You can always try OSCAR 5,
but it is totally different from OSCAR 4 and at this point documentation is
somewhat lacking:
http://oscar.openclustergroup.org/filebrowser/49/branch
http://oscar.openclustergroup.org/filebrowser/49/branch
Please make sure you read the
README first if you plan to try it out.
If you run into issues,
please post them to the development mailing-list. OSCAR 5 has TORQUE
2.0.0p8 and Maui 3.2.6p14+.
Alternatively, you can also
test drive a vmware headnode appliance based on Fedora Core 5 and an older
revision from trunk:
Cheers,
Bernard
From: Usman Ahmad [mailto:[EMAIL PROTECTED]
Sent: Fri 21/07/2006 03:18
To: Bernard Li
Subject: Re: [Oscar-users] No Free Nodes
Hi Bernard,
I tried to delete the nodes afterwards, as they appear in the log. But
would OSCAR Release 5 help? Ths issue with the Torque server has been solved in
this release?
Bets Regards
Usman Ahmad Malik
On 7/17/06, Bernard
Li <[EMAIL PROTECTED]> wrote:
Hi Usman:I'm a bit confused - it appers that you have deleted the nodes from the cluster - at least this is what the following messages is suggesting:
--> About to run /opt/oscar/packages/sis/scripts/post_clients for sis
using ODA to read the OSCAR database for node and adapters information ...
reading SIS database for node and adapters information ...
Node oscarnode1 is listed in the OSCAR database as using sis as the installer, but node oscarnode1 is NOT in the SIS database, DELETING node oscarnode1 from the OSCAR database ...
Done deleting node oscarnode1 from the OSCAR database.
Node oscarnode2 is listed in the OSCAR database as using sis as the installer, but node oscarnode2 is NOT in the SIS database, DELETING node oscarnode2 from the OSCAR database ...
Done deleting node oscarnode2 from the OSCAR database.
Node oscarnode3 is listed in the OSCAR database as using sis as the installer, but node oscarnode3 is NOT in the SIS database, DELETING node oscarnode3 from the OSCAR database ...
Done deleting node oscarnode3 from the OSCAR database.Do you have any ideas why those nodes have been removed from the SIS database?I suppose we can try to re-produce this if it poses an issue, but most developers are now focusing on getting OSCAR 5.0 released.Cheers,Bernard
From: Usman Ahmad [mailto:[EMAIL PROTECTED] ]Sent: Thu 13/07/2006 02:33
To: Bernard Li
Cc: Tyler Cruickshank; oscar-users@lists.sourceforge.net
Subject: Re: [Oscar-users] No Free Nodes
Hi Bernard,Sorry for the delay, I have attached the logs you have asked for.RegardsUsman Ahmad Malik
On 7/6/06, Bernard Li <[EMAIL PROTECTED] > wrote:Can you please post your oscarinstall.log (compress it) as well as the full output of "Test Cluster Setup"?Thanks,Bernard
From: Usman Ahmad [mailto: [EMAIL PROTECTED] ]Sent: Thu 06/07/2006 00:00
To: Bernard Li
Cc: Tyler Cruickshank; oscar-users@lists.sourceforge.net
Subject: Re: [Oscar-users] No Free Nodes
The new release does not fix this problem, as soon as I add subsequent nodes, the tests start failing (torque shell tests and all subsequent tests). Whereas, the ganglia sees the nodes OK.What is the solution?RegardsUsman Ahmad Malik
On 6/29/06, Usman Ahmad < [EMAIL PROTECTED]> wrote:I am using the SLC-3.0.6 (kernel 2.4.21-37.EL.cernsmp), Yes, I clicked the complete cluster setup buttn as well. It successfully added the third node.I will try the new release and will let you know.RegardsUsman Ahmad Malik
On 6/29/06, Bernard Li < [EMAIL PROTECTED] > wrote:After you add a new node, did you click on "Complete Cluster Setup" button?Which version of Scientific Linux are you using?This is definitely not the correct behaviour, and needs to be fixed. Any chance you can try the latest nightly tarball for 4.2.1 and see if the issue persists?
http://oscar.openclustergroup.org/filebrowser/49/branchThanks,Bernard
From: Usman Ahmad [mailto: [EMAIL PROTECTED] ]
Sent: Thu 29/06/2006 00:54
To: Bernard Li
Cc: Tyler Cruickshank; oscar-users@lists.sourceforge.net
Subject: Re: [Oscar-users] No Free Nodes
Hi,I have encountered the same problem, one node works fine but as soon as I add another node, its says: "Not enough free nodes" Tests Incomplete, then the torque shell tests and other tests start to fail. Then I also did (like Tyler) the start_over and did everything from the scratch. It worked fine and test were passed. But when I added third node, everything messed up again. this is definetely a bug inmy opinion. I am using Scientitfic Linux with OSCAR 4.2RegardsUsman Ahmad Malik
On 6/27/06, Bernard Li < [EMAIL PROTECTED] > wrote:Weird.Well at least it works now :-)oscartst user gets created when you run "test cluster setup".Cheers,Bernard
From: Tyler Cruickshank [mailto:[EMAIL PROTECTED] ]
Sent: Tuesday, June 27, 2006 10:40
To: Bernard Li; oscar-users@lists.sourceforge.net
Subject: RE: [Oscar-users] No Free Nodes
Bernard,I have success! Due to several problems I decided to reinstall via the ./start_over script. I began again with the ./configure script and moved through steps 1-8 with ease. Previous problems with testing the cluster were no longer an issue. Thanks for your help!P.S. Before reinstalling, I also seemed to have problems with /home/oscartst. I ended up deleting ./oscartst. It looks like it gets created between step 6 and 7?Thanks again for your help.Hi Tyler:
> >>Q: Torque related logs in /var/spool/pbs:
> A: No torque logs found in any of the directories.
> ./pbs:
> /aux /checkpoint /mom_logs /mom_priv
> pbs_environment
> /sched_logs
> /sched_priv /server_logs server_name /server_priv
> /spool /undelivered
Actually the TORQUE related logs I want you to look into reside in
/var/spool/server_logs - poke around there and see if you find anything
strange.
Otherwise, everything else seems fine - I'm not sure why it says that
there are not enough free nodes when "pbsnodes -a" says they are all
free... T he key is to determine why shell test is failing for TORQUE.
P.S. You can ssh to/from headnode/compute node with root/user fine
right?
Cheers,
Bernard
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Oscar-users mailing list
Oscar-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oscar-users
------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________ Oscar-users mailing list Oscar-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/oscar-users
