Hi, I finally figure out what is going on with my system. I have 2 NIC in my client node (integrated NIC and 3Com NIC I put in). The integrate NIC doesn't work. That causes my problem(cannot network boot or floppy boot).
Now I change 3COM NIC as my eth0. I was able to boot either from PXE or floppy. However after reboot from harddisk, my integrated NIC become eth0 again which doesn't match server MAC. I tried to change MAC address but it doesn't work. I want my 3COM NIC to be eth0! Does someone know how to prevent my integrated NIC boot as eth0 or disable it? Is it possible to modify boot floppy to make my 3COM NIC to be eth0? Thanks RX > Send Oscar-users mailing list submissions to > [EMAIL PROTECTED] > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.sourceforge.net/lists/listinfo/oscar-users > or, via email, send a message with subject or body 'help' to > [EMAIL PROTECTED] > > You can reach the person managing the list at > [EMAIL PROTECTED] > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Oscar-users digest..." > > > Today's Topics: > > 1. Re: Help PBS/Maui not working properly (Jose Manuel Benitez Sanchez) > 2. Re: Help PBS/Maui not working properly (Jose Manuel Benitez Sanchez) > 3. Re: problem to install 2.0 (Vann H. Walke) > 4. Re: problem to install 2.0 (Jason Brechin) > 5. Re: problem on changing Remote boot type (Thomas Naughton) > 6. Re: problem to install 2.0 (Jeff Squyres) > 7. Re: Help PBS/Maui not working properly (Jeremy Enos) > 8. Re: Help PBS/Maui not working properly (Jose Manuel Benitez Sanchez) > 9. Re: problem to install 2.0 ([EMAIL PROTECTED]) > > --__--__-- > > Message: 1 > Date: Thu, 12 Dec 2002 09:34:53 +0100 > From: Jose Manuel Benitez Sanchez <[EMAIL PROTECTED]> > To: Jeremy Enos <[EMAIL PROTECTED]>, [EMAIL PROTECTED] > Subject: Re: [Oscar-users] Help PBS/Maui not working properly > > > > Jeremy Enos wrote: > > I think you have to change your Maui config back to DEDICATED. > Not at all: > ---- > # /opt/maui/bin/showconfig | grep NODEAC > NODEACCESSPOLICY SHARED > ---- > > > -- > Jose Manuel Benitez Sanchez e-mail: [EMAIL PROTECTED] > http://decsai.ugr.es/~jmbs > Dpto. de Ciencias de la Computacion e Inteligencia Artificial > E.T.S. Ingenieria Informatica > Universidad de Granada Tel. +34 - 958 - 24 61 43 > 18071 - GRANADA (Spain) Fax: +34 - 958 - 24 33 17 > > > > > --__--__-- > > Message: 2 > Date: Thu, 12 Dec 2002 13:50:26 +0100 > From: Jose Manuel Benitez Sanchez <[EMAIL PROTECTED]> > To: Jeremy Enos <[EMAIL PROTECTED]> > CC: Mike Mettke <[EMAIL PROTECTED]>, [EMAIL PROTECTED] > Subject: Re: [Oscar-users] Help PBS/Maui not working properly > > Jeremy Enos wrote: > > I think you have to change your Maui config back to DEDICATED. > > > With Maui configured as DEDICATED, it won't allow more than one process > per node. So it assigns each new job to a free node, otherwise the job > is enqueued. That's the solution for having just ONE job running on a node. > > However, I wonder what would be more efficient: running two (or a small > number of) processes concurrently on a single node (taking advantage of > the time-sharing feature) or running one after another has finished. > > To allow for this second option, Maui has to be configured with > NODEACCESSPOLICY SHARED. But to get an efficient assignment policy, it > is necessary to assign every new job to the node with the lowest load. > Does anyone know how can this be achieved with Maui/PBS? > > > -- > Jose Manuel Benitez Sanchez e-mail: [EMAIL PROTECTED] > http://decsai.ugr.es/~jmbs > Dpto. de Ciencias de la Computacion e Inteligencia Artificial > E.T.S. Ingenieria Informatica > Universidad de Granada Tel. +34 - 958 - 24 61 43 > 18071 - GRANADA (Spain) Fax: +34 - 958 - 24 33 17 > > > > > --__--__-- > > Message: 3 > Subject: Re: [Oscar-users] problem to install 2.0 > From: "Vann H. Walke" <[EMAIL PROTECTED]> > To: [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED] > Organization: Presearch, Inc. > Date: 12 Dec 2002 07:55:54 -0500 > > After the desired MAC address shows up, click on it, then click on the > node in the right pane to which you want to assign it. Then click > "Assign MAC to node". Do this for any nodes you desire to use. Then > click the "Configure DHCP Server" button. > This is covered reasonably well in the Installation