Linux-Development-Sys Digest #715, Volume #6 Sat, 15 May 99 14:14:35 EDT
Contents:
Re: get client machine's IP-address (Iond Research Srl)
Re: A most peculiar bug! (Phil Howard)
Re: 2.2.8 - Evil behavior (Rodger Donaldson)
Strange behavior of SMP Linux 2.2.1 (Axel Farr)
Re: [ANN] CodeWarrior for Red Hat Linux, GNU ed. Shipping (M)
Re: Strange behavior of SMP Linux 2.2.1 (Axel Farr)
Re: 2.2.8 - Evil behavior (bryan)
Re: 2.2.8 - Evil behavior (bryan)
Re: USB Support (Christopher B. Browne)
Re: Destructive Erase? (Tony Smith)
Re: How to make linux boot/shutdown rapidly (Konrad Mieredorff)
Re: Strange behavior of SMP Linux 2.2.1 (David Fox)
Re: get client machine's IP-address (Craig Johnston)
Wie kann ich eine Minix Partition mounten? (root)
Re: Hostile Takeover of Linux (Marco Anglesio)
Re: 2.2.8 - Evil behavior ("G. Sumner Hayes")
----------------------------------------------------------------------------
From: Iond Research Srl <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development.apps,comp.programming,comp.unix.programmer,comp.os.linux.networking
Subject: Re: get client machine's IP-address
Date: Fri, 14 May 1999 18:13:49 +0200
Mark Hahn wrote:
> getpeername. this would be obvious if you had read even basic
> network programming books.
Thanks Mark for your answer, but I'm not sure getpeername can solve my
problem.
I'll try to explain myself better with an example:
You are connected from a linux machine (client) to a solaris machine
(server) with
a telnet session during which you want to execute this c-program
(complex.c)
main()
{
printf("Hi, Mark\n");
/* Now I wish to know the IP-address of the machine on which
the string "Hi, Mark" is displayed"
*/
}
This program runs in the server but thru the pseudo-tty its output is
displayed
at the client machine.
No sockets are used so we haven't the first arg to pass to the
getpeername.
You can try to pass 0 as socket-fd to getpeername but it returns the
local-host
IP-address (127.0.0.1).
So what's the matter ?
Maybe work around the slave part of the pseudo-tty ?
Sorry for this silly question but I'm not a network programmer, really
this is
my first networking problem, and also reading Stevens' I haven't yet
found a
solution.
Anyway, thanks a lot
Lucio
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Phil Howard)
Subject: Re: A most peculiar bug!
Date: Fri, 14 May 1999 17:28:56 GMT
On 14 May 1999 02:05:12 GMT Frampton Steve R ([EMAIL PROTECTED]) wrote:
| I have a Red Hat 5.2-based server running kernel 2.2.4 (custom kernel
| built from the kernel-source-2.2.4-4.i386.rpm file) on a P300 with an
| Adaptec AHA-131 SCSI controller, (using the aic7xxx driver that came
| with the 2.2.4 kernel package), driving four 9 Gb SCSI drives.
| Because this server is in a remote location, I usually do remote
| backups with tar to another disk partition, and then every so often
| move these to tape. The backup command I am using is:
|
| tar zcvpf /archive/backup-13-may-1999.tar.gz --exclude=archive
| --exclude=mnt --exclude=proc --exclude=var/spool/squid .
Have you considered using rsync? It can do compression, too, and
by skipping unchanged files (it can optionally check by checksum if
you want), reduce the network bandwidth (if that's an issue). What
you'd do is have the mirror disk where the tape drive is, and just
run rsync periodically to keep the mirror in sync. Then tar the
mirror to tape.
| This, in the past, has worked very well. However, the last couple
| of times I have done this, the tar process itself has hung while
| creating the archive. Even more strange, it is *impossible* to
| terminate the tar process, whether by <ctrl><c>, kill, or even
| with kill -9. Because I use "screen", I have to actually kill the
| screen processes and then use "screen -wipe" to clear out the sessions
| (note that this does *not* terminate the tar processes).
