0 INFO-VAX	Tue, 29 Jan 2002	Volume 2002 : Issue 55      Contents: Re: A position statement Re: A position statement Re: A position statement% Re: Alphaserver 4100: physical memory % Re: Alphaserver 4100: physical memory % Re: Alphaserver 4100: physical memory  Blade announcement Re: Blade announcement Re: Blade announcement Re: Blade announcement RE: Blade announcement Re: Blade announcement Re: Blade announcement! Blade Server for Network Security 1 Re: Can PC's read OpenVMS index-sequential files? 1 Re: Can PC's read OpenVMS index-sequential files?  Re: Capellas conference  Re: Capellas conference D Re: Capellas redefines Industry Standard to mean Windows *and* Linux Re: Compaq financial conference  Re: Compaq financial conference  Re: Compaq financial conference  Re: Compaq financial conference  Re: Compaq financial conference  Re: Compaq financial conference P Compaq site has link to Gradient Technologies DCE but not for VMS or NSK. NSK.NS" Re: Cross Compiler to SUN/Solaris?4 data-entry/keystroke job... howto display to screen?8 Re: data-entry/keystroke job... howto display to screen? DCE Programming  DCOM Programming Under OpenVMS$ How to get/put the symbol list in C?( Re: How to get/put the symbol list in C?( Re: How to get/put the symbol list in C?( Re: How to get/put the symbol list in C? Re: HP 35470 DAT drive question ! Re: I/O Performance on Alpha 4100  In Need of 9-track VMS Re: In Need of 9-track VMS Installing GNV RE: Installing GNV$ Intel Special issue of OpenVMS Times( Re: Intel Special issue of OpenVMS Times( Re: Intel Special issue of OpenVMS Times2 RE: Microsoft costs sucker windoze users billions!2 Re: Microsoft costs sucker windoze users billions! Re: mount problems on cluster  Re: mount problems on cluster  Re: mount problems on cluster  Re: mount problems on cluster ( Re: new partner also sent to openvms.org( Re: new partner also sent to openvms.org2 News Alert:Military Resale Group, Inc.(OTCBB-MYRG)+ Re: ON ERROR when READ SYS$INPUT?!?!?!??!?! + Re: ON ERROR when READ SYS$INPUT?!?!?!??!?!  Re: Open  PGPlot on OpenVMS and PNG Output$ Re: PGPlot on OpenVMS and PNG Output$ Re: PGPlot on OpenVMS and PNG Output Re: regarding HP regarding HP Re: STR$lowercase ?  Re: STR$lowercase ?  Re: STR$lowercase ?  Re: STR$lowercase ?  Re: STR$lowercase ?  Re: STR$lowercase ?  Re: STR$lowercase ?  Re: STR$lowercase ? ; Re: Younger recruits versus experienced veterans  ( was The ; Re: Younger recruits versus experienced veterans  ( was The   F ----------------------------------------------------------------------  % Date: Mon, 28 Jan 2002 20:39:26 +0100 1 From: John McLean <mcleanj@swissonline.delete.ch> ! Subject: Re: A position statement 5 Message-ID: <3C55A8EE.5F167741@swissonline.delete.ch>   D Last week I wrote this about Compaq's institutional shareholders ...  % > The NASDAQ reports of institutional J > shareholders tell us that for the period from June 30th to Sep 30th (ie.I > from just after the Alpha announcement until after the merger proposal)  > I > - The top 3 SELLERS were all based in Boston and together sold over 100  > million shares, > - The 4th highest seller was from Houston.B > - Of the top 14 sellers, 6 were based in either Massachusetts or
 > Connecticut G > - Number one seller was Putnam Investment Management who *used to be* E > the top shareholder.  Putnam sold 87.5% of their holding of over 64 D > million shares, dropping them to 26th highest stockholders.  (IIRC4 > Putnam invested heavily in Digital some time ago.)1 > - No buyer bought more than 14.6 million shares G > - Of the top 14 buyers, 1 was from Massachusetts (14.6 million) and 1  > from Connecticut (5 million) > F > It really looks like a vote of no-confidence in Compaq by people who > should know !     H I investigated all this in NASDAQ.com back on 1st of 2nd of January (andA even made a small Excel spreadsheet showing the top 14 buyers and F sellers).  Now when I look at the figures they are different.  PuttnamG is shown as BUYING shares not selling them, but all of the others in my  spreadsheet are correct.  F The current listing says that Putnams bought about 16 million shares. B The original figures reported a drop to 16 million, rather than anE increase of 16 million.  When I do the calculations it looks like the I change may have been incorrectly reported by NASDAQ and then corrected.     E Could I have been mistaken when I first collected the figures ?  It's E unlikely because reductions in stockholdings are in red but purchases / are shown in black.  I'm not that color blind !   + My apologies for misleading anyone on this.      John McLean    ------------------------------  % Date: Mon, 28 Jan 2002 15:07:20 -0500 % From: JF Mezei <jfmezei@videotron.ca> ! Subject: Re: A position statement , Message-ID: <3C55AF69.F42612E4@videotron.ca>   John McLean wrote:J > I investigated all this in NASDAQ.com back on 1st of 2nd of January (and  G > Could I have been mistaken when I first collected the figures ?  It's G > unlikely because reductions in stockholdings are in red but purchases 1 > are shown in black.  I'm not that color blind !   " NASDAQ.COM runs on windows.... :-)  / http://www.nyse.com/cgi-bin/ny_charts?sym=CPQ     L provides interesting charts. Look at the quarter and year charts, especiallyV the volume of shares traded. The stock has been going up in value, but down in volume.   ------------------------------  % Date: Mon, 28 Jan 2002 15:08:05 -0500 % From: JF Mezei <jfmezei@videotron.ca> ! Subject: Re: A position statement , Message-ID: <3C55AF96.2293F87E@videotron.ca>  8 Found this interesting article on the nyse.com web site:  6 H-P's Fiorina, Walter Hewlett Preach To The Undecided      1/28/02 10:15am    By Mark Boslet       Of DOW JONES NEWSWIRES  M   PALO ALTO, Calif. -(Dow Jones)- With a majority of major stockholders still I   undecided about Hewlett-Packard Co.'s (HWP) big merger, Chief Executive K   Carly Fiorina and opponent Walter Hewlett are on the road again hoping to (   sway this most important constituency.  R   They will find there is plenty of persuading to do, if interviews with 10 of theM   top 50 institutional shareholders are an indication of how much uncertainty I   remains over the proposed $24 billion Compaq Computer Corp. (CPQ) deal. I   Seven of these institutions contacted by Dow Jones Newswires last week, K   representing about 97 million shares, or 4.4% of H-P's outstanding stock, N   say they have not finalized plans on how to vote at a proxy tally that could   come as early as March.   F   Two of the holders indicated they would oppose the merger, while oneK   professed reluctant support. Combined, these more committed shares add up    to more than 2% of the stock.   