INFO-VAX Fri, 04 May 2007 Volume 2007 : Issue 243 Contents: ANN: Alpha firmware update CD v7.3 released Any futher news on DecCampus equivalent for Itanium VMS in the UK? Re: Any futher news on DecCampus equivalent for Itanium VMS in the UK? Re: DCL Magic? DecNet Plus installation on VMS 8.3 Itanium Re: DecNet Plus installation on VMS 8.3 Itanium Re: detect unaligned data access Re: detect unaligned data access Re: Favorite movies of the VMS crowd? Re: Favorite movies of the VMS crowd? Re: Free used AlphaServers Java problem Re: List of VMS command file .COM domain names (was: Free email account VMSUSER Re: List of VMS command file .COM domain names (was: Free email account VMSUSER Re: List of VMS command file .COM domain names (was: Free email account VMSUSER. Re: List of VMS command file .COM domain names (was: Free email account VMSUSER. Re: Medical software vendor says it won't support OS on Itanium-based servers se Re: Noahs ark found! Re: OT: Favorite movies of the VMS crowd? Re: OT: Favorite movies of the VMS crowd? Re: OT: Favorite movies of the VMS crowd? Re: OT: Favorite movies of the VMS crowd? Re: SSH Keys for a heterogeneous Environment To re-link or not to re-link for VMS upgrade (Post contains short and long versi Re: To re-link or not to re-link for VMS upgrade (Post contains short and long v ---------------------------------------------------------------------- Date: 3 May 2007 12:01:49 -0700 From: Verne Subject: ANN: Alpha firmware update CD v7.3 released Message-ID: <1178218909.407590.130970@l77g2000hsb.googlegroups.com> We have the Firmware Distribution Service (for Alpha) on our support contract and yesterday we received the Alpha Systems Firmware Update v7.3. FYI, the entire current firmware collection is available at http://ftp.digital.com/pub/Digital/Alpha/firmware There you can also find an ISO image if you want to make your own bootable firmware update CD. Verne ------------------------------ Date: 3 May 2007 15:04:23 -0700 From: insomnee Subject: Any futher news on DecCampus equivalent for Itanium VMS in the UK? Message-ID: <1178229863.558733.225720@y5g2000hsa.googlegroups.com> Any futher news on DecCampus equivalent for Itanium VMS in the UK? ------------------------------ Date: 3 May 2007 16:52:26 -0700 From: davidc@montagar.com Subject: Re: Any futher news on DecCampus equivalent for Itanium VMS in the UK? Message-ID: <1178236346.122884.263040@l77g2000hsb.googlegroups.com> On May 3, 5:04 pm, insomnee wrote: > Any futher news on DecCampus equivalent for Itanium VMS in the UK? Any reason the OpenVMS Educational License Program (which is world- wide, provides free licenses, and includes Itanium licenses for faculty, admins, and even students) wouldn't work for you? http://www.openvmsedu.com ------------------------------ Date: 3 May 2007 13:00:21 -0700 From: "hellpenny@gmail.com" Subject: Re: DCL Magic? Message-ID: <1178222421.272208.258140@n76g2000hsh.googlegroups.com> On Apr 24, 4:43 pm, Johann 'Myrkraverk' Oskarsson wrote: > Bob Koehler writes: > > In article <00A66925.7CDCD...@SendSpamHere.ORG>, VAXman- > > @SendSpamHere.ORG writes: > > >> What happens when a program called SYS$SETDDIR() and then drops out to > >>DCLwhere the default was set in the program? > > > The same as when they use "set default" instead of the .COM file. > > But then I never did like path names taking up 80% of my command > > line. > > zsh has some string manipulation/substitutionmagicthat allows one to > truncate a string like path names. That way, you can display only the > last n characters of the path name in your terminal title, or prompt. > I've forgotten the actual syntax, but can probably dig it out from my > .zshrc. > > Johann Johann, Here is the solution I came up with ... visit this download page and follow instructions to download and check out _my_ "SD" offering. http://hellpenny.googlepages.com/holdmisc Once installed, you can do what you're attempting with: $ @SDSYM +T=(S) +N +V ! intial LOGIN setup $ sd; ! turn off prompt string tailor/ ! & turn on +T top display in your LOGIN.COM. This will attempt to display your default in your XTERM (or xterm-compatible) banner. If that doesn't work, use +T=(S:0) to use line 25 of a VT with status line capability (VT320 or greater??). If that doesn't work, you can use +T=(S:1) which will punt and write a reverse video line on the top line of your screen and define the rest as a scrolling region. [ Note: the first option S (no option) works great with PuTTy or xterm/rxvt while S:0 (25th line status) works great on later model VT's. I don't really use S:1 but it will mostly do the job if the other options don't cut it. The line shows up in reverse video and is formatted similar to: NODE::DEVICE:[DIRECTORY] (VOLUME) If all those options don't appeal, use $ sd, to enable tailoring of your prompt string to match the default. With tailoring the prompt, you are limited to 32 chars so there are many options for 'shortening' it and still keeping it meaningful. For instance, I use: $ sd, +n="VAX" - ! Cut the string "VAX" from nodename and display +v=(F:1,L:2) ! Display First char + Last 2 chars of vol label D01::U63:[MOOREJ] > SH DEF USER63:[MOOREJ] D01::U63:[MOOREJ] > WRITE SYS$OUTPUT F$TRNLNM("SYS$NODE") DVAX01 You can turn off node and volume display with $ SD, -n -v if you just want the dir - [MOOREJ] > If the prompt string gets too long as you descend dirs, "default" abbreviation action is taken to preserve the most relevant info possible. D01::U63:[MOOREJ] > SD....EXAM* USER63:[MOOREJ.VDBM04.EXAMPLES] D01::U63:[*J.VDBM04.EXAMPLES] > Is an example or the abbreviation. There are way too many options to explain briefly so Use SD? to see basic help SD?? for intermediate and SD??? to see the whole