From: Digestifier To: Linux-Admin@senator-bedfellow.mit.edu Reply-To: Linux-Admin@senator-bedfellow.mit.edu Date: Sat, 24 Sep 94 23:14:07 EDT Subject: Linux-Admin Digest #102 Linux-Admin Digest #102, Volume #2 Sat, 24 Sep 94 23:14:07 EDT Contents: Re: PPP does not work (Bob Collie) Re: Linux Device Registration (H. Peter Anvin) Re: Need DL/Time Limiting ideas - Linux BBS (Patrick Killourhy) Re: Need DL/Time Limiting ideas - Linux BBS (Patrick Killourhy) Reinstallation (michael goshorn) Re: Tape Hassle (jon marcus madison) Re: Need DL/Time Limiting ideas - Linux BBS (Michael Dillon) Re: MultiTape Compressed Backups (Andreas Koppenhoefer) Re: What user interface to use??? (Jeff Kesselman) Re: Need DL/Time Limiting ideas - Linux BBS (Greg Corteville) Re: How to use a host as a router - READ THIS (Jay Ashworth) BusLogic EIDE (Roman Gollent) Re: DOMM 4 Linux /X is OUT !!! (Mark van Hoeij) Support for Diamond Stealth? (Alexander A Durzy) Re: HPFS Filesystem (Tim Cutts) HELP: makeing a SLIP connection (Khalil Foundy) Max MTU=1500? But NFS/Amd rsize=wsize=8192? (Shiu Wong) Re: [Q] ls -i gives 38857 inodes for empty dir (David Barr) Re: Has anyone gotten ftape to work? (Duncan Sutherland) ---------------------------------------------------------------------------- From: collieb@iia.org (Bob Collie) Subject: Re: PPP does not work Date: 22 Sep 1994 23:18:09 GMT ymlan@tstp1.seed.net.tw (Yue-mingLan) wrote: : Dear netters, : Could anybody out there can tell me how to make my PPP work. : I tried to use ppp-on script comes with Slackware 2.0 with : device name modified. But it still does not work at all. : I would like to learn more about pppd and chat commands. What does : the lock file mean ? : Any information welcome. : My e-mail address: ymlan@tpts1.seed.net.tw : Thanks in advance. : Lan Kindly cc: your answers to me, I am running the same version of Slackware. Bob Collie collieb@iia.org ------------------------------ From: hpa@ahab.eecs.nwu.edu (H. Peter Anvin) Subject: Re: Linux Device Registration Reply-To: hpa@nwu.edu (H. Peter Anvin) Date: Sat, 24 Sep 1994 21:06:54 GMT Followup to: <199409221150.HAA14999@cais.cais.com> By author: ericy@cais.cais.com In newsgroup: comp.os.linux.admin > > I spoke to Linus about this in Heidelberg, and the solution that > we came up with was to use dynamic allocation of major numbers. The only > tricky part is that you somehow need a way of refreshing the /dev/ entries > whenever you load the module. > > Perhaps a better solution would be for the iBCS2 stuff to have a > preferred major number of 30, and if it cannot get it, then it tries > to dynamicly allocate one. This way most of the time it will have the > same major. Even then, some simple script should be used after the module > is loaded to make sure that the /dev entries are correct. > > The iBCS2 emulator can already use dynamicly allocated majors, as > I recall, and the kernel support is already there. Thus for iBCS2, > the changes would be quite trivial. I am not sure about the sound driver, > however. > The problem is that of the /dev directory, and I for one don't want a script mucking with my carefully configured /dev directory, especially not one which might delete devices just because I boot a minimal kernel. Admittedly, the 16-bit device space inherited from Minix is a major problem. It would be relatively easy to fix on the filesystem side, but since dev_t is unsigned short it would mean a lot of programs would have to be recompiled, or there would have to be a hideously complicated compatibility library. Maybe something for libc-5 when it is a matter of upgrading binaries or keeping libc-4 around anyway. However, the major problem here is that the person who had promised to keep track of device number allocations disappeared off the face of the Earth. If it would help, I would volunteer to pick up the Device Number Registrar's position and allocate device major numbers (and possibly minor numbers as well for things like busmice). I have the resources to keep an updated list available for WWW and FTP, and even run a mailing list if so is desirable. /hpa -- INTERNET: hpa@nwu.edu --- Allah'u'abha --- IBM MAIL: I0050052 at IBMMAIL HAM RADIO: N9ITP or SM4TKN FIDONET: 1:115/511 or 1:115/512 STORMNET: 181:294/1 or 181:294/101 Windows 4.0 (a.k.a. Chicago): OS/2 with a Microsoft logo + 2 years late ------------------------------ From: killourh@lvl-sun684.usc.edu (Patrick Killourhy) Crossposted-To: alt.bbs,alt.bbs.unixbbs Subject: Re: Need DL/Time Limiting ideas - Linux BBS Date: 22 Sep 1994 23:26:19 -0700 In article , pizzi@nervous.com (Riccardo