Debian Official CD
Apologize for the cross post, but I figured there would be someone on these lists that can answer this one Is the Debian/68K tree included on ANY CD set that claims to be an "Official Debian CD" distribution, regardless of source? I'm particularly interested in staying with the 2.0.xx series right now, and I'd like to find a CD set so I can avoid all the ftp's to get X going, etc. I have several Mac's here running 2.0.36 successfully and I'd like to try X on my Quad 650 for starters... paul
Re: pcnet_cs on ThinkPad
>automatically loaded, i82365. That probably is the module for >pcnet_cs and not the one I was guessing was the correct one (similar name). Andrew, I think thats the controller for the PCMCIA slots (the i82365). The module used to drive the detected card is determined by the card id string that card services finds when it probes the slots. I may have missed some of the thread, sorry about that if this has been discussed. Check your /etc/pcmcia/config to find the mapping between detected cards and the drivers invoked when they are found. Paul
Re: Loading 2 LinuxOS with LILO
> >`LILO: slack' starts the Slackware OS but loads the Debian kernel image. >Note that whereas /vmlinuz in Debian is a symlink to /boot/vmlinuz-2.x.x, in >Slackware /vmlinuz is the kernel image itself. I've never been able to divine the solution to this either, when using a 2nd or 3rd disk, so I use a workaround that has never failed me... And that is, put the kernels of ALL those Linux's NOT on the 1st disk into the /boot directory that lives there, on whatever partition that happens to be. With Lilo on the MBR, which is what I use here, this has done the job for me! I presently have one Intel box here that has 95, NT 4.0 and two bootable Linux OS's on it, and they all boot fine with Lilo. There is no partition of the 2nd Linux that lives on the 1st disk, and only the root of the 1st one does, for that matter. Paul
Re: Wrong HD size in BIOS, Recovery?
> >I would suggest you tell the bios to auto-detect your disk, and then >cfdisk it, and make the free space into another partition. This should >avoid eating your data. > If you've installed Lilo, the system will not boot, I don't believe. Any change to the geometry will render it unable to access the disk. You probably will see the dreaded "LI" when you try. Paul
Re: dual boot?
> >Hello? >I am interest in Linux, but I don't decide yet if I choose it or not. >So I'd like to install Linux and Windows NT 4.0 together. >That means I want to make my computer dual bootable. >Is it possible? If so, how? >Please let me know. >After considering that, I intend to order one. > >Thank you. Certainly...I have a system here with W95/NT4.0 and two Linux's. No special SW is necessary IF you install things in a certain order, and even if you don't, as long as you've prepared a proper boot floppy for it, you can then install Lilo, and boot them all from there. If you only have a single disk in the machine, that likely means you will need to partition it, or use FIPS to shrink the partition that already exists...most PC's come with the disk as one giant partition which is wasteful, and not what you really want if you're gonna run Linux on it too, so the issues begin here. I don't think FIPS can shrink anything other than FAT, so if your disk is VFAT32 or NTFS, you're screwed there. If that is the case, you'll need to partition the disk, and reinstall the MS crap. In that case, install NT first after partitioning, and you'll want to be careful to ensure that the Linux kernel is fully below the 1024 cylinder limit, as imposed by most BIOS's and Lilo (I think this is still the case). So, this decision is dependent on how much space you need for NT and whether or not you have a single disk or not. Keep in mind that NT can live on another disk, as long as it can put its boot dependent stuff on the root of the 1st partition in the machine. Having Lilo on the MBR does not interfere with NT's booter, but, NT will want to put stuff there that its booter depends on. If that 1st partition is VFAT or FAT32 (whichever it is) NT can also put its stuff on the root of the partition where it lives, but in that case I think you need to use NT's booter to boot Linux also...not certain, but I believe thats the case. Anyway, I find the simplest thing to install NT first. Then, do the Debian install, and install Lilo to the MBR. The system will then go to Lilo's boot code after POST, and you'll have the opportunity to boot either Linux or NT, but you must tell Lilo where NT lives in the lilo.conf (other=/partition_where_NT_lives) and give it a label. Some SW like System Commander takes care of much of this for you, but its not really necessary, but it does a nice job for many folks. But Lilo does fine too, and you save about $90, because its really not required...Lilo can do it just fine...but the order of the install is important or you can cut yourself off from one or the other of the installed systems. Like installing Linux 1st, then NT, will clobber the MBR, and you won't be able to boot Linux unless you have a bootable floppy to get back to it. But, it can be done rather easily without hassle...and it... Works fine, lasts long time! :) Paul
Re: Can't start X on my Laptop ...
> >The Linux-Laptop Guide >(http://www.cs.utexas.edu/users/kharker/linux-laptop/) had no useful >information for me, at least none I could find ... > >I got the laptop with plain MSDOS 6.2 , so I cant look into the windows >configfiles ... > > >ok. Can anybody help me ? I'm lost at the moment. > Maybe. Been there, done that a number of times, but I did get lucky. Take another look at the Linux LT page. You'll find there many references to the timing parameters (the modeline) info. Thats the secret. I've managed to get X working on a number of LT's by 'borrowing' modelines from LT's that are cousins. Choose a card that forces the use of the SVGA Xserver, and plugin your borrowed modeline. Edit XF86Config so it only has one mode choice (640x480). If you're real lucky your LT might be able to do 800x600 but none I've tried could do that. One chipset that will definitely force the SVGA server is CL553x I think it is, and I use that selection in other cases to force that Xserver to be the one symlinked to /usr/X11R6/bin/X. Good luck! Paul
Re: LILO and two disks...
John, If I were you, I'd get that "linear" attribute outta there! Try it. Paul
Re: hplaserjet6L/pcmcia questions
> >I'm having trouble with my HP Laserjet 6L printer. APSfilter nor magifilter >have drivers for my printer. Is there any alternative? > Try HP IV. paul
RE: Getting a list of attached devices
>> I am trying to determine the name assigned to my CDROM drive so that I >> can mount it. Anyone have a suggestion ? dmesg | grep cd paul
Re: hosts.deny problem.
>hosts.allow >ALL: localhost,128.206.x.x : ALL > Whenever I've put an IP in, I always also add the mask, a la: ALL :localhost, 128.206.0.0/255.255.0.0 AND LEAVE OFF THAT SECOND ALL! >hosts.deny >ALL: ALL > Thats what you want. Paul
Re: file managers
> >So, anyone wanna chime in with why they like their favorite file managers >(under X)? I'm on the lookout for the cream of the crop. I hate to say it, >but the Winblows team has a good file manager, and I'd like similar I like the filemanager in TkDesk. paul
Re: Samba and NT Wkstn?
Leo, > >Well then, I guess wishing you good "luck" is all I can do! > OK, and I understand. I was hoping I could get an answer here, but I guess I have to try Samba's site because no one here seems to have any idea. Its crazy, 'cause everything I've read about it suggests that NT 3.51 Wkstn should just flat out work! In this case, machines running WfW, Win95 or NT 4.0 all work just fine, but NOT 3.51! And, the doc's just seem to say it should not be an issuebut it is! On three different versions of Samba, it refuses to connect to a resource, all giving the same "Access is Denied" error box. Actually, the 3.51 is installed on my laptop along with Linux and WfW. The WfW works fine, no problem, but not the NT 3.51. It does everything just fine except map a drive from Samba. Oh well... Interesting that NT 4.0 (non-SP3) has no trouble either... >If you have time, when you solve it I'd like to hear what went awry. > Sure will let you know. Thanks for your help. Paul
Re: Can anyone suggest a dialup terminal emulator?
>I think seyon can handle all of these requirements. > Actually, minicom is among the most accurate emulations I've ever found. Its as close to perfect as a VT220 as you'll find, at any price :) Paul
Re: Moving partition
>> In fact, my problem is : How do I move files from an old parttion to a new >one while >> ensure all links (and things like that) are kept ? > This is the way I've done it dozens of times...its from the days prior to cp having the correct attributes to do it correctly, and maintain last access times, etc... mount the target filesystem (lets say at /mnt) cd to the top of the one which you want to copy...lets say /usr then, this command: find . -depth | cpio -pdmv /mnt The result at /mnt will be AN EXACT duplicate of the filesystem. I understand that the more recent GNU cp will do the same, but this is the way I've always done it, and still do. Paul
Re: Midnight Commander hanging
> >I have seen a previous posting but no fix for Midnight Commander >hanging. > I'd bet its trying to find your DNS. Been there, done that. If you have a nameserver id'd in your /etc/resolv.conf if its not really accessible when you launch MC it'll do that every time. paul
Re: Moving partition
> >You could also make a new partition and mount it as /home, or /usr, or >some such. > Glad to hear that suggestion as that is precisely what I would do! Actually, I never put it all on one partition anyway. Old habits are hard to break. I like to get /var and /home OFF the root if possible, and generally use a single separate partition for /usr also. That also makes it easier to take advantage of multiple disks and the 2nd IDE controller on most m'bds these days. I typically put the 1st disk along with the CD (if ATAPI) and a 2nd disk, even if its small, on the 2nd IDE port. Put swap and /var on it, or even /home. Another advantage of this is, when you decide to try another distribution you just use /home with it, and swap, plus, you often have a partition or two around that comes in handy when it comes time to resize one, or move stuff around, etc. You'll find the system performance will be a little better too, 'cause it'll be able to take better advantage of the capabilities by keeping the I/O going on two disks instead of all of it on one, plus the additional controller channel. Paul
Re: Network cards.
>http://www.adaptec.com/products/index.html#fastethernet > Nathan, oopps, guess I missed that myself! paul
Re: resolve.conf specification
> >BTW, not only it would have to be able to choose the right resolv.conf file, >but also the right pap-secrets file, I believe... or I think there's a way >of writing different userIDs and passwords in pap-secrets... ? > You must have missed the earlier part of the thread. Supposedly, pppconfig will deal with it all, except for the /etc/resolv.conf; ie, the nameservers to use. Easiest thing to do about it is to use your own DNS, on your own machine. You could run a caching only nameserver which would work for any and all ISP's you might ever use. OR, as is done with NetBSD, use your /etc/options file to call a script that copies/renames or links the proper resolv.conf for each one. I choose to use my own DNS, in caching only mode, but thats mostly because my ISP sucks, and either their nameserver or mailserver is down every other day. paul
Re: Network cards.