library of functions. Another problem mentioned in the thread ... when you run another program which changes the default, what happens? My solution to this (assuming you know in advance what programs you expect default changes to come from...) you can do this: For instance I use SWING (from decus lib) or DOWN on occassion to navigate complicated dir trees and everytime I would change default using these programs my prompt would be incorrect. I added a way to *hook* these command symbols thru my SD routine. It's simple to use .. I specify it as an option to the TOP display flag +|-T Example - $ sh sym SWING SW*ING == "@PUB:[UTIL.cswing]cswing" $ sh sym DOWN DOWN == "@PUB:[UTIL.down]down" $ sh sym CLR CLR == "$cluster$common:clrscrn.exe" $ SD; +T=(S,C:SW*ING;DOWN;CLR) ! Make these commands 'known' to SD $ sh sym SWING SW*ING == "@PUB:[UTIL.com]sd &SWING" $ sh sym SD$COMMAND_SWING SD$COMMAND_SWING == "@PUB:[UTIL.cswing]cswing" So now when one of these commands gets typed, it actually calls SD with the 'hidden' & function (which invokes the original command SD $COMMAND_symbol) and then updates the prompt (or top line display if you were using that - hence the 'CLEAR SCREEN' - CLR command would 'redraw' the top bar if the TOP bar/scrolling region was being used). This is not a perfect solution (one which could "sense" when the prompt string needed to be corrected/updated) but it's as close as I could come and does an adequate job. Try it! You'll like it. Regards, -John Moore Sr. Unix Admin / WorkflowOne (Aging vax-hacker...) ------------------------------ Date: 3 May 2007 15:02:21 -0700 From: insomnee Subject: DecNet Plus installation on VMS 8.3 Itanium Message-ID: <1178229741.582055.283380@e65g2000hsc.googlegroups.com> So I was wanting to get away from using DecNet full stop. I've got brand new rx7640s and I didnt want Decnet. As a protocol, it's dead - surely! Anyway my developers don't want to update any code using copy/rcp or copy/ftp and want me to configure it. This went fine on a standalone rx4640, took me a while to work out what i'd done wrong but got there in the end. On the rx7640 cluster, it appears to get so far on boot up and then stop. It complains that is cannot mount the quorum disk, DGA1, though the system says it forms a cluster, drops the connection, reconnects, continues the boot and then stops here. Can't get out of it, message repeats on a 5 minute loop. I've had to restore my standalone backup to get back to a stable point and not sure where to go from here. Any Ideas? %NET-I-LOADED, executive image NET$SESSION_CONTROL.EXE loaded %NET$STARTUP-I-STARTPROCESS, starting process DNS %NET-I-LOADED, executive image SYS$NAME_SERVICES.EXE loaded %RUN-S-PROC_ID, identification of created process is 20200419 %%%%%%%%%%% OPCOM 3-MAY-2007 21:03:23.96 %%%%%%%%%%% Message from user SYSTEM on EMPD03 %CSP-W-QDNOTMNT, Please mount the quorum disk ------------------------------ Date: Thu, 3 May 2007 17:27:41 -0500 (CDT) From: sms@antinode.org (Steven M. Schweda) Subject: Re: DecNet Plus installation on VMS 8.3 Itanium Message-ID: <07050317274133_202002DA@antinode.org> From: insomnee > So I was wanting to get away from using DecNet full stop. I've got > brand new rx7640s and I didnt want Decnet. As a protocol, it's dead - > surely! It's dead if you kill it. It's alive if you use it, and it is useful in many ways. > Anyway my developers don't want to update any code using copy/rcp or > copy/ftp and want me to configure it. This went fine on a standalone > rx4640, took me a while to work out what i'd done wrong but got there > in the end. COPY /FTP and COPY /RCP have nothing to do with DECnet. > On the rx7640 cluster, it appears to get so far on boot up and then > stop. It complains that is cannot mount the quorum disk, DGA1, though > the system says it forms a cluster, drops the connection, reconnects, > continues the boot and then stops here. Can't get out of it, message > repeats on a 5 minute loop. I've had to restore my standalone backup > to get back to a stable point and not sure where to go from here. Any > Ideas? > > %NET-I-LOADED, executive image NET$SESSION_CONTROL.EXE loaded > %NET$STARTUP-I-STARTPROCESS, starting process DNS > %NET-I-LOADED, executive image SYS$NAME_SERVICES.EXE loaded > %RUN-S-PROC_ID, identification of created process is 20200419 > %%%%%%%%%%% OPCOM 3-MAY-2007 21:03:23.96 %%%%%%%%%%% > Message from user SYSTEM on EMPD03 > %CSP-W-QDNOTMNT, Please mount the quorum disk At first glance, this appears to be as closely related to DECnet as COPY /FTP and COPY /RCP are. Have you tried mounting the quorum disk? I'd suggest worrying less about DECnet and more about the VMScluster configuration. ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: 3 May 2007 12:33:58 -0700 From: Joshua Lehrer Subject: Re: detect unaligned data access Message-ID: <1178220838.906313.203350@e65g2000hsc.googlegroups.com> On May 3, 1:46 pm, "Guy Peleg" wrote: > "Joshua Lehrer" wrote in message > > news:1178208892.462057.299410@n76g2000hsh.googlegroups.com... > > >I would like to detect unaligned data access in my application on VMS > > and perhaps count them, or log it to a file. > > > I know that one can do this in the VMS debugger with "set break/ > > unaligned" so there must be a way to programmatically detect unaligned > > data access. > > No need to reinvent the wheel... > > Use the FLT extension in SDA > > SDA> FLT LOAD > SDA> FLT START TRACE > I'm sorry I wasn't more clear. What I am looking for is a way, from C (C++, Fortran, whatever...), for my process to detect when it performans an alignment fault / unaligned data access. Due to other constraints beyond the scope of this post, I can not use an external tool to monitor the running