Posts by Yury Velikanov
For last few months I was part of the Pythian’s team helping implement Oracle 11gR2 Clusters for different Oracle customers. All the implementations had different requirements and configurations, however, in all cases, the client’s DBAs made the same mistake over and over again. They used sqlplus and lsnrctl utilities to manage (start/stop) databases and listeners. This is totally wrong in the 11gR2 Cluster world. The following commands are just few examples on how you should start/stop Oracle processes
I just recently faced an issue on how to automate the Workflow Mailer override address for development and testing environments during the EBS cloning process. The problem is that the only official way to set the override address in the latest ATG releases is to go through the OAM portal and change the override email via the GUI interface. I came to this solution.
Just recently I came across another unclear error message while troubleshooting an Oracle Instance registration with SCAN Listener. This post is one of the posts which may save quite a bit of time for people who will hit the same issue.
If you ever had tried to switch Oracle net trace than you know that it is quite straight forward to switch it on. You would say that I need to set the following parameters in the sqlnet.ora file…
If you have tried to automate EBS cloning then you definitely know that you can’t get through without hacking and customizing the Oracle-provided Rapid Clone utility. One of the areas we should “hack” the code to make it running in batch scripts is the services startup bit. The problem is that by default straight after Rapid Clone is completed Oracle starts all the services on the Apps node and until now there wasn’t a way to change it.
The Oracle OpenWorld 2010 conference is over and life is getting back to normal. However, this is a traditional time for the OOW after-parties and local meetings to discuss the results of the biggest annual event in Oracle space. Our Sydney Oracle Meetup is not an exception.
One of the tasks that I was involved in was 700GB database emergency cloning for an important functional issue troubleshooting. It was 10.2.0.5 (10GR2) 64 Bit database on Linux. To make a long story short I gave the DUPLICATE command a try. I used the comand in this post to start the process. After 4h of the restore process I was horrified (you know how it is to work under time and management pressure) at seeing the following errors
I’ve just come across an old known issue again today. Was working in the greatest and latest 12.1.2 Oracle e-Business Suite environment. Surprisingly, I face the old issue when WF_LOCAL_ROLES and WF_LOCAL_USER_ROLES objects (among other objects) are set to NOLOGGING mode. The easiest solution is to switch the whole database to FORCE_LOGGING mode:
As part of my preparation for the Exadata OCP Exam, I was reading through the “Oracle® Exadata Storage Server Software User’s Guide”. I found the two famous events in the Exadata world, “db file scattered read” and “db file sequential read” that cause so much confusion, have been renamed to “cell multiblock physical read” and “cell single block physical read” accordingly. Hopefully Oracle will port those events’ names to the RDBMS world as well in version 12c.
If it happens and you are in and around the Sydney area 1-2 weeks from now then you might be interested in taking part in these events.