Received: from PACIFIC-CARRIER-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA19252; Tue, 12 Mar 96 18:19:28 EST Received: from SENATOR-BEDFELLOW.MIT.EDU by MIT.EDU with SMTP id AA27540; Tue, 12 Mar 96 18:18:43 EST Received: (from root@localhost) by senator-bedfellow.MIT.EDU (8.6.13/2.3JIK) id SAA06977 for Linux-Development-System-Dist@senator-bedfellow.mit.edu; Tue, 12 Mar 1996 18:13:51 -0500 Message-Id: <199603122313.SAA06977@senator-bedfellow.MIT.EDU> From: Digestifier To: Linux-Development-System@senator-bedfellow.MIT.EDU Reply-To: Linux-Development-System@senator-bedfellow.MIT.EDU Delivered-To: Linux-Development-System@senator-bedfellow.mit.edu Precedence: bulk Date: Tue, 12 Mar 96 18:13:45 EST Subject: Linux-Development-System Digest #498 Linux-Development-System Digest #498, Volume #2 Tue, 12 Mar 96 18:13:45 EST Contents: Shutdown problem (John Mellor) Re: Netscape broken under 1.3.70 (John Boyer) Re: Emacs 19.30 "M-x manual-entry" output is truncated in 1.3.71 (David Fox) Re: Logical Volumes in Linux someday ? (David Thomas Richard Given) Re: need secure OS to entrust millions to (Jim Nance) Re: Logical Volumes in Linux someday ? (David E. Hollingsworth) 1.3.7[01]: cannot open /dev/modem: no such device (sl14@crux1.cit.cornell.edu) Logical Volumes in Linux someday ? (Remko Stuik) Re: Chicken-or-the-egg-game with microkernel and FS drivers (Stanislav Meduna) Re: Logical Volumes in Linux someday ? (Craig Tierney) Re: Check your IRQ Setting on Everything !! Linux lives w/Caldera !!! (Ulrich Eckhardt) Problem making libc-5.3.5 (gethostby_r.c) (Michael N. Goldshteyn) Problem making libc-5.3.5 (gethostby_r.c) (Michael N. Goldshteyn) automating virtual console switch (Phil Howard) Re: Kernel Change Summary 1.3.72 (Marek Michalkiewicz) Re: Do we really need ttyS* and cua* ? (Miquel van Smoorenburg) ---------------------------------------------------------------------------- From: John Mellor Crossposted-To: comp.os.linux.misc Subject: Shutdown problem Date: 12 Mar 1996 16:48:29 GMT Recently I encountered a problem with shutdown. When I ran shutdown on two different linux machines (running different 1.3.x kernels), shutdown barfed and screwed up the system. I think I remember hearing about this. Can somebody please tell me how to fix this. Thanks, jp mellor ------------------------------ From: boyer@rd.bbc.co.uk (John Boyer) Subject: Re: Netscape broken under 1.3.70 Date: 12 Mar 1996 15:49:32 GMT John Burton (john@piper.gats.hampton.va.us) wrote: : In article <4hffe0$92q@aztec.co.za> dcook@aztec.co.za (David Cook) writes: : : > I have experienced a problem with Netscape v2.0 final release under : > 1.3.6something up to 1.3.70. : > : > As soon as Netscape tries to connect to a http server, it stops with a : > "Broken pipe" error. Booting Linux into v1.2.13 solves the problem. : > : > Anyone else expereinced this? : > : : Nope, works fine for me...Linux kernel 1.3.58, 1.3.68, 1.3.70 : P5-90 32MB memory, libc.so.5.2.18... : : John : -- : John Burton GATS, Inc. : j.c.burton@gats.hampton.va.us 28 Research Drive : j.c.burton@larc.nasa.gov Hampton, VA 23666 : (804) 865-7491 (voice) (804) 865-1021 (fax) : Works fine for me under 1.3.35. I left my machine compiling 1.3.72. this morning. I shall know tonight what it does to netscape 2.0. J. -- ************************************************************** * John Boyer, BBC Research & Development Department * * Kingswood Warren, Tadworth, Surrey, KT20 6NP * * email: john.boyer@rd.bbc.co.uk * * phone: 01737 836580 fax: 01737 836665 * ************************************************************** ------------------------------ From: fox@graphics.cs.nyu.edu (David Fox) Subject: Re: Emacs 19.30 "M-x manual-entry" output is truncated in 1.3.71 Date: 12 Mar 1996 08:23:37 -0500 In article dsf@mars.cs.nyu.edu (David Fox) writes: ] Don't use this, it causes a number of things not to work, like su. ] Noel Cragg suggests a much better solution: ] ] (setq process-connection-type nil) A further refinement - only use pipe connections for manual pages and compilations. This lets truly interactive processes work properly: *** lisp/man.el~ Fri Nov 24 17:19:36 1995 --- lisp/man.el Tue Mar 12 08:17:08 1996 *************** *** 542,544 **** (setq Man-arguments man-args)) ! (let ((process-environment (copy-sequence process-environment))) ;; Prevent any attempt to use display terminal fanciness. --- 542,545 ---- (setq Man-arguments man-args)) ! (let ((process-environment (copy-sequence process-environment)) ! (process-connection-type nil)) ;; Prevent any attempt to use display terminal fanciness. *** lisp/compile.el~ Mon Nov 13 22:48:02 1995 --- lisp/compile.el Tue Mar 12 08:17:28 1996 *************** *** 428,429 **** --- 428,430 ---- (let* ((process-environment (cons "EMACS=t" process-environment)) + (process-connection-type nil) (proc (start-process-shell-command (downcase mode-name) -- David Fox http://found.cs.nyu.edu/fox xoF divaD NYU Media Research Lab baL hcraeseR aideM UYN ------------------------------ From: dtrg@st-andrews.ac.uk (David Thomas Richard Given) Subject: Re: Logical Volumes in Linux someday ? Date: 12 Mar 1996 14:14:13 GMT In article , Remko Stuik wrote: >I was wondering if logical volumes will be in linux someday... >would be very nice to have it in 2.0 ! What do you mean by logical volumes? Unix doesn't really have the concept of `volumes' at all. -- =============================================================================== Why do people surf the Information Superhighway? Won't they get run over? http://www-hons-cs.cs.st-andrews.ac.uk/~dg Sun-Earther David Daton Given of Lochcarron ------------------------------ From: jlnance@avanticorp.com (Jim Nance) Crossposted-To: comp.os.linux.misc,comp.os.linux.networking,comp.unix.bsd.bsdi.misc,comp.unix.bsd.netbsd.misc,comp.unix.bsd.freebsd.misc Subject: Re: need secure OS to entrust millions to Date: 12 Mar 1996 13:50:19 GMT In article <4hsv73$d3h@park.uvsc.edu>, Terry Lambert wrote: >So you are arguing that the product of two prime numbers >renders the prime numbers themselves "secret". > >I claim that since the search space is known, they are "obscure". Would you make the same claim about any secret key algorythm that used a fixed length key. Jim ------------------------------ From: deh@atype.com (David E. Hollingsworth) Subject: Re: Logical Volumes in Linux someday ? Date: 12 Mar 1996 13:04:33 -0500 In article <4i40rl$119@calvin.st-and.ac.uk> dtrg@st-andrews.ac.uk (David Thomas Richard Given) writes: > In article , > Remko Stuik wrote: > >I was wondering if logical volumes will be in linux someday... > >would be very nice to have it in 2.0 ! > > What do you mean by logical volumes? Unix doesn't really have the concept > of `volumes' at all. You must never have used AIX, where one can define a set of "logical volumes" that don't have to correspond one-to-one with "physical volumes" (that is, hard drives). One thing that this gives you is the ability to declare that the logical volume is replicated multiple times across the physical volumes, so that a single drive failure can't cause loss of data. --deh! ------------------------------ Date: Wed, 6 Mar 1996 18:27:13 -0500 From: sl14@crux1.cit.cornell.edu Crossposted-To: list.linux-activists.kernel Subject: 1.3.7[01]: cannot open /dev/modem: no such device I tried to upgrade from 1.3.68, but I can't use the modem port with either 1.3.70 or 1.3.71. It says no such device. My /dev/modem is linked to /dev/cua1, and it works with 1.3.68 and all previous version of the kernel. I tried both the beta pgcc and gcc-2.7.2 (no SR bug) it doesn't affect the problem. Is there any fix for this? I can't load modules either, but I've seen that one being discussed already. sl14@cornell.edu Witty .sig under construction. ------------------------------ From: stuik@fys.ruu.nl (Remko Stuik) Subject: Logical Volumes in Linux someday ? Date: Thu, 7 Mar 1996 21:57:50 GMT Hi, I was wondering if logical volumes will be in linux someday... would be very nice to have it in 2.0 ! Regards, Remko -- General Protection Fault in Module <> Segmentation Fault Core Dumped Cannot Load Command.Com System Halted Windows, Unix, Dos, What's in the name, In the end they're all the same *********************************************************************** * stuik@fys.ruu.nl * http://www.fys.ruu.nl/~stuik * *********************************************************************** * GS d+ s+: a21 C++ ULUAHOS*++ P- L++ E--- W++ N+ o K? w+(w--) O- * * M- V-- PS PE Y PGP- t+ 5(5+) X+ R tv b+ DI D+ G e>++++ h r-- y? * ------------------------------ From: stano@entw-1.swh.sk (Stanislav Meduna) Subject: Re: Chicken-or-the-egg-game with microkernel and FS drivers Date: 12 Mar 1996 08:49:29 GMT In article , albert@krakatoa.ccs.neu.edu (Albert Cahalan) writes: >A> Last question: Can LILO boot Linux off of *any* filesystem >A> format? Even if I invent my own and never bother telling >A> LILO about it? > >Yes, LILO just keeps a list of blocks. Probably not. LILO over NFS would be quite funny :-) BTW, _how_ does LILO determine the block number, when given a file and an offset ? Does such function exist in the VFS layer, so it can be done independently of the actual FS (when supported) ? Regards -- Stano ------------------------------ From: tierney@rintintin.Colorado.EDU (Craig Tierney) Subject: Re: Logical Volumes in Linux someday ? Date: 12 Mar 