>>> On 2/28/2008 at 8:57 AM, in message <[EMAIL PROTECTED]>, David Boyes <[EMAIL PROTECTED]> wrote: >> When setting up the IP address for a hipersocket I am curious as to if >> people are giving it the same IP address as with the regular outside > of >> the mainframe (OSA or whatever) IP address. > > Absolutely DO NOT do this. Each interface needs a unique address (and > IMHO, a unique DNS name). The whole premise of IP routing and network > function is based on this concept. > >> We have TCP/IP stacks with >> hipersockets running on VSE, Linux and z/OS. On some of the VSE > stacks we >> use the same IP address for the hipersocket as we do for the OSA. On > a >> few other VSE stacks we give them separate IP addresses, and we do the >> same (different addresses) for all of the Linux and z/OS stacks. > > The only reason you get away with this on hipersockets is that they > never see the outside world (they act like a isolated hub). You are also > not likely to actually get any benefit -- in fact, I would suspect that > you're not using those interfaces because the OSA initialization is > silently overriding it due to IP address conflict. > >> How do >> other places do it? > > Every interface has a unique address and DNS name. If I want > interface-independent stuff, that's what VIPA is for -- and the VIPAs > all have unique IP addresses and unique names. > > Example: > > Linux guest with OSA and hipersocket interfaces, plus VIPA to identify > access to specific service, runs in VM userid VA1TRS89: > > (addresses replaced with letters) > > OSA interface: > > va1trs01-osa0.srv.va1.sinenomine.net: x.y.z.a > > Hipersocket interface: > > va1trs01-hs0.srv.va1.sinenomine.net: p.q.r.s > > VIPA: > > prod-smtp.app.va1.sinenomine.net: e.f.g.h > > > Note ALL are unique addresses, and the OSA and hipersockets are in > DIFFERENT subnets. The VIPA is also in a different subnet, advertised by > specific host routes. > > Users are only given the VIPA name (note, NAME, not address), and that > can be moved around by simply manipulating the DNS. > > >> And is there any particular reason? > > Any other solution is a network routing loop looking for a place to > happen. IP routing is based on knowing a destination that you CAN reach > to send packets to destinations that you don't know how to reach. > Without unique addresses, the whole thing falls apart. > > (the only reason that your setup hasn't already fallen is that some of > the older IBM stacks permitted this configuration error. The new ones > don't.)
Sounds good, David. Thanks for all of the information. Frank -- Frank Swarbrick Senior Systems Analyst - Mainframe Applications Development FirstBank Data Corporation - Lakewood, CO (303) 235-1403 The information contained in this electronic communication and any document attached hereto or transmitted herewith is confidential and intended for the exclusive use of the individual or entity named above. If the reader of this message is not the intended recipient or the employee or agent responsible for delivering it to the intended recipient, you are hereby notified that any examination, use, dissemination, distribution or copying of this communication or any part thereof is strictly prohibited. If you have received this communication in error, please immediately notify the sender by reply e-mail and destroy this communication. Thank you. ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
