Re: scsi-partitioning
On Wed, Oct 03, 2001 at 03:25:11PM +0200, Gerald Richter wrote: > > - cfdisk on /dev/sdb tells me: > FATAL ERROR: Bad signature on partition table > Press any key to exit cfdisk > > how do I proceed if I want to partition sdb? > > I have no idea why you are getting the error. Have you tried the "fdisk" command? I have seen it work in cases where cfdisk fails. Mike
Re: rdate can't connect socket
On Thu, Sep 13, 2001 at 08:54:55AM -0400, Bob Hilliard wrote: > I have the following command in a cron.daily script: > rdate time.mist.god > > This has worked consistently for a long time, but for the past > week or two it returns: > rdate: Could not connect socket: Connection refused > > This message returns almost instantly - too quickly, I believe, > to be a connection refused at the remote server. > > Can this be a socket on my machine that refuses a connection? > I cant say whether your theory is correct or not. It sounds plausible. However, I did have the same problem a while back. I am pretty sure it the remote server refusing. I resolved the problem by installing the nptdate package, and using my name server as a time host also. Mike
Re: a woodied gnumeric
On Tue, Aug 21, 2001 at 11:17:43PM -0700, Bill Wohler wrote: > Any word on when woody will have an installable gnumeric? > Gnumeric has been updated in sid a couple of times in the past few days. I am told that sid -> woody typically takes two weeks if all goes well, so maybe in two weeks. Mike
Re: OT: AMD chips cause kernel errors and hangs?
On Tue, Aug 21, 2001 at 11:59:06PM -0600, Rick Macdonald wrote: > > We have a Linux cluster of 1000 nodes. I wasn't involved in setting it up. > They use RedHat 6.2 kernel 2.2.19. Dual AMD 1.2GHz, 2GB memory, 2GB swap, > GB ethernet. > Quoting the latest Kernel Traffic (kt.zork.net), which summerizes the kernel development list: 'But Alan Cox pointed out, "Athlon SMP will actually not always work with 2.2.' I suggest pointing this out to the one who *did* set up the cluster, and perhaps persuading him/her to move to 2.4 and a distro that supports it. As for your hardware, I am jelous. Where do I sign up for an account? :) Mike
Re: Where to put stuff .. FHS
On Mon, Aug 13, 2001 at 08:15:03AM -0500, Kent Tenney wrote: > Howdy, > > I'm a newbie. > > I am installing potato on a box to use for a photo stock > house/ web / graphic design operation. > > I will be maintaining a fairly large and constantly increasing > collection of image files, as well as accounting stuff, metadata on > image files, directories for client projects etc. > > I've had a look at some FHS info and am still unclear where > the preferred location for my data is, and what the implications > are for partitioning disks. > > I prefer not to restructure later on. > > I would appreciate any help on how to set up for flexibility > in backing up, upgrading hard drives etc. > I think you should regard your image and accounting data as user data. Hence, the only guidance that FSH provides for this is that it goes somewhere under /home/joeuser/. Your image data will probably be the most valuable and space consuming data on your system. Plan your backup stratagy around that. Ideally, you should be able to backup your largest partition to one tape. You dont have to do this, but it makes things easier. Since 650 MB is prolly too small, you cant use cdrw, you need to spring for a tape drive. Big tape drives cost much more than big hard drives these days. Worse, tape drives are usually promoted using their "compressed" capacity, which is generally 2X their native capacity. Most of your data will be image files, which (I assume) is already compressed. In this case, the tape drive's hardware compression could actually make the backup file *bigger*. Make your purchase decison based on native capacity. I suggest you proceed as follows: - Decide on a tape drive. - Partition your system as follows: one partition to be mounted as /home, sized according to the native capacity of your backup drive. one swap partition a partition for everything else Well this is how I would do it. I am sure that there are plenty of other opinions and philosophies around. I hope this helps. Mike
Re: Unidentified subject!
On Mon, Aug 13, 2001 at 05:48:21AM -0500, gary p largess wrote: > Dear debian; > I would like to learn something about Linux, but I finding it very > frustrating, I have visited dozens of sites looking for a OS, and they > all seem to be too small (1 or 2 floppies) or too big (over 300 M), or > they simply don't specify the systems minimum requirements. > > I do not want to experiment with the computer I use for work/school, but > I have a "project" computer I would like to install some version of Linux > on. > 133 Mhz Pentium > less than 300 Mb available hard drive > 8 Mb mem. > All I want is a version of Linux I can install on this machine, that has > a graphic interface (I have old/bad memories of DOS), I'm even willing to > pay a small amount. > You will have a hard time getting Linux's graphical environment (X Windows) to work on such a machine, if it can be done at all. Not a job for a beginner. I think you can learn a great deal about Linux by installing a base Debian system and working with it. This can be done entirely by floppy. A more complete installation will require cdrom or some kind of network connection. Mike
Re: Cloning servers
One way to create a "mostly" identical server (in the sense that it has the same packages installed as the original) is to do dpkg --get-selections > mydebs on the original server, and dpkg --setselections < mydebs on the cloned server. (Or, if you are extra clever, pipe this over the network :P ). This, obviously, requires no shutdown. Mike On Thu, Aug 09, 2001 at 11:41:47AM -0300, Jordi S . Bunster wrote: > > I've got a Debian box running a lot of services, so there's lots of > disk activity. I was thinking about cloning it (the server) to a > redundant one, identical to the other. Just wondering Is there a > way to do this in a sane way while the system is running? Or should I > just power off and use dd? >
Re: potato to sid upgrade error 32 broken pipe
> --- > 90 packages upgraded, 17 newly installed, 0 toremove and 3 not upgraded. > > 3 packages not fully installed or removed. > Need to get 0B/19.5MB of archives. After unpacking 13.0MB will be used. > Do you want to continue? [Y/n] > perl: error while loading shared libraries: libdb.so.3: cannot open > shared object file: No such file or directory > E: Write error - write (32 Broken pipe) > E: Failure running sript /usr/sbin/dpkg-preconfigure --apt > sid:~# > > It is not finding some perl modules that it needs. The required packages are probably in your cache. I suggest trying to manually install perl packages till you get one to configure (dpkg -i someperlpack.deb). Then get the others. Once you have a complete perl installation, try apt again. Mike
Re: LaTeX editor
On Thu, Jul 26, 2001 at 11:57:35AM -0300, GARGIULO Eduardo INGDESI wrote: > Hi all. > > I was reading some messages from "TeX fonts" and "BibTex style" threads > on this list, and I think perhaps someone can help me to configure an > editor to write Tex docs. Currently I'm using raw vim, without any > tex feature added! > > How can I configure vim (or other editor) to easy write TeX docs? > I am very happy with emacs+auctex+bibtex+reftex for this. Here is the code that I use in my .emacs file to turn on the bells and whistles: ;; Auctex mode (require 'tex-site) (setq TeX-auto-save t) (setq TeX-parse-self t) (setq-default TeX-master nil) (add-hook 'LaTeX-mode-hook 'turn-on-auto-fill) ;if you want this ;; Reftex mode (add-hook 'LaTeX-mode-hook 'turn-on-reftex) (setq reftex-plug-into-AUCTeX t) (setq reftex-extra-bindings t) If you are using a newer version of xemacs, you may need to put this in ~/.xemacs/init.el instead. Mike
Re: Soundblaster 16
On Thu, Jul 19, 2001 at 09:22:17AM -0400, Adam Bell wrote: > I have the same card. > > Probably your problem is that it's in plug and play mode, and since it's > an ISA card that is suckland for Linux. > > You can avoid isapnp difficulties all together by installing a 2.4 kernel. (i.e. install woody, or go to potato+2.4kernel via the Bunk packages.) The 2.4 series has smarts that can auto-sense IRQ's and addresses and stuff. It doesnt work with every concievable isa card, but sb16 is one thing that it is well tested with. Mike
Re: How to set up a "prefect" router
On Tue, Jul 17, 2001 at 06:33:49PM +0200, Raffaele Sandrini wrote: > Hi > > I have an old 486 DX33Mhz PC. I'd like to set it up as a router. It has a > Ethernetcard, 250MB HDD and no CDROM. > What is importatnt to look at? Should i use a 1 floppy Linux? If yes wichone? > Should i use debian? Is it important to use kern2.4? Other usful tipps? > > I assume you want to set it up and have it just work, for an indefinite period. I also assume that if it breaks, *you* have to fix it :). In that case, consider not using the harddrive. Along with the power supply and fans, it is the least reliable component in the system, and well past its prime. I think some kind of 1 floppy linux, as you mentioned, might work fine for you. Other possibilities are floppy based tftp (aka diskless) boot and (if you have access to a burner) eprom based tftp boot. Mike
Re: How to set up a "prefect" router
On Tue, Jul 17, 2001 at 12:57:44PM -0700, Mike Pfleger wrote: > > Sorry to intrude, but this last bit was too good to pass up! I have access > to DATA I/O burner that will happily do EPROMs like those Award's BIOS came > on in those older MBs. Where are docs on how to muck with this to make an > intelligent ROM boot (like Sun hdwr does)? > http://sunsite.utk.edu/ftp/pub/linux/LDP/HOWTO/Diskless-HOWTO-8.html is a good starting point. (You may want to pick an LDP mirror closer to you, utk=Knoxville, TN). I think there may be a helpful doc or 2 in the kernel source tree also. Have fun Mike
Re: Swap fscked in 2.4.5?
On Thu, Jul 12, 2001 at 05:48:06PM -0400, Paul Wright wrote: > The fix was to have twice as much swap as RAM (or more). This may require > one to have multiple swap partitions, if you have more than 64M of RAM. > The 128 MB limit on swap partitions went away in the 2.2 series. I think it is 2 gig now. Mike
Re: MUAs that compare with Outlook (your chance to show how much better Linux is than MS!!)
On Thu, Jul 12, 2001 at 09:28:41AM -0700, Kurt Lieber wrote:
> OK, I've read with great amusement all the chest-thumping going on about
> MUAs, MTAs and how Microsoft email products are things that you scrape
> off the bottom of your shoe.
>
> I, for one, am brand-spanking new at Linux and have yet to find a Linux
> MUA that meets my needs. I really do like Linux and would like to
> transition over to it for my desktop machine, but because of it's
> weakness on the MUA side, I haven't been able to do so. ("weakness" is
> my perception - you can prove me wrong by continuing to read)
>
> So, here's a list of my requirements and I'm hoping you guys can point
> me to an MUA that meets them. If so, I'll gladly switch over to Linux
I think evolution, once it is stable and feature complete, will satisfy
your needs nicely. There are official debs for a beta 0.10 release in
sid/woody, and prolly Ximian debs available for potato. I use neither
evolution nor outlook, but it appears to me that evolution was designed
specifically as an outlook alternative/replacement.
Be aware that evolution is a gnome app, so there are *lots* of preqs.
In other words, you need a fairly full gnome installation to run
evolution.
Mike
Re: Via AC 97 Audio chipset
On Wed, Jul 11, 2001 at 02:01:39PM -0500, [EMAIL PROTECTED] wrote: > I have an AMD k7 800mhz computer. The mother board has on board sound. it's > a VIA AC 97 audio Chipset. Anyone have any luck getting this to work. On > the install I loaded the AWE32 module and it accepted it. now I can play > cd's, but I have to turn all volume levels to full, but the sound is very > low, and I get a lot of static. > Wow! I never would have guessed that that would work at all. I think AC 97 is supported in kernel 2.4.x. I suggest you upgrade to woody, or use the Bunk packages to get to potato+2.4. Mike
Re: Swap fscked in 2.4.5?
On Wed, Jul 11, 2001 at 11:57:33AM -0700, Geoffrey Romer wrote: > My swap partition is behaving very rudely. Specifically, swap space never > seems > to get freed- swap size only increases, never decreases. Once it maxes out, > the system becomes sluggish, and there's nothing to do but reboot (the notion > of rebooting a system to fix a problem is giving me Windows flashbacks...) > > I'm told that this is a bug in the 2.4 series- does anyone know if upgrading > to 2.4.6 will help this problem at all? If not, are there any other > workarounds? > > One work around is to make sure that you have a swap partition at least twice as big as physical memory. This has been just a rule of thumb prior to 2.4, but seems to be pretty important now. Mike
Re: Network throughput too low?
