From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Mon, 28 Feb 94 15:13:10 EST Subject: Linux-Development Digest #508 Linux-Development Digest #508, Volume #1 Mon, 28 Feb 94 15:13:10 EST Contents: Re: Why not put cluster diffs in nominal kernel before 1.0? (David L. Black) Patch to scheduler in developemnt : Simple Question (S. Joel Katz) Re: sysvinit-2.5 and bash-1.13.5 problem! (Miquel van Smoorenburg) Re: sysvinit-2.5 and bash-1.13.5 problem! [SOLUTION] (Miquel van Smoorenburg) Weird problems with sendmail 8.6.5 and 8.6.6 (Edvard Tuinder) TTY bugs (Frank Lofaro) Re: Specialix driver (Thomas Quinot) Linux/Windows (a920101@zipi.fi.upm.es) Re: YP or NIS for linux? (John F. Haugh II) Re: Alpha release of message catalog command and functions (John F. Haugh II) Re: NETWORK unreachable in pl15 (Alexander Lau) Re: dip at 14.4?? (Brian Cummings) X & Viper (Brian Cummings) Re: PLEASE use the GPL -- NOT (Jay Maynard) Re: bug in harddisk controller code?? (Nick Andrew) Re: Specialix driver (Jay Maynard) Adaptec 1542A and kernel panic (Sarat Vemuri) ---------------------------------------------------------------------------- From: dave@hh.sbay.org (David L. Black) Subject: Re: Why not put cluster diffs in nominal kernel before 1.0? Date: 28 Feb 1994 00:20:58 -0800 In pclink@qus102.qld.tne.oz.au (Rick) writes: >While cluster-08a on pl15a and pl15h boost the iozone performance on my >FD885+767Mb Seagate, fsck runs 3-4 times slower on a 75% full 200Mb >ext2 partition. Anybody else noticed this? >Rick. Yeah, noticed it here too. I was starting bdflush *before* fsck-ing. Start it *after* fsck-ing and speed is back to normal, with this change: in wakeup_bdflush() (fs/buffer.c), commented out the printk and sync_buffers if bdflush_running is false. Works for me, but not to say I didn't inadvertantly break something... :) Dave -- David L. Black dave@hh.sbay.org Hip-Hop BBS Sunnyvale, CA KE6AJC @ N0ARY.#NOCAL.CA.USA.NA ------------------------------ From: stimpson@panix.com (S. Joel Katz) Subject: Patch to scheduler in developemnt : Simple Question Date: 28 Feb 1994 05:24:45 -0500 I am working on a patch to the Linux scheduler to reserve a certain percentage of CPU time for the superuser. This would prevent mere users from creating so many processors that a superuser couldn't access the system quickly to fix some damage. The system works by keeping track of which processes deserve special treatement and incrementing an int everytime a process that is flagged gets a timeslice. It uses a decay function and modifies priorities such that the superuser processes get increased user level priorities whenever the decay function indicates cpu use below the threshold which is set by a system call. Now, my question: You don't want users who run setuid root programs to count against the superuser. You don't want logins, gettys and system daemons and stuff to count either (or do you?). I have to determine during the fork system call whether to mark the process as part of the special group or not. Is there some simple algorithm that I haven't thought off? I currently do it based on the real user id. Unfortunately this includes login and update and demon processes. I can't think of an easy fix. Any ideas? SJK ------------------------------ From: miquels@drinkel.nl.mugnet.org (Miquel van Smoorenburg) Subject: Re: sysvinit-2.5 and bash-1.13.5 problem! Date: Fri, 25 Feb 94 01:10:14 MET In article <1994Feb23.074508.29267@sifon.cc.mcgill.ca> isaac@physics.mcgill.ca writes: > Sysinit is actually much worse than that. If your booting fails > while still read-only, then saving ioctl.save restricts the > machine from being rebooted. Okay, it complains about not being able to write /etc/ioctl.save, but that's more an informative error. It just doesn't write the record, but it actually doesn't care. > And if the process is killed in > between (as in your case), the