Bugs item #996531, was opened at 2004-07-23 15:19
Message generated for change (Comment added) made by srumpel01
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=109368&aid=996531&group_id=9368
Category: Installation
Group: 3.0
Status: Open
Resolution: None
Priority: 5
Submitted By: SR (srumpel01)
Assigned to: Brian Elliott Finley (brianfinley)
Summary: Client installation don`t get finish
Initial Comment:
Hello at all,
I checked out Oscar version 3.0 with RH 9.0 on a little
testcluster with consists of an Dell Poweredge 1300 and
the clientnodes are Dell Optiplex 270 (P4 2,8 GHz) and
Dell Precision 340 (P4 1,8 GHz)
Everthing works fine, install Oscar, build a client
image, configure dhcp options.
The clientnodes boot from PXE, get the IP and load the
image, check the hardware at this Point it is fine also.
But now a message like this is shown on the screen:
We have connectivity to the Image Server!
start_syslogd
using multicast...
get_flamethrower_directory
flamethrower_client(flamethrower_directory)
-----------------------------------------------------
udp-receiver --interfrace eth0 --portbase 9000
2>/tmp/udp-receiver.flamethrower_directory | tar -x -c
/var/lib/systemimager/flamethrower 2>/tmp/tar.stderr.log
......
4 hours and many many Points(.) later i cancelt this.
I try it serveral time and it is evertime the same.
Know somebody something about this.
I want to use the real cluster with an NAG Parallel
Library Fortran 90 .
Do you have some experience with this.
----------------------------------------------------------------------
>Comment By: SR (srumpel01)
Date: 2004-07-26 12:55
Message:
Logged In: YES
user_id=1089755
I disabled the Multicast on Network Setup allways and build
my image without the Muticast option but everthing is the
same, i look in flamethrower.conf and this option are disabled.
----------------------------------------------------------------------
Comment By: Jason Brechin (brechin)
Date: 2004-07-26 02:51
Message:
Logged In: YES
user_id=274641
I have not heard of any successful installations using
multicast with OSCAR 3.0. It was included as an
experimental feature that had been reported as working.
Brian Finley should be able to elaborate more.
----------------------------------------------------------------------
Comment By: linux (jix)
Date: 2004-07-26 01:29
Message:
Logged In: YES
user_id=912954
Even I too confronted this problem,however, I build my
cluster using DHCP request simultaniously but I like to
know why OSCAR 3.0 fails at multicasting.Is it has any
problem with flamethrower or I am doing some thing wrong.Can
anybody enlighten me ??
Thanx in advance
--regards jix
----------------------------------------------------------------------
Comment By: Jason Brechin (brechin)
Date: 2004-07-23 16:36
Message:
Logged In: YES
user_id=274641
This is due to a non-working multicast image installation
software stack (part of SIS). To complete your installation,
you can choose to not use multicast in the Setup Networking
step and click Configure DHCP. Your clients should then be
able to build using the unicast method.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=109368&aid=996531&group_id=9368
-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
_______________________________________________
Oscar-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/oscar-devel
