Linux-Development-Sys Digest #652, Volume #8 Wed, 18 Apr 01 15:13:19 EDT
Contents:
Re: How do I initiate Gnome? ("ray")
Problems to view the IP_RECVTOS (Thomas Pfeiffenberger)
Checkpointer for linux (Muzaffer Ozakca)
Re: Booting Linux on a robot (Jan Eric Andersson)
Re: killing a parent process also kills childern ? (Thanh-Lam NGUYEN)
Re: Booting Linux on a robot ("Philip Van Hoof")
modify driver of kernel 2.2.x to kernel 2.0.x ("MrHyde")
Syscall, please help (Mikael Chambon)
Re: Capturing raw ethernet frames (Grant Edwards)
ide-scsi problem mounting cd (Fred Kitner)
Re: IO system throughput (William Morris)
Re: Booting Linux on a robot (Eggert Ehmke)
Re: A Linux emulator for Linux, does this exist? (Philip Armstrong)
Re: IO system throughput (Greg Copeland)
Re: A Linux emulator for Linux, does this exist? ([EMAIL PROTECTED])
Re: killing a parent process also kills childern ? ("Victor Martinez")
Re: Booting Linux on a robot (Andre Kostur)
Re: A Linux emulator for Linux, does this exist? (Grant Edwards)
----------------------------------------------------------------------------
From: "ray" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x
Subject: Re: How do I initiate Gnome?
Date: Wed, 18 Apr 2001 16:06:57 +0800
Thanks for the long and informative reply.
I have one problem. I cannot find /etc/CORBA/servers in my directory.
Does all these sound like I did not install gnome services?
Thanks
I have tried the ~/.xinitrc entry but it seems to fail too.
------------------------------
Date: Wed, 18 Apr 2001 10:09:41 +0200
From: Thomas Pfeiffenberger <[EMAIL PROTECTED]>
Subject: Problems to view the IP_RECVTOS
Hello!
We are working on an IP tester.
I have problems to read the IP_RECVTOS for an incoming packet with
getsockopt (..,IP_RECVTOS,...). I always
get 0 for the tos.
Can somebody help me please?
Ciao
Thomas
--
_________________________________________
Dipl.Ing (FH)
Thomas Pfeiffenberger
Salzburg Research
Advanced Networking Center
Jakob-Haringer-Str. 5/III
A-5020 Salzburg/Austria
Tel. :++43.662.2288.444
Fax. :++43.662.2288.222
[EMAIL PROTECTED]
_________________________________________
------------------------------
From: Muzaffer Ozakca <[EMAIL PROTECTED]>
Subject: Checkpointer for linux
Date: Wed, 18 Apr 2001 04:30:49 -0400
Hi,
I'm planning to implement a checkpointer/restarter for Linux (probably
for 2.2 for the beginning) in the user level i.e. as a checkpointing
library. I'll be grateful if anybody could direct me with some ideas how
it could be better implemented (and yet be implemented). I made a survey
and there are libraries and systems such as libckpt, libckp and condor
of uwisc that does this (and do much more).
The checkpointer will be used as a base for some future intentions to
build a bigger system to do process migration and fault tolerance
research.
Any help, ideas and -pointers to- information are welcome.
(Excuse me if you've seen this message before, I had hard time posting...)
--
Muzaffer Ozakca
[EMAIL PROTECTED]
============================================================
Get your FREE web-based e-mail and newsgroup access at:
http://MailAndNews.com
Create a new mailbox, or access your existing IMAP4 or
POP3 mailbox from anywhere with just a web browser.
============================================================
------------------------------
From: Jan Eric Andersson <[EMAIL PROTECTED]>
Subject: Re: Booting Linux on a robot
Crossposted-To: alt.linux,alt.os.linux,alt.os.linux.mandrake,comp.os.linux.misc
Date: Wed, 18 Apr 2001 04:15:48 -0500
Jasmin Letendre wrote:
> I'm currently developping a six legged robots. The robot has an on-board
> conmputer and we want it to boot from a flash disk. We use Linux
> Mandrake. After long hour of trial and error, we managed to boot from
> the flash disk and have the prompt for a logging name.
>
> The problem is that any user name we enter (even root!) gives a "login
> invalid" without even asking for a password. And if we press
> Ctrl+Alt+Delete, it says "You don't exist. Go away!"
Interesting project!
But as to why you're getting this response, well, I've never run into
anything like it...