On Wed, Jul 11, 2001 at 07:04:16PM +0200, Mart van de Wege wrote: > Folks, > > A minor question, possibly OT: for some reason I'm getting a very low > network throughput between my laptop and my desktop machine. I just ran a > test with netpipe-tcp, and the maximum speed is some 15Mbps. As I have a > 3Com 905C in the desktop box and a 3Com 574 PCMCIA card on the laptop > (both are 10/100 NICs) I would have expected a bit more speed here. Also, > my laptop complains about dropped interrupts during the test, and a quick > look with ifconfig shows that it is having buffer overruns. Is this > related? Anyone got a clue as to what is going on and how I can boost my > network performance? > I suspect a driver issue. Some NIC's are better supported than others under Linux. Perhaps the biggest reason for this is the varying degree of support that Linux recieves different the chipset manufacturers. People who need/want/demand near theoretical peak performance from fast ethernet under linux are usually rather picky about the NIC's they use. Back in ancient times, when dinosaurs roamed, and there was a company called DEC who made a processor called Alpha, the "Tulip" was the chip of choice. I really have no idea what the performance champion is wrt stuff you can actually buy today, but im told that inexpensive Realtek based cards are perhaps the best bang for the buck. Suggested possible approaches: Upgrade your kernel, hence your nic drivers if you are not using a fairly current kernel, and keep an eye on release notes of new kernels as they are released. Or, change your hardware. Or, just deal with it. Other possabilities: cable issues, router/switch issue Good luck Mike
Re: reiserfs
On Wed, Jul 11, 2001 at 12:14:48PM -0400, Peter Kok wrote: > Hi all > > I am new in debian. What is reiserfs > > It is a high performance file system, still somewhat in the experimental stage. However, it has already (as I understand) proven useful to some advanced users with very particular requirements. It has a number of advanced features/benifits, but perhaps the most mentioned is journaling. There are several other advanced/journaling file systems for linux under development. These would include XFS, JFS, ext3, and maybe others that I have not heard of. At present, the most widely used file system for linux is ext2, aka e2fs. That is probably what your are/should be using. Mike
Re: Cannot connect to a website from debian linux box
On Mon, Jul 09, 2001 at 11:40:12PM -0400, Shaji N V wrote: > Hi, > > I am facing a problem in connecting to a website from my linux box. I use > dialup connection through my ISP or log on to my work(via dialup). The > strange thing is that all machines at work are able to access the website > (sun or windows), by my linux box cannot. If I boot the machine in windows, > then I can connect to the website either thro work or ISP. The website is > running https, so I try the following on the sun box: > Perhaps this has something to do with the ecn. It seems that debian 2.4.x precompiled kernels ship with ecn enabled. A few firewalls are brokenly filtering (and dropping) ecn enabled packets, however. You can disable ecn by: echo 0 > /proc/sys/net/ipv4/tcp_ecn Mike
Re: kernel-image
On Tue, Jul 10, 2001 at 06:53:32AM +0200, Jesper Holmberg wrote: > I recently installed Potato from discs, and then dist-upgraded to > Woody. Now, I was looking into the subject of kernels, and I find I > have no package named kernel-image, although of course there exists > one (2.2.17 it seems) in /boot. > I dont think this is a problem. > Now, what would I do if I wanted to upgrade the kernel to 2.2.X or > 2.4.X, and why don't I have a kernel-image package? > Fetch and install the desired kernel-image package using your favorite method (dselect, apt-get, wget + dpkg -i, whatever). The installaion of the new kernel is nearly fully automated, if things are in the expected places. For instance, if the package system finds a sym link called "vmlinuz" in the standard spot, it make vmlinuz point to your new kernel and makes vmlinuz.old point to your old kernel. If you are going to 2.4, you may need to update your /etc/lilo.conf to point to an initrd image as well as the kernel image. (man initrd) The most important thing is to have an alternative way to boot your system if you screw up (boot floppy or rescue floppy, best to have both). Mike
Re: Acrobat Reader & Linux distributions
On Sun, Jul 08, 2001 at 02:18:46PM +0100, Ted Harding wrote: > Hi Folks, > > I'm having problems with Acrobat Reader (version 4.0) segfaulting > repeatedly on a SuSE-7.2. (This is with "Smooth text and images" > turned on -- turning this off seems to cure it). > > What are people's experiences with this version of Acrobat Reader on > other recent Linux distributions? > I have had trouble, stability wise, with pervious versions of Acrobat reader on debian. I think that may be in part because Adobe distributes binarys, hence the program is not always compiled against the exact libraries that it will be used with. My solution is to use xpdf. I have found xpdf to be fast, light, functional, and robust. If suse doesnt package it (that would suprise me), you can get the source and build it yourself. Mike
Re: creating a local mirror
On Sat, Jul 07, 2001 at 09:06:51PM +0200, Martin F. Krafft wrote: > we are using a local debian mirror for several reasons. we let rsync > run on one of the official servers once a week over the weekend, and > still have security.debian.org in the sources.list files on all > workstations. however, due to the nature of rsync, whenever packages > on the server are updated, they are first removed locally (i have to > use the --delete option since we only have a 40Gb drive for the > mirror), before everything is rsync'd. usually this works quite well, > but there are times when the connection is so bad that on monday > morning not all packages have been rsync'd so that Packages.gz says > they are there, but they aren't quite yet. > > apt-proxy is not an option, apt-move doesn't help either since it > doesn't ensure all packages to be there - and since we do a lot of > testing with random packages, we need to have access to all packages > at all times. > > any ideas on how to improve this? > Perhaps http would be a more dependable protocol. I have found, as I am sure others have, that http is better at cutting through net congestion than ftp, for the purpose of fetching debs from busy servers. I dont have experience to back it up, but I would speculate that http would work better than rsync too. You could use wget. I think there are switches that you could use to avoid fetching package files that you already have. It should be fairly easy to come up with a script to remove old packages. Mike script to remove old packages.
Re: Security manager (mozilla M18), trying to run as normal user
On Sat, Jul 07, 2001 at 11:17:36AM -0400, Carl Fink wrote: > And is there any way to get Mozilla to stop offering me an unwanted > "Netscape Search" pane when I type a URL into the address bar? Even with > "smart" URLs turned off, it still slows typing down so much it's very > irritating. > Yes. Use Galeon. As I am sure that most people know, Galeon is a front end to Mozilla. It seems to me that the Galeon user interface was designed with the needs and wants of the users at the forefront, as opposed to some other browser user interfaces that put the desires of the content creators (in particular, the ones that want to sell you something) at the top. My favorite galeon feature is the one that allows you to select when anamations are displayed. I select "never". I have been using the debanized galeon and underlying mozilla from sid as my browser at home for a couple of weeks now. If this build isnt in woody by now, I expect it will be soon. Stability-wise, I put it on par with Netscape 4, and well above any previous Mozilla release that I tried. Mike
Re: Swapping /usr and /
On Fri, Jul 06, 2001 at 09:09:35PM -0400, Andrew Overholt wrote: > Hi, > > After install potato on my laptop, I realize that I would like to swap the > partitions for /usr and / .. any idea how I can do this safely? I managed > to swap /home and /usr with little hassle but I'm kinda more concerned > about / and /usr. > You probably need to boot with a completly independent rescue disk(s). That way, you are just copying files, not doing surgery on a running system. Also, make sure you understand how to run lilo via chroot before you try this. Mike
Re: OutPut of Pon?
On Fri, Jul 06, 2001 at 09:00:52PM +0100, Keri wrote: > > /etc/ppp/peers/provider: unreconized option ' /dev/modem' > Did you use pppconfig to generate the /etc/ppp/peers/provider file? Have you hand edited this file? Most people, even experienced users who know enough to put this file together by hand with a text editor, use pppconfig because its just easier and quicker. Mike
Re: q ad migrating IDE -> SCSI, partition layout
On Fri, Jul 06, 2001 at 09:46:52PM +0200, Robert Waldner wrote: > 0: And, regarding the IDE-disk: Is this possible _without_ losing the > data on the current /mp3 (hda8)? Eg, deleting hda1-7, re-creating a > large hda1, ´cause I don´t have any means to backup 15 GB and I surely > don´t want to re-convert my whole cd-collection to mp3... > I am not sure, because I have never used the program, but it looks like "parted" might be able to non-destructively grow the partition for you. I know that parted is in sid, probably in woody, and probably not in potato. Mike
Re: how stable is the testing branch?
On Fri, Jul 06, 2001 at 04:51:25PM +, joe golden wrote: > Debian has koffice in the testing branch>> > this may be the answer to my need for a stable word processor and graphing > spreadsheet for our small school network of 10 machines. > > Will the testing branch be so unstable as to crash regularly? I don't want > to look bad and I don't want linux to look bad. > > Unstable is "unstable" because it changes often, not because it crashes often. As a whole, woody runs very reliably. It is very unlikely that a normal user could do something accedentally that would bring a woody system to its knees, requiring a reboot. The contents of woody changes daily, as developers squash bugs and upgrade versions, etc. Occasionaly, new bugs are introduced. Even more occasionally the new bugs are worse than the old bugs. Hence, there is more risk in tracking woody than in tracking potato. However, the risk is small, and very acceptable (to most people) if you find that you need features available in woody, but not in potato. Just because woody changes on ftp.debian.org or your friendly neighborhood mirror doesnt mean you have to install the new packages. You may wish to get your machines configured, and upgrade packages only during semester breaks. I dont know how old your machines are, but if they are "pretty old", you may wish to ensure that they have enough memory to run kde without swapping too often. I am running a fairly full gnome setup with 64 meg. It is enough, 128 would be much better. It would cost me about $10-20 to upgrade to 128, but I am not going to bother as this workstation is about to be demoted to mailserver/firewall. Mike
Re: 2.4.5-1 kernel in woody trouble
On Fri, Jul 06, 2001 at 09:04:23AM -0600, [EMAIL PROTECTED] wrote: > On Fri, Jul 06, 2001 at 08:45:06AM -0500, Jorge Santos wrote: > > > > Hi, > > > > I've installed kernel-image 2.4.5-1 on woody but I get the following > > errors when booting: > > > > request_module[block-major-3]: Root fs not mounted > > VFS: Cannot open root device "303"or 03:03 > > Please append a correct "root=" boot option > > Kernel Panic: VFS: Unable to mount root fs on 03:03 > > > > It looks like you may have compiled support for your hard disk as a > module instead of compiling it into the kernel. I did this once... > > Actually, you can get away with things like building ext2 support as a module when your root filesystem is ext2 now. This is exactly what is done in kernel-image package 2.4.5-1. Here is the output of lsmod, under 2.4.5-1 on my system (root fs is ext2 on a scsi drive) [EMAIL PROTECTED]:~$ /sbin/lsmod Module Size Used by parport_pc 19344 1 (autoclean) lp 5104 0 (autoclean) parport24672 1 (autoclean) [parport_pc lp] ppp_deflate39232 1 (autoclean) bsd_comp4160 0 (autoclean) ppp_async 6352 1 (autoclean) ppp_generic17296 3 (autoclean) [ppp_deflate bsd_comp ppp_async] rtc 5344 0 (autoclean) soundcore 4112 0 (autoclean) unix 15168 57 (autoclean) sd_mod 10736 3 (autoclean) aic7xxx 101152 3 (autoclean) scsi_mod 88256 2 (autoclean) [sd_mod aic7xxx] ext2 35952 2 (autoclean) As you can see, everything that can be a module is a module. Here is how it works: A temporary file system, containing kernel modules, a rudementry shell (ash) and maybe some other stuff is loaded into a ramdisk during the boot process. The kernel boots under lilo and uses the ramdisk fs as its root fs temporarily. It runs a temporary init from the ramdisk fs. The modules required to access the real root fs are loaded. Once the kernel is configured with the modules it needs, the temporary init chroots to the real root fs and the real init is run. You need to make some changes to lilo.conf to make all this happen. Here is my lilo.conf. The most important difference between it and what you may be used to is the initrd line. initrd contains the data that is loaded into the ramdisk fs. The docs are unfortunately a little sketchy/confusing on the correct used of initrd with lilo. boot=/dev/fd0 root=/dev/sda2 compact install=/boot/boot.b map=/boot/map vga=normal image=/vmlinuz initrd=/boot/initrd label=Linux Mike
Re: Digital camera and Linux
On Fri, Jul 06, 2001 at 06:11:53AM -0700, Michael Epting wrote: > On Fri, Jul 06, 2001 at 02:06:06PM +0400, Ilya Martynov wrote: > > Or any advice on another relatively cheap and good digital camera > > which can be used with Linux? > > Check out www.gphoto.org. There is a compatibility list there. gphoto > is packaged for Debian but supports very few current models. gphoto2 is > under development and supports many more cameras. I bought a Canon > Powershot A20 which is not specifically supported, but a simple patch to > gphoto2 did the trick and I can now move pictures to and from my camera > from my Debian machines. This is especially useful on my laptop so that > I now have capacious storage for pictures when I'm on the road. > -- Yes, gphoto is cool. I bought a Camedia 360L not too long after they came out. It wasnt specifically supported by gphoto, but by selecting a similar sounding model from the same manufacturer, I was able to get acceptable results. However, by attempting to use certian features (that I really didnt need anyway) I was able to get gphoto to crash repeatibly. Obvious solution: "Dont do that!" Anyway, I am happy with gphoto and am glad to hear that it continues to be developed. Addressing other comments in this thread, I too considered getting a floppy based camera, since floppys are obviously portable to Linux. I rejected that idea because it would be a slow, cumbersome way to handle that much data. I looked for a camera that communicated via a serial interface. Based on my experience with this, I would say that serial is marginally fast enought for my 1.3 Mpixel camera, but not fast enough for 3.1 Mpixels. I stayed away from usb at the time I bought my camera because I did not consider usb support in Linux to be stable/mature enough. If I were buying a camera today, however, I would look seriously at trying to put together a gphoto + usb solution under kernel 2.4.x. Firewire, if it could be made to work for this, would be even cooler. My 2 cents Mike >
Re: Problem with SCSI drive during installation
On Tue, Jul 03, 2001 at 07:26:52PM +0200, bgpaolo wrote: > > Debian is very OK distribution...but.I have a problem with my SCSI > magneto optical drive (by Fujitsu) during Debian GNU/Linux 2.2 (potato) > installation procedure. > My SCSI card is recognized immediatly (Adaptec 2920 - TMC 3660 Future Domain) > but as soon as installation finds MO drive, my computer hangs up. > If I disconnect MO drive and I connect another SCSI drive, for example YAMAHA > CDRW SCSI drive, installation proceeds regularly. > Now, since I have NOT problem with MO SCSI drive during Mandrake and RedHat > installation, why does it happen with Debian? (I love Debian and I wish use it > with advantage). > I would guess that this is a driver issue, hence a kernel issue. If you still have RedHat or Mandrake loaded, take a look at the boot messages (or review them with dmesg). Note the kernel version and scsi driver version. Compare these with versions on the Debian installation. If this leads you to believe that kernel 2.4.x might help, the Debian homepage has a link, I believe, to instructions on running potato with 2.4.x. However, I think it is cleaner just to upgrade to woody at this point. Hope this rather wild guess is helpful. Mike
pam broken in sid
As others have stated, the -26 version of pam in sid is broken, so if you install it, you wont be able to login at all. Unfortunately, I *have* installed it and don't have a session open. Question: What is the best way (using a rescue disk) to fix/circumvent/whatever pam well enough to be able to login once and repair the situation with the new debs? Thanks, Mike
Re: Most painless re-installation method
On Fri, May 25, 2001 at 10:48:10PM +1000, Kieren Diment wrote: > I have had some rather bad disk corrpution (laptop battery failed on > me I think). The easiest way for me to > fix it as far as I can see is to re-install as there are a few key > applications that don't work.. I have a root partition > and a /home partition. Only the /root partion is affected. I can > back up /etc which seems for the most part unaffected by the > corruption. > > Can anyone suggest the most painless way to do this. I can back up > /etc onto /home. > I am reminded of a line from "Aliens" to the effect of "I say we bug out and nuke the whole complex from orbit. It's the only way to be sure." Definitely back up /etc. You will mostly use it as an occasional reference, with a few selected files (like probably your X config file) you may copy outright. Also, try to get a copy of your current package selections with: dpkg --get-selections > myselections If you do this as root, make sure you move "myselections" to the home partition, as root's home directory is in the root partition. Later, when you have at least a minimal system freshly installed, you can dpkg --set-selections < myselections Download potato rescue.bin, root.bin, base2.2.tgz, drivers.tgz to your home partition. Make rescue.bin and root.bin floppies with the dd command per the installation instructions. Boot with rescue.bin and follow the bouncing ball. If you want woody, I think you have to at least start with potato since, as I understand, woody boot disks are not ready for prime time yet. Good Luck Mike
Re: /tmp trouble
On Sat, May 19, 2001 at 01:33:56AM +0200, Carel Fellinger wrote: > On Fri, May 18, 2001 at 10:34:51PM +0600, V.Suresh wrote: > > I just wanted to experiment with partitions. So, created a 30 MB > > partition, and modified /etc/fstab as > >/dev/hda7 /tmp ext2defaults 0 0 > > > >And rebooted. Now, mutt, lynx et al couldn't create tmp files, and won't > >budge. What's wrong? I had to revert back, to put /tmp under / partition. > >What's wrong with my setting? Help. > > just a stupid idea, but did you create a filesystem on that new > partition like with: "mke2fs /dev/hda7" ? > This may be a permissions problem. /tmp should be 777 with the sticky bit set, to keep userx from deleting usery's files. chmod 1777 /tmp Mike
Re: Want to set up Dual Boot (sort of)
Call me a rebel, or whatever, but I prefer not to put lilo in the harddrive boot sector on a machine that will dual (duel :P) boot with MS. I prefer to let MS have its way with the boots sector, an boot Linux with a lilo floppy, backed up by a spare lilo floppy and a rescue disk. Here is why: If you screw up lilo, you can just about always recover with your spare lilo. -- faster, unattended boots. If you are switching os's, you can select your new os by inserting or ejecting a floppy, starting the shutdown, and taking a break. You dont have to hang around and wait for lilo to ask you what to boot. -- safety: I had a friend who was running dual boot, and his Norton Anti- virus noticed that the boot sector had changed, and asked if he wanted to "fix" it. He said yes, and Norton put his system into an unbootable state. There are many ways to set up dual boot. What I have presented here is probably a minority opinion on the "best" way, but it is simple, fast, relatively fool proof, and works-for-me (tm). Mike
Re: potato or woody?