> >Just go for an Adaptec. You will have problems that you never dreamt >of. The topic was about ethernet cards! paul
Re: [fsck]: file system errors ...
> > Is anything going wrong !? It happens, sometimes, even when not having >failure power or mounted disks. > Nino, Are you actually shutting down with "shutdown -h now" and waiting for the System Halted message, or are you just killing power? paul
Re: adsl
> >The problem with DSL is that the router provided is usually configured >with Win95 and unless the installer brings a laptop, you are going to need >a Win95 system to set it up. The units they offer here can be logged into >and set up manually but I have not seen an installer that knows how to do >that. I ended up setting mine up myself. > George, Thanks for that info. In my bud's case down in Va. he has two Win95 boxes on his home net, along with a Mac and several Linux's of course, so he should be ok. Where are you? The reason I ask is I'm in Bell Atlantic area, as is my friend down there. He can get it and I can't. Paul
Re: Configuring scsi controller on ISA bus
>Adaptec 151x, 152xaha152x=iobase[,irq[,scsi-id[,reconnect]]] >also when I tried to access the cd from install program it errored and >showed >(maybe 'insmod driver'?) > Thanks for any help you may provide David, Yah, I've had trouble over the years myself with the syntax for the Lilo command line. But, sometimes the issue is that the particular driver for that specifc interface REQUIRES elements beyond the iobase, irq. Both there, and if you try to use insmod driver to get it into the running kernel. Problem I've had with that is that you never know which do and which don't NEED that additional info without trying to determine it from the source. Or, if there's some doc for it, I don't know where to find it. Paul
Re: Network cards.
Some PCI NICs may be troublesome, but you can count on most older 3Com's or Intel. Personally, I have several 3Com cards, and a few LinkSys that are NE2000 clones, and none of them have failed to work with any Linux. I'd check out the source of your rumor. Sounds like crap to me! Paul
Re: adsl
A friend of mine in Va. recently found out that he can get this service too! We discussed it at some length, and as near as I could tell the ADSL equipment provides a UTP ethernet port to which you connect it to your NIC. Sounds pretty much straight up to me. Paul
RE: Exchange Client
Hmmm, I can use plain old MicroSoft mail to pop from my Vax-cluster, or Linux...are you saying that Exchange doesn't provide a pop3 server, or can't? Paul
Re: help please! still unknown interface and SIOCSIFADDR!!!!!
>eth0: unknown interface > >I have setup my NE2000 to be IRQ=5 and IO=0x300. Please help!! Rino, The eth0 message means the kernel did not find the ethernet card during its last boot. You could try the "module" and see if it will initialize the card. Ie: insmod ne2 300,5 should insert the module in the running kernel. If that doesn't work, I would suspect that some kernel option has not been selected (when you rebuilt it). If you still have the original kernel around, boot it and see what gives. All kernels since 2.0.0 will find an NE2000 clone; at least those of the well-behaved category. Note that there are many that do not have the proper signature byte, but I've been using LinkSys cards for years, and they've always worked. Besides that, the Linux driver even deals with many of those of the poor clone class as well; usually. Paul
Re: Help - X meets my OLD monitor
> >Hope you find that site! Wish I could find the URL, sorry. > There may be others, but I suspect this is the one he couldn't find. Try: www.nashville.net/~griffin/monitors/ paul
Re: Connecting two computers via serial
> >Is it possible to connect two computers (desktop and notebook) via a null >modem >connection? And, if so, what software would be suitable to make the link? Anthony, Sure, if you must. You could use SLIP or PPP. Or, better yet, use the parallel ports; a connection known as PLIP. See the HOWTO's. Paul
Re: resolve.conf specification
>wouldn't it be possibile to make a bunch of files, called >/etc/resolv.cont.isp1 >/etc/resolv.cont.isp2 >.. There is a script that deals with this very problem as part of the ppp kit for NetBSD. It resolves this issue not by linking to, but by copying the proper resolv.conf for that connection to /etc at connect time. Paul
Re: Samba trouble
>The problem is most likely that your username in windows is not the same >as the debian username you are entering the password for. > Or, /usr is not a share. On systems where I use Samba, I always create the accounts with adduser and passwd. AND (this is important-if you want it to work that is) don't use mixed case in either the username or password. Windows can't deal with that. Make sure you are logged on to Windows 95 with the same account. Start/ Shutdown/Close all programs and logon as a different user, then logon to make absolutely certain. Paul
Re: Interesting Problem (PCMCIA ethernet and modem)
>> After some struggles I was able to install a PCMCIA ethernet card. It is >> working wonderfully. Unfortunately intslling it affected my modem card. >> Now when I dial using the modem it makes a connection but I can no longer >> telnet of ping outside. I changed my /etc/init.d/networks file to the old >> one without eth0 but it did not help. Any ideas? > Check your /etc/pcmcia/config filethere you may need to exclude IRQs or ports to use. You may have cardmgr setting your modem to, and using an IRQ that is being used by other hardware, especially if you have sound and/or other devices in the machine. In that file you can tell cardmgr what IRQs to use, or not use, to avoid that. You can't specify what it will use if an when it detects a serial port (modem) but you can juggle them to avoid conflicts. If its not already being used, enable it to use IRQ 5 and see if that doesn't help you out. paul
Re: problems+suggestions
> >Wouldn't it be nice if a configuration of all installed packages could be >saved in some data file that you could put on a disk and whenever you want >to install a new machine or go back to this good configuration you just >load the data-file into dselect or whatever program that gets/installs >packages. Some different default configs could also be shipped with >the dist to be used by new debian people. > >Maybe this is already possible? Tell me please! > SuSE Linux does precisely that using its YaST tool. Paul
Re: Samba
> >I've just installed the Samba-server (and read the man-pages). >I might be missing something but I can't figure out how to get a list >of what Wimpdos calls "the entire network". Can this be done without >knowing >any server-names? > I expect you're on a wimpdos 95 box, but you didn't say. Yes, there is, but you need to first make sure that your windows machine and the Linux/samba box are in the same WORKGROUP. On 95, double-click network in Control Panel, then check the ID tab to see the workgroup the machine is in. The server and client must be in the same workgroup with the normal defaults in place. To really do anything and access any "shares" as windows calls them, you must have some shares setup of course, or when you click the server in Network Neighborhood you won't see anything available. And, you probably won't be able to "map network drive" to assign any of them to a DOS drive letter unless you have an account on that server AND you are logged into windows as that same user. So, if you have a user account on Linux, login to windows as that same user using the same password. Be careful, and don't use mixed case passwords or user names because Windows gets awfully confused if you do... :) Voila! Paul
Re: PPP connection speed
>Is there anything wrong with the script? I use them in Redhat as well, and >it didn't cause trouble at all... > > Shao, I don't think so...but one thing, unless you've altered the /etc/ppp/options file, the default with RedHat has only the statement "lock" in it. That means the rest of the options will take their defaults as far as pppd goes; whatever they may be, when you use RH. The difference may be there. Paul
Re: PPP options for different providers - how ?
> >Probably similar to the "call" parameter? that is what is used with the >debian "pon" script. this reads the relevant file from /etc/ppp/peers/* >for each set of options. > Michael, Well, see, therein lies part of the problem with this ppp thing. It changes so much with each release, with refinements or whatever, that the use of it is altered. For me, I've been using it for so long and to avoid having to set things up differently for each round, I continue to use my old scripts. I've never used pon and until this thread didn't even know about /etc/ppp/peers! Over the years I've continued to use the scripts and such to avoid the turmoil of the thing changing every time. I need to catch up with it :) Paul
Re: PPP options for different providers - how ?
>The above file is only for global options to every ppp connection that >will be made on your box, which rules out username parameters. I generally >as a rule never touch it. the Debian defaults seem to work okay. Several pppd versions back there was an option (-u) that specifies such parameters. I believe it is now deprecated, but it does still work! :) When it will stop working is anyone's guess...but, it still does. I've been using it for years...it allows you to specify a file from which the login information is supplied. Short of using any supplied script to provide the proper options file for multiple logins, I've always maintained separate ones for each ISP and use a small script to ensure the correct options file and chatscript are used for each. Without the flexibility of the complete options file, it is possible that you may find an ISP whose link setup defaults simply do not work. For example, I've experienced using RH's default options file (which uses only the statement 'lock' and is otherwise devoid of any options) fails to establish a viable link to specific ISP hardware. Using an options file like that, essentially accepts all the defaults of the ISP setup and since you have no options file, all of the defaults are in force there...sometimes these will conflict! Specifically, using that setup may result in the wrong IP assigned to the link, but only when certain hardware is on the receiving end of the call, that sort of thing. The resulting link does not work in that case, of course. You may also find the options noipdefault and defaultroute are not set properly for a particular ISP. In short, its probably not reasonable to accept the defaults and one size does not fit all when it comes to the /etc/ppp/options file. You may find that out if you do use two or more ISP's! Paul
Re: How big is my H/D ?
>If you need to use DOS you'll have to upgrade the BIOS, twiddle the BIOS >to work (try something like CHS=1023/64/63), or use a hack like Ontrack >Disk Manager. > Not so. Just make sure that your DOS partition and any Linux root is completely within the limits of the 504MB. The problem comes from the BIOS being old and not permitting access to cylinders beyond the 1024 limit. Linux doesn't have any problem with that, UNLESS you use Lilo. Linux itself DOES NOT use the BIOS, but Lilo does, and that is why any root must be within the limits of the BIOS itself. Just use the rest of it for the other partitions, such as /usr and /home. Its kinda cool in a way because you can have everything other than the root itself beyond the limit, and DOS will never see it and therefore can't screw with it either. Not only DOS by the way, but NT won't like it either...the difference tho' with NT is that if it sees a partition out there, it will refuse to use it, but it will see it. Paul
Re: Lpd dying
Make certain there is a spool file, and the name is the same as the print queue. Check /var/log/messages too. Paul
Re: whois default
You could edit the named.ca file. Paul
Re: Two idential root partitions - how?
