We are not using shared ZFS (or HFS for that matter). We merged three plexes into 1 a couple of years ago (for charging reasons), and we only share the minimum dasd to allow the plex to work (Couple/logger etc). All systems in the plex are in the same MII-plex, but have their own MAS, RACF d/b, SMS environment, Unix environment. We cannot use shared ZFS, as all systems need to have access to any datasetst that are shared and this is not how our dasd is setup. We are actively working toward a single RACF d/b, single JES environment but slowly!
The common MII environment means that we occaisionally get some false contention for same named datasets but is manageable. We never saw contention with HFS - I need to check the propogation of SYSZDSN in MII, but now we are seeing contention with ZFS - I assume a Vsam major name is getting propogated for ZFS when (perhaps) SYSZDSN was not - -need to check tomorrow. That said, I can put a system name in the automount policy, but if a batch job runs on a different system it will store data into different datasets and therefore potentially not have the latest data. e.g. job ABC runs with USER=DEF on system A and the automount causes ZFS file UNIX.SYSA.DEF to be mounted. If the job saves data in this ZFS and later runs on system B (which will mount UNIX.SYSB.DEF), it will not have the latest version of the data. Andrew ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
