Java on condor
Dr. R
interested in setting up gridapplication condor cluster that can run a
large scale Java simulation. Are the vm condor workers configured to
run Java jobs? Can the vm condor workers be easily configured to do
so, if not?
Also, I'm interested in learning about anybody's experience setting up
a cluster using cloud resources (e.g. futuregrid nimbus, ec2, etc).
This is all very new to me, interesting, and looks like work that
fills a definite need.
Thanks
Prof Rossetti
Kyungyong Lee
--
You received this message because you are subscribed to the Google Groups "acis.p2p" group.
To post to this group, send email to aci...@googlegroups.com.
To unsubscribe from this group, send email to acisp2p+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/acisp2p?hl=en.
Renato Figueiredo
Dr. Renato J. Figueiredo
Associate Professor
ACIS Lab - ECE - University of Florida
UF Site Director, Center for Autonomic Computing
http://byron.acis.ufl.edu
ph: 352-392-6430
Dr. R
independent appliance pool:
http://www.grid-appliance.org/wiki/index.php/Deploying_independent_appliance_pools_using_PlanetLab_bootstrap
I'm having some issues testing the grid appliance. Issues like the
condor system not starting up automatically, no condor_collector being
found, etc. I need to hack at it some more.
Once I can get a pool set up, then I'll work on getting Java going. I
think it is a matter of configuring condor to run java jobs and
installing Java on the workers. I'll probably need to enable
something in condor. It would be good if the GridApplication had some
information about the default Condor configuration that comes with
it. From your response it sounds like the java universe is not set
up. I'll have to dig into it.
Then, I'll move on the FutureGrid. The links you provided look
useful. I want to establish a gridappliance condor pool that uses both
my local independent pool and resources from FutureGrid.
Manuel D. Rossetti, Ph.D., P.E.
Professor and Associate Department Head
University of Arkansas
Department of Industrial Engineering
4207 Bell Engineering Center
Fayetteville, AR 72701
Phone: (479) 575-6756
Fax: (479) 575-8431
email: ross...@uark.edu
www: www.uark.edu/~rossetti
On Jul 25, 8:10 am, Kyungyong Lee <k...@acis.ufl.edu> wrote:
> Hi Prof. Rossetti,
>
> Thank you for your interest in the grid-appliance. Regarding running java
> simulations, java is not installed by default at every virtual machine. What
> we usually do when we run experiments are to prepare a executable file with
> all necessary libraries embedded to it. I'm not sure if it is possible for
> java. Other ideas from others?
>
> For FutureGrid cluster setup, I'm not sure how much you are familiar with
> FutureGrid, Nimbus and your specific requirements for setups, but the
> following link can be a good starting point for a beginner. We also have
> pre-packaged grid-appliance images and tutorials in the FutureGrid tutorial
> site. If you need more helps or specific requirements, please let us know
> them.
>
> https://portal.futuregrid.org/tutorials
>
> In the above page, I recommend you to see
> Nimbus One-click Cluster Guide <https://portal.futuregrid.org/tutorials/nm2>
> Deploying virtual private Grid Appliance clusters using
Dr. R
Thank you for your response. It sounds like I need to follow the
condor instructions about configuring to run java jobs and set setup
my worker images accordingly.
As you can see from my response to Kyungyong Lee, I'm in the process
of setting up a local private condor pool using gridappliance. I'm
still having some issues but will then move onto FutureGrid.
I'm wondering if someone would be available to answer a few questions
if I can't get over the hump.
Thanks
Manuel
Manuel D. Rossetti, Ph.D., P.E.
Professor and Associate Department Head
University of Arkansas
Department of Industrial Engineering
4207 Bell Engineering Center
Fayetteville, AR 72701
Phone: (479) 575-6756
Fax: (479) 575-8431
email: ross...@uark.edu
www: www.uark.edu/~rossetti
> I believe it also works if you install Java in the read-only NSF-mounted
> partition each grid appliance exports. The image on FutureGrid which runs
> Hadoop has Java installed, if you want to give it a try.
