Re: [Oscar-users] PBS configuration failure during post_install (OSCAR4+FC2)

[Shiang-Tai Lin]

Hi Bernard, I am using Fedora Core 2. Yes, I can ssh to the client nodes with no problem. The rpm results are [BTW, I re-installed OSCAR the second time and chose the Torque package in step 1, hoping that the results would be different. But it was the same.] # rpm -qa | grep torque torque-client-1.0.1p5-2 torque-server-1.0.1p5-2 torque-gui-1.0.1p5-2 torque-mom-1.0.1p5-2 torque-1.0.1p5-2 # rpm -qa | grep pbs # I list the warning messages from the oscarinstall.log below. Hopefully this may help diagnose the problem. Thanks, Shiang-Tai ... update-rpms                 ################################################## WARNING: Appears oscar-httpd was not installed. Defaulting to local filesystem. Creating RPM cache database... update-rpms, version 1.1.14 ... Starting ntpd: ESC[60G[  ESC[0;32mOKESC[0;39m  ] WARNING: OSCAR does not know how to configure sendmail yet. WARNING: Please bug the OSCAR developers to finish the disable-servics package! WARNING: There will be no mail service running on the client nodes! --> Successfully ran server non-core package post_server_install scripts ... --> About to run /opt/oscar/packages/disable-services/scripts/post_install for disable-services ************************************ WARNING ************************************ OSCAR could not set up the configuration for any mailing service on the server. The current version of the disable-services packages in OSCAR only supports the Postfix mail transfer agent (MTA). It looks like you have another MTA installed (e.g, sendmail or exim); as such, please be aware that OSCAR will not automatically configure it. ************************************ WARNING ************************************ --> About to run /opt/oscar/packages/opium/scripts/post_install for opium ... Bernard Li wrote: Hi Shiang-Tai:   Yeah, by booting the nodes with a bootdisk it should re-image the nodes with the new image...   It appears that the PBS server is running...   After the nodes have been imaged, can you ssh to them without problems?  Also, you're encountering the problems in Step 7 (Complete Cluster Setup) right?   What does:   % rpm -qa | grep torque % rpm -qa | grep pbs   on the server node give you?   You _don't_ need Torque for the cluster, but without a resource manager the cluster isn't really intelligent in terms of managing workloads.  Besides, it should work, so something must be wrong here.   I forgot, what base Linux are you using?   Cheers,   Bernard