Pizzi) writes: |> >Why don't you just modify Uniboard so that it spawns a daemon that does |> >the online-time checking concurrently with the sz/sb/sx process? People |> >go out and write things for Unix, and never take full advantage of the |> >fact that it's a completely multi-user, mutli-tasking OS. |> |> Basically, I didn't like the idea of a user file transfer aborted |> in the middle because the time-for-call expired. This is why it works this way. |> A UniBoard sysop here in Italy runs a BBS for profit over a `144' line |> (I think it is 1-900 for you US people: a line which charges the user |> some money per minute, and some of the $$$ go into the sysop's pockets). |> He modified his zmodem to do exactly what Patrick suggested: the caller |> is brute-force disconnected when the timer for the call expires. He did |> this by forking a daemon from within zmodem (usual rz/sz by Omen Technology). |> Reason? The telco wouldn't pay to the sysop anything for calls over 30' |> in lenght :-) OK, granted, it's a tad rude, but here's two responses to that: 1) Most modern terminal programs have zmodem, and most people use it. Zmodem's transfer resumption takes care of this problem. Still not a great solution though, as you pointed out, so here's a better one: 2) record the time that the FIRST file transmission was begun. when the first file is completely transferred (at the point where you are currently checking the time limit) calculate the average throughput of the connection and use that to estimate the estimated time for transmission of the rest of the files. This could also be done by dumping (for instance) a 1000 CR characters to the user's terminal, and calculating the average cps from the time it takes for that. the more characters, obviously, the more reliable the estimate, but the longer it takes to get the estimate. This can also be readjusted after each file as per the above. An even better method would be to calculate the average throughput by measuring the time it takes to transmit a system banner or motd (or whatever). That way the user is getting useful text instead of a blinking cursor.. ------------------------------ From: killourh@lvl-sun683.usc.edu (Patrick Killourhy) Crossposted-To: alt.bbs,alt.bbs.unixbbs Subject: Re: Need DL/Time Limiting ideas - Linux BBS Date: 21 Sep 1994 21:20:47 -0700 In article , pizzi@nervous.com (Riccardo Pizzi) writes: |> In article <35l52k$3ju@xanax.apana.org.au> trevor@xanax.apana.org.au (Trevor Lampre) writes: |> |> >A BBS needs to know what speed the user connects at to estimate |> >if they have enough time to finish downloading a file. This requires |> >knowing the connect string. A standard Unix getty doesn't see this or |> >pass it onto the BBS. As a result Uniboard will let users download for |> >hours beyond their time limit. |> |> This is only partially true. UniBoard _does_ check for time expiration |> after each downloaded file (even during batch download) and will disconnect |> the user if his time limit is overridden. |> Why don't you just modify Uniboard so that it spawns a daemon that does the online-time checking concurrently with the sz/sb/sx process? People go out and write things for Unix, and never take full advantage of the fact that it's a completely multi-user, mutli-tasking OS. ------------------------------ From: goshorn@mozart.cs.colostate.edu (michael goshorn) Subject: Reinstallation Date: 24 Sep 1994 21:20:15 GMT Nate, I don't have the answers you're looking for but I would be interested in answers to your questions as well. In this area, you might find the following helpful. If anyone has better ideas, please advise. The slackware "installpkg" command has a "-warn" option that will show you what would happen if you were to install a given package on your existing system. I don't believe there is anything similar for disk sets as a whole; you'll have to check each package individually and then try to figure out how the installations affect things together by hand. M Goshorn goshorn@cs.colostate.edu ------------------------------ Subject: Re: Tape Hassle From: icqo409@iupui.edu (jon marcus madison) Date: 18 Sep 94 16:32:24 -0500 In article , Dennis Duffner wrote: >I've been trying to use the new jumbo120 tape drive to backup my system. > >Unfortunately, I don't know what's happened but I can't insmod drv_hello.o >because the .o file isn't anywhere! :sigh: uh, drv_hello.o is nothing but a test driver, so it's no use annyway... >Anyone get this working? I desperately need to back this thing up. yup. you need to get ftape, & depending on your kernel, search for