|
| A process list shows both tar processes, as well as a zombie gzip
| process (the first time, I tried it with the "z" (compress) parameter
| to tar, while the second time, I tried it without compression). While
| it looks like tar is still consuming some CPU time, the number gradually
| shrinks to 0 (the first time, gzip was consuming massive amounts of CPU
| time and took days to shrink to 0.00).
|
| root 24956 0.0 0.2 1048 664 ? D Apr 29 0:32 tar zcvpf /archive/bac
| root 30006 3.7 0.2 1048 660 a1 D 21:41 0:43 tar cvpf /archive/back
| root 24957 0.0 0.0 0 0 ? Z Apr 29 13:32 (gzip <zombie>)
It appears that the tar process is blocked in I/O that is not completing.
Look at it's /proc/30006 entry and see if you can figure out what file or
device it's currently reading. That might be where it's hung.
| What is very strange is that tar hangs at the very same location
| this time as it did last time!!! Here is a list of the last few
| files that it managed to add to the archive before hanging:
|
| usr/src/linux-2.2.4/Documentation/video4linux/radiotrack.txt
| usr/src/linux-2.2.4/Documentation/watchdog.txt
| usr/src/linux-2.2.4/Documentation/xterm-linux.xpm
| usr/src/linux-2.2.4/MAINTAINERS
| usr/src/linux-2.2.4/Makefile
| usr/src/linux-2.2.4/README
| usr/src/linux-2.2.4/Rules.make
| usr/src/linux-2.2.4/arch/
| usr/src/linux-2.2.4/arch/i386/
| usr/src/linux-2.2.4/arch/i386/Makefile
| usr/src/linux-2.2.4/arch/i386/boot/
| usr/src/linux-2.2.4/arch/i386/boot/Makefile
| usr/src/linux-2.2.4/arch/i386/boot/bootsect.S
| usr/src/linux-2.2.4/arch/i386/boot/compressed/
| usr/src/linux-2.2.4/arch/i386/boot/compressed/Makefile
| usr/src/linux-2.2.4/arch/i386/boot/compressed/head.S
| usr/src/linux-2.2.4/arch/i386/boot/compressed/misc.c
| usr/src/linux-2.2.4/arch/i386/boot/compressed/piggy.o
I would suggest a full fsck. There might be a bad block pointer.
Are you sure the tar file is not hitting the file size limit? Look
and see how big it is.
| I have never seen this behavior prior to kernel version 2.2.4 so I
| am wondering if something may have changed in the aic7xxx driver (I
| was using kernel version 2.0.36 before) that would cause this bug, or
| else something about the kernel itself is causing this hang.
That could be. If it is a driver bug doing it, the fsck should hang,
too. Try reading the raw partition:
dd if=/dev/whatever of=/dev/null bs=somepowerof2
| Obviously, I'm going to have to reboot to get rid of these processes
| as they absolutely refuse to be killed! This is strange and bothersome
| enough that I thought I'd mention it here.
It would bother me, too. See if it reproduces after reboot.
--
Phil Howard KA9WGN
[EMAIL PROTECTED] [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Rodger Donaldson)
Subject: Re: 2.2.8 - Evil behavior
Date: Sat, 15 May 1999 17:01:35 +1159
Reply-To: [EMAIL PROTECTED]
On Fri, 14 May 1999 23:12:37 GMT, bryan <[EMAIL PROTECTED]> wrote:
>2.2.7 is a loser on my system. the ps2 mouse is goofed up - when I
>startx, the video hangs and even gpm causes some kind of hang.
>
>drove me NUTS until I went back to 2.2.5 or so and the problem went away.
You're lucky. None of the 2.2.x kernels I've tried work on one of my UMC
486 PCI motherboards - the mouse fails on all of them. 2.0.x works
perfectly.
--
Rodger Donaldson [EMAIL PROTECTED]
Why is it I can find sex toys but no socks?