Mark, >I'm installing Debian. Good choice! >I'm partitioning as follows: > >100 Meg root partition >3 Gig /usr partition >100 Meg root partition - identical copy > 1st question; is this all one big disk? For what follows, I'm assuming it is. I like to do something like this 100M / 64M swap 100M /var (and I prefer putting this on a 2nd disk if possible) whatever's left for /usr if its all on a single disk. If there is a 2nd disk, I also like putting /etc and /home there, preferring to put any additional root or additional Linux root on the 1st. Just my preference...and the idea I have anyway about the separate /var is that it happens to be quite active and alot of write operations on that with a running system. I like the idea of keeping write operations off the root as much as possible, just in case a drive goes nuts..I know it doesn't happen as much as it used to maybe, but I still do it. I also think you're more likely to lose a disk, than you would blow up your root partition. Thats been my experience. Another reason for the 2nd disk. Also why getting /var off the root makes good sense to me. Also, by putting /etc and /home on the 2nd, if you should lose the 1st disk, then you won't be worrying about all the config changes you might have made, and so on. Now, to duplicate root...I know I do it the hard way, and partly because I'm not very clever with scripting, but I use cpio to duplicate filesystems. Have for years, but like I said, I do it the hard way...one at a time! For example, I do this: Mount the target where you want to duplicate a filesystem on at /mnt lets say. Here, lets duplicate /usr # cd /usr # find . -depth | cpio -pdmv /mnt This is old, and like I said, not very clever, but it does make an exact duplicate of /usr at /mnt, maintaining links, file access and update times, etc. When its done you can just umount /mnt and mount it over top of /usr and your system wouldn't know the difference! I know, I've done it hundreds of times...it just flat works! Even doing it the hard way, you'll have a duplicate system in a few minutes on that spare parition. Where the lack of sophistication comes in using this is, don't do it from /. I've done it without thinking about what I was doing, and if you think about it you'll see why its pretty dumb to do! If you do, when it gets down to /mnt on that root (where you've already recreated everything down to that point of course), it'll do it all over again! You get all that stuff twice on the new filesystem! Not what you wanted to do of course. Now, like I said, I'm not real clever with script programming, but I'm sure you'll get some other ideas too, that are a bit smarter than this. This is just a bit of history for me, and it works so well I've never bothered looking into doing it any other way. And it is pretty fast too! And simple. Since I fool around with several Linux distributions I'm always reconfiguring my disks, and I use this simple command to do it. Thats another reason for keeping /home on a separate partition, 'cause that way, its /home for several; they all mount it, and all the stuff from any of them that are on /home are then availabe on any and all of them. I always like having two bootable Linux's on any box! Saved my buns more than once. Actually, I have one box that has three; SuSE, Debian and RH. The RH is only bootable from floppy, and its a minimal sort of system, all on less than a 300M partition. Its sort of like the emergency system. I've found its easier to fix a Linux box if you can boot Linux on it :) Paul
Re: modem connection speed
>How can I tell at what speed I am connecting to my ISP under linux? > Check your modem manual. There is an AT command that will tell your modem to report not the DTE port speed (the speed of the interface between your serial port and the modem itself), but the actual carrier speed--the speed of the modem to modem connection. I'm not sure how to tell the ppp daemon or chatscript to tell you that but if you call them using minicom, it will report the speed when the connection occurs...or, it will report whichever the modem has been commanded to report, I should say. Here, if you've told it to report the actual carrier speed, thats what you'll see. Paul
Re: no pcmcia drivers on install
>Uh, it's on the _previous_ menu. When you are offered "Install Drivers" >select "Alternate: Install PCMCIA Support" _FIRST_. I may be wrong about this, but from the past, this option is intended for those that need pcmcia support during the install. For example, an ethernet pcmcia adapter will be used to mount an NFS partition or something like that. Support for pcmcia devices for the running system is in a package, which installs the card services, and config stuff for the cards that are supported. At least thats the way it used to be. Paul
Re: [Dual boot] dual boot NT and debian?
You don't need anything more than Lilo to do this. I tri-boot Win95, NT and Linux all without anything more than Lilo. With these three, the order that works the easiest is 1st Win95, then install NT, then Linux or if Linux is there already, boot it, edit lilo.conf, and then run lilo. If done in this order, you need only id one of the uSlop OS's, because the NT loader will find and have an entry for Win95. They both are booted from it in this case. Actually, on this same machine there is a second Linux (RH) but I boot it from floppy. Never have figured out how to do it right with Lilo when the root is not on the first disk, other than to put the kernel in /boot of whatever Linux IS on the 1st disk. I wasted alot of time trying to figure that out but soon tired of all the boot tries that failed! :) It was just easier to put the kernel in /boot than to struggle anymore trying to get it right. Either that, or put the root partition of both on that 1st disk. Used to be harder to get away with because of disk cyl limits, but thats not in the way these days. On systems with multiple disks, I like using them all, and therefore put the partitions all over the place...and I keep one disk on one IDE, and the 2nd on the other too, for any possible advantage performance wise. That being the case its easy enough to keep the root of all of them on the 1st disk too. Thats also why I keep the boot floppy handy just in case, and that floppy only bootable Linux is for emergencies more than anything else...and yes, its come in very handy more than once! Much easier to fix a system if you can boot a Linux up somehow; anyhow! ;) paul
RE: Win 95 + Linux
On 05:51:09 [EMAIL PROTECTED] wrote: >>Is it possible to connect a win95 machine to a linux with a serial cable >using PPP? > >-- The question in my mind is why you'd even bother? Just use slattach. On the Win95, the SLIP support is there, but its hidden. You'll have to get it off the CD distribution. It works fine, they just don't install it. Using slattach is far easier to deal with in my opinion. A small two-line maybe three, script and its done. Especially useful for laptops that you might not have a network card for, etc. Works fine, lasts long time. And a lot less hassle. Paul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
RE: debian-user needs a charter!
On 20:25:44 Jim Pick wrote: > >Now it has been hijacked for use as a batteground for a flame-war. >Not good. > >Bruce, could you please write up something simple - perhaps stating that >debian-user is a "user support" list only, and that from now on, >political discussions should go to debian-debate? (or some other list). >Also, name-calling, profanity, and baiting is just not acceptable. > >Please, make it clear that debian-user is not an anarchy like 'usenet'. >We don't have to tolerate inappropriate behaviour (like carrying on >political flamewars). > >Cheers, > > - Jim Well said Jim! End it. Now! If they won't knock it off, expell them from the list. Done. Paul pgpXwD1fJoqJ4.pgp Description: PGP signature
RE: EtherLink III
On 04:51:01 Matthew Tebbens wrote: >>Is anyone running Debian using an EtherLink III 3C589C pcmcia card ? >I'm looking for a module/support so I can use the card. > >Thanks, >Matthew > Matthew, Well, I use it on my old laptop but not running Debian. The laptop runs (of all Linux's :) Slakware! And its old, based on kernel 1.2.13. The 3Com is supported by the pcmcia 3Com driver module that has been in the pcmcia modules for some time. Oh, by the way, the reason the laptop doesn't run Debian is that I have no floppy drive for it, and Slackware, running on a umsdos filesystem (not supported in Debian) using loadlin, etc, offered me a way to get Linux onto it by ftp'ing (it has WFW on it, so I had ftp and uSlop tcp available). That way I was able to get various kernels on it and finished the install by ftp and so on. The 3Com works nicely with the 2.somthing.29 pcmcia stuff. Works like a charm, and I use it every day at work to check out trubls, net links, etc. Paul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
RE: dosemu w/ graphics mode in X
On 22:44:28 [EMAIL PROTECTED] wrote: >>Does anyone know how to get graphics mode to work in X with dosemu. I >have seen screenshots of windows 3.1 running under X in a xdos box. I >don't want to do anything that exotic. I want graphics mode in an >xdos box. > >Any hints? > Erv, If you get an answer in a private email, please pass it on to the list! I'd like to be able to run some ham radio programs under dosemu, and they use some of the graphic mode calls to build their display. I don't have the source for any of these, so can't identify the particular function in use, but some of it does display properly in VGA mode. One thing that I do see happen is once one of these offensive programs is run, and then exited, the cursor disappears until I kill that dos box and start it up again. One of them goes into some sort of hi-graphic mode, that doesn't function at all (displays a map of the world, with sun and moon position and track). These programs are widely used in the EME community (Earth-Moon-Earth) and no corresponding functionality is availabe directly under Linux. Paul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
RE: Aug 4 21:54:00 carlf modprobe: can't locate module net-pf-5
Carl, I have just recently changed my ISP to Bell Atlantic, and have had to make changes to my pppd options file in order to establish a working connection. This info may help in your case, I'm not at all sure, but it might apply. >Trying to make a PPP connection to AT&T's Worldnet service, I have >created a script that will dial. The modem connection is made, but >ifconfig never reports a ppp0 route. In my case, I could never get logged in, since there is no actual prompt or login process which is much different for me and both of my most recent ISP's. I've always had that process scripted with CHAT. Now that has to change also...argh! Anyway, this is what I was able to discover to get it working quickly. By the way, I'm still using Debian 1.1 (yeah, I know, a little behind the curve, but you now the story...if it works...). I had to alter the CHAT script of course, but that in itself was an interesting event. Since there was no prompt issued, I just ask Chat to expect the modem connect stringand then let it go. In the /etc/ppp/options file I had to turn 'passive' off, and enable 'silent'. The difference is supposed to be that passive mode of LCP means it solicits the authentication and then sets up the link. This didn't work at all with this ISP, and this is the mode I've always used with the other two over the last three years or so. In my case, this silent mode of LCP causes a rather pregnant pause in getting the connection going, but so far it has worked OK, so I haven't played with it anymore. Perhaps there's a better way to deal with this. The pause is about 30-40 seconds during which no traffic passes in either direction.It really means silent when it says 'silent'! Har! Anyway, once the ISP's LCP process gets tired of waiting around and spits out a packet, off it goesand starts up my link. When my options sets 'passive' the traffic goes back and forth, but the authentication fails and the line hangs upso this is what I do. Now, since they use PAP, I also had to use the '+ua' option (which is supposed to be obsolete) and bury my login/password in a file which that option will read during the authentication process, which does, of course succeed! As I said, not sure if any of this applies to your specific problem or not, but thought it was interesting anyway. Hope it does. Paul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: splitting up the debian-user mailing list
On 13:16:26 Randy Edwards wrote: >>On Fri, 1 Aug 1997, Bruce Perens wrote: > >> It's getting kind of loud here. I've been thinking of splitting the >> debian-user list into several lists: > > I know what you mean. Just this week I turned off a couple of other >general Linux lists I was subscribed to simply because I was dealing with >too much mail. > >> debian-user: user discussion >> debian-install: installation problems >> debian-novice: newbies are intimidated by other lists, so here's >> one of their own. > Here we go again! We just had this discussion a few months back. And my opinion hasn't changed one bit! I don't like the idea of splitting them up, said so before, and thats still where I am. Been on this list for several years, don't post much, but learn ALOT from the dialog, and splitting them up is not the answer, I don't think. Paul > > Regards, | Debian GNU/ __ o > .|/ / _ _ _ _ _ __ __ > Randy| / /__ / / / \// //_// \ \/ / > (FidoNet 1:325/805) | // /_/ /_/\/ /___/ /_/\_\ > ([EMAIL PROTECTED]) | ...because lockups are for convicts... > >-- >TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to >[EMAIL PROTECTED] . >Trouble? e-mail to [EMAIL PROTECTED] . -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Accessing NTFS partition.