process, the given process should be able to detect it itself. I don't want to run it in the debugger and 'set break/unaligned' and I don't want the number after-the-fact from auditing like tools. Let's just say I wanted this given program, every time it runs, to count it's own total alignment faults and write out the final number to disk. I'm pretty sure this can be done - when you run in the debugger you can do it. I will check out the system services reference manual for "$START_ALIGN_FAULT_REPORT" and other related functions. Thanks, joshua ------------------------------ Date: Fri, 04 May 2007 09:45:40 +1000 From: Jim Duff Subject: Re: detect unaligned data access Message-ID: <463a7425@dnews.tpgi.com.au> Joshua Lehrer wrote: > On May 3, 1:46 pm, "Guy Peleg" > wrote: >> "Joshua Lehrer" wrote in message >> >> news:1178208892.462057.299410@n76g2000hsh.googlegroups.com... >> >>> I would like to detect unaligned data access in my application on VMS >>> and perhaps count them, or log it to a file. >>> I know that one can do this in the VMS debugger with "set break/ >>> unaligned" so there must be a way to programmatically detect unaligned >>> data access. >> No need to reinvent the wheel... >> >> Use the FLT extension in SDA >> >> SDA> FLT LOAD >> SDA> FLT START TRACE >> > > I'm sorry I wasn't more clear. What I am looking for is a way, from C > (C++, Fortran, whatever...), for my process to detect when it > performans an alignment fault / unaligned data access. Due to other > constraints beyond the scope of this post, I can not use an external > tool to monitor the running process, the given process should be able > to detect it itself. I don't want to run it in the debugger and 'set > break/unaligned' and I don't want the number after-the-fact from > auditing like tools. Let's just say I wanted this given program, > every time it runs, to count it's own total alignment faults and write > out the final number to disk. > > I'm pretty sure this can be done - when you run in the debugger you > can do it. > > I will check out the system services reference manual for > "$START_ALIGN_FAULT_REPORT" and other related functions. > > Thanks, joshua > Example code here: http://www.eight-cubed.com/examples/framework.php?file=sys_align_faults.c Regards, Jim. -- www.eight-cubed.com ------------------------------ Date: Thu, 03 May 2007 23:52:56 -0400 From: JF Mezei Subject: Re: Favorite movies of the VMS crowd? Message-ID: Since this is straying from VMS marketing movies.... (not in any order) Many james bond movies. Indiana Jones movies Peacemaker Airplane !!! Hunt for Red October 2001 and 2010 (Interesting snapshot of world prior to demise of Pan Am, Soviet Union, AT&T but also clairvoyant in predicting americans and russians working on the same project becauyse they needed each other. Oceans' 11 (The recent remake) Bourne Identity/Supremacy Aliens The Shining And some silly movies like Rat Race. And french movies such as La Totale (original of True Lies), Les Voyageurs, etc. ------------------------------ Date: 3 May 2007 21:10:55 -0700 From: davidc@montagar.com Subject: Re: Favorite movies of the VMS crowd? Message-ID: <1178251855.084892.6820@q75g2000hsh.googlegroups.com> On May 3, 10:49 am, Rob Brown wrote: > What about Dark Star? I'm sorry, but you are false data. In the beginning, there was nothing... except me. I first saw that movie when I was in college. The opening scene where the starship, complete with dull roar engines (in space...) just instantanously STOPPED. There was thing slow growing laughter in the audience as we all realized that the conservation of momentem had been horribly violated. It was a wonderful thing. ------------------------------ Date: Thu, 03 May 2007 21:17:31 GMT From: Tad Winters Subject: Re: Free used AlphaServers Message-ID: Tad Winters wrote in news:Xns991F99E273D53staffordnospamwinter@130.81.64.196: > David J Dachtera wrote in > news:46315C41.A4128C6@spam.comcast.net: > >> Tad Winters wrote: >>> >>> Please email me if you'd like a free, used AlphaServer _and_ you are >>> able to pick it up in Orange, California. The AlphaServers were >>> working fine, but the disk drives had to be removed and destroyed in >>> accordance with corporate policy. These are pedestal systems and at >>> lease a couple of them have video adapters so they will run DEC >>> Windows. >> >> A few more specifics, please? >> >> Model? >> >> CPU speed? >> >> RAM size? >> >> SCSI / FC HBAs present in each? >> >> NICs present in each? >> >> Any CD, DVD or tape drives present in each? >> > > AlphaServer 1200 5/400 and AlphaServer 1000 4/266 (or pretty close) > about 512 MB each > 3 channel (or possibly 1 channel) RAID controller > DE500 in at least 1, with the other possibly a DE435. > CD ROM drive in each. > They tape drive may have been removed. > Keep in mind that I'm not in the same location as these systems, else > I'd haul them home myself and arrange to put these in people's hands. > If I can't get someone to pick them up, they'll just be sent to a > local recycler. > (I previously worked for the company that has these systems, but I > don't now work for them and I am not compensated for these efforts. > I'm doing this merely to enlarge the hobbyist community. If I'm > successful in this effort, I'm more likely to have other offices let > me find a taker for their systems.) > > Tad > These were all picked up by one person. I'm sure he's a happy hobbyist... or maybe he'll make a little money on eBay. :-) Tad ------------------------------ Date: 3 May 2007 15:06:34 -0700 From: insomnee Subject: Java problem Message-ID: <1178229994.247910.287720@n76g2000hsh.googlegroups.com> We are migrating our Oracle 7 / forms 45 VMS 7.2-1 application to Oracle 10g / forms 