From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Sun, 7 Nov 93 19:13:26 EST Subject: Linux-Development Digest #214 Linux-Development Digest #214, Volume #1 Sun, 7 Nov 93 19:13:26 EST Contents: Q: crond (el@lisse.NA) [SOLUTION] Re: TERM problems- Please help! (Simon J Ferrett) Re: WANTED: COBOL compiler (John R. Campbell) Sudden lockups when overloading a socket? (Daniel Y.H. Wong) Linux for mca? (selma flatmoe) UltraStor 24F/34F anyone? (Marvin L. Taylor) OS/2 and LINUX INSTALLATION (ANDROUTSOS DIMITRIOS) Re: WANTED: COBOL compiler ---------------------------------------------------------------------------- From: el@lisse.NA Subject: Q: crond Date: Sun, 7 Nov 1993 08:51:10 GMT Hi, I have a little problem with the crond. I use the SLS103 distribution of the Trans-Ameritech CD, but tried the Yggdrasil demon that comes with their old CD, the SLackware one and the vixie-cron sources. I use the batching and compress to email (smail3.1.28). This involves making changes to smails configuration to put the mails in a spool directory and run a job once in a while to pipe all the files trhough compress into uux. I noticed at some stage that some messages were sent twice or even three times. I checked smail's script (/usr/local/lib/smail/util/cbsmtp.sh), but if I run this manually it works just fine. I run top in one window and change the times in the crontab and find that I have two crond running. Hmm, check rc and rc.local, they only start crond once. I thought may be crond just started another copy because because the flock system may have worked not too well. Reboot the system. Now I run top for even longer and see occasionally something like this: PID USER PRI NI SIZE RES SHRD ST %CPU %MEM TIME COMMAND 20 root 29 0 72 108 204 S 0.0 1.4 0:16 /etc/crond 2107 root 29 0 72 108 204 S 0.0 1.4 0:16 /USR/BIN/CROND ^^^^^^^^^^^^^^ Note the Upper Case. I compiled the vixie-cron sources and get about a thousand :-)-O zombies while it is running. Is this known behaviour? Has this been fixed? Where do I get the fixes? greetings, el ps: I don't get c.o.l.help due to noise vs signal ratio and my feed is several thousand miles away. Flame as much as you like :-)-O -- Dr. Eberhard W. Lisse \ / Windhoek Central Hospital \ * | Department of Obstetrics and Gynaecology Private Bag 13215 \ / 61 203 2106/7 (Bleeper) 61 224014 (home) Windhoek, Namibia ;____/ ------------------------------ Crossposted-To: comp.os.linux.help,comp.os.linux,comp.os.linux.misc From: c9108932@peach.newcastle.edu.au (Simon J Ferrett) Subject: [SOLUTION] Re: TERM problems- Please help! Date: Sun, 7 Nov 1993 13:26:10 GMT bouchier@convex.com (Paul Bouchier) writes: >Here's a fix the problem. >Replace line 38 in misc.c, which was: >current_time = t.tv_sec * 20 + t.tv_usec / 50000; >with >current_time = (t.tv_sec & 0xffffff) * 20 + t.tv_usec / 50000; >Problem is t.tv_usec is currently around 0x2cdc459b, and will overflow >when multiplied by 20. I don't know why it was working before - I guess >it overflowed to a harmless number before, but now doesn't. If this isnt already a FAQ it should be - ter1.0.7 stopped working recenly due to the exact thing you just described above. The solution? Grab term 1.0.8 from you favourite linux ftp site and all problems will be solved.. Hope this helps... -- c9108932@cs.newcastle.edu.au - Simon Ferrett Alias Mcv on irc sometimes... BCompSci (or getting there anyhow...) ------------------------------ From: soup@penrij.UUCP (John R. Campbell) Crossposted-To: comp.os.linux.help Subject: Re: WANTED: COBOL compiler Date: 6 Nov 93 13:23:11 GMT raf4482@TAMUTS.TAMU.EDU (Reid Allen Forrest) writes: >In article <1993Nov2.224954.1702@penrij.uucp>, >John R. Campbell wrote: >>COBOL == Completely Obnoxious Boring Obsolescent Lanquage >>There are NO happy COBOL programmers (an obvious oxymoron). >Yes, I'm completely serious! COBOL is not as obsolescent as some may (like) >to think. It's in use worldwide in thousands of businesses, just like >FORTRAN is in use in thousands of engineering, space, and scientific >establishments. COBOL is generally the preferred language for business >programming. Granted. Of course, I've seen places where PL/I has replaced COBOL. When I did COBOL programming (both on MainFrames and even SCO XENIX) I didn't enjoy it. For instance, the Automated Bonds System for the New York Bond Market is written in FORTRAN rather than COBOL; the FORTRAN compiler for the UNIVAC 1100 generated better code for the application and was more easily written to process the (almost) free-form order messages. I had a lot of fun going through audit tapes using NOSC Pascal for the 1100... > Not to mention the fact that many of the businesses around >today have most of their software written in COBOL. While COBOL is good at what it's good at :-) I don't see a long future. Unfortunately, COBOL is _really_ designed for mainframe use and BATCH mentality. The "terminal" specific stuff has used what, on a mainframe, is the "operator's console" (i.e. ACCEPT xxx FROM CONSOLE). This makes COBOL programs rather environment specific. Of course, all my mainframe work was batch work (I date back to the days of card readers) so my attitude shows my age (or is it my age shows my attitude?). Don't take me _too_ seriously. I still think that COBOL's days are over; new programs should be written in a more useful language (but NOT "C": C combines the power of assembly language with the ease of use of assembly language). Ada, despite DoD backing, almost looks competent, and, at least, there IS an FSF effort to support Ada (GNAT). For NEW programs, I'd discourage use of COBOL. For programs that already exist that you want to maintain, I'd suggest retention of the existing platform (like LPI COBOL on SCO UNIX/XENIX). > And, yes, I do know >MANY happy COBOL programmers. :) Are you sure you aren't confusing it with drug use???? :-) (Sorry but my attitude is _not_ easily adjusted). COBOL is a GREAT lightning rod to start/sustain a flame war. I suspect that a LEX/YACC - based compiler would be possible (I don't know about desirability, though). Maybe you should consider a DOS COBOL compiler and run the application through DOSEMU... :-) Toodles... -- John R. Campbell soup%penrij@kd3bj.ampr.org or: uunet!kd3bj!penrij!soup -- John R. Campbell soup%penrij@kd3bj.ampr.org or: uunet!kd3bj!penrij!soup ------------------------------ From: wongda@eecg.toronto.edu (Daniel Y.H. Wong) Subject: Sudden lockups when overloading a socket? Date: 7 Nov 93 16:44:13 GMT Hi, My system lockup completely when I am using slip and X11 during heavy transfer. I can not always reproduce the crash, but in the linux pl13 src it say: REMAINING KNOWN BUGS AND PROBLEMS: ... + Sudden lockups when overloading a socket? This seems to occur with X11 sessions (as per Linus Torvalds, 05/28/93) and has its origin in the new timer.c code... I wonder anybody else experience the same problem, and is there some patches to fix it? Thanks -- Daniel Y.H. Wong UofT:(416)978-1659 wongda@picton.eecg.toronto.edu Electrical Engineering -- ------------------------------ From: sflatmoe@nde.unl.edu (selma flatmoe) Subject: Linux for mca? Date: 7 Nov 1993 17:30:25 GMT I have been told that linux will not run on the mca. I have a ps2 and would like to put linux on it sometime. Will this ever be possible? If so when? But mabee i should first ask if anyone has any versions of it that will run on the mca? Please let me know..thanx ------------------------------ Crossposted-To: comp.os.linux.admin From: marvint@csn.org (Marvin L. Taylor) Subject: UltraStor 24F/34F anyone? Date: Sun, 7 Nov 1993 18:25:07 GMT Greetings all, I am hoping that someone here can help me with the problems I'm having with LINUX. My primary problem is that I cannot rebuild the kernel. I am having other problems, too, but these may clear up if I manage to rebuild the kernel properly. It's difficult for me to check/post-to this news group (plus it costs $$'s), so I'll try to be really verbose and give all the information one might ever need to fix the problem. Fundamentally, trying to get a kernel that will support the 24F (and for the 34F, for my roommate's new PC), has proven to be a chore. The sources I've obtained won't compile (see below)! HARDWARE INFO (of note): AMI 486DX/33, EISA Diamond SpeedStar/1MB (ISA) UltraStor 24F (EISA) 2 HD's (Both are SCSI-2, Addr's 0, 1), 2 FD's (3.5", 5.25"), NEC-84 CD-ROM BOOT SEQUENCE: OS/2 2.1 Boot Manager boots, selects LINUX partition. LILO boots to the /dev/sda3 (180MB ext2). Applicable contents of the LILO config file are shown. Note that the "/Image" is an SLS 1.01 version of the kernel that I'm booting, "/zImage" is the SLS 1.03 version I actually hacked on enough to manage a compile/load (but won't boot). boot = /dev/sda3 install = /etc/lilo/boot.b delay = 100 compact image = /Image root = /dev/sda3 vga = 3 label = 1 image = /zImage root = /dev/sda3 vga = normal label = 3 KERNEL SOURCES: I had been running SLS release 1.01 for a number of months (it took me a while after I first downloaded to get