------------------------------
From: Axel Farr <[EMAIL PROTECTED]>
Subject: Strange behavior of SMP Linux 2.2.1
Date: Sat, 15 May 1999 12:35:13 +0200
Hello,
from time to time, I get a strange behaviour of my SMP machine running
Linux 2.2.1.
When I spawn 4 calculation tasks from within an master process (they
communicate with each other using PVM 3.4 beta6), then it happens that
the machine nearly locks up completely. It takes approximately 60
minutes to shut down the system, and you can watch "top" writing the
process status line-by-line. The machine runs slower than any 386 I have
ever seen.
But the system is still running !! It does not crash!
This does not happen everytimes, but it seems related to the fact that
the physical memory gets exhausted and the system has to swap out memory
pages. Tools like "top" or "xosview" show a system load of approximately
100%, where the user load is also around 100% (on a dual processor
machine, I normaly get around 190% user load and 10% system load for my
processes). The load is approx. 4, when the system is running in a
"normal" state, and around 5 or 6 when it is in the strange state I
described above.
To me, it seemes as if the two cpu's prevent each other from running
simultanously. Maybe one cpu is permanently locked, and only one
running. Is it possible that the paging algorithm in 2.2.1 has a bug
that causes the lock-up ?
Sometimes I get messages of the kind "eth0: something wicked happened:
000a / 000b" from the driver of the via-rhine-based 100MBit network
card. But I am not sure if that has something to do with the behaviour
of my machine.
The behaviour does not change when I switch to runlevel 1 (no network)
and the via-rhine.o gets unloaded.
My configuration:
2 x Pentium II - 400 MHz
256 MB RAM
2 x 10 GB IDE Hdd, with 3 Swap partitions on each drive
Via-Rhine-based 10/100 MBit card at a 100 MBit switch
Gigabyte GA686-BXD-Board
ATI XPert@Work graphics card
Suse 6.0
Linux Kernel 2.2.1 with SMP-Support
Greeting, Axel
------------------------------
From: M <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.unix.programmer
Subject: Re: [ANN] CodeWarrior for Red Hat Linux, GNU ed. Shipping
Date: Sat, 15 May 1999 23:09:17 +1000
[EMAIL PROTECTED] wrote:
>
> In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
> says...
> > > Personally, I still prefer a modern vi or emacs for productivity.
> > >
> > > --Sumner
> >
> > I agree. An ide slows me down, there's no way I would use one instead of
> > vi and emacs if I didn't have to - let alone pay for it. I'd rather have
> > the edge than look like I have it.
>
> Well if individual productivity rather than group productivity (in the
> greater professional domain) was the goal, then you'd have a leg to stand
> on.
>
> Ed
Yes, you're quite right, as are the above posters. I must say that I
think I was, how can I say...a little unreasonable at the time.
Michael
------------------------------
From: Axel Farr <[EMAIL PROTECTED]>
Subject: Re: Strange behavior of SMP Linux 2.2.1
Date: Sat, 15 May 1999 15:30:01 +0200
Some further notes on this problem:
When the calculation processes are finished (or killed), then the system
behaves "normal" again.
Axel
------------------------------
From: bryan <[EMAIL PROTECTED]>
Subject: Re: 2.2.8 - Evil behavior
Date: Sat, 15 May 1999 14:18:44 GMT
Brent Corbin <[EMAIL PROTECTED]> wrote:
: Strange... the ps2 mouse and startx both work fine on my 2.2.7 system...
let me elaborate. when the cpu switchbox I have (linksys - active
powered and buffered keybd/mouse/monitor switch) tries to switch to a
linux box running 2.2.7, it locks up (the switcher that is). only
happens with 2.2.7. weird, I know.
that box would also drop chars when I ran this little util that
flashes the keybd lites for tx/rx ether traffic. turns out it was too
much i/o thru that box and the little cpu inside couldn't keep up.
maybe something like that was happening when I switched linux boxes
with this switcher. don't know.
: On Fri, 14 May 1999 23:12:37 GMT, bryan <[EMAIL PROTECTED]> wrote:
: >2.2.7 is a loser on my system. the ps2 mouse is goofed up - when I
: >startx, the video hangs and even gpm causes some kind of hang.
