1 INFO-VAX	Thu, 03 Apr 2003	Volume 2003 : Issue 184       Contents:G Re: Any way to have a filename beginning with the ampersand character ?  Re: COV Sponsors Re: COV Sponsors Re: COV Sponsors Re: COV Sponsors Re: COV Sponsors Re: COV Sponsors! Re: DBI prebuilt modules for perl ! Re: DBI prebuilt modules for perl ! Re: DBI prebuilt modules for perl . Re: Do OpenVMS Alpha have a Year 2057 problem?. Re: Do OpenVMS Alpha have a Year 2057 problem?. Re: Do OpenVMS Alpha have a Year 2057 problem?. Re: Do OpenVMS Alpha have a Year 2057 problem?. Re: Do OpenVMS Alpha have a Year 2057 problem?. Re: Do OpenVMS Alpha have a Year 2057 problem?' Re: Earth to Andrew, do you get it now?  Re: Fortran Guru requested Re: Fortran Guru requested RE: Fortran Guru requested  Re: ftp_mirror and TCPIP servers! How does Shadow Copy REALLY work? 6 Re: How to connect Vaxstation 4000 VLC to LCD monitor? Re: How to get timing  Re: How to get timing 0 Re: Include an automated date within a printout.0 Re: Include an automated date within a printout.> Re: Inquirer: HP's Alpha RetainTrust programme a complete bust3 Re: latest Business Critical Systems Customer Times  Re: MS-Word from OpenVMS ? Re: MS-Word from OpenVMS ? Re: MS-Word from OpenVMS ?P Re: Mystery solved? (Re: [Q] Files not found during image backup, w/ and  w/o diP Re: Mystery solved? (Re: [Q] Files not found during image backup, w/ and w/o dir$ Re: Netscape gags on VMS FTP Servers$ Re: Netscape gags on VMS FTP Servers' Re: OpenVMS and Security - REALLY !!!!! ' Re: OpenVMS and Security - REALLY !!!!! ' Re: OpenVMS and Security - REALLY !!!!! ' Re: OpenVMS and Security - REALLY !!!!! ' Re: OpenVMS and Security - REALLY !!!!! ' Re: OpenVMS and Security - REALLY !!!!! G Re: OpenVMS v7.3-1 and Advanced Server v7.3A - Pathworks server crashes  Re: pathworks administer issue Re: pathworks administer issue Re: pathworks administer issue- Re: Samba on VMS Get_Hostbyname error message  SMTP authentification  Re: SMTP authentification  Re: SMTP authentification  Re: SMTP authentification  Re: SMTP authentification  Re: SMTP authentification  Re: SMTP authentification  Re: SMTP authentification  Re: SMTP authentification  Re: SMTP authentification  Re: SMTP authentification / Re: So much for Opteron 32bit compatibility ... / Re: So much for Opteron 32bit compatibility ... / Re: So much for Opteron 32bit compatibility ... / Re: So much for Opteron 32bit compatibility ... 2 Re: Space Invaders was: Re: Fortran Guru requested2 Re: Space Invaders was: Re: Fortran Guru requested- tcp/ip for vms 5.3 compat. w. vms 6.2 and 7.1 / Re: The Space Invaders minute of the day: /NOOP 9 trademarks (was: RE: OpenVMS and Security - REALLY !!!!!) = Re: trademarks (was: RE: OpenVMS and Security - REALLY !!!!!)  Re: TZ30's and Alpha's? & Re: who is procedure of backup for vms  F ----------------------------------------------------------------------   Date: 3 Apr 2003 12:12:01 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) P Subject: Re: Any way to have a filename beginning with the ampersand character ?3 Message-ID: <DT0NWgXHCAG3@eisner.encompasserve.org>   M In article <03040208362809@dscis6-0.dalsemi.com>, brandon@dalsemi.com writes:  >  > Users may wig out because the % > file they created no longer exists.   ;    I've never seen this.  Is this real or just a user's (or #    applications) misinterpretation?    ------------------------------   Date: 3 Apr 2003 04:28:05 -0800 ( From: qwqwqwqw70@hotmail.com (Jon Power) Subject: Re: COV Sponsors = Message-ID: <71367ac8.0304030428.38dff877@posting.google.com>   	 Dear all,   F Actually, everytime we end up in a comp.os.vms dialogue we do notice aD large increase in the downloads and visits to our VMS section of the	 web site.   F I want to remind you all that I was dragged into this discussion by anD ancient reference to an event on the FreeVMS and me being personallyE insulted. If you are going to open the door then you cannot expect me  not to get involved.  F As for selling on this site, well, yes, there is no doubt that. What IC said, will resonate with some people that are considering migration A and that may result in more business for Sector7. It seems a fair D exchange of value, I get slandered, and in return get a chance to be$ compensated by some 'reasonable' PR.  A However, regardless of what you feel about the subject, what I've < written is the truth and this story has only one conclusion.  @ Now, I will admit, that I dont normally spend this much time and effort on any other subject.  A I 'get off' seeing S7 complete super high compexity projects like + OpenVMS migrations - on time and on budget.   E I also remember being boo'd off stage at DECUS events many years ago, E and also the sheer agression (of the technical folks) when presenting 2 to corporations that were interested on migrating.  3 The situations changed now, we are always welcomed.   F So, the solution to me posting here is simple, dont drag me or S7 intoF these discussions and we typically wont post. Yes, we monitor the newsC groups, and we very rarely respond -- we never sell, but as I said, 5 YOU opened the door, dont be shocked by who walks in.   D By the way, as are doing a very large project for a well known name.D Someone there was reading the news groups and grabbed one of my PM'sC and said "I didnt know S7 did OpenVMS as well, we've been trying to ? get of this system for years and didnt know how to approach it"   F So --- thanks, and unless someome responds to this posting .. I've got real work to do.  D Since, you guys have focused me on VMS migrations again, I think its time for another campaign.   /Jon   ------------------------------   Date: 3 Apr 2003 04:50:24 -0800 ( From: qwqwqwqw70@hotmail.com (Jon Power) Subject: Re: COV Sponsors = Message-ID: <71367ac8.0304030450.27195c3e@posting.google.com>    "Main, Kerry" <Kerry.Main@hp.com> wrote in message news:<BE56C50EA024184DAF48F0B9A47F5CF4040ECF50@kaoexc01.americas.cpqcorp.net>...    OK - I cant resist this.   "Why do people migrate ?"   A The reason is to take their proven application and the associated ? business logic to a more widely accepted platform. Where people + resources are easier to find (and cheeper).   E Corporations dont hate operating systems, they hate staff costs. They  also hate risk.   # "Use J2EE, Java etc etc on OpenVMS"   = Yes, good idea, I agree that OpenVMS is an excellent platform D (provided the third party packages are still supported ... is Tuxedo$ still supported and 'up to date' ??.    < So, look at it from a typical CIO's point of view in a large corporation:  ?  * UNIX / NT systems are installed and running mission critical  applications without problems.C  * There is a significantly larger pool of expertise in UNIX and NT  than OpenVMS  * HP ditches ALPHA <  * DEC tells everyone to go to NT on ALPHA and then cans it.  * VMS people are expensive 3  * VMS people are difficult to find in an emergency   * Will ITANIUM work ?  B  * Will INTEL ditch it to compete with AMD's 64 bit x86 compatible) (not emulation mode for 32 bit)chip set ? E  * HP's primary marketing focus is HP/UX on PA-RISK / ITANIUM -- what ( happens if OpenVMS becomes a liability ?  F Before I get the 'its all about virus's message' back -- I dont recallB many UNIX virus's. and most organization have had no problems with# their UNIX based mid range servers.   D So, why not develop and implement on a system that has market share,: high avaiability to resources, relatively portable between manufacturers.  @ Whenever a CIO of a large enterprise talks to us, one comment isD common "We are fed up being held hostage by that group of VMS guys".  = YOU see this as a technology issue - THEY see it as a hostage 	 sitation.   
 /Jon Power   ------------------------------  # Date: Thu, 03 Apr 2003 14:54:51 GMT # From: "John Smith" <a@nonymous.com>  Subject: Re: COV Sponsors F Message-ID: <%WXia.25687$7Im.697@news04.bloor.is.net.cable.rogers.com>  A "Doc.Cypher" <Use-Author-Address-Header@[127.1]> wrote in message / news:20030403060955.3235.qmail@gacracker.org...    > > > On Wed, 2 Apr 2003, "Main, Kerry" <Kerry.Main@hp.com> wrote: > ? > >Please give us a break .. This is not some newsgroup you can  befuddle6 > >with pure fud to suit your own business strategies. > > > I'd have to agree with that, I can't see Jon winning any new business here.    C Not all current VMS customers frequent c.o.v.. All Jon has to do is C approach some of those customers and pitch them on the idea and I'm F sure he'd find a receptive audience at many of those customers becauseD HP does zero (0) to promote VMS and prove that it has any commercial
 life left.  D Example...Company X uses Sybase and thinks, for some unknown reason,? that it would be a good idea to install VMS clusters for DR and D business continuity. Buzz - wrong answer. No can get current versionA of Sybase on VMS. Think of many other applications...same answer.   E HP likes to say that applications sell hardware....well, count up the E applications that are mainstream on VMS. Next, count up the number of B those applications where you see HP and/or the ISV promoting thoseD applications on VMS - the fingers of one hand should be sufficientlyE large for the exercise - in fact, it may be an order of magnitude too  great.  B If HP isn't expanding the market for VMS then it's a dying market.  E Sue,..care to cc: Gorham/Marcello/Blackmore/Stallard/carly with this?    ------------------------------   Date: 3 Apr 2003 08:26:55 -0800 ( From: qwqwqwqw70@hotmail.com (Jon Power) Subject: Re: COV Sponsors = Message-ID: <71367ac8.0304030826.33ca4639@posting.google.com>    Kerry,   I just read your comments.  I Just a word on your testamonials. IBM just signed Cerner - looks like you  need to update the web site.  	 Feb 2003: L http://www-1.ibm.com/industries/healthcare/doc/content/news/pressrelease/385 240105.html   E As for the Indian Railways, last time I was in India (Training 40 DEC K engineers to do migrations from OpenVMS to Tru64, paid for by DEC) Dr Pandi 8 was working on reimplementing the system on NT with RTR)  D Thanks for the list of OpenVMS sites, we'll be sure to contact them.  ' Your "Key Message" is seriously flawed.   L The technology is irrelevent -- as long as the base technology 'works' -- it4 is market forces and costs that drive the decisions.  J As for your 'car' analogy,, if a car type, regardless of how 'good it is',J is difficult to find spares for, hard to find engineers to work on it, youJ will eventually buy a car that has easy access to spares, mechanics and is cheap to maintain.  E Once again, you are missing the point, corporations are NOT migrating G because they dont like VMS, they are migrating because they have easier A access to skilled resources on other platforms. Corporations dont L personalize operating systems, CIO's however, who have families to feed likeI the rest of us are being told to reduce the cost of IT by as much as 50%.   K By your own "Key Message", organizations should just buy the exact computer 5 systems that fits the exact needs of the application.   I However, in this day and age and financial environment, consolidation and = reduction on supported platforms is key to lowering IT costs.   J Corporations are trying to reduce the number of vendors and platforms to 1 or 2, not INCREASE the number.  E Take a typical situation at home, you have 10 remote controls for the H different systems, your wife is continually on your a** because she cantH remember which remote control does what. The kids are always leaving theK remotes the refridgerator and you've given up explaining why the ability to F do 'on the fly' adjustment of the gamma convergence is so important to watching "Law and Order".   L To simplify your life, you eventually buy 3 universal remotes, consolidatingJ the 10 highly efficient remotes into 3 usable ones that reduce your stress? levels and therefore take away the need to post on comp.os.vms.   D Its not about technology, its about simplification which brings cost
 reduction.  J If VMS was still mainstream, maybe the universal remotes would control VMS systems, but, it isnt.  C Here is the KEY MESSAGE: It's about COSTS and RISK, not technology.   E Today, a new technology sale of any size to a corporate is an ITO (IT  optimization) driven sale.   The key points are:      Server Count Reduction     Support staff consolidation      Ultimately cost reduction   K Why do you think so much effort is being put into dynamic partitioning down 9 to sub processors increments and individual I/O channels?   I Standard 'static' partitioning makes server consolidation when accounting F for peak loads average out at (say) 60% average utilization of a largeE server (if your lucky) and still keep up with peak load requirements.   G Dynamic Partitioning increases the average utilization to around 75% or 8 better (dynamically allocating resources for peak usage)  D 'Compute on Demand' as an adjunct to SCON with DLPAR's, will furtherJ increase the 'safe level' to 85% or better by offloading the peak to other? processors where the corporations just 'pay' for what they need ! (additionally) when they need it.   I Its all about COSTS -- doing more with less NOT DOING THE SAME WITH MORE.    ------------------------------   Date: 3 Apr 2003 09:30:48 -0800 ( From: qwqwqwqw70@hotmail.com (Jon Power) Subject: Re: COV Sponsors = Message-ID: <71367ac8.0304030930.14d79684@posting.google.com>   d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<A1udnaRqlcGuORejXTWc3w@metrocast.net>...  % With regard to the "stateless" stuff.   * What I meant was 'higher level' than NFS.   C When all applications use data capsules that have attached URL's to @ objects that can process the data, and the comms is sufficientlyC reliable to move large messages anywhere for service then it really E doesnt matter what the hardware or operating systems is that services ) the request. ("LINUX, UNIX, VMS, oh my").   D We are starting to see this happening more and more, especially with web based apps.   = The middleware (Tuxedo, MQSeries, RTR) handles failover, load ! balancing and reliable messaging.   C Once we see that, then the concept of hardware and operating system ( software handling failover is redundant.  B Its not practical today, to ship 20Mb of application code very farE (geographically) especially when that code is specifically tied to an 5 operating system or even an operating system version.    Its only a matter of time.   /Jon   ------------------------------   Date: 3 Apr 2003 12:29:58 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: COV Sponsors 3 Message-ID: <hhRrcQvtKpqS@eisner.encompasserve.org>   h In article <71367ac8.0304021011.69264b58@posting.google.com>, qwqwqwqw70@hotmail.com (Jon Power) writes: > 5 > I keep wondering why you guys dont get the 'point'.   &    Because you've got the wrong point.  H > The number of midrange UNIX(s) server out there must be 10 to 15 times > the  > installed base of VMS.      Not important.   @ > Most every major corporation is running their mission critical > applications% > or UNIX/VMS or zOS (used to be MVS)   !    They're all losing money, too.   E > Somehow, US industry keeps running without the incredibly important  > UNIQUE/ > technology that you believe VMS has to offer.  >  > How can this be ????  A    They don't.  It's costing them out the whazoo and they haven't     figured it out yet.  B > How can most of the enterprises keep an IT infrastructure up and	 > running  > WITHOUT VMS ??????????????      They don't.  F > Come to think of it, Why didnt HP just DUMP Tru64 and HP/UX and only > sell > OpenVMS ??  E    HP doesn't care what you run, they just want you to by it from HP.   H > Could it possibly be that the same capabilities offered thru differentD > channels is available on almost every computer system out there ??      Nope.  D > If we are going to discuss a truly unique offering in the industry	 > then we  > should be talking TANDEM.  > ? > FAULT TOLERANT.  Not just 5x9 highly available, but NON STOP.  > C > So, if technology is the 'key' to winning the war, I would expect  > every VMS 0 > crusader to suggest that everyone buys TANDEM.  6    VMS' 17 year uptime record is pretty damn non-stop.  G > After all, if all of this great VMS clustering technology beats UNIX,  > and E > its all about technology, then TANDEM obviously knocks VMS and UNIX  > and NTC > into obscurity. Who cares about cluster failover when the systems  > simply > don't stop running ?  G    In the minds of the spenders, it's not all about technology.  That's     why they keep spending.    H > Everyone will agree that OpenVMS has better base clustering capability > thanA > UNIX & NT (you'll have to dig deep to make the case against the  > mainframe) -D > but today's applications make use of the 3rd party middleware that
 > bypasses@ > the need for such an integrated base, all inclusive clusteringH > capability and uplevel the failover, load balancing to the message and > DB layer where it should be.  '    Why should an application be needed?   A > If you look at the progression of VMS layered products, you can 
