[Oscar-users] Help request - How to add device driver into Oscar's kernel image/initrd

Hi - I have been happily running Oscar 5.1b on RH 5.3 for some time, but now have some new hardware that contains a disk controller that is not recognised by the existing image. I have tested that the new (LSI) driver works by installing RH 5.3 from DVD, then adding the new driver with the

Re: [Oscar-users] Help

First of all, if you are new to OSCAR, I strongly recommend you use Fedora or RHEL instead of SUSE! and then, just follow these steps one by one and inform us if any error occurred! (with full log and the step number) (from: http://www.sfr-fresh.com/unix/privat/oscar-base-5.1rc1.tar.gz:a/oscar-5.

Re: [Oscar-users] Help

Sory, I mistake... I hava downloaded this http://sourceforge.net/projects/oscar/files/ ***Download oscar-repo-common-rpms-5.1rc1.tar.gz (106.3 MB) * I found this link in the official site: http://

Re: [Oscar-users] Help

Hello, Suse is not officially supported and yum will definitively need to be installed if you want to use OSCAR. Because Suse is not officially supported, the distro is not correctly detected (the error you had earlier). The devel doc actually give some details about how to support a new distro

Re: [Oscar-users] Help

On 3/9/2011 7:02 PM, Everton Lima wrote: I am using the last version on the site. 6.1 I have downloaded the OSCAR in the download area (112Mb) then I alter the /etc/zypp/repos.d/ adding the new repository that the guide tells, but the guide use yum and i am using yast, becase i am using Open

Re: [Oscar-users] Help

I am using the last version on the site. 6.1 I have downloaded the OSCAR in the download area (112Mb) then I alter the /etc/zypp/repos.d/ adding the new repository that the guide tells, but the guide use yum and i am using yast, becase i am using OpenSuSe. Since now, I do not use anything the I d

Re: [Oscar-users] Help

Hi, Siavash is right, these two questions are essential. It will also help if you tell us which Linux distribution you are using and also if you can send the /var/log/oscar/bootstrap.log file. Regards, - "siavash ghiasvand" a écrit : > Hi, > > > which version was used?! > can you expla

Re: [Oscar-users] Help

Hi, which version was used?! can you explain what you did step by step? Sincerely yours, Siavash Ghiasvand -- http://forum.senmurv.org/profile/siavash 2011/3/9 Everton Lima > Hi there, > could someone take me a help? > > I was trying to install the OSCAR, but I can't. >

[Oscar-users] Help

Hi there, could someone take me a help? I was trying to install the OSCAR, but I can't. I was following the tutorial of site The erros is: /var/log/oscar does not exist, we create it Verbosity: 0 = == OSCAR bootstrap =

[Oscar-users] [help] Oscar firewall

Hi, I'm trying to write a firewall with iptables under Oscar/CentOS 5.5. Is there anybody here, who could/would help me... here is the state where I am: *filter :INPUT ACCEPT [0:0] :FORWARD ACCEPT [0:0] :OUTPUT ACCEPT [24392:5034163] :RH-Firewall-1-INPUT - [0:0] -A INPUT -j RH-Firewall-1-INPUT -A

Re: [Oscar-users] help

Hello, Please update your system, new binary packages for OSCAR are now available. Then, let me know if you still have the problem. Thanks, - "hamid nosrati" a écrit : > i have a problem in installing oscar on "centos 5.4 x_86" > > whenever i trying to install oscar with command "oscar-c

Re: [Oscar-users] help

Hello, For OSCAR-6.x, if the distro you are looking for is not in the list of supported distros, it means that the distro is not supported for that specific version of OSCAR, even if we reference it later in the documentation. Regards, - "Alfonso Mendez Tenorio" a écrit : > Dear friends

[Oscar-users] help

Dear friends at Oscar: I am having problems installing OSCAR. I am have no experience in installing this software. First of all I'd like to know how to make specific questions about the installation of the problem. Where can I send my questions. I am using Fedora core 9. I note that there ar

[Oscar-users] help

i have a problem in installing oscar on "centos 5.4 x_86" whenever i trying to install oscar with command "oscar-config --bootstrap" , after a moment i see the following error: http://isoredirect.centos.org/centos/5/os/i386/repodata/repomd.xml: [Errno 14] HTTP Error 404: Not Found output of co

[Oscar-users] Help!no opkgs passed!