On 19:19:16 Philip Rangel wrote: >>[EMAIL PROTECTED] (Paulo Ramos) writes: > >> >> Hi people, >> >> Unfortunally, I have to work with Window$ NT (Not There). Four weeks ago, >> NT crash, freezing the computer (crash proof, they say), so I press the >> reset button. After that, when NT is starting up, in the blue screen >> appear a message saying that NT can't acess the disk. The hard disk as two >> partitions, 20 MB primary partition in FAT format and a 180 MB extended >> partition in NTFS. In the NTFS partition I have all my work (one year of >> work) and no backup. When this happen I contact Micro$oft support, but >> they know less then me. Since I read that debian is capable of access NTFS >> partition, can any one tell me how I can do that? >> >> You should be able to mount it and read it, but maybe not fix it! Welcome to what is known as "the blue screen of death"! I've been treated to that myself on more than one occasion. NT is very sensitive to hardware additions/failures and perhaps thats what you've had happen. Good luck! Paul >well i think i've seen a debian pkg in projekts/experimental >but haven't tried out this nor the original. >the original program is from: > >Martin von Lvwis >[EMAIL PROTECTED] > >and as sombody else said it is very alpha. > >I didn't get from your message if you can boot NT and not acces the >partition. >But if you just can't boot you could try to get another HD install NT >on it and try to acces. > >Philip Rangel <[EMAIL PROTECTED]> >c/o Bashford >Lokstedterweg 112 >20251 Hamburg Tel: 040/4808512 > >-- >TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to >[EMAIL PROTECTED] . >Trouble? e-mail to [EMAIL PROTECTED] . -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Debian-Lite : The Project
On 15:18:49 "Robert D. Hilliard" wrote: > > I think this should be the main object of a smaller version of >debian (Please not Debian Lite!). A normal debian installation loads >up a single user machine with a lot of unneeded and unwanted server and >network administration stuff. > >Bob > Man, do I like this idea! I really like Linux, and all it offers, and I really appreciate Debian, but just yesterday I was saying to myself, I wish that someone with the skills would put together a Linux better suited to the single user environment where many (most?) of us use our home systems. Free from all the hassles of permissions, root privelege to do this or that, etc! Right on! For a user like myself anyway, it sure makes a helluva lot more sense than all the multi-user protections/permissions and such. I for one, get really frustrated with such things, and it really ticks me off that if I ftp a file then I can't move it to some directory before I unzip it or thin gs like that. Everyone says don't run as root and use 'su' but damnit, some of this is nuts when the machine is really only an individual's workstation, or at least I think it gets in the way, and probably frustrates the hell out of a lot of people that finally give it up! Paul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: splitting up the debian-user mailing list
On 01:41:40 [EMAIL PROTECTED] wrote: >>Paul, > >I agree that we've had this argument before. What has changed between then >and now is that the number of postings has become 2 to 4 times what it was >before. I'm hoping that I can bring some clarity to various lists by >splitting them up. > > Thanks > > Bruce >-- Yeah, I suppose its inevitable really, and I understand (I read the list every day of course, and have for years now). I like the idea of Debain News being separate if that makes sensebut as for user, I feel that 'newbies' are user' s too, and ALL Debian user's should be there; newbie or not. Hell, I stil feel like a newbie myself, and compared to many of the real talents here, I am. But I wouldn't want to miss what they have to say either! I'd rather see less of the chatter from the maintainers actually, than miss the user stuff. Most of what they have to say goes way the hell over my head anyway :) Paul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
RE: PPP connection problem
On 02:07:28 Lazar Fleysher wrote: > >send (ATDT) >Serial Connection established >Using interface ppp0 >Connect:ppp0<-->/dev/ttyS2 >LCP: timeout sending Config-Requests >Connection terminated >Receive serial link is not 8-bit clean >Problem: all had bit 7 set to 0 > >Does anyone know what the problem is? > One thing that can cause this is Xon/Xoff flow control interfering with the setup of the low level link (the LCP process itself). However, it looks like your serial port may not be init'd correctly (not 8N1) by setserial. Paul >e-mail to [EMAIL PROTECTED] . -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: determining serial link speed?
On 07:06:32 Lawrence wrote: >>Martin Steigerwald wrote: >> >> Hi! >> >> Is there any easy way to find out, what speed my modem connected to >> the ISP? (using ppp & chat on Debian Linux m68k Amiga). >> >> I want to be sure that it connected at 28800 baud and not at 14400. >> > If its a Hayes modem, there is a command that will tell the modem to report the DCE (modem-to-modem) speed rather than the DTE port to modem speed. That value is returned on the local DTE, but how to get chat to echo that so you can see it is another matter. Paul PS: Oh, forgot, that Hayes command is: ATW1 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Directory permissions.
On 13:28:43 Nathan E Norman wrote: > >Using the correct tools is important. David gives you one such tool - I >personally type the following command in the directory I wish to copy: >"find . -print | cpio -p /target". This is of course a simplification; >find and cpio have a lot of powerful options, and people will argue the >merits of tar vs. cpio all day. It works for me. At any rate, mc is not >up to the task. > For an even more precise duplicate of a directory hierarchy I've used this: find . -depth | cpio -pdmv /target This will create an exact duplicate and traverse any sub-dir's, preserving symlinks, permissions, and dates of everything from the current directory down thru the tree. I've used this many times moving an entrie filesystem such as /usr from one partition to another, etc. Works slick. It is also true that the newer GNU tar offers the same functionality as well. But, I've used this so often, that I simply trust it implicitly. Paul >> > system structure standards. I don't remember if it gave ownerships but >> > it's worth a shot if someone knows what it's called and where I can find >> > it. >> >> No, it's far too short to cover ownerships, only locations. The 1.2 FSS >> doesn't give it's own location(!), but only that of its FAQ which is >> ftp://tsx-11.mit.edu/pub/linux/docs/linux-standards/fsstnd/FSSTND-FAQ >> > >I concur. You're in for an install - I can smile cos I've bombed >my machines a few times. > >> P.S. I don't see multiple copies of your postings here. > >Nor do I. > >Nathan > >-- >TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to >[EMAIL PROTECTED] . >Trouble? e-mail to [EMAIL PROTECTED] . -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: DEITY TEAM -- REQUEST FOR FUNCTIONALITY and COMMENTS
On 14:30:40 Leslie Mikesell wrote: >>One more idea to throw in the pot: > >How about including smbfs in the base kernel and allowing installation >from a Win95 or NT share? Almost every office is going to have one >of those around where you can share out a CDROM with a couple of >mouse clicks. You could even do from with Windows-for-WorkGroups if you >mangle the names to fit but that probably isn't worth the trouble. >This might help a lot of people get their first Linux system up on >machines that don't have their own CDROM drives. > >Les Mikesell Les, Now thats a great idea! Think about it guys, this is good! Paul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: syslogd will not start
On 15:25:22 Hamish Moffatt wrote: >>On Sat, Apr 05, 1997 at 11:04:39AM -0700, Rick Macdonald wrote: >> Speaking of syslogd, when I boot the machine it hangs for 5 ro 10 seconds >> when syslogd starts. It didn't used to. >> I'm running from unstable. Anybody know what it's doing? > >Does this here too ... Keeps the machine at high CPU load for that >time too. (*) > I see this here on 1.1 also, but I thought it might be due to my system losing the sense of "who" it is! I no longer see the hostname on the login prompt, or if I do a uname -a , for example. And, no info goes to /var/log/messages (since syslogd doesn't start I presume). Where else, other than the /etc/hostname does it keep what it believes to be the hostname of the system? If I knew that, I think I could fix this on my endI think!? Paul
Re: My screen gets messed up :-(
On 17:14:44 Paul Wade wrote: >> After I do a 'more (binary file)' often my command prompt and everything I >> type is unreadable. This has probably something to do with ANSI. How can I >> correct this 'problem'? >> >sometimes the clear command solves it > >sometimes I logout and login again > >Paul Wade, CEO Try the 'reset' command! Paul >*
Re: Why is PPP so screwed up!?!?!