: >
: >drove me NUTS until I went back to 2.2.5 or so and the problem went away.
: >
: >yes I built both kernels in the same way, and by me.
: >
: >--
: >Bryan
--
Bryan
------------------------------
From: bryan <[EMAIL PROTECTED]>
Subject: Re: 2.2.8 - Evil behavior
Date: Sat, 15 May 1999 14:19:29 GMT
Rodger Donaldson <[EMAIL PROTECTED]> wrote:
: On Fri, 14 May 1999 23:12:37 GMT, bryan <[EMAIL PROTECTED]> wrote:
: >2.2.7 is a loser on my system. the ps2 mouse is goofed up - when I
: >startx, the video hangs and even gpm causes some kind of hang.
: >
: >drove me NUTS until I went back to 2.2.5 or so and the problem went away.
: You're lucky. None of the 2.2.x kernels I've tried work on one of my UMC
: 486 PCI motherboards - the mouse fails on all of them. 2.0.x works
: perfectly.
its ps/2, also? weird. wonder what's different...
I'll try 2.2.9 and see if things got better.
--
Bryan
------------------------------
From: [EMAIL PROTECTED] (Christopher B. Browne)
Crossposted-To: redhat.general,comp.os.linux.misc
Subject: Re: USB Support
Reply-To: [EMAIL PROTECTED]
Date: Sat, 15 May 1999 14:25:16 GMT
On 15 May 1999 07:52:07 GMT, Michael Proto <[EMAIL PROTECTED]> posted:
>On Sat, 15 May 1999 01:33:54 GMT, [EMAIL PROTECTED] muttered:
>>USB-ZIP drives are a neat option; who knows when?
>
>IIRC, USB Zip drives *ARE* available. I've seen some that have a
>transparent-blue case that resembles a <shudder> iMAC.
Availability of *drives* is not the issue; I've seen them on store shelves.
The issue is the availability of a driver that knows how to talk to them.
*That* is the thing for which I wonder "when?"
--
Those who do not understand Unix are condemned to reinvent it, poorly.
-- Henry Spencer <http://www.hex.net/~cbbrowne/lsf.html>
[EMAIL PROTECTED] - "What have you contributed to free software today?..."
------------------------------
From: Tony Smith <[EMAIL PROTECTED]>
Subject: Re: Destructive Erase?
Date: Thu, 13 May 1999 19:10:08 +0100
Reply-To: [EMAIL PROTECTED]
If you want this to be quick, use mmap() to map the file into virtual memory,
and then you can use memset() or bcopy() to set the bytes to the values
suggested by Bill.
Tony
bill davidsen wrote:
> In article <nOEY2.549$[EMAIL PROTECTED]>,
> matthew gauthier <[EMAIL PROTECTED]> wrote:
> | I'm trying to implement a destructive file erase, however, I'm at a
> | loss as to the best way to insure that the same blocks are overwritten.
> | If I use stdio and overwrite a file with an equal amount of data and
> | close it, will the kernel put it on the same disk realestate at the
> | next sync? Or am I going to need to work at a much lower level here?
> |
> | Any thoughts and pointers on where to start looking are appreciated, as
> | are cc's, since my newsfeed sometimes drops articles.
>
> You open the file for read and write, seek to the end, get the current
> position (aka the length), seek to the beginning, and then write
> (length) bytes of zeros to the file. If you want security against a pro
> physically taking apart your disk and getting bits off under a
> microscope, write 0xFF, then zero, then 0xE5 to each byte, three passes
> with a sync (or fsync) at the end of each write.
>
> Look for a program called 'purge' in the old comp.sources.unix archives.
> I wrote it about 1991 and I believe I contributed it. I got the 0xE5
> from a security dude at DOE when I had clearance, don't ask why that
> value, other than many format programs use it for the initial low level
> format.
>
> --
> bill davidsen <[EMAIL PROTECTED]> CTO, TMR Associates, Inc
> One common problem is mistyping an email address and creating another
> valid, though unintended, recipient. Always check the recipient's
> address carefully when sending personal information, such as credit
> card numbers, death threats or offers of sexual services.