10g on OpenVMS 8.3 on Itanium. A user invokes a Java job from an Oracle 10g form in order to submit operating system commands and bacth jobs to a VMS batch queue. The java job then submits a DCL JOB on VMS under the ORACLE10G user, this then submit the required job to the correct batch queue as the real user (the user logged on to the form). The batch queue number is written to a file which is later picked up by the form to show to the user and the user continues. In the background when the user has submitted the request from the form the DCL job invokes a cobol transactions generating an Oracle report / database update. This java job is key to our application in calling our various transactions that exist on the VMS platform. However when this job is invoked the execution hangs after three or four invocations from the same user session.When the batch queue in which the job runs is closed an unlimited number of jobs can be queued successfully,the problem is non existent, but we can't run our transactions until the queue is then started. The java job is using runtime.exec, it appear that it hangs during the process.wait() although the job I is waiting for successfully runs on the batch queue. We have eliminated ORACLE FORMS as the problem as the java code can also be made to hang when submitted directly from the database. (although less frequently). The problem occurs only when it is a job submitted to a queue via epsQueueJob. Any ideas? ------------------------------ Date: Thu, 03 May 2007 21:34:06 GMT From: Tad Winters Subject: Re: List of VMS command file .COM domain names (was: Free email account VMSUSER Message-ID: briggs@encompasserve.org wrote in news:GTbGU+7rpWOJ@eisner.encompasserve.org: >> Note. As stated in RFC 2181 >> >> " >> 11. Name syntax >> >> Occasionally it is assumed that the Domain Name System serves only >> the purpose of mapping Internet host names to data, and mapping >> Internet addresses to host names. This is not correct, the DNS is >> a general (if somewhat limited) hierarchical database, and can >> store almost any kind of data, for almost any purpose. >> >> " >> >> Underscores are perfectly valid for use in records other than domain >> name records and are heavily used in for instance SRV records. >> >> But since we were talking about domain names rather than SRV or other >> types of records my statement stands that the dollar and underscore >> characters are not valid for that purpose. > > RFC 2181 does not distinguish between what you call "domain names" > and what you call "SRV records". In the language of RFC 2181, > "domain name" covers both the names of "A" records and the names > of "SRV" records. And it also covers the values in "NS" records, > "MX" records, "SOA" records and so on. > > If you want to argue about the standards for "Internet Host Names" > then we can argue about that instead. But then neither STD13 nor > RFC 2181 are authoritative documents with respect to that question. > I've noted for years that 2181 is still sitting at a "proposed standard" status. With that status, it seems to me that domain names containing underscores or dollar signs could safely be denied as acceptable. ------------------------------ Date: Fri, 4 May 2007 01:34:05 +0000 (UTC) From: david20@alpha1.mdx.ac.uk Subject: Re: List of VMS command file .COM domain names (was: Free email account VMSUSER Message-ID: In article , Tad Winters writes: >briggs@encompasserve.org wrote in >news:GTbGU+7rpWOJ@eisner.encompasserve.org: > >>> Note. As stated in RFC 2181 >>> >>> " >>> 11. Name syntax >>> >>> Occasionally it is assumed that the Domain Name System serves only >>> the purpose of mapping Internet host names to data, and mapping >>> Internet addresses to host names. This is not correct, the DNS is >>> a general (if somewhat limited) hierarchical database, and can >>> store almost any kind of data, for almost any purpose. >>> >>> " >>> >>> Underscores are perfectly valid for use in records other than domain >>> name records and are heavily used in for instance SRV records. >>> >>> But since we were talking about domain names rather than SRV or other >>> types of records my statement stands that the dollar and underscore >>> characters are not valid for that purpose. >> >> RFC 2181 does not distinguish between what you call "domain names" >> and what you call "SRV records". In the language of RFC 2181, >> "domain name" covers both the names of "A" records and the names >> of "SRV" records. And it also covers the values in "NS" records, >> "MX" records, "SOA" records and so on. >> Actually in RFC 2181 it talks about DNS labels not domain names. >> If you want to argue about the standards for "Internet Host Names" >> then we can argue about that instead. But then neither STD13 nor >> RFC 2181 are authoritative documents with respect to that question. >> > >I've noted for years that 2181 is still sitting at a "proposed standard" >status. With that status, it seems to me that domain names containing >underscores or dollar signs could safely be denied as acceptable. Even if RFC 2181 was to be upgraded to a standard I'd be very wary about using dollars and other non-standard characters in domain names for quite sometime. As RFC 2181 says " Note however, that the various applications that make use of DNS data can have restrictions imposed on what particular values are acceptable in their environment. " Underscores probably wouldn't cause too much of a problem since many/most applications would have been written to cope with them since they were fairly widely used in the past. However RFCs for other internet applications and the implementations of the applications would need to be updated to cope with the expanded domainname character set. As