E   The broad indecision suggests arguments from both sides have yet to M   convince many financial institutions and money managers. It also raises the N   possibility the momentum may be shifting in the fight. It does seem H-P "hasQ   leveled the playing field," says Jose Arau, an analyst at the California Public M   Employee's Retirement System. First, "you heard one side get all fired up," F   then the other comes forward with its arguments. "It's fascinating."  K   In December, when the Packard Foundation joined Hewlett and the trust and K   foundation he helps run to oppose the deal, 18% of H-P the stock lined up ;   against the transaction. The deal looked dead on arrival.   H   Since then, the company has launched a campaign to undermine Hewlett'sN   credibility, referring to him as a "musician and academic" with conservativeK   investment goals differing from those of other shareholders. Last Monday, P   Alliance Capital Management Holding L.P., previously identified by insiders asE   a supporter of the transaction, came out publicly backing the deal.   I   On Friday, Compaq Chief Executive Michael Capellas tried to create more J   sway by saying he believed Putnam Investment Management Inc. favored the;   deal. Putnam later declined to discuss how it would vote.   O   Clearly, H-P executives feel better about their odds. "Every day that goes by K   I am less worried," says H-P Chief Financial Officer Robert Wayman. "I am H   broadly encouraged." Wayman, who is assisting Fiorina in the company'sN   latest round of presentations begun two weeks ago, says, "There are a lot ofO   investors who want to talk to us. A number of them have indicated support for 
   this deal."   N   Hewlett's camp likewise isn't ready to concede an inch and suggests it, too,K   can carry the day. "The painting of Walter Hewlett as not having the same P   interests (as institutional shareholders) is obviously wrong, but it prompts aF   question at some meetings," says Spencer Fleischer, vice chairman ofK   Hewlett advisor Friedman Fleischer & Lowe. "Shareholders want to know who 4   he is," and they find out "he is very determined."  N   Interviews suggest portfolio managers are looking beyond the attack. SeveralP   called Hewlett's and Fleischer's joint presentation "logical" and containing "F   valuable points." "Personally I am a little disappointed they had to (resort) to B   attacks on him," adds one manager who asked to remain anonymous.  M   "We continue to focus on the merits of the deal, or the lack of them," says L   Fleischer, who has conducted about 30 presentations, the most recent round"   beginning in the past two weeks.  P   With the pressure rising for both sides, the current presentations could proveN   decisive. H-P's top 50 shareholders own about 41% of its stock, according toH   Securities and Exchange Commission filings dating mostly to September,.   enough to determine the outcome of the vote.     The Heart Of The Matter   P   At the heart of H-P's presentation is its goal of using the merger to become aF   more competitive supplier of the vast array of products and servicesK   corporations need, says Wayman. "The enterprise story was the hardest for 3   people to relate to," he says. "Now they get it."   M   Fleischer counters that the risks of integrating a company as big as Compaq K   are greater than H-P claims. Compaq has tried for several years to better M   compete with more efficient Dell Computer Corp. (DELL) and still Dell pulls J   away. Compaq also doesn't fill many of the strategic gaps H-P has in itsH   business-computing lineup. "H-P shareholders are hugely overpaying for   Compaq," he says.   P   Investors are eager to listen, but they are skeptical. They come at you with "M   all these estimates of synergies and growth," Arau points out. "You have to M   take them with a grain of salt." More than with other mergers, "I find fund O   managers are taking an exceptionally diligent view of this" combination, says N   Scott Winkler, chief executive at business consultant D.H. Brown Associates,M   who has received lots of calls from portfolio managers. "The smart ones are     looking at the customer base."  J   Also awaited is the report from the Institutional Shareholders Services,#   expected 20 days before the vote.   L   "It is not an easy decision," says one money manager who remains undecidedN   and "very unhappy H-P's most profitable business (its printers) shrinks as aF   percent of the (new) total." At the same time, the combined servicesL   businesses won't compare to the successful upscale services and consultingH   operations at International Business Machines Corp. (IBM), the manager   says.   N   For many, the analysis is boiling down to dollars and cents. Benjamin Easow,N   an analyst at Northern Trust Global Investments, says he is still "crunchingJ   the numbers" after meeting with H-P management. His organization remains   officially undecided.   N   Wall Street experts say investors have warmed to the notion that merging theI   two companies' PC operations will let their larger size produce greater N   efficiencies. Some also don't like the alternative of turning an independentO   H-P back over to a board that might not be willing to make necessary job cuts    and cost reductions.  M   "I get both arguments," said one investor who opposes the deal, and who has J   listened to the 60- to 90-minute presentations from both sides. Compaq'sO   lower margin hardware is not the right combination for H-P, the manager said. I   "They should be buying a business that increases the return on capital. G   (Instead), they are buying one that decreases the return on capital."   O   Examined closely are two sides' differing positions on expected revenue loss. I   H-P argues the combined businesses will lose about 5% of their revenue. M   Hewlett puts its closer to 10%. The difference can change the deal's bottom    line.   P   "This is an area of high, high interest," says Wayman. He adds that H-P's goalO   is to do better than its assumption. Fleischer fires back: "We think they are    overly optimistic."   O   There is no way to be sure of the assumptions, says Dwight Blazin, an analyst J   and portfolio manager at Davis Selected Advisers. Portfolio managers areN   looking at the revenue risk from many angles. Blazin said one of his biggestN   concerns is whether distributors of products such as PCs will keep all theirN   current business with the combined companies. If a distributor orders 80% ofL   its PCs from H-P and Compaq today, will the customer concentrate that muchH   business in a single company? asks Blazin. The answer is hard to know.  L   Some investors have already cut their exposure to the deal by beginning toM   sell their H-P and Compaq stock. But eliminating large positions takes time K   and might not be completed before the vote. This is especially true since 7   about 79% of H-P shareholders also have Compaq stock.   N   The financial markets are convinced that if the deal goes through, H-P stockM   is likely to fall further as investors worry the combination will go badly.  If theE   deal falls apart, Compaq shares could slip as investors rethink its *   opportunities as an independent company.  L   It becomes a question of "which you are selling down the river," quips one