--
================================================================
Tony Smith
[EMAIL PROTECTED]
Reply address is corrupted to prevent spam. Remove
the spurious underscore to reply.
================================================================
------------------------------
From: Konrad Mieredorff <[EMAIL PROTECTED]>
Subject: Re: How to make linux boot/shutdown rapidly
Date: Sat, 15 May 1999 15:19:38 +0000
Michael Hirsch wrote:
>
> Michael Hirsch <[EMAIL PROTECTED]> writes:
>
> > I was thinking that one could write a kernel module that would imitate
> > my ThinkPad. When loaded it would immediately write the contents of
> > memory to a disk file, then shutdown the CPU. At boot time, you'd
> > need a new boot loader that would load the file back into memory, then
> > return execution to the appropriate place. Thus, shutdown would only
> > take as long as it takes to write RAM to disk (and if you are clever
> > you would only have to write the buffers that aren't just images of
> > one on the disk). Start up would take as long as it takes to read
> > them off disk.
>
> Sorry to followup to my own post, but I've since found a kernel patch
> that seems to do almost exactly what I was talking about:
>
> http://falcon.sch.bme.hu/~seasons/linux/swsusp.html
>
> Here's a quote from the homepage:
>
> Enable the possibilty of suspendig machine. It doesn't need APM. You
> may suspend your machine by either pressing Sysrq-d or with 'swsusp'
> or 'shutdown -z (patch for sysvinit needed). It creates an image which
> is saved in your active swaps. By the next booting the kernel detects
> the saved image, restores the memory from it and then it continues to
> run as before you've suspended. If you don't want the previous state
> to continue use the 'noresume' kernel option.
>
> I gotta try this.
I'm wondering if it also saves the hardware state and/or reinitializes
the hardware befor loading the image.
- Konrad Mierendorff
------------------------------
From: d s f o x @ c o g s c i . u c s d . e d u (David Fox)
Subject: Re: Strange behavior of SMP Linux 2.2.1
Date: 15 May 1999 07:57:09 -0700
Axel Farr <[EMAIL PROTECTED]> writes:
> Some further notes on this problem:
>
> When the calculation processes are finished (or killed), then the system
> behaves "normal" again.
The first thing I would do is see how it performs under 2.2.9.
--
David Fox http://hci.ucsd.edu/dsf xoF divaD
UCSD HCI Lab baL ICH DSCU
------------------------------
From: [EMAIL PROTECTED] (Craig Johnston)
Crossposted-To:
comp.os.linux.development.apps,comp.programming,comp.unix.programmer,comp.os.linux.networking
Subject: Re: get client machine's IP-address
Date: 15 May 1999 16:50:50 GMT
In article <[EMAIL PROTECTED]>, Iond Research Srl <[EMAIL PROTECTED]> wrote:
>
>I'll try to explain myself better with an example:
>
>You are connected from a linux machine (client) to a solaris machine
>(server) with
>a telnet session during which you want to execute this c-program
>(complex.c)
>
>main()
>{
> printf("Hi, Mark\n");
>/* Now I wish to know the IP-address of the machine on which
> the string "Hi, Mark" is displayed"
> */
>}
You'll have to grab the info from /var/wtmpx or the equivalent
on your system.
--
Craig Johnston
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (root)
Subject: Wie kann ich eine Minix Partition mounten?
Date: Sat, 15 May 1999 19:17:43 +0200
Hallo Alle!
Ich habe auf hda3 Minix installiert. Unter Linux probiere ich diese
Partition mit mount /dev/hda3 -t minix /minix zu mounten, aber ich
bekomme eine Fehlermeldung, dass es keinen Superblock auf dieser
Partition gibt.
Was kann ich tun?
Danke.
