From oliveira at fadom.br  Tue Sep  5 00:39:08 2006
From: oliveira at fadom.br (=?iso-8859-1?Q?Rodrigo_Oliveira_Ara=FAjo?=)
Date: Mon, 4 Sep 2006 19:39:08 -0300
Subject: [Udpcast] Problems with Sata
References: <mailman.0.1157408198.26265.udpcast@udpcast.linux.lu>
Message-ID: <000801c6d072$efb259d0$0565a8c0@si.lab>

Hello, i´m from a college at Brazil

I had some problems with a SATA driver and updcast. My Western Digital Sata 
doesn´t have support, what could I do ?

Sinceraly
Rodrigo Oliveira Araujo




From bdykstra at virtualwar.com  Tue Sep 12 04:53:25 2006
From: bdykstra at virtualwar.com (Brian Dykstra)
Date: Mon, 11 Sep 2006 22:53:25 -0400 (EDT)
Subject: [Udpcast] Suspected SATA problem with Toshiba Tecra M5
Message-ID: <2929.192.168.2.100.1158029605.squirrel@webmail.virtualwar.com>

Hi,

I've been using Udpcast for about a year know with from both CD and PXE
boot. It's been working great for us.  We've been imaging a bunch of old
IBM A31 Thinkpads for a training course from a high-end Toshiba Qosmio G35
laptop.  Everything has been gold.

Recently we were able to purchase a bunch of new Toshiba Tecra M5-S433
laptops with gigabit nics and 100GB SATA drives.  We were all psyched up
to image at lightspeed but then the pain started.

When we boot the systems from CD the network adapter is properly detected
as a Intel Pro/1000 using the e1000 driver.  It gets a DHCP address with
no problem.

Because the disks are SATA we load the optional SATA drivers.  The CD
detects the hardware and offers the ata_piix - SCSI low-level driver for
Intel PIIX/ICH ATA controllers.  This seems to match what we've found in
the Toshiba documentation and it sucessfully finds the hard disk.

The problem comes when we start imaging (sending or receiving).  After
about 1.5GB at roughly 130Gbps everything starts spirally downward until
eventually we are transferring data at about 14Gbps.  On a 100GB disk this
is painful.

We've swapped out as much stuff as possible (various switches, cables,
even used a PCMCIA GB network adapter in the Tecra M5).  We've done all
the normal stuff with capping the transfer rate at speeds down around
30-40Mbps but the transfer still spirals down to the teens.

Since we've swapped out all the server hardware (which works fine imaging
other laptops) and all the network gear we are left with the Tecra SATA
drive and the ata_piix driver.  We've tried loading other SATA drivers but
no luck.

The only interesting part of the output from the Kernel Message after
loading the driver looks like this:

<driver version stuff>
<6>PCI: Found IRQ 11 for device 0000:00:1f.2
<6>PCI: Sharing IRQ 11 with 0000:00:1d.1
<4>ata: 0x170 IDE port busy
<7>PCI: Setting latency timer of device 0000:00:1f.2 to 64

It then continues with hard disk configuration information.  The latency
timer bothers me.

I hope I've provided enough information to be useful.

Thanks
-- 
Brian Dykstra
bdykstra at virtualwar.com
http://www.virtualwar.com



From Gilles.Galboud at ec-lyon.fr  Tue Sep 12 08:45:26 2006
From: Gilles.Galboud at ec-lyon.fr (Gilles Galboud)
Date: Tue, 12 Sep 2006 08:45:26 +0200
Subject: [Udpcast] problem with cloning pc
Message-ID: <45065786.9000108@ec-lyon.fr>

Hi all,

I'm trying to cloning hd images, but it doesn't work with
two computers anymore ( sender+receiver).
What's going on ?

thanks

-- 

Gilles Galboud                          Email:Gilles.Galboud at ec-lyon.fr
Ecole Centrale de Lyon                  Phone:(4)72-18-61-02
Cegely - Batiment H9	                
36 Avenue Guy de Collongue
69134 Ecully cedex



From dmaclachlan at Hatsize.com  Tue Sep 12 21:19:20 2006
From: dmaclachlan at Hatsize.com (Darin MacLachlan)
Date: Tue, 12 Sep 2006 13:19:20 -0600
Subject: [Udpcast] Kernel rebuild
Message-ID: <4506B3EE.B22F.005C.0@Hatsize.com>

Hello,

I have been testing UDPCast to replace an existing Opforce server for imaging Windows Boxes.

I have been able to successfully run my tests in a VM Environment.  How I'm working on the actual physical boxes that the system will be using.  There are HP DL360 servers.  I've run across a problem where I'm unable to access the Hard drive connected to the HP SCSI Array.  I've found the proper block (it uses the cciss driver) in the modules folder from the kernel download.  I have been unsuccessful with getting the cciss driver to load in order to access the harddrive.  So i found in the archives a series of emails that discuss how to build a custom kernel.  I have been able to create the kernel with the SCSI drive enabled and the cciss block enabled as part of the kernel.  When I use the makeImage script, I alway run into an error where the script is building the modules from the old 2.6.16.18updcast folder and not the 2.6.16.20udpcast folder where the modules are located.  

I have had a look at the script, but cannot seen to find the area where the modules folder is specified.  I noticed that there is a -N option listed in the script ( to point to the modules folder), but this errors out as well.

Can you supply any insight or can you modify the Castomatic web site to allow this driver to be loaded as part of the kernel?

Regards,




Darin MacLachlan
System Engineer
Hatsize Learning Corporation






From kroe at rs-schesslitz.de  Tue Sep 12 22:51:40 2006
From: kroe at rs-schesslitz.de (Dieter Kroemer)
Date: Tue, 12 Sep 2006 22:51:40 +0200
Subject: [Udpcast] copy more than one device
Message-ID: <45071DDC.3030908@rs-schesslitz.de>

Hi,

is there a possibility to copy several devices instead of one.

I would like to copy /dev/hda1 /dev/hda2 and /dev/hda3 but not /dev/hda4.
In the online tool I can't find anything to do this.

Btw. I'm using udpcast in my school for nearly 100 Computers - it works
great, thanks :-))

Kind regards
Dieter




From breuer.jens at googlemail.com  Wed Sep 13 10:43:38 2006
From: breuer.jens at googlemail.com (Jens Breuer)
Date: Wed, 13 Sep 2006 10:43:38 +0200
Subject: [Udpcast] copy more than one device
In-Reply-To: <45071DDC.3030908@rs-schesslitz.de>
References: <45071DDC.3030908@rs-schesslitz.de>
Message-ID: <76ce20420609130143t531016aerc66b518b6a80be95@mail.gmail.com>

> Hi,

Hi Dieter,

> is there a possibility to copy several devices instead of one.
>
> I would like to copy /dev/hda1 /dev/hda2 and /dev/hda3 but not /dev/hda4.
> In the online tool I can't find anything to do this.

As far as I know you can use UDPCAST to broadcast/multicast exactly
one file/device.
I don't know which method you use for imaging the machines, but if you
use the PXE-method it should be easy to set up some kind of batch-job
that does the imaging in three steps. But that means that you would
need three initrd-images, one for each partition you want to image.