the correct patches. linuxftp.caltech.iupui.edu:/pub/Linux/sunsite-Incoming/ftape* "punks be wallowin' in weed for tha know how, but we be swallowing these in a showdown..." -lpg, Real hip hop -- jon madison oit consultant in training ------------------------------ From: mpdillon@halcyon.com (Michael Dillon) Crossposted-To: alt.bbs,alt.bbs.unixbbs Subject: Re: Need DL/Time Limiting ideas - Linux BBS Date: Wed, 21 Sep 1994 10:00:51 +0100 > >>basically, you need no "BBS Software" for a Linux machine. just setup > >>the users like yourself, turn on the modem and away you go. > > > >>well, er, there are a few other things (of course), but thats about it. > > > >Umm... Sure, but not every BBS user even has a clue about UNIX. They'd > >be 100% lost. > > This is so. I used to run an OS/2 based BBS but switched to Linux/Uniboard > to give my users access to the Net. Before I switched I did a mail-out to > see what interest there was. About 30% of my users responded, of those only > about 50% said they would like a Unix shell account. OF the shell accounts > I created only about 50% of those ever bothered to log in. > > Switching from the more friendly DOS type BBS's to a Unix one can be very > risky. I haven't had a new user in months, before I switched things were > ticking over quite nicely. > > Many people can't even log in. I've changed the login message so many times > it's not funny, people still can't seem to get the idea of logging in as 'bbs' > before they get to the bbs. Even giving out my home phone number didn't do > the trick. Well, since you are running Linux, just change your getty program so that it runs the bbs program directly rather than running /bin/login. > >My intention is to create a system that combines the positives of both > >operating systems. MS-DOS's ease of use with Linux's power and > >flexibility. Things don't have to be complicated for the user to make it > >powerful. > > Getting a good Unix BBS has been the bane of my existence. Uniboard isn't > bad but it still isn't as good as the DOS based ones. Unix doesn't help > either. A BBS needs to know what speed the user connects at to estimate > if they have enough time to finish downloading a file. This requires > knowing the connect string. A standard Unix getty doesn't see this or > pass it onto the BBS. As a result Uniboard will let users download for > hours beyond their time limit. Get the source code for getty_ps and add this feature into it. Just get it to dump the connect string into a file and have a separat program deal with the time limit problem. cruisin' down the information highway, lookin' for a blast breakin' all the speed limits as I come zoomin' past! -- Michael Dillon Internet: mpdillon@halcyon.halcyon.com C-4 Powerhouse Fidonet: 1:353/350 RR #2 Armstrong, BC V0E 1B0 Voice: +1-604-546-8022 Canada BBS: +1-604-546-2705 ------------------------------ From: koppenas@tick.informatik.uni-stuttgart.de (Andreas Koppenhoefer) Subject: Re: MultiTape Compressed Backups Date: Tue, 20 Sep 1994 08:21:25 GMT In article tom@metronet.com (Tom Griffing) writes: Jason Sokolosky wrote: >Is there ANY WAY to do a MultiTape Compressed Backup???????? Use GNU tar and add the flags: [ -L, --tape-length N ] [ -M, --multi-volume ] [ -Z, --compress ] [ -z, --gzip, ] Hey Tom, that's not enough! % tar -L 1440k -Mzcf /dev/fd0 /home tar: cannot use multi-volume compressed archives % Andreas -- Andreas Koppenhoefer, Student der Universitaet Stuttgart, BR Deutschland prefered languages: German, English, C, perl ("Just another Perl hacker,") SMTP: koppenh@trick.informatik.uni-stuttgart.de privat: Belaustr. 5/3, D-70195 Stuttgart, Germany, Earth, Sector ZZ9 plural Z alpha phone: +49 711 696378 and +49 711 694111 (19-22h MEZ=GMT+1) ------------------------------ Crossposted-To: comp.os.linux.development From: jeffpk@netcom.com (Jeff Kesselman) Subject: Re: What user interface to use??? Date: Sat, 24 Sep 1994 21:50:28 GMT In article , Tony Schwartz wrote: >When do dial into your local ISP using a standard terminal connection, what >software is used to providet the menuing, ability to do internet functions >like telnet, ftp, gopher, etc??? telnet provides telnet. ftp provides ftp gopher provides gopher etc. there is no menuing as such, they axre all command-line driven. All these UNIX utilities are designed to be run from the command line. The menuing systems you might be familiar with are all front ends that call these programs. Oh, and manual pages are veiwed byt typing man. (Type 'man man' for an explaination of how to use the online manual.) Welcome to the real world of