 > clearly see F > the movement away from the old style 'everything in the application'	 > to what   > we see in the 3/4 tier systems      Duh, what I just said.   B > SMG & RMS (data and screen control tightly embedded in the code)D > FMS  (attempt to move the screen maintenance out of the code - but > still  > having tightly bound UAR's; > TDMS & Rdb - failed attempt to combine UI and data access H > DEC FORMS, ACMS, Rdb - multi tiered abstracted tool set, allowing true@ > separation of the UI, business logic, transaction control (and > balancing) > and database.  > & > So, where are these products today ?  )    SMG, RMS, Rdb, ... are alive and well.    F > DEC FORMS - replaced with Oracle forms or client UI/HTML/Slim Client- > ACMS - history books - replaced with Tuxedo & > Rdb - faded and replaced with Oracle > G > DEC's best product that should have been a world killer was RTR. This  > piece A > of middleware handled shadow servers and load balancing between E > geographically separated servers with complete "non clustered" load 
 > balancedD > functionality in the middleware. Guess what -- it was available on
 > OpenVMS,$ > Tru64, AIX, SOLARIS, HP and NT ...  F    Real work takes more than RTR, and those applications sound way too"    much like the few Andrew knows.  F > We had to do this at Sector7 -- we were the king of the hill when it	 > came to B > VMS migration, we felt comfortable, knew everything - but we are	 > finding 7 > less and less VMS systems available for our services.   C   Fewer and fewer people are dumb enough to want to get off VMS and !   you think that's a VMS problem?    >  > /Jon Power  	    Troll.    ------------------------------  # Date: Thu, 03 Apr 2003 07:05:33 GMT & From: "Rob Boor" <rob.boor@compaq.com>* Subject: Re: DBI prebuilt modules for perl/ Message-ID: <13Ria.208$9l3.75@news.cpqcorp.net>    Hi,   H for a testing project with MySQL server running on OpenVMS, I managed to build F DBI V1.35 & DBD-MYSQL V1.026 with OpenVMS V7.3-1 and TCPIP V5.3 (eco1)< with some minor modifications. System is running Perl V5.6-1  L Drop me a mail if you are interrested in e.g. a backup saveset of this build8 and/or instructions how to built it yourself from source   regards,     Rob Boor     OpenVMS Support Specialist.    Hewlett-Packard Nederland B.V. Utrecht, The Netherlands  6 * robDOTboor@hpDOTcom         (replace "DOT" with .'s)  8 "Joseph Norris" <jozefn@bolt.sonic.net> wrote in message> news:Pine.LNX.4.40.0304021151570.3210-100000@bolt.sonic.net... >  > Hello, > A > Does anyone know where I can get pre-built DBI modules for perl 	 > on VMS?  >  > Thanks to all. > = > #Joseph Norris (Perl - what else is there?/Linux/CGI/Mysql) K > print @c=map chr $_+100,(6,17,15,16,-68,-3,10,11,16,4,1,14,-68,12,1,14,8, E > -68,4,-3,-1,7,1,14,-68,-26,11,15,1,12,4,-68,-22,11,14,14,5,15,-90);  >  >    ------------------------------  $ Date: Thu, 3 Apr 2003 11:13:43 -0500. From: "Jerry Alan Braga" <jabraga@flanagan.ca>* Subject: Re: DBI prebuilt modules for perl1 Message-ID: <M5Zia.65$7b6.168370@news.on.tac.net>   5 does anyone know if php on vms supports mysql as well     1 "Rob Boor" <rob.boor@compaq.com> wrote in message ) news:13Ria.208$9l3.75@news.cpqcorp.net...  > Hi,  > J > for a testing project with MySQL server running on OpenVMS, I managed to > build H > DBI V1.35 & DBD-MYSQL V1.026 with OpenVMS V7.3-1 and TCPIP V5.3 (eco1)> > with some minor modifications. System is running Perl V5.6-1 > H > Drop me a mail if you are interrested in e.g. a backup saveset of this build : > and/or instructions how to built it yourself from source > 
 > regards, >     Rob Boor! >     OpenVMS Support Specialist.  >   > Hewlett-Packard Nederland B.V. > Utrecht, The Netherlands > 8 > * robDOTboor@hpDOTcom         (replace "DOT" with .'s) > : > "Joseph Norris" <jozefn@bolt.sonic.net> wrote in message@ > news:Pine.LNX.4.40.0304021151570.3210-100000@bolt.sonic.net... > > 
 > > Hello, > > C > > Does anyone know where I can get pre-built DBI modules for perl  > > on VMS?  > >  > > Thanks to all. > > ? > > #Joseph Norris (Perl - what else is there?/Linux/CGI/Mysql)  > > print @c=map chr8 $_+100,(6,17,15,16,-68,-3,10,11,16,4,1,14,-68,12,1,14,8,G > > -68,4,-3,-1,7,1,14,-68,-26,11,15,1,12,4,-68,-22,11,14,14,5,15,-90);  > >  > >  >  >    ------------------------------   Date: 3 Apr 2003 08:47:18 -0800 : From: craig.berry@SignalTreeSolutions.com (Craig A. Berry)* Subject: Re: DBI prebuilt modules for perl= Message-ID: <7f15589f.0304030847.32973a6f@posting.google.com>   w Joseph Norris <jozefn@bolt.sonic.net> wrote in message news:<Pine.LNX.4.40.0304021151570.3210-100000@bolt.sonic.net>...   A > Does anyone know where I can get pre-built DBI modules for perl 	 > on VMS?   9 The pre-built Perl 5.6.1 kit available here includes DBI:   * http://www.sidhe.org/vmsperl/prebuilt.html  E It is possible to build DBI yourself, but dependiing on Perl version, @ DBI version, and some other variables, it can be a real headacheA because DBI uses a highly customized build procedure wherein many E things can go wrong.  The vmsperl mailing list archive includes quite $ a bit of discussion of this problem:  1 http://www.xray.mpe.mpg.de/mailing-lists/vmsperl/     B You'll also, of course, need a DBD module that is specific to your: database, and you'll probably have to build that yourself.   ------------------------------  % Date: Thu, 03 Apr 2003 07:11:33 +0000 7 From: David B Sneddon - bigpond <dbsneddon@bigpond.com> 7 Subject: Re: Do OpenVMS Alpha have a Year 2057 problem? ( Message-ID: <3E8BDEA5.50005@bigpond.com>   Rudolf Wingert wrote:  > Hello, > L > today I did show an OpenVMS novize the SET TIME command. Within the HELP IH > did see that a legal year specification for Alpha is 1957 to 2056 (forG > OpenVMS vax from 1858 to 9999). In case of this, my question is, does J > OpenVMS AXP have a year 2057 problem? Or why do I have this restriction? > % > TIA and best regards Rudolf Wingert  >  >   B This is what the year defaults to if you ONLY SUPPLY TWO DIGITS...    From the help:   9           Alpha systems also allow a 2-digit year format:   &           [dd-mmm-yy[:]] [hh:mm:ss.cc]  I           The century applied to the 2-digit year format is determined by I           the corresponding year in the supported Alpha date range (1957-            2056).   Regards, Dave.  --  I David B Sneddon (dbs)    VMS Systems Programmer     dbsneddon@bigpond.com I Sneddo's quick guide ...          http://www.users.bigpond.com/dbsneddon/ I DBS freeware at ...   http://www.users.bigpond.com/dbsneddon/software.htm I "Life is what happens to you while you're busy making other plans" Lennon    ------------------------------   Date: 3 Apr 2003 09:37:40 +0200 ' From: huber@mppmu.mpg.de (Joseph Huber) 7 Subject: Re: Do OpenVMS Alpha have a Year 2057 problem? + Message-ID: <yDUAqLTfSiSQ@vms.mppmu.mpg.de>   e In article <MDEJJFGEEOPAFONJONBKOEPNCOAA.win@fom.fgan.de>, "Rudolf Wingert" <win@fom.fgan.de> writes: L > today I did show an OpenVMS novize the SET TIME command. Within the HELP IH > did see that a legal year specification for Alpha is 1957 to 2056 (forG > OpenVMS vax from 1858 to 9999). In case of this, my question is, does J > OpenVMS AXP have a year 2057 problem? Or why do I have this restriction?  P Certainly VMS has no problem, but probably the Alpha firmware (something called  "interval system clock"). * Do You expect to have Alphas in 2057 ? :-)   --  N Joseph "Sepp" Huber   mailto:joseph.huber@web.de   http://www.huber-joseph.de/   ------------------------------  % Date: Thu, 03 Apr 2003 12:48:52 +0200 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> 7 Subject: Re: Do OpenVMS Alpha have a Year 2057 problem? ' Message-ID: <3E8C1194.295CEDE6@aaa.com>   3 You might have to handle e.g. a 50 year house loan, & then you might get problems in 2007...   Jan-Erik Sderholm.    Joseph Huber wrote:  > , > Do You expect to have Alphas in 2057 ? :-)   ------------------------------   Date: 3 Apr 2003 12:59:57 +0200 ' From: huber@mppmu.mpg.de (Joseph Huber) 7 Subject: Re: Do OpenVMS Alpha have a Year 2057 problem? + Message-ID: <pnqTnE4FiL2U@vms.mppmu.mpg.de>   c In article <3E8C1194.295CEDE6@aaa.com>, Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> writes: 5 > You might have to handle e.g. a 50 year house loan, ( > then you might get problems in 2007... > Jan-Erik Sderholm.   3 Of course, and VMS has no problem with such dates.  H The question was for the SET TIME command only, which has the limitation (on Alpha).     --  N Joseph "Sepp" Huber   mailto:joseph.huber@web.de   http://www.huber-joseph.de/   ------------------------------  % Date: Thu, 03 Apr 2003 05:09:56 -0600  From: brandon@dalsemi.com 7 Subject: Re: Do OpenVMS Alpha have a Year 2057 problem? 1 Message-ID: <03040305095611@dscis6-0.dalsemi.com>   R > Certainly VMS has no problem, but probably the Alpha firmware (something called  > "interval system clock"). , > Do You expect to have Alphas in 2057 ? :-)  N 2057?  I will be 98.  I wonder if my own SET TIME algorythm will be functional at that time...    John Brandon VMS Systems Administrator  Dallas Semiconductor john.brandon@dalsemi.com 972.371.4172 wk  972.371.4003 fx    ------------------------------  # Date: Thu, 03 Apr 2003 14:02:30 GMT 3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond) 7 Subject: Re: Do OpenVMS Alpha have a Year 2057 problem? 0 Message-ID: <W9Xia.215$0C3.206@news.cpqcorp.net>  e In article <MDEJJFGEEOPAFONJONBKOEPNCOAA.win@fom.fgan.de>, "Rudolf Wingert" <win@fom.fgan.de> writes:  >Hello,  > K >today I did show an OpenVMS novize the SET TIME command. Within the HELP I G >did see that a legal year specification for Alpha is 1957 to 2056 (for $ >OpenVMS vax from 1858 to 9999). ...5                                                       K This is a documentation error.  Alpha will accept dates outside this range.  Here is an example:        $ sho time        2-APR-2003 08:58:27     $ set time = 02-apr-4321     $ sho time        2-APR-4321 00:00:02  J As explained further along in the help entry, if you enter a 2 digit year,/ it will be assumed bo be between 1957 and 2056.    --  J       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------   Date: 3 Apr 2003 10:21:31 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 0 Subject: Re: Earth to Andrew, do you get it now?3 Message-ID: <Iqnq5hQ1aczP@eisner.encompasserve.org>   h In article <d7791aa1.0304021028.21bdba3f@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:o > qwqwqwqw70@hotmail.com (Jon Power) wrote in message news:<71367ac8.0304011251.24376d14@posting.google.com>...  >>  < >> In the old days --- VMS had much more to offer than UNIX.      Still does.  I >> These days -- who cares. Everyone has their favorite editor, most work B >> on VMS and UNIX, make/mms/sccs,rcs/cms --- is a 30 minute RTFM.  E    Nope.  CMS had features I've never found in sccs, RCS, or CVS, but A    I like CVS and I'm still learning it (OBTW, runs on VMS, too).   H    make is just plain wrong.  It's got more bugs than any other softwareE    tool I've ever used.  Not to mention MMS has features make doesn't     even try to do.  H >> The issue is not "which is best" -- the issue is which system provide8 >> the best opportunity for keeping personally employed.  F    The issue is which is best for the job I'm trying to do.  On all myG    jobs VMS is best, even thought I have been forced to do some of them     on other systems.  D    VMS has features UNIX doesn't even try to have.  Some folks dnt's5    need all of them on every job, but they are there.   I >> I see thousands of mission critical systems running just fine on UNIX, I >> I even see many working just as well on NT (yup - I know thats hard to  >> swallow) and VMS.  G    I see mission critical systems that don't run correctly because they #    need features UNIX doesn't have.   D >> Does anyone these days truly believe that VMS has some magic thatG >> makes it light years ahead ?? -- even if it did -- would anyone take  >> the risk of using it ?   &    Yes, and of coure there is no risk.  C >> Take RTR (Reliable Transaction Router) -- probably the best kept A >> secret in the world. Reliable messages transfered with minimal F >> overhead, asychronous event handling ... good old DEC technology ..F >> its still around today (I think) but who would take the risk of newF >> development on RTRV3?. It may be better than Tuxedo or MQSeries ...I >> but they are mainstream and the same problems can be solved with all 3  >> products.  '    People who know what they are doing.   I >> Who cares which is better ?, the question is do you want keep employed D >> and which system offer the best and most lucrative opportunities.  D    No, the question is do I want to do the job, or just hack around.C    Do I want to have real work or just be another dime-a-dozen UNIX     programmer?  D >> On that subject --- as more VMS systems are abandoned and the VMSE >> skill sets become hader to acuire - the hourly rate for VMS people F >> will go up -- but its temporary --- the highest paid people will beF >> those that have BOTH skill sets and can help organizations maintain >> AND TRANSITION.  @    We pay higher for those who actually know what they're doing.   ------------------------------   Date: 3 Apr 2003 06:03:14 GMT / From: "Dave Weatherall" <djweath@attglobal.net> # Subject: Re: Fortran Guru requested 5 Message-ID: <DTiotGxQ0bj6-pn2-y3dE2EpRHdmw@localhost>   / On Wed, 2 Apr 2003 08:42:54 UTC, Paddy O'Brien  $ <paddy.o'brien@tg.nsw.gov.au> wrote:  I > Happily, it seems that we have not screwed up, nor has the compiler or  	 > linker.  > K > Didier posted a source which was not that of the executable.  So none of  ) > us "gurus" are able to see the problem.   + I've compiled the source Didier posted on :     	Alpha/VMS 6.2 with	Fortran V7.1 	Alpha/VMS 7.1-2	Fortran 7.5 	Vax/VMS		Fortran 6.4   # No crashes on any of the platforms.   F By adding terminal type '112' to the list of valid terminals it works  just fine on my Decterms.   E Stylistically, it does not conform with my own ideas but that's just  C personal preference. (Using doubles to hold integers is not one of  ; 'em, particularly since Alpha became the main VMS platform)   E I was taken aback by the use of the SCR$ cursor routine. I assume it  F comes from the SCRRTL shareable image which the .EXE invokes. I can't E find any documentation for it in the VMS 7.1 manuals. I have a vague  C recollection from decades ago. Just shows how good VMS has been at  C backward compatibility. Anybody ready to try and get it running on  
 IPF/VMS?     --   Cheers - Dave.   ------------------------------  $ Date: Thu, 3 Apr 2003 11:32:50 -0500; From: "Brian Tillman" <Brian.Tillman@notnoone.notnohow.com> # Subject: Re: Fortran Guru requested $ Message-ID: <3e8c622f$1@news.si.com>  , >I've compiled the source Didier posted on :  K I've looked at all the posts from Didier on my news server and I can't find , any source, except for a couple of excerpts. --  I Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot com 5 Smiths Aerospace  Addresses modified to prevent SPAM. @ 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991p8        This opinion doesn't represent that of my company   ------------------------------  $ Date: Thu, 3 Apr 2003 08:38:28 -0800# From: "Tom Linden" <tom@kednos.com>5# Subject: RE: Fortran Guru requested 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIGEOKGOAA.tom@kednos.com>   