   manager.  3   By Mark Boslet, Dow Jones Newswires; 650-496-1366   >   (This story was originally published by Dow Jones Newswires)   ------------------------------  % Date: Mon, 28 Jan 2002 11:29:10 -0800 C From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam> . Subject: Re: Alphaserver 4100: physical memory0 Message-ID: <3C55A686.A8BA638E@intel.com.nospam>   Fabio Cardoso wrote:   > Dear Sirs  > - > Is there a way to check the position of the ' > memory  in the Alphaserver 4100 slots $ > using ANALYZE/SYSTEM or DECEvent ? > ( > We have two 4100 (one with 4GB and the) > other with 6 GB) and we need to upgrade + > both to 8 GB (max memory in this server).  > + > So, I dont want to open the machines, and & > I need to make the specifications of > the current memory positions.   6     I don't believe you can find that information from9 running system.  What you want is the console output from 3 powering on the system.  Is the system connected to 3 PCM, or do you log the console output to a printer?  That's where you need to look.  ;     OTOH, (1) why do you need to know the "memory position" 5 since (2) there are rules as to which slots pair each 7 sucessive memory option is installed in, as well as the ; ordering of larger-to-smaller options.  Not only should the 7 FCE who does the installation know all this and be able 5 to reconfigure the options as required, but there are 7 instructions in the system box that tell the same thing  when you finally do open it up.   7     The worst problem you may face is if the 4GB system 9 has four 1GB memory options, and the 6GB has two 2GB plus 6 two 1GB options.  In these cases, you'll need to order7 four 2GB options in the first case, and two 2GB options 0 in the second, to come up to a full 8GB in both.       -Ken --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfield ! F20 Automation VMS System Support  kenneth.h.fairfield#intel.com    ------------------------------  % Date: Mon, 28 Jan 2002 19:34:59 -0000 = From: "David McKenzie" <david.mckenzie@spitfire0.demon.co.uk> . Subject: Re: Alphaserver 4100: physical memoryB Message-ID: <1012246518.23617.0.nnrp-13.c1edba74@news.demon.co.uk>  @ Hmm don't have a system close, but Ithought clue might show this  F "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam> wrote in2 message news:3C55A686.A8BA638E@intel.com.nospam... > Fabio Cardoso wrote: > 
 > > Dear Sirsm > > / > > Is there a way to check the position of then) > > memory  in the Alphaserver 4100 slotsd& > > using ANALYZE/SYSTEM or DECEvent ? > >l* > > We have two 4100 (one with 4GB and the+ > > other with 6 GB) and we need to upgradek- > > both to 8 GB (max memory in this server).u > >l- > > So, I dont want to open the machines, andn( > > I need to make the specifications of! > > the current memory positions.s >d8 >     I don't believe you can find that information from; > running system.  What you want is the console output from 5 > powering on the system.  Is the system connected tol5 > PCM, or do you log the console output to a printer?p  > That's where you need to look. > = >     OTOH, (1) why do you need to know the "memory position"S7 > since (2) there are rules as to which slots pair eacht9 > sucessive memory option is installed in, as well as theo= > ordering of larger-to-smaller options.  Not only should ther9 > FCE who does the installation know all this and be able 7 > to reconfigure the options as required, but there arep9 > instructions in the system box that tell the same thingt! > when you finally do open it up.t > 9 >     The worst problem you may face is if the 4GB systemS; > has four 1GB memory options, and the 6GB has two 2GB plusG8 > two 1GB options.  In these cases, you'll need to order9 > four 2GB options in the first case, and two 2GB optionsS2 > in the second, to come up to a full 8GB in both. >n
 >     -Ken > --8 > I don't speak for Intel, Intel doesn't speak for me... >l > Ken Fairfield # > F20 Automation VMS System Supporte > kenneth.h.fairfield#intel.coms >  >s   ------------------------------  % Date: Mon, 28 Jan 2002 12:07:13 -0800oC From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam> . Subject: Re: Alphaserver 4100: physical memory0 Message-ID: <3C55AF71.A61C5AC6@intel.com.nospam>   David McKenzie wrote:   B > Hmm don't have a system close, but Ithought clue might show this       Good catch!e  @     Fabio, go into SDA ($ ANAL/SYSTEM) and do SDA> CLUE CONFIG .D That will list out the memory options on the first screen of output.   --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfieldr! F20 Automation VMS System Support  kenneth.h.fairfield#intel.comv   ------------------------------  % Date: Mon, 28 Jan 2002 15:27:35 -0500-% From: JF Mezei <jfmezei@videotron.ca>  Subject: Blade announcement2, Message-ID: <3C55B427.536A4DBA@videotron.ca>  H From a PR press release by Compaq which unveiled iots Blade stuff today:   ## Revolutionize Data Center   G   HOUSTON, Jan. 28 /PRNewswire-FirstCall/ -- Delivering on its adaptivehH   infrastructure strategy, Compaq Computer Corporation (NYSE: CPQ) todayR   announced the availability of the ProLiant BL e-Class, the first tier of the newM   Compaq ProLiant BL Line of industry-standard server blades. The ProLiant BLnP   e-Class is the first ultra-dense, server blade architecture engineered for theN   enterprise, enabling customers to pack 280 servers into a standard 42U rack.  oO  The ProLiant BL e-Class was designed for achieving efficiencies in traditionaleM   corporate data centers, as well as xSP environments, and extends customers'iO   critical IT resources through rapid deployment, virtual presence and control,eM   and plug-and- play serviceability. These new capabilities can save valuablenM   IT staff time and reduce new server deployment times from hours to minutes.  ##      H Yep, managing 280 windows servers will definitely save valuable IT staffA time....  