UNIX. ------------------------------ From: gcortevi@nyx10.cs.du.edu (Greg Corteville) Crossposted-To: alt.bbs,alt.bbs.unixbbs Subject: Re: Need DL/Time Limiting ideas - Linux BBS Date: 24 Sep 1994 14:29:06 -0600 Since I got so many people who run BBSs into the thread, how about answering this question for me. In setting up the download protocols on my board, they work fine if I download over the modem from another UNIX machine. However, if I try to download to a DOS machine, the transfer doesn't even start correctly. The filename is in 8.3 format. What are the command line parameters I need to give on the Linux side? -- Greg Corteville | "If privacy is outlawed, only gcortevi@nyx10.cs.du.edu | outlaws will have privacy." gcortevi@trident.lbs.msu.edu | - PGP Documentation ---[PGP Key available via finger]------------------------------------ ------------------------------ From: jra@zeus.IntNet.net (Jay Ashworth) Crossposted-To: comp.os.linux.misc,comp.os.linux.development Subject: Re: How to use a host as a router - READ THIS Date: 22 Sep 1994 21:28:49 -0400 ianm@qualcomm.com (Ian McCloghrie) writes: >This is common practice (and, in fact, required by many TCP/IP protocl >stacks). Whether or not it is "correct" is unclear. It's quite >possible to implement routing using the same IP address on two >interfaces, if one of them is a point-to-point link (namely, >a slip line). The idea of every physical network having its own >IP network is ideologically pure. Ideological purity, while clean >and elegant, is often discarded in favour of optimizations. Given >the current state of the IP address space, it could easily be argued >that wasting an entire network on a 2-host point-to-point slip line >is incorrect behaviour :) True. But you'll note I didn't say anything about where those 2 addresses need to reside. Common sense would seem to suggest putting your "router's" PPP port on your host's net, and it's ether on your own, and in fact, this works. At worst, external incoming connections will get aimed at your ether IP number, but you don'e lost a _whole_ there... Cheers, -- jra -- Jay R. Ashworth Ashworth Designer High Technology Systems Consulting & Associates ka1fjx/4 jra@baylink.com Linux: The Choice of a GNU Generation +1 813 790 7592 ------------------------------ From: rgollent@force.stwing.upenn.edu (Roman Gollent) Subject: BusLogic EIDE Date: 24 Sep 1994 19:32:19 GMT I was wondering if anyone has a BusLogic EIDE controller and wether or not it is compatible with Linux. The thing that worries me the most about this controller is that it has its own BIOS. The version I am looking at comes with a 512k cache and 4 HD / 4 FLOPPY support and is compatible with VESA Local Bus. Regards, Roman ------------------------------ From: hoeij@sci.kun.nl (Mark van Hoeij) Subject: Re: DOMM 4 Linux /X is OUT !!! Date: Sun, 25 Sep 1994 00:36:23 GMT In <1994Sep12.122248.12528@taylor.infi.net> mark@taylor.infi.net (Mark A. Davis) writes: >stephenb@scribendum.win-uk.net (Stephen Benson) writes: >OK- I have both MS-"DOS" and Linux DOOM. My machine is a 486 33 DX with >16 MB of 0 wait X 32 RAM. Adaptec 16 bit SCSI controller, Orchid >Pro Designer II ET-4000 16bit SVGA card, Sound Blaster Original, etc. >The game absolutely flies under MS-"DOS", I can't even tell how many >frames per second, but at least 6. >The game is completely unplayable under Linux, with a frame rate of 0.5 >per second (the is the fastest in 640x480 X, in 1024x768 it is slower). I get about 4 to 10 (depending on where I walk in the game) frames/sec on a 386-40 with an S3 card (with ISA bus). The bottle neck is your graphics card. Mark van Hoeij ------------------------------ From: durzya@titan.ucs.umass.edu (Alexander A Durzy) Subject: Support for Diamond Stealth? Date: 21 Sep 1994 22:23:52 GMT Does anyone know if the current version of Linux supports the S3 chipset in the Diamond Stealth video cards?? ------------------------------ From: tjrc1@cus.cam.ac.uk (Tim Cutts) Subject: Re: HPFS Filesystem Date: 23 Sep 1994 07:51:05 GMT hpernu@delta.hut.fi (Heikki Johannes Pernu) writes: > Anyone working on the HPFS filesystem? A friend of mine was considering it a few months ago, but I've heard nothing since. > Is anyone trying to make HPFS read/write? I'm not the only > one interested. Said friend, but I don't know whether he was really serious. There is going to be a problem, too, with OS/2 3.0. As anyone who's already tried the beta will already know, the current read-only hpfs filesystem does not work with Warp's HPFS. > Also I seem to get strange error messages about missing > files of pattern 'eadata.