From: Digestifier To: Linux-Admin@senator-bedfellow.mit.edu Reply-To: Linux-Admin@senator-bedfellow.mit.edu Date: Tue, 18 Oct 94 02:13:29 EDT Subject: Linux-Admin Digest #211 Linux-Admin Digest #211, Volume #2 Tue, 18 Oct 94 02:13:29 EDT Contents: SCSI: T130B card (Mark Spencer) getty problem (Jim Carroll) Re: SCSI vs IDE (Matthew Dillon) Re: Help with mounting cdrom & floppy drive (Miles Lott) Re: Slackware color xterm (Edward Doolittle) Re: shadow-332: -f bug present (Gert Doering) Re: ftape driver for 1.1.49 ? (Another Totoro) Re: Extreme delays telnetting into linux box (David - Morris) Running out of pty's, how do i add more? (Daniel Garcia) Re: New Motif lib's for use with XFree 3.1 ? (Adam J. Richter) Re: more LEDs? (phillyboy) Re: Whats wrong with sunsite.unc.edu? (J V Sneeringer) Re: SCSI vs IDE (Orc) Re: more LEDs? (H. Peter Anvin) ---------------------------------------------------------------------------- From: myspence@mit.edu (Mark Spencer) Subject: SCSI: T130B card Date: 17 Oct 1994 23:16:48 GMT Reply-To: mspencer@eng.auburn.edu I've got a Trantor 130B SCSI card in my computer at school. As far as I know, there's no support from Linux. Anybody got a patch? I used the T128, but as the source says, it doesn't work since the architecture is different. -Mark Spencer ------------------------------ From: jimc@e-Commerce.Com (Jim Carroll) Crossposted-To: comp.os.linux.help Subject: getty problem Date: 17 Oct 1994 23:31:36 GMT My goal is to set up a pppd server, patiently waiting for a remote ppp process to connect. However, I'm trying to get getty to work properly, and it ain't happening. To me, this shouldn't be a difficult problem, so either I've overlooked something, or I've got a piece of broken software. The Gory Details: - Linux 1.1.47 - ppp-2.1.2a - getty_ps-2.0.7e I'm testing the connection from another Unix box (SPARC, running SunOS 4.1.3_U1) using direct cabling. One cable is a null modem cable, the other is a straight-through cable. I've succeeded in verifying that they can talk. That is, I can cu from the Sun, and while connected, I can run 'date >/dev/ttyS0' on the Linux host, and it displays properly. (Well, not entirely. It 'barber-poles'. But at least the speed is right.) But I *never* get a login prompt. If I hit newline a couple times, I get a segmentation fault from getty listed in the logfile. And this was *after* I created the file /etc/getty/getty/getty.ttyS0 with some parameters (see below). Before I set this file up, getty was segfaulting and respawning every couple seconds (according to syslog), finally disabling it: "respawning too fast: disabled for 5 minutes". This is my getty.ttyS0 file: DELAY=1 WAITCHAR=YES Just for fun, I added the following line, and the desired string actually displays just prior to giving me a segfault: CONNECT=\r \shello\sthere!\n\r On a lark, I've tried setting it to use /dev/cua0, but it's no better. So, what am I up against? 1. User Brain Damage. 2. broken software 3. bad wiring 4. some/all of the above Any/all helpful suggestions will be warmly received. Request responses via email, as I regrettably rarely have the luxury of reading news. :-( TIA. -- Jim Carroll -- jimc@e-Commerce.Com e-Commerce, Inc., 1030 Kamato Road, Suite 201 Mississauga, Ontario, Canada L4W 4B6 Tel: +1 905 602 0863 Fax: +1 905 602 8402 ------------------------------ From: dillon@apollo.west.oic.com (Matthew Dillon) Subject: Re: SCSI vs IDE Date: 17 Oct 1994 11:16:40 -0700 :In article drierp@mail1.its.rpi.edu (Peter Drier) writes: :>>Even with SCSI you often have two scsi controllers as CD-ROMS often tie the :>>bus :>>for ages with their low transfer rate. I'm in the middle of finishing :>>sorting out our computer society system once we get a bigger case - then it :>>will be 2 IDE controllers, 3 IDE drives, 2 SCSI drives on a SCSI controller, :>>4 serial ports, 3 parallel ports, ethernet and floppy on one ISA bus. :> :>>I suspect bus contention becoming a performance issue 8) :> :>This is where bus mastering comes into play. The buslogic 445s for example :>buffers requests so a slow drive will not slow down the system. :> :>Peter Drier It isn't so much DMA as it is reselection that allows slower SCSI devices to coexist with faster ones. While the overall transfer rate of a CDRom is slow, it still packetizes the data so each 'sector' is collected and then transfered at full speed. Data from other SCSI devices can be transfered inbetween the data coming from the CDRom. At least theoretically. What bus mastering / DMA does is free up the cpu to do other things while a transfer is going on, rather then have to sit in a loop. -Matt -- Matthew Dillon dillon@apollo.west.oic.com 1005 Apollo Way ham: KC6LVW (no mail drop) Incline Village, NV. 89451 Obvious Implementations Corporation USA Sandel-Avery Engineering [always include a portion of the original email in any response!] ------------------------------ From: mlott@starbase.neosoft.com (Miles Lott) Crossposted-To: alt.uu.comp.os.linux.questions,comp.os.linux.help Subject: Re: Help with mounting cdrom & floppy drive Date: 17 Oct 1994 18:25:27 GMT Narayan Moni (anish@unix1.sncc.lsu.edu) wrote: : Another thing is, the way the A drive is mounted seems to be giving : trouble. When we put in a floppy and check its contents by entering : "ls -la" it gives us the contents . But if we change the floppy the : the listing does not change. So it always gives us the listing of : the floppy which is first inserted into the drive. You might try remount. : -- : Anish Moni : Off : 210, System Network Computer Center, Louisiana State University, : Baton Rouge, Louisiana - 70808. : Tel : (504) 388-1928 (off) Email : anish@unix1.sncc.lsu.edu -- ============+==============+==============+============ Miles Lott, A.A.S. E.E.T. Pro-Video Service, Houston, TX SBE Member - Broadcast TV Certification ============+==============+==============+============ ------------------------------ From: dolittle@math.toronto.edu (Edward Doolittle) Subject: Re: Slackware color xterm Date: Mon, 17 Oct 94 16:42:13 GMT >Daniel M. Coleman wrote: > I recently downloaded the slackware distribution, and I noticed that > some of the default settings on the color_xterm are not quite right. > Whenever I press the backspace key, instead of backspacing, it acts > as if I had pressed the enter key. The normal (noncolor) xterm works > fine. What could be wrong? In article <37r5ba$mmn@Tut.MsState.Edu>, Stormy@Grand.Mother.Com writes: >I've had this problem too, except it's not acting like I've pressed enter, >because it ignores my input, more like just a newline and carriage return >while clearing the input buffer. Ummm, sounds like an intr (usually ^c) ... try running a program and pressing backspace; it should behave as if you "pressed ^c". Enter the command "stty -a" to see the status of the special characters. >Anyway, I noticed that the ONLY time this happend was when I logged out of >the shell calling openwin BEFORE the xterm was started, as in exec openwin or >simply typing openwin & and then hitting control-d to logout. The defaults for the special characters are in /usr/include/linux/tty.h. They may be changed by getty, login (see /etc/login.defs), or by stty or tset commands in /etc/profile, in $HOME/.profile (or $HOME/.bash_profile etc. etc.), in $HOME/.bashrc (that is a bad idea, though). In particular, mgetty sets intr=^?. I guess that if you log out of the shell calling openwin, there is no controlling tty, and the terminal emulators revert to their defaults, which for some strange reason may be different between xterm and color_xterm (someone please correct me on all this...). You might try grepping the color_xterm source for "erase" and "intr" to see what it thinks the "defaults" are, if any. Also, look through ..../lib/X11/app-defaults/color_xterm or whatever. >I never found a solution, cept for waiting till X was done booting >before I exited my console shell. In the case of xterm, you can add the option -tm "intr ^c erase ^?" to the command line to change the intr and erase characters to those values. You can also set the resource ttyModes. RTFM xterm for more information. Followups redirected to comp.os.linux.help. Ed ================================================================== "There are no differences but | dolittle@math.toronto.edu differences of degree between degrees | dolittle@math.utoronto.ca of difference and no difference." | --William James | Ed Doolittle ------------------------------ From: gert@greenie.muc.de (Gert Doering) Subject: Re: shadow-332: -f bug present Date: Sun, 16 Oct 1994 00:16:35 GMT hm@ix.de (Harald Milz) writes: >In comp.os.linux.admin, Harald Milz (hm@ix.de) wrote: >> I installed the shadow-3.3.2 suite a couple of minute ago and found >> that the old -f bug is again in there. What's wrong? >I hate to followup myself, but I found that getty_ps-2.0.7e solves >the problem by denying login names starting with a '-'. Same for mgetty. Anyway, the proper way would be to fix the shadow suite instead of having to work around in every getty program. gert -- Yield to temptation ... it may not pass your way again! -- Lazarus Long //www.muc.de/~gert Gert Doering - Munich, Germany gert@greenie.muc.de fax: +49-89-3243328 gert.doering@physik.tu-muenchen.de ------------------------------ From: kkfong@netcom.com (Another Totoro) Subject: Re: ftape driver for 1.1.49 ? Date: Mon, 17 Oct 1994 01:20:55 GMT Yes, ftape is in another package, and you also need yet another package (modutil) to load the ftape. They are all in sunsite under /pub/Linux/kernel/tapes. Since you have a pretty new kernel, you should grab that dma-irq patch too. With you grab all those packages, you will have yet another hell of a time trying to figure out how does the darn thing work. root@nimir.demon.co.uk wrote: : I have compiled 1.1.49 - make config asked if I wanted ftape : (I chose yes) but I do not seem to have a /dev/ftape or : for that matter any ftape source code. : Is ftape now a module - if so where is it, and why does : make config still ask about it ??? : Steve Hunt hunt@nimir.demon.co.uk ------------------------------ From: dwm@shell.portal.com (David - Morris) Crossposted-To: comp.os.linux.development Subject: Re: Extreme delays telnetting into linux box Date: 15 Oct 1994 04:44:53 GMT I'll say it again. When I telnet from a LAN workstation whose IP address can't be resolved into a name I experience a VERY long delay. When I add the host to the /etc/hosts, the connection is almost instant.... If this doesn't help you, it would be helpful to know. Dave Morris ------------------------------ From: kender@leviathan.ccnet.com (Daniel Garcia) Subject: Running out of pty's, how do i add more? Date: 17 Oct 1994 20:11:47 GMT Reply-To: kender@leviathan.ccnet.com Hello alls, I've been a happy linux user/sysadmin for almost two years now, even happier now that my machines have access to the net (thank god for PPP and low cost service providers :), however, now that my machine is supporting more than two simultaneous users (more like 4-5 at any given time, i know, still not a lot, but read on), most of whom are running screen (in addition to me running X on the console), i'm starting to run out of ptys! I've used MAKEDEV to try to make more, but that doesn't seem to have quite worked. Any ideas on what I should do? I'm running slackware 1.0, but my kernel has been up'd to 1.1.50. Thanks in advance! D -- ___________________________________________________________________________ /Daniel Garcia/Soon to be PhD Student/dgarcia@cohl.llnl.gov/kender@esu.edu / /Linux Hacker/C Programmer for Hire /#include /The Answer's 42/| ,-------------+----------------------+---------------------+-------------- + | |I wake up scared, I wake up strange, I wake up wondering if anything in my| | | life is ever gunna change, I wake up scared, I wake up strange, and | / | everything around me stays the same... |/ `--------------------------------------------------------------------------' ------------------------------ From: adam@adam (Adam J. Richter) Crossposted-To: comp.windows.x.i386unix,comp.windows.x.motif Subject: Re: New Motif lib's for use with XFree 3.1 ? Date: 17 Oct 1994 19:42:30 GMT In article , Peter Mutsaers wrote: >To Adam Richter: nothing stops you to recompile XFree 3.1 yourself and >try to keep the version number [...] I know we can do this, but we want a concensus within the Linux community to go one way or the other. It would be better to switch to XFree86's incompatable libraries than have half of the Linux community using our R5-compatable binding's and half using XFree86's R5-incompatable bindings. I agree that if we're going to make a change, it might be better to go with ELF a month or two early than to switch twice. -- Adam J. Richter Yggdrasil Computing, Incorporated (408) 261-6630 "Free Software For The Rest of Us." ------------------------------ From: rkoffler@ux4.cso.uiuc.edu (phillyboy) Subject: Re: more LEDs? Date: 15 Oct 1994 05:06:23 GMT wcreator@kaiwan.com (Steven M. Doyle) writes: >In <37e6si$v0@eddy.frmug.fr.net> zarkdav@eddy.frmug.fr.net (Benjamin Ryzman) writes: >>>>>>> Au sujet de "Re: SCSI vs IDE", Alan Cox écrivait récemment: >>AC> In article <36t84s$pb0@dhp.com> panzer@dhp.com (Panzer Boy) >>AC> writes: >>>> Umm, I have three LED's on my computer, plus a panel for stupid >>>> speed. I have 3 drive controllers hooked up to the "HD, Turbo, >>>> and Power" LED's. I don't anyone who uses turbo for more than >>>> minor kicks, and if you can't tell you're computer is on, maybe >>>> you need your head examined. Of course, if you only have 1 led, >>>> this could be a problem. >>AC> Definitely we need more LEDs. I've got the turbo wired to the >>AC> carrier detect on the amateur radio PI2 card 8) >>Yep! >>In fact, we'd need a user programmable LCD (2x40?) screen for: >>-disk status >>-serial port status >>-system load >>-custom peripheral status (like your amateur radio...) If anyone is interested, Dell used to (still does?) make computers with a four character display. I have a 486D/33 with one. Dell included a small program for DOS that allows is to either display a clock, or it can scroll a message specified on the command line. Without the program, I think it shows the sector and the drive being accessed. Since it is a DOS program, my display just shows the time of day when I booted Linux using loadlin. Rob -- ****************************************************************** |Couldn't roll me a seven philly@uiuc.edu | |if you gave me loaded dice.