 posted 29-MARO   >-----Original Message-----rA >From: Brian Tillman [mailto:Brian.Tillman@notnoone.notnohow.com]V' >Sent: Thursday, April 03, 2003 8:33 AMa >To: Info-VAX@Mvb.Saic.Com$ >Subject: Re: Fortran Guru requested >o >a- >>I've compiled the source Didier posted on :e >tL >I've looked at all the posts from Didier on my news server and I can't find- >any source, except for a couple of excerpts.o >--gJ >Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot com6 >Smiths Aerospace  Addresses modified to prevent SPAM.A >3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "."  >Grand Rapids, MI 49512-19919 >       This opinion doesn't represent that of my companyW >  >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). @ >Version: 6.0.467 / Virus Database: 266 - Release Date: 4/1/2003 >m ---a& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.467 / Virus Database: 266 - Release Date: 4/1/2003-   ------------------------------   Date: 3 Apr 2003 08:53:18 -0800 - From: goathunter@goatley.com (Hunter Goatley)e) Subject: Re: ftp_mirror and TCPIP serversu= Message-ID: <3ff5fed3.0304030853.6cdc2477@posting.google.com>.  \ Alder <PGDEHMKOKIMD@spammotel.com> wrote in message news:<3E8BBBF9.9000908@spammotel.com>...M > I googled this issue to no avail, so perhaps it's safe to ask :-)  Can the  P > FTP_MIRROR procedure maintained by Hunter Goatley be used against an HP TCPIP  > Services for VMS server? > Q > The documentation is the FTP_MIRROR.COM file itself suggest "NO" as an answer, ME > but I just thought I'd check if anyone had found a way around this.t > I Yes, the current version of FTP_MIRROR.COM works with the TCP/IP Services.H FTP server, too (thanks to Dr. Martin P.J. Zinser).  The readme needs to8 be updated to reflect that.  I'll do that sometime soon.   Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/  goathunter@goatley.com   ------------------------------   Date: 3 Apr 2003 08:17:29 -0800 . From: Jack.Trachtman@vmmc.org (Jack Trachtman)* Subject: How does Shadow Copy REALLY work?< Message-ID: <69d784c4.0304030817.5cf246c@posting.google.com>  @ I've just reread the V7.3 "Volume Shadowing for OpenVMS" manual.  D The text implies that there is some intelligence in a copy operation@ (e.g. see 2nd column in table 7-1).  I've always presumed that aF "copy" was a physical copy, so the time to complete would be primarilyC based on disk size and CPU load.  (This would also cover a possiblegC security situation on the target disk where no blocks would be left8? untouched.)  But the above mentioned table and some of the text 8 seems to say that the copy time is also dependant on howC "different" the source and target disks are (which I can understand'* in a "merge" situation, but not a "copy").  ? Could someone elucidate on what algorithm "copy" actually uses?0   Thanks   ------------------------------  # Date: Thu, 03 Apr 2003 18:06:28 GMTD# From: "John Smith" <a@nonymous.com>n? Subject: Re: How to connect Vaxstation 4000 VLC to LCD monitor?aE Message-ID: <EK_ia.1329$az1.829@news01.bloor.is.net.cable.rogers.com>n  6 "Patrick Young" <P.Young@unsw.EDU.AU> wrote in message7 news:55f85d77.0303172305.2b5cef7f@posting.google.com...b. > "pbc" <pbc@informchaos.com> wrote in message+ news:<b55cn1$7em$1@alpha2.radio-msu.net>...aC > > I wonder is there any way to connect my Vaxstation 4000 VLC (itl has RGBsD > > cable) to a regular LCD monitor or any other monitor without RGB input., > > Any converters? Any other possibilities? > D > My workstation at work is an OpenVMS PWS600AU (Alpha) connected to an9 > 18 inch LCD monitor and it is *sweet*, crisp and clear.  >hC > These monitors however tend to want to run at the resolution theysE > were designed for (in my case 1280x1024), at other resolutions theya > are plain ugly.n    F The 4000/60 had video card options than did the following. I think the VLC had similar video options.   8-plane low-reso 1024x768 @ 60Hzp 1024x768 @ 72Hzo andt 1024x864 @ 60Hz      8-plane high-res 1280x1024 @66Hzg  E Typical 15" LCD panels have 1024x768 native resolution, while the 17"aB one have 1280x1024 (typical). Just make sure you get one that willC synch anywhere between 60-72Hz and you should be fine, once you geti the BNC-VGA adapter.   ------------------------------   Date: 3 Apr 2003 06:03:16 GMTs/ From: "Dave Weatherall" <djweath@attglobal.net>  Subject: Re: How to get timing5 Message-ID: <DTiotGxQ0bj6-pn2-2VBLMdXc4J7j@localhost>d  B On Mon, 31 Mar 2003 04:03:36 UTC, texascheryl@hotmail.com (Cheryl  Hoefelmeyer) wrote:   H > In a previous post, to demonstrate the speed of a spawn, W.F.J Mueller > wrote: >  > >In one session execute. > >e> > > $                         write sys$output "1 - "+f$time()> > > $ spawn/nolog/nosym/nokey write sys$output "2 - "+f$time()> > > $                         write sys$output "3 - "+f$time()> > > $ spawn                   write sys$output "4 - "+f$time() > >yB > >In the other execute the certainly CPU bound program, e.g. this > Fortran code > >n > >         program allcpu > > 100     continue > >         goto 100 > >         endh > >lH > >Now measure the time of a SPAWN with and without a competing process. > On a= > >DEC 3000 - M300LX and with a heavy symbol and logical name 
 > environmenteC > >(about 140 process logicals and 700 symbols) I got the followingc	 > timing:  > >K) > >   Node idle CPU bound process runningg > >?! > > SPAWN  0.720 sec  146 sec !!!r% > > SPAWN/NO.. 0.270 sec   19 sec !!!u > >s > E > What command(s) is/are used to get the timing? I'd like to try thisi > experiment myself.  	 Hi CherylrF                     no-one seems to have responded so a quick one fromF me. There is a pair of LIB$ routines, which I can't check 'cos I'm at B home, that you can use in code to obtain statistics about process 9 resources between calls. It might be LIB$START_TIMER and rF LIB$STOP_TIMER. You can can then obtain CPU and Elapsed time usage as @ well as the no. of Page Faults. HELP LIB$ might get you started.   -- d Cheers - Dave.   ------------------------------  % Date: Thu, 03 Apr 2003 12:06:21 +0100g9 From: Alan Adams <alan.adams@orchard-way.freeserve.co.uk>i Subject: Re: How to get timing? Message-ID: <c2e151dd4b.Alan.Adams@orchard-way.freeserve.co.uk>e  4 In message <DTiotGxQ0bj6-pn2-2VBLMdXc4J7j@localhost>:           "Dave Weatherall" <djweath@attglobal.net> wrote:  D > On Mon, 31 Mar 2003 04:03:36 UTC, texascheryl@hotmail.com (Cheryl  > Hoefelmeyer) wrote:m > J > > In a previous post, to demonstrate the speed of a spawn, W.F.J Mueller
 > > wrote: > >  > > >In one session execute. > > > @ > > > $                         write sys$output "1 - "+f$time()@ > > > $ spawn/nolog/nosym/nokey write sys$output "2 - "+f$time()@ > > > $                         write sys$output "3 - "+f$time()@ > > > $ spawn                   write sys$output "4 - "+f$time() > > >tD > > >In the other execute the certainly CPU bound program, e.g. this > > Fortran code > > >n > > >         program allcpu > > > 100     continue > > >         goto 100 > > >         ends > > >EJ > > >Now measure the time of a SPAWN with and without a competing process. > > On a? > > >DEC 3000 - M300LX and with a heavy symbol and logical nameo > > environment:E > > >(about 140 process logicals and 700 symbols) I got the followingt > > timing:. > > > + > > >   Node idle CPU bound process running1 > > >e# > > > SPAWN  0.720 sec  146 sec !!!5' > > > SPAWN/NO.. 0.270 sec   19 sec !!!  > > >0 > > G > > What command(s) is/are used to get the timing? I'd like to try thisT > > experiment myself. >  > Hi CheryldH >                     no-one seems to have responded so a quick one fromH > me. There is a pair of LIB$ routines, which I can't check 'cos I'm at D > home, that you can use in code to obtain statistics about process ; > resources between calls. It might be LIB$START_TIMER and nH > LIB$STOP_TIMER. You can can then obtain CPU and Elapsed time usage as B > well as the no. of Page Faults. HELP LIB$ might get you started. >     It's LIB$INIT_TIMER to start it,@ LIB$SHOW_TIMER will output the elapsed statistics to SYS$OUTPUT,G LIB$STAT_TIMER will return one of the 5 statistics for each call to it.B   Alan   -- w
 Alan Adams& alan.adams@orchard-way.freeserve.co.uk http://www.nckc.org.uk/t   ------------------------------  $ Date: Thu, 3 Apr 2003 11:29:19 -0500; From: "Brian Tillman" <Brian.Tillman@notnoone.notnohow.com>.9 Subject: Re: Include an automated date within a printout.u$ Message-ID: <3e8c615c$1@news.si.com>  1 >Could you elaborate on "not _that_ hard" please.d  J Well, I wrote a modified print symbiont (in Fortran, no less), and it onlyL took a couple of days to get it working, using the VMS documentation.  So, II guess it means "if I can do it, and if you can read, you can do it, too."t -- sI Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot comr5 Smiths Aerospace  Addresses modified to prevent SPAM.V@ 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991r8        This opinion doesn't represent that of my company   ------------------------------   Date: 3 Apr 2003 10:55:03 -0800 , From: JohnEllicottington@lycos.co.uk (Johno)9 Subject: Re: Include an automated date within a printout. = Message-ID: <10822590.0304031055.1662334e@posting.google.com>g  t Rob Brown <brown@gmcl.com> wrote in message news:<Pine.LNX.4.44.0304021124320.24240-100000@localhost.localdomain>...J > > >Does anyone know a way I can get a standard ASCII text file to outputJ > > >to the printer on a VAX VMS/DCL system and contained with the output, > > >regardless when run,l >  > How about $ PRINT/HEADER > ?  RobsD This will just produce a header page prior to printing of the actual file.gC I do not think I have been very clear here. I have a standard asciiiE text file which runs to about 40/50 lines of instructions. In the toplD right had corner of the file is 'Day: Friday only' and below this isD 'Date: __________'. This is where I would write the appropriate dateD on the actual printout. This is what I was hoping could be automatedD and the date actually inserted into the file at this point each time it is produced.  Johnol   ------------------------------  $ Date: Thu, 3 Apr 2003 09:55:48 -0800= From: "Lee Courtney" <lcourtney@-REMOVETHISFILTER-mvista.com>sG Subject: Re: Inquirer: HP's Alpha RetainTrust programme a complete busta/ Message-ID: <v8otgtbtat4ued@corp.supernews.com>f  J I have been lurking in this news group for a couple years and observe thatI this sounds like the HP3000 and MPE/iX a few years ago. Where is MPE now?c  H I strongly suspect the same sad MPE story is playing out with VMS. FirstJ many heart-rending assurances that all is well, we're behind the platform,J we're porting MPE to Itanic, we're doing more advertising - all were true.D Then HP drops a bomb that they are end of life'ing the platform. TheK result - droves of loyal HPcustomers headed for the AS400 and non-HP NT ande Unix boxes.l   Lee Courtney  . "John Smith" <a@nonymous.com> wrote in message@ news:Xhpia.2368$7Im.1820@news04.bloor.is.net.cable.rogers.com...
 > March 28+ >  http://www.theinquirer.net/?article=8585e >c? > ....Perhaps the BCS layoffs taking place this week, initiallytB > targeting marketingg resources, will help set things straight byF > dispensing with ineffectual marketeers whose continued employment atA > HP has ben nothing but a drain on stock value and HP resources.y >tE > Given the shaky position of the soon-to-be pasture-ized Tru64 UNIX,iF > and HP's conscious decision to NOT promote OpenVMS to new customers,H > perhaps the lack of knowledge of the RetainTrust Program will serve as' > a wake-up call to aneemic marketeers.  >  > ...... >t@ > We can only hope this is the case. A quick read of HP's recentG > financials and some informed extrapolation will prove that OpenVMS isaG > a $800USD-in-profit cash cow that is being milked to the tune of mores> > than $100M USD per annum to prop up Tru64 UNIX and the RiskyH > Consolidated Enterprise UNIX scheme (made even riskier since the Tru64E > UNIX kernel developers, whose participation is essential to the new F > and improved UNIX offering, have been cashiered). Mo Hello, HP. Have > you woken up yet?e >r > -------------------- >sF > If I were the average VMS customer, I'd be really pissed about this. >PD > If I were the average VMS customer who was fence-sitting about theE > future of VMS and considering unix, I'd consider AIX more seriouslyfG > than HP-UX after reading this. And if I was considering Linux, I'd beO7 > waiting for the Dell Itanics before I'd buy a HP one.  > B > If I was the average VMS ISV, I'd be furiously porting to linux. >m >w >w   ------------------------------  $ Date: Thu, 3 Apr 2003 18:04:34 +0200$ From: Michael Unger <unger@decus.de>< Subject: Re: latest Business Critical Systems Customer Times* Message-ID: <00A1DD93.F255D36B.7@decus.de>  4 "Sue Skonetski" <susan_skonetski@hotmail.com> wrote:  D > here is the pointer to the latest BCS Customer newsletter.  If you" > need the PDF please let me know. >  > [...]h    I Apparently links to PDF versions (whole document as well as selected, OS m; specific, parts) are available from the web site mentioned.h   Michaelw   ------------------------------   Date: 3 Apr 2003 09:53:43 +0200X' From: huber@mppmu.mpg.de (Joseph Huber)e# Subject: Re: MS-Word from OpenVMS ?a+ Message-ID: <6O$9nmQO7H+x@vms.mppmu.mpg.de>(  ` In article <f%Eia.38675$cB3.239672@nnrp1.uunet.ca>, "Chris Moore" <moore_mc@hotmail.com> writes:J > I did manage to get CATDOC from Process....(though so far I haven't beenD > able to get it to actually do anything....but that's probably just > me....I'll keep at it) > L > What I really SHOULD have included in the post was that the requirement isI > for character-cell terminal use only (factory environment), so graphicsJ > suggestions are a no-go. >   ? There is also ANTIWORD, it does both, Ascii text (like catdoc),a: and Postscript output. Find the bits to build it on VMS at(   http://wwwvms.mppmu.mpg.de/~huber/pds/   ------------------------------  % Date: Thu, 03 Apr 2003 11:20:22 +0200o' From: JOUKJ <joukj@hrem.stm.tudelft.nl>i# Subject: Re: MS-Word from OpenVMS ?g* Message-ID: <b6gud1$843$1@news.tudelft.nl>   Patrick Spinler wrote:H > Abiword <http://www.abisource.com/> is a fairly nice lightweight word J > processor which has filters for incoming MS docs, modulo some features. / >  It might be possible to port abiword to VMS.  > E > The major dependancy I recall in the Abiword source is for the GTK sF > toolkit.  There exists a port of the GTK graphics toolkit to VMS at K > <http://h71000.www7.hp.com/openvms/products/ips/gtk.html>.  I don't know  ' > if it's recent enough to work or not.oG It not. abiword needs gtk2.x. Presently I'm working on a port of gtk2.xu/ Some part are already available via my web-pagea)      http://nchrem.tnw.tudelft.nl/openvmsI See glib2 , atk and pango." I did not do any runtime test yet.> Note : -everything is pre-alpha. and some parts may be broken.E         -I have no timescale planned to finish it up (need some spare           moments)o > J > Unfortunately, Abiword uses GNU autoconf so you'd have to manually hack G > together a config.h and some Makefiles. :-(  Fortunately, it doesn't  + > appear to use the unixy fork() at all.:-)f > B And Abiword supports presently only a subset of the MSword syntax.  E I know that there is a group (I do not know who) which are working on 8 a port of Openoffice. Does anybody know more about this?                       Jouk   ------------------------------  # Date: Thu, 03 Apr 2003 18:41:01 GMT # From: "John Smith" <a@nonymous.com>l# Subject: Re: MS-Word from OpenVMS ?gE Message-ID: <1f%ia.1577$az1.182@news01.bloor.is.net.cable.rogers.com>E  5 "Chris Moore" <moore_mc@hotmail.com> wrote in messagei- news:Tsmia.38277$cB3.235406@nnrp1.uunet.ca...yD > Does anyone know of VMS software that can read and display MS-Word docs?i > ; > We have WP 5.1+ for OpenVMS, which allows (via Pathworks). interchange of WPlE > docs between PCs and the REAL computer (unashamed VMS bigotry), butN not Word@ > stuff.  At this point, display-only would be more than enough. > C > I went through this search about 2 years ago without success, but  thea > PC-types are at me again.i    C It's been 10+ years since I've worked with WP on a VAX - still havei the keyboard overlay someplace.a  E If WP 5.1+ can handle .rtf files then that's an option on the MS-Word 
 save menu.  C If WP can handle an older Word format, that too is on the Word savel= menu (Office 97 will save in Word 2.0 and  95 and Works (wps)i	 formats).c  C Word can also save documents in some WordPerfect formats (Office 97j5 will save in WP 5.0 and 5.1 formats for Dos and Win).t  F You may lose some formatting on the conversion/save to an earlier Word or WordPerfect format    ------------------------------  % Date: Thu, 03 Apr 2003 12:14:08 +0100i9 From: Alan Adams <alan.adams@orchard-way.freeserve.co.uk>eY Subject: Re: Mystery solved? (Re: [Q] Files not found during image backup, w/ and  w/o dio? Message-ID: <239852dd4b.Alan.Adams@orchard-way.freeserve.co.uk>h  < In message <b096a4ee.0304021446.6ad45983@posting.google.com>9           spamsink2001@yahoo.com (Alan E. Feldman) wrote:U  [ > Nic Clews <sendspamhere@127.0.0.1> wrote in message news:<3E8A8ACD.85760065@127.0.0.1>...w > > "Alan E. Feldman" wrote: > > > y > > > spamsink2001@yahoo.com (Alan E. Feldman) wrote in message news:<b096a4ee.0303211500.3ab58fcf@posting.google.com>...w > >  > > 5 > > Clarification: we're talking BACKUP/IMAGE here...e > >  > > C > > > 1. BACKUP performs a scan of INDEXF.SYS to build a file list.i > > L > > Yes, and this doesn't change. BACKUP "snapshots" the disc when it scans,A > > and this list does not alter even if files are added/removed.f > > H > > > 2. BACKUP begins walking the directory structure and saving files,K > > > marking the corresponding entry for each saved file in the index list   > > > as saved as it goes along. > > I > > Not quite. BACKUP does not take account of any files added or removedd3 > > since the BACKUP command and the scan happened.h >  > ? > Yes, I know. I thought that was consistent with what I said. - >  >   L > > > 3. BACKUP encounters directory XX.DIR and starts reading its contents. > > > B > > > 4. BACKUP reads the file id for XXMONREM.LOG;2087 in XX.DIR. > > J > > No, it got this from the BACKUP initialization right at the beginning.L > > File IDs come from the indexf.sys, the directory contents are arbitrary. >  > H > But doesn't it have to match the file id from INDEXF.SYS with the fileF > id in the .DIR file? How else would BACKUP know during the directory1 > walk which file to match up with in INDEXF.SYS?i  o; I don't know whether Backup does this, but it seems likely.p  ' From the INDEXF.SYS obtain the file-id.n Open the file by IDNK Read the directory back-link pointer, which points to the parent directory.p3 Note that name, open it, and find its dir backlink.e Repeat until 000000.DIR found.( Build the directory path from the names.   e.g. file Z.DAT found by ID. Its backlink points to W.DIR W.DIR points to V.DIR  V.DIR points to P.DIR  P.DIR points to 000000.DIR  # File spec therefore is [P.V.W]Z.DAT   A The interesting bit is with alias directories. A file in multiple4L directories only has one backlink, therefore only one true parent. Using the? above in SYS$COMMON for example will show the "official" paths.   H This feature was I believe behind a problem with ANALYZE/DISK around theK time of VMS5, which kept "correcting" the backlinks in SYS$COMMON. It couldn' prevent an upgrade happening correctly.p   Alan   -- w