------------------------------
From: [EMAIL PROTECTED] (Marco Anglesio)
Subject: Re: Hostile Takeover of Linux
Reply-To: [EMAIL PROTECTED]
Date: Sat, 15 May 1999 17:24:11 GMT
On Fri, 14 May 1999 15:09:11 -0700, Tom Emerson <[EMAIL PROTECTED]> wrote:
>Marco Anglesio wrote in message ...
>To be honest, I don't know if folks from Sylvan proper "create" the tests or
>not, but I have heard [and I'll admit this is merely "rumor", but it fits
>well] that the tests are created by people with little or no actual/working
>knowledge of the material being tested -- they simply take "the text" used
>in the study course and look for what _appear_ to be "testable" points and
>that's what ends up on the exam.
I wouldn't think so; the people creating the test are the people that
create the study course for it, for obvious reasons. That doesn't mean
that the test is any good. It typically takes quite a bit of time for a
test to be validated. For example, MS and Lotus (for companies that I've
taken technical tests from) both put tests through a beta period where
they evaluate the validity and difficulty of questions to real-world
examinees. The College Board's process is even more involved, and rightly
so, given the importance of their exams. For examples of the most strongly
evaluated tests, think of the Weschler or Stanford-Binet "IQ" tests. And
even the Weschler or Stanford-Binet are still controversial (of course,
this has much more to do with what they're testing as what the test design
is).
>that people in the decision-making position about who gets hired or not tend
>to favor a "certificate" over "field experience" -- I'm not saying this is
>universal, but I think that was the original poster's point-of-view)
Well, that's entirely true. I think that the assumption in a test is that
you've gained the experience from practical work as well as study. You can
always "beat" that assumption by studying extra-hard (although this is
less and less feasible with broader question sets). More to the point, it
is up to the employer to figure out the correct balance between practical
experience and academic/theoretical experience. The point of the test is
to provide a basis for judging academic/theoretical experience.
>question -- when I asked which question I missed, they refused to answer.
With most computer-based tests, they "refuse" to answer because they can't
answer. Sylvan's test administrators are just that, test administrators;
they aren't experts on the test material.
>I think you've muddled the point in your last statement -- don't you mean
>that the difference in score between someone in the 50th vs. the 55th will
>be GREATER and therefore more reliable? (in other words, I think you have a
>doubly-negating phrase there with "is much less so")
Less variance rather than less score gap. I'm thinking of it like a
statistician (which is entirely legit; I studied statistics and psychology
at university, which leads directly to test design).
>Unfortunately, this is a relatively impractical method of testing...
Unless you're willing to pay several thousand dollars for each, relatively
limited, test. You'll probably also sacrifice test reliability. It's
hardly practical; in fact, it runs against the advantages of
computer-based testing (quick, cheap, widely available, and grades
available in rt rather than after a lengthy delay). Essay-type questions
aren't necessarily good ones, either; the rubric for grading said essay
has to be pretty standard, in which case you run into the same problems as
before.
marco
--
,--------------------------------------------------------------------------.
> Marco Anglesio | Love is a perky elf dancing a merry <
> [EMAIL PROTECTED] | little jig and then suddenly he turns <
> http://www.the-wire.com/~mpa | on you with a miniature machine gun. <
> | --Matt Groening, Life in Hell <
`--------------------------------------------------------------------------'
------------------------------
From: "G. Sumner Hayes" <[EMAIL PROTECTED]>
Subject: Re: 2.2.8 - Evil behavior
Date: Sat, 15 May 1999 13:46:18 -0400
bryan wrote:
>
> Rodger Donaldson <[EMAIL PROTECTED]> wrote:
> : You're lucky. None of the 2.2.x kernels I've tried work on one of
> : my UMC 486 PCI motherboards - the mouse fails on all of them. 2.0.x
> : works perfectly.
>
> its ps/2, also? weird. wonder what's different...
ps/2 mice in general work fine with 2.2.x. Linus has a ps/2 mouse,
so you can be sure that it'd get fixed quickly if it broke. Maybe
you have a bit of an odd ps/2 mouse? The driver did change somewhat
from 2.0.x -> 2.2.x.
Anything special about your mouse or keyboard?
--Sumner
------------------------------
** 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: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