Shawn, Well, I respectively disagree. This thread has been going for so long, most of the chatter is NOT about how to establish a connect. What I was seeing here was all this crap about how screwed up Debian was, and how screwed up PPP was, and The real truth is that if someone sees a NEED, like I said, then they should do something about it, if they can, and to Debian for the benefit of all. One message after another, that if uSlop this, or uSlop that just ticked me off! I've been on this list for over two years now, don't post many messages to be sure, but I had seen my fill the last few weeks with all the attacks on Bruce, for this and that, and this damn thing got started. I'd seen enough. If you think the doc is lacking than write some, if you're capable. I read the list to learn something and to follow the Debian development, and personnally I think these guys do a terrrific job with it. If you are capable and able, help them out by adding something where you see a need. I'm not capable of that, or knowledgeable enough. But all this whining don't get it either. Paul >> Why can't you guys give it up! ITs not a Debian thing dude! Its >> compiled into the kernel itself (ppp support). It is NOT a Debian >> issue! I do agree that someone (how about you?) could make it >> better by writing something, from your perspective that might >> help. BUT STOP PUTTING IT ON DEBIAN, OR ANY OTHER >> DISTRIBUTION! If you want to see more of it, monitor the FreeBSD >> questions forum sometime! FreeBSD supports two (different) ways >> of invoking PPP to establish a net link. If you want to do some >> serious whinning about it, try FreeBSD! Now, quit the bitchin' and >> get busy putting your ideas down so it will benefit everyone else. >> Thats the idea here. You guys been bantering this back and forth >> for over a week now. All the whinin' and cryin' won't make a damn >> bit of difference. So, knock it off, and put your energy to good use. >> Or, go somewhere else and bitch.
Re: Why is PPP so screwed up!?!?!
>thats intuitive Why can't you guys give it up! ITs not a Debian thing dude! Its compiled into the kernel itself (ppp support). It is NOT a Debian issue! I do agree that someone (how about you?) could make it better by writing something, from your perspective that might help. BUT STOP PUTTING IT ON DEBIAN, OR ANY OTHER DISTRIBUTION! If you want to see more of it, monitor the FreeBSD questions forum sometime! FreeBSD supports two (different) ways of invoking PPP to establish a net link. If you want to do some serious whinning about it, try FreeBSD! Now, quit the bitchin' and get busy putting your ideas down so it will benefit everyone else. Thats the idea here. You guys been bantering this back and forth for over a week now. All the whinin' and cryin' won't make a damn bit of difference. So, knock it off, and put your energy to good use. Or, go somewhere else and bitch. Paul >> >>Second, why should someone have to "cat /usr/bin/pon" to find out to >>edit /etc/ppp.chatscript anyway? I mean, I have to say that the Debian >>installation routine is so simple it's silly. So why doesn't it just go >>that one extra yard and ask a simple question like "Will this system be >>connected to the net via a PPP connection?" and then let me specify >>which com port and what the dialup number/login-id/password are? Even the >>people who claim that "ppp was a breeze" admit that they had to hunt >>around a little before they found the right files to edit (either that or >>they cheated and asked a friend). >> >>It's silly. There's no mention in the installation guide that ppp is even >>included in the base system, or that someone with only dial-in access >>to the net can use ppp with dselect's ftp method to add new packages. >>Instead, you get tossed into dselect without even being given the chance >>to get ppp going for the first time (even if you *did* know the files >>to edit). Now *that's* intuitive. >> >>Don't get me wrong, Debian is great and it's got some really cool features, >>but those features are not made apparent to a newcomer to the distribution. >> >>Here's an example. A buddy of mine in San Francisco recently e-mailed me >>that he had just installed Debian (his first Linux) and that he needed to >>know how to read a DOS floppy because he had used Win95 to download some >>tar'd and gzip'd msql source or whatnot and wanted to copy it onto the >>Debian machine. >> >>I explained to him the Debian packaging system (which he hadn't been made >>aware of in the docs he read or the install program)... and how he should >>go get the .deb files and that, when he's feeling brave, he should >>get ppp going and use dselect. So, he went and ftp'd (with Win95) some >>deb's and copied them over to the Debian machine and used dpkg to install >>them. Then, he got gcc and was stuck because it was bigger than a 1.44M. >>So, I told him that it was time for the "rite of passage", that he was >>going to have to go with ppp and dselect (which, again, he was not made >>aware of). It took about 3-4 days of e-mails before I got a message from >>him with the subject "I'M ACTUALLY DOWNLOADING WITH DSELECT!". >> >>The ordeal shouldn't be such that it would cause him to act like he had >>won the lottery. I guess my point is that Debian is not doing a good job >>at embracing the "first-time linuxer" and probably even the "first-time >>Debian linuxer". As a result, Debian is THE thing to be running as long >>as you've got a friend who has installed it before who will help >>point you in the right direction. >> >>And it doesn't have to be that way. I'd be glad to help... but in order to, >>I need at least SOME indication from others that they agree there's a >>problem here. Up until now, all I've heard is denial. >> >>- Joe >>
Re: Why is PPP so screwed up!?!?!
On 14:06:41 Gary Lee wrote: >> I think you all have very good points. I have used 3 different >ISPs and had to set them up 3 different ways. I would be lovely if PPP >could figure out how to connect to the ISP and what I want to do with >it--without me telling it (but thats not FUN). I don't think anyone is >saying that PPP is broken, just that it could use a little polish. Until >some dear soul has the time and inclination to add to the good work >already done, we will have to help each other. I do believe that's why >this list exist. >And thats all I have to say about that... > >Gary Lee I do the same thing here. Its pretty simple. Create the proper chatscript and options files with whats needed for those connects, and write a little script to copy them to options/chatscript when needed. Could be a little smarter, to figure out who you want to connect to at some point, but for me thats not neccessary, so I just do it the 'easy' way. Its not pretty, with little buttons and bows, it just works and does the job thats needed. Its no bid deal, really. But these folks that want it all handed to them, and don't care to do anything for themselves, should go along, follow uSlop right over the edge, just like little lemmings. Nuff said! Paul
Re: Why is PPP so screwed up!?!?!
Look, you don't seem to get the idea here! If you see a NEED, and are so smart to be able to solve the problem for everyone, WRITE IT YOURSELF and submit it to Debian! Furthermore, saying its NOT is the documentation is just totatlly wrong! If you chose NOT to install the HOWTO'S or INFO, thats your problem AND not the guys that put it all together. Have you read the Net and Serial HOWTO'S? OR, GASP! the PPP HOWTO? Linux, any of the distributions, comes with enough doc to keep you busy, and make you real smart, if you just take the time to look at it! Its all there man, all of it. Saying its not is like saying the sun isn't gonna come up tomorrow because Debian is so screwed up or PPP is at fault for aids for Pete's sake! If you want it to be 'easy' just go back to uSlop, and let them take away all the options and make all your decisions for you! If you don't want to learn what makes it tick, and find your problem, then you'll fit right in with their philosophy. Paul >they cheated and asked a friend). > >It's silly. There's no mention in the installation guide that ppp is even >included in the base system, or that someone with only dial-in access >to the net can use ppp with dselect's ftp method to add new packages. >Instead, you get tossed into dselect without even being given the chance >to get ppp going for the first time (even if you *did* know the files >to edit). Now *that's* intuitive. > >Don't get me wrong, Debian is great and it's got some really cool features, >but those features are not made apparent to a newcomer to the distribution. > >Here's an example. A buddy of mine in San Francisco recently e-mailed me >that he had just installed Debian (his first Linux) and that he needed to >know how to read a DOS floppy because he had used Win95 to download some >tar'd and gzip'd msql source or whatnot and wanted to copy it onto the >Debian machine. > >I explained to him the Debian packaging system (which he hadn't been made >aware of in the docs he read or the install program)... and how he should >go get the .deb files and that, when he's feeling brave, he should >get ppp going and use dselect. So, he went and ftp'd (with Win95) some >deb's and copied them over to the Debian machine and used dpkg to install >them. Then, he got gcc and was stuck because it was bigger than a 1.44M. >So, I told him that it was time for the "rite of passage", that he was >going to have to go with ppp and dselect (which, again, he was not made >aware of). It took about 3-4 days of e-mails before I got a message from >him with the subject "I'M ACTUALLY DOWNLOADING WITH DSELECT!". > >The ordeal shouldn't be such that it would cause him to act like he had >won the lottery. I guess my point is that Debian is not doing a good job >at embracing the "first-time linuxer" and probably even the "first-time >Debian linuxer". As a result, Debian is THE thing to be running as long >as you've got a friend who has installed it before who will help >point you in the right direction. > >And it doesn't have to be that way. I'd be glad to help... but in order to, >I need at least SOME indication from others that they agree there's a >problem here. Up until now, all I've heard is denial. > >- Joe
Re: Why is PPP so screwed up!?!?!
Look, if you can't get a connect, its NOT Debian, and its NOT PPP in all likelihood! If you do get logged into your provider, and then can't communicate to the net, its NOT either Debian OR PPP. Its probably YOUR setup (most likely), or your ISP (least likely). PPP is a link level protocol, responsible for setting up and maintaining the low level link; NOT the tcp/ip network! Different animal! If your scripts and options file doesn't setup the link properly, then its PPP and its setup at fault. If it does establish the link, and your script gets you logged in, then its not PPP, its your tcp/ip net setup. None of them are the DISTRIBUTION's fault, NONE! More often than not its probably your own, since it works, and works well for so many of us. Can an ISP create an environment that simply won't work with the default options file ? Absolutely! But again, THAT IS NOT THE DISTRIBUTION! So knock off all the smacking these guys for your own problem, or that which your ISP has created! They can't do anything about it! >By that standard, we should discuss only dselect and dpkg. Isn't Right, but its NOT productive to blame Debian or any other distr. for this, especially when the problem is more likely than not your own or your script. >Why not try to make it less daunting? Is it supposed to be some sort of a Then why don't you write it and stop all the bitching if you're so damn smart? >Judging from what I see on the linux newsgroups, many are using Windows95. >Microsoft evidently makes it easy. Why can't Debian? Actually, I rather like access to all the options that you DON'T get with uSlop! Paul >--- >John HaslerThis posting is in the public domain. >[EMAIL PROTECTED]Do with it what you will. >Dancing Horse Hill Make money from it if you can; I don't mind. >Elmwood, Wisconsin Do not send email advertisements to this address.