On Thu, May 17, 2001 at 02:45:35PM +0200, Stefano wrote: > Hi all. > > I'm going to install debian gnu/linux on a new computer and I'm wondering > if woody is stable enough. What would you suggest me: potato or woody? > > The system is a PC workstation used for statistical analysis and > "office-like" activity (and a bit of C++ programming too). > If you have to ask, you probably want potato. potato is one of the most bug-free, secure-out-of-the-box os distributions on the planet. If your overriding objective is to spend more time getting work done and less time futzing with your computer, then potato is the way to go. On the other hand, you can play with the latest toys (xfree86 4.0.3, gcc 3.0 prerelease, lots more) and still get work done with woody. Woody changes daily, and has more bugs than potato. If you are going to install it, you should have some tolerance for instability, and have some ability to work around and through problems. As a minimum, you should be comfortable running dpkg from the command line (or enough *nix experience be able to get comfortable fast). my 2 cents Mike
Re: Is S3 Savage4 video supported?
On Mon, May 14, 2001 at 01:57:17PM -0700, Lars Jensen wrote: > Is the S3 Savage4 video card supported in Debian potato? Or is there any > way of making it work? > > Thanks for any help. > I got one of these to play. Use the SVGA server. According to the docs, the card is very picky about which mode lines it will run. It would not work at all with the old modeline that I used with the Matrox card (till it died :( ). Here is the line that I am running: # nameclock horizontal timing vertical timingflags #Modeline "1280x1024" 135 1280 1312 1456 1712 1024 1027 1030 1064 Modeline "1280x1024" 135 1280 1312 1416 1664 1024 1027 1030 1064 /usr/share/doc/xserver-svga/README.S3V.gz, where I stole the above line from, gives a few more that might work. Mike
Re: magicfilter problem!
On Tue, May 15, 2001 at 05:58:22AM +1000, Steve Kieu wrote: > Sorry if it is too silly to ask but I am totally new > to debian! :-) When I run magicfilter to configure > printer, it says > > kieu:/home/ray# magicfilter > magicfilter: No configuration file specified > kieu:/home/ray# > > How can I run it? pls help. > /usr/sbin/magicfilterconfig Just run it as root. This builds a /etc/printcap for you that automaticaly runs magicfilter when you print. I know what they say about never, but there is no situation that I can think of that you would want to invoke magicfilter directly from the command line. The lpr command runs it for you. Mike
Re: upgrade to testing
On Sat, May 12, 2001 at 01:08:03PM -0400, MaD dUCK wrote: > hey guys, > i would finally like to pull one of my systems up to testing, but i > still can't really afford losing it, so i am a little cautious. i > guess i would simply like to know if anyone of you had previous bad > experiences with the dist-upgrade. i am on a laptop, currently running > potato with a custom kernel 2.4.4 and pcmcia-cs build. should i just > do it? it isn't reversible, is it? > I went potato -> woody on my home system about a week ago, and got sidetracked for a bit by a couple of bugs, both of which are documented in the Debian bug system. I was able to work through both, and am very happy with woody on that machine. However, my production server is gonna stay potato for now :P. Early in the install, perl got into a state where it could not find Glob.pm. This hosed installation scripts on several packages, preventing them from configuring, which blocked many more packages. I was able to work through this one by figuring out which perl packages *would* install and installing them manually via dpkg -i . This went quickly, since the required packages were alrealy sitting in apt's cache. I had a hard time getting an xserver to configure (either version 3 SVGA or the version 4 server). The postinstall script failed. I saw in the bug data base where Brandon directed the person who reported this bug to use a specific commandline to run the postinstall script manually. I did this on my machine. The script still exited with an error, but something changed, and one more trip through configure resulted in a correctly configured server. I am using ximian gnome 1.4, and got into a dependency conflict with gnomeprint or something to do with gnumeric. I just ditched gnomeprint and gnumeric rather than spending time figuring out a fix. Hope this helps. Mike rather than spending time figuring out a fix.
Re: unable to mount 2nd FAT16 partition
It is also possible that you are supplying the wrong type, just as the error message indicates. While Win95 can be made to format and install itself on a fat16 partition, I think it will prefer a newer type partition, which I think linux recognizes as "vfat". Try mount -t vfat Mike On Sun, Mar 25, 2001 at 10:57:41AM -0500, Andrew Perrin wrote: > Forgive me if this is obvious, but is your first one mounted on /mnt? You > can of course have only one filesystem mounted at a given mountpoint. > > ap > > -- > Andrew J Perrin - Ph.D. Candidate, UC Berkeley, Dept. of Sociology > (Soon: Asst Professor of Sociology, U of North Carolina, Chapel Hill) > [EMAIL PROTECTED] - http://www.unc.edu/~aperrin > > On Sun, 25 Mar 2001, William Staniewicz wrote: > > > I have no problem mounting my initial FAT16 partition, > > but am unable to mount the 2nd one. The 2nd one was added > > recently following the installation of Windows95 and Debian > > Linux (in that order). > > > > I use this to mount: > > > > mount -t msdos /dev/hda3 /mnt > > > > This is what I get back: > > > > mount: wrong fs type, bad option, bad superblock on /dev/hda3, > >or too many mounted file systems > > > > Below is my partitioning: > > > >Device BootStart EndBlocks Id System > > /dev/hda1 * 169278176+ 6 FAT16 > > /dev/hda2 124 250512064 39 Unknown > > /dev/hda370 1232177286 FAT16 > > /dev/hda4 251 782 21450245 Extended > > /dev/hda5 251 266 64480+ 82 Linux swap > > /dev/hda6 267 782 2080480+ 83 Linux > > > > What am I doing wrong? > > > > -Bill > > The "unknown" partition is a Plan 9 one. Also, when in > > Windows I am able to access the second partition. > > > > > > -- > > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > > > > > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] >
Re: Netgear NIC card support
Bay used to use DEC tulip chips in these cards. Now they are being made with a chip by Lite-on. If you have the newest card, you will need the newest driver from: http://cesdis.gsfc.nasa.gov/linux/drivers/tulip.html This driver is known to work with the new chips. Mike On Tue, Dec 01, 1998 at 07:11:27PM -0500, rathon wrote: > Hi, > > I bought a Netgear Fast Ethernet PCI Adapter card - FX310tx by Bay Networks. > > Then I looked thru the EthernetHOW-TO and this card is not listed anywhere. > > I am using Debian2.0. Does this mean no driver, no support ??? > > Can anyone confirm for me pls ? > > I need to know if I should return this and get something else. The price > was good though. > > Regards, > Rathon > > > Get free e-mail and a permanent address at http://www.netaddress.com/?N=1 > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: 128 MB RAM
On Tue, Dec 01, 1998 at 07:48:15AM +0100, Daniel Elenius wrote: > Dana G Haugli writes: > >Hi! > > > >I have 128 MB of RAM on my computer, but Linux only recognizes up to 64 MB. > >I have tried adding "mem=128M" to my lilo config file as recommended in the > >HARDWARE HOWTO, but that doesn't seem to work. Any suggestions? > > There's a patch that fixes that. You can find it...somewhere.:-)..where > you usually find your patches (dont remember). Or, you can try the > newest development kernels, they fix it too. > The newest stable kernel (2.0.36) also recognizes >64 meg without help. Mike
Re: 3COM 3c509B port selection
The cards that I have used (some tulips and some ancient Western Digitals) autosense the media. Are you sure this one doesnt do that too? Here is what I would try if you cant get it to work and nobody who actually knows something about this NIC has a better idea: Get the newest 3com driver from www.beowulf.org. Look at the comments at the beginning of the source to see if you see anything helpful. If this one is newer than the module than the one you are using, consider compiling it and sticking it in. Sometimes new chipsets break old drivers, but sometimes there is a fix in the new driver. Mike On Fri, Nov 27, 1998 at 07:40:51PM -0500, Stan Brown wrote: > I am trying to install on a freinds machine. It has a board lables 3c50 > (b), which has all 3 conetctor types (AUI, 10 base 2, 10 base T) on it > unfortunately it *does Not* have an jumpers. the 3c509 module > recognizes it, and figures out the IRQ and memory location, butI can;t > figure out how to tell it which port to use. >
Re: Installtion of xserver for debian 2.0
I think the XBF server will run this card. I am told that you can get it at : ftp://ftp.redhat.com/pub/XBF/ Mike On Sat, Nov 14, 1998 at 10:52:14AM -0800, Rakesh Mohan wrote: > I have installed debian 2.0, but having problems installing X Window > system. > > I have i740 vedio card, and it is VGA compatible. I think server for > i740 chipset is still not avaible for Debian, so i selected Unsupported > VGA compatible card during configuration. After configuration I got the > message that server has been started successfully. But it fails during > startx, please check the attached file for startx output. > > Thanks, > -Rakesh > > startx output > > /usr/X11R6/lib/X11/xinit/xinitrc > -auth /root/.Xauthority > > > XFree86 Version 3.3.2.3 / X Window System > (protocol Version 11, revision 0, vendor release 6300) > Release Date: July 15 1998 > If the server is older than 6-12 months, or if your card is newer > than the above date, look for a newer version before reporting > problems. (see http://www.XFree86.Org/FAQ) > Operating System: Linux 2.0.35 i686 [ELF] > Configured drivers: > VGA16: server for 4-bit colour VGA (Patchlevel 0): > ET4000, ET4000W32, ET4000W32i, ET4000W32i_rev_b, ET4000W32i_rev_c, > ET4000W32p, ET4000W32p_rev_a, ET4000W32p_rev_b, ET4000W32p_rev_c, > ET4000W32p_rev_d, ET6000, ET6100, et3000, ncr77c22, ncr77c22e, > ati, > sis86c201, sis86c202, sis86c205, tvga8200lx, tvga8800cs, > tvga8900b, > tvga8900c, tvga8900cl, tvga8900d, tvga9000, tvga9000i, tvga9100b, > tvga9200cxr, tgui9400cxi, tgui9420, tgui9420dgi, tgui9430dgi, > tgui9440agi, cyber9320, tgui9660, tgui9680, tgui9682, tgui9685, > cyber9382, cyber9385, cyber9388, cyber9397, cyber9520, 3dimage975, > 3dimage985, oti067, oti077, oti087, oti037c, cl6410, cl6412, > cl6420, > cl6440, generic > MONO: server for interlaced and banked monochrome graphics adaptors > (Patchlevel 0): > hgc1280, sigmalview, apollo9, hercules > (using VT number 7) > > XF86Config: /etc/X11/XF86Config > (**) stands for supplied, (--) stands for probed/default values > (**) XKB: rules: "xfree86" > (**) XKB: model: "pc101" > (**) XKB: layout: "us" > (**) Mouse: type: PS/2, device: /dev/mouse, resolution: 100 > (**) Mouse: buttons: 3, 3 button emulation (timeout: 50ms) > (**) VGA16: Graphics device ID: "Primary Card" > (**) VGA16: Monitor ID: "Primary Monitor" > (--) VGA16: Mode "640x480" needs hsync freq of 31.57 kHz. Deleted. > (--) VGA16: Mode "320x240" needs hsync freq of 63.15 kHz. Deleted. > (**) MONO: Graphics device ID: "Primary Card" > (**) MONO: Monitor ID: "Primary Monitor" > Warning: The directory "/usr/X11R6/lib/X11/fonts/100dpi" does not exist. > Entry deleted from font path. > Warning: The directory "/usr/X11R6/lib/X11/fonts/Type1" does not exist. > Entry deleted from font path. > Warning: The directory "/usr/X11R6/lib/X11/fonts/Speedo" does not exist. > Entry deleted from font path. > Warning: The directory "/usr/X11R6/lib/X11/fonts/100dpi" does not exist. > Entry deleted from font path. > (**) FontPath set to > "/usr/X11R6/lib/X11/fonts/misc:unscaled,/usr/X11R6/lib/X11/fonts/75dpi:unscaled,/usr/X11R6/lib/X11/fonts/misc,/usr/X11R6/lib/X11/fonts/75dpi" > (--) VGA16: PCI: Unknown vendor (0x8086) Unknown chipset (0x7800) rev > 33, Memory @ 0xe300, 0xe200 > (**) VGA16: chipset: generic > (--) VGA16: videoram: 256k (using 256k) > (**) VGA16: clocks: 25.26 28.32 25.26 25.26 > (--) VGA16: Maximum allowed dot-clock: 90.000 MHz > (--) VGA16: There is no mode definition named "640x480" > (--) VGA16: Removing mode "640x480" from list of valid modes. > (--) VGA16: There is no mode definition named "320x240" > > Fatal server error: > No valid modes found. > > > When reporting a problem related to a server crash, please send > the full server output, not just the last messages > > _X11TransSocketUNIXConnect: Can't connect: errno = 111 > giving up. > > xinit: Connection refused (errno 111): unable to connect to X server > > xinit: No such process (errno 3): Server error. > > > > __ > Get Your Private, Free Email at http://www.hotmail.com > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Adaptec SCSI Controller?