One different (and maybe more clever) thing that comes to mind is to
use the "Other commands to be executed after transfer"-option in
cast-o-matic to execute another two set-up-by-hand
udp-sender/udp-receiver processes that will be responsible for
/dev/hda2 and /dev/hda3.

Anyway, you will end up with 3 image files - one for each partition.

Please keep us informed on how you solved your issue. :-)

> Kind regards
> Dieter

Kind regards and greetings from Saarland
Jens


From Maciej.Nadolski at cs.put.poznan.pl  Wed Sep 13 11:04:30 2006
From: Maciej.Nadolski at cs.put.poznan.pl (Maciej Nadolski)
Date: Wed, 13 Sep 2006 11:04:30 +0200
Subject: [Udpcast] Executing command after transfer / NO cast-o-matic
Message-ID: <4507C99E.10402@cs.put.poznan.pl>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I'm wondering how to make udpcast to execute certain commands after
transfer is completed. I know that it can be achieved using
cast-o-matic, but I'm using commandline version of udpcast. Is it
possible to do just something like that? :
udp-sender --parm1 --parm2 --file /foo/bar ; NextCommand
or
udp-receiver --parm1 --parm2 --file /foo/bar ; NextCommand

Thanks for help.

- --
Maciej Nadolski
Address: Poznań, Berdychowo 2, Room 42
Phone: +48616652951
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.4 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFB8mef7ObEm2Z9hARApIpAJ942ejGKbYF3WMPALgMaKYxytokqgCeNj0c
1NEYqDgX2eE56/WfMyItSoI=
=hxMk
-----END PGP SIGNATURE-----


From alain at knaff.lu  Wed Sep 13 12:48:14 2006
From: alain at knaff.lu (Alain Knaff)
Date: Wed, 13 Sep 2006 12:48:14 +0200
Subject: [Udpcast] Suspected SATA problem with Toshiba Tecra M5
In-Reply-To: <2929.192.168.2.100.1158029605.squirrel@webmail.virtualwar.com>
References: <2929.192.168.2.100.1158029605.squirrel@webmail.virtualwar.com>
Message-ID: <4507E1EE.6020006@knaff.lu>

Brian Dykstra wrote:
> The problem comes when we start imaging (sending or receiving).  After
> about 1.5GB at roughly 130Gbps everything starts spirally downward until

I think you mean Mbps here, or else you have a very fast network and 
disk there ;-)

> eventually we are transferring data at about 14Gbps.  On a 100GB disk this
> is painful.

The bandwidth displayed is the network bandwidth needed by udpcast, and 
if you have chosen compressed transfer, udpcast will need to transfer 
much less data once it hits the very compressible areas of the hard 
disks which are empty.

In that case, the disk speed, not the network speed, become the limiting 
factor (because the program read the *un*compressed data from the disk 
(needing a high bandwidth), but transfers compressed data over the 
network (needing a lower bandwidth).

The uncompressed amount of data is still displayed at the end of the 
status line. If this moves normally, you don't need to worry.

Regards,

Alain


From alain at knaff.lu  Wed Sep 13 13:01:14 2006
From: alain at knaff.lu (Alain Knaff)
Date: Wed, 13 Sep 2006 13:01:14 +0200
Subject: [Udpcast] Kernel rebuild
In-Reply-To: <4506B3EE.B22F.005C.0@Hatsize.com>
References: <4506B3EE.B22F.005C.0@Hatsize.com>
Message-ID: <4507E4FA.5020108@knaff.lu>

Darin MacLachlan wrote:
> [...] When I use the makeImage script, I alway run into an [LF by AK]
> error where the script is building the modules from the [LF by AK]
> old 2.6.16.18updcast folder and not the 2.6.16.20udpcast folder [LF by AK]
> where the modules are located.  

You can pick the kernel using the -k flag.
For example, if you prefer the 2.6.16.20 kernel, do:

makeImage -k /boot/vmlinuz-2.6.16.20udpcast ...


N.B. The script is *not* building the modules, it rather expects them to 
already be present in /lib/modules/2.6.16.20udpcast beforehand.
You can do that by calling "make modules_install" in your kernel source 
directory before calling makeImage

Regards,

Alain

P.S. When sending mails, please watch your line lengths. Most people 
read their mails in a 80 character-wide window, and using longer lines 
makes it inconvenient for them to read your mails or reply to them. Thanks


From alain at knaff.lu  Wed Sep 13 13:02:46 2006
From: alain at knaff.lu (Alain Knaff)
Date: Wed, 13 Sep 2006 13:02:46 +0200
Subject: [Udpcast] problem with cloning pc
In-Reply-To: <45065786.9000108@ec-lyon.fr>
References: <45065786.9000108@ec-lyon.fr>
Message-ID: <4507E556.1070209@knaff.lu>

Gilles Galboud wrote:
> Hi all,
> 
> I'm trying to cloning hd images, but it doesn't work with
> two computers anymore ( sender+receiver).
> What's going on ?
> 
> thanks

Could you describe what you see (error messages, if it hangs somewhere, 
how far does it get, etc.).

Regards,

Alain



From dmaclachlan at Hatsize.com  Wed Sep 13 15:41:51 2006
From: dmaclachlan at Hatsize.com (Darin MacLachlan)
Date: Wed, 13 Sep 2006 07:41:51 -0600
Subject: [Udpcast] Kernel rebuild
In-Reply-To: <b3c053c463a8850fc4873b956b98c16546754ac5@localhost>
References: <4506B3EE.B22F.005C.0@Hatsize.com>
	<b3c053c463a8850fc4873b956b98c16546754ac5@localhost>
Message-ID: <4507B656.B22F.005C.0@Hatsize.com>

Thanks for the response,
I found the problem, I was using the old version of the linux file so it was looking for the old file structure for the module drivers.

I'm still fighting with trying to get the kernel to load the cciss scsi module, not as a module but part of the kernel.  The HP systems need the cciss as part of the starting kernel.  

Regards,



Darin MacLachlan
System Engineer
Hatsize Learning Corporation



>>> 
From: 	Alain Knaff <alain at knaff.lu>
To:	Darin MacLachlan <dmaclachlan at Hatsize.com>
Date: 	09/13/06 5:01 AM
Subject: 	Re: [Udpcast] Kernel rebuild

CC:	<udpcast at udpcast.linux.lu>
Darin MacLachlan wrote:
> [...] When I use the makeImage script, I alway run into an [LF by AK]
> error where the script is building the modules from the [LF by AK]
> old 2.6.16.18updcast folder and not the 2.6.16.20udpcast folder [LF by AK]
> where the modules are located.  

You can pick the kernel using the - k flag.
For example, if you prefer the 2.6.16.20 kernel, do:

makeImage - k /boot/vmlinuz- 2.6.16.20udpcast ...