RE: Mail list problems??
On 19:29:46 Scott Stanley wrote: >>Every time I post to the debian user mail list I am getting 5-10 error >messages saying the mail could not be delivered. Although, I do get a >copy of the mail sent back to me from the list. I am wondering if this >is related to the problems with the mail list, or if I am the only one >getting these errors > Scott, You're not the only one. I'm seeing them here to. Paul
RE: Why is PPP so screwed up!?!?!
I agree! This is getting pretty boring, with all the silly ranting and raving. For Pete's sake, the Debian guys didn't create PPP in the first place! Take it to those that did, if you're really that stuck! It's really not so damn difficult to be honest, but the first time or two it might seem daunting. I mean c'mon, what do you think the rest of us are using to get to the net? Good grief, this thread has been a marathon already! Almost as bad as all the grief thrown at Bruce lately over the publicity thing, or the mess about some damn contest or other..gezzz! >I don't see what all the whining is about. > Paul > >It is not difficult for you or me to get PPP working. It *is* difficult >for many people. There is a real problem, and saying over and over "It >isn't hard for me" is no solution. >-- >John HaslerThis posting is in the public domain. >[EMAIL PROTECTED]Do with it what you will. >Dancing Horse Hill Make money from it if you can; I don't mind. >Elmwood, Wisconsin Do not send email advertisements to this address.
Re: Good buy or not?
On 19:39:15 Paul Christenson wrote: >It better be a VERY good price, since it's basically a dead-end machine. Well, it was. Original selling price was $1649. I got it for $698. Why? Not this one. I can plug in several different Pentium or even a Cyrix upgrade chips. And it'll go to 128MB of RAM. Three empty drive bays too. > >You can put in a bigger hard drive, MAYBE more RAM, but that's about it. > The RAM is standard, 72 pin simms. Plenty of space for my Conner 850's also. >As for it having a P75, I don't see much difference between my P133 at >home, my P100 at work, or the P75 on my test bench. > I agree with this. Thats why I said originally that I knew this cpu would NOT be much better, if at all, as compared to my AMD 133. But, my goal was to get into a Pentium, which could be readily upgraded. I did that. And I'm pleased with the machine, their support thus far, and all the software that came with it also. Oh, yeah, it is for Win 95, but I already took that off to m ake room for Debian. :) Paul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Good buy or not?
I wonder if anyone on the list may have experience with Packard Bell machines, running Debian ? The reason for the question is, a local store is selling excess stock that didn't sell during the holiday season at a price that is very, very, tempting. Actually, the price is only a little more than the cost of a decent motherboard, and it is for a complete system, including monitor, 4X CD-ROM, etc. The processor is a 75Mhz Pentium. I don't know the whole story of the Pentium line, but would it be reasonable to assume that the cpu could be readily upgraded? Thats a detail that might depend entirely on the capability of the board in the machine, and I know little about PB, other than the adverse stories of the recent past. Are they still using "refurbished" stuff and selling it as new? Do their machines use standard memory components, so they could be easily upgraded with parts from other vendors, etc? They don't mention the vendor, but the machine includes an video accelerator type of card, and the machine is billed as a "multimedia home PC". The monitor is one of those goofy looking things with speakers glued to its sides. It is a model 4240. Anyone have any comments on its insides, and whether it might be as good a buy as it appears to be? My current machine is an old 486 box, and I need space for an additional HD, etc. Running an AMD 486/133, so this machine would not really be much of an upgrade in itself in terms of performance, but if possible, I would quickly upgrade its cpu and memory. Would this be a decent platform to build on or not? Thanks Paul -- This message was delayed because the list mail delivery agent was down.
RE: PPP Dial up help
On 00:22:43 Knight wrote: >>Ok Mr Newbie here again. I have been trying to get Debian to connect to my >local ISP threw an external modem. When I issue the PON command it dials >the connection fine, the modems connect, but then the line drops. The PLOG >stated that I was dropped by the peer. Eddie, Well, yes, very likely, the option file is the key. Make sure it is set for r ts/cts and not xon/xoff flow control. The option file also has about a bazillion options f or the control of the LCP or IPCP processes themselves. You should check your messages file (/var/log/messages) by using "tail /var/log/messages" right after trying a connect. There you should see a more descriptive (hopefully) reason the modems are hangin g up. It could be your option file is set for PAP or CHAP and your ISP isn't, or whatever. In my case, I have a chatscript set up that logs me in and starts the PPP sessio n. It is called from the option file itself. After starting PPP up the chatscript ret urns to the option file where the actual LCP and IPCP negotiations are controlled. If you a re by chance not using rts/cts, you may see LCP timeouts in the messages file, and that would be the reason the line hangs up. If LCP cannot establish the proper contr ol over the link, nothing else happens, so it hangs up the line right now! :) Paul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
NFS ?
(Sorry for the formatted message the first time) Hi Deb'lers! I'm having a really confusing problem with NFS. Running Debian 1.1 here. The problem is in mounting filesystems on our Vax, which is running TGV's Multinet. The problem appears to be related to the uid/gid passed to the server from a mount attempt. Is there some way of passing a uid/gid of other than root when attempting an NFS mount? I've tried the -o options and specified the gid and uid, but apparently those options are not permitted when the target is an NFS filesystem, as the error "unknown option" results. The same filesystem is mountable by PC's using Ipswitches Acadia product, which defaults to a uid/gid of -2 (65534). Oh, the mount fails the same way when using either Debian, Redhat 2.1 or 3.0.3. Multinet does provide a uid/gid translation function, but it apparently does not accept root's default of 0,0. It seems I'm stuck again in that proverbial hard place...Any ideas? Geez! Paul -- This message was delayed because the list mail delivery agent was down.
Re: accessing files over serial cable?
On 09-Nov-96 John Rulnick wrote: >>> " " == Bruce Perens <[EMAIL PROTECTED]> writes: > >Now I've got a PPP link between my laptop and desktop, so I would like >to run dselect, etc., accessing file systems on the desktop as if they >were local to the laptop -- I assume NFS is the thing. I will plow >forward into these uncharted waters now, but if anyone would care to >offer NFS advice, I'm all ears (or is it eyes?)... > Yes, NFS mount the disks and CD etc, that should be the ticket. However, there really was no need to do PPP connect between them. I use SLIP to do the very same thing here. You'll find slattach far easier, and it will likely be a bit better on thruput. There is nothing truly magic about PPP but one thing it does do is that it has more overhead per frame. For local machines connected by a cable, SLIP is easier and probably better. I say probably because as long as both PPP implementations support all the options, it really doesn't matter, and it would be essentially the same as SLIP. The overhead is essentially a don't care. Paul -- E-Mail: [EMAIL PROTECTED] Date: 11/09/96 Time: 15:46:29 This message was sent by XF-Mail -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: time to split the list?
On 06:20:26 Boris Yati Beletsky wrote: >>On Thu, 5 Sep 1996, Bruce Perens wrote: > >|>I would propose: >|> >|> debian-install >|> debian-nontechnical >|> debian-technical >|> >it's a very good idea , but i don't understand whats debian-nontechnical >would do? > Well, I don't like it. I've learned alot from the dialog, but I must admit that often you guys get talking about issues that are unique to the packaging or something, and has nothing to do with the user. Its more like the maintainers often dialog here when that may be better done elsewhere. Like, what I meant to do with this or that, but didn't have the time.Then the response goes, well, why not do it this way, and we're offMaybe there should be a separate debian-install, but the other stuff is pretty much debian-user (minus the developer or maintainer nits this and nit that) if you ask me. Paul >___ >Boris Beletsky <[EMAIL PROTECTED]> >(finger [EMAIL PROTECTED] for pgp public key) >___ >Commorandi natura deversorium nobis, non habitandi dedit
RE: Dosemu
On 04:17:39 [EMAIL PROTECTED] wrote: >>Has anyone succeeded in getting the latest versions of >Dosemu to work? I can successfully compile, but can't >insmod emumodule. Insmod dumps fifteen or so lines saying: > >x undefined >... >Loading failed! The module symbols (from linux-2.0.7) don't >match your linux-2.0.7 > >What's going on?? Seems as if there are some system modules >( I use a very modularized kernel ) which are not loaded. >However, I use kerneld, so all neccessary mods should be there. > >Someone please help!! > >Thanx... :-) No, it doesn't work here either. My I-Connect CD was cut on 12 june. I haven't fussed with it at all, but just wanted to respond that in my case it doesn't work either. Paul
RE: fvwm-95 - Making progress, but not there yet
On 15:51:24 Jim Worthington wrote: > It's complaining about a missing "rplay" library >Where can this package be found? Don't know, and didn't see this when brought up fvwm95. >I tried running fvwm95 but it crashed on startup producing no messages. >Is it likely that the missing rplay library is the cause? > I don't have sound either, and didn't see this. My first attempt did bring up fvwm95, but with no taskbar, no icons, only an empty xterm. I had to edit the file locations (I used the example.fvwm95) for the module path, icon path and pixmap or bitmap, whatever. Then it worked fine. Using it right now in fact. Paul
RE: How to migrate a Debian system to another hard drive?
