an interpreted language :-). -D
Re: lilo- from Sarah
Are you trying to boot from the second IDE bus? My Linux disk is on the second IDE bus in my computer, and my BIOS is too crappy to boot from it. It would always hang with "LI". I had to use loadlin.exe instead. If you want more info about loadlin, jusk ask. HTH, -D On Sat, Dec 30, 2000 at 05:31:45AM -0800, samtara leumas wrote: > I am new debian user (and a newbie)- My debian distro is not booting > properly with LILO- when I turn on the computer it gets to LI and nothing > else- have to then turn off the computer and boot from the CDRROM. I can't > find the LILO file to alter it! Which directory is it in? > > Also- I have yet to learn how to subscribe to USENET- I've read all about it > but how do I subscribe? >
Re: shell to /dev/ttySxx
Try: $ cat > /dev/ttySxx If you don't give any args to cat, it reads stdin. (aka, "interactive") HTH, -D On Sat, Dec 30, 2000 at 07:15:47PM +0100, Matthieu Paindavoine wrote: > Hello, > > I have a device connected to a serial port. There is no protocol, I just > send out characters. > I would like to know how to have a shell where every thing I typed is > sent to this device once I type [enter], pretty much like irc. > Currently I can output files to this device (Cat xxx > /dev/ttySxx), but > I am looking for a more interactive mode. > > Thank you, and have a happy, happy, New Year! > > Matthieu >
Re: remote x via ssh question
This thread has invoked some curiosity in me. If I use ssh to forward X connections, does that mean I can use X through an IP masquerading router? I cannot use X in the normal way right now (by setting DISPLAY to my IP address) because my IP doesn't really exist beyond the router. Thanks, -D
Re: C compiler.
When compiling C/C++ code you must tell the compiler exactly which libraries you are using. The first step is with the preprocessor to include the header files. The second step is with the linker to tell it where the functions/classes are actually implemented. If you really want to use gcc instead of g++ for compiling C++ code make sure you specify the options -Xc++ and -lstdc++ on the command line. The g++ wrapper does this for you. HTH, -D On Tue, Jan 02, 2001 at 06:09:56PM -0800, Xucaen wrote: > huh??? it doesn't mention anything about this in > the man pages (man gcc) > why does it do this? I am having similar problems > with iostream.h >
Re: OT - Virus?
On Wed, Jan 03, 2001 at 09:26:20AM -0900, Ethan Benson wrote: > NOTICE: you have just been infected with Cooperative UNIX Email Virus, > to cooperate please run rm -rf / as root. Thank you for your > cooperation. > > ;-) > My brother got a similar e-mail from a friend of his once. It said "This virus works on the honor system. The subject said "This is a virus". Please send this to everyone you know, then delete all the files on your HD". :-) -D
Re: Official Debian
Potato is the current stable (version is 2.2r2) but a large number of packages are out of date. Woody is 'testing' now (not unstable any more) so I expect that it is now quite reasonable and will be 'stable' soon. If you don't need to install right away, I would recommend waiting a while and getting Woody instead. (though I doubt you would find a book with woody bundled in it, an older book should work just fine ; also there's this list :-)) HTH, -D
Re: Memory leakage
On Wed, Jan 03, 2001 at 06:05:13PM -0800, kmself@ix.netcom.com wrote: > > $ fuser -v bigfile# show process(es) using bigfile > $ fuser -vk bigfile # kill process(es) using bigfile > $ fuser -v bigfile# verify the kill worked > $ cat /dev/null > bigfile # "empty" the file > $ rm bigfile > > In general, you want to *empty* a file (cat /dev/null > file) before you > delete it, and you don't want to delete an open file. > Why would you cat /dev/null into the file before removing it? -D
boot problems
I can't boot Debian the right way. I use loadlin.exe to boot, thus I copy my kernel images to C:\ so that loadlin can find them. This has worked fine for me with RH6.1 (custom kernel) and RH7.0 (stock kernel). I copied my Debian kernel to C:\, but loadlin tells me it's not a kernel. I can boot the Debian system if I use the RH kernel (2.2.16). Debian Potato, 2.2.17 idepci kernel Any ideas as to why loadlin doesn't think the debian kernel is a kernel? Also, what is the proper command to upgrade the kernel? I want to upgrade to the "normal" kernel (I used the idepci kernel to do a network install) and if I can I want to upgrade to 2.2.18. Thanks, -D
Re: Network Throughput
On Fri, Jan 05, 2001 at 01:36:15AM +0100, Philipp Schulte wrote: > Could you please give me an URL, where this is defined? > I searched the SI units but it does not mention any difference between > MB and Mb. > Phil > I don't know where this is defined, but this is a consistency I've noticed as well. Your 56k modem is 56 kb == 56 kilobits. -D