shutdown.allow fails since all > shells are killed... Yup. I only put in shutdown.allow as a feature, if you have problems with it then just don't create shutdown.allow - then all will be as it was in init 2.4 (aren't the docs clear about this?) > had the same problem you have with kill. I heard. There is a patch on sunsite and tsx-11 for bash. It reckognizes that the shell it is running on is actually bash, and re-enables the builtin kill command (you can do that. GNU software can do everything :) > In conclusion SysVinit 2.5 is VERY fragile and I don't recommand > it to anyone before it is made a bit more resistant. Stick > to 2.4 which is still superior IMHO. As you know, I'm off the net for half a year. Just wanted to put out a new version. Some things have changed, but I think for the better. I hope some distribution makers will pick it up, then it will crystallize eventually. Things aren't as bad as they seem. I am convinced sysvinit 2.50 is as good, or better, than 2.4. > (As said in French: "Better is sometimes enemy of good") :-) Q: It's blonde, male, a very good lover, and speaks French. A: Moi. :) >Martin > Mike. -- | Miquel van Smoorenburg | "I know one million ways, to always pick | | miquels@drinkel.nl.mugnet.org | the wrong fantasy" - the Black Crowes. | ------------------------------ From: miquels@drinkel.nl.mugnet.org (Miquel van Smoorenburg) Crossposted-To: comp.os.linux.admin,comp.os.linux.misc Subject: Re: sysvinit-2.5 and bash-1.13.5 problem! [SOLUTION] Date: Sat, 26 Feb 94 22:05:27 MET In article <2kjqjo$8so@usenet.INS.CWRU.Edu> chet@hela.ins.cwru.edu (Chet Ramey) writes: >In article <940222371.212734@drinkel.nl.mugnet.org>, >Miquel van Smoorenburg wrote: > >Now, where did you get a version of 1.14? Ask Marco van Wieringen how he manages to get the weirdest version numbers into programs. My bash 1.12.1 seems to be a 1.13.5. Hmm. :) :) >>There is a better solution. With bash, it is possible to turn the >>commands back on (I read the source to find this). You can use >>"enable ". > >Gee, you could also have read the manual to find it :-). Hey, you know the saying: "When all else fails, read the manual". >Chet Mike. -- | Miquel van Smoorenburg | "I know one million ways, to always pick | | miquels@drinkel.nl.mugnet.org | the wrong fantasy" - the Black Crowes. | ------------------------------ From: Edvard.Tuinder@delirium.NL.MugNet.ORG (Edvard Tuinder) Crossposted-To: comp.mail.sendmail Subject: Weird problems with sendmail 8.6.5 and 8.6.6 Date: 27 Feb 1994 18:19:43 +0100 Hi, I'm experiencing a real weird problem with my sendmail. I recently upgraded from 8.6.4 to 8.6.5. Since that time I am unable to send local mail to any user, except when they have a .forward file. Today I've tried 8.6.6, but the same thing happens. Sample session: (I've changed the non-ascii char's like CTRL-D in ^D etc.) #p0 18:07> rmail alex < test-mail #p0 18:08> mailq Mail Queue (1 request) --Q-ID-- --Size-- -----Q-Time----- ------------Sender/Recipient------------ SAA02009 2 Sun Feb 27 18:08 ed \3702^D #p0 18:08> rmail ewout alex root < test-mail #p0 18:09> mailq Mail Queue (1 request) --Q-ID-- --Size-- -----Q-Time----- ------------Sender/Recipient------------ SAA02011 2 Sun Feb 27 18:08 ed Subject Subject root #p0 18:08> rmail ed < test-mail #p0 18:08> mailq Mail Queue (1 request) --Q-ID-- --Size-- -----Q-Time----- ------------Sender/Recipient------------ SAA02029 2 Sun Feb 27 18:08 ed "|/staff/bin/ed" #p0 18:15> sendmail -d0,15 -d20,1 -d24,5 alex < test-mail Version 8.6.6 SYSTEM IDENTITY (after readcf): (short domain name) $w = delirium (canonical domain name) $j = delirium.nl.mugnet.org (subdomain name) $m = nl.mugnet.org (node name) $k = delirium.nl.mugnet.org --parseaddr(root) parseaddr-->30f1c=root: mailer 3 (local), host `', user `root', ruser `' next=0, flags=0, alias 0, uid 0, gid 0 