Jeff, We used to do this until recently, it worked well and we would be still if someone had not done a silly that would have caused too much effort to resolve.
When you are running stable versions of RACF on both Z/OS and Z/VM then there is no issue. The thing that you have to be aware of is that the database has to be updated to the level of the highest level of RACF using the database. Up until recent years it was always Z/OS that was ahead so they always applied the RACF version updates but, with Z/VM 5.4, this was ahead of the RACF server on Z/OS so we had to do the database update. What happened to us is that the Z/OS folks came to upgrade to 1.7 and applied their RACF database updates (because they always had and they didn't need to coordinate). This, in fact, downgraded the RACF database and caused some pointer corruption. Their reaction was, of course, that we had corrupted "their" database and that we should get off it and sort out the corruption "that we had caused". That is the only gotcha I know about. Colin Allinson VM Systems Support Amadeus Data Processing GmbH