N.B. The script is *not* building the modules, it rather expects them to 
already be present in /lib/modules/2.6.16.20udpcast beforehand.
You can do that by calling "make modules_install" in your kernel source 
directory before calling makeImage

Regards,

Alain

P.S. When sending mails, please watch your line lengths. Most people 
read their mails in a 80 character- wide window, and using longer lines 
makes it inconvenient for them to read your mails or reply to them. Thanks




From alain at knaff.lu  Wed Sep 13 17:22:19 2006
From: alain at knaff.lu (Alain Knaff)
Date: Wed, 13 Sep 2006 17:22:19 +0200
Subject: [Udpcast] Problems with Sata
In-Reply-To: <000801c6d072$efb259d0$0565a8c0@si.lab>
References: <mailman.0.1157408198.26265.udpcast@udpcast.linux.lu>
	<000801c6d072$efb259d0$0565a8c0@si.lab>
Message-ID: <4508222B.8050507@knaff.lu>

Rodrigo Oliveira Araújo wrote:
> Hello, i´m from a college at Brazil
> 
> I had some problems with a SATA driver and updcast. My Western Digital Sata 
> doesn´t have support, what could I do ?
> 
> Sinceraly
> Rodrigo Oliveira Araujo

I pretty sure can tell you that it is *not* an issue with your hard disk 
(Western Digital), but rather with the hard disk controller (usually 
integrated in the motherboard). Indeed, the SATA protocol is 
standardized, so it should work for all disks.

However, the I/O API of SATA controllers is not standardized. On some 
motherboards you can change a BIOS setting to make it work. Look for 
"SATA support", and try changing it from "native" to "legacy" or vice-versa.

On some boards, both settings work, but legacy is much slower. I've also 
heard of a case where initially SATA didn't work (it was set to native), 
but where changing it to "legacy" and then *back* to "native" suddenly 
made it work...

Regards,

Alain



From alain at knaff.lu  Thu Sep 14 09:32:10 2006
From: alain at knaff.lu (Alain Knaff)
Date: Thu, 14 Sep 2006 09:32:10 +0200
Subject: [Udpcast] Executing command after transfer / NO cast-o-matic
In-Reply-To: <4507C99E.10402@cs.put.poznan.pl>
References: <4507C99E.10402@cs.put.poznan.pl>
Message-ID: <4509057A.5040001@knaff.lu>

Maciej Nadolski wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi,
> 
> I'm wondering how to make udpcast to execute certain commands after
> transfer is completed. I know that it can be achieved using
> cast-o-matic, but I'm using commandline version of udpcast. Is it
> possible to do just something like that? :
> udp-sender --parm1 --parm2 --file /foo/bar ; NextCommand
> or
> udp-receiver --parm1 --parm2 --file /foo/bar ; NextCommand
> 
> Thanks for help.

Yes, this should work.

Regards,

Alain


From alain at knaff.lu  Thu Sep 14 10:33:27 2006
From: alain at knaff.lu (Alain Knaff)
Date: Thu, 14 Sep 2006 10:33:27 +0200
Subject: [Udpcast] pxe images error
In-Reply-To: <20060717180424.M88704@cepedi.org.br>
References: <20060717180424.M88704@cepedi.org.br>
Message-ID: <450913D7.2070001@knaff.lu>

Sara Nery wrote:
> Hi all, 
> 
> I'm testing the pxe images on site to clone images, but it is bringing me this
> error on the receiver machine:
> 
> "Warning: unable to open an inital console.
> Kernel panic - not syncing: No init found. Try passing init=option to kernel.
> "
> I'm using the same initrd and linux kernel on site.
> 
> Sometime ago, I had the same error in a minilinux project I was working on.
> The problem was that I was using the busybox package instead of
> busybox-static. The busybox needs the glibc and the busyboox-static does not
> need. I wonder if this is the same error. Have anybody here passed the same
> problem?May somebody help me with this issue?
> 
> Regards,
> 
> Sara Nery
> 
> ps - sorry for my poor english

Yes, the busybox shipped with udpcast is indeed statically linked. A 
dynamic version does not work, unless you also include glibc into your 
image.

If for whatever reason you want to compile your own busybox, you can 
start with the patch at http://udpcast.linux.lu/20060619/busybox.diff.gz 
. This contains not only the changes to integrate udpcast and lzop into 
busybox, but also two configuration files config.mini and config.maxi, 
which specify all configuration choices, including static compilation.

Regards,

Alain



From dmaclachlan at Hatsize.com  Thu Sep 14 19:12:06 2006
From: dmaclachlan at Hatsize.com (Darin MacLachlan)
Date: Thu, 14 Sep 2006 11:12:06 -0600
Subject: [Udpcast] HPDL360 and CCISS Kernel Module
Message-ID: <4509391E.B22F.005C.0@Hatsize.com>

I've been struggling with a problem that originally i thought was due to the hardware and the driver, but not sure now.

I have compiled the kernel on the target machine with the latest CCISS driver preloaded in the kernel as the device needs to be loaded at boot.  I then create an new INITRD using the makeImage script.  Again all goes well.  I transfer the needed files to the tftp server and boot the test server.  The kernel download, the initrd is run, all appears to be going well.  The menu system loads and i step through the option all the way to the "Devices to be copied" screen.  The is nothing displayed there.  I access the shell (built with --fullbox option) and in the dev folder I have cciss!c0d0 not what I was expecting (cciss/c0d0).  The dmesg reports:
Compaq SMART2 Driver (v 2.6.0)
HP CISS Driver (v 2.6.10)
cciss: Device 0xb178 has been found at bus 0 dev 4 func 0
cciss: using DAC cycles
blocks= 71122560 block_size= 512
heads= 255, sectors= 32, cylinders= 8716

blocks= 71122560, block_size= 512
heads= 255, sectors= 32, cylinders= 8716

cciss/c0d0: p1 p2
Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
.....

as the dmesg shows, the cciss module is being loaded.  When viewed using the shell, there appears to be a mix up with the displaying of the folder.

Any assistance would be appreciated.  I would like to be able to use UDPCast to replace an existing Opforce server.  

Regards,



Darin MacLachlan
System Engineer
Hatsize Learning Corporation






From Maciej.Nadolski at cs.put.poznan.pl  Fri Sep 15 10:33:44 2006
From: Maciej.Nadolski at cs.put.poznan.pl (Maciej Nadolski)
Date: Fri, 15 Sep 2006 10:33:44 +0200
Subject: [Udpcast] Udpcast + firewall
Message-ID: <450A6568.4050702@cs.put.poznan.pl>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,
I'm aware of using udpcast threw firewall, certainly isn't a best
practice, but I'm sending images from firewalled server, to pcs in
computer laboratory.
Server is ofcourse firewalled. I opened ports 9000 and 9001 in firewall
 for inbound udp connections. All outbound connections are allowed. I'm
using OpenSuse 10.1, but it shouldn't be the case.
udp-receiver or udp-sender on server, always have to be executed at the
end. If it isn't done like that, the sender and receiver doesen't find
themselves threw broadcast.
Firewall is blocking something for sure, but I have no idea, what. Could
you tell me what rule in fw I missed?
Thanks.