Have you tried booting in single user maintenance mode (runlevel 1 - type
"lilo init 1") to see if you can get in that way? It should give you a root
prompt without asking for a password.
Jan Eric
------------------------------
From: Thanh-Lam NGUYEN <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development
Subject: Re: killing a parent process also kills childern ?
Date: Wed, 18 Apr 2001 11:25:59 +0200
Idon't know how it really works. I needed to create a monitor on my box
(solaris 2.6). As soon as I quit the shell that started the monitor it was
killed, when I killed the shell my monitor was killed too. As soon as I ignore
the SIGUP I could kill the shell and my monitor was still running reparenting
itself to thread which pid is 1.
That's why I think that sigup is send to all running thread too.
I'm not sure that this will work on a linux box. As there also a nohup shell
command, that should work too.
Yoav Zach wrote:
> Thanks for the answer.
> As far as I can see - the SIGHUP signal is sent only to child processes that
> are STOPPED when the parent is killed. What about children that are running
> ?
>
> "Thanh-Lam NGUYEN" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > Yoav Zach wrote:
> >
> > > I wrote the simplest piece of code, in which a process creates a new one
> and
> > > they both wait for input from the keyboard. Now, when killing the parent
> > > process ( by sending it SIGTERM ), the child process is killed too. I
> can't
> > > figure out who kills it and why. Does anyone knows the answer ?
> >
> > When the parent process is killed, it sends a SIGUP to every child.
> > As there's no default handler for this signal, the child process kills
> itself.
> > In order to keep the child and not the parent, you'll have to handle this
> > signal.
> >
> >
> >
------------------------------
From: "Philip Van Hoof" <[EMAIL PROTECTED]>
Subject: Re: Booting Linux on a robot
Crossposted-To: comp.os.linux.misc
Reply-To: [EMAIL PROTECTED]
Date: Wed, 18 Apr 2001 10:56:49 GMT
Heh, try booting the kernel with the init=/bin/sh argument
--
Philip van Hoof aka freax (http://www.freax.eu.org)
irc: irc.openprojects.net mailto:freax @ pandora.be
------------------------------
From: "MrHyde" <[EMAIL PROTECTED]>
Subject: modify driver of kernel 2.2.x to kernel 2.0.x
Date: Wed, 18 Apr 2001 14:03:54 +0200
I'm trying to use a driver for a SiS 900 NIC on an
2.0.36 linux kernel. I can't upgrade the kernel so
it has to use a 2.0 kernel.
I've noticed the driver sis900.c only works on 2.2.x or
2.4.x kernel.
I need to modify this driver for works with the 2.0.x, so I don't know what
I must to do, and what are the diferences between 2.0.x drivers and 2..2
/2.4 drivers.
Thank you.
------------------------------
From: Mikael Chambon <[EMAIL PROTECTED]>
Subject: Syscall, please help
Date: Wed, 18 Apr 2001 14:45:50 +0200
Hi,
Is there way to forbid a syscall to be interrupted by another
syscall ???
If I send 3 syscalls to the same process at the same time, I assume that
it will be the ret_from_syscall of the third syscall that would be
executed at the first time? Am I right ?
Thanks for all,
--
Mikael Chambon
------------------------------
From: [EMAIL PROTECTED] (Grant Edwards)
Subject: Re: Capturing raw ethernet frames
Date: Wed, 18 Apr 2001 13:45:45 GMT
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] wrote:
>On Mon, 16 Apr 2001 20:26:06 GMT Grant Edwards <[EMAIL PROTECTED]> wrote:
>
>|>What if someone wanted to write their own protocol based
>|>on ethernet? How would it be done?
>|
>| You write a driver module that installs a handler for the
>| Ethernet protocol number in question.
>
>What if they want to do it in a process instead of the kernel,
>in much the same way a serial port protocol implemented in a
>process would start by opening the serial port device.
Then you create an AF_PACKET,SOCK_RAW socket using the
appropriate Ethernet protocol number.
--
Grant Edwards grante Yow! My pants just went to
at high school in the Carlsbad
visi.com Caverns!!!
------------------------------
From: Fred Kitner <[EMAIL PROTECTED]>
Subject: ide-scsi problem mounting cd
Reply-To: [EMAIL PROTECTED]
Date: Wed, 18 Apr 2001 10:34:56 -0400
Hi,
I hope someone can help me out here. I am running SuSe 7.1
with kernel 2.4.2
I compiled same and setup ide-scsi and scsi generic support.