owner=(none), home="(none)", fullname="(none)" --parseaddr(alex) parseaddr-->44d40=alex: mailer 3 (local), host `', user `alex', ruser `' next=0, flags=0, alias 0, uid 0, gid 0 owner=(none), home="(none)", fullname="(none)" From person = "root" #p0 18:15> mailq Mail Queue (1 request) --Q-ID-- --Size-- -----Q-Time----- ------------Sender/Recipient------------ SAA02036 12 Sun Feb 27 18:15 root \3702^D #p0 18:15> I really have no clue as to what is going on. At first I thougth the getopt function that comes with the sendmail src somehow broke the address, but after some debugging that didn't appear to be the problem. I'm running linux with gcc-2.5.8 with libc-4.5.19. I've compiled 8.6.6 with optimizing, without optimizing, static linked, but that didn't solve the problem. Has anybody got an idea as to what may be going on? Thanx in advance, Ed -- Edvard Tuinder ed@delirium.nl.mugnet.org In the beginning was the Word - And the Word was: ``Hey, you!'' ------------------------------ From: ftlofaro@unlv.edu (Frank Lofaro) Subject: TTY bugs Date: Mon, 28 Feb 94 07:00:29 GMT These tty bugs exist both in standard pl15h and in the new tty code by tytso@athena.mit.edu. 1. /dev/tty0 points to the vc that was active at OPEN time, not read/write time. 2. If /dev/tty0 is the tty of the process, all access to /dev/tty returns ENXIO: No such device or address. 3. If /dev/tty0 is the tty of the process, control-c, etc are IGNORED. This exists in the new tty code: 4. The new tty code hoses the TTY column of procps. It shows con for processes with no tty, for example. Any ideas on what needs to be done? I am willing to help fix these bugs. The fixes will be relative to the new tty code. I have switched to using the new tty code, it seems to be a much cleaner interface, etc. Good work! P.S. I did send a subscribe message to the new-tty mailing list, but have not got a confirmation so I am sending this post here for the time being. (I don't trust mailing lists too much anymore, ever since I got dropped from the KERNEL channel by a software glitch). Also, 3 of the 4 bugs listed above exist in the standard pl15h distribuution. ------------------------------ From: thomas@melchior.frmug.fr.net (Thomas Quinot) Subject: Re: Specialix driver Date: 22 Feb 1994 14:17:56 +0100 Jon Brawn (jonb@specialix.com) wrote: : IF Specialix were to write a driver for SI on Linux, we could NOT release the : source of the download code into the public domain AT ALL. We COULD supply a : binary file with it in. The Linux driver source would be made available. But... Why couldn't you license it under the GNU General Public License ? This has nothing to do with public domain, and Linux is _not_ public domain software... : WOULD this be legal? Yup. But it wouldn't be portable/customizable/hackable... : Comments? Here's my 3 cents worth :-) Cheers, Thomas. -- ThoThoThoThoTho Totolitoto ! ------------------------------ Subject: Linux/Windows From: a920101@zipi.fi.upm.es Date: 28 Feb 94 10:32:49 +0100 I heard something about a MS-Windows emulator under Linux. Can anybody tell me about it? (E-mail, please). THANKS! Juanje. ------------------------------ From: jfh@rpp386 (John F. Haugh II) Subject: Re: YP or NIS for linux? Reply-To: jfh@rpp386.cactus.org (John F. Haugh II) Date: Mon, 28 Feb 1994 01:13:58 GMT In article <2kola2$e3p@bambi.zdv.Uni-Mainz.DE> kubla@goofy.zdv.Uni-Mainz.DE (Dominik Kubla) writes: >There are already two implementations: > yp-linux > NYS >The first one can be found in any Slackware source archive, while the >second one is (at least) available fron nic.funet.fi in /pub/BETA >yp-linux is a port of the NetBSD implementation, while NYS is written >from scratch and features service switching between files, nis, nis+ and >hesiod. But it is not yet completed and will force you to recompile all >affected binaries while the yp-linux support is included in libc since >release 4.5.x I picked up a copy of the latest NYS and checked out some of the functions. There are two serious deficiencies -- 1). /etc/shadow is mishandled when it or a single entry doesn't exist. 