 How to solve this question? By the way, does this warning, which refers oscar_server is not defined in /etc/hosts, affect the setup? my content in /etc/hosts: 127.0.0.1  localhost.localdomain localhost 192.168.0.1  gpu-cluster-5.localdomain gpu-cluster-5-

Re: [Oscar-users] Help with netwrok setup in oscar wizard

- États de l'Est Objet: [Oscar-users] Help with netwrok setup in oscar wizard Hello I have bootstrapped oscar sucessfully on my head node running ubuntu 8.04. I am having trouble with step 6 (Setup Networking) tho.I am trying to setup the network boot option so i can collect the mac adresses b

[Oscar-users] Help with netwrok setup in oscar wizard

Hello I have bootstrapped oscar sucessfully on my head node running ubuntu 8.04. I am having trouble with step 6 (Setup Networking) tho.I am trying to setup the network boot option so i can collect the mac adresses but i am getting a fake stop-start daemon error. r...@ben-cluster:/home/ben# oscar_

Re: [Oscar-users] help installing Oscar 6.0.1 on CentOS 5.2 i386

USA/Canada - États de l'Est Objet: [Oscar-users] help installing Oscar 6.0.1 on CentOS 5.2 i386 I am trying to install Oscar 6.0.1 on CentOS 5.2 i386, i followed the manual, 1) while performing yum update i got the following error : Error Downloading Packages: kernel-xen-devel - 2.6.18-9

[Oscar-users] help installing Oscar 6.0.1 on CentOS 5.2 i386

I am trying to install Oscar 6.0.1 on CentOS 5.2 i386, i followed the manual, 1) while performing yum update i got the following error : *Error Downloading Packages: kernel-xen-devel - 2.6.18-92.1.22.el5.i686: failure: RPMS/kernel-xen-devel-2.6.18-92.1.22.el5.i686.rpm from updates: [Errno 256] N

[Oscar-users] Help installing head node

Hey I've been trying to install OSCAR on fedora 9 but have run into these issues I don't know what the problem could be: --> Setting env variable OSCAR_UI to gui Can't locate package Exporter for @OSCAR::SystemSanity::ISA at ./scripts/system-sanity line 22. Checking your system... Argument "FAILURE

Re: [Oscar-users] help

In /etc/yum.conf set gpgkey=0 or gpgcheck=0 On Tue, Nov 4, 2008 at 4:24 PM, Ben Holmes <[EMAIL PROTECTED]> wrote: > Hey when trying to install yume i get an error saying it is unsigned any > help? > > - > This SF.Net email is

[Oscar-users] help

Hey when trying to install yume i get an error saying it is unsigned any help? - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win grea

Re: [Oscar-users] help me about OS

You should use the most recent supported distro. Especially with Fedora which changes very fast. I believe the beta supports some more recent versions of fedora than 5 if you have trouble finding it. On Wed, Oct 1, 2008 at 11:00 PM, S.Ali mirsoleimani < [EMAIL PROTECTED]> wrote: > Hello dear

[Oscar-users] help me about OS

Hello dear all I don't know which distribution of fedora is suitable for me. 4,5 or even higher. As 5 is fully supported I'm looking for this but I'm not sure. Please let me know about it. S.Ali mirsoleimani Web: www.linuxclusters.ir <>

Re: [Oscar-users] [HELP ME PLEASE!!!]JOB IN NoDE

I have a situation where I can no longer net/pxe boot the compute nodes on a cluster. Because I can get to it physically and reboot it manually. I need to push a new image out to the nodes can I use cpushimage to do this and if so How exactly? Anthony Walls Compuware Corporation [EMAIL PROTECTED]

Re: [Oscar-users] [HELP ME PLEASE!!!]JOB IN NoDE

To reimage a node, you need to use the admin wizard currently. There are some notes about it here: http://svn.oscar.openclustergroup.org/trac/oscar/wiki/AdminGuide/Wizard#Overview As far as installing programs and rpms in general, there are lots of ways to do this, the main thing is to make sure