Can you alt-ctrl-del all 280 servers at the same time ?l    N When one considers the RAM used by the windows kernel, having 280 instances ofJ windows surely starts to waste sufficient resources (both disk and ram) toM become less cost effective than a real Sun or IBM server ? (Alpha being dead, 1 it is pointless to present it as an alternative).c   ------------------------------  % Date: Mon, 28 Jan 2002 22:01:45 +0100 1 From: John McLean <mcleanj@swissonline.delete.ch>n Subject: Re: Blade announcementm5 Message-ID: <3C55BC39.43363797@swissonline.delete.ch>    JF Mezei wrote:n > J > From a PR press release by Compaq which unveiled iots Blade stuff today: >  > ## > Revolutionize Data Center  > I >   HOUSTON, Jan. 28 /PRNewswire-FirstCall/ -- Delivering on its adaptive J >   infrastructure strategy, Compaq Computer Corporation (NYSE: CPQ) todayT >   announced the availability of the ProLiant BL e-Class, the first tier of the newO >   Compaq ProLiant BL Line of industry-standard server blades. The ProLiant BLRR >   e-Class is the first ultra-dense, server blade architecture engineered for theP >   enterprise, enabling customers to pack 280 servers into a standard 42U rack.  6 Ultra-dense ?  Yep, that sounds like Compaq all right.   > J > Yep, managing 280 windows servers will definitely save valuable IT staffC > time....  Can you alt-ctrl-del all 280 servers at the same time ?A  0 This must be where Compaq "service" comes in ...  - They are probably using this new keyboard... r1 http://www.bbspot.com/News/2001/08/keyboard.html    G or maybe the one with just three keys with a piece of plastic to bridgeuH all three (Ctrl, Alt & Del) so you only need to press one big button...       John McLean0   ------------------------------  # Date: Mon, 28 Jan 2002 21:40:12 GMTt4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk> Subject: Re: Blade announcemento0 Message-ID: <3C55C44D.BEADF47A@blueyonder.co.uk>   JF Mezei wrote:h  n > (Alpha being dead,3 > it is pointless to present it as an alternative).R  > JF, please come back when HP, Compaq or whoever is the current2 owner of alpha stop taking orders for new systems.   regardsP --   Tim.Llewellyn@cableinet.co.uk     C Standard disclaimer applies. My views in no way represent those of  ! my employers or service provider.c   ------------------------------  # Date: Mon, 28 Jan 2002 22:38:32 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: Blade announcementt; Message-ID: <Ipk58.2613$LH1.3720@rwcrnsc52.ops.asp.att.net>e  A QuickBlade + Marvel + Adaptive Infrastructure software + Galaxy =J
 FIRE/ICE/WINDl  
 Charlie Matcor   ------------------------------  % Date: Mon, 28 Jan 2002 16:40:21 -0600a+ From: Christopher Smith <csmith@amdocs.com>e Subject: RE: Blade announcementoL Message-ID: <3B55D7F383B0D31197D9009027541CBF1170E1EE@cmiexch1.cmi.itds.com>   > -----Original Message-----; > From: Terry C. Shannon [mailto:terryshannon@mediaone.net]u  C > QuickBlade + Marvel + Adaptive Infrastructure software + Galaxy =  > FIRE/ICE/WINDt  , You're sure it's not EARTH/WIND/AND FIRE? ;)   Chriso    ! Christopher Smith, Perl Developerh Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");, 'a  J   ------------------------------  # Date: Mon, 28 Jan 2002 22:52:11 GMTy4 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: Blade announcementt; Message-ID: <vCk58.2707$LH1.4869@rwcrnsc52.ops.asp.att.net>h  8 "Christopher Smith" <csmith@amdocs.com> wrote in messageF news:3B55D7F383B0D31197D9009027541CBF1170E1EE@cmiexch1.cmi.itds.com... > > -----Original Message-----= > > From: Terry C. Shannon [mailto:terryshannon@mediaone.net]  >nE > > QuickBlade + Marvel + Adaptive Infrastructure software + Galaxy =  > > FIRE/ICE/WINDa >e. > You're sure it's not EARTH/WIND/AND FIRE? ;)  J Umm, that was a rock group circa 1970 or so (used to listen to their stuffL whilst enjoying my all-expense-paid vacation in Viet Nam). FIRE/ICE/WIND areE codenames for various instantiations of the post-Marvel, post-Cyclonea" Blade-and-switched-fabric servers.   ------------------------------  # Date: Tue, 29 Jan 2002 02:25:56 GMTg* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: Blade announcement C Message-ID: <UKn58.173366$TC1.12172744@bin5.nnrp.aus1.giganews.com>e  ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in messagel5 news:vCk58.2707$LH1.4869@rwcrnsc52.ops.asp.att.net...W >a: > "Christopher Smith" <csmith@amdocs.com> wrote in messageH > news:3B55D7F383B0D31197D9009027541CBF1170E1EE@cmiexch1.cmi.itds.com...  > > > -----Original Message-----? > > > From: Terry C. Shannon [mailto:terryshannon@mediaone.net]t > >aG > > > QuickBlade + Marvel + Adaptive Infrastructure software + Galaxy =a > > > FIRE/ICE/WINDn > >o0 > > You're sure it's not EARTH/WIND/AND FIRE? ;) > L > Umm, that was a rock group circa 1970 or so (used to listen to their stuffJ > whilst enjoying my all-expense-paid vacation in Viet Nam). FIRE/ICE/WIND are G > codenames for various instantiations of the post-Marvel, post-Cyclone $ > Blade-and-switched-fabric servers.  K You keep referring to them, but I've yet to be able to find any descriptiverF material.  Do you have a pointer to anything more than marketing fluff# (though I haven't even found that)?r   - bill   ------------------------------    Date: 28 Jan 2002 22:48:25 -0800( From: steve@merilus.com (Steve Hemenway)* Subject: Blade Server for Network Security< Message-ID: <8faf9401.0201282248.b4939e8@posting.google.com>  9 FireCard - security on the leading edge... of the blade!    E Blade Server technology is new to the industry but not to us! We have C been developing the technology for the past two years and have been0B actively marketing our products for well over a year. In fact, our> award winning technology is already being used by governments,< schools, health services, manufacturing, financial services,F professional services, research companies, accounting services, ISP's,( media and high profile federal agencies.  D Merilus FireCard is a shining example of how blade server technologyB can benefit your business! FireCard is a complete network securityD system engineered onto a system independent PCI card and designed toC function in any workstation or server on your network. Just plug itaE into any available PCI slot and FireCard will be ready to protect ando optimize your network.  A FireCard comes with a sophisticated state-of-the-art firewall, ansD IPSec compliant VPN endpoint, its own Global Management System and aE huge assortment of network security and management features. FireCard"A has moved ahead of the industry by offering more functionality ataD lower prices than any other security product in the market. Here areC just a few of the advantages the Merilus FireCard offers over othert+ blade servers and network security systems.   @ Fits into regular PCI bus - Can be installed into just about any server or workstation E No additional software to buy or install onto the blade - Comes ready  to configureE No management software to buy - Comes with it's own Global Managementk System? Developed over a year ago - Our technology is tried and tested t9 Draws only power from the PCI bus for increased security hE Comes with no seat licensing to lower costs and make network planningr easier' Already available through distribution pB MSRP - Just 765USD! - Priced less that most "Naked" Blade Servers   # MORE INFORMATION AT WWW.MERILUS.COMe   ------------------------------    Date: 29 Jan 2002 06:19:10 +0800, From: Paul Repacholi <prep@prep.synonet.com>: Subject: Re: Can PC's read OpenVMS index-sequential files?- Message-ID: <87ofjedspt.fsf@prep.synonet.com>l  / Kilgallen@SpamCop.net (Larry Kilgallen) writes:t  a > In article <9946d62e.0201280311.59679c3a@posting.google.com>, br@b-con.dk (Bendix Riis) writes:a  7 > > Does anyone know a method for PC's to read (by key) + > > index-sequential files on a VMS system?r   C > Perhaps the DECnet support in Pathworks provides this.  Certainlye? > they can be read by key over DECnet from a remote VMS system.o  ! Well, DECnet-DOS V1 could do it. n   -- n< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.g@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Mon, 28 Jan 2002 17:47:59 -0800s+ From: "xenman" <xenman@sprynet.nospaam.com> : Subject: Re: Can PC's read OpenVMS index-sequential files?2 Message-ID: <a34v0k$lc5$1@slb0.atl.mindspring.net>  * Bendix Riis <br@b-con.dk> wrote in message7 news:9946d62e.0201280311.59679c3a@posting.google.com...fF > Does anyone know a method for PC's to read (by key) index-sequential > files on a VMS system? >i
 > Bendix Riis  > B-cona > email: br@b-con.dk  - I suggest that you try the ConnX product from $ CONNX Solutions.  Try: www.connx.com   ------------------------------  % Date: Mon, 28 Jan 2002 21:48:28 +0100 1 From: John McLean <mcleanj@swissonline.delete.ch>i  Subject: Re: Capellas conference5 Message-ID: <3C55B91B.73579B58@swissonline.delete.ch>e   "John N." wrote: >  > I got it.   E Well, you should have taken something for it.  Prozac would be a goodu idea...r  H A very gentle interview from Lehmans who, by the way, bought $10 millionF worth of Compaq stock back in Q3 to add to the $2 million they alreadyH held.  Now maybe it is a different part of Lehman but it sure looks like@ they had a vested interest in letting Mike advertise the merger.   Some snippets ..  H Capellas quoted plenty of figures from memory and said things like a lotE of people don't realize that Compaq power the top ten stock exchangeslD and that we are a leader in supercomputing.  So, why don't they knowG Mike ?  Is it your crap marketing managers who won't allow the high-endo* segments to have a voice, or is it you ???    E I thought he was actually going to use the "V" word when he mentionedo< clustering ... but it was only to talk about supercomputing.    G Not only did he completely avoid mentioning VMS but he talked about thetF future being "standard building blocks and standard operating systems" with Compaq adding value.o    A It's not hard to see why VMS is in the razor-thin marketing nicheyF between Tadem and Tru64.  It's also not hard to see why MC ignores any correspondence about VMS.9  E He's wants to be inferior standard (sorry, industry standard) and addoF value.  Sorry Mike but there's plenty of people in that game already. B You'll have to work on razor-thin margins because of the amount ofH competition and all it needs is one bad IT-spending year and CPQ will go	 belly-up.r  G So Mike, you like the idea of being able to sell right across the board(D - hand-helds to fault-tolerant (which BTW aren't standard).  But youG admit in the interview that you don't expect someone to buy an iPaq andhC a Tandem.  It looks more like you stand a very good chance of beingnF beaten at every level of "industry standard-ness" simply because there@ are companies who cater to specific markets and do it very well.  9 Mike has made it damn clear where he stands in regards tooG whats-its-name.  I am seriously starting to wish for a new owner ratherlF than have the good ship VMS go rusty while it is tied up at the docks.    + For your amusement gentle reader, check outcM http://www.idg.net/idgns/1999/07/22/UPDATE2CompaqNamesCapellasPresident.shtmlrF which reports the appointment of Capellas (and is dated 22 July 1999).  - Some quotes (and jokes) from that article ...o    G One analyst said Capellas seems to have the "meat-and-potatoes" abilitytB to run a company, but questioned whether he has the "star quality"= necessary to make Compaq stand out in the eyes of the public.p  F "Has he got the necessary presence to be a leader in an industry whereC presence counts?" wondered Roger Kay, a desktop market analyst withpF International Data Corp. (IDC) in Framingham, Massachusetts. "He's gotE the meat-and-potatoes basic industry, daily decision-making power youlH need to do the job, but the other piece of it is the public persona," he said.-  F "It's increasingly important to have a real character at the top," KayF added. "If it's not a founder, like Steve Jobs or Bill Gates, then youC need an industry leader with the polished resume and a sort of star1	 quality."n ...eC In listing the products that Compaq will continue to support movingHB forward, which include OpenVMS, Tru64 Unix  and Tandem, mention ofH Microsoft's Windows NT software was conspicuously absent, Enderle noted.  E "What we're hearing from Microsoft is that Compaq's not getting along H with them ... (Capellas) appears to be a part of that thinking," Enderle said.e ....E "While we clearly have challenges in front of us, we are on the rightnD path again, and this company can be a truly great company," Capellas said.  ...kG "We want to offer services with everything we do, as a component of our = consumer business as well as the enterprise," Capellas added.  ...eH Compaq stock was down 1 point to $25 on the Nasdaq stock exchange today.    & meat-and-potatoes or potato-headed ??  charisma or charisma-bypass ??' right path or lost in the wilderness ??i "You want services with that ?"t< Compaq stock price was at $25 compared to the current $11.70       John McLean    ------------------------------  # Date: Mon, 28 Jan 2002 22:41:02 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>  Subject: Re: Capellas conference+ Message-ID: <2sk58.3065$MR3.4978@rwcrnsc53>l  > "John McLean" <mcleanj@swissonline.delete.ch> wrote in message/ news:3C55B91B.73579B58@swissonline.delete.ch...t >b >n > "John N." wrote: > >d