- --
Maciej Nadolski
Address: Poznań, Berdychowo 2, Room 42
Phone: +48616652951
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.4 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFCmVof7ObEm2Z9hARAjLeAJ0T0Mh3wP39JQ2bRsoH4kOQ1v58igCbBTPM
v2rzh6KdVtTVWsch622FUQM=
=+u3c
-----END PGP SIGNATURE-----



From alain at knaff.lu  Fri Sep 15 13:44:38 2006
From: alain at knaff.lu (Alain Knaff)
Date: Fri, 15 Sep 2006 13:44:38 +0200
Subject: [Udpcast] HPDL360 and CCISS Kernel Module
In-Reply-To: <4509391E.B22F.005C.0@Hatsize.com>
References: <4509391E.B22F.005C.0@Hatsize.com>
Message-ID: <450A9226.5030203@knaff.lu>

Darin MacLachlan wrote:
> I've been struggling with a problem that originally i thought was due to the hardware and the driver, but not sure now.
> 
> I have compiled the kernel on the target machine with the latest CCISS driver preloaded in the kernel as the device needs to be loaded at boot.  I then create an new INITRD using the makeImage script.  Again all goes well.  I transfer the needed files to the tftp server and boot the test server.  The kernel download, the initrd is run, all appears to be going well.  The menu system loads and i step through the option all the way to the "Devices to be copied" screen.  The is nothing displayed there.  I access the shell (built with --fullbox option) and in the dev folder I have cciss!c0d0 not what I was expecting (cciss/c0d0).  The dmesg reports:
> Compaq SMART2 Driver (v 2.6.0)
> HP CISS Driver (v 2.6.10)
> cciss: Device 0xb178 has been found at bus 0 dev 4 func 0
> cciss: using DAC cycles
> blocks= 71122560 block_size= 512
> heads= 255, sectors= 32, cylinders= 8716
> 
> blocks= 71122560, block_size= 512
> heads= 255, sectors= 32, cylinders= 8716
> 
> cciss/c0d0: p1 p2
> Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
> .....
> 
> as the dmesg shows, the cciss module is being loaded.  When viewed using the shell, there appears to be a mix up with the displaying of the folder.
> 
> Any assistance would be appreciated.  I would like to be able to use UDPCast to replace an existing Opforce server.  
> 
> Regards,
> 
> 
> 
> Darin MacLachlan
> System Engineer
> Hatsize Learning Corporation

And what happens if you just chose the "badly named" cciss!c0d0 device? 
Does it work anyways, or do you get an error?

Thanks,

Alain



From alain at knaff.lu  Fri Sep 15 13:47:21 2006
From: alain at knaff.lu (Alain Knaff)
Date: Fri, 15 Sep 2006 13:47:21 +0200
Subject: [Udpcast] Udpcast + firewall
In-Reply-To: <450A6568.4050702@cs.put.poznan.pl>
References: <450A6568.4050702@cs.put.poznan.pl>
Message-ID: <450A92C9.1060903@knaff.lu>

Maciej Nadolski wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi,
> I'm aware of using udpcast threw firewall, certainly isn't a best
> practice, but I'm sending images from firewalled server, to pcs in
> computer laboratory.
> Server is ofcourse firewalled. I opened ports 9000 and 9001 in firewall
>  for inbound udp connections. All outbound connections are allowed. I'm
> using OpenSuse 10.1, but it shouldn't be the case.
> udp-receiver or udp-sender on server, always have to be executed at the
> end. If it isn't done like that, the sender and receiver doesen't find
> themselves threw broadcast.
> Firewall is blocking something for sure, but I have no idea, what. Could
> you tell me what rule in fw I missed?
> Thanks.

Port 9000 and 9001 in both directions should be enough.

Try to set up your firewall to log all rejected packets (-j LOG target), 
and check what is blocked.

Also, make sure that it is indeed a firewalling issue (i.e. does it work 
with firewall off?), and not something else, like a mismatched broadcast 
address or subnet mask on senders and receivers.

Regards,

Alain


From dmaclachlan at Hatsize.com  Fri Sep 15 15:34:26 2006
From: dmaclachlan at Hatsize.com (Darin MacLachlan)
Date: Fri, 15 Sep 2006 07:34:26 -0600
Subject: [Udpcast] HPDL360 and CCISS Kernel Module
In-Reply-To: <9db64f8240a8f95efce29e3044f13ac79c6ec3be@localhost>
References: <4509391E.B22F.005C.0@Hatsize.com>
	<9db64f8240a8f95efce29e3044f13ac79c6ec3be@localhost>
Message-ID: <450A579C.B22F.005C.0@Hatsize.com>

Sorry, I should have added that,
If you use the cciss!c0d0 and add it manually to the device to use option box, it always returns an error message << File "/dev/cciss!c0d" not found (No such file or directory) >>




Darin MacLachlan
System Engineer
Hatsize Learning Corporation


>>> 
From: 	Alain Knaff <alain at knaff.lu>
To:	Darin MacLachlan <dmaclachlan at Hatsize.com>
Date: 	09/15/06 5:45 AM
Subject: 	Re: [Udpcast] HPDL360 and CCISS Kernel Module

CC:	<udpcast at udpcast.linux.lu>
Darin MacLachlan wrote:
> I've been struggling with a problem that originally i thought was due to the hardware and the driver, but not sure now.
> 
> I have compiled the kernel on the target machine with the latest CCISS driver preloaded in the kernel as the device needs to be loaded at boot.  I then create an new INITRD using the makeImage script.  Again all goes well.  I transfer the needed files to the tftp server and boot the test server.  The kernel download, the initrd is run, all appears to be going well.  The menu system loads and i step through the option all the way to the "Devices to be copied" screen.  The is nothing displayed there.  I access the shell (built with -- fullbox option) and in the dev folder I have cciss!c0d0 not what I was expecting (cciss/c0d0).  The dmesg reports:
> Compaq SMART2 Driver (v 2.6.0)
> HP CISS Driver (v 2.6.10)
> cciss: Device 0xb178 has been found at bus 0 dev 4 func 0
> cciss: using DAC cycles
> blocks= 71122560 block_size= 512
> heads= 255, sectors= 32, cylinders= 8716
> 
> blocks= 71122560, block_size= 512
> heads= 255, sectors= 32, cylinders= 8716
> 
> cciss/c0d0: p1 p2
> Uniform Multi- Platform E- IDE driver Revision: 7.00alpha2
> .....
> 
> as the dmesg shows, the cciss module is being loaded.  When viewed using the shell, there appears to be a mix up with the displaying of the folder.
> 
> Any assistance would be appreciated.  I would like to be able to use UDPCast to replace an existing Opforce server.  
> 
> Regards,
> 
> 
> 
> Darin MacLachlan
> System Engineer
> Hatsize Learning Corporation

And what happens if you just chose the "badly named" cciss!c0d0 device? 
Does it work anyways, or do you get an error?