Re: help debugging a compiler error (cpp)
On Fri, Jan 05, 2001 at 01:02:01PM -0500, Jonathan Markevich wrote: | | --- | bash-2.03$ make | rm -f nv.o os-interface.o os-registry.o Module-linux NVdriver | cc -c -Wall -Wunknown-pragmas -Wno-multichar -O -D__KERNEL__ -DMODULE | -D_LOOSE_KERNEL_NAMES -DUNIX -DLINUX -DNV4_HW -DNTRM -DRM20 -D_X86_=1 | -Di386=1 -D_GNU_SOURCE -DRM_HEAPMGR -D_LOOSE_KERNEL_NAMES-I. | -I/usr/src/linux/include nv.c | cpp: -lang-c: linker input file unused since linking not done | cc: installation problem, cannot exec 'cc1': No such file or directory | make: *** [nv.o] Error 1 | --- | | Does anyone know what I need installed here? Thanks. I don't know what you need for a debian system, but the Solaris system's in the lab have cc1. It is part of Sun's C++ compiler. -D
Re: Debian and Pentium 4
www.linux.com/hardware/newsitem.phtml?sid=26&aid=11457 Very interesting article. Shows why working together and free updates are important to quality systems. -D On Fri, Jan 05, 2001 at 12:37:39PM -0700, Ray Percival wrote: | I can't get to linux.com just now to find the article but basically | it is Microshaft's fault. It seems that had Intel used the same | versioning for cpuid that they have used up untill now it would | have broken NT/2000 in a very bad way. (All the details are in | the linux.com article I can't seem to get to right now) In any case | Intel hacked the P4 to not break winders but they broke everything | else in the meantime. This is one of the reasons why I will only | use AMB for x86 boxen. When I can find the link I'll post it.
boot problems take 2
Does nobody have any suggestions? I see lots of helpful suggestions on the list daily, but no comments on my problem yet. Should I forget about Debian and continue with RedHat? Please offer suggestions, even if it is a wild guess. Anything would be helpful. -D On Wed, Jan 03, 2001 at 09:40:42PM -0500, D-Man wrote: | | I can't boot Debian the right way. I use loadlin.exe to boot, thus I | copy my kernel images to C:\ so that loadlin can find them. This has | worked fine for me with RH6.1 (custom kernel) and RH7.0 (stock | kernel). I copied my Debian kernel to C:\, but loadlin tells me it's | not a kernel. I can boot the Debian system if I use the RH kernel | (2.2.16). | | Debian Potato, 2.2.17 idepci kernel | | Any ideas as to why loadlin doesn't think the debian kernel is a | kernel? | | Also, what is the proper command to upgrade the kernel? I want to | upgrade to the "normal" kernel (I used the idepci kernel to do a | network install) and if I can I want to upgrade to 2.2.18. | | Thanks, | -D
Slow responses
Just an FYI, it seems that my outgoing mail takes quite a while to reach the list. (I don't believe it is the list's fault) My message Re the c++ compiler was sent several hours ago, but just arrived in my folder. After sending my message I received the message from other people sent around the same time as me (even after!). This is just to avoid any flames about being behind schedule and just mimicking other peoples messages. (I have already received such a message) -D
Re: GCC and EGCS
As I understand the history, the egcs people had a disagreement with the gcc people so they split off and continued to develop gcc on their own. Later, egcs was in a better state than gcc so the gcc people got together with the egcs people and merged the two compilers to make a best of both techniques compiler. Thus the egcs split no longer exists and gcc 2.95.2 is the latest compiler. I'm not sure, but I think the kernel was built/tested with gcc 2.7.2 which is why they recommend it. However, I have read comments that say for kernel 2.4 gcc 2.91.6 is recommended and 2.7.2 is no longer supported. I have built my own kernel without trouble before, on a RH 6.1 system (I think it had egcs 1.1.?). Other people have commented on building their kernels with gcc 2.95.2 without trouble. I had upgraded my RH system to gcc 2.95.2 without any major trouble (once I got around having all the rpms correctly installed). I think you will be fine with gcc 2.95.2. Just don't try and use 2.96 ;-). HTH, -D On Sat, Jan 06, 2001 at 05:00:14PM +0100, 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 | 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! | | Could someone solute my confusion? | | Thanks for every help. | | cheers, | Raffaele
Re: boot problems take 2
Thank you Brian and Gary for replying. It is working now. I recently discovered rsync and used it to copy my home directory from RH to Debian. Just now I used it instead of cp to put the kernel image on C:\ and it is working now. Apparently cp messed up the file. Now I'm upgrading to 2.2.18 kernel and configuring the modules for it. I also had to reconfigure exim and mutt so that my return address would be right and the list would accept my messages :-). This message is coming from debian. -D