--Doug Supernaw | ****************************************************************** ------------------------------ From: jsneerin@pinot.callamer.com (J V Sneeringer) Subject: Re: Whats wrong with sunsite.unc.edu? Date: 15 Oct 1994 20:09:29 GMT Stephen Parkinson (stephen@zmemw16.demon.co.uk) wrote: : In article : mcclung@owens.ridgecrest.ca.us (Scott McClung) writes: : > : > >>Second: Does anyone know a mirror for sunsite, preferably close to : > >>California? : > I'd like to know that as well. Anyone know any mirror sites closer : > (network-wise) to the west coast of North America? : : how about 'linuxftp.caltech.edu' There's also ftp.cdrom.com (SF Bay area). Look in /.4/linux. Not quite as extensive as sunsite or mit, but they do have the whole Slackware distribution, alternate/new kernels, etc. ============================================================================== James Sneeringer CrowMan @ EFnet jsneerin@nike.calpoly.edu crowman@netcom.com jsneerin@slonet.org ------------------------------ From: orc@pell.com (Orc) Subject: Re: SCSI vs IDE Date: Thu, 13 Oct 1994 07:13:15 GMT In article , Alan Cox wrote: >Even with SCSI you often have two scsi controllers as CD-ROMS often tie the bus >for ages with their low transfer rate. I'm in the middle of finishing >sorting out our computer society system once we get a bigger case - then it >will be 2 IDE controllers, 3 IDE drives, 2 SCSI drives on a SCSI controller, >4 serial ports, 3 parallel ports, ethernet and floppy on one ISA bus. > >I suspect bus contention becoming a performance issue 8) To say nothing of timeout problems. I've got a two-controller setup on my system, and I've had recurrent problems with the aha1542 driver locking up my system and dying horribly because when it gets a reset interrupt from one of the cards, it then has to go out and reset every gawdforsaken device it can find. About halfway through, the interrupt handler gets swapped out and the scheduler freaks, converting a perfectly usable 486 box into an inefficient space heater. ____ david parsons \bi/ Of course I've commented that code out now. \/ ------------------------------ From: hpa@ahab.eecs.nwu.edu (H. Peter Anvin) Subject: Re: more LEDs? Reply-To: hpa@nwu.edu (H. Peter Anvin) Date: Mon, 17 Oct 1994 01:16:35 GMT Followup to: <37rugg$kqm@kisa.seanet.com> By author: blane@seanet.com (Brian Lane) In newsgroup: comp.os.linux.admin > > I don't think we need to do a device for it. Just a simple daemon to > wake up every xx seconds, read info to display, write to LCD, adn go back > to sleep should work. I have a 16x2 LCD that I wired to my parallel port > last night working okay. I'm working on the daemon to display > loadaverages, users and memory usage. > > It's also possible that I may be able to connect it so that the printer > can be used as well(LCD code would have to make sure the printer was not > in use before writing to the LCD). > Another thing that may be considered is that many cases sold today have a 2 or 2 1/2-digit LED display, usually jumpered to show the CPU speed. If it can be removed from the jumper setup, it would be quite useful for showing the load average (0-20, with one decimal place). Seeing the system load is definitely something useful, rather than the CPU speed, which doesn't change (surprise!). The easiest way to feed it would probably be through a parallel port, using latches hooked to the status pins to load a digit at a time, alternatively using decoder circuitry to pack 2 1/2 digits into 8 data pins + 3 status pins (1/2 digit, 2 decimal points), but thet way it would not be possible to use it for anything but digits. It probably would be necessary to dedicate a parallel port, but that shouldn't be that big of a deal given what I/O cards cost these days. /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 First hug free; all subsequent ones free. ------------------------------ ** 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 ******************************