Thanks,

Alain





From alain at knaff.lu  Sun Sep 17 22:51:23 2006
From: alain at knaff.lu (Alain Knaff)
Date: Sun, 17 Sep 2006 22:51:23 +0200
Subject: [Udpcast] Windows Version corrupt on Website?
In-Reply-To: <504930132@web.de>
References: <504930132@web.de>
Message-ID: <450DB54B.3080807@knaff.lu>

Bastian Stock wrote:
> Hello there,
> 
> I hope I get no dubble Post :-) If so "SORRY" (I am new in Using Mailling Lists)
> 
> today I Downloaded the latest Windows Version of udpcast from the Website.
> When I start the Programm on a PC with only one Network Interface it works
> fine (both sender and receiver).
> 
> But when I start it on the same PC (a VMware) with two Network Interfaces
> the Programm crashes immediately. I tryed to bind the Interface with: 
> "--interface 192.168.1.2" (Yes the IP is the correct) but this also dosent work.
> 
> The only Message that appears is: "Udp-sender 2006-06-19"
> and for the receiver: "Udp-receiver 2006-03-25"
> 
> If the Windows Error Reporting Service is active it will send a Error Report to MS.
> 
> I hope this is only a small Error...
> 
> When it works it is a really cool Programm!
> 
> ----------------------------------
> Greetings from Germany
> 
> Bastian Stock

This is fixed in today's version (was due to a buffer overrun... Windows 
seems to need *huge* amounts of memory to store device information, and 
2 ethernet cards were just too much... sigh!)

Regards,

Alain



From alain at knaff.lu  Sun Sep 17 22:53:27 2006
From: alain at knaff.lu (Alain Knaff)
Date: Sun, 17 Sep 2006 22:53:27 +0200
Subject: [Udpcast] typo in "busyboxUsage" ?
In-Reply-To: <44F705CD.3010503@bio.umass.edu>
References: <44F705CD.3010503@bio.umass.edu>
Message-ID: <450DB5C7.1000704@knaff.lu>

Tom Carpenter wrote:
> The "busyboxUsage" file in "udpcast-20060619.tar.gz" contains
> the following line:
> 
> "\t--slice-sice\tinitial slice size\n" \
> 
> is "sice" a typo, i.e. should it read "size"?
> 
> -Tom Carpenter

Indeed, it was a typo.
Fixed in today's version

Regards,

Alain



From alain at knaff.lu  Sun Sep 17 22:56:41 2006
From: alain at knaff.lu (Alain Knaff)
Date: Sun, 17 Sep 2006 22:56:41 +0200
Subject: [Udpcast] Finding the slowpoke machine when ID numbers scroll
 away
In-Reply-To: <20060726030726.6ogrbfldwk0wc8wg@www.lll.lu>
References: <Pine.LNX.4.64.0607141039390.32257@artistic.localdomain.domain>	<44B94D5E.9090303@knaff.lu>	<Pine.LNX.4.64.0607250911390.20260@artistic.localdomain.domain>
	<20060726030726.6ogrbfldwk0wc8wg@www.lll.lu>
Message-ID: <450DB689.4070102@knaff.lu>

Alain.Knaff at lll.lu wrote:
> Quoting D G Teed <dteed at artistic.ca>:
>> We had initially used the --rexmit-hello-interval flag due to
>> older hardware that wasn't always connecting on the first
>> attempt.  If you think it could be related to multiple
>> "press any key to start" prompts, we can peel it off.
> 
> Yes, as this option's purpose is to attempt multiple connections (in 
> case the first one is missed), it definately causes multiple connection 
> messages. If you drop this option, you should only get a single 
> connection message.
> 
> Alain

In today's version, I changed it in such a way that the prompt is only 
printed once, even with --rexmit-hello-interval

Regards,

Alain



From alain at knaff.lu  Sun Sep 17 23:12:36 2006
From: alain at knaff.lu (Alain Knaff)
Date: Sun, 17 Sep 2006 23:12:36 +0200
Subject: [Udpcast] Compiling udpcast on NetBSD 3.0
In-Reply-To: <53216.203.11.167.254.1152689452.squirrel@webmail.exetel.com.au>
References: <53216.203.11.167.254.1152689452.squirrel@webmail.exetel.com.au>
Message-ID: <450DBA44.3000805@knaff.lu>

elekktretterr at exemail.com.au wrote:
[...]
> Stop.
> make: stopped in /usr/local/src/udpcast-20060619
> 
> Thanks to anyone who has a look at this.
> 
> Cheers.
> 
> Petr

Could you retry with today's version? This is now autoconf-based, and 
tries to autodetect a number of OS features, and use alternatives where 
these are not available.

Regards,

Alain



From steph2607 at tiscali.fr  Wed Sep 20 16:58:02 2006
From: steph2607 at tiscali.fr (Stephane)
Date: Wed, 20 Sep 2006 16:58:02 +0200
Subject: [Udpcast] Udpcast cast-o-matic
Message-ID: <200609201658.03512.steph2607@tiscali.fr>

Hi,

A suggestion about: http://udpcast.linux.lu/cast-o-matic/

Would it be possible to add javascript buttons to check/uncheck all drivers?

Here is my proposition:
===============================
<script type="text/javascript">
	function checkAll(){
		champs_input=document.getElementsByTagName('input');
		for(i=0;i<champs_input.length;i++){
			type=champs_input[i].getAttribute('type');
			name=champs_input[i].getAttribute('name');
			if(type=="checkbox"){
				champs_input[i].checked=true;
			}
		}
	}
	function UncheckAll(){
		champs_input=document.getElementsByTagName('input');
		for(i=0;i<champs_input.length;i++){
			type=champs_input[i].getAttribute('type');
			name=champs_input[i].getAttribute('name');
			if(type=="checkbox"){
				champs_input[i].checked=false;
			}
		}
	}
</script>
===============================

An to test:
===============================
<p><a href="javascript: checkAll();">Check All</a> / <a href="javascript: 
UncheckAll();">Uncheck All</a></p>
===============================

Thank you very much!
-- 
Stéphane.



From steph2607 at tiscali.fr  Wed Sep 20 17:40:27 2006
From: steph2607 at tiscali.fr (Stephane)
Date: Wed, 20 Sep 2006 17:40:27 +0200
Subject: [Udpcast] Building offline with RPM and 'Do not blank screen'?
Message-ID: <200609201740.28216.steph2607@tiscali.fr>

Hi,

I don't find any parameter to avoid blanking screen when buiding bootcd-image 
offline with rpms...

Is it the default way it works?

Thanks!
-- 
Stéphane.



From steph2607 at tiscali.fr  Wed Sep 20 18:50:48 2006
From: steph2607 at tiscali.fr (Stephane)
Date: Wed, 20 Sep 2006 18:50:48 +0200
Subject: [Udpcast] Building offline and LZOP...
Message-ID: <200609201850.49393.steph2607@tiscali.fr>

Hi,

I build udpcd.iso offline using rpm->alien->deb and lzop compression doesn't 
seem to work.
(I use 20060917 version)
When I choose lzop, at the moment I press a key to launch opéreation, I get:
	lzop: applet not found
And then it seems sot start but it send 0 bit and then print: Transfer 
complete

/bin/lzop is present on live system running.

What do you think about it?

Thanks!
-- 
Stéphane.