Re: [Oscar-users] Help on qsub problem

PROTECTED] on behalf of Zhimin Xiong Sent: Tue 4/29/2008 8:16 AM To: oscar-users@lists.sourceforge.net Subject: [Oscar-users] Help on qsub problem Dear OSCAR users, I had a question on qsub. I found the job exited with -2 from the file /var/spool/server_priv/accounting/20080429 when I submit a job

Re: [Oscar-users] Help on qsub problem

You can remove a problem node from the cluster using the "Delete Cluster Node" in the installer. That function is also in the management wizard. This removes the node from all the configuration files as well. Then once you have fixed the node you can just add it back again. As for the qsub issue

[Oscar-users] Help on qsub problem

Dear OSCAR users, I had a question on qsub. I found the job exited with -2 from the file /var/spool/server_priv/accounting/20080429 when I submit a job via a script, and no other output file. It may be ascribed that I marked one node offline due to some problem, that it reported " EDAC MC0: CE -

[Oscar-users] Help

Hello Mr/Mrs, Thanks for your excellent and convenient software oscar.Now I have a problem in my installation.I try to install oscar 5.0 on a fedora core 5 x86-64 platform.The process goes on very well,but in step 2 and step 5 some value can't be changed which prevent my installation.The uncha

Re: [Oscar-users] Help! Installing OSCAR 4.2 on RHEL 5

The current version of OSCAR is 5.0 and the list of supported distributions is here http://oscar.openclustergroup.org/supported_distros If you would like to do some development work in getting other distributions up and running, thats fine, but I would recommend installing OSCAR on a supported dis

[Oscar-users] Help! Installing OSCAR 4.2 on RHEL 5

Hi Friends, I am trying to install OSCAR 4.2 (the 250MB tar file) onto my RHEL5. Following the instructions on the README file, I get this output after I run ./install_cluster eth0 (FYI, I have tried OSCAR 5 on RHEL 5 and Ubuntu 7.0.4 but it just exits with an error message that it did not iden

Re: [Oscar-users] help: lib missing on slave nodes

; Hua-Jun Fan, Ph.D. > > Email: [EMAIL PROTECTED] > Tel: (936) 261-3111 > Fax:(936) 261-3117 > > Department of Chemistry > Mail Stop 2215 > PO Box 519 > Prairie View, TX 77446-0519 > > > ____ > > From: [EMAIL PROTECTED] on

Re: [Oscar-users] help: lib missing on slave nodes

1:00 AM To: oscar-users@lists.sourceforge.net Subject: Re: [Oscar-users] help: lib missing on slave nodes That rpm is probably not on the rpmlist for the compute nodes. So you could do this one of several ways. 1) You can copy /opt/oscar/oscarsamples/rhel-4-x86_64.rpmlist to a new file, say custom.r

Re: [Oscar-users] help: lib missing on slave nodes

______ > > From: [EMAIL PROTECTED] on behalf of Michael Edwards > Sent: Wed 5/9/2007 11:25 PM > To: oscar-users@lists.sourceforge.net > Subject: Re: [Oscar-users] help: lib missing on slave nodes > > > > Did you install those packages on the head node? > > If

Re: [Oscar-users] help: lib missing on slave nodes

othically, after copying these files, how can I install on each compute node? Thanks for these tedious questions. Huajun From: [EMAIL PROTECTED] on behalf of Michael Edwards Sent: Wed 5/9/2007 11:25 PM To: oscar-users@lists.sourceforge.net Subject: Re: [Osc

Re: [Oscar-users] help: lib missing on slave nodes

Did you install those packages on the head node? If so, copy the required packages to the image directory on the head node (by default this is /var/lib/systemimager/images/oscarimage) and then do chroot /var/lib/systemimager/images/oscarimage Then install the packages as you did on the head node

[Oscar-users] help: lib missing on slave nodes

Hi All I have two slave nodes in a cluster running OSCAR5.0. One of the software failed with message of can't find libstdc++.so.5 when running on slave nodes. This file is under /usr/lib/ directory at master node, but not in slave nodes. on slave node, it shows libstdc++.so.6. I am totally new

