1 INFO-VAX	Fri, 26 Dec 2003	Volume 2003 : Issue 714       Contents:P Re: DCL extensions Was ( How to make exe return value to dcl symbol &         Fl DynDNS.org client , Re: Floating point arithmetic support in DCL, Re: Floating point arithmetic support in DCL, Re: Floating point arithmetic support in DCL Re: Merry Christmas! Mirror-Set vs. Partition-Set My Holiday Wishes & Re: scrambled terminal output on OPA0:( Re: Singapore Exchange to run on OpenVMS Re: sysman$node_table   F ----------------------------------------------------------------------  % Date: Thu, 25 Dec 2003 23:17:32 -0600 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>Y Subject: Re: DCL extensions Was ( How to make exe return value to dcl symbol &         Fl 6 Message-ID: <3FEBC46C.5E33D2DB@NeOaSrPtAhMlNiOnWk.net>   Tom Linden wrote:  > @ > I believe adding decimal arithimetic might be useful for many,K > but to some extent it is reinventing the wheel and doesn't go far enough.  > J > In thinking about how one might implement such features, it occurs to me > thatJ > far more useful would be the ability to have a facility for user-defined > lexical functions, e.g.  > 7 > $ define lexical foo(sdt1, sdt2, ...) [returns(sdtr)] ' > with default return being char string - > where sdt := SDL data_type, so for example.  > ? > $ define lexical payment(word,decimal precision(10,2),decimal  > precision(10,2)) -% > _$ returns(decimal precision(10,2))  > 
 > then you > $ years=15 > $ interest=6.0 > $ principal=500000.22 3 > $ write sys$out payment(years,interest,principal)  > A > By making it a lexical function  then one can include in pipes.  > I > Now if that were available I could implement all the math functions you , > would most likely want in less than a day!  D I agree that user-defined lexicals would be very useful, but I worry" that it might compromise security.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Thu, 25 Dec 2003 22:06:02 -0600 - From: "MARVIN JOHNSON" <m68k@email.uophx.edu>  Subject: DynDNS.org client0 Message-ID: <4LmdnfLp3PEJLnaiRVn-hQ@comcast.com>  * Is anyone using a VMS box with DynDNS.org?  L I am wanting a client for their service. I found a Perl client available and figured that may work (?).   -- MARVIN JOHNSON   ------------------------------  % Date: Thu, 25 Dec 2003 14:43:34 -0500 * From: JF Mezei <jfmezei.spamnot@istop.com>5 Subject: Re: Floating point arithmetic support in DCL ) Message-ID: <3FEB3DDB.33AC78AC@istop.com>    Guy Peleg wrote: > I will look at native  > floating point supportH > and the amount of efforts involved. Who knows maybe we will be able to > provide it for V8.2......   L Wouldn't the average VMS engineer be able to implement this in one evening ?H And I fully expect you to backport this to VAX VMS 7.2 as well, but I amK willing to wait an extra day or two before that patch becomes available :-)  :-) :-) :-) ;-) :-)   M Out of curiosity, what information do you store in the definition of a symbol L ? Is there a publicly available structure hidden in a library that describes2 it, or is that available only in VMS source code ?   ------------------------------  % Date: Thu, 25 Dec 2003 22:57:51 -0600 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>5 Subject: Re: Floating point arithmetic support in DCL 6 Message-ID: <3FEBBFCF.5D0A2816@NeOaSrPtAhMlNiOnWk.net>   Guy Peleg wrote: > " > Thank you all for your feedback. > G > Here is what I would like to suggest to summarize all the responses :  > M > * For V8.2 we will introduce new lexical function F$MATH. While this is not 6 > the best approach, it will provide better arithmetic; > capabilities in DCL without breaking existing procedures.  > N > * For a post V8.2 release, we will implement "native" floating point support4 > in DCL. V8.2 is expected to ship in H204 so we areF > probably talking about mid-late 2005 for native floating point. When: > available, I will publish the spec to get your feedback/ > L > Please let me know if you agree with this strategy and again thank you for* > your feedback, it is highly appreciated. >  > Happy Holidays,  >  > Guy  > = > "Guy Peleg" <guy.peleg@hp.com_remove_this> wrote in message / > news:EWTFb.11204$bD4.3662@news.cpqcorp.net... 