 Alan Adams& alan.adams@orchard-way.freeserve.co.uk http://www.nckc.org.uk/9   ------------------------------  % Date: Thu, 03 Apr 2003 16:54:09 +0200sE From: Jan C. =?iso-8859-1?Q?Vorbr=FCggen?= <jvorbrueggen@mediasec.de>QY Subject: Re: Mystery solved? (Re: [Q] Files not found during image backup, w/ and w/o dir + Message-ID: <3E8C4B11.3E90F1EC@mediasec.de>i  @ > > 4. BACKUP reads the file id for XXMONREM.LOG;2087 in XX.DIR. > H > No, it got this from the BACKUP initialization right at the beginning.J > File IDs come from the indexf.sys, the directory contents are arbitrary.   That's not quite correct.l  L When it scans the index file, BACKUP has an in-memory copy of all valid fileL ids, plus some flags for each. When it does the directory scan, it will takeL the file id from the directory entry, and use that as an index into its listL of file ids. It will use the flags associated with this file id to do house-L keeping such as whether this file should be backed up at all, whether it hasK already been backed up (for aliased files), whether it should be deleted or6J the backup pass recorded. And, of course, it will at the end of processingJ backup all files that haven't had their "done" bit set yet - those are theH "lost" files which do not have any directory entry (this is for /IMAGE).  K BACKUP used to employ some bits in the file id itself which it thought were3I empty for this housekeeping. A number of years back (IIRC, before the bignE re-write) a bug turned up in that some action (IIRC, deleting) wasn't.H performed on large volumes...because in the mean time, those "free" bits> had been added to the file index portion of the file id. Ouch!   	Jan   ------------------------------   Date: 3 Apr 2003 09:00:59 -0800o- From: goathunter@goatley.com (Hunter Goatley) - Subject: Re: Netscape gags on VMS FTP Serversi= Message-ID: <3ff5fed3.0304030900.246e9320@posting.google.com>t  X Alder <PGDEHMKOKIMD@spammotel.com> wrote in message news:<3E84B4E7.106@spammotel.com>... > Joseph Huber wrote: [ > > In article <3E7FDCA0.6020801@spammotel.com>, Alder <PGDEHMKOKIMD@spammotel.com> writes:' > > U > > I have given up to use the VMS TCPIP server because of the different assumptions  < > > different clients have about (mis-)handling VMS servers.R > > I have excellent experience with Hunter Goatleys HGFTP server: it can be setupQ > > to completely appear as a Unix server, and then works even with most Windows -+ > > and macintosh clients, Mozilla as well.n > >  o > U > I've now switched over to Hunter's HGFTP and configured it for full UNIX mode with:E > 	HG_FTP_DO_UNIX_LS ALWAYSy > 	HG_FTP_DO_UNIX_LS_INIT TRUE > 	HG_FTP_HIDE_VMS_SYST TRUE > S > Netscape 7.01 for Windows can now navigate (no pun intended) the sub-directories 0S > correctly, but will display a "0" where the name of the current directory should g: > be.  Is this "normal" behaviour under the circumstances? > P > MSIE (version 6 anyway), is now the one which gags and refuses to display the M > anonymous root directory at all.  Pathetic.  Even the MS-DOS FTP client in e > Windows ME will work!g > G Did you use the ANONYMOUS_LOGIN.TEMPLATE file in HG_ROOT:[COM]?  It haspI an example showing how to define a concealed logical for the FTP session. K Long story short, you have to FTP into ANONYMOUS_ROOT:[000000] (or whatevernG logical name you want to use) because of the $^&%$&#ed up commands MSIE F will send it.  If you follow the template example, along with the UNIZF emulation you described setting up above, it should work OK with MSIE.   Hunter ------& Hunter Goatley, goathunter@goatley.com   ------------------------------  * Date: Thu, 3 Apr 2003 17:31:25 +0000 (UTC)+ From: david20@alpha2.mdx.ac.uk (David Webb)g- Subject: Re: Netscape gags on VMS FTP Serverst+ Message-ID: <b6hr5d$cko$1@aquila.mdx.ac.uk>a  m In article <3ff5fed3.0304030900.246e9320@posting.google.com>, goathunter@goatley.com (Hunter Goatley) writes:tY >Alder <PGDEHMKOKIMD@spammotel.com> wrote in message news:<3E84B4E7.106@spammotel.com>...  >> Joseph Huber wrote:\ >> > In article <3E7FDCA0.6020801@spammotel.com>, Alder <PGDEHMKOKIMD@spammotel.com> writes: >> > aV >> > I have given up to use the VMS TCPIP server because of the different assumptions = >> > different clients have about (mis-)handling VMS servers.hS >> > I have excellent experience with Hunter Goatleys HGFTP server: it can be setuphR >> > to completely appear as a Unix server, and then works even with most Windows , >> > and macintosh clients, Mozilla as well. >> >   >> lV >> I've now switched over to Hunter's HGFTP and configured it for full UNIX mode with: >> 	HG_FTP_DO_UNIX_LS ALWAYS >> 	HG_FTP_DO_UNIX_LS_INIT TRUE  >> 	HG_FTP_HIDE_VMS_SYST TRUEu >> tT >> Netscape 7.01 for Windows can now navigate (no pun intended) the sub-directories T >> correctly, but will display a "0" where the name of the current directory should ; >> be.  Is this "normal" behaviour under the circumstances?  >> jQ >> MSIE (version 6 anyway), is now the one which gags and refuses to display the .N >> anonymous root directory at all.  Pathetic.  Even the MS-DOS FTP client in  >> Windows ME will work! >> rH >Did you use the ANONYMOUS_LOGIN.TEMPLATE file in HG_ROOT:[COM]?  It hasJ >an example showing how to define a concealed logical for the FTP session.L >Long story short, you have to FTP into ANONYMOUS_ROOT:[000000] (or whateverH >logical name you want to use) because of the $^&%$&#ed up commands MSIEG >will send it.  If you follow the template example, along with the UNIZ-G >emulation you described setting up above, it should work OK with MSIE.n >@  N Unfortunately unless HGFTP now works with ODS-5 disks it still isn't a general solution to the problem.L Since there are now a number of public domain software ports (and commercial5 products - latest version of Oracle) which work best  J (if not actually require) ODS-5 disks HGFTP needs to be updated to support ODS-5.  
 David Webb VMS and Unix team leader CCSS Middlesex University   >Hunter  >------d' >Hunter Goatley, goathunter@goatley.comb   ------------------------------  * Date: Thu, 3 Apr 2003 10:05:25 +0000 (UTC)+ From: david20@alpha1.mdx.ac.uk (David Webb)M0 Subject: Re: OpenVMS and Security - REALLY !!!!!+ Message-ID: <b6h115$9ru$1@aquila.mdx.ac.uk>Y  q In article <857e9e41.0304021807.2e9d3bb2@posting.google.com>, susan_skonetski@hotmail.com (Sue Skonetski) writes:fO >Just so the newsgroup is aware, HP Law has been informed.  We are trademarked.d  G Personally I think CISCO's use of the term VMS for their VPN ManagementeL Solution is the bigger problem. We have IBM here redesigning our network andN they will be putting in CISCO VPNs. It caused me a minutes confusion when they- said they would be putting in a 'VMS' system.c  
 David Webb VMS and Unix team leader CCSS Middlesex University e   >  >Warm Regards, >Sue > w >Fabio Cardoso <fabiopenvms@yahoo.com.br> wrote in message news:<20030402112209.49269.qmail@web20206.mail.yahoo.com>...f >> Click >> d& >> http://www.consolut.com/openvms.htm >> 1
 >> Regards >>   >> FC  >> h >> ===== >> ==========================- >> Fbio dos Santos Cardoso  >> OpenVMS System Manager. >> Rio de Janeiro - Brazil >> fabiopenvms@yahoo.com.brs >> ==========================e >> r5 >> __________________________________________________  >> Do you Yahoo!?:@ >> Yahoo! Tax Center - File online, calculators, forms, and more >> http://tax.yahoo.comr   ------------------------------  # Date: Thu, 03 Apr 2003 14:45:04 GMTa# From: "John Smith" <a@nonymous.com>00 Subject: Re: OpenVMS and Security - REALLY !!!!!H Message-ID: <QNXia.25521$7Im.20044@news04.bloor.is.net.cable.rogers.com>  8 "David Webb" <david20@alpha1.mdx.ac.uk> wrote in message% news:b6h115$9ru$1@aquila.mdx.ac.uk...-? > In article <857e9e41.0304021807.2e9d3bb2@posting.google.com>,T3 susan_skonetski@hotmail.com (Sue Skonetski) writes:dD > >Just so the newsgroup is aware, HP Law has been informed.  We are trademarked. > > > Personally I think CISCO's use of the term VMS for their VPN
 ManagementB > Solution is the bigger problem. We have IBM here redesigning our network and F > they will be putting in CISCO VPNs. It caused me a minutes confusion	 when theyN/ > said they would be putting in a 'VMS' system.e >    David,  C The real reason Cisco named their system VMS is that EVERYONE knowso that 'VMS' from HP is dead.    ------------------------------  # Date: Thu, 03 Apr 2003 15:32:36 GMT # From: "John Smith" <a@nonymous.com>j0 Subject: Re: OpenVMS and Security - REALLY !!!!!H Message-ID: <ouYia.26230$7Im.24759@news04.bloor.is.net.cable.rogers.com>  8 "David Webb" <david20@alpha1.mdx.ac.uk> wrote in message% news:b6h115$9ru$1@aquila.mdx.ac.uk...>? > In article <857e9e41.0304021807.2e9d3bb2@posting.google.com>, 3 susan_skonetski@hotmail.com (Sue Skonetski) writes: D > >Just so the newsgroup is aware, HP Law has been informed.  We are trademarked. > > > Personally I think CISCO's use of the term VMS for their VPN
 ManagementB > Solution is the bigger problem. We have IBM here redesigning our network andsF > they will be putting in CISCO VPNs. It caused me a minutes confusion	 when theyi/ > said they would be putting in a 'VMS' system..     David,  C The real reason Cisco named their system VMS is that EVERYONE knows ; that 'VMS' from HP is dead. How could anyone have any otheroD impression? No marketing. No advertising. No announcements about VMSF other than to the converted. No effort made to acquiring new customersA for VMS. No on-going mentions in the press. No publication in the.+ press of carly in her 'I love VMS' t-shirt.b  D Cisco is to be forgiven. I'm sure that they can put the trademark to better use.g   ------------------------------  $ Date: Thu, 3 Apr 2003 11:16:44 -0500; From: "Brian Tillman" <Brian.Tillman@notnoone.notnohow.com> 0 Subject: Re: OpenVMS and Security - REALLY !!!!!$ Message-ID: <3e8c5e69$1@news.si.com>   >This would be a twofer: >o >http://www.alphavax.com/z  8 That page comes up a blank for me (javascript disabled). -- sI Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot comh5 Smiths Aerospace  Addresses modified to prevent SPAM.m@ 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991a8        This opinion doesn't represent that of my company   ------------------------------   Date: 3 Apr 2003 08:43:49 -0800-( From: bob@instantwhip.com (Bob Ceculski)0 Subject: Re: OpenVMS and Security - REALLY !!!!!< Message-ID: <d7791aa1.0304030843.92b0711@posting.google.com>  ^ david20@alpha1.mdx.ac.uk (David Webb) wrote in message news:<b6h115$9ru$1@aquila.mdx.ac.uk>...s > In article <857e9e41.0304021807.2e9d3bb2@posting.google.com>, susan_skonetski@hotmail.com (Sue Skonetski) writes: Q > >Just so the newsgroup is aware, HP Law has been informed.  We are trademarked.  > I > Personally I think CISCO's use of the term VMS for their VPN ManagementoN > Solution is the bigger problem. We have IBM here redesigning our network andP > they will be putting in CISCO VPNs. It caused me a minutes confusion when they/ > said they would be putting in a 'VMS' system.k >  > David Webb > VMS and Unix team leader > CCSS > Middlesex University n  & sounds like good advertising to me ...   ------------------------------   Date: 3 Apr 2003 12:42:06 -0600l; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)b0 Subject: Re: OpenVMS and Security - REALLY !!!!!3 Message-ID: <J52WgNXFUp1l@eisner.encompasserve.org>h  q In article <857e9e41.0304021807.2e9d3bb2@posting.google.com>, susan_skonetski@hotmail.com (Sue Skonetski) writes:eP > Just so the newsgroup is aware, HP Law has been informed.  We are trademarked. >  > Warm Regards,i > Suen >   G    Thanks.  It's always good to see some non-Palmer-like behaviour from     the owners of VMS.1   ------------------------------  # Date: Thu, 03 Apr 2003 15:23:36 GMT < From: "Brad McCusker" <Brad.remove.McCuskerNOSP@MCompaq.com>P Subject: Re: OpenVMS v7.3-1 and Advanced Server v7.3A - Pathworks server crashes0 Message-ID: <YlYia.219$xH3.189@news.cpqcorp.net>  E With any Advanced Server crash, you really should go straight to yourl Customer Support Center.  E There is an Advanced Server V7.3A-ECO1 available, please get that andiJ install it as soon as possible.  But, it probably won't help you with this crash.  K Look in the PWRK$LMLOGS:PWRK$LMMCP*.LOG file that was active at the time of L the crash.  Do you see a SYSTEM-F-ACCVIO, followed by a traceback indicatingG the cash occurred in a module called VCDROPPED?  If so, that is a knowne? crash. It is caused by services other than Workstation("00") ormL Server("20") trying to connect (e.g. Messenger Service "03"). Not that thereI is probably anything you can do about it, unless you know of which client E might be doing that and making it stop.   A PWRK$LMMCP.EXE with identt* V7.3-110A-98539 will need to be installed.  F If its not in VCDROPPED, then, its something new, and, we will need to analyze the dump.i   --
 Brad McCuskerp OpenVMS Engineeringe Hewlett-Packard Companym Nashua, NH USAC "Andrew Rycroft" <andrew.rycroft@intrinsitech.com> wrote in messaged7 news:58ba0101.0303280653.6763e29e@posting.google.com..., > Hi,l > H > We have just upgraded to openVMS v7.3-1, and Pathworks Advnaced Server v7.3A. >eJ > The Pathworks Server crashes without any apparent reason every two days.; > No entry in the OPA0 nor in any Pathworks event log file.sF > There are two dumpfiles in PWRK$LMROOT:[LANMAN.LOGS]PWRK$LMMCP.DMP;2 >3) > Has anybody had any similar behaviour ?s >e > Thanks > Andrew   ------------------------------  % Date: Thu, 03 Apr 2003 11:56:24 +0200s6 From: Serge ZANGHERI <szangheri@grenoble.sema.slb.com>' Subject: Re: pathworks administer issue05 Message-ID: <3E8C0548.97DA9CBE@grenoble.sema.slb.com>u   Hi,R8 thanx again to all of you for answers or preconisations.. Here is the result of new, new investigations.  = I don't have errors about my problem in logs or event viewer.9 I'm using decnet (not TCP/IP).( My admin/config is correctly configured.C I reinstalled, and the sysgen checking phase is OK. No parameter toe change.    I resume my problem :IA The ONLY issue is to start admin. The admin command says : cannotMA continue, server is not running. (admin /config is running well).o  D I remind that I can connect drives with this VS. So no problem about licence.H But connection only on sharable directory defined  by default I presume.C I mean I can create drive on my NT on netlogon, pwutil and rplfilese directories and it works.oJ Where are defined these defaults directories (in which file), may be I can2 add others directory in file instead to use admin.  @ One way is scsnode is not equal to the name of the station .....   Thanx  for any information Serge      "serge.zangheri" a crit :   > Hi,?, > I installed Pathworks for the second time.5 > But when I start admin I got the following answer :h- > Cannot continue - the server is not runnings >aH > I can connect pwrk$root:netlogon etc, on my pc, I can do admin/config. > No problem about licence.u > swapfile put to 150000 > pagefile put to 300000
 > 80Mb RAM% > other system paramters semmes good.a >C > Someone could help ? > Thanks > Sergeg   ------------------------------  % Date: Thu, 03 Apr 2003 14:16:12 +0200o6 From: Serge ZANGHERI <szangheri@grenoble.sema.slb.com>' Subject: Re: pathworks administer issuea5 Message-ID: <3E8C260C.A87BBBCB@grenoble.sema.slb.com>   G Actually, I can start administer, I put username and password, and just"C after click on the OK button, there is a long blank and the "cannot00 continue : server is not running" error appears. Thanx. Serge    "serge.zangheri" a crit :   > Hi,<, > I installed Pathworks for the second time.5 > But when I start admin I got the following answer :a- > Cannot continue - the server is not runningc >nH > I can connect pwrk$root:netlogon etc, on my pc, I can do admin/config. > No problem about licence.o > swapfile put to 150000 > pagefile put to 300000
 > 80Mb RAM% > other system paramters semmes good.. >  > Someone could help ? > Thanks > Serge    ------------------------------  # Date: Thu, 03 Apr 2003 14:56:09 GMTr< From: "Brad McCusker" <Brad.remove.McCuskerNOSP@MCompaq.com>' Subject: Re: pathworks administer issuei0 Message-ID: <dYXia.216$Um3.205@news.cpqcorp.net>  C "Serge ZANGHERI" <szangheri@grenoble.sema.slb.com> wrote in messageh/ news:3E8C260C.A87BBBCB@grenoble.sema.slb.com... I > Actually, I can start administer, I put username and password, and justrE > after click on the OK button, there is a long blank and the "cannott2 > continue : server is not running" error appears.  # OK, now we are getting somewhere...o  J I think you might have a domain configuration problem.  How is this serverB configured with respect to domains?  Are you trying to be a domainD controller?  Are you trying to participate in a domain?  Standalone?  K Is NETLOGON running (maybe try the NET SERVICE command - this is V5, been an# long time since I had to use that.).  A Can you use the NET commands to login and administer your server?    --
