Do I understand you correctly that if I use IUCV *IDENT <resname> the resname should be the filepoolid? And if I do that for any *local* pool, and then the fact that there is a global pool of the same name will not cause any problem?
What about in the other direction? Is the REMOTE sfs server affected when a LOCAL server is known in the group? I seem to remember that earlier today, when BOTH systems got the HCPACR2733E msg, only the pool defined as LOCAL crashed (after meeting the REMOTE pool), but the one defined as REMOTE kept on working as usual (even though it also got that msg). Is that correct? Thank you for your help! Shimon > > > When I left in RESANY GLOBAL the two systems still seem to > > interact, even if I have LOCAL in the DMSPARMS file. > > The other system got: > > HCPACR2733E IDENTIFY REQUEST FROM NODE <othernode> FOR > > RESOURCE YPOOL IS A DUPLICATE -- RESOURCE REJECTED. > > So, LOCAL in the parms file does NOT seem to prevent network-wide > > announcement of the pool. > > Correct. Only VMSYSxxx filepools are local from CP's perspective. All > others create global APPC/VM resources. The LOCAL parameter in DMSPARMS > causes the SFS server (not CP) to reject connections from remote users. > > > One of the servers (the 'production' one) *must* be defined > > as REMOTE/GLOBAL. Is there ANY way to run a local service > > with the same name and tell VM that I *only* want to > > access the local one? > > No. COMDIR is the only way to redirect the connection. > > If you want to prevent resource revocation, use IUCV *IDENT <resname> > rather than RESANY. > >