docs. > > Good luck, > Vann > > On Thu, 2002-12-12 at 03:05, [EMAIL PROTECTED] wrote: > > Dear helper: > > > > I am new here and I had a problem to install oscar 2.0. > > I try to build a cluster myself. I was confused on step 6 setup networking. > > I was able to collect mac, then I "stop collecting" and click > > "configure DHCP server". After that I am not sure what to do next so I > > tried click "setup network boot" and switch to my client node and > > started PXE again. However no matter what I do, I cannot install > > client. I also tried boot from floppy. The boot log is attached. > > I have 3Com 3c905TMX NIC. > > > > Would you please help me on this one? What is the sequence I have to > > do? When should I switch to client to second boot from network or > > let the first network boot rnning? > > > > Thanks for help and best regards > > RX > > > > --__--__-- > > Message: 4 > Subject: Re: [Oscar-users] problem to install 2.0 > From: Jason Brechin <[EMAIL PROTECTED]> > To: "Vann H. Walke" <[EMAIL PROTECTED]> > Cc: [EMAIL PROTECTED], [EMAIL PROTECTED] > Organization: > Date: 12 Dec 2002 08:49:31 +0000 > > In addition, if you'd like to use PXE booting (aka network booting), you > need to click Setup Network Boot near the bottom of the window. > > Jason > > On Thu, 2002-12-12 at 12:55, Vann H. Walke wrote: > > After the desired MAC address shows up, click on it, then click on the > > node in the right pane to which you want to assign it. Then click > > "Assign MAC to node". Do this for any nodes you desire to use. Then > > click the "Configure DHCP Server" button. > > This is covered reasonably well in the Installation docs. > > > > Good luck, > > Vann > > > > On Thu, 2002-12-12 at 03:05, [EMAIL PROTECTED] wrote: > > > Dear helper: > > > > > > I am new here and I had a problem to install oscar 2.0. > > > I try to build a cluster myself. I was confused on step 6 setup networking. > > > I was able to collect mac, then I "stop collecting" and click > > > "configure DHCP server". After that I am not sure what to do next so I > > > tried click "setup network boot" and switch to my client node and > > > started PXE again. However no matter what I do, I cannot install > > > client. I also tried boot from floppy. The boot log is attached. > > > I have 3Com 3c905TMX NIC. > > > > > > Would you please help me on this one? What is the sequence I have to > > > do? When should I switch to client to second boot from network or > > > let the first network boot rnning? > > > > > > Thanks for help and best regards > > > RX > > > > > > > > ------------------------------------------------------- > > This sf.net email is sponsored by: > > With Great Power, Comes Great Responsibility > > Learn to use your power at OSDN's High Performance Computing Channel > > http://hpc.devchannel.org/ > > _______________________________________________ > > Oscar-users mailing list > > [EMAIL PROTECTED] > > https://lists.sourceforge.net/lists/listinfo/oscar-users > > > > --__--__-- > > Message: 5 > Date: Thu, 12 Dec 2002 09:56:40 -0500 (EST) > From: Thomas Naughton <[EMAIL PROTECTED]> > Reply-To: Thomas Naughton <[EMAIL PROTECTED]> > To: Feng Yang <[EMAIL PROTECTED]> > cc: <[EMAIL PROTECTED]> > Subject: Re: [Oscar-users] problem on changing Remote boot type > > Feng Yang, > > Typically we have the boot order: Flpy, CD, HD, PXE. > > If nothing is on the HD (or if you 'dd' the MBR) it will fall back to PXE, > and after the install it will boot from the HD. > > There are other methods depending on your hardware's capabilities for > remote admin but those are obviously more advanced questions and depend on > the hardware. > > --tjn > > _________________________________________________________________________ > Thomas Naughton [EMAIL PROTECTED] > Research Associate (865) 576-4184 > > > On Thu, 12 Dec 2002, Feng Yang wrote: > > > dear support: > > I have one problems: > > when I install the client node completely, I must change the bootlist > > manually to downlevel the PXE boot. I wonder if there has the solution to > > change the network boot package, with which client node can direct the PXE > > boot to localdisk? > > because if I have 64 nodes, it seems I can affort it, if I have 128 > > or more, ...... > > > > Best regards; > > Yang Feng > > ISMS_ Open System Solution > > Integrated Technology Services, IBM China > > 23/F , IBM Tower Pacific Century Place 2A GongTi Bei Road Chao Yang > > District Beijng 100027 > > Tel: 86-10-65391188 Ext. 4138 > > Fax: 86-10-65391692 > > > > > > > > > > ------------------------------------------------------- > > This sf.net email is sponsored by: > > With Great Power, Comes Great Responsibility > > Learn to use your power at OSDN's High Performance Computing Channel > > http://hpc.devchannel.org/ > > _______________________________________________ > > Oscar-users mailing list > > [EMAIL PROTECTED] > > https://lists.sourceforge.net/lists/listinfo/oscar-users > > > > > > --__--__-- > > Message: 6 > Date: Thu, 12 Dec 2002 09:58:29 -0500 (EST) > From: Jeff Squyres <[EMAIL PROTECTED]> > To: Jason Brechin <[EMAIL PROTECTED]> > Cc: "Vann H. Walke" <[EMAIL PROTECTED]>, "" <[EMAIL PROTECTED]>, > "" <[EMAIL PROTECTED]> > Subject: Re: [Oscar-users] problem to install 2.0 > > On Thu, 12 Dec 2002, Jason Brechin wrote: > > > In addition, if you'd like to use PXE booting (aka network booting), you > > need to click Setup Network Boot near the bottom of the window. > > Also be sure to check the full installation guide; this is detailed on > pages 20-21 and 33. > > -- > {+} Jeff Squyres > {+} [EMAIL PROTECTED] > {+} http://www.lam-mpi.org/ > > > --__--__-- > > Message: 7 > Date: Thu, 12 Dec 2002 11:25:44 -0600 > To: Jose Manuel Benitez Sanchez <[EMAIL PROTECTED]> > From: Jeremy Enos <[EMAIL PROTECTED]> > Subject: Re: [Oscar-users] Help PBS/Maui not working properly > Cc: Mike Mettke <[EMAIL PROTECTED]>,[EMAIL PROTECTED] > > Jose- > It depends on the nature of the jobs you're running. Typically, if you're > running the same type of jobs on a shared node, then they just split the > bottleneck and run less efficiently. In most cases, I would also recommend > setting the number of virtual procs = physical procs for the PBS > config. How many physical procs/box does your cluster have? > Also, are the jobs you are running CPU, memory, or I/O bottlenecked? > thx- > > Jeremy > > At 01:50 PM 12/12/2002 +0100, Jose Manuel Benitez Sanchez wrote: > >Jeremy Enos wrote: > >>I think you have to change your Maui config back to DEDICATED. > > > > > >With Maui configured as DEDICATED, it won't allow more than one process > >per node. So it assigns each new job to a free node, otherwise the job is > >enqueued. That's the solution for having just ONE job running on a node. > > > >However, I wonder what would be more efficient: running two (or a small > >number of) processes concurrently on a single node (taking advantage of > >the time-sharing feature) or running one after another has finished. > > > >To allow for this second option, Maui has to be configured with > >NODEACCESSPOLICY SHARED. But to get an efficient assignment policy, it is > >necessary to assign every new job to the node with the lowest load. > >Does anyone know how can this be achieved with Maui/PBS? > > > > > >-- > >Jose Manuel Benitez Sanchez e-mail: [EMAIL PROTECTED] > >http://decsai.ugr.es/~jmbs > >Dpto. de Ciencias de la Computacion e Inteligencia Artificial > >E.T.S. Ingenieria Informatica > >Universidad de Granada Tel. +34 - 958 - 24 61 43 > >18071 - GRANADA (Spain) Fax: +34 - 958 - 24 33 17 > > > > > > > > --__--__-- > > Message: 8 > Date: Thu, 12 Dec 2002 18:38:18 +0100 > From: Jose Manuel Benitez Sanchez <[EMAIL PROTECTED]> > To: Jeremy Enos <[EMAIL PROTECTED]> > CC: Mike Mettke <[EMAIL PROTECTED]>, [EMAIL PROTECTED] > Subject: Re: [Oscar-users] Help PBS/Maui not working properly > > Jeremy Enos wrote: > > Jose- > > It depends on the nature of the jobs you're running. Typically, if > > you're running the same type of jobs on a shared node, then they just > > split the bottleneck and run less efficiently. In most cases, I would > > also recommend setting the number of virtual procs = physical procs for > > the PBS config. How many physical procs/box does your cluster have? > The cluster has 8 client nodes, each one has a single PIII 800 MHz > Processor, with 128 MB of main memory and a 40 GB HDD. By now it's a > little outdated. We intend to upgrade it early next year. > > > Also, are the jobs you are running CPU, memory, or I/O bottlenecked? > Well, we have a little of everything. Most of them are heavy computing > tasks (i.e. Neural Nets training, Neuro-Fuzzy Systems modelling, Genetic > Algorithms executions); some others have large memory requirements > (feature selection processes). Finally, I/O isn't being a very big issue > by now. > > > So it's hard to find a perfect, common solution for everyone!! > > > > -- > Jose Manuel Benitez Sanchez e-mail: [EMAIL PROTECTED] > http://decsai.ugr.es/~jmbs > Dpto. de Ciencias de la Computacion e Inteligencia Artificial > E.T.S. Ingenieria Informatica > Universidad de Granada Tel. +34 - 958 - 24 61 43 > 18071 - GRANADA (Spain) Fax: +34 - 958 - 24 33 17 > > > > > --__--__-- > > Message: 9 > From: [EMAIL PROTECTED] > To: Jason Brechin <[EMAIL PROTECTED]> > Cc: [EMAIL PROTECTED],[EMAIL PROTECTED],[EMAIL PROTECTED] > Subject: Re: [Oscar-users] problem to install 2.0 > Date: Thu, 12 Dec 2002 19:12:10 +0000 > > I did all steps Vann mentioned. Here is what I need to do, please correct me. > > For network boot: > Collect MAC for each node and stop collecting > click "configure DHCP server" > click "setup network boot" > then(???): > click "close", switch to client node, start a new network boot(I > just push reset button on my client, now power down) > or > switch to client node, starting a new network boot, after all > clients are up, switch back to server, click "close" > > For boot from floppy, I need to do: > Collect MAC for each node and stop collecting > click "configure DHCP server" > click "build autoinstall floppy" > click "close" ??? (or wait for all clients are up) > Insert floppy to client node, and start new boot > hope client is boot > click "close" ??? > > Does anyone know 3C905 TMX NIC + Linux 7.2 works OK for OSCAR? There > is a sentence inside doc mentions there are problems for some 3Com cards. > > Thanks for help > > Season greatings > RX > > > > > In addition, if you'd like to use PXE booting (aka network booting), you > > need to click Setup Network Boot near the bottom of the window. > > > > Jason > > > > On Thu, 2002-12-12 at 12:55, Vann H. Walke wrote: > > > After the desired MAC address shows up, click on it, then click on the > > > node in the right pane to which you want to assign it. Then click > > > "Assign MAC to node". Do this for any nodes you desire to use. Then > > > click the "Configure DHCP Server" button. > > > This is covered reasonably well in the Installation docs. > > > > > > Good luck, > > > Vann > > > > > > On Thu, 2002-12-12 at 03:05, [EMAIL PROTECTED] wrote: > > > > Dear helper: > > > > > > > > I am new here and I had a problem to install oscar 2.0. > > > > I try to build a cluster myself. I was confused on step 6 setup > networking. > > > > I was able to collect mac, then I "stop collecting" and click > > > > "configure DHCP server". After that I am not sure what to do next so I > > > > tried click "setup network boot" and switch to my client node and > > > > started PXE again. However no matter what I do, I cannot install > > > > client. I also tried boot from floppy. The boot log is attached. > > > > I have 3Com 3c905TMX NIC. > > > > > > > > Would you please help me on this one? What is the sequence I have to > > > > do? When should I switch to client to second boot from network or > > > > let the first network boot rnning? > > > > > > > > Thanks for help and best regards > > > > RX > > > > > > > > > > > > ------------------------------------------------------- > > > This sf.net email is sponsored by: > > > With Great Power, Comes Great Responsibility > > > Learn to use your power at OSDN's High Performance Computing Channel > > > http://hpc.devchannel.org/ > > > _______________________________________________ > > > Oscar-users mailing list > > > [EMAIL PROTECTED] > > > https://lists.sourceforge.net/lists/listinfo/oscar-users > > > > > > --__--__-- > > _______________________________________________ > Oscar-users mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/oscar-users > > > End of Oscar-users Digest ------------------------------------------------------- This sf.net email is sponsored by: With Great Power, Comes Great Responsibility Learn to use your power at OSDN's High Performance Computing Channel http://hpc.devchannel.org/ _______________________________________________ Oscar-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/oscar-users