 > > Hi Folks,  > > : > > Here is your chance to influence the future of VMS.... > > D > > I have investigated the possibility of supporting floating point > arithmetic1 > > in DCL. The intention is to support something  > > similar to the following:  > > 
 > > a=1.23
 > > b=2.44 > >  > > write sys$output a+b > > 3.67 > > C > > My intention is to support the following operands : +,-,/ and *  > > J > > If this feature is important to you please drop me a line with a short: > > justification. Assuming the response will be positive,* > > I will try to implement this for V8.2. > > & > > Merry Christmas and happy new year   Thanx, Guy!   < Your enthusiasm and can-do attitude are greatly appreciated!   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Thu, 25 Dec 2003 23:09:32 -0600 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>5 Subject: Re: Floating point arithmetic support in DCL 6 Message-ID: <3FEBC28C.D8AEC48B@NeOaSrPtAhMlNiOnWk.net>  / Phillip Helbig---remove CLOTHES to reply wrote:  > A > In article <8fdGb.6079$xm.245164@nasal.pacific.net.au>, Phaeton " > <spameater@spam.invalid> writes: > H > >       Will we see a DCL compiler in the future ? What I have in mindG > >       is that we could create small, fast and tight .EXE-s from DCL  > >       code.  > A > There is Charlie Hammond's DCL_DIET, which will speed things up J > somewhat.  I've heard of a tool which converts DCL procedures to Fortran< > code then compiles and links that.  Should be pretty fast.  7 There was a third-party product which did exactly that.   D Trouble is, the vendor built-in a poison-pill: The .EXE's required a) proprietary RTL which required a license.   ' Needless to say, the poison back-fired.   H So, I'll say now what I said back then: DCL -> Macro/32 is the way to go8 - always has been. Even Alpha still ships with Macro/32.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Thu, 25 Dec 2003 20:30:04 +0100 " From: Didier Morandi <no@spam.com> Subject: Re: Merry Christmas! 4 Message-ID: <3feb3b10$0$24046$626a54ce@news.free.fr>  / To you too and to all readers, Merry Christmas! B The Lord be with us (and Europe, and Australia, and and and... :-)   D.
 www.vaxus.org    Bob Ceculski wrote:   5 > wishing every VMS user a joyful and blessed holiday 1 > season to you and your families ... OpenVMS has 5 > given us another fantastic year, and there are many 2 > more waiting for tose who use the best OS on the > planet ... Merry Christmas!    ------------------------------  % Date: Thu, 25 Dec 2003 17:20:59 -0600 ( From: brandon@dalsemi.com (John Brandon)% Subject: Mirror-Set vs. Partition-Set 1 Message-ID: <03122517205931@dscis6-0.dalsemi.com>    HSG80 dual controllers.     5 Scenario # 1 ----------------------------------------   & 6 each 36-GB disk drives on one shelf.$ 3 each Mirror-Sets of those 6 disks.D Each Mirror-Set assigned as a Unit and becomes a logical disk drive.   [d1][d2][d3][d4][d5][d6] [  m1  ][  m2  ][  m3  ] [  u1  ][  u2  ][  u3  ] [ dga1 ][ dga2 ][ dga3 ]  5 Scenario # 2 ----------------------------------------   & 6 each 36-GB disk drives on one shelf.$ 3 each Mirror-Sets of those 6 disks.) 1 each Stripe-Set of those 3 Mirror-Sets. 5 3 each Partitions are created from that 1 Stripe-Set. C Each Partition assigned as a Unit and becomes a logical disk drive.    [d1][d2][d3][d4][d5][d6] [  m1  ][  m2  ][  m3  ] [          s1          ] [  p1  ][  p2  ][  p3  ] [  u1  ][  u2  ][  u3  ] [ dga1 ][ dga2 ][ dga3 ]  G My understanding is that Scenario # 2 allows for higher performance.  6 G channels versus 2 channels.  Given you do not over-populate your drives  (obvious duh).  J Is it also more efficient because the I/O to the drives in a partition areP combined?  For example, applications write to the devices DGA1, DGA2, and DGA3.   