On 11:20:51 Tim Egbert wrote: >>Is there a simple way to migrate a Debian system to another hard drive? > >On our Debian system, we have two hard drives. The primary drive has the >root directory and swap partition and is bootable. It also contains the >various Debian software packages, programs, libraries, etc. The second >drive just has the /home directory tree. > Tim, This question of replicating a filesystem is one that comes up often. I use cpio to do this, and so far its been flawless for me. It creates an exact duplicate of a filesystem, and the following command preserves file access times, symlinks, etc. Works nice, and its simple. Just cd to the root of the filesystem you wish to duplicate, and issues the following command: # find . -depth | cpio -pdmv /target_directory This is based on cpio's "pass thru" mode. Everything appearing on stdin goes to stdout Paul
RE: Do you use SLIP or a variant with Debian?
On 16:47:57 Mike Taylor wrote: >>If so, I'd appreciate a short note from you. I'd like to know if you >use SLIP because PPP is unavailable, more expensive, or otherwise >inconvenient. > >Thanks >Mike Yes, I do. At home I use SLIP (slattach) to connect two other machines (both running MiniLinux :) to my Debian box. Also, I do use SLIP to access our network at work. Actually I use both for that, but since some of the terminal servers I have there do not support PPP I use SLIP. The only real advantage I see with PPP is its a bit simpler to configure as far as the provider of the connection is concerned. The protocol does the work, for the most part. Not so with SLIP. For the uses I make of it the authentication is not an issue, nor is the support of protocols other than tcp/ip. So I use SLIP. As for the Debian box, the changes with PPP and its implementation on Linux over the past two years have been a little difficult at times, but once its working, its fine. The real advantages it offers over SLIP are not very important if you don't need or use them. Performance wise, there's little difference. Paul
RE: help ?
Shankar, > >- >That's it; here the system hangs. This looks like it might be the "auto-config" setting in the setserial script. There is a caution there about using it, but I've never had any trouble with it myself. However, I do edit the file and turn off the "auto". Uncomment the manual config lines for the appropriate ports that you actually have in the machine, and try that. I must tell you though that with Debian I don't do this anymore because it works properly! Other distributions fail in setting the proper IRQ for my 3rd serial port (cua2 set at irq 5). The "auto" setting does work for me on Debian, and properly sets even this port with no problem. However Redhat, versions 2.1 and the new 3.0.3 release, have never handled this properly using the "auto" and must be set manually. Try turning off the auto and see what happens. Paul
RE: Qs about Linux setup on Compaq 486 notebook
Re: Qs about Linux setup on Compaq 486 notebook I'm want to use Linux on a Compaq 486/25SL notebook. I'm stuck on the following issues and would appreciate any pointers. 1 I believe the video ROM is being shadowed. The BIOS setup has options to shadow the video ROM at two different addresses, but none to turn it off. I've seen the cautions about shadowing, but have never been bothered by the existence of it on my system (an AMI BIOS, several years old now). I've run all the Linux's available including Redhat, Debian, Slackware, Minilinux, etc, and had no problem from it with any of them. The reason: Linux doesn't use the normally shadowed memory addresses. It avoids the total 384K of memory completely (!) on most distributions of the kernel it is unused and marked reserved. It is not used. However, there are some that attempt to use portions of it, such as LinuxLite, to enable them to run on very low memory configurations. Otherwise, you should not have any grief from this at all. 2 Some memory is reserved, presumably for shadowing arcane stuff, including but not limited to battery management pop-ups (or is that in the video RAM?). Now, you mention a potential for problems with some laptops. Many vendors DO reserve portions of the "upper memory area" (that 384K normally used for hardware options, shadowing the ROM's, etc, for special purposes that are intended to improve the performance of the laptop or notebook. In some cases this memory can be recovered by the user, and in others it cannot. With some laptops, this could render the machine useless I suspect, but with most it should be of little consequence. 3 Who is reserving this memory? Does the BIOS do it? Or is it DOS? Its almost always the BIOS functionality that does this, and on many it is done to provide additional cache to improve the disk performance, etc. Since Linux (Debian) install instructions ask you to turn off all shadowing, I can do that automatically if DOS is doing the shadowing, but I get the impression that the BIOS does this. The how do I go about disabling shadowing and not reserving memory? Well, as I said above, it might not be a debilitating problem. On some models it could be. 4 These machines were supposed to have a small partition with diagnostics programs on it. I don't see anything but a Big-DOS partition with FIPS and other HDD utilities. Can there be a parition that such an utility cannot see? Sure, DOS can only see certain types. For instance, it can't see a Linux partition either! Its just like it doesn't exist. However, you should be able to see it using Linux's fdisk. The problem there of course is you'll have to be running Linux to be able to use it! You could install something like Minilinux using an umsdos file system (installs to a directory on an existing FAT partition). Its actually quite small (only will use about 40 megs). 5 Some utilities I have used show 256k video RAM, but there's supposed to be 512k. The same utilities see 16450 UARTs on my serial port and a 16550A on the modem board, but MSD sees 8250s on the serial port. How reliable is this stuff? Are there definitive h/w utilities in Linux? Video RAM is troublesome. Always has been, for any system. Linux's setserial utility will tell you exactly what they are. It works. MSD does report certain "emulations" of 16450 as 8250, and many do not pass muster for the 550 either. There are many 16550 "compatibles" and emulations that are not really 16550's. Thats also common. 6 After resizing partitions using FIPS, the installation boot disk vomits a hda: read_intr: 0x40, but I figure that's because of wrong information in the BIOS. I don't have any idea what you mean by this, so can't comment. > information in the BIOS. > >I'm looking at the Troubleshooting info in the LDP and the PC h/w faqs. >Any information will be highly appreciated. >Thanks, >-bhaskar
RE: how in the world?........
On 13:32:56 Mike& Candy List wrote: >>can I delete unnecessary partitions?Ibegan to install debian on C drive >but decided to wait until I could install a discrete hard drive to >dedicate to debian or other linux.This accomplished, I attempted to >delete the partitions I created on C drive, but fdisk tells me I can't >delete extended partition while logical drives exist but when I attempt >to delete the logical drives I'm told that no logical drives are defined >both disks are relatively small so it is of substantial concern that I >am able to reclaim this space. any suggestions?...Mike List Mike, Seen this one before too! I think you probably created the partitions with Linux fdisk, and are attempting to delete them using DOS fdisk. Doesn't work! You'll have to remove them with Linux fdisk or format the drive, I think. You could try setting the partition type to a DOS FAT (I think its type 6), but you'll even have to use Linux fdisk to do that. DOS will see the partition, and calls it NON-DOS, but it won't let you manipulate the table entries for it at all. If you can, set it do a DOS type, and then you'll be able to delete it with DOS fdisk. Paul
Re: ipfwadm?
> >Actually, either of the above tips by themselves may fix the problem as I >think they are essentially doing the same thing. > >For what it's worth, this is how I set up masquerading on my machine, but >I know there are several ways to do it. The way you are doing it sounds >like it is working. > > /sbin/ipfwadm -F -a masquerade -S 192.168.100.0/24 -D 0.0.0.0/0 > >Good luck, > >Gerry >[EMAIL PROTECTED] Gerry (and all the other respondents too) Thanks! You all are saying pretty much the same thing and it sounds like I was really looking at or misinterpreting what I was seeing from tcpdump. I think the idea that I should have been using a 3rd machine to do that may be key here. And Gerry, the command line above is one I had seen in a previous post either here (maybe it was you then too) or some other Linux list and that is what I tried originally, in fact. However, as your line is constructed its saying that masquerade is a "policy" and the man page for ipfwadm (on Debian 1.1) doesn't mention that being a valid policy (is the page not current perhaps)? Anyway, I think, with the help of the Debian list, that if I don't have it right now, the info you've provided will help me get it right. Thanks all! Paul
Re: NFS trouble
> >My 2 cents' worth on this one is that my (new) system had full NFS >capability until I installed the netstd package, at which point I got the >above message. > >Poking around showed that /etc/init.d/netstd__nfs had all 5 of the _start_ >lines commented out. I am a Unix dumbkov but uncommenting the nsfd and >mountd lines got things going again. This may not be the way to fix it. > >HTH > >Lindsay That is precisely what my problem was also None of the daemons were started 'cause they were all commented out Paul
ipfwadm?
Anyone here on the Debian-L know the secrets of using the ipfwadm utility to set up masquerading? I've built a kernel with the proper options but I'm concerned about whether I'm really masquerading, or just forwarding packets. How do I prove it? There was a recent LJ article on using masquerading, but it was primarily based on the software prior to the recent advances. In particular, the method used to establish the masquerading ruleset, and verifying just what has been set is my concern. There is no longer a ipfw, but now the management utility is ipfwadm. The concern comes from the setting of the masquerade rule. The ipfwadm has an option (-M) for masquerading, but, this is NOT used for setting the rule, and the only valid option is -l, for listing of the masquerading rules. The only way I can get a rule set is to use the following command (does this really result in masquerading or not is the question): #ipfwadm -F -i masquerade -P all -S 192.168.210.0/0 -D 0.0.0.0/0 The reason for the question is this; the ipfwadm -M -l shows no masquerade rule set. And, I got on the net using this last nite, and sure enough was able to get out to the net from my laptop, and using Lynx, got out to the Web. I discovered that several links on various pages were not accessible from the LT, but they were if I ran a browser directly on wb2oyc (my Deb1.1 box). This got me wondering if perhaps my reserved net address was getting thru my ISP to the net, and thats why I wasn't able to get to some of the links. So, I ran tcpdump on wb2oyc while doing this. Sure enough, there I see packets sent from the Web host directly to the address of the laptop (!) which is assigned the address in the 192.168 reserved space and shouldn't ever get thru my ISP's router! In other words, I was not masquerading for its address; I don't think. Bummer! Worse, my ISP is not stopping those packets. Paul
Re: NFS trouble
Sherwood, >> I keep getting the following message when trying to NFS mount any of the >> exported filesystems. >> >> mount clntupd_create: RPC: Program not registered >> > >Check that on the nfs server that mountd is running. This one is usually >NOT started from inetd but instead is started from one of the rc files > Thanks! I think you're right! Just got your repsonse, and looked, and no mountd...There are four nfsiod's running, but no mountd. Also, I tried the old Debian 0.93R6 last nite, and it still works with that system, so this is probably it. Thanks alot for your answer! Paul PS: My distribution came from I-Connect; the CD was cut on Jun 16. Your suggestion led me to the problem; thanks again. I found all the lines that execute the daemons commented out in the script in /init.d. The script was there, and being called on entry to runlevel 2 by the link S25netstd_nfs, but all the daemons were not executed by init because they were all commented out!