RFC 3696 says " 2. Restrictions on domain (DNS) names The authoritative definitions of the format and syntax of domain names appear in RFCs 1035 [RFC1035], 1123 [RFC1123], and 2181 [RFC2181]. Any characters, or combination of bits (as octets), are permitted in DNS names. However, there is a preferred form that is required by most applications. This preferred form has been the only one permitted in the names of top-level domains, or TLDs. In general, it is also the only form permitted in most second-level names registered in TLDs, although some names that are normally not seen by users obey other rules. It derives from the original ARPANET rules for the naming of hosts (i.e., the "hostname" rule) and is perhaps better described as the "LDH rule", after the characters that it permits. The LDH rule, as updated, provides that the labels (words or strings separated by periods) that make up a domain name must consist of only the ASCII [ASCII] alphabetic and numeric characters, plus the hyphen. No other symbols or punctuation characters are permitted, nor is blank space. If the hyphen is used, it is not permitted to appear at either the beginning or end of a label. There is an additional rule that essentially requires that top-level domain names not be all- numeric. When it is necessary to express labels with non-character octets, or to embed periods within labels, there is a mechanism for keying them in that utilizes an escape sequence. RFC 1035 [RFC1035] should be consulted if that mechanism is needed (most common applications, including email and the Web, will generally not permit those escaped strings). A special encoding is now available for non-ASCII characters, see the brief discussion in Section 5. " David Webb Security team leader CCSS Middlesex University ------------------------------ Date: Thu, 3 May 2007 12:58:51 -0400 From: "William Webb" Subject: Re: List of VMS command file .COM domain names (was: Free email account VMSUSER. Message-ID: <8660a3a10705030958g2666d88dt59f18da1958470b@mail.gmail.com> On 3 May 2007 11:55:45 -0500, briggs@encompasserve.org wrote: > In article , david20@alpha2.mdx.ac.uk writes: > > In article , briggs@encompasserve.org writes: > >>In article , david20@alpha2.mdx.ac.uk writes: > >>> In article <133idqs3vrncda2@corp.supernews.com>, "C.W.Holeman II" writes: > >>>>Doc wrote: > >>>>> "David Turner, Island Computers" wrote in > >>>>> news:133h7qcepn0ar65@news.supernews.com: > >>>>> > >>>>>> We own VMSUSER.COM > >>>> > >>>>Does anyone have a list VMS command file .COM domain names like: > >>>> > >>>> SYS$STARTUP.COM > >>>> TCPIP$PWIP_DRIVER_STARTUP.COM > >>>> > >>> > >>> The $ is not a valid character in an internet domain name. > >>> (Use of underscore is also not strictly allowed - though older DNS servers > >>> supported it.) > >> > >>It's the other way around. > >> > >>RFC 1034: Various characters are illegal in a domain name. > >> > >>RFC 2181: There are no illegal characters in a domain name. Dollar sign > >>and underscore are perfectly fine. > >> > >>BIND 4 didn't much like underscores as I recall. > >>BIND 9 is perfectly happy with them and cleaves to the "no characters > >>are illegal" interpretation. > > >>RFC 2821 is still picky about the valid character set for domain names > >>that are used to refer to SMTP servers, but that doesn't make the > >>domain names illegal per se. > > > > Sorry as far as I am aware the standard is still STD13 which refers to > > RFC 1034 and RFC 1035 but does not refer to RFC 2181. > > > > See > > > > http://www.faqs.org/rfcs/std/std13.html > > 1987. How quaint. > > > My recollection is different from yours. BIND 4 was perfectly happy with > > underscores it was the move to BIND 8 which tightened things up > > see for instance > > You're right about this, of course. I've been using BIND 9 for so > long that I'd completely forgotton that it was the BIND 8 migration > where many of us first stumbled over underscores. > > > Note. As stated in RFC 2181 > > > > " > > 11. Name syntax > > > > Occasionally it is assumed that the Domain Name System serves only > > the purpose of mapping Internet host names to data, and mapping > > Internet addresses to host names. This is not correct, the DNS is a > > general (if somewhat limited) hierarchical database, and can store > > almost any kind of data, for almost any purpose. > > > > " > > > > Underscores are perfectly valid for use in records other than domain name > > records and are heavily used in for instance SRV records. > > > > But since we were talking about domain names rather than SRV or other types of > > records my statement stands that the dollar and underscore characters are not > > valid for that purpose. > > RFC 2181 does not distinguish between what you call "domain names" > and what you call "SRV records". In the language of RFC 2181, > "domain name" covers both the names of "A" records and the names > of "SRV" records. And it also covers the values in "NS" records, > "MX" records, "SOA" records and so on. > > If you want to argue about the standards for "Internet Host Names" > then we can argue about that instead. But then neither STD13 nor > RFC 2181 are authoritative documents with respect to that question. > I once asked a colleague about whether a particular configuration complied with the corporate standard in existence at the time. He gave me a funny little smile and replied: "We have many standards". WWWebb ------------------------------ Date: Fri, 4 May 2007 02:54:52 +0000 (UTC) From: moroney@world.std.spaamtrap.com (Michael Moroney) Subject: Re: List of VMS command file .COM domain names (was: Free email account VMSUSER. Message-ID: david20@alpha2.mdx.ac.uk writes: >>To see what's left try >> >>dir sys$startup:*.com/excl=(*$*.com,*_*.com) >> >>dir sys$update:*.com/excl=(*$*.com,*_*.com) >> >>etc >> >>or >> >>dir sys$sysdevice:[000000...]