 WRITE DGA1
 WRITE DGA2
 WRITE DGA3  ? Would the three events be one I/O or three I/O for a partition?   P Or can someone kind of explain that process to me?  Or point me to a good link!  Thanks!        J*o*h*n B*r*a*n*d*o*n  VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Thu, 25 Dec 2003 23:25:14 -0600 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> Subject: My Holiday Wishes6 Message-ID: <3FEBC63A.E7645A04@NeOaSrPtAhMlNiOnWk.net>  B My wish for the group and for all with whom we share this world is simple:    Peace in our days.   Pax Domini sit semper vobiscum.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Thu, 25 Dec 2003 22:48:51 -0600 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>/ Subject: Re: scrambled terminal output on OPA0: 6 Message-ID: <3FEBBDB3.6B171EFD@NeOaSrPtAhMlNiOnWk.net>  / Phillip Helbig---remove CLOTHES to reply wrote:  > A > In article <3FE90401.471DEF6@NeOaSrPtAhMlNiOnWk.net>, "David J. : > Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes: > O > > > > I have a VT320 connected by an MMJ cable to a VAXstation 4000/60.  When K > > > > in an editor, the terminal output is scrambled, like there had been M > > > > dropouts, line noise etc.  (The cable is quite short.)  This persists M > > > > even after SET TERM/INQ etc.  If I then SET HOST/LAT or TELNET to the P > > > > same machine or another one (in the cluster or outside it), the terminal > > > > works fine.  > > > > M > > > > I also have a VT320 as OPA0: on a VAX 4000-100A, i.e. essentially the ; > > > > exact same setup.  I don't have the problems there.  > > > > ; > > > > What could be the problem and how can it be solved?  > > > N > > > This sounds like a flow control problem.  Check the various SET TERMINALK > > > SYNC flags (hostsync, TTsync, etc.) for the port.  Make sure that the ; > > > terminal is generating XON/XOFF flow control signals.  > > L > > Check grounding as someone suggested, check flow control as someone elseI > > suggested, and also check for dirty contacts on the cable ends and in G > > the jacks. Make sure all outlets are wired correctly. Make sure the G > > computer and the terminal see the same potential on the ground line @ > > (isolated ground?), and make sure that potential is in spec. > C > Considering that once I do SET HOST/LAT or TELNET (to the same or I > another machine) things work fine, doesn't that rule out bad grounding, H > dirty cables, missing flow control etc?  After all, if the information/ > is garbled, how can SET HOST/LAT ungarble it?   G I think that question pretty much answers itself. You're going from the ? console port on the machine to a terminal port on a server (two G different physical interfaces), no? ...or are you saying that you can - D from OPA0: - SET HOST/LAT to another node or TELNET to 127.0.0.1 and have the problem go away?   H > The terminal setup is the same as my other terminals, which work fine,) > so that doesn't seem to be the problem.   