 > > I got it.e >oG > Well, you should have taken something for it.  Prozac would be a goodt	 > idea...o   Celexa has its advocates, too.   > J > A very gentle interview from Lehmans who, by the way, bought $10 millionH > worth of Compaq stock back in Q3 to add to the $2 million they alreadyJ > held.  Now maybe it is a different part of Lehman but it sure looks likeB > they had a vested interest in letting Mike advertise the merger. >  > Some snippets .. >sJ > Capellas quoted plenty of figures from memory and said things like a lotG > of people don't realize that Compaq power the top ten stock exchangeseF > and that we are a leader in supercomputing.  So, why don't they knowI > Mike ?  Is it your crap marketing managers who won't allow the high-endw, > segments to have a voice, or is it you ??? >s >PG > I thought he was actually going to use the "V" word when he mentioneda> > clustering ... but it was only to talk about supercomputing. >a >oI > Not only did he completely avoid mentioning VMS but he talked about thetH > future being "standard building blocks and standard operating systems" > with Compaq adding value.r >f  J This is indeed troubling. MC obviously can read spreadsheets and financialB roll-ups, hence he MUST know where the profit margins come from...   ------------------------------    Date: 29 Jan 2002 05:41:00 +0800, From: Paul Repacholi <prep@prep.synonet.com>M Subject: Re: Capellas redefines Industry Standard to mean Windows *and* Linuxa- Message-ID: <87wuy2duhf.fsf@prep.synonet.com>r  ' JF Mezei <jfmezei@videotron.ca> writes:e  F > This is actually great for hobbyists. Within a few years, a WildfireA > class system will be available for a few dollars (actually, the-4 > movers may charge more than the machine is worth).  8 I'll put one next to the KL. KL-10E, $2. Moving, $100...   -- -< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.e@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Mon, 28 Jan 2002 19:18:26 GMTw4 From: "Terry C. Shannon" <terryshannon@mediaone.net>( Subject: Re: Compaq financial conference+ Message-ID: <6uh58.1584$9U4.1554@rwcrnsc54>i  2 "JF Mezei" <jfmezei@videotron.ca> wrote in message& news:3C559BD5.61659D88@videotron.ca... > Jan-Erik Sderholm wrote:- > >a? > > You would only get this error if some *other* user had been0< > > changing *the very same record* at the same time as it'sA > > beeing processed at one place, may it be a ticket, a seat, or 
 > > whatever.t >aH > When you book a flight, it is not only your reservation record that isE > modified, but also the fligh'ts inventory of available seats in the2
 particular( > booking class applicable to your fare. >cF > And the flight's inventory gets constant updates both from passengerE > reservations and the airline itself which will adjust the inventoryk	 accordingmH > to demand, seat sales etc. Some airlines such as AA have sophisticatedL > programs that analyse the inventory fo seats on each flight and adjust theL > number of available seats for various booking classes (fares) depending on2 > many variables and past history for that flight. >rK > Similarly, fares are also constantly adjusted on a real time basis. Prior2 toK > e-tickets, I had an intesting experience once. I booked a flight over theC net,H > called the airline to confirm the fare (the on-line systems were never reallyJ > good at quoting the right taxes in canadaian dollars). Then hopped on my bikeJ > for the 15 minute ride to airport to pickup my ticket. By the time I got. > there, the fare had changed by $25 (higher).  J How true! I've seen that happen while logged into an airline site. Look atL one fare, wander off for a mere 60 seconds to check the fare on an alternateK site, and when you return to the original fare, lo and behold: it's gone upn
 $50 or so.  I As an aside, when you decide to book a flight online, do be sure to check F multiple sources. Last year I needed to take a trip from BOS to MIA. I? checked Expedia, and they showed me an American flight at $400.n  $ The same flight on Expedia was $300.  ( The same flight on www.ual.com was $198.  E This is a level of wagering rarely achieved by the Wall Street Casinoe
 Analysts. ;-}.   Caveat Booker!  
 charlie matcop   ------------------------------  # Date: Mon, 28 Jan 2002 19:19:00 GMTt* From: "Bill Todd" <billtodd@metrocast.net>( Subject: Re: Compaq financial conference@ Message-ID: <Euh58.87701$%b.5239158@bin4.nnrp.aus1.giganews.com>  L "Jim Johnson" <Jim.Johnson@software-exploration.nospam.com> wrote in message) news:3c550b70.1143003@news.demon.co.uk...e > JF,V >9C > I cannot emphasize enough that you really need to read Lampson at , > least, and almost certainly Gray & Reuter. >oH > You're describing an example of the Byzantine Generals problem.  Ayup,C > the last message can always get lost.  And if it can get lost anddB > everything work, then it didn't need to be there.  But, the last% > message can always get lost, so ...o  D Actually, lost messages are a common occurrence and any (reasonable)E distributed transaction mechanism handles them as a matter of course. L Byzantine faults (at least according to Gray & Reuter) involve not just lostK but possibly malicious messages, and most systems try to avoid (rather thanpC deal with) them via a combination of fail-fast (to varying degrees)m< hardware/software plus at least limited trust relationships.   - bill   ------------------------------  % Date: Mon, 28 Jan 2002 14:39:49 -0500 % From: JF Mezei <jfmezei@videotron.ca> ( Subject: Re: Compaq financial conference, Message-ID: <3C55A8F8.28D67E93@videotron.ca>   "Terry C. Shannon" wrote:lA > checked Expedia, and they showed me an American flight at $400.i > & > The same flight on Expedia was $300. > * > The same flight on www.ual.com was $198. > G > This is a level of wagering rarely achieved by the Wall Street Casino  > Analysts. ;-}e     Flight != fare.   J And web based systems don't show you real-time availability of seats. ThatL gets updated from time to time whenever the airline sends out updates to the various CRS systems.  N So when you check on one site, it may not have access to the inventory that isM the same "age" as the inventory another system shows you. And that is why youiN see "price" differences. And if you call the airline (or use the airline's webK system) you are more likely to have access to the "live" inventory and havee" greater access to the lower fares.  N But there are times where this works against you because some web based systemM may still think there are seats available at a low fare and allow you to bookoI it and they are bound to honour the fare they quoted you because they nowa issue tickets when you book it.e  M In the good old days of easy sabre, you would see the actual fare basis (suchiL as BKPAP4) and then book your flight to have that one. And you could force aM reservation in a specific booking class, and see if the airline would confirmYF it or not. (in case the airline's real time inventory would have seats" available while the CRS wouldn't).    J The newer dumbed down systems just try to find you "a" fare that seems theH lowest without really telling you what you're getting.  Many fares don'tJ "autoprice" and those new web based systems won't offer those fares to youK because the computer won't see it as a valid fare for your itineray and youo8 have no way of forcing it like we used to be able to do.   ------------------------------  # Date: Mon, 28 Jan 2002 20:02:14 GMTy? From: Jim.Johnson@software-exploration.nospam.com (Jim Johnson) ( Subject: Re: Compaq financial conference/ Message-ID: <3c55ae28.4047339@news.demon.co.uk>e  E Yes, true enough.  That was a detail I didn't really want to get intol
