I was building my own Oracle RAC 11g cluster (18.104.22.168) with Openfiler as my shared storage. Two VM's as my rac-cluster, one VM with Openfiler as my own SAN. Followed the instructions of the article of Jeffrey Hunter how to build a RAC on OEL and iSCSI, however this article is not based on VM. My Openfiler-VM was ready, given access to my both nodes on iSCSI-level, and now I want my rac1-node to discover the iSCSI-drives on my Openfiler.
And now for something quite different…. Some monitoring-scripts are mailing to me, for example the backup-script (will do it later on in Grid Control….). For a few nodes in the DMZ this didn’t work anymore. And actually I don’t want to know what is going on, but at the place I work there’s no Linux admin in the neighbourhood. So I can give it a try…
Appeared that the network-boys (no girls..) changed some configuration (mx-records or something like this). Logging and “mailq” gave ‘Name service error’.
Suse 9, x64-bit. Solution:
Een datawarehouse-achtige database heeft zo’n 278 scheduler_jobs lopen. Database gaat down voor onderhoud, blijft een avondje uit. Na opstarten van de database blijkt dat de next_run_date van de meeste jobs kleiner is dan de sysdate, en de jobs niets meer doen.