From steph2607 at tiscali.fr  Wed Sep 20 19:32:55 2006
From: steph2607 at tiscali.fr (Stephane)
Date: Wed, 20 Sep 2006 19:32:55 +0200
Subject: [Udpcast] Udpcast cast-o-matic
In-Reply-To: <200609201658.03512.steph2607@tiscali.fr>
References: <200609201658.03512.steph2607@tiscali.fr>
Message-ID: <200609201932.55917.steph2607@tiscali.fr>

Hi,

Le Mercredi 20 Septembre 2006 16:58, Stephane a écrit :
> Hi,
>
> A suggestion about: http://udpcast.linux.lu/cast-o-matic/
>
> Would it be possible to add javascript buttons to check/uncheck all
> drivers?
>
> Here is my proposition:
> ===============================
> <script type="text/javascript">
> 	function checkAll(){
> 		champs_input=document.getElementsByTagName('input');
> 		for(i=0;i<champs_input.length;i++){
> 			type=champs_input[i].getAttribute('type');

The following line isn't usefull...
> 			name=champs_input[i].getAttribute('name');
... I put it when I was testing my script...


> 			if(type=="checkbox"){
> 				champs_input[i].checked=true;
> 			}
> 		}
> 	}
> 	function UncheckAll(){
> 		champs_input=document.getElementsByTagName('input');
> 		for(i=0;i<champs_input.length;i++){
> 			type=champs_input[i].getAttribute('type');

Idem:
> 			name=champs_input[i].getAttribute('name');

> 			if(type=="checkbox"){
> 				champs_input[i].checked=false;
> 			}
> 		}
> 	}
> </script>
> ===============================
>
> An to test:
> ===============================
> <p><a href="javascript: checkAll();">Check All</a> / <a href="javascript:
> UncheckAll();">Uncheck All</a></p>
> ===============================
>
> Thank you very much!

Thanks!
-- 
Stéphane.



From Dale.Hofkens at diditrade.com  Wed Sep 20 18:13:09 2006
From: Dale.Hofkens at diditrade.com (Dale Hofkens)
Date: Wed, 20 Sep 2006 18:13:09 +0200
Subject: [Udpcast] cast-o-matic image gives error
Message-ID: <C2C7BA112F2AD24B8782B4405184FACD666D34@sv-mail.diditrade.com>

Hi All,
 
We are using Udpcast for some time now to clone pc's, and it's working
excellently, thanks for all the work that went into this project.
 
Recently we have changed hardware on our systems we clone, so i went to
the cast-o-matic site and made a new image.
We built a new client image on 20/9/2006 and download the new kernel
(2.6.17.13) to include on our pxe server.
 
We use LZOP compression, and this is what gives an error.
 
On the clients we receive the following message LZOP: Applet not found
as an error. 
After that the system gives a message it wont reboot and ends.

We dont't have this error with gzip compression, but the image that
resides on our server is LZOP and i read on the udpcast website that
lzop is the preferred compression method. Could it be that we do
something wrong, or is there a bug in the latest udpcast
version/cast-o-matic site.
 
Thanks for looking into this.
 
Kind regards,
Hofkens Dale
 
 
-----------
Settings used : 
pxe image
 
e100 , ata_piix
dhcp used
port 9000
no udpcast paramaters
compression  : lzop
transfer direction : receiver
use ff shell
reboot system
do not blank screen



From stephane.boireau at ac-rouen.fr  Wed Sep 20 21:50:27 2006
From: stephane.boireau at ac-rouen.fr (Stephane Boireau (Animateur Secteur Bernay/Pont-Audemer))
Date: Wed, 20 Sep 2006 21:50:27 +0200
Subject: [Udpcast] Building offline and LZOP...
In-Reply-To: <200609201850.49393.steph2607@tiscali.fr>
References: <200609201850.49393.steph2607@tiscali.fr>
Message-ID: <200609202150.27952.stephane.boireau@ac-rouen.fr>

Hi,

Le Mercredi 20 Septembre 2006 18:50, Stephane a écrit :
> Hi,
>
> I build udpcd.iso offline using rpm->alien->deb and lzop compression
> doesn't seem to work.
> (I use 20060917 version)
> When I choose lzop, at the moment I press a key to launch opéreation, I
> get: lzop: applet not found
> And then it seems sot start but it send 0 bit and then print: Transfer
> complete
>
> /bin/lzop is present on live system running.
>
> What do you think about it?

I bet the same error with an image build online using 
http://udpcast.linux.lu/cast-o-matic/

What's wrong?

Thanks!
-- 
Stéphane Boireau
Animateur TICE sur le Secteur de Bernay/Pont-Audemer (27)



From alain at knaff.lu  Wed Sep 20 23:50:19 2006
From: alain at knaff.lu (Alain Knaff)
Date: Wed, 20 Sep 2006 23:50:19 +0200
Subject: [Udpcast] Building offline and LZOP...
In-Reply-To: <200609202150.27952.stephane.boireau@ac-rouen.fr>
References: <200609201850.49393.steph2607@tiscali.fr>
	<200609202150.27952.stephane.boireau@ac-rouen.fr>
Message-ID: <4511B79B.60000@knaff.lu>

Stephane Boireau (Animateur Secteur Bernay/Pont-Audemer) wrote:
> Le Mercredi 20 Septembre 2006 18:50, Stephane a écrit :
[...]
>> get: lzop: applet not found
[...]
> I bet the same error with an image build online using 
> http://udpcast.linux.lu/cast-o-matic/
> 
> What's wrong?
> 
> Thanks!

The applets were mistakenly in a non-alphabetic order, which confused 
busybox' dychotomic search algorithm.

This is now fixed (20060920).

Regards,

Alain


From alain at knaff.lu  Wed Sep 20 23:53:44 2006
From: alain at knaff.lu (Alain Knaff)
Date: Wed, 20 Sep 2006 23:53:44 +0200
Subject: [Udpcast] Building offline with RPM and 'Do not blank screen'?
In-Reply-To: <200609201740.28216.steph2607@tiscali.fr>
References: <200609201740.28216.steph2607@tiscali.fr>
Message-ID: <4511B868.6030206@knaff.lu>

Stephane wrote:
> Hi,
> 
> I don't find any parameter to avoid blanking screen when buiding bootcd-image 
> offline with rpms...
> 
> Is it the default way it works?
> 
> Thanks!

Yes, indeed. By default the screensaver kicks in after a while. You need 
to use cast-o-matic (or invoke makeImage manually) if you want different 
behavior.

Regards,

Alain



From alain at knaff.lu  Wed Sep 20 23:59:50 2006
From: alain at knaff.lu (Alain Knaff)
Date: Wed, 20 Sep 2006 23:59:50 +0200
Subject: [Udpcast] cast-o-matic image gives error
In-Reply-To: <C2C7BA112F2AD24B8782B4405184FACD666D34@sv-mail.diditrade.com>
References: <C2C7BA112F2AD24B8782B4405184FACD666D34@sv-mail.diditrade.com>
Message-ID: <4511B9D6.1070308@knaff.lu>

Dale Hofkens wrote:
> On the clients we receive the following message LZOP: Applet not found
> as an error. 

The applets were mistakenly in a non-alphabetic order, which confused
busybox' dychotomic search algorithm.

This is now fixed (20060920).

Regards,

Alain



From alain at knaff.lu  Thu Sep 21 00:06:11 2006
From: alain at knaff.lu (Alain Knaff)
Date: Thu, 21 Sep 2006 00:06:11 +0200
Subject: [Udpcast] Udpcast cast-o-matic
In-Reply-To: <200609201658.03512.steph2607@tiscali.fr>
References: <200609201658.03512.steph2607@tiscali.fr>
Message-ID: <4511BB53.7030209@knaff.lu>

Stephane wrote:
> Hi,
> 
> A suggestion about: http://udpcast.linux.lu/cast-o-matic/
> 
> Would it be possible to add javascript buttons to check/uncheck all drivers?