an US-24F boot image so I could install). At this point, I could rebuild the kernel without problems, using the 1.7 version of the UltraStor ALPHA code. Then, in August, I grabbed the SLS 1.03 files from TSX-11. I also grabbed the UltraStor sources and I found they would not compile. I *think* that the UltraStor sources that I got with 1.03 would also not compile. In any event, I tried again this weekend -- I grabbed new files off of TSX-11 yesterday: /pub/linux/sources/system/linux-0.99.13.tar.gz /pub/linux/ALPHA/scsi/ultrastor.tar.z /pub/linux/ALPHA/scsi/ultrastor-1.11d.tar.gz I re-installed them in the above order, and then ran: cd /linux make dep make clean make zImage The compile failed under the "linux/kernel/blk_drv/scsi" directory, with the following errors: gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c hosts.c -o hosts.o hosts.c:113: `MAX_SCSI_HOSTS' undeclared here (not in a function) hosts.c:113: conflicting types for `host_busy' hosts.h:182: previous declaration of `host_busy' hosts.c:114: `MAX_SCSI_HOSTS' undeclared here (not in a function) hosts.c:115: `MAX_SCSI_HOSTS' undeclared here (not in a function) hosts.c:115: conflicting types for `scsi_device_type' hosts.h:193: previous declaration of `scsi_device_type' hosts.c:116: `MAX_SCSI_HOSTS' undeclared here (not in a function) hosts.c:117: `MAX_SCSI_HOSTS' undeclared here (not in a function) hosts.c:117: conflicting types for `host_queue' hosts.h:189: previous declaration of `host_queue' hosts.c:118: `MAX_SCSI_HOSTS' undeclared here (not in a function) hosts.c:118: variable `host_wait' has initializer but incomplete type hosts.c:118: conflicting types for `host_wait' hosts.h:191: previous declaration of `host_wait' hosts.c:119: `MAX_SCSI_HOSTS' undeclared here (not in a function) hosts.c: In function `scsi_init': hosts.c:128: `MAX_SCSI_HOSTS' undeclared (first use this function) hosts.c:128: (Each undeclared identifier is reported only once hosts.c:128: for each function it appears in.) make[3]: *** [hosts.o] Error 1 gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c scsi.c -o scsi.o scsi.c: In function `scsi_times_out': scsi.c:433: warning: zero-length format string scsi.c:435: warning: zero-length format string scsi.c: In function `scsi_do_cmd': scsi.c:671: warning: zero-length format string scsi.c: In function `scsi_done': scsi.c:900: warning: format argument is not a pointer (arg 3) scsi.c:1061: warning: zero-length format string gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c scsi_ioctl.c -o scsi_ioctl.o gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c constants.c constants.c:11: warning: `reserved' defined but not used constants.c:13: warning: `vendor' defined but not used gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c st.c -o st.o gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c sd.c -o sd.o sd.c: In function `requeue_sd_request': sd.c:440: warning: zero-length format string sd.c: In function `sd_init_onedisk': sd.c:794: warning: too many arguments for format gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c sd_ioctl.c -o sd_ioctl.o gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c sr.c -o sr.o sr.c: In function `requeue_sr_request': sr.c:403: warning: zero-length format string gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c sr_ioctl.c -o sr_ioctl.o gcc -D__KERNEL__ -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer \ -m486 -c ultrastor.c -o ultrastor.o ultrastor.c: In function `ultrastor_14f_detect': ultrastor.c:377: warning: implicit declaration of func `check_region' ultrastor.c:440: warning: implicit declaration of func `snarf_region' ultrastor.c: In function `ultrastor_abort': ultrastor.c:927: warning: control reaches end of non-void function ultrastor.c: In function `ultrastor_reset': ultrastor.c:984: warning: control reaches end of non-void function ultrastor.c: In function `ultrastor_interrupt': ultrastor.c:1014: warning: unsigned int format, pointer arg (arg 2) ultrastor.c:1028: warning: unsigned int format, pointer arg (arg 4) ultrastor.c:1057: warning: unsigned int format, pointer arg (arg 3) ultrastor.c:1108: warning: unsigned int format, pointer arg (arg 3) make[3]: Target `scsi.a' not remade because of errors. At this point, I added the following line to "scsi.h": #define MAX_SCSI_HOSTS 10 And the compile completed successfully, but with the warnings shown above (implicit declaration & format stuff). So, I installed the new kernel "zImage", ran LILO install and tried to boot to this kernel. It died with the following errors on the screen: UltraStor 24F SCSI @ Slot 2 IRQ14 Unable to handle kernel paging request at address c0000002 Oops: 0000 EIP: 0010:00167d62 EFLAGS: 00010206 eax: 00167d61 ebx: 000000069 ecx: 00187ab8 edx: 000003d5 esi: 00000002 edi: 000171f24 ebp: 00000001 ds: 0018 es: 0018 fs: 0018 gs: 0018 Pid 0, process nr: 0 6f 6e 64 69 74 69 6f 6e 20 47 Task[0] (swapper) killed: unable to recover Kernel panic: trying to free up swapper memory space In swapper task - not syncing And, doing the "nm tools/zSystem | sort", I see the following pertinent information: 00167ca4 t constants.o 00167ca4 t gcc2_compiled. 