2). There is no support for DBM files. The first bug makes an application incapable of determining if an entry actually exists since getspent() and friends will return the entry from /etc/passwd instead. While this might seem to be a favor, it is imcompatible with SVR[>3]. The second isn't really a bug and the regular C library has this same shortfall. The result of this is that if you have a large /etc/passwd file it takes longer for the last user in the file to login than the first. I've not had a chance to look at yp-linux, but if it is based on NetBSD it probably doesn't have /etc/shadow support at all, though it should have DBM file support since BSD is where DBM file support comes from ... -- John F. Haugh II [ NRA-ILA ] [ Kill Barney ] !'s: ...!cs.utexas.edu!rpp386!jfh Ma Bell: (512) 251-2151 [GOP][DoF #17][PADI][ENTJ] @'s: jfh@rpp386.cactus.org There are three documents that run my life: The King James Bible, the United States Constitution, and the UNIX System V Release 4 Programmer's Reference. ------------------------------ From: jfh@rpp386 (John F. Haugh II) Subject: Re: Alpha release of message catalog command and functions Reply-To: jfh@rpp386.cactus.org (John F. Haugh II) Date: Mon, 28 Feb 1994 01:28:53 GMT In article <2kq8g0$fas@senator-bedfellow.MIT.EDU> Mitchum.Dsouza@mrc-apu.cam.ac.uk writes: >You should have consulted before you started this. NLS has been implemented >and integrated into libc ever since libc-4.4.4. It seems you have wasted you >time somewhat. How bizarre. I got this same response in the mail ... Anyhow, I needed that for something I wrote and it only took a few hours to whip up. I will look at the referenced package closer, but when I looked at it last weekend it appeared to be incomplete and I wasn't convinced in the least that it would work. Based on your posting I suspect I need another look at it. -- John F. Haugh II [ NRA-ILA ] [ Kill Barney ] !'s: ...!cs.utexas.edu!rpp386!jfh Ma Bell: (512) 251-2151 [GOP][DoF #17][PADI][ENTJ] @'s: jfh@rpp386.cactus.org There are three documents that run my life: The King James Bible, the United States Constitution, and the UNIX System V Release 4 Programmer's Reference. ------------------------------ From: kclau@harbor.tem.nctu.edu.tw (Alexander Lau) Subject: Re: NETWORK unreachable in pl15 Date: 28 Feb 1994 15:23:57 GMT David Sisson (daves@megavolt.cc.vt.edu) wrote: : In article <2k14hf$d6c@pascal.csid.gmeds.com>, : Mario Camou wrote: : >madge@shell.portal.com (Nick Tingle) writes: : >: >camou@csid.gmeds.com (Mario Camou) writes: : >: >: I'm having the same problem. The message I get is: : >: >: : >: >: SIOCADDRT: Network is unreachable : >: > : >: >After reading several other posts in this newsgroup and : >: >comp.os.linux.help, I gathered that I needed a new route command (it's : >: >on its way via ftpmail). Meanwhile, I have to do route commands for each : The new route command didn't help me. The problem is actually occuring : at the 'ifconfig eth0 ...' line. At least four errors show up for that : (starting with the one listed above). If I type just 'ifconfig eth0', I get : interface unknown. Either there is a problem with ifconfig or with the : kernel. I've compiled pl14 before with no problems. Have you ever try net-0.32 ? At least it work for me ! But strange that after using net-0.32, pl15 work but pl14 fail. ------------------------------ From: brianc@myhost.subdomain.domain (Brian Cummings) Subject: Re: dip at 14.4?? Date: 28 Feb 1994 15:44:58 GMT Jim O'Quinn (oquinn@vern.bga.com) wrote: : Is there a set of patches for dip that will run my modem at 14.4? : 9600 is wwwaaayy tttooo ssslllloooowwwww..... I'm using DIP 3.3.7 and have my port set to 38400. Try that. ------------------------------ From: brianc@myhost.subdomain.domain (Brian Cummings) Subject: X & Viper Date: 28 Feb 1994 15:46:29 GMT Does anyone know if the XFree Viper project has reached color yet? :) I'm anxious to see my victims in Netrek in their full-color :) ------------------------------ From: jmaynard@nyx10.cs.du.edu (Jay Maynard) Subject: Re: PLEASE use the GPL -- NOT Date: Mon, 28 Feb 94 15:24:51 GMT In article , Craig Burley wrote: >Fortunately, that gun is mythical, since GPL lawyers aren't telling >you you can't use their software in your programs. In fact, GPL >supporters, and the GPL itself, actually encourage you to do so. Of course they do. The more GPV-infected code there is out there, the closer the world gets to RMS' software communist utopia. -- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can jmaynard@oac.hsc.uth.tmc.edu | adequately be explained by stupidity. "The difference between baseball and politics is that, in baseball, if you get caught stealing, you're out!" -- Ed Shanks ------------------------------ From: nick@kralizec.zeta.org.au (Nick Andrew) Subject: Re: bug in harddisk controller code?? Date: 23 Feb 1994 01:17:20 +1100 In <1994Feb16.065724.4788@bogomips.ee.ubc.ca> jmorriso@bogomips.ee.ubc.ca (John Paul Morrison) writes: >I recently had a serious hang with Linux 0.99.15c, and I think >something got buggered with the IDE controller & the block device code. >(I had to power down to get harddisk IO back; the reset button was not >enough) I have had the same sort of hang - except with a SCSI disk rather than EDI. I posted a couple of articles about it in c.o.l.help and here entitled something like "Heavy swapping hangs Linux". Nick. -- Kralizec Dialup Unix (Public Access) Data: +61-2-837-1183, 837-1868 Zeta Microcomputer Software v.42bis v.32bis 14.4k 24 hours P.O. Box 177, Riverstone NSW 2765 Plan: To beat Gnuchess 4.0 fairly! ------------------------------ From: jmaynard@nyx10.cs.du.edu (Jay Maynard) Subject: Re: Specialix driver Date: Mon, 28 Feb 94 15:20:17 GMT In article , Charles Hedrick wrote: >Why are we having arguments over absurd interpretations of the GPL >when Linus has clarified the interpretation that actually applies to >Linux? Because others are exhorting people to use the GPV to license their code, and so the question "Is the GPV a Good Thing?" is relevant. -- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can jmaynard@oac.hsc.uth.tmc.edu | adequately be explained by stupidity. "The difference between baseball and politics is that, in baseball, if you get caught stealing, you're out!" -- Ed Shanks ------------------------------ From: sarat@crl.com (Sarat Vemuri) Crossposted-To: comp.os.linux.help Subject: Adaptec 1542A and kernel panic Date: 28 Feb 1994 09:19:26 -0800 Hello all, I am running Linux on a AMD 386/40. It has IDE hard disk and I recently added the 1542A SCSI controller and a LMS CM214 CD-ROM. Previously, my kernel didn't have SCSI in it. I recompiled kernel and everything is working fine. But, the kernil panics after being idle for a while. The panic message goes some thing lik "Swapper killed", "Swapper Not syncing" etc. This panic occurs even with the kernel without SCSI compiled in it. Mere presence of the SCSI controller causes this. I've tried with pl13 and pl15 with same results. Is this a known problem with Adaptec 1542A? I also have the following, USR SPortster, Actix GE 32 video card. The 1542 is using IRQ11, DMA 6 and port 0x330. I appreciate any help you can offer. I am going nuts with these panics since I leave the machine on all the time to process news and mail. Thanks Sarat sarat@crl.com ------------------------------ ** 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-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.development) via: Internet: Linux-Development@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 Digest ******************************