I can burn cd's no problem but I can't mount the cdrom
I have tried many ways to mount the device.
I have tried just about all the devices in /dev like:
brw-r----- 1 root disk 11, 0 Jan 19 03:36 /dev/scd0
or
brw-r--r-- 1 root root 11, 0 Apr 17 10:08 /dev/scc0
there major , minor numbers are all the same .
However when i issue the command: mount /cdrecorder
i get the following message:
bigtime:~ # ln -s /dev/scd0 /dev/cdrecorder
bigtime:~ # mount /cdrecorder
mount: /dev/scd0: unknown device
Can anyone help me here? What device am i looking for? Do i need to do a
mknod and if so what device should it be?
Thanks in advance for any help. I am able to send more info if its
nessassary.
But i have included all the information that I know how to get.
fred
Please foward help to [EMAIL PROTECTED]
I have an entry in fstab as follows:
/dev/scd0 /cdrecorder auto ro,noauto,user,exec 0 0
>From dmesg:
****************************************************************
hda: WDC AC24300L, ATA DISK drive
hdb: ST34325A, ATA DISK drive
hdd: Hewlett-Packard CD-Writer Plus 8100, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: 8421840 sectors (4312 MB) w/256KiB Cache, CHS=524/255/63
hdb: 8405775 sectors (4304 MB) w/512KiB Cache, CHS=8895/15/63
Partition check:
hda: hda1 hda2
hdb: hdb1 hdb2 hdb3
Floppy drive(s): fd0 is 1.44M
****************************************************************
>From log after modprobe ide-scsi
****************************************************************
Apr 17 10:16:29 bigtime kernel: scsi0 : SCSI host adapter emulation for IDE
ATAPI devices
Apr 17 10:16:29 bigtime kernel: Vendor: HP Model: CD-Writer+ 8100
Rev: 1.0g
Apr 17 10:16:29 bigtime kernel: Type: CD-ROM
ANSI SCSI revision: 02
****************************************************************
>From cdrecord -scanbus
bigtime:~ # cdrecord -scanbus
Cdrecord 1.9 (i686-suse-linux) Copyright (C) 1995-2000 J�rg Schilling
Linux sg driver version: 3.1.17
Using libscg version 'schily-0.1'
scsibus0:
0,0,0 0) 'HP ' 'CD-Writer+ 8100 ' '1.0g' Removable CD-ROM
0,1,0 1) *
0,2,0 2) *
0,3,0 3) *
0,4,0 4) *
0,5,0 5) *
0,6,0 6) *
0,7,0 7) *
******************************************************************
>From cdrecord checkdrive
bigtime:/lib/modules/2.4.3/kernel/drivers/scsi # cdrecord dev=0,0,0
-checkdrive
Cdrecord 1.9 (i686-suse-linux) Copyright (C) 1995-2000 J�rg Schilling
scsidev: '0,0,0'
scsibus: 0 target: 0 lun: 0
Linux sg driver version: 3.1.17
Using libscg version 'schily-0.1'
Device type : Removable CD-ROM
Version : 0
Response Format: 1
Vendor_info : 'HP '
Identifikation : 'CD-Writer+ 8100 '
Revision : '1.0g'
Device seems to be: Generic mmc CD-RW.
Using generic SCSI-3/mmc CD-R driver (mmc_cdr).
Driver flags : SWABAUDIO
**********************************************************************
I tried changing fstab entry to /dev/sg0 but its a type c device and i get
not a block device , which of course it isent its a c type device.
>From scsi_info /dev/sg0
bigtime:/lib/modules/2.4.3/kernel/drivers/scsi # scsi_info /dev/sg0
SCSI_ID="0,0,0"
MODEL="HP CD-Writer+ 8100"
FW_REV="1.0g"
------------------------------
From: William Morris <[EMAIL PROTECTED]>
Subject: Re: IO system throughput
Date: Wed, 18 Apr 2001 15:08:22 GMT
Paul Repacholi <[EMAIL PROTECTED]> wrote:
> William Morris <[EMAIL PROTECTED]> writes:
> Are both PCI64s on one internal bus?
The system has a single bus with two 66MHz 64 bit slots and a number
of 33MHz 64 bit slots.