*' even with ls... Those are the extended attributes (in other words, your settings noterbook for the file). I think they're stored in an odd way on DOS FAT filesystems, which is where you see this error. However, I seem to remember seeing somewhere an announcement of a msdos fs update that fixed this. Am I making this up, someone? > I might begin to work with the project myself, but I'm > not really a kernel hacker and know nothing about the > internals of any filesystem. I might do some testing though. > Thanks in advance ( I hope :-) Sorry I can't really help, but I think it would be best to wait for Warp's official release next month before anyone really commits themselves to such a project. Tim. -- =============================================================================== 194, Vinery Rd, Cambridge, CB1 3DS, UK (+44) 223 572622 http://mole.bio.cam.ac.uk/~tjrc1/ =============================================================================== ------------------------------ From: kronos@CAM.ORG (Khalil Foundy) Subject: HELP: makeing a SLIP connection Date: 24 Sep 1994 01:50:47 -0400 I am trying to slip connection manualy. I dialup my provider using cu, when the connection is established I suspend cu and then I execute slattach in order to turn my serial connection into CSLIP mode. But slattach give me the followin error message: SLIP_set_disc: Invalide argument tty_get_name: Invalide argument cslip started I tryed dip and get the first error message. I am using Linux 1.1.45 Any hit will bw appreciated. Thanks in advance. Khalil Foundy. P.S. would you please respond by email. ------------------------------ From: shiu@cae.wisc.edu (Shiu Wong) Crossposted-To: comp.os.linux.help Subject: Max MTU=1500? But NFS/Amd rsize=wsize=8192? Date: 24 Sep 1994 20:00:13 GMT Hi all, We have several Linux boxes running Linux 1.1.18, which use our Sun SPARCstations as the NIS/NFS servers. We were trying to change the NFS packet read/write size to 8192 bytes (as SunOS does). We read the fine Linux Network Admin book and it seems to us that the UDP packet size is limited to 3500 bytes and hence the NFS read/write size is limited to 3500 bytes. Also we were trying to increase the read/write size in the automounter to 8192 bytes too. Even if we change it, will it work? Moreover, when we do a `netstat -i' the MTU is by default 1500 bytes. If the size is only 1500 bytes, how can NFS/Automounter can communicate with 8192 bytes? Is it possible that with kernel 1.1.18 we can increase the MTU size to 8192 with `ifconfig' or it doesn't matter because NFS will overwrite the default MTU size? We are wondering if there are any public domain packages that allow us to monitor the packet size when our Linux boxes are communicating with the Sun servers? Anyone has any experience? Any help is highly appreciated. - Shiu K. Wong shiu@cae.wisc.edu ------------------------------ From: barr@pop.psu.edu (David Barr) Subject: Re: [Q] ls -i gives 38857 inodes for empty dir Date: 24 Sep 1994 05:46:32 -0400 In article <459@scribendum.win-uk.net>, Stephen Benson wrote: >I've done a bit of reading about directories/files/inodes, but I'm not >completely clear on it. I've read about inodes not being properly deallocated, >which seems to be a black hole for resources: anyway I have an empty directory >/temp with an inode count of 38857. Does it matter? No. --Dave ------------------------------ From: dasu@mercury.sfsu.edu (Duncan Sutherland) Crossposted-To: utah.linux Subject: Re: Has anyone gotten ftape to work? Date: 23 Sep 1994 15:43:49 GMT : : tape drives (SCSI or proprietary bus, since I don't have IDE in this : : machine) that are pretty affordable (under $500) that are close to : : 1GB? : If you want backup device of that size, you will need to raise your : $$$$ a bit. Right now the QIC-80 devices are about the only thing : around for less than $750 US or so. Also, I dont think you can split : a backup across multiple tapes with the current ftape, you need to make : seperate tar files on different tapes. Kind of a pain but better than : no backup at all. Actually I saw a 500 mb tape drive selling for about $200 at a recent computer show in these parts. I have no information about this, but it might be worth investigating. I believe it was by Colorado, but I could be wrong - Hope this helps - Duncan Sutherland dasu@sfsu.edu ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Admin-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.admin) via: Internet: Linux-Admin@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Admin Digest ******************************