Dave > I missed "session manager" if it was in your original post.
It wasn't. Very little was! > I don't know much about running "cross domain" or Enterprise Extender of the Hipersockets, but as it is just another TCPIP, if it's possible over wire, it's possible over "memory". Actually Enterprise Extender (EE) is not really "just another TCPIP" - or even "just another UDPIP" to respect better that the transport protocol is UDP rather than TCP. In fact, to the IP component of z/OS Communications Server, EE appears to be another CS IP instance - "same host". Other than that all I can say is that it's all a bit special. It is nowhere near as simple as AnyNet SNA over IP was where VTAM really did make the logical link look like "just another" TCP application. HiperSockets is a facility which operates between LPARs within a CEC - quite unrelated as to whether or not the LPARs happen to have been defined to reside in the same sysplex but which relies on the definition structures created for communication between members of the same sysplex - just to keep everyone on their toes! It is normally allowed to be set up dynamically using the IPCONFIG statement DYNAMICXCF parameter and its subparameters. "Cross-domain" as a term used by SNA and thus by VTAM is any communication between nodes which are not traditional SNA peripheral nodes supported by a "boundary function". There are not so many of this type of SNA node left any more. An example of a traditional SNA peripheral node supported by a "boundary function" would be a 3274 or 3174 supporting only (the appearance of) display and printer devices. > And as has been pointed out, a short fiber out one FICON (or ESCON if older) and into another can support CTC between your LPARs and get you connections just like you had before with the CTCA. If there is only the one CEC involved here, HiperSockets is the obvious way to go I would have thought so that all this discussion of bits of wire/glass external to the "box" ought to be moot. > And, Chris, actually I meant, just fire up another QWS3270 (or whoever you use) session on the desktop and connect directly to the next LPAR. Did I manage to cover what you had in mind as a TN3270E server on presumably every system so that TSO could be accessed on every system? Incidentally, the TN3270E server could be an OSA-ICC feature. > If we didn't already have a session manager (Netpass) here, I wouldn't spend a dime to acquire one. As it is, as long as it runs, we will keep it do to the disruption removing it from the existing user community would cause. If you had a reliable way to distribute software to your user community, you could arrange that the sort of selection that the user performs using a session manager could be performed using GUI-based programming on the workstation. This is a bit of a guess since I have never done - nor indeed seen - such an arrangement but surely it is possible, isn't it? > ... we will keep it do to the disruption ... That'll be "due" not "do" although I appreciate in the American dialect the "y" before the vowel has atrophied! It is "credit where credit is due" isn't it even in Microsoft-land? Google supports "due" over "do" 10 to 1. Chris Mason On Wed, 14 Jul 2010 13:12:58 -0700, Gibney, Dave <[email protected]> wrote: >I missed "session manager" if it was in your original post. > >I don't know much about running "cross domain" or Enterprise Extender of >the Hipersockets, but as it is just another TCPIP, if it's possible over >wire, it's possible over "memory". > >And as has been pointed out, a short fiber out one FICON (or ESCON if >older) and into another can support CTC between your LPARs and get you >connections just like you had before with the CTCA. > >And, Chris, actually I meant, just fire up another QWS3270 (or whoever >you use) session on the desktop and connect directly to the next LPAR. > >If we didn't already have a session manager (Netpass) here, I wouldn't >spend a dime to acquire one. As it is, as long as it runs, we will keep >it do to the disruption removing it from the existing user community >would cause. > >Dave Gibney >Information Technology Services >Washington State University > > >> -----Original Message----- >> From: IBM Mainframe Discussion List [mailto:[email protected]] On >> Behalf Of Neal Eckhardt >> Sent: Wednesday, July 14, 2010 12:38 PM >> To: [email protected] >> Subject: Re: VTAM Cross domain >> >> On Wed, 14 Jul 2010 11:39:57 -0700, Gibney, Dave <[email protected]> >> wrote: >> >> >Hipersocket and then you can use TSO telnet to reach the other >LPAR. >> > >> >But, I have to ask, why not just have them start another 3270 >emulator >> >session to the other LPAR? >> > >> >> Because it's easier and cleaner to just put an entry into the session >> manager, >> and let it find the software running in another LPAR through the cross >> domain >> connection. >> >> Neal ---------------------------------------------------------------------- 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