I think a typical 7200 rpm drive is mechanically capable of about 7MB/s, so I doubt if you will see any performance advantage for U2W unless you will be hanging several drives on the controller. Support for the AHA-2940U2W is pretty new for Linux. It is not in the stock hamm kernels. If you want to use it now, you will need to know how to compile a custom kernel. Some drives come in UW and U2W flavors. You can use the U2W drive on an UW controller, but termination is trickier. (The drive can't be terminated, but the scsi chain *must* be termimated). At least, I found that to be true for Atlas III's. I, and others, have had driver related problems with 2940's under Linux. Since Adaptec recently started giving tech data to the driver writers, these are being fixed rapidly. If you want a no-fuss, works out of the box solution, get a Buslogic. The driver has been stable for a long time. Their best controller is comparible to the 2940UW in performance and price. I have not used these myself, but all of the reports that have reached my ears have been very positive. Just about any ultra or ultra-wide drive should work great for you. Mike On Sat, Nov 14, 1998 at 03:03:41PM -0500, SEGV wrote: > I'm currently deciding which Adaptec SCSI controller to get: > > AHA-2940UW (Ultra Wide, 40MB/s) > AHA-2940U2W (Ultra 2 80MB/s) > > Will Debian Linux 2.0 take full advantage of the U2W? Will I get the full > transfer rate (ie, as fast as 98/NT would allow)? > > PS: For harddrives I am looking at the Quantum Viking II drives. Good choice? > > -- > SEGVhttp://www.cgocable.net/~mlepage/ > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Debian and AMI MegaRAID
On another list, Alan Cox pointed out that the kernel 2.0.36pre series includes the latest Megaraid driver. To install this kernel to a Debian system, get a pristine 2.0.35 source and unpack it in the usual manner. Dont use the Debian source package, I think it is patched already. Go to Alan's site, get the 2.0.36pre patch de jour (I think they are up to 20 now) and apply it. Compile the kernel. I suggest using the Debian kernel package package. Look for the documentation under /usr/doc once you have the kernel package package installed. If you dont hurry, you will miss the fun and adventure of running an unreleased kernel. All humor aside, I have been running 2.0.36pre15 for several days now with no problem. 2.0.36 should be in full release any day now. If you need the Megaraid driver, I suggest giving the prepatch a shot. Mike On Thu, Nov 12, 1998 at 07:33:57PM +0100, Fabio Belletti wrote: > > We have a dual Pentium II server with AMI Mega RAID Controller but > we don't know if it works with Debian Linux 2.0 because the 'recently > released' Linux driver from AMI seems to work only with the latest Red Hat > version (5.2) or the previous Red Hat version (5.0/5.1) with the driver v > 0.92 . > > If it works could you please send us the information about the correct > installation procedure ? > > > Thanks! > > > Byte Byte > > > Fabio Belletti > Omega Generation > > > > [=] > [ FABIO BELLETTI ] > [ -- ] > [ Office: OMEGA GENERATION s.r.l. ] > [ Via Murri, 39 ] > [ I-40137 Bologna (Italy) ] > [ phone : +39 051 30.66.44] > [ fax : +39 051 39.03.10] > [ mobile: +39 347 22.61.753 ] > [ e-mail: [EMAIL PROTECTED] ] > [ web : http://www.omega.it ] > [=] > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Moving Files from Windows
If your NT box is networked to a machine with a CD writer, consider making a CD. There is support in Debian for .zip files (look for the zip package). I think some zip programs can split up files for multiple floppies. Mike On Wed, Nov 11, 1998 at 08:42:55AM -0500, Costa, Michael J. wrote: > I have just installed LINUX on another machine in my office. The powers that > be are reluctant at this moment to let me put in on the network. I have > loaded some items down to my NT machine and now need to ship them to my > LINUX box. However they are too big to fit on a floppy. Anyone got any ideas > of a utility I can use to compress on NT and uncompress on LINUX?
Re: Help, with Debian 2.0 install from CD-ROM not part of HDD card
It looks like your cdrom is one of the early ones that runs off a propitary interface and reqires a special driver. Later cdroms run off the ide controller and use a common driver, regardless of brand. Later drives are recognized automatically by Debian, yours will require just a little more work. As root, run modconf. Go to the cdrom modules and choose a module to install. Mike On Mon, Nov 09, 1998 at 03:55:03AM -0500, Craig M. Kimmer wrote: > Hello, > > > I am having problems with Debian 2.0 to install the the important, > extra, > and/ or packages, which include the kernal source and patches. I have got > a operating system, but it does not recongize the CDROM drive, thus I can > not add or upgrade any program packages to the system. I have tried to > move the CDROM drive to run off the HDD controll but the system will not > even do look at the BIOS to startup. I am including at the base of this > message the system specs. of this machine. > > CPU:Cyrix 5x86 100MHz > Hard Drives:BigFoot 1.2 Gb > WD 4.0 Gb > Floppy Drives: 3.5" > Bus Type: PCI > Extra Drives: TEAC CD-55 tray ROM 4x > Mouse and style:Bus on COM1 > modem: on COM2 > Memory: 24 Megs > Root Directory: hdc7 > O/S on system: Windows 95 > Kernal Version: 2.0.34 > Sound Card: Drives CDROM - Sound Blaster Pro 16 compatible > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Disk problems
On Tue, Nov 03, 1998 at 07:32:19AM -0500, Biciunas, Paul John wrote: > Hello, all. > > I installed Debian 2.0 (2.0.34) Greenbush distribution. > My disks are 2 IDE drives, a 540M and 2.5G slave. > > The partitions are (df output) > /dev/hda1 99029 ... / > /dev/hda3 348873 ... /home > /dev/hdb1 495714 ... /var > /dev/hdb2 1926659 ... /usr > > I had problems making a kernel, but finally managed to build a bzImage. > The kernel booted, and upon testing, realized that I needed to rebuild > the kernel. > That's when the fun started. The make failed, when it couldn't process > some .c files in /usr/src/linux/lib/ - "file" said they were MPEG files. > > Firing up emacs, it complained about not being able to find > /usr/local/share/emacs/... > files, and sure enough, /usr/local/share was no longer a directory, but > some .c file. > > Running fsck was a nightmare. > Instead, I rebooted from the installation cdrom, and repartitioned the > disks, checked > for bad blocks (passed), and started dselect. On Install, what I got was > multiple > > EXT_fs error (device 03:42): ext2_find_entry : bad entry in directory > #8193 : rec_len > is too small for name_len - offset 0, inode 538976288, rec_len=8224, > name_len=8224 > > Is my disk toast? That would be my guess. Unfortunately, it looks like the newer 2.5G drive is the one going south. To test this theory, unplug the data cable from the 2.5G and jumper the 540M as a solo master. Reinstall from scratch, including repartitioning. That way, you can be sure that you are not starting out with any corrupt files. I think, however, that IDE drives can fail in ways that induce errors in drives mounted on the same controller, so it *could* be that the 540M is failing. If the above does not clear up the problem, unplug the 540M, set up the 2.5G as solo master, and try again. If the problem still presists, it is prolly a controller problem, or something I missed. If you are not using the kernel-package for your kernel compiles, I suggest you check it out. It makes things much easier. The documentation will be in /usr/doc. Mike
Re: transferring large files from DOS/Windoze to Debian
If your home machine is dual boot, you could transfer to the MS side with laplink, reboot to Debian, and mount the FAT partition. If you resort to floppies, at least some zip programs in MS-land allow you to split big files into many, floppy-sized, files. Debian can mount floppies, and with the zip package, process zip files. Your mini-com, kermit idea can prolly be made to work, but I have no experience with this. Mike On Sun, Nov 01, 1998 at 10:58:33AM -, Chris Evans wrote: > Is there a way to transfer large files from DOS/Windoze machines > to Debian if you only have a null modem or floppies to "connect" > the two? > > I want to download some large files for Debian (staroffice and a > newer kernel source). My Debian machine is now at home and it > will cost a fortune to download them at 14.4k over the 'phone but I > can catch them with my NT machine at work for free and put them > on my Win95 portable (conversion to Debian for this is > planned!!!) with laplink. > > Is there a good way to transfer them from the portable to the > Debian machine? I have a feeling that kermit or another protocol at > each end of the null modem cable is the best answer. I can get > kermit or the like for the portable and I have minicom on the Debian > machine. > > Failing that, anything that will split files under DOS/doze in a way > that they can be reassembled after floppy transfer under Debian?
Re: Install to removable hdd
It is possible for a disk to be in a state where cfdisk (the program used on the installation disk) can't work on it, but fdisk can. fdisk is a bit harder to use, but I suggest that you give it a shot. -choose the option on the installation program that lets you exit to a console -type 'fdisk /dev/hdb' at the prompt (no quotes) -to keep from getting confused, use the p command as often as you need to -the program will ask you if you want to make extended or primary partitions. You prolly want to make primary. -Warning Will Robinson! If you just type 'fdisk', you will be working on /dev/hda, which you prolly dont want to do. Mike On Sat, Oct 31, 1998 at 06:55:10PM -0700, Neoklis Kyriazis wrote: > I have been trying to install (sneak in!) Linux from cdrom on to a PC at work > so I have to do this on a removable hdd. I originally failed to install from > the cdrom for some reason so I downloaded resc1440 and copied to floppy. This > worked and I went as far as disc partitioning but after specifying hdb > (removable) I got the error message FATAL ERROR: Bad primary partition.
Re: installation problem
I have seen circumstances where cfdisk (used by the install program) would fail, but fdisk would succeed. Try going to the console and typing 'fdisk'. Mike On Tue, Oct 27, 1998 at 12:19:52PM -0800, Artin Nebel Rebekale wrote: > while installing hamm to my thinkpad laptop i had this odd problem. > I've already checked the linux/laptop resource pages and they had no > solution for this. Basically i get to the screen to partition the > drive, /dev/hda is the only drive i have (and therefore the only one > available) but when i choose ok, it just takes me back to the menu. > i've tried all the different rescue disks and this same thing > happens. i switched to console 2 and was able to successfully mount > the drive and look through it, so the drive does work. anyone have > any ideas?
Re: Mail; how to do remote... (was Re: need help getting X server running...)
