For your Dirmaint question: you need to"merge" the old and the new CP directory. I've got execs to help with this task, as usual(?) documentation is scarce: comments in the exec. I'll send them. - MRGDIR is the main one - it splits the new (5.3) directory in seperate files (one for each userid) - it checks which minidisks exist in the current CP directory and changes the address to avoid conflicts. - Dirmaint commands and RACF commands are prepared: ADD user for new userids AMDISKs for exsitng users. You are the master and decide what gets executed. - CHGMDPPF is the next one, it can update the PPFs to replect the changed minidisk addresses in te PPF files. Use only if you understand the principles.
2008/3/21, Kris Buelens <[EMAIL PROTECTED]>: > There may be extra migration steps, refer to RACF's program directory > to find out. There were for each major RACF release change, and I > would be surprised if there were now: pass phrase support may require > extra fields in the RACF database. > > 2008/3/21, Thomas Kern <[EMAIL PROTECTED]>: > > > Very Carefully... > > > > My feeling is the best way to move such pieces of data is to violate all > > security and do it in a standalone mode without any other security packag > > es > > involved. This might mean having to have management or security people > > looking over your shoulder at every keystroke. > > > > 1) Find out the volume, cylinder locations for your source and target, th > > ey > > match in size, right? > > > > 2) Shut down your old production system so there is NO chance of the RACF > > > > database being updated during its move. > > > > 3) IPL your rescue system (single volume disaster recovery system or your > > > > 3-5 volume maintenance system, some system where you have PRIVILEGES with > > out > > security). > > > > 4) Attach the source volume and the target volume to your running rescue > > system. > > > > 5) Use the DEFINE MDISK command to define both source and target minidisk > > s. > > > > 6) Use DDR to copy all cylinders from source minidisk to target minidisk. > > > > > > 7) ACCESS both minidisks read-only (ACCESS 200 C/C, ACCESS 201 D/D). > > > > 8) Compare the contents as best as you can, same number of files, same > > number of blocks, same MD5 Cyclical Redundancy Check? > > > > 9) Shut down your rescue system. > > > > 10) IPL your new production system. > > > > 11) Print out as much of your RACF database as possible and compare to th > > e > > printouts from before the move. (That's not part of your weekly reports? > > Then make that step 0, print out all of your RACF database.) > > > > 12) Backup your old system to tape. > > > > 13) Several weeks later, securely erase the DASD from your old system. > > > > 14) Several months later, securely erase those backup tapes. > > > > /Tom Kern > > /U.S. Dept of Energy > > /301-903-2211 > > > > > > > > On Fri, 21 Mar 2008 15:37:03 -0400, Hamilton, Brian <[EMAIL PROTECTED] > > om> > > wrote: > > > > >What is the best way to copy over my vm 5.2 racf database to my new vm > > >5.3 system? > > > > > >Also what is the best way to migrate dirmaint and its directly from vm > > >5.2 to 5.3? > > > > > >Thanks > > > > > > -- > Kris Buelens, > IBM Belgium, VM customer support > -- Kris Buelens, IBM Belgium, VM customer support