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006 - APOLOGIES

I'm really sorry for this mistake - I've switched e-mail address/messages. -- Fernando Laudares Camargos Révolution Linux http://www.revolutionlinux.com --- * Tout opinion et prise de position exprimée dans ce message est celle de son auteur et pas néce

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

Pat, j'ai tes clés encore avec moi. Je serais au bureau demain très tôt, genre 7:00. A+ -- Fernando Laudares Camargos Révolution Linux http://www.revolutionlinux.com --- * Tout opinion et prise de position exprimée dans ce message est celle de son aute

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

Let us know if you have to set up the home directories by hand again. It seems like the nfs_oscar alias is not getting defined, which is a problem. On 3/9/07, Marco Bayas <[EMAIL PROTECTED]> wrote: > Ok. > > It seems nfs is active. I made a mistake last time. I did not use the name > of the head n

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

If nfs is broken, then nothing is going to work (unless you fiddle around a lot by hand) because OSCAR sets everything up to assume you have a shared home directory. Can you ssh from the head to the compute node, and from the compute node to the head node (try as root, those passwords should be in

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

I got nothing This is the error message in the client node: "mount: can't get address for nfs_server" Does that mean the nfs service is not active? I was looking a some of the features included with OSCAR and I concluded that the only package I need is Torque (I want to run a biomolecular

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

The easiest way to test nfs is by going to the client and doing the following mount -t nfs nfs_server:/home /home This should result in your home directory mounted or, hopefully, an error message in either the terminal or /var/log/messages on either the client or the server. Check and make sure

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

The nfs service appears to be active. I checked the "/var/log/ messages" and all the processes related with nfs are marked as "succeded". Perhaps you can tell me about a more direct way to test the activity of nfs. I have been looking in the internet for information but I got lost :( I am

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

Is the nfs service active on the head node? Is there a firewall service of any kind blocking ports? NFS is peculiar and runs over a somewhat random range of ports so is hard to deal with in a firewalled environment. Check and make sure on the compute node that nfs_oscar is not pointing to "loca

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

Hi I only found this : " oscarnode01 mount: mount: failed to probe ports on NFS server nfs_oscar oscarnode01 netfs: Mounting NFS filesystems: failed " in "/var/log/messages" with the command "dmesg" I got this: "...Installing knfsd .. eth0: no IPv6 routers present " On Mar 5, 2007, at 4

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

Check the output of the "dmesg" command as well as in "/var/log/messages" for information about nfs. On 3/5/07, Marco Bayas <[EMAIL PROTECTED]> wrote: Hi I have tried that several times. It does not work. There is always an error message when I reboot the client. The error appears for "Mountin

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

Hi I have tried that several times. It does not work. There is always an error message when I reboot the client. The error appears for "Mounting NFS File systems...". Do you know if there is a "reboot log"? so I can check the error message in detail. On Mar 5, 2007, at 9:39 AM, Michael

Re: [Oscar-users] Help with OSCAR5.0 on Mandriva2006

Have you tried rebooting the head node, waiting for it to completely boot, and then rebooting the compute node? Then run the install_cluster script again and rerun the tests. OSCAR doesn't seem to have refreshed your shell, or possibly the nfs daemon is not running, or both. Why? I have no ide

[Oscar-users] Help with OSCAR5.0 on Mandriva2006

Hello.I am trying to use OSCAR 5.0 with Mandriva 2006 in order to cluster two computers. I got the mandriva 2006 Powerpack+ from the mandriva store and I installed the system using the configuration suggested in the OSCAR5.0 installation manual. I have followed all the steps indicated in the OSCAR

Re: [Oscar-users] Help regarding Installation of OSCAR 5.0 !!

If you could tell us a little more about the hardware you are installing on and post a compressed copy of the oscarinstall.log file, it would be helpful. On 12/17/06, Jan Muhammad <[EMAIL PROTECTED]> wrote: > Dear All, > > We are trying to install/configure OSCAR5.0 on Fedora core 5 and facing the

[Oscar-users] Help regarding Installation of OSCAR 5.0 !!