 Brad McCuskerl OpenVMS Engineeringn C RTL Project Leader Hewlett-Packard CompanyD Nashua, NH USA   ------------------------------  $ Date: Thu, 3 Apr 2003 11:21:54 -0500; From: "Brian Tillman" <Brian.Tillman@notnoone.notnohow.com>i6 Subject: Re: Samba on VMS Get_Hostbyname error message$ Message-ID: <3e8c5f9f$1@news.si.com>  3 >We are tyring to run samba version 1.9.7v4 on VMS.$  K Do yourself a favor and get at least V2.0.6.  2.2.8 is the current version.$ -- )I Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot comw5 Smiths Aerospace  Addresses modified to prevent SPAM. @ 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991 8        This opinion doesn't represent that of my company   ------------------------------  + Date: Thu, 03 Apr 2003 09:44:20 +0100 (MET)i9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>d Subject: SMTP authentification; Message-ID: <01KUAA9D1ZK0AH2P35@sysdev.deutsche-boerse.com>0  G I have come upon an obstacle in moving my hobbyist system from ISDN to  G DSL.  I used to have everything---"physical" connection to the ISP via .C ISDN, "logical" internet connection via the ISP, IP addresses, DNS gD services, domain registration---at a one-stop shop, namely my ISP.  H However, I'm forced to drop him because the costs are too high compared E to DSL.  With DSL, the "physical" connection to the ISP, the logical  F internet connection, DNS services and domain registration are usually < separate (though in my case the last two might be combined).  G The disadvantage of this move is that up until now I have been using a   mail gateway at my old ISP.L  D My new ISP requires either one of "his" email addresses in the From:F header, or SMTP authentification.  I had hoped that access to the mailH gateway would be IP-based (this has been my experience in the past), butG alas that is not the case.  It's not a valid option for me to be forcedmB to use a certain style of email address in the From: header.  I'veE registered a couple of domains with the parent company of the new ISP:E and hope that I could at least use these in the From: header (perhaps H the parent company can offer me a mail gateway---I'm still waiting for aG reply for this). I've registered another domain with the parent company G of a company I am also using for dynamic DNS services, so perhaps again/E I could use a mail router there, at least for that domain; again, I'm  still waiting for a reply. e  : Of course, I would like to use one gateway for everything.  A One option would be SMTP authentification.  Can that be done withgH out-of-the-box DEC/Compaq/HP TCP/IP Services?  Can it be done with some H additional software (MX?)?  Can it be done with a different TCPIP stack?  D Another option would be to not use a mail gateway at all, but try toH deliver the mail directly myself.  Would this be an option?  The dynamicE IP addresses I use do translate to something, but there is no obviouss> connection to the From: header.  (There have been threads hereH discussing rejecting email on the basis of reverse DNS lookups, though IE don't recall if the criterion for acceptance was that it translate todC something at all or that it translate to something connected to theeI From: header.  On the other hand, the mail gateway of my old ISP doesn't sG have a reverse DNS entry, and I've never had any problems sending mail  A through that.)  Another worry would be that certain ranges of IP aH addresses are known to be used for "dial-up" access and are blocked for E that reason by some folks.  If I go this route, is it enough to just p9 undefine the alternate gateway in the SMTP configuration?L  ? I don't want to use some open relay somewhere, since many folks B (understandably) black-list those as sources of spam.  However, ifD someone has a gateway I can use, with verification via stealth (i.e.G no-one else knows that it exists; probably not practical), addresses inpF the From: header (I could send you a list of valid ones) or IP addressG (though this would change on average every 20 hours or so), please let nI me know.  I'd be willing to pay a reasonable fee for such a service if I s7 can't solve my problem more easily through other means.o  F What about the Reply-To: field?  Perhaps the ISP would be happy if theD From: header was in his style, but the Reply-To: field contained theG address I want people to reply to.  (I'm assuming that all mail clients D honour this these days.)  How can one set this with VMS MAIL?  Years@ ago, I seem to recall that /SUBSTITUTE_DOMAIN with the NOHIDDEN I attribute would put the substitute domain in the Reply-To: field and the pE real address in the From: field (and HIDDEN would put the substitute  G domain in the From: field and not set the Reply-To: field), but IIRC I -F couldn't reproduce this later and, since /SUBSTITUTE_DOMAIN is mainly H intended for something else, this might have been just a can-go-away-at- any-time side effect.8   ------------------------------  * Date: Thu, 3 Apr 2003 10:18:55 +0000 (UTC)+ From: david20@alpha1.mdx.ac.uk (David Webb)2" Subject: Re: SMTP authentification+ Message-ID: <b6h1qf$9ru$2@aquila.mdx.ac.uk>   w In article <01KUAA9D1ZK0AH2P35@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:l; >Of course, I would like to use one gateway for everything.h >wB >One option would be SMTP authentification.  Can that be done withI >out-of-the-box DEC/Compaq/HP TCP/IP Services?  Can it be done with some eI >additional software (MX?)?  Can it be done with a different TCPIP stack?  > E >Another option would be to not use a mail gateway at all, but try to_I >deliver the mail directly myself.  Would this be an option?  The dynamicnF >IP addresses I use do translate to something, but there is no obvious" >connection to the From: header.    M Many ISPs force users to send through their mail gateway by registering their   dial-up address pool in the DUL.J If they have done this then any system using the DUL list will reject mail% coming directly from those addresses.m  N You can check whether your ISP has added it's addresses to the DUL by going to  $ http://mail-abuse.org/cgi-bin/lookup  M and entering in the public address your connection has been given by the ISP.   r  
 David Webb VMS and Unix team leader CCSS Middlesex University     >(There have been threads hereI >discussing rejecting email on the basis of reverse DNS lookups, though InF >don't recall if the criterion for acceptance was that it translate toD >something at all or that it translate to something connected to theJ >From: header.  On the other hand, the mail gateway of my old ISP doesn't H >have a reverse DNS entry, and I've never had any problems sending mail B >through that.)  Another worry would be that certain ranges of IP I >addresses are known to be used for "dial-up" access and are blocked for hF >that reason by some folks.  If I go this route, is it enough to just : >undefine the alternate gateway in the SMTP configuration? > @ >I don't want to use some open relay somewhere, since many folksC >(understandably) black-list those as sources of spam.  However, ifhE >someone has a gateway I can use, with verification via stealth (i.e.sH >no-one else knows that it exists; probably not practical), addresses inG >the From: header (I could send you a list of valid ones) or IP address H >(though this would change on average every 20 hours or so), please let J >me know.  I'd be willing to pay a reasonable fee for such a service if I 8 >can't solve my problem more easily through other means. >aG >What about the Reply-To: field?  Perhaps the ISP would be happy if thenE >From: header was in his style, but the Reply-To: field contained thesH >address I want people to reply to.  (I'm assuming that all mail clientsE >honour this these days.)  How can one set this with VMS MAIL?  YearsaA >ago, I seem to recall that /SUBSTITUTE_DOMAIN with the NOHIDDEN  J >attribute would put the substitute domain in the Reply-To: field and the F >real address in the From: field (and HIDDEN would put the substitute H >domain in the From: field and not set the Reply-To: field), but IIRC I G >couldn't reproduce this later and, since /SUBSTITUTE_DOMAIN is mainly dI >intended for something else, this might have been just a can-go-away-at-e >any-time side effect.   ------------------------------   Date: 3 Apr 2003 09:16:52 GMT-3 From: gartmann@immunbio.mpg.de (Christoph Gartmann) " Subject: Re: SMTP authentification0 Message-ID: <b6gu64$t8m$1@n.ruf.uni-freiburg.de>  w In article <01KUAA9D1ZK0AH2P35@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:rH >I have come upon an obstacle in moving my hobbyist system from ISDN to H >DSL.  I used to have everything---"physical" connection to the ISP via D >ISDN, "logical" internet connection via the ISP, IP addresses, DNS E >services, domain registration---at a one-stop shop, namely my ISP.  -I >However, I'm forced to drop him because the costs are too high compared pF >to DSL.  With DSL, the "physical" connection to the ISP, the logical G >internet connection, DNS services and domain registration are usually 6= >separate (though in my case the last two might be combined).: > H >The disadvantage of this move is that up until now I have been using a  >mail gateway at my old ISP. >nE >My new ISP requires either one of "his" email addresses in the From:t" >header, or SMTP authentification.  3 [stuff about From:-header and Reply-To: deleted...]l  K SMTP authentication has nothing to do with the header of an e-mail. SMTP asgO such has no authentication, authentication is an extension to the standard SMTPoO protocol. Thus, you need a TCP/IP stack that supports this extension. Accordinge: to the manual here Multinet doesn't support it, PMDF does.  E >Another option would be to not use a mail gateway at all, but try tou< >deliver the mail directly myself.  Would this be an option?  O This requires a static IP address. The IP address doesn't need a DNS entry. ButsM it is a typical feature of spam to originate from a SMTP server without a DNS N entry. Thus, I would strongly recommend to have a DNS entry for the IP address@ used by the mail gateway. So I think you have three options hereD a) get a static IP address for your gateway and run your own gatewayH b) use a product that supports SMTP authentication in order to send your	    e-mailwK c) find some mail-gateway that will releay the mail for you and doesn't userI    smtp authentication. Usually these relays require a static IP address.i   Regards,    Christoph Gartmannt  H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, Germany                                           |H +------------- http://www.immunbio.mpg.de/home/menue.html -------------+   ------------------------------  + Date: Thu, 03 Apr 2003 13:08:04 +0100 (MET)a9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> " Subject: Re: SMTP authentification; Message-ID: <01KUAIDYAFIGAH2P35@sysdev.deutsche-boerse.com>-  H > > My new ISP requires either one of "his" email addresses in the From:% > > header, or SMTP authentification.M > F > SMTP authentication has nothing to do with the header of an e-mail.   A Right.  As I said, the ISP (1&1, with whom I am otherwise happy) eF requires EITHER a "standard" email address in the From: header OR the ' use of SMTP authentification (or both).l  G > SMTP as such has no authentication, authentication is an extension totI > the standard SMTP protocol. Thus, you need a TCP/IP stack that supportss > this extension.    Right.  G > According to the manual here Multinet doesn't support it, PMDF does.    I For someone familiar with the basics of configuring DEC/Compaq/HP TCP/IP  I Services, how long would it take, all told, to move to PMDF?  Is there a W0 hobbyist (i.e. free or cheap) version available?  F > Another option would be to not use a mail gateway at all, but try to= > deliver the mail directly myself.  Would this be an option?t  % > This requires a static IP address. e   Why?  I > The IP address doesn't need a DNS entry. But it is a typical feature ofdI > spam to originate from a SMTP server without a DNS entry. Thus, I would.G > strongly recommend to have a DNS entry for the IP address used by ther > mail gateway.   @ The dynamic addresses I have DO translate to some name (such as H p508181F4.dip0.t-ipconnect.de).  Based on David's post, however, I have I discovered that addresses from my ISP's pool have been registered on the oF "black-sheep" list.  I don't know how many folks reject mail based on 
 this, though.d  ( > So I think you have three options here > F > a) get a static IP address for your gateway and run your own gateway  G Any idea how to do this at a reasonable cost.  I DO have my own static  F addresses now (but I use my old ISP's mail gateway).  One reason I am / moving to DSL and dynamic IPs is to save money.E  J > b) use a product that supports SMTP authentication in order to send your >    e-mail   ? If I want to continue to use VMS MAIL, is PMDF the only option?   M > c) find some mail-gateway that will releay the mail for you and doesn't use K >    smtp authentication. Usually these relays require a static IP address.,  B Probably both the static address and finding a gateway would be a  problem.   ------------------------------   Date: 3 Apr 2003 12:12:16 GMT,3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)d" Subject: Re: SMTP authentification0 Message-ID: <b6h8f0$4a4$1@n.ruf.uni-freiburg.de>  w In article <01KUAIDYAFIGAH2P35@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:,J >For someone familiar with the basics of configuring DEC/Compaq/HP TCP/IP = >Services, how long would it take, all told, to move to PMDF?n  N PMDF sits on top of it. In principle you don't need to know anything about theK underlying TCP/IP stack to configure PMDF. How much time you'll need to getnJ familiar with PMDF depends on the features you intend to use. Just to have9 SMTP authentication it should be done within a few hours.f  = >Is there a  hobbyist (i.e. free or cheap) version available?l  7 Don't know, have a look at Process Software's homepage.t  G >> Another option would be to not use a mail gateway at all, but try toi> >> deliver the mail directly myself.  Would this be an option? >t& >> This requires a static IP address.  >k >Why?r  M Assume an address "myself@my.home". In order for someone to send an e-mail toiL this address, someone's computer has to ask a namserver for a mail-exchangerN (MX) record for "my.home". The answer is an IP-address. Now someone's computerK will establish a direct connection to this address and deliver the mail viaeI SMTP. It is quite obvious that the IP address must not change very often.eK Changing the IP address requires updating of the primary nameserver and all F caching nameserver. Usually it takes a few days until everybody on the3 Internet receives the new translation of "my.home".   ) >> So I think you have three options hereo >> 5G >> a) get a static IP address for your gateway and run your own gateway  >uH >Any idea how to do this at a reasonable cost.  I DO have my own static G >addresses now (but I use my old ISP's mail gateway).  One reason I am -0 >moving to DSL and dynamic IPs is to save money.  K Ask your provider. It will be a bit more expensive but most providers offerc
