In response to the email... " The Oracle DBA indicates that Oracle is easier to implement under Red Hat" - I'm interested in what this means precisely, as we should be pretty similar in each environment - we only have the one code set and installer. If there are significant differences that people are finding (on Z) then I'm happy to look into it. I got the following response...
Part of the reason for my preference is "the Devil you know". I've managed servers with RHAS and RHEL for a number of years. Also, RHEL and RHAS use commands and configurations that are very close to Solaris which I also managed for many years. There are just differences in the way that SUSE works from RHEL and Solaris which are different than the systems that I know well. SUSE is one that I do not know well. The problem that I found with Oracle on SUSE on ZVM is that Oracle did not have a clean install. The base version that was available for SUSE was Oracle 10.2.0.2 and it did not install cleanly on SUSE on ZVM. I did not get a usable database until I installed the 10.2.0.4 upgrades. Also, there was a large amount of manual verification that I had to do because the installation could not properly verify the environment. Now, I'm not sure if this was because of SUSE or ZVM, but it is time consuming and frustrating, regardless. The second problem occurred when I was trying to perform a datapump import using the parameter REMAP_TABLESPACE. This parameter allows me to do a full import of a database and create the tablespaces that do not exist in different locations from the database where the export was performed. Being able to use this functionality is a time saver for me because I do not manually have to create the tablespaces before the import is performed. Some of my database contain many tablespaces and creating them manually before an import is very time consuming. When I tried to use the REMAP_TABLESPACES parameter on the Oracle database that was created on the SUSE OS, it failed miserably. I tried for a day and a half to make it work, but it just did not. I've used this parameter and process on Windows, Solaris, and RHEL OS's with no problems and so I thought SUSE was the issue. However, after much thought it may just be the Z/OS platform that has this bug in the datapump. If it is the platform and not the OS, then it really does not matter which OS is used. If the platform is the issue, then porting new databases to this platform will just be a more manual effort than on our existing platforms and on any other platform that I've used in the past, too. [Ruddy A. Melancon] ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 ---------------------------------------------------------------------- For more information on Linux on System z, visit http://wiki.linuxvm.org/