00167cad t _unknown 00167cb5 t _vendor 00167cd4 t _print_opcode 00167cf4 T _print_command 00167db4 T _print_status 00167dd4 T _print_msg 00169584 T _print_sense 00169820 t ___gnu_compiled_c 00169820 t _bios_segment_table 00169820 t gcc2_compiled. 00169820 t ultrastor.o DIRTY DISKS Another problem that I see is with the root filesystem "CLEAN" flag. I'm using the EXT2 file system on a ~180MB partition for the root disk, plus I mount three FAT16 partitions, and use one swap partition. The problem is that whenever LINUX boots (*), I see a warning to the effect of this: warning, mounting dirty filesystem, running e2fsck is recommended Even when I've shutdown by running "halt", or even if I've run the commands: "cd / ; umount -a ; swapoff -a ; halt". The only way I've managed to get around this is to enter run-level 'S' first, then MANUALLY do a halt. And furthermore, "init s" within a script causes that script to be terminated -- so I cannot setup a simple way (for novices) to cleanly halt the system! Any help will be appreciated, (Response by e-mail is preferred), - Marvin Taylor ------------------------------ From: androut@ecf.toronto.edu (ANDROUTSOS DIMITRIOS) Subject: OS/2 and LINUX INSTALLATION Date: Sun, 7 Nov 1993 22:17:41 GMT I wish to install LINUX on my system but I am not sure how I can go about it since I have a boot manager for OS/2 installled. I have 2 IDE drives C and D. D is just a bunch of data so is of no consequence. Presently I have my C drive (245 MB) partitioned at 60 for OS/2 and the rest for DOS...I wish to repartition th eDOS partition to install Linux, however I want to also add the Linux partition on the Boot Manager. Could someone please give me some insight ...the FAQs are not very helpful when it comes to installing Linux on OS/2. Jim ------------------------------ Crossposted-To: comp.os.linux.help From: dke@maroon.tc.umn.edu () Subject: Re: WANTED: COBOL compiler Date: Sun, 7 Nov 1993 22:51:20 GMT In <1993Nov6.132311.5549@penrij.UUCP> soup@penrij.UUCP(John R. Campbell) writes: >>to think. It's in use worldwide in thousands of businesses, just like >>FORTRAN is in use in thousands of engineering, space, and scientific >>establishments. COBOL is generally the preferred language for business >>programming. >Granted. Of course, I've seen places where PL/I has replaced COBOL. PL/I replaceing cobol... Ick... Doing lots of business applecations, I wish i could use cobol as opposed to pascal quit frequently.... [ Lines deleted ] >For instance, the Automated Bonds System for the New York Bond Market >is written in FORTRAN rather than COBOL; the FORTRAN compiler for the >UNIVAC 1100 generated better code for the application and was more >easily written to process the (almost) free-form order messages. [ Lines Deleted ] >While COBOL is good at what it's good at :-) I don't see a long future. >Unfortunately, COBOL is _really_ designed for mainframe use and BATCH >mentality. The "terminal" specific stuff has used what, on a mainframe, >is the "operator's console" (i.e. ACCEPT xxx FROM CONSOLE). This makes >COBOL programs rather environment specific. I think this is the crux of the matter, Cobol has not kept up with the times like fortran has... if some adaptions could be made to it, it might find it self being used for more new applecations. overhauling the compiler to generate better code wouldn't hurt eather :) >For NEW programs, I'd discourage use of COBOL. For programs that already >exist that you want to maintain, I'd suggest retention of the existing >platform (like LPI COBOL on SCO UNIX/XENIX). I think this is good counsel.... -- David K Eggen | --Consultant at large-- dke@Staff.tc.umn.edu | You will be remembered only by the problems (612) 527 8152+| you solve and the problems you create! ------------------------------ ** 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 ******************************