 this feature.s  K >> b) use a product that supports SMTP authentication in order to send yourD >>    e-mail >1@ >If I want to continue to use VMS MAIL, is PMDF the only option?  L You could write your own smtp server for outgoing mail. The protocol is pureM text. You'll need to write a program what is able to establish some form of arH telnet link to port 25 on the remote system. Alternatively you could useL C-Kermit and its scripting language for the task, perhaps combined with someL DCL. Recently I did something similar to interact with a webserver, completeO some form there, retrieve a document and repeat this in a loop for the other 99a documents.    N >> c) find some mail-gateway that will releay the mail for you and doesn't useL >>    smtp authentication. Usually these relays require a static IP address. >tC >Probably both the static address and finding a gateway would be a w	 >problem.e  K The gateway isn't a big deal. Every SMTP-server is able to perform the tasktK (SMTP relaying). A properly managed SMTP-server has the feature disabled or M restricted. It is a task of a few minutes tp enable a certain domain-name forgO relaying. But of course I would this only for a specific IP address or at leasti% for a limited range of IP addresses. o   Regards,    Christoph Gartmann   H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, Germany                                           |H +------------- http://www.immunbio.mpg.de/home/menue.html -------------+   ------------------------------  + Date: Thu, 03 Apr 2003 15:29:09 +0100 (MET)a9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> " Subject: Re: SMTP authentification; Message-ID: <01KUAM5JGYJYAH2P35@sysdev.deutsche-boerse.com>i  F > PMDF sits on top of it. In principle you don't need to know anythingD > about the underlying TCP/IP stack to configure PMDF. How much timeJ > you'll need to get familiar with PMDF depends on the features you intendI > to use. Just to have SMTP authentication it should be done within a fewi	 > hours. t  ? Right.  Process sells so many things---Multinet, TCPware, PMDF,_E SSH---some of which are contained in others (or at least some of the VF functionality is contained within others) that it gets confusing.  :-)  E TCPware and Multinet are stacks, PMDF runs on top of any stack.  How  % does the MX package figure into this?d  ' > > This requires a static IP address.   >  > Why?  E > Assume an address "myself@my.home". In order for someone to send ansI > e-mail to this address, someone's computer has to ask a namserver for a , > mail-exchanger (MX) record for "my.home".   D Actually, only if my.home itself doesn't translate to an IP address.  F > The answer is an IP-address. Now someone's computer will establish aH > direct connection to this address and deliver the mail via SMTP. It isH > quite obvious that the IP address must not change very often. ChangingD > the IP address requires updating of the primary nameserver and allH > caching nameserver. Usually it takes a few days until everybody on the6 > Internet receives the new translation of "my.home".   F Right, but these days with services like dyndns.org, dynaccess.de etc G one can update the DNS and MX records within seconds.  That's the sort j of setup I'm thinking about.  G > Ask your provider. It will be a bit more expensive but most providersf > offer this feature.   E As far as I know, 1&1 doesn't offer this.  I can check, but of course G this only makes sense if they are less restrictive if I have a fixed IPpB address.  However, if I only need a fixed IP address for technicalE reasons, i.e. due to keeping MX and DNS records current, then this is>G not an issue.  I don't think 1&1 would be less restrictive if I have a rC fixed IP address, since they know now already when I am using what S address.  L > > b) use a product that supports SMTP authentication in order to send your
 > >    e-mail  >pA > If I want to continue to use VMS MAIL, is PMDF the only option?w  I > You could write your own smtp server for outgoing mail. The protocol iswJ > pure text. You'll need to write a program what is able to establish someJ > form of a telnet link to port 25 on the remote system. Alternatively youE > could use C-Kermit and its scripting language for the task, perhapsnF > combined with some DCL. Recently I did something similar to interactE > with a webserver, complete some form there, retrieve a document andr4 > repeat this in a loop for the other 99 documents.   $ That seems like quite a bit of work!  O > > c) find some mail-gateway that will releay the mail for you and doesn't use.M > >    smtp authentication. Usually these relays require a static IP address.t > D > Probably both the static address and finding a gateway would be a 
 > problem.  H > The gateway isn't a big deal. Every SMTP-server is able to perform theF > task (SMTP relaying). A properly managed SMTP-server has the featureC > disabled or restricted. It is a task of a few minutes tp enable a I > certain domain-name for relaying. But of course I would this only for asG > specific IP address or at least for a limited range of IP addresses. l  D Right.  Technically, it's not a problem.  What I need is a friendly F gatekeeper who will let me use his gateway based on IP address (which H could change once a day), domain names, or some combination.  Actually, H all email I would send would contain a domain name which resolves to an C IP address (many will resolve to the same IP address), though this d% address will change once a day or so.    ------------------------------  * Date: Thu, 3 Apr 2003 14:07:39 +0000 (UTC)+ From: david20@alpha1.mdx.ac.uk (David Webb)m" Subject: Re: SMTP authentification+ Message-ID: <b6hf7b$bea$1@aquila.mdx.ac.uk>L  w In article <01KUAIDYAFIGAH2P35@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:aI >> > My new ISP requires either one of "his" email addresses in the From: & >> > header, or SMTP authentification. >> eG >> SMTP authentication has nothing to do with the header of an e-mail. M >gB >Right.  As I said, the ISP (1&1, with whom I am otherwise happy) G >requires EITHER a "standard" email address in the From: header OR the t( >use of SMTP authentification (or both). >tH >> SMTP as such has no authentication, authentication is an extension toJ >> the standard SMTP protocol. Thus, you need a TCP/IP stack that supports >> this extension. r >  >Right.e >wH >> According to the manual here Multinet doesn't support it, PMDF does.  >>  L PMDF is mailhub (and some mail client) software not a separate TCP/IP stack.M It can work with all the main VMS stacks (DEC TCP/IP services, Older versionsS of UCX, Multinet, TCPWARE).e  L Although the mail server (PMDF MTA) supports SASL and SMTP AUTH I don't know whether the PMDF Mail clients   A PMDF MAIL - (an extended but compatible replacement for VMS MAIL)F andn. PMDF PINE - (a supported port of PINE for VMS)   support SMTP AUTH for sending.M (You would normally set things up so that these were internal clients sendingyF to the PMDF MTA - hence they wouldn't normally need to use SMTP AUTH).  O I've cross posted this to the vmsnet.mail.pmdf newgroup to get a fuller answer.i    
 David Webb VMS and Unix team leader CCSS Middlesex University   PS.-I Unfortunately due to the fact that Process develop and support PMDF underkN license from SUN - since SUN bought the original owners Innosoft - there is as. far as I am aware no Hobbyist version of PMDF.         J >For someone familiar with the basics of configuring DEC/Compaq/HP TCP/IP J >Services, how long would it take, all told, to move to PMDF?  Is there a 1 >hobbyist (i.e. free or cheap) version available?e >mG >> Another option would be to not use a mail gateway at all, but try tot> >> deliver the mail directly myself.  Would this be an option? >i& >> This requires a static IP address.  >i >Why?  > J >> The IP address doesn't need a DNS entry. But it is a typical feature ofJ >> spam to originate from a SMTP server without a DNS entry. Thus, I wouldH >> strongly recommend to have a DNS entry for the IP address used by the >> mail gateway. b ><A >The dynamic addresses I have DO translate to some name (such as aI >p508181F4.dip0.t-ipconnect.de).  Based on David's post, however, I have oJ >discovered that addresses from my ISP's pool have been registered on the G >"black-sheep" list.  I don't know how many folks reject mail based on   >this, though. >C) >> So I think you have three options herea >> eG >> a) get a static IP address for your gateway and run your own gatewaye >rH >Any idea how to do this at a reasonable cost.  I DO have my own static G >addresses now (but I use my old ISP's mail gateway).  One reason I am a0 >moving to DSL and dynamic IPs is to save money. >iK >> b) use a product that supports SMTP authentication in order to send your  >>    e-mail >n@ >If I want to continue to use VMS MAIL, is PMDF the only option? >nN >> c) find some mail-gateway that will releay the mail for you and doesn't useL >>    smtp authentication. Usually these relays require a static IP address. >MC >Probably both the static address and finding a gateway would be a h	 >problem.    ------------------------------   Date: 3 Apr 2003 14:44:30 GMTA3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)T" Subject: Re: SMTP authentification0 Message-ID: <b6hhce$74f$1@n.ruf.uni-freiburg.de>  w In article <01KUAM5JGYJYAH2P35@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:f@ >Right.  Process sells so many things---Multinet, TCPware, PMDF,F >SSH---some of which are contained in others (or at least some of the G >functionality is contained within others) that it gets confusing.  :-)h >"F >TCPware and Multinet are stacks, PMDF runs on top of any stack.  How & >does the MX package figure into this?  J It is similar to PMDF, it sits on top of VMS-Mail and/or the TCP/IP stack.  ( >> > This requires a static IP address.  >> >> Why?r > F >> Assume an address "myself@my.home". In order for someone to send anJ >> e-mail to this address, someone's computer has to ask a namserver for a- >> mail-exchanger (MX) record for "my.home". e >0E >Actually, only if my.home itself doesn't translate to an IP address.e  M The question, whether my.home translates to an IP address, is the question to.O the DNS! The only case where a computer doesn't need to contact the DNS is whentH it has the address in its cache because it has been used shortly before.  G >Right, but these days with services like dyndns.org, dynaccess.de etc  H >one can update the DNS and MX records within seconds.  That's the sort  >of setup I'm thinking about.M  O This is slightly different from what I thought. If you run your own SMTP-serveriO and this one has a volatile IP-address you'll have problems. "dynaccess.de" can0N avoid these problems in a way that THEY provide the SMTP server to the outsideI world. The principle is still the same, there must a server with a staticaI address for a give domain. But their approach could be a solution to your>L problem. You run your smtp-server, tell their smtp-server your e-mail domainO and your current ip-address. As soon as your address changes you tell them your 
 new address.    H >> Ask your provider. It will be a bit more expensive but most providers >> offer this feature. - >-F >As far as I know, 1&1 doesn't offer this.  I can check, but of courseH >this only makes sense if they are less restrictive if I have a fixed IP	 >address.   M Once you have the fixed address you don't need their smtp-server anymore. YounO may run your own one. Their restrictions won't bother you anymore. I must admit K though that they need to do a little configuration in their DNS to point toa your domain and static address.n  J >> You could write your own smtp server for outgoing mail. The protocol isK >> pure text. You'll need to write a program what is able to establish someeK >> form of a telnet link to port 25 on the remote system. Alternatively youaF >> could use C-Kermit and its scripting language for the task, perhapsG >> combined with some DCL. Recently I did something similar to interact F >> with a webserver, complete some form there, retrieve a document and5 >> repeat this in a loop for the other 99 documents. y >k% >That seems like quite a bit of work!a  M If it is only for sending mail it is not that much. I would expect one or twoo days.t  E >Right.  Technically, it's not a problem.  What I need is a friendly nG >gatekeeper who will let me use his gateway based on IP address (which eI >could change once a day), domain names, or some combination.  Actually, tI >all email I would send would contain a domain name which resolves to an .D >IP address (many will resolve to the same IP address), though this & >address will change once a day or so.   This should be possible.   Regards,    Christoph Gartmanne  H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, Germany                                           |H +------------- http://www.immunbio.mpg.de/home/menue.html -------------+   ------------------------------  + Date: Thu, 03 Apr 2003 17:02:36 +0100 (MET)h9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> " Subject: Re: SMTP authentification; Message-ID: <01KUAQ9UZBJEAH2P35@sysdev.deutsche-boerse.com>r  I > > TCPware and Multinet are stacks, PMDF runs on top of any stack.  How r) > > does the MX package figure into this?  > L > It is similar to PMDF, it sits on top of VMS-Mail and/or the TCP/IP stack.  C OK.  Does anyone know if MX supports SMTP authentification from VMSa MAIL?   + > > > > This requires a static IP address. t > > >c
 > > > Why? > > I > > > Assume an address "myself@my.home". In order for someone to send aneM > > > e-mail to this address, someone's computer has to ask a namserver for ao0 > > > mail-exchanger (MX) record for "my.home".  > > H > > Actually, only if my.home itself doesn't translate to an IP address. > C > The question, whether my.home translates to an IP address, is thenE > question to the DNS! The only case where a computer doesn't need to?H > contact the DNS is when it has the address in its cache because it has > been used shortly before.   E Right.  But if I send an email to user@machine.subdomain.domain, and rG machine.subdomain.domain translates to an IP address, then no explicit cA MX record is needed.  Of course, normally explicit machine names RF shouldn't be in email addresses, and thus normally there should be MX F records to allow generic email addresses.  (With dynamic IP addresses A being IP addresses of NAT/PAT DSL routers which forward incoming mD connections to an internal IP address (which can be a TCPIP cluster E alias), then this is less important---the stuff after the "@" in the tH email address could translate to the WAN address of the router, and the I router would take care of sending that on to a specific machine (perhaps r via a TCPIP cluster alias).)  G >Right, but these days with services like dyndns.org, dynaccess.de etc oH >one can update the DNS and MX records within seconds.  That's the sort  >of setup I'm thinking about.u  E > This is slightly different from what I thought. If you run your ownmJ > SMTP-server and this one has a volatile IP-address you'll have problems.H > "dynaccess.de" can avoid these problems in a way that THEY provide the$ > SMTP server to the outside world.   G Do you have experience with them?  I have an account there and plan to M host a domain with them.  D > The principle is still the same, there must a server with a staticG > address for a given domain. But their approach could be a solution toHE > your problem. You run your smtp-server, tell their smtp-server your.D > e-mail domain and your current ip-address. As soon as your address* > changes you tell them your new address.   G They certainly do this for other DNS records (including CNAME, MX etc.)t@ I'm hoping they will offer a similar service for a mail gateway.A Dyndns.org, which offers a similar service, does NOT offer a mail F gateway.  I'm waiting for an answer from dynaccess.de.  (I've had goodF experience with both of these organisations so far.  Dynaccess.de is aH bit more expensive, but offers a bit more.  Also, since .de (and .nl andC dk) domains require the DNS server to be in the country, dyndns.org 9 cannot host such domains with their Custom DNS service.) e  B > Once you have the fixed address you don't need their smtp-serverH > anymore. You may run your own one. Their restrictions won't bother youJ > anymore. I must admit though that they need to do a little configuration; > in their DNS to point to your domain and static address. _  G I still don't see the TECHNICAL reason why I need a static IP address, yE as long as my dynamic one is updated often enough and quickly enough.eF As far as I can tell, there are only POLITICAL reasons why running an F SMTP server on a dynamic IP is a bad idea---because some mail servers % reject mail from "dial-up" addresses.P  I Of course, if I use my own SMTP server to deliver mail directly, I don't   need them at all.9  H > If it is only for sending mail it is not that much. I would expect one > or two days. -  D Perhaps we can have a contest here as to who can post the best SMTP-? server-with-SMTP-authentification code reasonably quickly.  :-)i  H > > Right.  Technically, it's not a problem.  What I need is a friendly J > > gatekeeper who will let me use his gateway based on IP address (which L > > could change once a day), domain names, or some combination.  Actually, L > > all email I would send would contain a domain name which resolves to an G > > IP address (many will resolve to the same IP address), though this o) > > address will change once a day or so.e >  > This should be possible.   I'm open for offers!   ------------------------------   Date: 3 Apr 2003 15:31:03 GMTg3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)h" Subject: Re: SMTP authentification0 Message-ID: <b6hk3n$95r$1@n.ruf.uni-freiburg.de>  w In article <01KUAQ9UZBJEAH2P35@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes: H >I still don't see the TECHNICAL reason why I need a static IP address, F >as long as my dynamic one is updated often enough and quickly enough.  M You'll need the static IP address if you run a SMTP server to be reachable byaO the outside world. There are technical reasons for that: your e-mail domain hasfK to be registered in some DNS specifying the IP address of your SMTP server.aM Once you change this IP address all other DNSs that cache your domain need toaJ be updated as well. This takes some time in the range of hours, not in the range of seconds.p  G >As far as I can tell, there are only POLITICAL reasons why running an eG >SMTP server on a dynamic IP is a bad idea---because some mail servers a& >reject mail from "dial-up" addresses.  