Dear All, We are trying to install/configure OSCAR5.0 on Fedora core 5 and facing the following error message. Any body can suggest/help in this Could not restart 'systemimager_server_rsyncd' at /opt/pscar/oscar5.0/packages/sis/scripts/post_server_install line 62 Thanks in advance for help.

Re: [Oscar-users] help

Please set the following environment variable and re-try, it will provide you with additional debugging info: export DEBUG_UPDATE_RPMS=y BTW, chances are you forgot to copy some RPMs to /tftpboot/rpm - I suggest you go through the release notes carefully to make sure you have not missed anything.

Re: [Oscar-users] help! full install

ists.sourceforge.net Subject: Re: [Oscar-users] help! full install If you mean you want all the FC3 rpms installed on the client nodes you can either make a very long rpm list instead of using the default one, or you can use the tools included with Systemimager to make an image off a full install

Re: [Oscar-users] help! full install

27;t think this currently works yet, we'll see if we can get the functionality finalized before 5.0 is released. Cheers, Bernard From: [EMAIL PROTECTED] on behalf of Michael Edwards Sent: Thu 27/07/2006 20:33 To: oscar-users@lists.sourceforge.net Subje

Re: [Oscar-users] help! full install

If you mean you want all the FC3 rpms installed on the client nodes you can either make a very long rpm list instead of using the default one, or you can use the tools included with Systemimager to make an image off a full install instead of using the one automatically generated by OSCAR. If you'r

[Oscar-users] help! full install

hi: FC3  +  oscar4.2    I want to use oscar to manage my cluster. I want oscar client to install  all the RPMs.But I do not know how to do. help!!   cheers zhangjilin - Take Surveys. Earn Cash. Influence the Future of IT J

Re: [Oscar-users] help NFS stale file

Hi, do you have pfilter enabled? Try "service pfilter stop" on the master and the client nodes. Normally pfilter does a good job but you never know... Regards, Erich On Saturday 08 April 2006 10:32, genie lfs wrote: > hello all... > sorry for reiterating the problem... but some body help please

RE: [Oscar-users] help NFS stale file

Title: [Oscar-users] help NFS stale file Have you looked at /var/log/messages for any error messages?   You could also look at the NFS mount options and modify fstab accordingly (search google for help on this).  Another solution is to eliminate NFS altogether, i.e. copy the necessary

[Oscar-users] help NFS stale file

hello all... sorry for reiterating the problem... but some body help please... we are having a oscar cluster of 4 nodes built on fc3 using oscar 4.2.1.. the installation was perfect and i ran some sample applications also on them... mpi-povray and tachyon are all working smoothly on the cluster..

RE: [Oscar-users] help!. building client image (scientific linux 305)

> it reported lamd wasn't runnign on a compute node. > but isn;t this started by the pbs ? No, actually for TORQUE, the user is responsible for starting/stopping your own daemon in the script, search for "lamboot" in the following: http://svn.oscar.openclustergroup.org/oscar/branches/branch-4-2/p

Re: [Oscar-users] help!. building client image (scientific linux 305)

Bernard Li wrote: Can you show us the output of pbsnodes -a now that you have resolved the problem with /etc/hosts? I think your guess is right, and that the two new nodes cannot communicate with your pbs_server - did you check their error logs located in /var/spool/pbs? Also, post lamtes

RE: [Oscar-users] help!. building client image (scientific linux 305)

Title: Re: [Oscar-users] help!. building client image (scientific linux 305) Can you show us the output of pbsnodes -a now that you have resolved the problem with /etc/hosts?   I think your guess is right, and that the two new nodes cannot communicate with your pbs_server - did

RE: [Oscar-users] help!. building client image (scientific linux 305)

Title: Re: [Oscar-users] help!. building client image (scientific linux 305) Can you try to restart pbs_mom on all your compute nodes and then run pbsnodes -a to see if the state becomes "free"?   # cexec /etc/init.d/pbs_mom restart   # pbsnodes -a   Cheers,   Bern

Re: [Oscar-users] help!. building client image (scientific linux 305)