Done.

Thanks,

Alain


From steven at nchc.org.tw  Thu Sep 21 07:38:46 2006
From: steven at nchc.org.tw (Steven Shiau)
Date: Thu, 21 Sep 2006 13:38:46 +0800
Subject: [Udpcast] Unable to make udpcast in gcc 3.2.2
Message-ID: <45122566.8090005@nchc.org.tw>

Just noticed that 20060920 was released, so as usual, I compiled it in
my RedHat 9 Linux with gcc (GCC) 3.2.2 20030222 (Red Hat Linux 3.2.2-5).
I can not compile it successfully. However, udpcast-20060619 works.
Here is my procedure to compile version 20060920:
--------
~/tmp/udpcast/udpcast-20060920>./configure
checking for gcc... gcc
checking for C compiler default output file name... a.out
checking whether the C compiler works... yes
checking whether we are cross compiling... no
checking for suffix of executables...
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ANSI C... none needed
checking how to run the C preprocessor... gcc -E
checking for egrep... grep -E
checking whether gcc needs -traditional... no
checking for a BSD-compatible install... /usr/bin/install -c
checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
checking target system type... i686-pc-linux-gnu
checking for an ANSI C-conforming const... yes
checking for inline... inline
checking for pthread_create in -lpthread... yes
checking for ANSI C header files... yes
checking for sys/wait.h that is POSIX.1 compatible... yes
checking for sys/types.h... yes
checking for sys/stat.h... yes
checking for stdlib.h... yes
checking for string.h... yes
checking for memory.h... yes
checking for strings.h... yes
checking for inttypes.h... yes
checking for stdint.h... yes
checking for unistd.h... yes
checking fcntl.h usability... yes
checking fcntl.h presence... yes
checking for fcntl.h... yes
checking limits.h usability... yes
checking limits.h presence... yes
checking for limits.h... yes
checking sys/ioctl.h usability... yes
checking sys/ioctl.h presence... yes
checking for sys/ioctl.h... yes
checking sys/time.h usability... yes
checking sys/time.h presence... yes
checking for sys/time.h... yes
checking arpa/inet.h usability... yes
checking arpa/inet.h presence... yes
checking for arpa/inet.h... yes
checking sys/param.h usability... yes
checking sys/param.h presence... yes
checking for sys/param.h... yes
checking for memory.h... (cached) yes
checking malloc.h usability... yes
checking malloc.h presence... yes
checking for malloc.h... yes
checking signal.h usability... yes
checking signal.h presence... yes
checking for signal.h... yes
checking netinet/in.h usability... yes
checking netinet/in.h presence... yes
checking for netinet/in.h... yes
checking winsock2.h usability... no
checking winsock2.h presence... no
checking for winsock2.h... no
checking termios.h usability... yes
checking termios.h presence... yes
checking for termios.h... yes
checking whether llseek declared in unistd.h... no
checking whether lseek64 declared in unistd.h... yes
checking for htons... yes
checking for an ANSI C-conforming const... (cached) yes
checking whether struct tm is in sys/time.h or time.h... time.h
checking return type of signal handlers... void
checking for atexit... yes
checking for on_exit... yes
checking for tcsetattr... yes
checking for lseek64... yes
checking for snprintf... yes
checking for 64-bit off_t... no
checking whether gcc supports loff_t type... yes
checking whether gcc supports offset_t type... no
checking whether gcc supports long long type... yes
configure: creating ./config.status
config.status: creating Makefile
config.status: creating config.h
config.status: config.h is unchanged
~/tmp/udpcast/udpcast-20060920>make
gcc -g -O2 -Wall -Wall -Wshadow -DBB_FEATURE_UDPCAST_FEC
-D_FILE_OFFSET_BITS=64 -DUSE_SYSLOG -DUSE_ASSEMBLER -O6 -DNO_BB -I.  -I.
  -c -o udp-receiver.o udp-receiver.c
gcc -g -O2 -Wall -Wall -Wshadow -DBB_FEATURE_UDPCAST_FEC
-D_FILE_OFFSET_BITS=64 -DUSE_SYSLOG -DUSE_ASSEMBLER -O6 -DNO_BB -I.  -I.
  -c -o socklib.o socklib.c
gcc -g -O2 -Wall -Wall -Wshadow -DBB_FEATURE_UDPCAST_FEC
-D_FILE_OFFSET_BITS=64 -DUSE_SYSLOG -DUSE_ASSEMBLER -O6 -DNO_BB -I.  -I.
  -c -o udpcast.o udpcast.c
gcc -g -O2 -Wall -Wall -Wshadow -DBB_FEATURE_UDPCAST_FEC
-D_FILE_OFFSET_BITS=64 -DUSE_SYSLOG -DUSE_ASSEMBLER -O6 -DNO_BB -I.  -I.
  -c -o receiver-diskio.o receiver-diskio.c
gcc -g -O2 -Wall -Wall -Wshadow -DBB_FEATURE_UDPCAST_FEC
-D_FILE_OFFSET_BITS=64 -DUSE_SYSLOG -DUSE_ASSEMBLER -O6 -DNO_BB -I.  -I.
  -c -o receivedata.o receivedata.c
receivedata.c:80: field `data_iov' has incomplete type
receivedata.c:134: confused by earlier errors, bailing out
make: *** [receivedata.o] Error 1

Anyone has same error ?
Thanks in advance.

-- 
Steven Shiau



From alain at knaff.lu  Thu Sep 21 10:00:27 2006
From: alain at knaff.lu (Alain Knaff)
Date: Thu, 21 Sep 2006 10:00:27 +0200
Subject: [Udpcast] Unable to make udpcast in gcc 3.2.2
In-Reply-To: <45122566.8090005@nchc.org.tw>
References: <45122566.8090005@nchc.org.tw>
Message-ID: <4512469B.50105@knaff.lu>

Steven Shiau wrote:
> Just noticed that 20060920 was released, so as usual, I compiled it in
> my RedHat 9 Linux with gcc (GCC) 3.2.2 20030222 (Red Hat Linux 3.2.2-5).
> I can not compile it successfully. However, udpcast-20060619 works.
> Here is my procedure to compile version 20060920:
> --------
[...]
> receivedata.c:80: field `data_iov' has incomplete type
> receivedata.c:134: confused by earlier errors, bailing out
> make: *** [receivedata.o] Error 1
> 
> Anyone has same error ?
> Thanks in advance.
> 

I made a new version (20060921), with <sys/uio.h> include added.
Could you check whether this one works better?

Thanx,

Alain


From steven at nchc.org.tw  Thu Sep 21 10:56:31 2006
From: steven at nchc.org.tw (Steven Shiau)
Date: Thu, 21 Sep 2006 16:56:31 +0800
Subject: [Udpcast] Unable to make udpcast in gcc 3.2.2
In-Reply-To: <4512469B.50105@knaff.lu>
References: <45122566.8090005@nchc.org.tw> <4512469B.50105@knaff.lu>
Message-ID: <451253BF.8030807@nchc.org.tw>

Yes, http://udpcast.linux.lu/download/udpcast-20060921.tar.gz works
great here.
Thanks for the quick response.