 It depends...    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Thu, 25 Dec 2003 22:33:32 -0600 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>1 Subject: Re: Singapore Exchange to run on OpenVMS 6 Message-ID: <3FEBBA1C.B1960573@NeOaSrPtAhMlNiOnWk.net>   Rob Young wrote: > { > In article <3FE90142.3EBFB4E3@NeOaSrPtAhMlNiOnWk.net>, "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes: 2 > > Hhmmm... Nice job of side-stepping the issues. > > J > > This being Christmas, I'm feeling generous. Let's try one more time... > >  > % > >> > Are there any other questions?  > >> > > >> > Suggestion: > >> >M > >> > Take a sabbatical from the VMS nest and come see how you fare out here I > >> > in the real world of EDP where the rubber meets the road and where I > >> > careers are made or broken by what top management (i.e., "decision F > >> > makers") believe is a safe long term investment in hardware andM > >> > operating software. My money says you won't last 30 seconds out here - ? > >> > the sharks will eat you alive in the first five seconds.  > >> > > >>" > >>         Not even close laddy. > > C > > I'm thinking dead on, based on the consistent "pie in the sky", I > > PollyAnna attitudes we see here from you and some other VMS insiders.  > >  > J >         No.  Your assertion above is that I live in a "VMS nest."  WhileK >         entertaining yourself with your cute little metaphors, you aren't  >         even close ... laddy.    Try again.    G Within the limits of your VMS-related job responsibilities, you exhibit G - here - a naivete that can only be termed "frightening" by someone who F abides out here in the front-line trenches - a place I dare say you'veD never seen, or at least have not seen in recent history - 24 hours a day, 7 days a week.   6 >         I actually have 4 or 5 job respnsiblities.    D I'm glad for you. You are still a VMS "PollyAnna", none the less. InH VMS-land, life "is". Nothing especially good, and mostly very worrisome.  H >         To suggest I "won't even last 30 seconds out here - the sharksJ >         will eat you alive in the first five seconds" sounds oh so scary: >         but isn't very accurate.  But self-entertaining?  H "Frightening" would be a better word. As relates to VMS, you have so farG displayed precious little that even intimates a solid grasp of anything B related to the business issues facing those of us out here on "the front".   E Your other skills may carry you as relates to those areas. In the VMS H arena, you're dead meat before you've even begun. Don't believe me? I'llA get you an appointment with our IT Director and our V.P. of I.T.    D Oh, be sure to bring a flame-proof suit to wear over your chain-mail! armor and Kevlar-reinforced vest.    > > G > > Let VMS learn the lesson of the Ghost of Christmas Yet To Come. The K > > epitaph was written years ago, but is as truthful today as it was then. ; > > You know what I mean - I don't HAVE to repost it, do I?  > >  > O >         I know what you mean, and if you say it long enough, you and Tom will + >         eventually be right, or will you?   D Rather depends on whether your management ever learn what we already? know, what most of our employers know, what most companies more E successful than hp already know, and what all of these have known for  many years.   I > "OpenVMS customers heavily dependent on third-party applications should K > immediately begin exploring migration options with an objective of moving N > most key packaged applications off the OpenVMS platform by the end of 2001." > / >    Gartner Group senior IT analyst Tom Henkel 	 >    1999   D Ah, so you *DO* see why advertising is so vital! For the less astute however, I will clarify:  F IDC and Gartner both carry a certain degree of credibility out here inE the market place. Hp's credibility is currently a negative value, and F silence has no expressible credibility at all, save that it proves theH nay-sayers' point. Thus, to leave no room for doubt, VMS's position mustD be made unmistakably clear through widespread mainstream advertising over and over and over again.   B Ask yourself: what Compaq do counter such press? Remember to quoteF publication name, date of publication or air and page number or market where it was aired.   I > > "Mens' actions lead to certain ends. If those actions be changed, are  > > not the ends also changed?"  > > 