I've managed to solve the issue with the pbs_nodes. the client nodes had (in /var/spool/pbs/mom_priv/config) a name for the head which didn't match the hosts file. My best guess is this had to do with the fact that my heads nodes public interface's IP/name was loaded by DHCP and when I swapped

Re: [Oscar-users] help!. building client image (scientific linux 305)

Bernard Li wrote: What's the output of 'pbsnodes -a'? pbsnodes -a returns that all are unknown or down [EMAIL PROTECTED] oscar]# pbsnodes -a cc001.pg-207.computing.dcu.ie state = state-unknown,down np = 1 properties = all ntype = cluster cc002.pg-207.computing.dcu.ie

RE: [Oscar-users] help!. building client image (scientific linux 305)

What's the output of 'pbsnodes -a'? Is pbs_mom running on all your client nodes? Cheers, Bernard > well it was going well > > i added two more nodes > and now it fails > > [EMAIL PROTECTED] oscar]# testing/test_cluster > Performing root tests... > Maui service > check:maui

RE: [Oscar-users] help!. building client image (scientific linux 305)

Hi Neil: > on the SL4.2 > I don't believe I *had* to use SL_* rpms. I did as I had on SL3. > I definitly saw the conflicts betweem Twisted and > python-twisted. which > resolved by removing Twisted (and the dependecny listed with apitest) As mentioned, the python-twisted/Twisted issue is resol

Re: [Oscar-users] help!. building client image (scientific linux 305)

Bernard, looking at the instalation with fresh eyes this Monday morning. I noticed that one of the head nodes Ethernet controllers (nVidia force2 on motherboard) was flaky under SL4.2. It was always slow to start up and sometimes failed causing, ntp problems, ccsd to fail, and a very very sl

RE: [Oscar-users] help!. building client image (scientific linux 305)

Hi Neil: > To get this far I needed to > - add all sl_release*.rpm and all SL_* rpms to the detected > redhat-4EL-i386.rpmlist > these are, I guess, SL extensions to redhat 4 default rpms. So you're saying using the default rpmlist you cannot create an image on SL 4.2? Can you enable debuggi

Re: [Oscar-users] help!. building client image (scientific linux 305)

lete --name Cheers, Bernard From: Neil Costigan [mailto:[EMAIL PROTECTED] Sent: Thu 30/03/2006 06:00 To: Bernard Li Cc: Neil Costigan; oscar-users@lists.sourceforge.net Subject: RE: [Oscar-users] help!. building client image (scientific linux 305) > Hi Neil: > > > Does Scienti

RE: [Oscar-users] help!. building client image (scientific linux 305)

m > <== OK screen /tftpboot/rpm/screen-3.9.15-10.i386.rpm > <== OK sed /tftpboot/rpm/sed-4.0.7-3.i386.rpm > <== OK sendmail/tftpboot/rpm/sendmail-8.12.11-4.RHEL3.1.i386.rpm > <== OK setserial /tftpboot/rpm/setserial-2.17-12.i386.rpm > <== OK setup /tftpboot/rpm/setup-2

RE: [Oscar-users] help!. building client image (scientific linux 305)

ednesday, March 29, 2006 6:36 > To: oscar-users@lists.sourceforge.net > Subject: [Oscar-users] help!. building client image > (scientific linux 305) > > oscar 4.2 > > > to start out with I'm using a non-oscar-standard linux > (scientific linux > 3) but its rep

[Oscar-users] help!. building client image (scientific linux 305)

oscar 4.2 to start out with I'm using a non-oscar-standard linux (scientific linux 3) but its reported to work. its a RH3 clone. I need to use it to keep compatibility with a GRID I hope to join. i'm at the build client image stage it detects the distro as redhat-3as-i386 after choosing the

RE: [Oscar-users] Help, I broke OSCAR!

4/08/2005 10:00To: oscar-users@lists.sourceforge.netSubject: RE: [Oscar-users] Help, I broke OSCAR! Many thanks Bernard and FernandoTwo things:1) The reason I thought I needed a second oscar image is that half my nodes now have IDE drives and the other half SATA (=SCSI?). Doesn't this mean they

RE: [Oscar-users] Help, I broke OSCAR!