Regards
--
William Morris
[EMAIL PROTECTED]
------------------------------
From: Eggert Ehmke <[EMAIL PROTECTED]>
Crossposted-To: alt.linux,alt.os.linux,alt.os.linux.mandrake,comp.os.linux.misc
Subject: Re: Booting Linux on a robot
Date: Wed, 18 Apr 2001 18:51:24 +0200
Reply-To: [EMAIL PROTECTED]
On Wed, 18 Apr 2001 02:49:41 GMT, Jasmin Letendre
<[EMAIL PROTECTED]> wrote:
>The problem is that any user name we enter (even root!) gives a "login
>invalid" without even asking for a password. And if we press
>Ctrl+Alt+Delete, it says "You don't exist. Go away!"
I had this problem once when my /etc/passwd and /etc/shadow files were
messed up. Try to log in single user mode and reinstall them.
--
Eggert Ehmke
Email: [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Philip Armstrong)
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: A Linux emulator for Linux, does this exist?
Date: 18 Apr 2001 16:45:46 +0100
In article <[EMAIL PROTECTED]>,
Jonadab the Unsightly One <[EMAIL PROTECTED]> wrote:
>How much does a 390 cost?
How long is a piece of string ?
:)
I'm told IBM mainframe pricing is of the "turn them upside and shake
them until all the spare money falls out" variety, but having never
been in a position to actually want or need one of the beasts, I can't
speak from personal experience!
Phil
--
http://www.kantaka.co.uk/ .oOo. public key: http://www.kantaka.co.uk/gpg.txt
------------------------------
Subject: Re: IO system throughput
From: Greg Copeland <[EMAIL PROTECTED]>
Date: 18 Apr 2001 13:02:05 -0500
There was recently an article in Open mag., whereby, they seems to of
founds lots of issues with fiber controllers and Linux. It seems that
Linux currently has pretty poor support. I don't remember the other
items, but it seems there may be several issues here. The article that
I'm talking about was specifically a SAN article, but it addressed
trying to use fiber to talk to the SAN (using several approaches). At
any rate, you may want to look for it and see what you can find. I've
also read elsewhere that there seems to be some stability issues with
Linux's fiber drivers. I'm not sure if this is specific to fiber type
SAN implementations or not, but again, it may be worth looking into.
Greg
Paul Repacholi <[EMAIL PROTECTED]> writes:
> William Morris <[EMAIL PROTECTED]> writes:
>
> > We are trying to achieve a disk transfer rate of 90-100MB/s using a
> > pair of Qlogic ISP2100 cards each attached to a fibrechannel disk
> > array capable of around 90MB/s each. The required performance is
> > possible on NT (using just one disk) but on Linux is causing us
> > problems. We can achieve around 70MB/s from one disk but only 45MB/s
> > each when both disks are active at the same time.
>
> Seems like a PCI arbitration problem is the most likley.
>
> BTW, a real-time rule of thumb is that you need 10X performace on the
> raw iron to garentee X performance. You do not say what the system is.
> Are both PCI64s on one internal bus?
>
> --
> Paul Repacholi 1 Crescent Rd.,
> +61 (08) 9257-1001 Kalamunda.
> West Australia 6076
> Raw, Cooked or Well-done, it's all half baked.
--
Greg Copeland, Principal Consultant
Copeland Computer Consulting
==================================================
PGP/GPG Key at http://www.keyserver.net
DE5E 6F1D 0B51 6758 A5D7 7DFE D785 A386 BD11 4FCD
==================================================
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: A Linux emulator for Linux, does this exist?
Date: Wed, 18 Apr 2001 18:16:59 GMT
[EMAIL PROTECTED] (Philip Armstrong) writes:
> In article <[EMAIL PROTECTED]>,
> Jonadab the Unsightly One <[EMAIL PROTECTED]> wrote:
> >How much does a 390 cost?
>
> How long is a piece of string ?
>
> :)
> I'm told IBM mainframe pricing is of the "turn them upside and shake
> them until all the spare money falls out" variety, but having never
> been in a position to actually want or need one of the beasts, I
> can't speak from personal experience!
The expensive part is that you likely need a staff of a goodly dozen
people to manage the OS and hardware, a room with serious power and
air conditioning, and that doesn't buy you the hordes of
administrators needed to manage DB/2, CICS, and the applications.
In effect, you need to have a million dollars a year to spend on
staff, and hundreds of thousands to spend on real estate, and that's
_without_ IBM's participation in the "pocket shaking."