Steven.

Alain Knaff wrote:
> Steven Shiau wrote:
>> Just noticed that 20060920 was released, so as usual, I compiled it in
>> my RedHat 9 Linux with gcc (GCC) 3.2.2 20030222 (Red Hat Linux 3.2.2-5).
>> I can not compile it successfully. However, udpcast-20060619 works.
>> Here is my procedure to compile version 20060920:
>> --------
> [...]
>> receivedata.c:80: field `data_iov' has incomplete type
>> receivedata.c:134: confused by earlier errors, bailing out
>> make: *** [receivedata.o] Error 1
>>
>> Anyone has same error ?
>> Thanks in advance.
>>
> 
> I made a new version (20060921), with <sys/uio.h> include added.
> Could you check whether this one works better?
> 
> Thanx,
> 
> Alain

-- 
Steven Shiau <steven _at_ nchc org tw> <steven _at_ stevenshiau org>
National Center for High-performance Computing, Taiwan.
http://www.nchc.org.tw
Public Key Server PGP Key ID: 9762755A


From Dale.Hofkens at diditrade.com  Tue Sep 26 17:02:09 2006
From: Dale.Hofkens at diditrade.com (Dale Hofkens)
Date: Tue, 26 Sep 2006 17:02:09 +0200
Subject: [Udpcast] cast-o-matic reboot checkbox
Message-ID: <C2C7BA112F2AD24B8782B4405184FACD667211@sv-mail.diditrade.com>

Hi,

Is it possible that the Reboot system after udpcast? (only available
with full-featured shell) option is not working on the website at the
moment? We built various images lately but none of them reboot after
receiving an image.

Kind regards,
Hofkens Dale



From ygingras at ygingras.net  Wed Sep 27 23:20:12 2006
From: ygingras at ygingras.net (Yannick Gingras)
Date: Wed, 27 Sep 2006 17:20:12 -0400
Subject: [Udpcast] Windows and duped machine names
Message-ID: <m21wpxknw3.fsf@ogre.ygingras.net>


Hi I'm wondering if there is a way to set the machine name
automatically after the broadcast.  I heard that the last version of
Ghost do something like that.  For the GNU/Linux boxen I can autoset
the hostname from a small script in init.d.  I don't know much about
Windows networking.  Is the hostname in some file that we can write to
by just mounting the partition?  Do we have to mess with the registry?

All the boxes connect to a Samba PDC.  Can I copy the machine accounts
and let them share the SID?

-- 
Yannick Gingras


From alain at knaff.lu  Wed Sep 27 23:26:34 2006
From: alain at knaff.lu (Alain Knaff)
Date: Wed, 27 Sep 2006 23:26:34 +0200
Subject: [Udpcast] cast-o-matic reboot checkbox
In-Reply-To: <C2C7BA112F2AD24B8782B4405184FACD667211@sv-mail.diditrade.com>
References: <C2C7BA112F2AD24B8782B4405184FACD667211@sv-mail.diditrade.com>
Message-ID: <451AEC8A.9050704@knaff.lu>

Dale Hofkens wrote:
> Hi,
> 
> Is it possible that the Reboot system after udpcast? (only available
> with full-featured shell) option is not working on the website at the
> moment? We built various images lately but none of them reboot after
> receiving an image.
> 
> Kind regards,
> Hofkens Dale

Apparently, this is due to a new bug in busybox' init.

In order to workaround this, I added the -f flag to reboot in 
cast-o-matic, this makes reboot bypass init.

Regards,

Alain




From jnewbigin at ict.swin.edu.au  Thu Sep 28 01:54:22 2006
From: jnewbigin at ict.swin.edu.au (John Newbigin)
Date: Thu, 28 Sep 2006 09:54:22 +1000
Subject: [Udpcast] Windows and duped machine names
In-Reply-To: <m21wpxknw3.fsf@ogre.ygingras.net>
References: <m21wpxknw3.fsf@ogre.ygingras.net>
Message-ID: <451B0F2E.80209@ict.swin.edu.au>

You need a SID changer, which can normally also set the host name.

There is no SID changer for linux.... yet.  It so happens that I am 
working on one at the moment.  I hope it will be ready 'real soon now'.

I'll post to this list once I have a test version ready.

John.

Yannick Gingras wrote:

> Hi I'm wondering if there is a way to set the machine name
> automatically after the broadcast.  I heard that the last version of
> Ghost do something like that.  For the GNU/Linux boxen I can autoset
> the hostname from a small script in init.d.  I don't know much about
> Windows networking.  Is the hostname in some file that we can write to
> by just mounting the partition?  Do we have to mess with the registry?
> 
> All the boxes connect to a Samba PDC.  Can I copy the machine accounts
> and let them share the SID?
> 


-- 
John Newbigin
Computer Systems Officer
Faculty of Information and Communication Technologies
Swinburne University of Technology
Melbourne, Australia
http://www.ict.swin.edu.au/staff/jnewbigin



From K.MacDonald at ed.ac.uk  Thu Sep 28 10:20:22 2006
From: K.MacDonald at ed.ac.uk (Kenneth MacDonald)
Date: Thu, 28 Sep 2006 09:20:22 +0100
Subject: [Udpcast] Windows and duped machine names
In-Reply-To: <451B0F2E.80209@ict.swin.edu.au>
References: <m21wpxknw3.fsf@ogre.ygingras.net> <451B0F2E.80209@ict.swin.edu.au>
Message-ID: <1159431622.18113.64.camel@penguin.ucs.ed.ac.uk>

You can bring down a sysprep'd image and write a sysprep.inf file
containing the name, network settings, etc. for Windows to read during
its minisetup.

See our stuff at http://pie.ucs.ed.ac.uk/pie-1.2/ and in particular
http://pie.ucs.ed.ac.uk/pie-1.2/bin/image-build-xp for how we do this,
optionally using udpcast (or http or CDROM) to fetch the actual image.

Cheers,

Kenny.

On Thu, 2006-09-28 at 09:54 +1000, John Newbigin wrote:
> You need a SID changer, which can normally also set the host name.
> 
> There is no SID changer for linux.... yet.  It so happens that I am 
> working on one at the moment.  I hope it will be ready 'real soon now'.
> 
> I'll post to this list once I have a test version ready.
> 
> John.
> 
> Yannick Gingras wrote:
> 
> > Hi I'm wondering if there is a way to set the machine name
> > automatically after the broadcast.  I heard that the last version of
> > Ghost do something like that.  For the GNU/Linux boxen I can autoset
> > the hostname from a small script in init.d.  I don't know much about
> > Windows networking.  Is the hostname in some file that we can write to
> > by just mounting the partition?  Do we have to mess with the registry?
> > 
> > All the boxes connect to a Samba PDC.  Can I copy the machine accounts
> > and let them share the SID?
> > 
> 
> 
-- 
Desktop Services Team, EUCS, The University of Edinburgh.