Database CI pipeline idea

1. Prod release
2. Remote online clone of Prod database using this technic. See the link above to the list of restrictions, but most influenceable are: source database must be in archivelog mode and have local undo (otherwise it should be in read-only mode), same endianness in source and target. The database link should exists from target to source.
3. Keep created clone as golden source. Use PDB snapshot copy to create databases for testing
   – parameter clonedb=true should be set
   – filesystem should supports sparse files (ext4, NTFS, … are ok)
   – as spare files are used, whole process is very fast and don’t consumes a lot of space