 just then.   Jim.  F On Mon, 28 Jan 2002 19:19:00 GMT, "Bill Todd" <billtodd@metrocast.net> wrote:   >iM >"Jim Johnson" <Jim.Johnson@software-exploration.nospam.com> wrote in messaget* >news:3c550b70.1143003@news.demon.co.uk... >> JF, >>D >> I cannot emphasize enough that you really need to read Lampson at- >> least, and almost certainly Gray & Reuter.t >>I >> You're describing an example of the Byzantine Generals problem.  Ayup,2D >> the last message can always get lost.  And if it can get lost andC >> everything work, then it didn't need to be there.  But, the lasta& >> message can always get lost, so ... > E >Actually, lost messages are a common occurrence and any (reasonable).F >distributed transaction mechanism handles them as a matter of course.M >Byzantine faults (at least according to Gray & Reuter) involve not just lostiL >but possibly malicious messages, and most systems try to avoid (rather thanD >deal with) them via a combination of fail-fast (to varying degrees)= >hardware/software plus at least limited trust relationships.  >s >- billo >c >h >    Jim Johnson  Software Exploration, Ltd.) (remove '.nospam' from the reply address)t   ------------------------------  # Date: Mon, 28 Jan 2002 20:03:47 GMT ? From: Jim.Johnson@software-exploration.nospam.com (Jim Johnson)a( Subject: Re: Compaq financial conference/ Message-ID: <3c55ae5c.4099194@news.demon.co.uk>   B Mitigate is the right word here.  Actually, the normal way to dealC with long commit-coordinator outages is through heuristic outcomes.DB It doesn't add extra messages in the normal case, and still allows. businesses to avoid infinite in-doubt periods.   Jim.  F On Mon, 28 Jan 2002 18:03:09 GMT, "Bill Todd" <billtodd@metrocast.net> wrote:   >sM >"Jim Johnson" <Jim.Johnson@software-exploration.nospam.com> wrote in messageh* >news:3c550b70.1143003@news.demon.co.uk... >f >... > I >> Two phase commitment 'solves' this problem at some cost -- namely thath5 >> there is an in-doubt period that may be unbounded.v >bD >And, of course, there are ways to mitigate even that risk when it'sD >necessary to do so.  Unbounded comm failures can be avoided through< >appropriate levels of interconnection redundancy, unboundedM >commit-coordinator failures by having a fail-overable Commit mechanism, etc.i >  >- billI >e >r >    Jim Johnsone Software Exploration, Ltd.) (remove '.nospam' from the reply address)b   ------------------------------  % Date: Mon, 28 Jan 2002 21:55:07 +0100e1 From: John McLean <mcleanj@swissonline.delete.ch> ( Subject: Re: Compaq financial conference4 Message-ID: <3C55BAAB.2631334@swissonline.delete.ch>   JF Mezei wrote:l >  ...s > L > And web based systems don't show you real-time availability of seats. ThatN > gets updated from time to time whenever the airline sends out updates to the > various CRS systems.  C If you go through the airlines' websites, most put you onto Amadeusn< booking system and it seems pretty good.  It certainly shows< availability and if you tweak things a bit you can often get< point-to-point services with a number of different airlines.  F I've also accessed some of the highly-rated US online booking services< and found them quoting a lowest price of USD 3000 for ZurichD (Switzerland) to Melbourne (Australia) return.   I can usually buy aE return out of Australia for about AUD 2000 or CHF 1800, ie. about USDa 1300, not 3000 !     John McLean    ------------------------------  # Date: Tue, 29 Jan 2002 04:09:35 GMTy& From: "C.W.Holeman II" <cwhii@ACM.org>Y Subject: Compaq site has link to Gradient Technologies DCE but not for VMS or NSK. NSK.NSn' Message-ID: <3C562076.E6768DFC@ACM.org>    On page:   	Compaq - Operating Systemsa6 	http://www.compaq.com/products/software/index.html#os   There is a link for DCE:  ( 	DCE [Distributed Computing Environment]$ 	http://www.tru64unix.compaq.com/dce  J which states that Gradient Technologies has taken "over responsibility for development,K maintenance and support of DCE on Tru64 UNIX and Windows"  and that "DCE onm OpenVMSe9 and Compaq NonStop Kernel are not part of the agreement".   H There is no other link for DCE on the page, just a link for the Gradient Technologies& product but not for the one they sell!   -- i C.W.Holeman II  cwhii@acm.org				http://emle.org- http://also.as/cwhii		http://JulianLocals.comn   ------------------------------  % Date: Tue, 29 Jan 2002 06:59:17 +01001, From: Didier Morandi <Didier.Morandi@gmx.ch>+ Subject: Re: Cross Compiler to SUN/Solaris?n% Message-ID: <3C563A35.94945E8@gmx.ch>y   Larry Kilgallen wrote: > W > In article <3C54DAA4.A199BB38@gmx.ch>, Didier Morandi <Didier.Morandi@gmx.ch> writes: N > > So, if I understand you well, and it seems easy, my new question would be: > >gP > > Solaris equivalences to the RTL routines? (and SS calls, and SMG and and...) > L > Highly unlikely, since Solaris is C-centric rather than using descriptors.J > Unless some third party specifically markets a VMS emulation package forI > Solaris (as distinguished from VAX or Alpha emulation), they are betteru > off to stick with VMS.  K Larry, as all of you here, my spirit - that I learned at DEC - is "CustomereM Satisfaction". The (known) problem is that the "satisfaction" in a Customer'srM mouth is not "what is the best for him", it is "what has been decided by that . kind of people who fired KO one day". You see?  P So, if the Customer decided to move from VMS to SUN, or to Winsh*t, my job is to help him to move.c   Sad, but reality.-   D. -- eH   ----------------------------------------------------------------------E MORANDI Consulting.  WEB: http://Didier.Morandi.Free.fr/index_us.html E Pflanzschulstrasse 53, 8004 Zurich, Switzerland. GSM: +41 79 705 4670i/ 19, chemin de la Butte, 31400 Toulouse, France.s  H Disaster Recovery Plans, Computer Security Audits, DEC OpenVMS ExpertiseH On parle franais Man spricht Deutsch se habla Castellano English spoken   ------------------------------  % Date: Mon, 28 Jan 2002 12:17:14 -0800n From: Pearlz <pearlz@efn.org>a= Subject: data-entry/keystroke job... howto display to screen?sD Message-ID: <Pine.GSU.4.21.0201281215440.3332-100000@garcia.efn.org>  < I have a file that is performing a data-entry type operation8 on 650K records, all with the same keystrokes.  How do I: display this to the screen?  It would look like a ghost is doing the work.e   pearlz pearlz@efn.org   ------------------------------  % Date: Mon, 28 Jan 2002 21:08:45 -0500H1 From: Michael Austin <maustin@firstdbasource.com>rA Subject: Re: data-entry/keystroke job... howto display to screen?l2 Message-ID: <3C56042D.33837ECE@firstdbasource.com>  
