Rob Yes, I know I'm not dealing with topics you have raised sequentially.
I could be accused of being facetious if I were to reply that, from my home PCs, I use a name server implementing "BIND" for the purposes of resolving the names in the URLs I and my wife select throughout the day. But, assuming my ISP's name server follows "BIND", it would be true. That's a fair assumption because use of "BIND" for name resolution is a very popular choice: http://en.wikipedia.org/wiki/BIND <quote> BIND ... was as of 2004[update] the most commonly used Domain Name System (DNS) server on the Internet, and still proclaims itself to be so.[1] On Unix-like operating systems it is the de facto standard. Originally written by four graduate students at the Computer Systems Research Group at the University of California, Berkeley (UCB), the name originates as an acronym from Berkeley Internet Name Domain,[2] reflecting the application's use within UCB. </quote> Note that we have another "de facto" standard - ho hum ... It is of course evident from this post that you have in mind the name server function provided by the z/OS Communications Server (CS) IP component which will cease to be supported after V1R13, the function, that is! In any case, I see the others responding to the initial question - I think including Edward Jaffe and maybe including Mike Ward - are referring to the z/OS CS BIND server as you intended. The problem, as I see it - and it may well be that I have not seen something all potentially affected customers have, is that the guidance on "plans to replace functionality" hasn't been spelled out fully yet. This reference to "Linux for System z" looks like a stop-gap. I would hope to see a redbook before too long which goes through a few most likely choices with a worked example of an implementation in each case. I looked over my now obviously rather less useful write-up for a customer describing how to set up a name server on OS/390 Communications Server from nearly 10 years ago. Out of 8 pages, 5 concern the files and how they interrelate so I guess that part still has some validity in whatever UNIX-like environment is selected for a replacement for a *primary* or *secondary* name server on a platform other than z/OS. > Customers who currently use or plan to use the z/OS BIND 9.2.0 function as a caching-only name server should use the resolver function, which became generally available in z/OS V1.11, to cache Domain Name Server (DNS) responses. As usual - as very usual - ambiguity abounds. "Caching" is an additional function which became available in z/OS V1R11 and is *added* to the existing resolver function as implemented though the RESOLVER address space which has been with us - without the option *not* to use it - since z/OS V1R2. What's interesting about both the RESOLVER address space *and* this "caching" function is that you get both "at no extra charge", that is, no additional energy is required to be fed through the brain of the systems programmer responsible for the IP component of z/OS Communications Server in order, should he or she care to look, to introduce caching of resolved names. [1] It just comes with the potentially ignored RESOLVER address space. I guess this is an example of IBM dealing with a customer's problem before the customer has a chance to realise there is - or will be - a problem. On the surface this looks like a "mushroom" story but, in the case of "caching", it's more like getting the plants into the greenhouse before the storm hits. There is a very extensive - at the highest redbook standard one could say - section on "Resolver caching" in the z/OS CS Configuration Guide (CG). I have chosen the V1R11 CG which introduces the topic as the URL below: http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/F1A1B391/1.2.6 Although, perhaps unwittingly, you have been force-fed the resolver function you can reject the additional generosity of the resolver *caching* function, if you insert a NOCACHE statement into whatever data set contains the generically named TCPIP.DATA statements, often by means of a SYSTCPD DD- statement, for whichever individual typically server applications you feel do *not* need resolved names to be cached. Chris Mason [1] Some may have noticed that recently I had a compliment regarding my contributions where Communications Server topics potentially needed to be addressed - "potentially" since the most recent has turned out to have been largely a waste of effort due to insufficient preliminary research. However, the word "brittle" intervened. I was a bit puzzled by what this might mean until I found myself composing the above. I guess, as a teacher, I have got into the habit of trying to detect where emphasis is needed and so a slightly combative form of words is chosen. So, if your reaction is "that's a bit sharp", it's more likely that you will see the point. I don't think it works so well the other way round and in any case, I'm not in a position to run an interactive quiz and give out sweets or little mouse- mats or whatever to the "students"! On Tue, 15 Feb 2011 13:29:54 -0500, Rob Schramm <rob.schr...@gmail.com> wrote: >I was curious about plans to replace the functionality. > >Statement of Direction >"z/OS V1.13 is planned to be the last release in which the BIND 9.2.0 >function will be available. Customers who currently use or plan to use the >z/OS BIND 9.2.0 function as a caching-only name server should use the >resolver function, which became generally available in z/OS V1.11, to cache >Domain Name Server (DNS) responses. Customers who currently use or plan to >use the z/OS BIND 9.2.0 function as a primary or secondary authoritative >name server should investigate using BIND on Linux for System z or BIND on >an IBM blade in an IBM zEnterprise BladeCenter® Extension (zBX)." > > >On Tue, Feb 15, 2011 at 1:26 PM, Ward, Mike S <mw...@ssfcu.org> wrote: > >> Yes >> >> -----Original Message----- >> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On >> Behalf Of Rob Schramm >> Sent: Tuesday, February 15, 2011 12:25 PM >> To: IBM-MAIN@bama.ua.edu >> Subject: Re: z/OS 1.13 preview >> >> Is anyone using BIND DNS to provide DNS resolution? ---------------------------------------------------------------------- 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