1996 19:18:36 GMT In article , David E. Hollingsworth wrote: > >In article <4i40rl$119@calvin.st-and.ac.uk> dtrg@st-andrews.ac.uk (David Thomas Richard Given) writes: > >> In article , >> Remko Stuik wrote: >> >I was wondering if logical volumes will be in linux someday... >> >would be very nice to have it in 2.0 ! >> >> What do you mean by logical volumes? Unix doesn't really have the concept >> of `volumes' at all. > >You must never have used AIX, where one can define a set of "logical volumes" >that don't have to correspond one-to-one with "physical volumes" (that is, >hard drives). > >One thing that this gives you is the ability to declare that the logical >volume is replicated multiple times across the physical volumes, so that a >single drive failure can't cause loss of data Under Netware this is referred to a mirroring, and a logical volume is a number of physical partitions (drives) that can be strung together. Thus if you have two 1 gig drives, the operating system would make it look as if there is one 2 gig drive. Under this description of logical volumes, if you lose one of the volumes, you lose the whole logical partition. So which definition of 'logical volumes' are we talking about? Craig Craig Tierney (tierney@orbit.colorado.edu) | 3 - My favorite Colorado Center for Astrodynamics Research(CCAR) | integer between University of Colorado at Boulder | e and pi. ------------------------------ From: Ulrich Eckhardt Crossposted-To: comp.os.linux.setup,comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.admin Subject: Re: Check your IRQ Setting on Everything !! Linux lives w/Caldera !!! Date: Tue, 12 Mar 1996 20:58:19 +0100 74201.3202@compuserve.com wrote: > > I am also loading Linux...Caldera's new CND with an Xserver front end > Netscape brower and Web Server the whole nine yards...I have it > working with Win'95 via LILO and in works great except for two > things...... > > Any way sounds like you have a conflict with your IRQ's best > bet take out as many NON-essentials devices then try to load it. > > I didn't even think TI was still making CPU's that could be a source > of your problems.....The linux I discussed above is cheap check it > out at WWW.caldera.com > > PJH Chicago IL Hi, is Word Perfect available also ? -- ========================================================== Ulrich Eckhardt Ulrich.Eckhardt@darmstadt.netsurf.de W.-Liebknecht Str. 53 64846 Gross - Zimmern Germany ------------------------------ From: mikeg@sashimi.wwa.com (Michael N. Goldshteyn) Subject: Problem making libc-5.3.5 (gethostby_r.c) Date: 9 Mar 1996 09:39:50 -0600 I am trying to build libc-5.3.5 and it is getting some errors in the module libc/sysdeps/pthreads/inet . The error messages (warnings excluded) when the module is compiled are the following: ============================================================================ ../../../../include/pthread/mit/sleep.h: In function `machdep_gettimeofday': In file included from ../../../../include/pthread/mit/pthread.h:54, from ../../../../include/pthread.h:9, from gethostby_r.c:1: ../../../../include/pthread/mit/sleep.h:57: dereferencing pointer to incomplete type ../../../../include/pthread/mit/sleep.h:57: dereferencing pointer to incomplete type ../../../../include/pthread/mit/pthread.h: At top level: In file included from ../../../../include/pthread.h:9, from gethostby_r.c:1: ../../../../include/pthread/mit/pthread.h:131: field `wakeup_time' has incomplete type =========================================================================== I looked in sleep.h to see what this "incomplete type" was and saw structures of type "struct timeval" being used although the struct is never defined. I believe that this may be the problem. Has anyone else experienced this? If you've been able to get around this problem, please let me know how. Thank You Michael Goldshteyn mikeg@wwa.com ------------------------------ From: mikeg@sashimi.wwa.com (Michael N. Goldshteyn) Crossposted-To: comp.os.linux.misc Subject: Problem making libc-5.3.5 (gethostby_r.c) Date: 9 Mar 1996 09:41:46 -0600 I am trying to build libc-5.3.5 and it is getting some errors in the module libc/sysdeps/pthreads/inet/gethostby_r.c. The error messages (warnings excluded) when the module is compiled are the following: ============================================================================ ../../../../include/pthread/mit/sleep.h: In function `machdep_gettimeofday': In file included from ../../../../include/pthread/mit/pthread.h:54, from ../../../../include/pthread.h:9, from gethostby_r.c:1: ../../../../include/pthread/mit/sleep.h:57: dereferencing pointer to incomplete type ../../../../include/pthread/mit/sleep.h:57: dereferencing pointer to incomplete type ../../../../include/pthread/mit/pthread.h: At top level: In file included from ../../../../include/pthread.h:9, from gethostby_r.c:1: ../../../../include/pthread/mit/pthread.h:131: field `wakeup_time' has incomplete type =========================================================================== I looked in sleep.h to see what this "incomplete type" was and saw structures of type "struct timeval" being used although the struct is never defined. I believe that this may be the problem. Has anyone else experienced this? If you've been able to get around this problem, please let me know how. Thank You Michael Goldshteyn mikeg@wwa.com ------------------------------ From: phoward@sunsrv13.clr.com (Phil Howard) Subject: automating virtual console switch Date: 12 Mar 1996 17:08:27 GMT I am setting up a Linux console for a network and server monitoring system for the machine room operators at the main console area. The plan is to have the primary virtual console show the status of all the monitored services, and other virtual consoles showing more detail for some of them. The operators may switch the virtual console and leave it and then not notice the red flashing indications on the primary virtual console. So what I'd like to do is have it automatically switch after some idle time. I'm thinking that the best place to do this is to modify the kernel console code where it normally blanks the screen, and instead have a means to make it switch to a certain screen instead. This might involve a new escape sequence or a new system call to indicate where to switch to, the timeout period, etc. Has anyone done this, either this way or any other way? If I do this, would you be interested in it if it had some special feature? Any other or alternate suggestions? Note that merely managing everything from my program within one virtual console is not adequate as I want to allow the operators (or myself) to make use of regular virtual consoles with getty running to login and do whatever needs to be done in certain cases, yet to ensure that even from these virtual consoles the idle switchback will still take place. Note that my plan is to have my monitoring program open non-getty consoles. ------------------------------ From: marekm@i17linuxb.ists.pwr.wroc.pl (Marek Michalkiewicz) Subject: Re: Kernel Change Summary 1.3.72 Date: 12 Mar 1996 21:00:32 GMT Michael Elizabeth Chastain (mec@treflan.shout.net) wrote: : Kernel Change Summary : Linux 1.3.72 (25957 lines) : Sat 9 Mar 1996 : : My KCS archive: ftp://ftp.shout.net/pub/users/mec/kcs Sounds like you forgot this one (important!): arch/i386/kernel/ldt.c: fixed a big security hole in sys_modify_ldt There is an exploit program which gives root to any user, works on 1.3.x (x<72) and (after trivial modification) on 1.2.13 too! Since this file was not changed since Feb 5, 1995, it should be possible to copy it to 1.2.13 to plug this hole. Marek ------------------------------ From: miquels@drinkel.ow.org (Miquel van Smoorenburg) Subject: Re: Do we really need ttyS* and cua* ? Date: Mon, 11 Mar 1996 01:21:11 +0100 (MET) In article , Ian T Zimmerman wrote: >If you want to callin and callout on the same port, use uugetty which >understands the two-device scheme. It is not unique to Linux - SunOS >does a similar thing. [SNIP] >I think there is an _almost_ correct solution with mgetty and one >device, but it involves polling and not locking the port until polling >determines there is actually a connection. To this person the >two-device scheme is cleaner. We should have an ioctl that one can do on an already open ttyS port that blocks the read() from the device instead of the open(). That way, any dialout program can open the cua device if the ttyS device is not busy, and as soon as the modem starts to send RING messages the ttyS port is opened. It should be trivial to implement; the reason that it hasn't been done is that the scheme is modeled after a design of a few years old, in a time where modems were assumed to but put in auto-answer mode and there weren't any "chatty" getties talking hayesspeak around yet. Tytso, are you reading this? Mike. -- + Miquel van Smoorenburg + Cistron Internet Services + Living is a | | miquels@cistron.nl (SP6) | Independent Dutch ISP | horizontal | + miquels@drinkel.ow.org + http://www.cistron.nl/ + fall + ------------------------------ ** 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-Development-System-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.development.system) via: Internet: Linux-Development-System@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-Development-System Digest ******************************