On Sun, Oct 25, 1998 at 05:11:28PM -0800, Jesse Evans wrote: > Now, on to the next item... How do I deal with mail? I've got a dial-up > connection through > an ISP, so I need to somehow configure linux to connect to my POP account. > Where do I look and what do I do? > You could install a mail client that can access a pop account on its own. Netscape mail is one program that I know of that can do this. Pine is another, but due to license problems, you can't get ready-to-go Pine binaries for Debian any more. If you want to use a mail client that cant do its own pop access, you can use fetchmail to get your mail, and then read it with anything you want. I use Mutt. I suspect that it is the most popular mail client among Debian users. I like it cuz it is fast, snazzy lookin, and can live in an xterm. But I suggest you read the dselect descriptions, install a few likely looking suspects, and do your own evaluation. > Finally, what do you all use for your web browsers? The only one that > shows up in Dselect is Netscape. > Are there any others available? I use Netscape 3.04. There *are* others available. I checked some of them out about a year ago. To me, they all seemed kinda lame compaired to Netscape. Mike
Re: your mail
My guess would be that Adaptec's latest hardware tweak broke the driver in the 2.0.34 kernel used in hamm. I had a less severe driver problem with my 2940UW, which was apparently fixed by using 2.0.36_pre15 kernel. Another way to get the latest AIC78xx drivers is to go to ftp://ftp.dialnet.net:/pub/linux/aic7xxx/ and , I suppose, apply a patch from there to a released kernel source. (I haven't done this.) If you have access to a running Debian system, and you can produce a kernel with updated scsi, you might try making your own boot floppy using the boot-floppie package from admin. (I havent done this either.) A few hints if you are inclined to try this: When you download the 2.0.36 pre patch from Alan Cox's site, the file will be gziped, but will not have a .gz extention (at least if you download through Netscape, as I did). I had to rename the file to have a .gz at the end and then gunzip it before the patch would work. Debian kernel source packages are already patched, so applying further patches may have unpredictable results. Use the pristine source if you want to patch. The Debian kernel-package package is pretty handy for automating kernel compiling. Hope this is helpful. Mike On Sun, Oct 25, 1998 at 05:32:53PM +0100, Marcus Geiger wrote: > Hi, > > I am new to debian linux (not new to linux in general), and still trying to > install debian 2.0. > > All kernels I tried to boot on my machine, hang after they found my CD-ROM > drive (FreeBSD boots clearly) > > My setup is: > ASUS SP98AGP-X > K6-233 > Adaptec 2940UW (rev 1.32) > 2 x SCSI hardisks > IDE-CDROM > 128MB > Diamon Fire GL1000PRO (AGP) > > Does anyone of you know such a problem or have any ideas. > > Pleas help. > > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Problems installing Debian
I have never seen this particular behavior before. However, strange problems during installation from floppys is often attributable to corupt floppys. The rawwrite or dd process usually used to make installation floppys from downloaded images does not tolerate bad media very well. I suggest remaking your rescue disk on a different floppy. But first, check the floppy for bad sectors by formatting it under dos or whatever, or any other method you find convienient. Even better, use a new floppy, if you have one handy, checking it for bad sectors first. Hope this helps Mike On Sun, Oct 11, 1998 at 02:59:21PM -0500, Alexander Bugeja wrote: > I am trying to install Debian from floppy (I have the rescue, driver > and 5 base floppies set up) on my AMD K6 machine. However > when I try to start the installation, with the rescue disk, the > machine just reboots. More specifically, I first get the > following messages > > Loading root.bin... > Loading Linux... > Uncompressing Linux > > followed by a string of messages too fast to read, and then > the machine reboots all over again, leaving me right where I started. >
Re: Wordperfect + Debian 2.0
Sure. You can run pretty much anything in hamm that ran in bo. You just need the proper packages from old-libs. Since I am running Netscape 3.04, I have several packages from old-libs installed, but I would guess that libc5 is all that WP needs. Mike On Tue, Sep 22, 1998 at 01:49:02PM +0200, Bostjan JERKO wrote: > Hello ! > > Has anybody tried to install Wordperfect on Debian 2.0. I know it is possible > to install it on 1.3 with libc5, but what about 2.0 (with libc6). > > Bostjan > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Your friends...... whatever.
On Sat, Sep 05, 1998 at 01:56:57PM -0400, Igor Grobman wrote: > > I am the debian anti-spam person, (much deleted) Debian-user is amazingly spam free. I could tell that someone was doing something, but I had no idea who or what. Thanks, Igor Mike
Re: AGP Video Cards?
SuSE recently released a server for the G200. This will probably make it into the next release of XFree86, but for now it is "beta". My previous experience with using SuSE servers with Debian (the then new nVidia server) was favorable. But unless you are pretty good at setting up X, I suggest you stick with with the official XFree86 servers, all of which, I think, are available in hamm. An excellent source of Linux oriented hardware data (including video cards) is Net Express's web site: http://www.tdl.com/~netex/ Mike On Mon, Aug 31, 1998 at 12:21:17PM -0400, [EMAIL PROTECTED] wrote: > I'm trying to get a good AGP video card to work with Debian 2.0. Looking at > the list of supported cards in the xf86config utility I see the "matrox > Milenium II (AGP)." However this card seems hard to find now since it's been > replaced by the new one G200. Does anybody know if this new card is supported? > > Any advice on any other AGP card? >
Re: reply to USR ?
I dont think you need to mess with setserial at all. Please be a little more specific in describing "all the steps". Mike On Sun, Aug 30, 1998 at 02:18:42PM -0400, spOOL wrote: > To: spOOL <[EMAIL PROTECTED]> > Subject: Re: 3 com(USR) modem > > > Now I can't figure out how to configure my modem so that I can connect to > > the internet and go to the debian ftp site. > > Did you run pppconfig? yesI went through all the steps to create the > connection. > When it tried to connect it couldn't find the modem. I checked and the > modem is on Com 2 so I used setserial to set it to /dev/ttyS1 which should > be Com 2right > > any ideas??? > > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null >
Re: 3 com(USR) modem
There is a set of jumpers on the front of this card to set com port and irq, and a diagram on the back of the card showing how to use the jumpers. com1 corresponds to /dev/ttyS0 in Linux, and this usually uses irq 4. com2 coresponds to /dev/ttyS1 in Linux, and this usually uses irq 3. Unless you are using a ps2 mouse, your mouse is probably set up to use one of these, so you will want to use the other for your modem. If you are running the machine dual boot with a MS operating system, you may want to boot to MS and find out what com port the modem is using there. The modem comes from the factory jumpered for plug and play. You can make it work that way under Linux, but you will need the isapnptools package. Things will be much easier under Linux if you just hard jumper it to a com port. You will need ppp, net_std, and net_base packages. I think ppp is now ppp_pam or something. You will need to set up your ppp options and your chat script. I think there is now a package that automates this task, but I cant remember what it is called. I hope this answers at least some of your questions. Mike On Sun, Aug 30, 1998 at 12:54:28PM -0400, spOOL wrote: > OkI admit itI am feeling like a total idiot!!! and I am > getting very frustrated with my systemThis list seems to be my only > source for helpso to everyone that has already answered questions for > me.thanks...and here's another one > > I tried to install Debian 1.3 but couldn't get X going with the right video > settings, so I downloaded Hamm. I have gotten help from this list on my > mouse, ethernet card, video card, monitor, and such. Now I can't figure > out how to configure my modem so that I can connect to the internet and go > to the debian ftp site. > > I have a US Robotics Sportster 56k Internal...not the win only model. I > can't find any info anywhere for this modem.. anyone have any ideas??? > > Once I can get this working and can at least be able to connect to the ftp > site and get the .deb packages that I keep forgetting to download or have > errors downloadingI should be able to get this thing > goingRIGHT > > HELP > > THANKS AGAIN EVERYONE > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null >
Re: ping: sendto: Operation not permitted
IP masquerading is compiled into the stock 2.0.34 hamm kernel and works just fine for me. Remember to install the ip masq ftp module so that ftp works right behind the firewall. I too installled the ipmasq package and ended up removing it. It would probably be a nice, useful package if there were any docs at all for it. You dont need this package to do ip masquerading. Im barely knowlegable enough to get ip masqing working on my own system, so instead of offering specific advice, I refer you to the How-To and the ipfwadm man page. Mike On Thu, Aug 27, 1998 at 07:31:50PM -0400, Bob Bernstein wrote: > For the record, removing the debian 'ipmasq' package - which had an empty > configuration and therefore was defaulting everything to deny - solved the > ping and dns problems, solved everything in fact except how to get ipmasqing > working in 2.0.34. > > I'm suspicious of that 'IPAUTOFW' option. Does anybody have a kernel config > for 2.0.34 that produces a working ip masq? > > > > > Bob Bernstein <[EMAIL PROTECTED]> wrote: > > > Has anyone ever seen the Subject: line's message before? > > > > I just compiled 2.0.34 to get ip masqing going, and I appear in the process > > thereof to have trashed something really basic. Ecch! I get this whether I > > ping a local host, a remote host, or the machine itself. > > > > The system's ifconfig and routing tables look ok. It will dial up my ISP and > > get an IP assigned which duly appears in both those tables. But it then acts > > as if there's no DNS available, which there is, as I can verify by > > connecting with another machine to the same ISP. > > > Bob Bernstein at Esmond, R.I. > > > > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null >
Re: boot probs with aic7xxx driver... urgent...
Check your cables. A similar problem on one of my systems was caused by a bent pin on an external scsi cable. Mike On Mon, Aug 24, 1998 at 12:47:08PM +0200, Alexander Koch wrote: > Hi folks. > > We've a shiny new server and the resc1440 image from the current > diskset is freezing on boot while "downloading sequencer code ... > 413 instructions loaded" ... > > It's an Adaptec AIC 7880... > > Please, if anyone has an idea, please tell me. > Please always cc me on replies... > > Thanks, > Alexander
Trouble with IP forwarding
I am trying to get ip forwarding working between a hamm box and a Win95 box. Both machines have permanent ip addresses recognized by the campus name server. I compiled a custom kernel with the Debinized 2.0.34 source, with ip forwarding enabled. I installed a second net card in the hamm box and configured eth1 as 192.168.1.1. I established a route to the W95 box through eth1. I can get traffic between eth1 and the W95 box, but I cant get the hamm box to route traffic from the W95 box to the outside world through eth0. I have tried configuring the W95 box's gateway as 192.168.1.1, and also the public (eth0) address of the hamm box. Obviously I am missing something here. Any deposits to my clue bag greatly appreciated. Mike
Re: Large Paritions... advice?
I just got through partitioning a 9 gig disk myself. I found that fdisk worked better than cfdisk for me. I rebooted after partitioning, as suggested by the fdisk prompts. I had no trouble making and formating a 7 gig partition. (I was unable to format large partitions using cfdisk and not rebooting) Since you are making 5 partitions, at least one will have to be a logical partition. You can only have 4 physical partitions, numbered 1-4. Logical partitions are numbered 5-8. (you can only have four of those too). I dont know much about logical partitions cuz I dont use them, but I think that they have to be created from extended partitions. cfdisk handles this for you transparently, but I think this is a manual operation in fdisk. If it was me, I would combine the /usr and root partitions and save myself some head scratching. Mike On Mon, Aug 17, 1998 at 02:38:10PM -0400, Nebu John Mathai wrote: > Hello, > I would like to install Debian 2.0 on a 3 Gig partition which I do not > want to split > further. > 1. Are the problems of possible filesystem corruption realistic? > 2. Are there any performance losses with ext2fs on a 3 gig partition? > 3. For a single user workstation, hooked to the internet via cable > modem (and with > telnet and ftp servers enabled) would a single 3 gig partition be ok, > or would it be wiser > to split it up? > 4. Hardware-wise, are larger partitions worse on the drive than > multiple small ones? Or > is the drive indifferent? > > I have a single 8.4 Gig drive and I planned on having: > 100 MB Linux root > 2.9 Gig Linux /usr > 100 MB Linux swap > 3.0 Gig NT (unfortunately) > 2 Gig FAT32 /home (common for Linux and NT) > > But CFDISK would not let me parition it like this so I had to incorporate > Linux into one > massive 3 Gig partition. > > I'd appreciate hearing from anyone who has had horrible experiences with > filesystem > corruption (hope not to hear from anyone, I guess). > > Thanks for any help, and if this has been discussed elsewhere please point > me to a FAQ > and I'll shut up. >
Re: ATI Rage 3D Pro AGP with X
On Sat, Aug 15, 1998 at 08:55:13PM -0500, Matthew Myers wrote: > Does anyone know how to successfully configure X to run with an ATI Rage 3D > Pro AGP? > > When I try to run X it seems to start okay, but the screen is divided into > thirds vertically, and in each section I can only see the far right third of > the screen. it scrolls, but show the same thing in each third instead of > covering the whole screen. > > My Setup: > All of the following comes from the Official CD with no updates. > Debian Linux 1.3.1 > XF86_Mach64 xserver > > I have tried the Mach64 server from the CD and even downloaded the last > available libc5 version with the same results. Please help me. > I have a card by that description running pretty successfully under hamm. This is a new system, it has never seen bo. Have you tried steping through all the configured resolutions with cntrl-alt-+ or cntrl-alt--? (use the + and - keys on the keypad, not the ones near the backspace key). By default, X comes up in the highest configured resolution that passes its sanity checks. Hopefully, the next resolution down looks better. If it does, comment out bad modeline in your configuration file and you are set. HTH Mike
Re: xterm problems
If you are using X, consider xon instead of telnet. You will need to fix up an .rhosts file at the other end so that you can log on without a password for this to work. Obviously logging on in this manner has security implications, and you should consider them before you implement this. Using ssh would be a better choice, but this probably isnt an option unless the remote machine has it or you have root priviliges to put it there. xhost + remotehost; xon remotehost causes the remote host to display an xterm on your screen, using its own terminal database, avoiding the problems you describe below. Mike On Wed, Aug 12, 1998 at 01:52:22PM +0200, Michael Sicher wrote: > hello, > > i like debian 2.0 very much but now i have problems with xterm: > > - i am not able to log into some systems using telnet (connection closed > by foreign host). > > - on some systems i cannot start vi when logged in via xterm/telnet (no > terminal database found - debian 1.3 system) > > - backspace does not work > > how it is possible to solve this problems or to use xterm instead of > xterm-debian? > > thanks a lot! > > bye, > michael > > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null >
Re: Error using FAT32 in install
On Mon, Aug 10, 1998 at 10:53:22AM -0500, Ed Cogburn wrote: > To anyone else, does the install kernel for Debian 2.0 have FAT32 > support, or does that only show up in kernel-source 2.0.33/34? > I recently tried to install hamm by downloading the driver disk and the base tarball to a FAT32 partition and just making the rescue disk. No dice. The rescue disk could not mount the FAT32 partition. I downloaded these images just a few days before full release of hamm. The installed system, however, is able to mount FAT32. Mike
Re: Several questions before I run Debian
On Wed, Aug 05, 1998 at 11:53:34PM -0400, Matthew A. Reklau wrote: > I do not subscribe to this list and do not currently run Debian. Before I > run Debian I want to know: > > What kernel version does the current Debian 2.0 come with? > 2.0.34 > What support is included for Fat 32? > kernel 2.0.34 supports FAT32. Just mount it, works out of the box. I posed this same question to this list myself recently. Thanks to Bob Nielsen for providing me this info. > Do the X servers include support for the Diamond Viper V330? > Dunno, but I think so. Debian's X software is Xfree86 3.3.2, so if Xfree86 supports it, Debian supports it. If you dont get a better answer then that, you might check the Xfree86 web site. > If I can get acceptable answers to these questions I will begin to run > Debian ASAP. > > Thanks for your time and help. > > Respond to > > [EMAIL PROTECTED] > > since I am not on the list. > > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
How do I mount a FAT32 partition?