Many thanks Bernard and Fernando Two things: 1) The reason I thought I needed a second oscar image is that half my nodes now have IDE drives and the other half SATA (=SCSI?). Doesn't this mean they will need different partition tables and /etc/fstab files at a minimum? I realise that I don't need

RE: [Oscar-users] Help, I broke OSCAR!

[EMAIL PROTECTED] on behalf of Giles LesserSent: Wed 03/08/2005 16:20To: oscar-users@lists.sourceforge.netSubject: [Oscar-users] Help, I broke OSCAR! Hi AllHere's the problem: I had a happily running OSCAR (v4.1 on RHEL3) cluster based on Dell Precision 470 machines with IDE drives. I

[Oscar-users] Re: [Oscar-users]Help, I broke OSCAR!

Hello Giles, just a quick answer to your question (maybe somebady else in the list can write a better one). I don`t know if your procedure was right - I mean, I`m not sure if you should or should not have to build a new image, sata-supported, and push it to your new nodes. Anyway, if you execute t

[Oscar-users] Help, I broke OSCAR!

Hi All Here's the problem: I had a happily running OSCAR (v4.1 on RHEL3) cluster based on Dell Precision 470 machines with IDE drives. I went and bought some new (additional) nodes but IDE wasn't an option any longer, so I got SATA drives instead. The new boxes also seem to have come with an upda

RE: [Oscar-users] help

, May 28, 2005 4:48To: oscar-users@lists.sourceforge.netSubject: [Oscar-users] help Hello , I want to install oscar on fedora but when I put ./install_cluster eth0 , I have this error message: Couldn't set up oda database at ./wizard_prep line 172 Oscar Wizard preparation script f

[Oscar-users] help

Hello , I want to install oscar on fedora but when I put ./install_cluster eth0 , I have this error message: Couldn't set up oda database at ./wizard_prep line 172 Oscar Wizard preparation script failed to complete at ./install_cluster line 124 I don't konw what I should do?---

[Oscar-users] help newbie here

Hii friends This is abhinav from India. We at our lab tried Oscar3.3.0 on RH 9.0 and Fedora c1 but the installation does not get completed. during the install_cluster script it gives an erroe cant find opa.dw in @INC .   What is this problem all about. we tried a lot but i dot knw whats the problem

Re: [Oscar-users] help with insatll MDK9.0 and OSCAR 2.3.1

Jeff Squyres wrote: Actually, I'm wondering if the ODA password stuff got confused if there was a space in the password (see below). -- {+} Jeff Squyres {+} [EMAIL PROTECTED] {+} http://www.lam-mpi.org/ yes my password with space I have changed it. Now oda looks good.

[Fwd: Re: [Oscar-users] help with insatll MDK9.0 and OSCAR 2.3.1]

Benoit des Ligneris wrote: Hello, It looks like you don't have a valid interface. What is the output of "ifconfig" on your server ? to me everythink looks normal: eth0 Link encap:Ethernet HWaddr 00:07:E9:15:62:A8 <- cluster interface inet addr:10.0.0.1 Bcast:10.0.0.255 Mask

Re: [Oscar-users] help with insatll MDK9.0 and OSCAR 2.3.1

On Wed, 22 Oct 2003, Benoit des Ligneris wrote: > It looks like you don't have a valid interface. What is the output of > "ifconfig" on your server ? Actually, I'm wondering if the ODA password stuff got confused if there was a space in the password (see below). > > I am newbie with OSCAR and I

Re: [Oscar-users] help with insatll MDK9.0 and OSCAR 2.3.1

Hello, It looks like you don't have a valid interface. What is the output of "ifconfig" on your server ? Ben * Darius Kucinskas <[EMAIL PROTECTED]> [03-10-22 04:54]: > Hi, > > I am newbie with OSCAR and I am tired of trying to install OSCAR 2.3.1 > on Mandrake 9.0 everything gos smoofly until .