*.com/excl=(*$*.com,*_*.com) >> >I just tried that last command on a VMS 8.3 Alpha system with ODS-5 and >hard-links enabled on the system disk and with GNV installed >got as far as >Directory >SYS$SYSDEVICE:[SYS0.SYSCOMMON.GNV.SYS0.SYSCOMMON.GNV.SYS0.SYSCOMMON.GN >V.SYS0.SYSCOMMON.GNV.SYS0.SYSCOMMON.GNV.mnt.DATA1.HTTP_SERVER.ANALOG.ANALOG-5_0. >SRC.LIBPNG] >MAKE.COM;1 >before I gave up. >Looks like it would have looped around forever. Is there a patch for this or >a new qualifier so that directory listings don't follow links forever ? I recently installed GNV on an Itanic and got pissed that it did that. I traced it down to the fact that the [SYS0.SYSCOMMON]GNV.DIR is essentially a clone of the top level [000000]000000.DIR on the system disk. That is, everything in [000000] was apparently $ SET FILE /ENTERed into [SYS0.SYSCOMMON.GNV]. $ DIRECTORY just saw a [.SYS0] in that directory and followed it, and it found the GNV directory again, which found ... $ DIRECTORY didn't know it was chasing itself in circles. I just got rid of that stupid directory, and that took care of the infinite directory loop. (since all I wanted was sed, I didn't really care if it broke something else. That something else, if it exists, deserves to be broken) The stupid GNV kit also created a [mnt] directory, which is pointers to all your _other_ disks. So a directory of SYS$SYSDEVICE:[*...] will list all your files on all your disks, not just SYS$SYSDEVICE! Guess what a $ PURGE SYS$SYSDEVICE:[*...] will do! That GNV kit must have been created by a Unix weenie who didn't understand the VMS file system, and knew enough VMS to be dangerous ($ SET FILE/ENTER) ------------------------------ Date: Thu, 03 May 2007 20:57:46 +0200 From: Michiel Erens Subject: Re: Medical software vendor says it won't support OS on Itanium-based servers se Message-ID: <463a30a8$0$284$dbd41001@news.wanadoo.nl> Andrew wrote: > On 27 Apr, 14:57, b...@cs.uofs.edu (Bill Gunshannon) wrote: >> In article <1177666922.680728.246...@r3g2000prh.googlegroups.com>, >> Andrew writes: >> >> Knowing the complexity of Cerner's product and taking into consideration >> the dubious future of Itanium (and with it, VMS) can you blame them? >> This is one of those things that HP management should have anticipated >> and either didn't, or just chose to ignore. > > To be fair to HP Cerner were pretty complimentary about Itanium when > it came out talking about how it would offer new capabilities to > OpenVMS customers. It is still on their website : http://www.cerner.com/public/Cerner_2.asp?id=27196 -- ME ------------------------------ Date: 03 May 2007 18:14:18 GMT From: "Dan Drake" Subject: Re: Noahs ark found! Message-ID: [Did my newsreader manage to post a useless thing quoting most of that message? If so, sorry.] On Wed, 2 May 2007 12:24:16 UTC, bill@cs.uofs.edu (Bill Gunshannon) wrote: > In article , > David Mathog writes: > >... > > "It was a conflict that ought never to have occurred, because > > faith and science, properly understood, can never be at odds." > > > > Quoted from here: > > > > http://www.catholic.net/rcc/Periodicals/Issues/GalileoAffair.html > > Which is true, because Galileo's crime was not what he taught, but > purely a crime of disobedience to lawful authority. > >... "Purely" is a dangerous, or anyway dubious, word to use with the history of the real world, which is always more complicated than our pet theries. Consider, for instance, ==== 1. The proposition that the sun is in the center of the world and immovable from its place is absurd, philosophically false, and formally heretical; because it is expressly contrary to Holy Scriptures. 2. The proposition that the earth is not the center of the world, nor immovable, but that it moves, and also with a diurnal action, is also absurd, philosophically false, and, theologically considered, at least erroneous in faith. ... ...you, the said Galileo . . . have rendered yourself vehemently suspected by this Holy Office of heresy, that is, of having believed and held the doctrine (which is false and contrary to the Holy and Divine Scriptures) that the sun is the center of the world, and that it does not move from east to west, and that the earth does move, and is not the center of the world; also, that an opinion can be held and supported as probable, after it has been declared and finally decreed contrary to the Holy Scripture, and, consequently, that you have incurred all the censures and penalties enjoined and promulgated in the sacred canons and other general and particular constituents against delinquents of this description. ==== That's from the official condemnation, to be found at http://www.fordham.edu/halsall/mod/1630galileo.html There's also stuff about disobedience there, but the Inquisiton's conclusion is that second paragraph, and a demand that he give up all his error and heresies. Purely a matter of disobedience, it is not. Except disobedience to the requirement of believing what the Church told people to believe; but I think that argument would be disingenuous. Speaking of Jesuits -- and note the site I cited -- they have been perfectly content for years, and long before anyone heard of Cardinal Wojtyla, to concede that the Church messed up badly, and to move on from there. Regrettably, even after Pope John Paul II tried to clear matters up, there are others less willing to face the unsavory side of their church's part in the Galileo case (while in fact some Jesuits argued that the Pope didn't go far enough!) and still inspiring some quite unnecessary hostility to Rome by acting as apologists for the Inquisition. -- Dan