How do I mount a FAT32 partition on a hamm system? FAT16 has worked out of the box for years, and I know there is a way to mount NTFS partitions read only, but I havent seen anything on FAT32. Any pointers appreciated. Mike -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Hamm problem. (XDM, maybe)
Boot with your rescue disk. Mount the partition that contains your /etc directory. Edit /etc/X11/config, commenting out the line "start-xdm". Reboot normally. Mike On Thu, Jul 30, 1998 at 04:44:39PM -0700, James Brown Ender/Gcc" wrote: > Hi, >I've just upgraded my system to Hamm, but after a reboot the screen > flashs from text to graphics > mode, and back. I think it might be XDM, as X Windows doesnt load... > > Any suggestions on how I can stop this? I dont want to format my H/D, > and I cant shutdown XDM > because the keyboard doesnt work while switching. And as the screen > switchs about 5 times a second... > > -- Ender -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: AGP video card
If Permedia cards are not yet supported by XFree86, check the SuSE website. If you cant find it, let me know and I will dig it up for you. Mike On Tue, Jul 28, 1998 at 09:32:00PM -0600, [EMAIL PROTECTED] wrote: >[snip] > Part 2. > > I have a Fire GL 1000 , or also called Diamond Permidia 2 8MB AGP 2d/3d > combo video card. I didn't see this in the list of video cards to select. > Could someone please help me in setting up the correct server options for > this. > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Help: Upgrade to Hamm failed !?
autoup.sh and dselect are pretty smart, but they are not brilliant. You do need to clean up after dselect occasionally by running dpkg manually. I suggest you start to clean up your system by getting the instructions for manually upgrading to hamm (from the same place you got autoup.sh prolly) and checking to see that all the listed packages were successfully installed. If not, you will need to fix these problems first. Try to identify the most critical broken packages. Try to find out why they did not install cleanly. Sometime, you can fix a dependency problem by fetching a package manually and installing it manully with dpkg -i foobar.deb. Once you fix one problem, see what else will conifure with dpkg --configure. There is such a thing as a ppp package in hamm, but I think thats been downgraded from "standard" to "optional" now. ppp-pam is the "standard" package. Mike On Mon, Jul 27, 1998 at 10:29:42AM +0200, Jan Speich wrote: > Once upon a time I was running a nice stable system: Debian 1.3 with > Kernel 2.0.35 on a Pentium2 > But then I needed PPP 2.3.5 and the "disaster" was born: > > First I got the stupid I idea just to take the PPP 2.3.5 tar.gz follow > the instructions in the included > readme file and recompile the kernel. But the compiling errors and a > short view on the debian > page teached me that I would need libc6 for PPP 2.3.5. > > So, I decided to get it all and to upgrade to Debian 2.0 (Hamm), which > ,the package page promised, > would already include PPP 2.3.5. > > Upgrading seamed to be quiet easy: Having a ftp connection to a mirror > site, knowing thhe path to > "hamm", running autoup.sh, moving wtmp file , truncate utmp file and > rebooting. (At least that was > what I kept in mind after reading the "autoupREADME".) > > Everything went fine with these steps but I was suprised when I saw > while rebooting that > I was still running PPP 2.2.0. Well, I wasn't really suprised, as I > always thought PPP has > to be compiled into the kernel and I didn't compile the kernel after > upgrading. > > So, here is my first question: > > Question 1 > How do I install a package like a new PPP version or PPTP, as > dselect does not offer > instlalling these things (or am I blind). A step by step guide would > > be great. > > > O.K. the "disaster" goes on: I started "dselect" (don't remember the > actual reason but I found "splay" that I > wanted to have) when pressing RETURN to tell dselect that I have decided > > what to take it showed me some > conflicts concerning libc5, libc5.dev? (sorry, don't remember the > details). I just pressed RETURN > (dselect help said, that by pressing RETURN I would accept dselects > proposal to solve the problem an > eberything would be fine) but I only could switch between the help page > and the page showing the conflict. > So I decide to stop this mess by pressing X. > > As this problem didn't disapaer (How could it, I didn't change anything) > > I decide to use dselects: > "Remove unwanted packages" to have a clean system. > But, Ooops, dselect removed more than I expected (sorry, don't remember > the details). > But one thing that I'm really missing is "ftp" especially as I use ftp > as dselects access method. > So here are some more questions: > > > Question2 > How do I get back "ftp" (I'm still running Netscape with it's ftp) > > Question3 > Can I run autoup.sh just one more time or is there the risk to get > the system > into an unstable state. > > Question4 > How do I know if I really managed to upgrade to hamm. (Any command > to the version) > > Question5 > Does anyone know what went wrong after, while upgrading to hamm. > What do I have to > do to avoid this dselect conflict problem? > > > So, after all it isn't a real disaster, as the system is still up an > running but I still need PPP2.3.5... > > Thanks in advance > > Jan > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: ppp change?
I suggest you put "debug" into your /etc/ppp/peers/provider file so that the details of the negotiations get logged. Mike On Sun, Jul 26, 1998 at 09:28:06AM -0700, Richard Sevenich wrote: > I just installed kernel 2.0.35 and have noticed a change in ppp behavior, > if my memory serves me right. > > I normally login as a user and start ppp by going su momentarily to issue > command 'pon' and then exit back to user. This still works as usual. > > However, if I log in as root and issue 'pon' the modem still dials, but > the negotiations for the ppp link perhaps fail. I never get net access. > > Is this an administrative change or has something gone astray in my > recompile? > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Bo crashes under heavy disk load
On Fri, Jul 24, 1998 at 03:28:48PM -0400, Peter S Galbraith wrote: > > Torsten Hilbrich wrote: > > > You can try the aic-driver which is available at > > > > ftp://ftp.dialnet.net/pub/linux/aic7xxx/ > > > > I made some good experience, this driver seems to be more stable than > > the one included in 2.0.33 (which I currently use), especially when > > activating special features such as SCB paging and tagged command > > queueing. > > > > BTW: the aic7xxx driver changed alot in the last kernel releases, you > > might also give 2.0.34 or 2.0.35 a try. > > Stock 2.0.35 won't boot on an 2940U anyway. > It appears to work fine with the aic7xxx-5.1.0-pre5-2.0.35.patch > on ftp://ftp.dialnet.net:/pub/linux/aic7xxx/testing Stock 2.0.35 booted and appeared to run fine on my system with an on-the-motherboard 2940UW. I could not get it to find my scsi cdrom though, so I went back to the Debian packaged 2.0.34 kernel. Mike -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Lan Tcp/ip Question
On Mon, Jul 20, 1998 at 10:35:43AM -0500, Tomt wrote: > At 10:02 PM 7/20/1998 -0400, you wrote: > I have no gateway defined on either machine do I need one? route add -net...eth0 is sufficient to tell the Debian box to look for any machine on your local network through eth0. You also need some kinda routing info on the NT machine. Unless both machines can find each other, ping (other than self ping) wont get back to the originating machine. Try setting the Debian machine's address as gateway on the NT box. > Don't I need samba to access the NT machine? If your needs dont extend past transfering files, no. Mike -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Lan Tcp/ip Question
I have a Debian<-->NTsp3 network working. I am certianly no expert, but I will be glad to work with you on this. I assume you have tcp/ip installed on the NT machine. It doesnt come that way out of the box. I think you have to install tcp/ip to even assign an ip address. Make sure you reran SP3 after installing tcp/ip from the NT cd. My Debian box is set up with route add -net, no gateway, the same as yours. The NT box has the Debian box's address in as a gateway, but I dont know if that makes any difference. Somebody told me that NT quits listening after not hearing any traffic for a while. Try rebooting the NT box after setting the route on the Debian box. You dont need samba just to transfer files or print, if the Debian box is hosting the printer. NT does ftp out of the box (assuming you can get the network up of course). NT can print to a Debian box using the "lpr" method. Mike On Sun, Jul 19, 1998 at 12:06:49PM -0500, Tomt wrote: > Hi Everyone, > > Question for anyone who knows much about networking under Debian. > > I have two computer one is running Debian Linux. Kernel 2.0.29 and the > other is running Windows Nt Workstation Sp3. Now heres what I'm trying to > get done is the two machines to connect up and let the debian download > stuff off the Nt machine(I'm in the process of replacing the NT machine and > need to do some data transfer.) I've already figured out that I need > samba, and have it installed but it doesn't work. So I was reading through > some of the FAQs when it said to check and see if I can ping the other > machine, so I try that. Both machines CANNOT get any response when they > ping each other, I know the NT machine works because I can ping another > Windows machine on the network. I get what looks like an activity light on > the hub everytime I try something like ping from either machine. But > machines can ping themselves. > > Heres the setup > Windows NT machine - ip address 192.168.1.2 subnet 255.255.255.0 no gateway > defined > Debian Linux machine - ip address 192.168.1.1 subnet 255.255.255.0 no > gateway defined > > Debian Machine is running a 3com 3c509b(Very close to the best isa Nic I've > ever used.) > > I've narrowed my problem down to have to do something with the debian > routing setup. > > Heres route and ifconfig information > > route -n output > > __ > Kernel IP routing table > Destination Gateway Genmask Flags Metric RefUse Iface > 127.0.0.0 0.0.0.0 255.0.0.0 U 0 02 lo > > __ > > This is the way the output appears after login. I've been manually adding > this line > 192.168.1.0 0.0.0.0 255.255.255.0 U 0 01 eth0 > By typing > route add -net 192.168.1.0 eth0 > (I do know that this can be done through a script at startup, but I want to > get it working correctly before doing that.) > > ifconfig > > __ > loLink encap:Local Loopback > inet addr:127.0.0.1 Bcast:127.255.255.255 Mask:255.0.0.0 > UP BROADCAST LOOPBACK RUNNING MTU:3584 Metric:1 > RX packets:57 errors:0 dropped:0 overruns:0 > TX packets:57 errors:0 dropped:0 overruns:0 > > eth0 Link encap:10Mbps Ethernet HWaddr 00:20:AF:72:43:7C > inet addr:192.168.1.1 Bcast:192.168.1.255 Mask:255.255.255.0 > UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1 > RX packets:215 errors:0 dropped:0 overruns:0 > TX packets:8476 errors:0 dropped:0 overruns:0 > Interrupt:10 Base address:0x300 > > ___ > Local Loopback is working because I ping myself or 127.0.0.1 and I get a > response. > > I have tcpdump installed(very handle util.) And when I try to ping I get a > message about icmp: echo Request but thats it. > > Like I said I'm very sure this has something to do with my routing table, > but everything I've read or seen about the routing table confuses me more. > > Any ideas what I'm doing wrong > > Do I have to have a default gateway? And what is that used for? A router?(I > don't have one if so) > > Thanks in advance > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
xbase broken?