Re: Don't use Matrox cards (Was re:dosemu)
On 12:45:58 "Richard G. Roberto" wrote: >>On Mon, 19 Aug 1996, Richard G. Roberto wrote: > >Ah, this may have changed. When 3.0.3 was out in beta, the >Metro X server was also beta. The Xfree beta drivers blew >it away in performance tests posted on Xfree's web page. >I know this because I was looking for an alternative to >Slackware for my home machine and looked into RedHat and >Debian both. When I asked Redhat about the Metro X server, >they said it was a beta server that had a lot of debug code >in it, thus the slow numbers. Do you know how the full >release performs? > I tried the Metro X server also on my RH system. It sucks! Came up with the darndest wierd color combinations, and it didn't seem to pay any attention to fvwm at all--not the colors anyway! Some of the gaudiest, gosh awful combinations you could imagine. Most of them useless, due to the lack of contrast or the color choices of fg & bg. And it was (is) slow. Real slow. Paul >it but there is no "From" field (no colon). That's strange. > >Anyway, thanks for the info! > >Richard G. Roberto >[EMAIL PROTECTED] >201-739-2886 - whippany, nj > >-- >*** >Bear Stearns is not responsible for any recommendation, solicitation, offer or >agreement or any information about any transaction, customer account or account >activity contained in this communication. >***
NFS trouble
Debian guru's, Here I am again in need of some assistance with this new release. I've been using NFS over a SLIP link to my laptop to access resources on the Debian box (like the CDROM, etc) with the R6 distribution for several months. Now, with 1.1 my laptop and another machine both using MiniLinux I keep getting the following message when trying to NFS mount any of the exported filesystems. mount clntupd_create: RPC: Program not registered What's going on here? I've checked the /etc/rpc file, and they're almost identical on the two machines. Have I forgotten to move something from the old R6 system to the new one, or is there something else I need to be doing to have this work? I've copied the /etc/hosts.allow, hosts.deny and so on from the R6 system. To make matters worse, it DOES NOT ALWAYS NOT WORK! I understand this may not be (or is it probably not) strictly a Deb1.1 issue, but was hoping to here something from the collective wisdom here that might help. Thanks Paul
Re: GOOD TIMES
On 01:15:00 [EMAIL PROTECTED] wrote: >>Only Kidding! > > Bruce >-- > Clinton isn't perfect, but I like him a lot more than Dole. >Please register to vote, and vote for Democrats. >Bruce Perens AB6YM [EMAIL PROTECTED]http://www.hams.com/ I'm not kiddking! He IS a VIRUS! ;-) Paul
RE: GOOD TIMES
On 01:10:00 [EMAIL PROTECTED] wrote: >>Your system has now been infected with the GOOD TIMES virus. Have a nice day! >-- > Clinton isn't perfect, but I like him a lot more than Dole. >Please register to vote, and vote for Democrats. >Bruce Perens AB6YM [EMAIL PROTECTED]http://www.hams.com/ You're right about that; Clinton IS a VIRUS! Please register AND Vote. For the sake of the country, NOT for Clinton! Paul
RE: how to boot single-user mode?
On 11:44:51 Ken Gaugler wrote: >>Last night I tried and tried to get my system to come up in >single user mode (so I could do some major filesystem changes). Ken, I'm not certain, but you can get to it by issuing the command 'telinit 1' while logged in as root. Thats one way. >For background, I need to change the size of some partitions, so >I need to copy whole partitions to a temporary location while I >re-do the partition table on the target disk. I hope to avoid >having to re-install from tape or from scratch this way. Perhaps a suggestion for you to consider to accomplish this is using cpio. For example, the command 'find . -depth | cpio -pdmv /target_dir' will make an exact copy of a filesystem, preserving links and file access times, etc. If you execute the above from the top dir of a filesystem, it will make a duplicate at the 'target_dir'. I have used cpio like this to copy the entire root and/or /usr partitions to new locations with no problems whatsoever. Paul
Re: catch 22?
Thanks to all the Debian L'ers for the suggestions on how to deal with this problem, especially Dwarf, Guy, Jim and Heiko. I did manage to correctly get a fresh source and image installed by using Heiko's suggestion (or a very close version of it)! I changed the status file to read " purge ok not-installed" and that did fake out dpkg. Thanks guys! Appreciate all the help. But, now, another has shown up! You guys are gonna get tired of me stumbling around. Seems if I didn't have bad luck, I'd have none at all. My CD is on a SB16 card. I've told the new kernel NOT to load modules (N to kerneld) and somehow, what is now happening is SB driver loads during boot, and then later in the boot process I see loading modules, and you guessed it, it loads sbpcd as a module also! Only the module doesn't see the line passed to the kernel specifying the parameters for sbpcd, so it polls for every darn possible address and clone drive! Now my systems takes about ten minutes to boot! Almost as long as my Vax's at work! :-) Anyone tell me what question I answered wrong to cause this? Paul
Re: catch 22?
On 14:19:59 Guy Maor wrote: >>On Tue, 6 Aug 1996 [EMAIL PROTECTED] wrote: > >> Is there some way to force dpkg to reinstall (or remove) in spite of the >> error it encounters attempting to remove the older package first? > >Type `dpkg --force-help' for instructions on forcing options. I think >you want --force-remove-reinstreq. Be forewarned that you can >seriously damage your installation with some of these options. > >Guy Thanks for the try Guy, but no help there. Been there, done that. The remove (or forced install) doesn't work either, for the same reason. The prerem and/or the postrem script fails. Actually, I thought the force option would be the answer, but I haven't been successful with it yet. Paul
RE: catch 22?
On 23:36:54 [EMAIL PROTECTED] wrote: >> masquerading, somehow or other both my kernel-image and kernel-source >> packages have gotten to a state where I'm stuck fast! > >Before you try anything drastic, check the #!/... line of the >post-install scripts for the kernel. Mine pointed to a non- >existent perl. Either modify the script(s) or create a link >to the correct Perl. Once I had created the link, the install >went forward without a hitch. > Owen, Thanks for suggesting caution. Others have suggested the scripts having that error also. However, I have managed to resolve the dilemma. I tried Heiko's suggestion and edited my status file. I changed the kernel-source and image status to "purge ok not-installed". Then ran dselect, since I wanted to install a few other packages as well. All went fine! Except that now the postinst script fails. This is not dibilatating however as the source is now intact and the original kernel image (and the compressed) are where they should be. I believe this same error occured on the initial install of the kernel stuff as well. Paul
Re: catch 22?
> >manually created the symlink Jim, THANKS! I'll give that a try.best suggestion I've heard today :-) Paul > >- Jim
catch 22?
Oh Debian L'ers, I've managed to get myself in a catch-22 kind of dilemma. In an effort to get a working 2.0.0 kernel with the proper options to support IP masquerading, somehow or other both my kernel-image and kernel-source packages have gotten to a state where I'm stuck fast! I cannot successfully build a kernel at this point, and I can't remove or reinstall either of the packages. Attempting to reinstall results in errors during the prerem or postrem scripts for both the source and image package. Attempting to remove, errors with a recommendation to reinstall (which fails of course!--hence the dilemma) before attempting to remove! I am stuck in that proverbial hard place. Is there some way to force dpkg to reinstall (or remove) in spite of the error it encounters attempting to remove the older package first? Or is there some reasonable way to fake it out by creating the .deb files in some magic place where the remove will work, for example, or even by copying them from the CD to an appropriate place (I do have the I-Connect CD cut on 6_16)? I haven't been successful attempting to use dpkg -i either, but that could be me not getting the syntax correct perhaps. Any suggestions would be appreciated. Thanks. Stuck fast Paul
re: Questions
On 05:47:14 [EMAIL PROTECTED] wrote: > >>??? 'df' will show you what filesystems are mounted, not mount. > >Just type > >mount{Enter} > >without any parameters, and see what output you get. > >"Live and learn", eh? >>Sure! Typical of Linux there is always more to learn, and very often more >>than one way to do most anythingI use 'df' because it shows more info, >>and never have used just 'mount' but I'll try it. Well, I tried it, and it is quite different than using 'df' thats for sure. And it really just is reading the current /etc/mtab file. It is poorly formatted and kinda ugly. I'll continue using 'df' thank you. eh? Paul
RE: Questions
On 05:47:14 [EMAIL PROTECTED] wrote: And when I use "mount" command, it doesn't show CDROM is mounted. > >>??? 'df' will show you what filesystems are mounted, not mount. > >Just type > >mount{Enter} > >without any parameters, and see what output you get. > >"Live and learn", eh? Sure! Typical of Linux there is always more to learn, and very often more than one way to do most anythingI use 'df' because it shows more info, and never have used just 'mount' but I'll try it. eh? Paul
RE: Questions
Hung, >>Hi, >My CDROM is panasonic CR-562 CDROM. I select >the "sbpcd" module. Is that correct? I believe "sbpcd" is specifically for Sound Blaster and its clones. There may be several different vendors supplying the CD drive and Panasonic is only one of those. Do you have a SB or not? >boots, the busy light in CDROM flashes. But when I use >"dselect", it ask me the block device name. How do I answer it? /dev/sbpcd0 >And when I use "mount" command, it doesn't show CDROM is mounted. ??? 'df' will show you what filesystems are mounted, not mount. To mount the CD filesystem, type: mount /dev/sbpcd0 -t iso9660 /mnt or wherever you want it mounted. Paul