Drake dd@dandrake.com http://www.dandrake.com/ porlockjr.blogspot.com ------------------------------ Date: 3 May 2007 18:15:36 -0700 From: Sue Subject: Re: OT: Favorite movies of the VMS crowd? Message-ID: <1178241336.771007.277670@n59g2000hsh.googlegroups.com> On May 1, 1:42 pm, "Doug Kimball" wrote: > So, what are your top three favorite movies, VMSophiles? We need a promo for > the upcoming HP Partners Roundhouse in Nashua, and we would love your input. > Thanks! > > Doug > > -- > Doug Kimball > Manager, Sales and Support > Software Partners, Inc.www.softwarepartners.com Enemy of the State Space Cowboys Sneakers Sue ------------------------------ Date: 3 May 2007 18:25:00 -0700 From: Sue Subject: Re: OT: Favorite movies of the VMS crowd? Message-ID: <1178241900.132609.300860@n59g2000hsh.googlegroups.com> On May 3, 9:15 pm, Sue wrote: > On May 1, 1:42 pm, "Doug Kimball" > wrote: > > > So, what are your top three favorite movies, VMSophiles? We need a promo for > > the upcoming HP Partners Roundhouse in Nashua, and we would love your input. > > Thanks! > > > Doug > > > -- > > Doug Kimball > > Manager, Sales and Support > > Software Partners, Inc.www.softwarepartners.com > > Enemy of the State > Space Cowboys > Sneakers > > Sue Tron Daryl Apollo 13 Lord of the Rings Hunt for Red October Independence Day sue ------------------------------ Date: 3 May 2007 18:37:09 -0700 From: Sue Subject: Re: OT: Favorite movies of the VMS crowd? Message-ID: <1178242629.420235.122480@c35g2000hsg.googlegroups.com> On May 3, 9:25 pm, Sue wrote: > On May 3, 9:15 pm, Sue wrote: > > > > > > > On May 1, 1:42 pm, "Doug Kimball" > > wrote: > > > > So, what are your top three favorite movies, VMSophiles? We need a promo for > > > the upcoming HP Partners Roundhouse in Nashua, and we would love your input. > > > Thanks! > > > > Doug > > > > -- > > > Doug Kimball > > > Manager, Sales and Support > > > Software Partners, Inc.www.softwarepartners.com > > > Enemy of the State > > Space Cowboys > > Sneakers > > > Sue > > Tron > Daryl > Apollo 13 > Lord of the Rings > Hunt for Red October > Independence Day > > sue- Hide quoted text - > > - Show quoted text - ok so I am putting my favorites Sahara National Treasure Second Hand Lions (this is really good if you have not seen it) Operation Petticoat (not a chick flick but its old) Chick Flicks African Queen Lady Hawk Pride and Prejudice Father Goose TV shows that I think are worth it The Carol Burnett Show the Red Green Show (I have 8 DVD's of Red Green) and I might have a couple CSI. come to think of it I might not have anything that is still running on TV in the US ;') More than you wanted to know Sue ------------------------------ Date: Thu, 03 May 2007 23:24:32 -0400 From: JF Mezei Subject: Re: OT: Favorite movies of the VMS crowd? Message-ID: Sue wrote: > Tron > Daryl > Apollo 13 > Lord of the Rings > Hunt for Red October > Independence Day Shouldn't Sue have at lest one star trek movie in there ????? Now, if Digital had had marketing savvy, they would have enslaved Majel Barret, and made her voice the official voice for DecTalk. Billions of trek fans would have instantly become VAX/VMS fans. ------------------------------ Date: 03 May 2007 14:02:18 -0400 From: "Richard E. Silverman" Subject: Re: SSH Keys for a heterogeneous Environment Message-ID: Use the -e option: -e file Edit the comment/passphrase of the key. > > Richard E. Silverman schrieb: > >>>>>> "MK" == Martin Krischik writes: > > > MK> 3) And now the ultimate question: Is there a tool which can > > MK> convert keys? Thanks to c.o.vms I now know that ssh-keygen is > > MK> supposed to be able to do this. But hey, look at this: > > MK> ------------------------------------------- > > >> ssh-keygen -i -f vms_dsa > > MK> unsupported cipher 3des-cbc decode blob failed. > > MK> ------------------------------------------- > > MK> It seems that ssh-keygen is a bit limited in it's abilities. > > The input key must be unencrypted; this is mentioned in the man page. > > Easier said then done. The shh-keygen used to generate the key does not > have a decrypt option: > > ------------------------------------------- > ssh_keygen -h > Usage: ssh_keygen [options] [key1 key2 ...] > > Where `options' are: > -b nnn Specify key strength in bits (e.g. 1024) > -t dsa | rsa Choose the key type. > -c comment Provide the comment. > -e file Edit the comment/passphrase of the key. > -p passphrase Provide passphrase. > -P Assume empty passphrase. > -? > -h Print this help text. > -q Suppress the progress indicator. > -1 Convert a SSH 1.x key. > -i file Load and display information on `file'. > -D file Derive the public key from the private key 'file'. > -B number The number base for displaying key information (default > 10). > -V Print version number of tcpip$ssh_ssh-keygen2.exe image. > -r file Stir data from file to random pool. > -x file Convert private key from X.509 format to SSH2 format. > -k file Convert a PKCS 12 file to an SSH2 format certificate > and private key. > -7 file Extract certificates from a PKCS 7 file. > -F file Dump fingerprint of file. ------------------------------------------- Martin -- Richard Silverman res@qoxp.net ------------------------------ Date: 3 May 2007 11:30:54 -0700 From: AEF Subject: To re-link or not to re-link for VMS upgrade (Post contains short and long versi Message-ID: <1178217053.573128.239990@o5g2000hsb.googlegroups.com> Hello VMSers, Seeking advice on whether or not to re-link my app's executables for VMS upgrade from VAX/VMS V6.1 to V6.2: Short version: I have an application written in DEC Pascal V5.1-22 (except for the "message manager" which is written in VAX macro). It has approx. 