 See above.  J >Of course, if I use my own SMTP server to deliver mail directly, I don't  >need them at all.  O You'll have technical problems running your own SMTP server without a static IP E address unless you use an additional service like the one provided by 3 dynaccess.de . That's why they provide the service.m  I >> If it is only for sending mail it is not that much. I would expect one  >> or two days.  >nE >Perhaps we can have a contest here as to who can post the best SMTP-i@ >server-with-SMTP-authentification code reasonably quickly.  :-)   What could one win? ;-)a   Regards,    Christoph Gartmannu  H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, Germany                                           |H +------------- http://www.immunbio.mpg.de/home/menue.html -------------+   ------------------------------  # Date: Thu, 03 Apr 2003 18:31:16 GMT % From: Don Sykes <alphase@pacbell.net>M" Subject: Re: SMTP authentification+ Message-ID: <3E8C7DEE.B7D1F14C@pacbell.net>o  G I just went through a similar exercise, moving from one DSL to another.oH After lots of trepidations, everything seems to be working fine now. TheD only distinction I can see is I requested a static IP (over here you= have a choice of dynamic or static for about the same price).d  B For what it's worth, I'm running DEC TCPIP Services 5.1 and, afterE registering my domain name with the DSL's name service, this is all Ie
 had to do:   1. Setup my LAN IPs - result:i TCPIP> sho hostm      LOCAL databaset Host address    Host names& 127.0.0.1         LOCALHOST, localhost@ 67.125.xxx.xxx    MYGATE		(Address of my DSL router/modem combo) 67.125.xxx.xxx    NTONEy 67.125.xxx.xxx    VMS2    + 2. Establish the BIND Name routes - result:  TCPIP> sho namew BIND Resolver Parameters  Local domain: ALPHASE.COM  SystemM   State:     Started, Enabledt     Transport: UDP   Domain:    ALPHASE.COM   Retry:     4   Timeout:   4G   Servers:    206.13.xxx.xxx, 206.13.xxx.xxx	(These were provided by myh DSL ISP)   Path:       No values definedl    & 3. Setup the default gateway - result: TCPIP> sho route/ful/permD&                              PERMANENT< Type           Destination                           GatewayE PN  * 0.0.0.0                               67.125.xxx.xxx     MYGATE   / 4. Setup a substitute domain and zone - result:k TCPIP> sho config smtp SMTP ConfigurationB                                                                    OptionsZ@ Initial interval:   0 00:30:00.00       Address_max:    16       NOEIGHT_BIToF Retry interval:     0 01:00:00.00       Hop_count_max:  16       RELAY@ Maximum interval:   3 00:00:00.00                                TOP_HEADERSh  G Timeout             Initial       Mail    Receipt       Data  TerminateeG   Send:                   5          5          5          3         10k   Receive:                5    Alternate gateway:  not defined, General gateway:    not definedi  ' Substitute domain:  HIDDEN, ALPHASE.COMh Zone:               ALPHASE.COMo	 ---------t  E Now all mail is routed through the default gateway. I did not have to E configure any special mail gateways. I know you were asking how to donG this using *dynamic* IPs, but I thought this might help you, or others,n& who may give up on the dynamic method.   Cheers -- r   Have VMS, Will Travels Wire paladin, San Franciscoo   (paladinATalphaseDOTcom)     -------------------------------m Phillip Helbig wrote:o > H > I have come upon an obstacle in moving my hobbyist system from ISDN toH > DSL.  I used to have everything---"physical" connection to the ISP viaD > ISDN, "logical" internet connection via the ISP, IP addresses, DNSD > services, domain registration---at a one-stop shop, namely my ISP.I > However, I'm forced to drop him because the costs are too high comparedaF > to DSL.  With DSL, the "physical" connection to the ISP, the logicalG > internet connection, DNS services and domain registration are usuallyr> > separate (though in my case the last two might be combined). > H > The disadvantage of this move is that up until now I have been using a > mail gateway at my old ISP.d > F > My new ISP requires either one of "his" email addresses in the From:H > header, or SMTP authentification.  I had hoped that access to the mailJ > gateway would be IP-based (this has been my experience in the past), butI > alas that is not the case.  It's not a valid option for me to be forcedhD > to use a certain style of email address in the From: header.  I'veG > registered a couple of domains with the parent company of the new ISPoG > and hope that I could at least use these in the From: header (perhapshJ > the parent company can offer me a mail gateway---I'm still waiting for aI > reply for this). I've registered another domain with the parent companylI > of a company I am also using for dynamic DNS services, so perhaps againoG > I could use a mail router there, at least for that domain; again, I'ms > still waiting for a reply. > < > Of course, I would like to use one gateway for everything. > C > One option would be SMTP authentification.  Can that be done with I > out-of-the-box DEC/Compaq/HP TCP/IP Services?  Can it be done with some J > additional software (MX?)?  Can it be done with a different TCPIP stack? > F > Another option would be to not use a mail gateway at all, but try toJ > deliver the mail directly myself.  Would this be an option?  The dynamicG > IP addresses I use do translate to something, but there is no obvious @ > connection to the From: header.  (There have been threads hereJ > discussing rejecting email on the basis of reverse DNS lookups, though IG > don't recall if the criterion for acceptance was that it translate totE > something at all or that it translate to something connected to the$J > From: header.  On the other hand, the mail gateway of my old ISP doesn'tH > have a reverse DNS entry, and I've never had any problems sending mailB > through that.)  Another worry would be that certain ranges of IPI > addresses are known to be used for "dial-up" access and are blocked forSF > that reason by some folks.  If I go this route, is it enough to just; > undefine the alternate gateway in the SMTP configuration?m > A > I don't want to use some open relay somewhere, since many folkstD > (understandably) black-list those as sources of spam.  However, ifF > someone has a gateway I can use, with verification via stealth (i.e.I > no-one else knows that it exists; probably not practical), addresses incH > the From: header (I could send you a list of valid ones) or IP addressH > (though this would change on average every 20 hours or so), please letJ > me know.  I'd be willing to pay a reasonable fee for such a service if I9 > can't solve my problem more easily through other means.. > H > What about the Reply-To: field?  Perhaps the ISP would be happy if theF > From: header was in his style, but the Reply-To: field contained theI > address I want people to reply to.  (I'm assuming that all mail clientseF > honour this these days.)  How can one set this with VMS MAIL?  YearsA > ago, I seem to recall that /SUBSTITUTE_DOMAIN with the NOHIDDENlJ > attribute would put the substitute domain in the Reply-To: field and theF > real address in the From: field (and HIDDEN would put the substituteH > domain in the From: field and not set the Reply-To: field), but IIRC IG > couldn't reproduce this later and, since /SUBSTITUTE_DOMAIN is mainly J > intended for something else, this might have been just a can-go-away-at- > any-time side effect.s   ------------------------------   Date: 3 Apr 2003 05:36:33 -0800p( From: bob@instantwhip.com (Bob Ceculski)8 Subject: Re: So much for Opteron 32bit compatibility ...= Message-ID: <d7791aa1.0304030536.52b6b28e@posting.google.com>u  d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<xCSdnd02joTg3hajXTWc3w@metrocast.net>...7 > "Bob Ceculski" <bob@instantwhip.com> wrote in message-9 > news:d7791aa1.0304021136.3bcaf50d@posting.google.com...4: > > this eweek article confirms what I expected, just like7 > > when you run emulation mode on an AS400 w/system3x,6: > > you kill the processor by 50% ... and it looks this is > > also the case with AMD > F > No, Bob:  you're just continuing your near 0.000 batting average forK > competency.  If you actually *read* the article, you'd have seen that the L > 32-bit performance being denigrated is Itanic's, not Opteron's (and if youI > had the slightest clue about Opteron you'd know that its 32-bit SPECintDL > performance at 2 GHz is faster than the fastest current Pentium's, whereasM > according to HP Itanic2's performance on 32-bit code is more like a 300 MHzi > Pentium Pro).M >  > - bill  < if you read the paragraph before, it states itanium does not9 have 32 bit compatibility, therefore suggesting that they = are talking about opteron ... if what you state is true, thenO< the author needs to go back to grammer school, and only some7 one of your grammatical caliber could decipher that ...n    F We are not seeing or hearing anything from our customers and ISVs thatC indicates they want or need Itanium. But we are seeing interest forMA the upcoming Opteron processor family, essentially because it hastB 32-node compatibility, which Itanium doesn't," said John Loiacono,2 vice president of Sun's operating platforms group.   ------------------------------   Date: 3 Apr 2003 08:46:35 -0800 ( From: bob@instantwhip.com (Bob Ceculski)8 Subject: Re: So much for Opteron 32bit compatibility ...= Message-ID: <d7791aa1.0304030846.744caf16@posting.google.com>.  d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<xCSdnd02joTg3hajXTWc3w@metrocast.net>...7 > "Bob Ceculski" <bob@instantwhip.com> wrote in messagel9 > news:d7791aa1.0304021136.3bcaf50d@posting.google.com...r: > > this eweek article confirms what I expected, just like7 > > when you run emulation mode on an AS400 w/system3x,0: > > you kill the processor by 50% ... and it looks this is > > also the case with AMD > F > No, Bob:  you're just continuing your near 0.000 batting average forK > competency.  If you actually *read* the article, you'd have seen that theuL > 32-bit performance being denigrated is Itanic's, not Opteron's (and if youI > had the slightest clue about Opteron you'd know that its 32-bit SPECintDL > performance at 2 GHz is faster than the fastest current Pentium's, whereasM > according to HP Itanic2's performance on 32-bit code is more like a 300 MHza > Pentium Pro).p >  > - bill   I did read it, and8 the article states here that itanium doesn't have 32 bit compatibility ...e  B "We are not seeing or hearing anything from our customers and ISVsD that indicates they want or need Itanium. But we are seeing interestE for the upcoming Opteron processor family, essentially because it hasaB 32-node compatibility, which Itanium doesn't," said John Loiacono,2 vice president of Sun's operating platforms group.   ------------------------------  % Date: Thu, 03 Apr 2003 09:51:41 -0700a+ From: "Barry Treahy, Jr." <Treahy@MMaz.com>i8 Subject: Re: So much for Opteron 32bit compatibility ...' Message-ID: <3E8C669D.6090001@MMaz.com>d   Bob Ceculski wrote:   e >"Bill Todd" <billtodd@metrocast.net> wrote in message news:<xCSdnd02joTg3hajXTWc3w@metrocast.net>...  >    >t7 >>"Bob Ceculski" <bob@instantwhip.com> wrote in messaget9 >>news:d7791aa1.0304021136.3bcaf50d@posting.google.com...n >>     >>9 >>>this eweek article confirms what I expected, just likee6 >>>when you run emulation mode on an AS400 w/system3x,9 >>>you kill the processor by 50% ... and it looks this isn >>>also the case with AMDI	 >>>      h >>>sF >>No, Bob:  you're just continuing your near 0.000 batting average forK >>competency.  If you actually *read* the article, you'd have seen that thelL >>32-bit performance being denigrated is Itanic's, not Opteron's (and if youI >>had the slightest clue about Opteron you'd know that its 32-bit SPECintsL >>performance at 2 GHz is faster than the fastest current Pentium's, whereasM >>according to HP Itanic2's performance on 32-bit code is more like a 300 MHzM >>Pentium Pro).. >> >>- bill >>     >> >d >I did read it, ando9 >the article states here that itanium doesn't have 32 bita >compatibility ... >eC >"We are not seeing or hearing anything from our customers and ISVsoE >that indicates they want or need Itanium. But we are seeing interestrF >for the upcoming Opteron processor family, essentially because it hasC >32-node compatibility, which Itanium doesn't," said John Loiacono, 3 >vice president of Sun's operating platforms group.a >    >tE Are you folks confusing '32-node' with '32-bit' because I do not see eD 32-bit anywhere in your snippet.  It it is my understanding the the B Opteron processor has built onto the chips the ability to support H multi-way processor configurations coming out of the gate.  Is that not  what they are referring to???b   Barryr     -- f  @ Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028t   ------------------------------  $ Date: Thu, 3 Apr 2003 19:50:10 +0200" From: "Hans Vlems" <hvlems@iae.nl>8 Subject: Re: So much for Opteron 32bit compatibility ...5 Message-ID: <b6hstb$5eg5o$1@ID-143435.news.dfncis.de>p  8 "Barry Treahy, Jr." <Treahy@MMaz.com> schreef in bericht! news:3E8C669D.6090001@MMaz.com...o > Bob Ceculski wrote:b >d8 > >"Bill Todd" <billtodd@metrocast.net> wrote in message. news:<xCSdnd02joTg3hajXTWc3w@metrocast.net>... > >i > > 9 > >>"Bob Ceculski" <bob@instantwhip.com> wrote in messagee; > >>news:d7791aa1.0304021136.3bcaf50d@posting.google.com...a > >> > >>; > >>>this eweek article confirms what I expected, just likey8 > >>>when you run emulation mode on an AS400 w/system3x,; > >>>you kill the processor by 50% ... and it looks this is  > >>>also the case with AMD  > >>>M > >>>eH > >>No, Bob:  you're just continuing your near 0.000 batting average forI > >>competency.  If you actually *read* the article, you'd have seen thate theeJ > >>32-bit performance being denigrated is Itanic's, not Opteron's (and if you K > >>had the slightest clue about Opteron you'd know that its 32-bit SPECinteF > >>performance at 2 GHz is faster than the fastest current Pentium's, whereas-K > >>according to HP Itanic2's performance on 32-bit code is more like a 3001 MHz4 > >>Pentium Pro).m > >>
 > >>- bill > >> > >> > >  > >I did read it, and ; > >the article states here that itanium doesn't have 32 biti > >compatibility ... > >9E > >"We are not seeing or hearing anything from our customers and ISVs-G > >that indicates they want or need Itanium. But we are seeing interest-H > >for the upcoming Opteron processor family, essentially because it hasE > >32-node compatibility, which Itanium doesn't," said John Loiacono,:5 > >vice president of Sun's operating platforms group.2 > >y > >dF > Are you folks confusing '32-node' with '32-bit' because I do not seeE > 32-bit anywhere in your snippet.  It it is my understanding the theeC > Opteron processor has built onto the chips the ability to support I > multi-way processor configurations coming out of the gate.  Is that noth > what they are referring to???r >t Yes.   ------------------------------  % Date: Thu, 03 Apr 2003 16:53:09 +0100o From: Atanas <atanas@iec.co.il>o; Subject: Re: Space Invaders was: Re: Fortran Guru requestedt) Message-ID: <3e8c4a99@news.hai.iec.co.il>r   Barry Treahy, Jr. wrote:   >  >> PRSTSC::DTL wrote:  >>G >>> I encounter some trouble porting a fortran/vax program to Alpha (noI; >>> warnings at compile time, either with or without /F77):m >>>n >>> $ run spcinv	 >>> ../..s >>>-? >>> %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtualo >>> address=000000000000* >>> 0030, PC=FFFFFFFF808E5380, PS=0000001B3 >>> %TRACE-F-TRACEBACK, symbolic stack dump follows = >>>   image    module    routine             line      rel PCt
