Dave I suspect, on reflection, Richard would be the first to say that his post lacked a bit of precision. He didn't so much mean the "Resolver" as what the resolver "pointed to", either the use of a name server or a "local file", of whichever is your preferred format, or both and which first and which second.
Actually, assuming you are the only one who would mess about with such a thing as the "local file", your left hand should have some idea what your right hand is doing and so you really should know whether you had cast a local version of "localhost" to the four winds. So, on this assumption and the assumption that folk over whom you have little control are supposed to take care of the name server, my money is on name server carelessness. Incidentally, I see that, prompted by Richard's post, you have shown us how your resolver is configured and unfortunately with NOCOMMONSEARCH which, if you were managing a "local file" would mean you were still using the tedious antique format requiring a conversion utility before it could be used. This also tends to suggest that the antique "local file" is not used and you rely upon a name server for name to address resolution. I also discovered in scanning the Configuration Guide for "localhost" that what I covered in my last post is to be found in Chapter 13, "Domain Name System". Chris Mason On Mon, 25 Oct 2010 15:02:35 -0700, Gibney, Dave <[email protected]> wrote: >> -----Original Message----- >> From: IBM Mainframe Discussion List [mailto:[email protected]] On >> Behalf Of Richard L Peurifoy >> Sent: Monday, October 25, 2010 2:41 PM >> To: [email protected] >> Subject: Re: EntireX and LOCALHOST >> >> On 10/25/2010 4:23 PM, Gibney, Dave wrote: >> > I am hoping someone can shed some light and also, I might be able to >> > warn anyone about to take this path. >> > >> > We did a POR to activate MCL maintenance on our z9BC-L03. >> > >> > >> > After the POR and IPL, we had problems with EntireX from Software >AG. >> > This system consists of a "broker" address space and several RPC >> address >> > spaces. The communication is via TCPIP loopback. >> > >> > The RPC servers were configured to use LOCALHOST to make the >> connection. >> > Changing to 127.0.0.1 solved our problem. >> > >> > This software had apparently been running since 2007 with the >> LOCALHOST >> > specified in the parms. >> > >> > It took us several hours to determine what was wrong. Fortunately, >we >> > found this application was working in another LPAR where 127.0.0.1 >> and >> > also our internet facing IP address was used instead of LOCALHOST. >> >> For LOCALHOST to work, it needs to be defined in RESOLVER or in a DNS. >> Have you dropped a definition that defined LOCALHOST? > >Unchanged since 2008: >GLOBAL *Browsed 7 2008/05/27 2008/06/04 16:28:32 GIBNEY >RESOLVER *Browsed 3 2008/05/27 2008/06/18 14:18:54 GIBNEY >TRESDATA *Browsed 4 2008/05/27 2008/06/18 14:19:04 GIBNEY >TCPIP.WSUMVS1.PARMLIB(RESOLVER) >DEFAULTTCPIPDATA('TCPIP.WSUMVS1.PARMLIB(TRESDATA)') >GLOBALTCPIPDATA('TCPIP.WSUMVS1.PARMLIB(GLOBAL)') >NOCOMMONSEARCH > BROWSE TCPIP.WSUMVS1.PARMLIB(TRESDATA) >********************************* Top of Da >TCPIPJOBNAME TCPIP >HOSTNAME WSUMVS1 >; TRACE RESOLVER >DATASETPREFIX TCPIP > BROWSE TCPIP.WSUMVS1.PARMLIB(GLOBAL) - >********************************* Top of D >DOMAINORIGIN IT.WSU.EDU >NSINTERADDR 134.121.139.10 >NSINTERADDR 134.121.80.36 >NSPORTADDR 53 >RESOLVEVIA UDP >RESOLVERTIMEOUT 30 >RESOLVERUDPRETRIES 1 > >> >> Can you resolve other names from you z/OS system (try PING validname)? >Yes > >> >> -- >> Richard ---------------------------------------------------------------------- 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
