Kees Vernooy writes: >This will only synchronize the time between the boxes, not >with the 'real'time. I suppost the windows/unix server is >synchronized with the 'real' time and the OP want to synchronize >his z/ system to the real time this way, which is not possible
Good point. Yes, agreed. It depends on what the original poster wants. If he wants time consistency across systems -- to the degree NTP can provide consistency anyway -- then System z as master NTP server, even without STP, will accomplish that. (And a lot of organizations run that way.) The System z clock is much better than what you'd find in most distributed UNIX and Windows servers, and if the occasional manual adjustment is OK this all works fine. (Running on UTC is also a good idea.) If he wants the System z hardware clock to maintain "true" time, whether obtained via NTP or other time source (e.g. NIST dial-up), then he will need STP. It is possible for one, some, or all Linux guests running on System z to be NTP clients without STP. But their "time scope" will only be within the individual Linux instances. They won't be setting and resetting the System z master hardware timeclock. It may be possible to do the same for other mainframe operating systems. For example, it may be possible to take the NTP client code, tweak it slightly, compile it for z/OS, authorize it, and have it fiddle with the clock. Again, that's probably just for that instance of z/OS where it's running. One would also have to do a bit of thinking before authorizing such a clock-adjusting program. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Based in Tokyo, Serving IBM Japan / Asia-Pacific E-Mail: timothy.sipp...@us.ibm.com ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