 >>> abs PCB >>>                                             0 0000000000000000 >>> FFFFFFFF808E5380B >>>  SPCINV  INVADERS  SLEEP                  141 0000000000000ADC >>> 0000000000030ADCB >>>  SPCINV  INVADERS  SPECIAL_GRAPHIC        579 0000000000002BB8 >>> 0000000000032BB8B >>>  SPCINV  INVADERS  MOVE                    42 00000000000003D0 >>> 00000000000303D0B >>>  SPCINV  INVADERS  INVADERS                27 0000000000000000 >>> 0000000000000000B >>>                                             0 FFFFFFFF802513D4 >>> FFFFFFFF802513D4 >>>.% >>> 132         SUBROUTINE SLEEP(TAG)sI >>> 133 C       Set timer and wait for TAG seconds.  TAG is a CHARACTER*5 
 >>> stringG >>> 134 C       of the form ##.##, which is the seconds, and hundredthse >>> of seconds.dB >>> 135 C       To delay game for 5 seconds, TAG would be '05.00'.( >>> 136         IMPLICIT INTEGER*4 (A-Z)# >>> 137         EXTERNAL SS$_NORMALo% >>> 138         DOUBLE PRECISION TIME , >>> 139         CHARACTER DELTA_TIME*9,TAG*5 >>> 140 E >>> 141         DELTA_TIME='0 ::'//TAG          !Concatenate strings.pH >>> 142         RET_STAT=SYS$BINTIM(DELTA_TIME,TIME)    !ASCII to binary	 >>> time. D >>> 143         IF(RET_STAT.NE.%LOC(SS$_NORMAL))STOP'BINTIM failed.'I >>> 144         RET_STAT=SYS$SETIMR(,TIME,,)            !Set timer, event  >>> flag 0.uD >>> 145         IF(RET_STAT.NE.%LOC(SS$_NORMAL))STOP'SETIMR failed.'H >>> 146         RET_STAT=SYS$WAITFR(%VAL(0))            !Wait for flag 0 >>> to be set.D >>> 147         IF(RET_STAT.NE.%LOC(SS$_NORMAL))STOP'WAITFR failed.' >>> 148         RETURN >>> 149         ENDa >>> & > Try to recompile with /NOOPTIMIZE... >  > Barrya > J May be the problem is in other place - I have buid the program on VAX and  Alpha and it works on both:        PROGRAM TEST       CHARACTER TAG*5e       TAG='02.00'o       CALL SLEEP(TAG) 	       ENDh         SUBROUTINE SLEEP(TAG)sF C     Set timer and wait for TAG seconds.  TAG is a CHARACTER*5 stringI C     of the form ##.##, which is the seconds, and hundredths of seconds.e8 C     To delay game for 5 seconds, TAG would be '05.00'.       IMPLICIT INTEGER*4 (A-Z)       EXTERNAL SS$_NORMAL        DOUBLE PRECISION TIMEi"       CHARACTER DELTA_TIME*9,TAG*5       5       DELTA_TIME='0 ::'//TAG    !Concatenate strings.hA       RET_STAT=SYS$BINTIM(DELTA_TIME,TIME) !ASCII to binary time.g:       IF(RET_STAT.NE.%LOC(SS$_NORMAL))STOP'BINTIM failed.'=       RET_STAT=SYS$SETIMR(,TIME,,) !Set timer, event  flag 0.e:       IF(RET_STAT.NE.%LOC(SS$_NORMAL))STOP'SETIMR failed.'>       RET_STAT=SYS$WAITFR(%VAL(0)) !Wait for flag 0 to be set.:       IF(RET_STAT.NE.%LOC(SS$_NORMAL))STOP'WAITFR failed.'       RETURN	       ENDl   ------------------------------  % Date: Thu, 03 Apr 2003 17:01:26 +0200aE From: Jan C. =?iso-8859-1?Q?Vorbr=FCggen?= <jvorbrueggen@mediasec.de> ; Subject: Re: Space Invaders was: Re: Fortran Guru requestede+ Message-ID: <3E8C4CC6.FC980E1C@mediasec.de>s  G The other problem is that it really shouldn't be DOUBLE PRECISION TIME,sK but INTEGER*4 TIME(2) or INTEGER*8 TIME. Also, you shouldn't be using eventm flag 0.a   	Jan   ------------------------------  % Date: Thu, 03 Apr 2003 11:21:25 -0500hA From: Richard Hammersley <richard.hammersley%spam%@dartmouth.edu>n6 Subject: tcp/ip for vms 5.3 compat. w. vms 6.2 and 7.1, Message-ID: <3E8C5F85.2010606@dartmouth.edu>  K Is anyone running tcp/ip for VMS ver. 5.3 on alpha vms 7.1 or vax vms 6.2 ?   F The spd says you cannot buy software support for below vax vms 7.2 or B alpha vms 7.2-2 but does not say what min. version it will run on.   Thanks.o   Richard,   ------------------------------  $ Date: Thu, 3 Apr 2003 09:22:23 -05005 From: "Dale A. Marcy" <dqmunicorn@y12unicorn.doe.gov>s8 Subject: Re: The Space Invaders minute of the day: /NOOP+ Message-ID: <b6hg2v$r4$1@sws1.ctd.ornl.gov>c  I Just a wild thought.  Was this program keyed in from listings?  If so, itiJ could be possible that the code is allowing implicit variable declarationsL and there is a typo in the variable name being passed to the Sleep function.J It would implicitly declare the typo-ed variable as either integer or realI and pass a 0 by reference.  I have seen these problems before and that is G why I always use the IMPLICIT NONE command when I code Fortran modules.d   --
 Dale A. Marcy  VMS System Manager5 SAIC - Science Applications International Corporation 
 (865)576-4942n   Berra's Universal Law -t2 "Before everything changed, it was all different."   Hint to reply by e-mail:% Everyone knows there are no Unicorns.t  4 "Carl Perkins" <carl@gerg.tamu.edu> wrote in message& news:2APR200318265749@gerg.tamu.edu...: > Didier Morandi <Didier.Morandi.nospam@Free.fr> writes...= > }and here is the run when the source is compiled with /NOOPt > }s> > }%SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=000000000000) > }0030, PC=FFFFFFFF808E5380, PS=0000001Bi2 > }%TRACE-F-TRACEBACK, symbolic stack dump followsK > }   image    module    routine             line      rel PC           abst PCB > }                                             0 0000000000000000 FFFFFFFF808E5380B > }  SPCINV  INVADERS  SLEEP                  141 0000000000000A0C 0000000000030A0CB > }  SPCINV  INVADERS  SPECIAL_GRAPHIC        579 0000000000003150 0000000000033150B > }  SPCINV  INVADERS  INVADERS                42 000000000000038C 000000000003038CB > }                                             0 FFFFFFFF802513D4 FFFFFFFF802513D4	 > }DTL02>. >p > This makes more sense. >sG > This indicates that the failure is, as I suspected, in the very firstaG > call of SPECIAL_GRAPHIC, and it is the call that is SLEEP('10.00') ino > there. >iF > This is a very short distance into the run. You should run it in theG > debugger and go through the program via STEP, not GO. (Using SET MODEwJ > SCREEN so you can see where you are might be helpfull, since this shouldH > all be before the thing gets set to graphics mode - there is the minorF > annoyance of it clearing the screen, but that doesn't really cause aG > real problem). When you get to that first call to SPECIAL_GRAPHIC, do J > a STEP/IN and step until it gets to the call to SLEEP. Then STEP/IN and,F > before doing anything else check to see what the TAG variable holds. >a: > For me, EXAM TAG shows "INVADERS\SLEEP\TAG:     '10.00'" > E > If this is not what you get, then somthing is definately not right. & > See what the call used as arguments: > 
 > What I get:- >- > DBG> show stack 1  >p > invocation block 0 >m >     FP: 000000007AE47930. >     Procedure Descriptor (00000000000101A8):& >         Flags:                  3089; >           KIND:                 PDSC$K_KIND_FP_STACK (09)e >           FP is Base Registerg& >         Rsa Offset:             0050& >         Signature Offset        00000 >         Entry Address:          INVADERS\SLEEP5 >         Ireg Mask:              2000000C <R2,R3,FP> : >           RA Saved @ 000000007AE47980:  0000000000033248: >           R2 Saved @ 000000007AE47988:  00000000000104D8: >           R3 Saved @ 000000007AE47990:  000000007B0397F6: >           FP Saved @ 000000007AE47998:  000000007AE479A0* >         Freg Mask:              00000000* >         Size:                   00000070 >>% > Commands that may be useful to you:  >eG > You can do a SHOW SYMB/FULL TAG to find out where it's data is cominga from,kI > if you can't determine it from the above (I can't - I have no clue whattC > R2 is being used for and why it isn't the argument, for example).  >  > DBG> show symb/full tagr > data INVADERS\SLEEP\TAGi >     address: %R3A >     string descriptor type, character-coded text, size: 5 bytes  > F > Examine what is at the address given in R3 as 5 bytes of ASCII text. >b > DBG> exam/ascii:5 @%R3 > 66848:  '10.00's >kF > Or examine the data at the address stored in R3 as if it were of theH > type that matches the TAG variable. Since yours doesn't seem to match, > you may not find this useful.I >o > DBG> exam/type=(tag) @%R3P > 66848:  '10.00'o >h% > Then do a STEP and see if it is theoG >    141:         DELTA_TIME='0 ::'//TAG          !Concatenate strings.dG > line that fails. It may be. It may not be - it might be the next line C > (the line number given for the last stage of the traceback is noteF > necessarily completely accurate, being off by 1 is not too unusual). >,@ > If that first STEP command in this routine does not fail, then >s > DBG> exam delta_time- > INVADERS\SLEEP\DELTA_TIME:      '0 ::10.00'  >>F > to make sure that it matches what I just got here, which is correct. >oF > Then step until failure to locate the problem (if you havn't alreadyH > found it), examining the contents of all relevent variables before and > after every function.  > DBG> s% > stepped to INVADERS\SLEEP\%LINE 143 F >    143:         IF(RET_STAT.NE.%LOC(SS$_NORMAL))STOP'BINTIM failed.' > DBG> exam ret_stat# > INVADERS\SLEEP\RET_STAT:        1h > DBG> exam time0 > INVADERS\SLEEP\TIME:    0.603762814651698E-159 > 9 > is what I get after the step when SYS$BINTIM is called.  > L > BTW: I would really change the TIME variable to INTEGER*4 TIME(2). Really. >e
 > --- Carl   ------------------------------  + Date: Thu, 03 Apr 2003 10:16:01 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>cB Subject: trademarks (was: RE: OpenVMS and Security - REALLY !!!!!); Message-ID: <01KUACGXQQF6A9LL1B@sysdev.deutsche-boerse.com>r  A >>"OpenVMST is a proven state-of-the-art, easy-to-implement vaults >>management system "r >>E >>I guess HP doesn't own the trademark "OpenVMS". These guys, Controll  F A trademark only replies to a specific area.  Thus Apple Records (the D Beatles' company) and Apple Computer aren't infringing each other's I trademarks.  There is a cigarette called Prince, and a musician who used h  to go by that name.  No problem.  F Now, I couldn't use VMS as a name for an operating system, or perhaps ' not even for anything computer related.   B IIRC correctly, there is some sort of "? Milking System" with the D initials VMS.  And the most intriguing manuscript in the world.  No  problem.  F What I don't understand is why Brian, with 4 children and a last name H which begins with "S", doesn't have a Victor(ia) Michael(a) or whatever G among his progeny.  :-)  (Or maybe there IS a possibiity for confusing  E here, since with 4 children I suspect that the "up time" might rival t( that of our favourite operating system!)   ------------------------------  # Date: Thu, 03 Apr 2003 11:49:23 GMT-" From:   VAXman-  @SendSpamHere.ORGF Subject: Re: trademarks (was: RE: OpenVMS and Security - REALLY !!!!!)0 Message-ID: <00A1DD35.9F879C39@SendSpamHere.ORG>  w In article <01KUACGXQQF6A9LL1B@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:vB >>>"OpenVMST is a proven state-of-the-art, easy-to-implement vault >>>management system " >>>wF >>>I guess HP doesn't own the trademark "OpenVMS". These guys, Control >,G >A trademark only replies to a specific area.  Thus Apple Records (the kE >Beatles' company) and Apple Computer aren't infringing each other's hJ >trademarks.  There is a cigarette called Prince, and a musician who used ! >to go by that name.  No problem.r >rG >Now, I couldn't use VMS as a name for an operating system, or perhaps s( >not even for anything computer related. >gC >IIRC correctly, there is some sort of "? Milking System" with the tE >initials VMS.  And the most intriguing manuscript in the world.  No y	 >problem.l >sG >What I don't understand is why Brian, with 4 children and a last name cI >which begins with "S", doesn't have a Victor(ia) Michael(a) or whatever tH >among his progeny.  :-)  (Or maybe there IS a possibiity for confusing F >here, since with 4 children I suspect that the "up time" might rival ) >that of our favourite operating system!)   F I tried passing that by my wife when each was born but she wasn't too 3 keen on the name choices available begining with V.l  E I do, however, have a Sheltie registered "Starlet dot MLB" -- Starletd
 for short.  E As for uptime, MicroVAX IV -- code name Kevin -- has a remarkable up-gD time.  Boottime has been at about 4:30am in recent weeks and remainsC running until a forced crash (and that's not easy) at about 10:00pmtD or laters.  I, on the otherhand, find that I am needing the downtime of a PeeCee.  ;)   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM             R5   "Well my son, life is like a beanstalk, isn't it?" a   ------------------------------   Date: 3 Apr 2003 09:45:08 +0100.K From: pmoreau@ath.cena.fr (Patrick MOREAU, CENA Athis, Tel: 01.69.57.68.40)e  Subject: Re: TZ30's and Alpha's?! Message-ID: <nLhrVz+3vWgI@sinead>s  , In article <b6di7q0bl1@enews3.newsguy.com>, 4 "Zane H. Healy" <healyzh@shell1.aracnet.com> writes:H > Out of curiousity can a TZ30 drive be connected to an Alpha such as an' > AlphaStation 200 4/233, or PWS 433au?u  H I have a TZ30 (into a BA pizza box) hooked up to a 255/233. No problems.  F I also have a TZ86 hooked up to a DS10 (with a DDS3 DAT). No problems.   Patrick  --O ===============================================================================rN pmoreau@ath.cena.fr  (CENA)     ______      ___   _           (Patrick MOREAU)4 moreau_p@decus.fr (DECUS)       / /   /     / /|  /|J CENA/Athis-Mons FRANCE         / /___/     / / | / |   __   __   __   __  N BP 205                        / /         / /  |/  |  |  | |__| |__  |__| |  |N 94542 ORLY AEROGARE CEDEX    / /   ::    / /       |  |__| | \  |__  |  | |__|N http://www.ath.cena.fr/~pmoreau/            http://www.multimania.com/pmoreau/O ===============================================================================    ------------------------------   Date: 3 Apr 2003 12:09:29 -0600e; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)h/ Subject: Re: who is procedure of backup for vms 3 Message-ID: <qlHmUwVw8cbb@eisner.encompasserve.org>m  T In article <oQ21FuMtfxIx@eisner.encompasserve.org>, briggs@encompasserve.org writes: > D > Personally, I'm in the "go ahead and use /IGNORE=INTERLOCK" crowd.  F    I am, but only because I know what's going on on my system disk and)    I know I won't lose anything painfull.y  E    If I had a DBMS or active critical files on my system disk I would "    not count on /IGNORE=INTERLOCK.  ;    And I always get a standalone backup after an OS update.    ------------------------------   End of INFO-VAX 2003.184 ************************