[Oscar-users] help with insatll MDK9.0 and OSCAR 2.3.1

Hi, I am newbie with OSCAR and I am tired of trying to install OSCAR 2.3.1 on Mandrake 9.0 everything gos smoofly until ... log: -- --> Installing server non-core RPMs (core RPMs already installed) --> Successfully installed serve

[Oscar-users] help us

we install OSCAR 2.2.1,at the time of installation we face some problem.These are as follows--- 1.Is it necessary that nodes are homogeneous? 2.Can node contain duel booting such as(winXP,linux). 3.how can we give input a program in OSCAR 2.2.1 for testing the parallel processing? f

Re: [Oscar-users] Help !!!(MPICH + PBS)

no i did not. I did and finally everything is ok. I am using MPICH to compile an run my jobs.I am using mpirun to execute the jobs. Here i am sending a simple mpi job that compile and execute a cpi.c #!/bin/bash #PBS -N caca4 #PBS -o outputfile #PBS -e errorfile #PBS -l nodes=2:ppn=2 #PBS -m a

Re: [Oscar-users] Help !!!

David- Have you purged the server of the job via the method described previously, and restarted all the moms? thx- Jeremy At 12:35 PM 7/12/2003 -0400, David Gutierrez wrote: Hi: I am sending the following logs. What is wrong? Why i am geeting the following error each time i qsub a job?

[Oscar-users] Help !!!

i finalli got rid of the job in the qstat but now after reboot everything i am getting the following messages: PBS Job Id: 109.oscarnode1.oscardomain Job Name: PBSMPIcpiTEST Aborted by PBS Server Job cannot be executed See Administrator for help I saw the Mom logs in /var/spool/pbs/and t

Re: [Oscar-users] Help !!!

David- where are you seeing this message below? Jeremy At 05:44 PM 7/11/2003 -0400, David Gutierrez wrote: i finalli got rid of the job in the qstat but now after reboot everything i am getting the following messages: PBS Job Id: 109.oscarnode1.oscardomain Job Name: PBSMPIcpiTEST Abort

Re: [Oscar-users] Help Putting the MPIEXEC stdio Patch in the OSCAR version of PBS

I'm not sure what advantage you'll gain by using the PBS TM interface. You might want to visit to find answer to this question. We switched and are glad we did. -- Wendy Lin - IT Research Computing Services [EMAIL PROTECTED] http://www-rcd.cc.purdue.e

Re: [Oscar-users] Help Putting the MPIEXEC stdio Patch in the OSCARversion of PBS

Also note that the next version of LAM/MPI -- which, I *swear* we'll release soon -- will have many of these features: 1. mpiexec to start processes 2. use PBS's native TM to start/stop processes (instead of rsh/ssh) 3. funnel all MPI processes' I/O back to mpirun/mpiexec All of this is included

Re: [Oscar-users] Help Putting the MPIEXEC stdio Patch in the OSCAR version of PBS

That patch should work on the PBS included in OSCAR. You can get the src rpm from the CVS tree if you want to try it. http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/oscar/oscar/packages/pbs/SRPMS/ However, I'm not sure what advantage you'll gain by using the PBS TM interface. mpirun itself sho

[Oscar-users] Help Putting the MPIEXEC stdio Patch in the OSCAR version of PBS

Hello, I would like to use the PBS queue system to run mpi jobs. From what I understand, mpiexec will do this. I've compiled and installed mpiexec-0.74. When I run jobs, I get horribly garbled output. The README that comes with the mpiexec tar ball says ... [SNIP] 1. For sufficiently ne

Re: [Oscar-users] Help with mpirun

Bart, You might want to post your code, and how you linked it. Are you using LAM/MPI? Did you start the LAM deamons? Regards, Joe Griffin Bart Ciak wrote: I am new to OSCAR and could really use some help with this problem I'm having. To test my cluster I am using a program that calculates pi

[Oscar-users] Help with mpirun

I am new to OSCAR and could really use some help with this problem I'm having. To test my cluster I am using a program that calculates pi. I can run the program on Node1 with the command: mpirun n1 ./pi and it works fine. I can also run the program on Node2 with the same command, except n1

[Oscar-users] help,need in urgent

hi,   I am arthi from india. HEY i am trying to implement a 2 node cluster using PCQLinux 8.0 given along with the PCQuest magazine.   we have installed the linux in server machine even got booted up with the client but while we undergo the step of Test CLuster Setup   i get the follwoing me

  1   2   >