I am trying to install hamm on a friend's system. All went well till I started on X. The xbase install script fails with the following: mv: /usr/X11R6/lib/X11/xinit : no such file or directory Is there some work around, or should I just wait for it to be fixed? Mike -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: problems with X in hamm
On Thu, Jul 16, 1998 at 04:16:08PM -0500, Carlo U. Segre wrote: > > I have been trying to install a new machine with the frozen hamm > distribution and I am running into 2 problems with X11. > > 1. No user outside root is allowed to start X11. This may be a > configuration issue but I know that this did not happen in Debian 1.3 Take a look at the config and Xserver files in /etc/X11. There are some short instructions there on how to edit these files to your liking. I dont think the installation scripts would have overwritten these files unless you answered Y to the configuration file question for the applicable packages. > > 2. I use a S3 ViRGE video card and I am noticing that I have a corrupted > text console screen when I exit X (running it as root, of course). This > happens both when I am using the SVGA server and the S3V server. > > I am about to drop back to 1.3 out of frustration but wondered if these > problems are just my incompetence... > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: How to build a custom kernel with kernel-source-2.0.34_2.0.34-4.deb
No clue on question 1. In answer to question 2, I suggest you get the kernel-package package and then check /usr/doc/kernel-package for detailed instructions. Mike On Wed, Jul 15, 1998 at 09:56:31PM +0800, Alex Kwan wrote: > Dear debian fans, > > My system is hamm, and I have downloaded the > kernel source (kernel-source-2.0.34-2.0.34-4.deb) > from ftp.debian.org > > (1) What is the difference between linux-2.0.34.tar.gz > and kernel-source-2.0.34_2.034-4.deb. > > (2) How to build a custom kernel with > kernel-source-2.034_2.0.34-4.deb > step by step please (I knew how to build > it with linux-2.0.34.tar.gz) > > Thanks a lot! > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: problems with PPP & Debian 2.0 beta
"auth" is selected in /ppp/options now. It wasnt before. I had to put "noauth" in /ect/ppp/peers/provider to override it and get a connection. Mike On Tue, Jul 14, 1998 at 09:28:53PM -0500, Chris R. Martin wrote: > I recently upgraded my 'base' system to 2.0 beta, and also installed PPP > 2.3.5-2. However, when I did this, ppp stopped working. It will dial, and > it fails with the message "peer refused to authenicate". Obviously this > didn't happen before... > > I've checked pap-secrets and that looks the same.. in fact everything > looks the same that it did before. > > Any ideas? > > Thanks, > Chris > > please cc me by email > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Debian 1.3 with AHA2842B
Maybe the reason that the RH installation disks work on your box and Debian doesnt is kernel related. Make a Debian 2.0 (hamm) rescue disk and try to boot. I am pretty sure that the hamm installation disks have a later kernel than the 1.3 installation disks. If this works, I suggest you make the rest of the 2.0 installation set and install hamm. Mike On Tue, Jul 14, 1998 at 08:37:15PM +0200, [EMAIL PROTECTED] wrote: > Hello all! > > I have a problem, installing debian 1.3 with AHA2842B VLB controller. > > Hardware: > VLB motherboard with AWARD 4.50G bios > AHA2842B controller with bios enabled > Quantum ProDrive LPS 270 MB > Sony CDU 76S > > When I'm booting the root.bin, I see this on the screen: > > scsi0: AHA284x/ ... 4.0/3.2/4.0 > scsi : 1 host > scsi0: Scanning channel A for devices > scsi : aborting command due to timeout > aic7xxx: Aborting scb 0, TCL 0/0/0 > .. > scsi : BRKADRINT error(0x1): > Illegasl Host Access > Kernel panic : scsi0: BRKADRINT, error 0x1, seqaddr 0x0 > > In swapper task - not syncing. > > How can I solve this problem, please help me asap! > > I can isntall only the redhat 5.0 distribution without this error, but > I want to use debian. > > Please send a copy to my email address too. > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Loadable Modules and Configuration
Running modconf as root will prolly fix your problems/answer your questions Mike On Mon, Jul 13, 1998 at 06:20:29PM -0500, Len Cumbow wrote: > Hi, > > I am running Debian 1.3.1. I have a 3c509B that USED to work > in a previous installation of Debian. I have obviously hosed > things in the new install. The card is no longer detected > during boot. > > How do I get the card detected at boot? I think > a loadable module (3c509.o?) is involved. > > How do I get it loaded at boot time? > > What config files are involved and what do I put in them? > > Also, how do I configure things without booting from the > rescue diskette and running though a lot of unnecessary > installation steps? > > Thanks, > Len Cumbow > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: pascal.+development
I think there are some IDE's for Linux, but none seem to be as popular as Emacs/Xemacs. Emacs is not an integrated development environment per se, but it has many of the capabilities of one plus other things. Emacs has modes for C, C++, LaTeX, shell scripts, Matlab, and yes, even Pascal. font-lock-mode will give you syntax highlighting. C-h m will show key bindings for the mode that you are in. So that the vi lovers dont flame me too badly, I have to point out that vim has "modes" too, but I dont know if it has one for Pascal. Mike On Mon, Jul 13, 1998 at 04:46:37PM -0400, Alexander Gutfraind wrote: > Hello fellow users! > It's a weird newbie question I'm about to ask. but what > about Pascal? > you all seem to write in C or PERL, but I like pascal. > when I checked the pascal compiler I found it required all > types > of libraries, libc5. but shouldn't it cause some problems to > libc6? > I am not an experienced programmer, but would like to > improve that in Debian linux environment. > Is there (rather what are) C and Pascal complete > developments environments > like TurboPascal and TurboC from borland I'm using? > > TIA. -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Unknow pci device
Ignore it for now. It is harmless. As you suspect, it is kernel related. Both bo and hamm have 2.0.34 kernel packages. You can select a more recent kernel later in your installation process, or compile your own. Mike On Mon, Jul 13, 1998 at 08:27:26PM +, [EMAIL PROTECTED] wrote: > I'm having a problem with debian install disks. > Although the last kernel stable version > is 2.0.34, the installation disks comes > with 2.0.27 that causes an error in my > computer. > > Unknow pci device: 'address' : Read include/linux/pci.h > > With slackware 3.5 boot disks, the kernel version > is 2.0.34 and this error don't occur. With Slackware > 3.2 (kernel 2.0.27) a receive the same error. > > How can I fix this? Something like generate another > boot disk, or install first slackware and > debian after. > > Thanx > Anderson R. Fernandes > > > > > Get free e-mail and a permanent address at http://www.netaddress.com/?N=1 > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
setting mru in ppp
I am under the understanding that modem ppp links perform better with packet sizes somewhat less than the default of 1500. Is this still true? I am trying to set my incomming packet size to 542, but have been unable to get this to work. Am I doing it wrong, or is my provider's machine simply ignoring the request? I am running hamm (up to date). mapleloop# cat /etc/ppp/peers/provider # The chatscript (be sure to edit that file too) connect "/usr/sbin/chat -v -f /etc/chatscripts/provider" mru 542 defaultroute noauth /dev/ttyS0 38400 persist Thanks Mike -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: stupidity and disaster
My suggestion: Get a rescue disk (like the one you used to originally install debian). If you dont have one handy, use another computer to download an image from somewhere like www.debian.org and make one, following the instructions. Boot up with the rescue disk. Dont activate a swap, partition or format anything. (he he, but you prolly know that). Look for the option that lets you mount a partition. Mount the partition that bash lives on and undo what you did. Remove the rescue disk and reboot. I dont know why what you did did not work, but I am sure we will slap our foreheads and say, "no wonder" when one of the gurus explains it to us. Mike On Sat, Jul 11, 1998 at 03:56:10PM -0700, Jim McCloskey wrote: > Hello .. > > I seem to have done something very stupid. If anyone can tell me what > exactly it was that I did wrong, or better still, help me recover what > I'm very afraid might be a hopelessly trashed filesystem, I'd be very > grateful. > > The system is Debian 1.3, kernel version 2.0.30 > > I wanted to upgrade from bash-2.0 to bash-2.02 (to solve a problem > with getting Netscape 4.05 and the Real Audio Player to work > together). I downloaded bash-2.02 from the GNU archive and installed > it in /usr/local. The binary in /usr/local/bin . The next step must > have been where I committed my stupidity I renamed /bin/bash to > /bin/bash-2.0 and make a symbolic link from /usr/local/bash-2.02 to > /bin/bash (hoping that this would just drop the new bash in in place > of the old) . > > Everything seemed fine; my problem was solved; I was happy. > > When I went to boot again today, I was not happy. Thr boot proceeded > normally until: > > VFS: Mounted root (ext2 filesystem) readonly > INIT: version 2.71 booting > INIT: cannot execute "/etc/init.d/boot" > INIT: entering runlevel: 2 > INIT: cannot execute "etc/init.d/rc" > > Debian GNU Linux 1.3 (none) tty1 > (none) login root: root > Password: > Jul 11 15:03:20 login[8]: unable to change tty `dev/tty1' for user `root' > Unable to change tty /dev/tty1: Bad file number > > Trying to shutdown gracefully with Ctl-Alt-Delete gives: > > INIT: Switching to runlevel: 6 > INIT: Sending processess the TERM signal > INIT: cannot execute "/etc/init.d/rc" >Give root password for maintenance >(or type Ctl-D for normal startup): >/bin/bash: No such file or directory >/bin/sh: No such file or directory > > The root password seems to be recognized, but it seems that I've left > the system with no way to find a shell. > > Have I destroyed this nice thing totally? I'd be very grateful indeed > for any help or advice > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: URGENT hamm install help needed
On Mon, 6 Jul 1998, Ivan wrote: > Thanx for the quick reply Mike. You're welcome :) > > I can understand what you are saying re:install slowly rather than trying > to do everything at once (some people just have no patience!) > > My plan at the moment is to format my h/d (2.5Gb) put in a 12Mb vfat > partition to hold base2_0.tgz (loaded as a split file from floppies) and > then partition the balance as linux partitions. At least I can be certain > that I don't have a space problem. > > BIG, HUGE, QUESTION : Access to this list is the single most important > thing to me at the moment. If I take the default home-user packages, visit > netscape and download 4.05, will I have at least functional net access ??? > (in your opinion that is !) > I use Netscape as a browser, but not for email, but that's just personal taste. The main problem with you using Netscape for access to this list while installing is that Netscape needs X, so you will have to get X working first. You might want to consider starting out with pine, which is what I use. Advantages: easy to use, even for a beginner, and it will run in an xterm or on a console. I assume you are getting your mail via POP. I dont know if pine can fetch mail via POP on its own. If not, you will need fetchmail or something. I think elm is the Debian default mail reader, so I assume it will come in the 'home' installation. It too will work in an xterm or console. > If not, then I have lost access to all my help and will have to reinstall > windoze (yuck!) 2.5 gig is enough space for both M$ and Linux if you want both. I have two systems, a 486 and a PII. 500Mb (the whole disk) is a little cramped on the 486, but still reasonable. I run Linux in a 1 gig partition on the PII, which is plenty for me. > > After writing the first message, I read in the May or June archives of a > person who found a file that had been filled with the line " #padding ". > Having read that I remember finding something close to 20 of these files > (sorry, can't remember the path) on the first install which I think was the > final cause of "no space" error. > > It is the dependancy and conflict problems that worry me. > > I can't comment on the obsolecence of libg++ but it seemed that an awful > lot of hamm packages were very unhappy without !!! > > > At 08:19 AM 06-07-98 -0400, you wrote: > > > > > > > > > >If the size of your installed system is 400Mb, the disk space required is > >400Mb + space for the .deb files + working space. You can delete the .deb > >files after they are installed (dselect does this for you). So > >downloading fewer packages might help you. > > > >Here is the approach that I use: I divide my installation into parts: > >emacs, tetex, X, and everything else. I do my installation in phases, > >starting with 'everythging else'. > > > >Unless you have deleted them, the .deb that dselect downloaded for you are > >in /var/lib/dpkg/methods/ftp/debian/dists/frozen/main/binary-i386 > >If you feel adventurous, you could try deleting stuff that you dont need > >to install right away (say stuff in x11, tex, and editors) to give > >yourself some working room. Then install some other packages by hand > >using dpkg -i nameofpackage.deb. Delete any .deb files that you > >successfully install by hand. Once you feel that you have deleted enough > >.deb files (one way or another), give dselect another shot at it. dselect > >will know which packages were installed with dpkg and will not try to > >download them again. > > > >I think libg++ is obsolete. If you have the replacement (libstdc++ or > >something I think) you may not need libg++. > > > >Mike > > > > > > > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null > > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: URGENT hamm install help needed
If the size of your installed system is 400Mb, the disk space required is 400Mb + space for the .deb files + working space. You can delete the .deb files after they are installed (dselect does this for you). So downloading fewer packages might help you. Here is the approach that I use: I divide my installation into parts: emacs, tetex, X, and everything else. I do my installation in phases, starting with 'everythging else'. Unless you have deleted them, the .deb that dselect downloaded for you are in /var/lib/dpkg/methods/ftp/debian/dists/frozen/main/binary-i386 If you feel adventurous, you could try deleting stuff that you dont need to install right away (say stuff in x11, tex, and editors) to give yourself some working room. Then install some other packages by hand using dpkg -i nameofpackage.deb. Delete any .deb files that you successfully install by hand. Once you feel that you have deleted enough .deb files (one way or another), give dselect another shot at it. dselect will know which packages were installed with dpkg and will not try to download them again. I think libg++ is obsolete. If you have the replacement (libstdc++ or something I think) you may not need libg++. Mike On Mon, 6 Jul 1998, Ivan wrote: > > Being a home user at the moment I selected the home user download option ( > ~404Mb ), chose ftp as the install method, upgraded packages, bypassed > selection (as suggested) and went straight to install. > > As i'm on a time-limited ISP the download was interrupted several time but > seemed to complete OK ( in about 8 hours @ 33,3K ) > > The configuration phase failed completely - apparently most of the > applications are seeking libg++.so.27 & libc5 and bitching about not > finding them. Also had dependancy and conflict problems - probably caused > by an earlier package failing to install ??? > > I also got complaints about the device being full - on an 800Mb partition > am I running out of room ??? > > On the assumption that I messed something up - 8 hours later I had download > the whole lot again - and got the same problem (and, in the middle of the > night, didn't think to write down every one of the million+ errors - even > dselect gave up on me and sulked !) > > It seems almost as if the default selection for dselect by choosing the > home user option is to load everything ! > > On the second installation, noting the problems, I chose 'select' from the > dselect menu, didn't do anything and went back to the menu - no dependancy > or conflict problems were pointed out. > > 1. > > 2. Do i need to manually select packages to get libg++ and libc5? > > 3. Maybe I should stick to bo (but, if hamm is going stable at "any minute > now" would I gain anything ? would I lose much ?) > > 4. There seems to be as many problems using autoup (gauging by this list) > so I have discounted installing bo for the sole purpose of upgrading to hamm > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Using 2.1.108 kernel
Maybe you will find some helpful info at the NIC driver author's site: http://cesdis1.gsfc.nasa.gov/linux/drivers/ Mike On Sun, 5 Jul 1998, Peter Iannarelli wrote: > Hello all: > > Yesterday I tried to get the 2.1.108 kernel up and running. > I did get it running however I hand one problem, may ethernet > card (3Com 3C905 vortex) stopped. I am currently running > Debian's latest released kernel 2.0.34 and all is find. > > With the 2.1.108 kernel, ifconfig reports the NIC is up and running. > Additionally I can ping the network interface from the machine that > is running the 108 kernel, but I can't access anything on the > network. One note, the kernel seems to add a route and > the /etc/init.d/network file adds a route for that network interface. > > Any ideas or thought on this would be greatly appreciated. > > Thanks in advance. > > Peter > > > -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
Re: Modem slowdown after hamm upgrade?
