Slow data selection after RHel7 server inclusion

[SAPWorkerL]

We recently changed our production environment to run on a mix of Linux RHEL6 and RHEL7 servers, and at about the same time, we began to experience significant performance issues with our batch jobs, with table row extraction times increasing four to tenfold.

The affected programs have not changed, so that is not a likely source of the problem. Our basis team has followed all the available SAP guidance on implementing RHEL7 but to no effect. Database reorgs/stats improve performance slightly for some items, but it then rapidly degrades.

Does anybody have any experience with RHEL7 and RHEL6/RHEL7 mixed environments that might help us? Any guides to using the conf files for use with Oracle might help also. We have checked on indexes already too.

[SAPWorkerL]

Thanks for replying! We figured it out. The exception list did not exist for the filesystem backup so it was trying to back-up the ever changing database files (over 13TB) and filling up the network card so nothing could get through. An exception list was created exclude database files and everything runs as it did before.

[JamesZ]

Hi Liza,

A quick judge, is the bottleneck from Oracle database or SAP? What is the st03 telling us?

If it is oracle slow, please do check:

1. whether you patched SBP during migration
2. whether oracle parameter are tuned during migration

Best regards,
James