There is no problem in sharing same osa device addresses and osa card between lpars. . We share them for years, but between z/os lpars. It is simply like sharing dasds. I'm sure, because I do the hardware definitions (iocp, mvscp, hcd definitions). In my opinion, using the same device number for the same device is a simple and good method , prevents errors.
What I suspect is, linux/390 does not like to have the same device number with others. Am I right in this??? (question 1) By the way, is there an equivalent of mvscp OS config ,/EDT(eligible device numbers) for linux/390???? (question 2) Where do I define them? I mean, where do I say the Osa device number ? I say it in answering the network related questions during ipl, is it enough? Should I mention it in the parmline file. (question 3) In z/os we do define the device numbers for each operating system(lpar) with mvscp or hcd. This is different from iocp, iocp is more hw. Mvscp defines the operating system the EDT list and type of device addresses it uses. I learn a lot from this discussion, thanks a lot. inci ----- Original Message ----- From: "Vic Cross" <[EMAIL PROTECTED]> To: <LINUX-390@VM.MARIST.EDU> Sent: Wednesday, September 06, 2006 10:28 AM Subject: Re: can LINUX/390 and z/OS LPARS share devices? > Post, Mark K wrote: > > Doesn't that require "playing games" with genning different subchannel > > addresses to different LPARs, but using the same device numbers in those > > different LPARs? > > Not that I've been the bunny responsible for IOCDS, but as I recall it > was pretty simple and "standard EMIF" to map the same device address > numbers from each LPAR to different UNITADDR/subchannel. I think it was > documented in the OSA Express Redbook/Redpaper and might even be > mentioned in the "zSeries Connectivity Reference". > > I once helped a hardware buddy do an IOCDS for EMIF CTCs, which was > significantly more mind-warping but along the same lines. I think it's > just a case of how much complexity you put into the IOCDS to make it > easy for the network people to use the devices. > > Alan Altmark wrote: > > You *can* have the same device number in different LPARs, but sysprogs > > who > > do that are eventually found at the bottom of the nearest lake. > > To quote a former politician from down this way, "please explain?" I've > never found any need to over-complicate things by having different > device numbers for different LPARs (over-complicating things for z/OS > network people is a dangerous exercise IMO, which comes mainly from the > time when I was one :D ). If they want to talk to the first OSA, they > put in E000 (e.g.) in the VTAM TRLE for TCPIP -- works for all LPARs > (also can be done across different CECs, which is great for DR). Second > OSA is E100, and so on. Of course, like I said, I wasn't doing the > IOCDS... :) > > In fact, if my sysprogs *had* tried to push different device addresses > on me, I would have been looking for that lake you mention and telling > said sysprogs how fashionable concrete shoes are this season. :) > > When you're sharing DASD, you don't have different device numbers to > refer to a pack from different LPARs, right...? Why is it a bad idea > for OSAs? > > Cheerio, > Vic Cross > > -- > This message has been scanned for viruses and > dangerous content by MailScanner, and is > believed to be clean. > > ---------------------------------------------------------------------- > 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 > ---------------------------------------------------------------------- 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