> --rf
>
>
>
>
>
>
>
>
>
> On Mon, Jul 25, 2011 at 9:10 AM, Kyungyong Lee <k...@acis.ufl.edu> wrote:
> > Hi Prof. Rossetti,
>
> > Thank you for your interest in the grid-appliance. Regarding running java
> > simulations, java is not installed by default at every virtual machine. What
> > we usually do when we run experiments are to prepare a executable file with
> > all necessary libraries embedded to it. I'm not sure if it is possible for
> > java. Other ideas from others?
>
> > For FutureGrid cluster setup, I'm not sure how much you are familiar with
> > FutureGrid, Nimbus and your specific requirements for setups, but the
> > following link can be a good starting point for a beginner. We also have
> > pre-packaged grid-appliance images and tutorials in the FutureGrid tutorial
> > site. If you need more helps or specific requirements, please let us know
> > them.
>
> >https://portal.futuregrid.org/tutorials
>
> > In the above page, I recommend you to see
> > Deploying virtual private Grid Appliance clusters using Nimbus<https://portal.futuregrid.org/tutorials/ga4>
>
> > Best,
> > Kyungyong Lee
Renato Figueiredo
Kyungyong Lee
Dr. R
Great, that would be very helpful. I'm assuming you are referring to
the process briefly mentioned in this presentation:
http://www.grid-appliance.org/files/docs/edu-docs/GridApplianceCustomize.pdf
Manuel
Dr. R
the montepi example. Everything seemed to work.
I'm having some issues with the VMs. Stopped everything and restarted
but condor just "configures" for a long time. I'll keep
investigating. Doesn't seem like the condor collector is running.
I'm wondering how to log into the server and worker nodes. I didn't
see anything about their default user/password settings. Perhaps a
pointer to some docs...
Once things stabilize, I'll work on the java...
Kyungyong Lee
Did you try to log in using your grid-appliance ID and password
through VMWare player screen? If you could not log in using your
grid-appliance id, can you try using an id "griduser" and the password
that I talked?
Regarding VMWare server problem, I did not use VMWare before, so I
have no idea about it. Does anyone else have an idea?
Best,
Kyungyong Lee
On Sun, Jul 31, 2011 at 6:48 PM, Rossetti Manuel
<manuelr...@gmail.com> wrote:
> Kyungyong,
> Thanks for the note. I have been able to get the pool back up. I seem to
> be having an issue using VMWare Server. I can get it working when using
> VMWare Player but when I try to use VMWare Server for the worker and server
> nodes, condor never runs (i.e. the collector never responds). If you have
> any ideas about that it would be useful.
> I have attempted to log directly into the server or worker vm instances
> using my grid-appliance user id and password but it doesn't work. I think
> I need to log in using the local host via ssh??? but I don't know what IP
> (address) to use. Is there some documentation?
> Thanks
> Dr. Rossetti
> On Jul 30, 2011, at 8:44 PM, Kyungyong Lee wrote:
>
> Hi Dr. Rossetti,
> The default login name is your ID at the grid-appliance.org. The password is
> . If it does not work, please let me know. Because of the very
> easy password, I sent the e-mail only to you at this time. We will make a
> policy to reveal the default id and password. BTW, does your condor pool
> still not work after reboot? Thanks.
> Best,
> Kyungyong Lee
David Wolinsky
The only way to access a worker and server is by pre-installing your
ssh key into authorized_keys in the floppy.
unzip floppy.zip
mount -oloop floppy.img /mnt
cat key.pub >> /mnt/authorized_keys
umount /mnt
zip -r9 floppy.zip floppy.img
Also have you checked /var/log/monitor.log ? Which version of the
Grid Appliance are you using? I am not sure if I got side tracked
when I was upgrading the last version....
Cheers,
David
Kyungyong Lee
Regarding logging in problem to worker and server virtual machines, if
you still work on that issue, I recommend you follow below steps.
1. back up your worker and server floppy image.
2. Replace the worker and server floppy image to the default
floppy.img that was originally embedded in the grid-appliance zip
file.
3. Boot the VM, upon booting completion, turn it off.
4. Switch the default floppy.img file to worker or server floppy image
and restart VMs.
5. You can log in using "griduser" account.
If it does not work for you, please let us know. Thanks.
Best,
Kyungyong.