 > > - Scrooge  > > G > > Even *HE* eventually learned. Whether hp will remains to be seen...  > >  > G >         Learn what?  They are running an $80 billion dollar business.   & Hoopty friggin' doo! Your point is...?   > It'sH >         a chuckle to read the rantings of folks that would marginalizeH >         HP's handling of their business - the outside in looking at an& >         $80 billion dollar business.  G ...that is hemorrhaging jobs, it's flagship software product languishes H in the hell of lies and deceit foisted upon the user base and the marketH by past management incompetence while current management remains blindedF to the needs to undo past damage and shore-up hp's sagging share price and VMS's market position.  ? Sure - I'm gonna bet *MY* life savings on this bunch - *NOT*!!!   F You seem frighteningly quick to hang your hopes on hp that their otherA businesses will keep VMS alive even in the face of the relentless G onslaught of fiscal irresponsibility, governant malfeasance and stealth  marketing.    E I'm sure I don't have to mention how many lay off victims thought the B same of their position, business unit, product line, etc. in other7 companies prior to the sudden downturn in their career.    ...or do I?   H >         Suppose for the sake of argument VMS folded the tent, ok fine.  H ...or got it's tent folded for it by the market, hp management, etc. ...   > I 9 >         wouldn't be any less busy than I am right now.     I'm happy for you.   > Would a lot of the3 >         VMS talent get snatched up?  Absolutely.    = ...if their jobs don't get outsourced to India or Asia first.    > Fringe players may have I >         a tough time but fringe players have a tough time regardless of  >         segment.  A Working for one of hp's more visible VMS customers in the Chicago 7 market, I'd hardly call myself "fringe", but rave on...   # >         But VMS isn't going away:  > : > The IDC Quarterly Server Tracker data shows that OpenVMSE > server customer revenues for Q3CY2003 (HP's fiscal Q4) increased by G > more than 7% over the same quarter a year ago, and units shipped were  > up by more than 23%.  F ...and, as I've asked multiple times but you've consistently attemptedG to skirt the issue, how's VMS doing now as compared to three years ago? $ ...five years ago? ...ten years ago?  B >         Those are facts presented by IDC.  Care to dispute them?  F The numbers have never been in dispute. I'm glad to see VMS doing wellB versus prior year, for what little that may be worth. 23% of what?F ...and the value of "what" three, five or ten years ago was ...? (Fill in the blank.)  E The issue is the time frames used for comparison: what little VMS has G left versus the plentitude it had when it was where it needs to be once * again to resume healthy growth once again.  ! Starting to get the picture now?    H That's three(3) - count 'em: three(3) - times now that I've asked you toE produce that information. Not for my sake, mind you (I already know), F rather for your own. Got the balls to "leave the nest" and face a grim reality?   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Thu, 25 Dec 2003 23:19:40 -0600 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> Subject: Re: sysman$node_table6 Message-ID: <3FEBC4EC.FA4ABE22@NeOaSrPtAhMlNiOnWk.net>   Michael Austin wrote:  > 1 > Phillip Helbig---remove CLOTHES to reply wrote:  > J > > I'm looking for a way to specify logical names within sysman commands. > > $ > > Ideally, I want to be able to do > >   > >    SYSMAN> DO @LOCATION:FILE > > 5 > > where LOCATION is a logical name.  Alternatively,  > > + > >    SYSMAN> SET PROFILE/DEFAULT=LOCATION  > >    SYSMAN> DO @FILE  > >  > > would be OK. > > 6 > > If I fire up SYSMAN with just the defaults then do > >  > > SET PROF/DEF=LOCATION  > > K > > this seems to work, even though LOCATION is defined only in the process J > > table of the process which started SYSMAN.  If I first do the SET PROF  > > then SET ENV/CLUSTER, I get: > > R > >    %SYSMAN-I-PROFRESET, profile on remote nodes has been reset to UAF defaults > > J > > However, if I do a DIR command, I am indeed in LOCATION (which was not6 > > where I was when I fired up SYSMAN) on every node. > > ; > > If I first do SET ENV/CLUSTER then SET PROF/DEF, I get:  > > ? > >    %SYSMAN-I-NODERR, error returned from node <remote_node>  > > J > > for the non-local nodes.  Nevertheless, the DIR command shows that the8 > > default directory is indeed LOCATION, on every node. > > # > > DO @LOCATION:FILE doesn't work.  > > L > > None of the above is changed if I do SET PROF/PRIV=ALL.  The account has7 > > READALL and OPER enabled by default and has SETPRV.  > > L > > I would have thought that SYSMAN$NODE_TABLE would be the place to defineH > > logicals which SYSMAN knows about, but this doesn't seem to work for< > > directory specifications (it works fine for node lists). > >  > > What am I missing? > >  > & > I have used this trick many times... >  > SYSMAN> set env/clust D > SYSMAN> do pipe (define location disk:[dir] ;  @location:filename) > 3 > notice the spacing around this semi-colon ^^^^^^^   H Likewise, I find writing SYSMANINI files on the fly to be useful in some& cases, if a rather cumbersome to code.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------   End of INFO-VAX 2003.714 ************************