I too have noticed a ppp slow down on my bo system. A while back I updated a bunch of packages, and I think ppp was one of them. Here is what I think I know about that, even though I dont understand it. I think that /etc/ppp/options used to have: mru 542 That line is commented out in the most recent bo ppp package. I tried to uncommenting 'mru 542' in /etc/ppp/options, but cant establish a ppp link with that setting. I know that mru has been setable in the past and a setting of 542 has given much better performance for me than the default of 1500. I dont know if my problem is related to yours, but both look possibly ppp version related. Mike On Tue, 30 Jun 1998, Simon Blake wrote: > Hi there > > I've a dual PPro unit that has been a 33K6 PPP PAP dialin server for the > last year (hamm for the last 3 months), and worked fine, so long as I ran > pppd v2.2 - I kept the old binary lying around and just symlinked it back > in after every ppp update from hamm, because the v2.3 binary didn't work, > and I couldn't be bothered working out why :-). > > Since installing the hamm beta a few days back, my PPP performance has > plummeted - everything still works, but I get ~1Kbps rather than the > ~33Kbps I'd normally see. Since then I've tried 2.0.34 with ppp 2.2, and > 2.1.107 with ppp 2.3.5, to no avail. Everything seems to work fine (users > authenticate OK etc), just deathly slow. There are two modems in the box, > showing identical behaviour (both are setserial'd to spd_vhi, and both > are set to 115200 in the ppp options file), and a 14K4 in another hamm > machine shows similar poor performance. > > >From my (hamm) box at home, with a 56K modem, connected to a local 56K > ISP, I get (running 'pppstats -w 10' while an FTP is going on): > > IN PACK VJCOMP VJUNC VJERR | OUT PACK VJCOMP VJUNC >47905 35 35 0 0 | 268 19 18 0 >47981 36 35 0 0 | 213 20 19 0 >49461 38 36 0 0 | 269 19 18 0 >47981 36 35 0 0 | 258 18 17 0 >49038 35 35 0 0 | 201 19 18 0 > > Same machine dialed into the 33K server at work, with an FTP happening: > > IN PACK VJCOMP VJUNC VJERR | OUT PACK VJCOMP VJUNC > 1029 3 2 0 0 | 94 2 1 0 > 1017 2 2 0 0 | 120 4 2 0 >0 0 0 0 0 |0 0 0 0 > 509 1 1 0 0 | 94 2 1 0 > 508 1 1 0 0 | 26 2 1 0 > > I have no idea what is going on here - that's a 10 second gap when nothing > happened at all! I've also run a test with minicom and sz, and that > showed similar performance - around the ~1Kbps mark, so I'm pretty > confidant the problem doesn't lie with ppp, rather that something has > changed in the serial port or mgetty setup in the new system? > > Any ideas welcome - my users are about to send out the lynch mob, and the > idea of moving the modems back to a bo box doesn't really appeal. > > TIA > > Cheers > Si -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Linux system Instalation - CRC Error
Sounds like you there may be a media defect on the floppy you used for the rescue disk. Use another floppy. This is pretty common. Here is what I do do make a Debian boot floppy set. It is kinda time consuming, but it has pretty much eliminated this kind of problem for me. DOS (Win95, whatever) format the stack of floppys that you plan to use for your installation set. Watch the output messages. If the format program marks any sectors as bad, toss the disk or at least dont use it to make your installation set. Mike On Thu, 25 Jun 1998, Romilson Cruz de Carvalho wrote: > Gentlemen, > > I got the following files of instalation of Linux Debian: > > Root.Bin > Resc1440.Bin > Drv1440.Bin > Base-1.Bin Base-2.Bin > Base-3.Bin Base-4.Bin > Base-5.Bin Base1_3.Tgz > > The RAWRITE2.EXE was used to make the instalation disks. > > When I put the RESCUE DISK in the drive and reboot my computer, the > Linux initialization starts, > but at the time that ROOT volume will be mounting, I got the folllowing > error message: > > CRC ERROR: Mounting Root File System. > KERNEL PANIC. > > Any help will be appreciate. > > Best regards. > > Romilson C. Carvalho > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: PPP or no PPP ....
I am still using bo, but most, if not all, of my reply will be valid for hamm too. On Wed, 24 Jun 1998, Gutfraind wrote: > During hamm installation I entered the details of my modem > and of my ISP (IP, user, password..). > I have started pon and have heard the modem hand-shake, but > what's next? > I didn't get any details of what have happened. Appearently, > the connection was successful, > because the modem did not disconnect. > 1) How can I check my connection status? Type: /usr/sbin/ifconfig This will show the status of ppp and eth net connections. > I reasoned that if the connction was indeed successful, > dselect's ftp would work. You could have a good connection, but it your system would be unable to resolve host names to numerical addresses unless you had nameserver access set up correctly. Try to ping a well known host by name. If that doesnt work, try by numerical address. If you can ping by numerical address, but not by name, your name service is prolly misconfigured. > I used the default connection options, and dselect simply > wrote: "failed". > What's next? 2) What could be the problem? To see what your chat script did, and what your ppp server said back, look at /usr/adm/ppp.log This location valid for bo, I dont know about hamm. > I would like to get most of the standard packages, if not > through debian, than by windoz'ing, > but, 3) where, appart from dselect, I can find the list of > the standard packages, or to know which > packs are standard (using more than newbie educated guessing > mechanism...) > You can see the package decriptions on the website. Mike -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Can't get X to Run, Please help!
On Thu, 18 Jun 1998, David Miner wrote: > I recently installed DEBIAN Linux and downloaded some packages. > My problem is when I try to start X it wont start and I get this message > back. > > Fatal server error: > No valid modes found. > _X11TransSocketUNIXConnect: Can't connect: errno = 111 > giving up. > xinit: Connection refused (errno 111): unable to connect to xserver > xinit: No such process (errno 3): server error. > > I'm running the "SVGA" server and ran XF86Setup as "root". I also had to > edit > /etc/X11/Xserver because it originally had "None" for the server Are you sure that you installed this server? For every Debinized Xserver that I have installed, the post-install script has asked something like "Do you want to make this your defalut Xserver?" Answering "yes" causes the script to make the proper entry in /etc/X11/Xserver. Mike -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [NC013@aol.com: New User]
Check out the Linux Documentation Project at: http://sunsite.unc.edu/LDP/ There are many other mirrors of this site, you may wish to check the 'mirrors' section on the site to find one nearer to you. In particular, check out the on-line book "Installation and Getting Started Guide" If you want dead-tree books, you cant go wrong with O'Reilly (www.ora.com) Mike On Wed, 10 Jun 1998, Martin Schulze wrote: > I have only one slight problem. I don't have a clue how to use Linux. I'd > greatly appreciate any suggestions on how to learn the interface, as well as > what software I ought to have on my system and how to get and install it. > > I checked the Debian site's FAQs and howtos, but didn't find them very > helpful. Anything you can tell me wuld be extremely helpful. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: XFree86 + Netscape
You definitely need to get X working before attempting to install Netscape. It is possible to download X from XFree86.org, install it on your Debian system and get it working, but it would be alot easier, especially for a beginner, to install the equivelent .deb packages. I suggest you work on getting dselect to work via the ftp method. It will resolve dependencies, fetch the files, and install them for you. Failing that, consider getting a CDROM copy of Debian. As a minimum, you will need the Xbase package, an Xserver package appropriate for your card, and some other stuff that dselect will identify for you. Mike -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Compiling a kernel
Get the kernel-package package (in misc I think). This is a wrapper for kernel compilation process that will allow you to produce a custom kernel-image package that may be installed with dpkg. On Sun, 7 Jun 1998, Chris wrote: > > Hi, > > There was some discussion a while ago about installing a new kernel. I > was just wondering if someone had some definate instructions for > installing a 'debian compliant' kernel (with correct system.maps, etc, > etc). Following the instructions in the README for the kernel source does > not work correctly (there is no mention of a system.map file for > instance), and I've got a funny feeling I should be putting in a vmlinux > (or vmlinuz ?) file instead of zImage. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
IP masq was:PS/2 mouse device
On Thu, 4 Jun 1998, Nick Gillam wrote: > Also, what Debian package is required for IP Masquerading. I have a Debian System that does IP masquerading, here is how I did it. First, echoing an earlier comment, consult the HOWTO on this subject. I read somewhere that the masquerading code in kernel 2.0.33 is more stable than in earlier kernels, so I got the source for 2.0.33 and compiled it using the Debian kernel package. You will need to compile in support for firewalls, ect, per the HOWTO. Here is my setup: mapleloop connects to the outside world using a modem and ppp. It has a regular internet ip address assigned to that connection. mapleloop connects to a dual boot NT/Debian machine through an ethernet card (eth0). I have assigned ip=192.168.1.1 to that connection. I have assigned 192.168.1.2 and 192.168.1.3 to the ethernet card in the dual boot machine. It uses one address for NT, the other for Debian. I am allowed to assign these ip's because this part of the network is private; these ip's will never be seen by the internet. I run bind on mapleloop for the benifit of the dual boot machine. This is available as a Debian package. I added the following to /etc/init.d/network in mapleloop. This is probably the wrong place for it, but it works. (Maybe someone more knowledgeable can tell us where it *should* go.) ifconfig eth0 192.168.1.1 up route add -net 192.168.1.0 eth0 ipfwadm -F -a m -S 192.168.1.0/3 -D 0.0.0.0/0 insmod ip_masq_ftp There is also ip_masq module for irc I think. You will need this module too if you use irc behind the firewall. Before I installed the ip_masq_ftp module in the firewall machine, ftp clients behind the firewall would sometimes drop connection and abort in the middle of a download. I understand you can also fix this problem by configuring your clients to work in passive mode, but it is easier for me to just use the module. I configure the dual boot machine to point to 192.168.1.1 for both gateway and DNS services for both operating systems. Not exactly on the subject but related: My only printer is connected to mapleloop. I connect to it from the Debian remote machine by configuring lp as a remote printer in the printcap of the remote Debian machine. The remote NT machine obtains print services from mapleloop by using the LPR method in printer setup. See the NT help utility for details. Hope this helps Mike -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
smail upgrade, machine no longer accepts mail
I recently upgraded my smail to the new version (from 'stable', not the deep frozen stuff) and my machine stopped accepting mail. I think I found the cause. I found this in /etc/inetd.conf # smtpstream tcp nowait root/usr/sbin/tcpd /usr/sbin/in.smtpd (will be restored by smail postinst) Uncommenting this line and rebooting put me back in business. I guess the smail postinstall script forgot to restore it. Mike -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: 440LX chipset
> I am considering buying a new computer and would like to get a > Pentium II with the new 440LX chipset. My concern is over the AGP > graphics. Would I be able to use debian with AGP? Are there any > compatibility issues with 440LX? The 440LX chipset works fine with Linux, according to benchmark data that Net Express posted at their web site: http://www.tdl.com/~netex/mb/440lxdata.html BTW, I have no relationship with Net Express. I suppose that AGP support for Linux will depend on when (and if) the people who write vidio drivers for Linux get the tech data they need to do the job. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: signify
> > Does anybody know how to use signify with pine? I've copied one example > as my .signify in my $HOME dir., but I'm clueless on how to make it work > with pine. Try .signature Mike -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: forwarding mail
[Charset iso-8859-1 unsupported, filtering to ASCII...] > How I can automatically forward my mail to some other address (to > Internet) in debian? Make a file named '.forward' in you home directory and put the address you wish to forward to in that file. If you want to do something fancier than forwarding all mail to one address, look into procmail. Mike -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