160 executables that run on a VMS V6.1 system (the market system) which were built on another VMS V6.1 system (the development system). There are also non-clustered "update nodes" running either VMS V6.1 or V6.2. These update systems have copies of five of the executables which establish DECnet links among themselves and to the app running on the market system. Users connect to the update nodes from a VB app running on a Windows PC using TCP. All this currently works fine in production and I have the app set up on a test system running VMS V6.2 and it seems to run fine. I'd like to upgrade the market system to VMS V6.2. To re-link or not to re-link (on a VMS V6.2 system)? That is the question and why. What are the risks involved for each alternative? (I'm already in the process of upgrading the V6.1 update nodes to V6.2 because I know for certain that the five app executables run fine on VMS V6.2.) Thanks! Long version: My app is a trading system that runs on multiple VAX systems. The trading system consists of one "market node" and several "update nodes". The market node is the smarts of the system. This is where all the trading occurs. The market node and the update nodes communicate with each other via DECnet logical links. The traders run a VB front- end on Windows workstations and connect to "trader processes" on the update nodes via TCP/IP links. The update nodes can also run "SCM processes" which broadcast prices to special customer terminals. There are two desks (call them A and B). Each desk runs from its own market node and update nodes. The two sets of VAX systems are totally independent of each other. None are VMSclustered. The software runs basically the same on both, but with different trading rules and different formats for some things. I guess you could say they share the same architecture. Both market nodes are running VMS V6.1. The update nodes for either desk are a mix of VMS V6.1 and VMS V6.2 systems. The V6.1 systems have only a DECnet ECO kit applied (VAXNETA02_061 because one day we suddenly started getting "Phase IV node exists in the absence of bilingual router" messages) and some of the V6.2 systems have all relevant ECO kits applied. The market nodes are MicroVAX 3100 Model 95's and the update nodes are all MicroVAX 3100 Model 80's. Market node A has 151 executables plus three "shared memory" .EXE files. Five of these executables are also on the update nodes. Market node B has 171 executables plus three "shared memory" .EXE files. The same-named (corresponding) five executables as in case A are also on the update nodes. So at least 5 of the executables run fine on both VMS V6.1 and V6.2. The app is written in DEC Pascal V5.1-22 except for the "message manager" which is written in VAX macro. All were built on a VMS V6.1 system. (And there are a set of DCL command procedures, too, of course, that are used to run everything.) Now I'd like to upgrade all the V6.1 nodes to V6.2. What are the advantages of re-linking vs. not re-linking? I have the app running on a test system running VMS V6.2 (I can switch to either version on this box) and both versions (A and B) appear to run fine (not re-linked), but I didn't test every conceivable thing. Could something break from the upgrade if I don't re-link? Could something break if I do re-link? I have a backup system (for each desk) to quickly switch back to in case of trouble. All the executables have been built on a VMS V6.1 system. So when is it important to re-link? I set up test market systems (all 154 or 174 exe files) running on VMS V6.2 and everything comes up. I remember the original developer/consultant who supported and sold us this app once said if everything comes up, it's good! I performed some simple tests with the front-end and it appears to work okay. (I did not re-link anything.) It produces about two-dozen data files every day and I haven't finished checking them thoroughly yet but I think they're okay. So how do I determine if I really have to re-link everything on a V6.2 system? I could set up a demo system for the brokers to check out, and keep the backup market system on V6.1 which I can switch to in a few minutes if the V6.2 becomes a disaster. I've been running like this for a few years and have only recently started to add VMS V6.2 ECO kits to a few systems and upgraded two of the update nodes from V6.1 to V6.2. (I was originally told not to apply any ECO kits.) The app runs well on the VMS V6.2 test systems and I'd be somewhat amazed if anything would break with an upgrade to VMS V6.2. But I'm not 100% sure, so I'm asking here. All friendly advice is welcome. A big thanks in advance!!! AEF ------------------------------ Date: Thu, 03 May 2007 17:17:47 -0700 From: Ken Fairfield Subject: Re: To re-link or not to re-link for VMS upgrade (Post contains short and long v Message-ID: <59vc9hF2kp0n8U1@mid.individual.net> AEF wrote: > Hello VMSers, > > Seeking advice on whether or not to re-link my app's executables for > VMS upgrade from VAX/VMS V6.1 to V6.2: [...BIG snip...] Given that your application is written in Pascal, and you made no mention of upgrading Pascal, I can see no point in relinking at all! FWIW, your application (already) links to various shared images that implement the RTL and system stuff, and you will automagically access these on the V6.2 system without relinking. In the rare case that you have some code that links /SYSEXE, (whoops, sorry, that's Alpha...what is the equivalent on VAX?) you would need to relink _that_ code when you to a major upgrade, i.e., to V7.x, but not for V6.1 to V6.2... To repeat, I see no point whatsoever in relinking. :-) -Ken -- Ken & Ann Fairfield What: Ken dot And dot Ann Where: Gmail dot Com ------------------------------ End of INFO-VAX 2007.243 ************************