I expect that it doesn't make sense to build a Sysplex unless you're
planning to invest a goodly $10M USD just on hardware...
--
(concatenate 'string "cbbrowne" "@acm.org")
http://vip.hyperusa.com/~cbbrowne/resume.html
"People need to quit pretending they can invent THE interface and walk
away from it, like some Deist fantasy." -- Michael Peck
------------------------------
From: "Victor Martinez" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development
Subject: Re: killing a parent process also kills childern ?
Date: Wed, 18 Apr 2001 18:27:37 GMT
See manual page for setsid().
This will create a new process group.
If you are trying to run as a daemon, you need to also fork to detach from
your tty.
if ((pid=fork()) == -1)
{
/* fork error */
}
if (pid == 0)
{
/* child*/
if ( setsid() == -1 )
{
/* setsid error */
exit(1);
}
/* continue with child code */
}
else
{
/* parent*/
exit();
}
"Villy Kruse" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Wed, 18 Apr 2001 08:50:31 +0300, Yoav Zach <[EMAIL PROTECTED]> wrote:
> >Thanks for the answer, but I do not understand it -
> >What do you mean by 'inherited signals' ? What is inherited is the
> >signal's - not the signal itself. The sigterm sent to the parent is not
sent
> >to the child, that's for sure.
> >And anyway, what is this special action I need to take in order to
> >disinherit it ?
> >
>
>
>
> There is a difference between killing a job and killing by process id
> number. Killing a jobw will indeed send a signal to all members of that
> job, but killing the process id will only kill that process id, and only
> by catching the signal and propagate the signal to the child processes
> can the children be affected by that. The shell will quite often do that.
>
> A special case occurs if you kill a session leader, that is, a process
> with process id the same as the session id. I don't recall the exact
> rule in this case, though.
>
>
>
> Villy
------------------------------
Crossposted-To: alt.linux,alt.os.linux,alt.os.linux.mandrake,comp.os.linux.misc
Subject: Re: Booting Linux on a robot
From: [EMAIL PROTECTED] (Andre Kostur)
Date: Wed, 18 Apr 2001 18:31:10 GMT
Jasmin Letendre <[EMAIL PROTECTED]> wrote in
<[EMAIL PROTECTED]>:
>I'm currently developping a six legged robots. The robot has an on-board
>conmputer and we want it to boot from a flash disk. We use Linux
>Mandrake. After long hour of trial and error, we managed to boot from
>the flash disk and have the prompt for a logging name.
>
>The problem is that any user name we enter (even root!) gives a "login
>invalid" without even asking for a password. And if we press
>Ctrl+Alt+Delete, it says "You don't exist. Go away!"
>
>Does anybody has an idea about this problem?
>
>Thanks.
>
>Jasmin
>
You may be interested in looking at the Linux From Scratch project
(www.linuxfromscratch.org). Offhand it looks like login can't figure out
who you are. Which means either your /etc/passwd file is missing, or login
can't find it (your /etc/nsswitch.conf is missing or corrupted). Or things
are a little more complex if you're using pam (it's configuration files may
be missing or corrupted).
------------------------------
From: [EMAIL PROTECTED] (Grant Edwards)
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: A Linux emulator for Linux, does this exist?
Date: Wed, 18 Apr 2001 18:38:45 GMT
In article <9bkcra$5u0$[EMAIL PROTECTED]>, Philip Armstrong wrote:
>>How much does a 390 cost?
>
>How long is a piece of string ?
>
>:)
You've also got to remember we're talking about an equivalent
number of PC's running with 99.99% (or some such number)
uptime. By the time you add hot-swappable disk arrays,
redundant, hot-swappable power-supplies, hot-swappable ECC
memory, etc. a Wintel PC isn't all that cheap. Neither is the
24/7 support contract for when the automatic fail-over doesn't
work or the triple-point failure that "can't" happen does
happen. According to folklore, you can pretty much replace an
entire 390 one piece at a time without ever rebooting -- I
imagine that's a bit exagerated.
But, since the OP was talking about something to use in place
of two run-of-the-mill desktop machines, I'm pretty sure that a
390 isn't a financially viable solutions.
--
Grant Edwards grante Yow! QUIET!! I'm being
at CREATIVE!! Is it GREAT
visi.com yet? It's s'posed to SMOKEY
THE BEAR...
------------------------------
** 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 by posting to the
comp.os.linux.development.system newsgroup.
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
******************************