 Pearlz wrote:  > > > I have a file that is performing a data-entry type operation: > on 650K records, all with the same keystrokes.  How do I< > display this to the screen?  It would look like a ghost is > doing the work.> >  > pearlz > pearlz@efn.org  F What is the "file"? DCL? C? FORTRAN?  You have to give us more to workC with here...  Be a bit more specific.  Also. What OS? what version?o  7 If this is DCL, use $set verify or "$write sys$output" x5 If C, then use "debug" type statements such as printfr --     Regards,   Michael Austin7 First DBA Source, Inc. -- http://www.firstdbasource.coma Sr. Consultant   ------------------------------  % Date: Mon, 28 Jan 2002 20:24:07 +0000d/ From: Tony  Barker <tony@dartfactordata.ltd.uk>n Subject: DCE Programming6 Message-ID: <B87B63E7.8449%tony@dartfactordata.ltd.uk>  J I wonder if anyone can help me with my DCE problem.  I have an applicationJ happily running on ovms vax 7.1 and dce v1.4 eco 3.  I have been trying to/ port this to an alpha ovms v7.2-1 dce 3.0 eco 1sL The problems start when, after getting a valid and certified login context IH call this routine sec_login_inquire_net_info.  On the vax it returns theJ cell name, principal and group names and uuids, but on the alpha the groupL name is null but the num_groups field shows the correct value.  I've checkedG the returned uuid for the group using other sec_rgy calls and that doesCL return the expected group name.  So what happened to the group name and does4 it make any difference if I've got the correct uuid?   Thanks in advancei   Tony   ------------------------------    Date: 28 Jan 2002 13:28:02 -0800( From: Craig.Dedo@abbott.com (Craig Dedo)' Subject: DCOM Programming Under OpenVMSa= Message-ID: <83afd6f7.0201281328.2de4b174@posting.google.com>e  C I am working on a project that will require the development of DCOMa? server objects under OpenVMS.  The client objects will be underaE Windows NT.  The client DCOM server object will need to wrap a singlesA function in an interface that will be called from the client DCOMs object.e  .     I have already been doing some homework.  ? * I have already read the OpenVMS Connectivity Developer Guide.r6 * I already searched the Compaq OpenVMS DCOM Web site.? * I am reading Inside COM by Dale Rogerson.  I am about halfway  through this right now.g9 * I am also reading DCOM Explained by Rosemary Rock-EvanstC * I also have acquired two other books, both by Don Box:  Essentiale@ COM and Effective COM: 50 Ways to Improve Your COM and MTS-Based Applications  A     I would like to benefit from the experience of anyone who hassA already developed DCOM server objects under OpenVMS.  Please helpnB answer the following questions.  I also would appreciate any other0 relevant advice that people would like to offer.  B 1.  What, if any, simplified step-by-step procedures are there forD wrapping existing functions in a DCOM server object?  Where are such procedures documented?  @ 2.  From what I have read so far, DCOM programming appears to beF hideously complex and done at an extremely low level, much like most CC language programming.  Are there any higher-level interfaces to theD@ DCOM procedures?  E.g., something like the LIB$ Run Time Library= procedures as opposed to the SYS$ System Services procedures?n  E 3.  Must the developer do the programming in C++?  Or, is it possiblea< to create DCOM server objects in OpenVMS using other OpenVMS# programming languages?  If so, how?e  D 4.  Are there any good examples of how to create DCOM server objects under OpenVMS?  If so, where?   @ 5.  What are the best resources for learning how to use MIDL for creating interface definitions?   E 6.  Other than the books I listed, what other highly useful resourcesC should I consult?o  F 7.  I tried searching Usenet about DCOM and did not come up with much.@  What, if any, Usenet newsgroups are there that are suitable for dicussion about DCOM?d  A     Thanks in advance for any help you can give me.  Speaking forh' myself and not for Abbott Laboratories."   ------------------------------  # Date: Mon, 28 Jan 2002 19:26:53 GMTn! From: Ernest <wmozart5@yahoo.com>-- Subject: How to get/put the symbol list in C?mC Message-ID: <1Ch58.170125$TC1.11801328@bin5.nnrp.aus1.giganews.com>s  F Is the only way to get the list of *all* defined symbols (the closest J thing DCL has to environment variables, AFAIK) to run an external program G ("SHOW SYMBOL *") and parse the output?  I'm using Compaq C/C++ 6.4 on  K Alpha VMS 7.2.  I noticed the "environ" global exists, but just contains a rF few contrived variables that aren't really symbols or logicals at all.  G Assuming I can get the full list captured into a C data structure, and rH add/modify/delete entries as I need, how do I then output these symbols  for use by a child process?a   Thanks for any tips, Ernest   ------------------------------  # Date: Tue, 29 Jan 2002 00:39:42 GMTo From: danco@pebble.org1 Subject: Re: How to get/put the symbol list in C?h- Message-ID: <slrna5brqd.ngh.danco@pebble.org>s  C In article <1Ch58.170125$TC1.11801328@bin5.nnrp.aus1.giganews.com>, 
 Ernest wrote:i   > Thanks for any tips, > Ernest  D Do you honestly consider (local?/global?) DCL CLI symbols  to be the% equivilent of the UNIX "environment?"h  5 They're _somewhat_ similar, but by no means the same.i  > What you probably want is LIB$GET_SYMBOL() / LIB$SET_SYMBOL().B See HELP RTL LIB$ LIB$GET_SYMBOL and HELP RTL LIB$ LIB$SET_SYMBOL.B So far as I know, you can't do wildcard DCL symbol lookups via the DCL callable interface though.  C (On a related note, I'd take OpenVMS logicals, logical name tables,t( etc. over the UNIX environment any day.)   - Danu   ------------------------------  # Date: Tue, 29 Jan 2002 04:21:55 GMTo- From: "John E. Malmberg" <wb8tyw@qsl.network> 1 Subject: Re: How to get/put the symbol list in C? * Message-ID: <3C562783.7000906@qsl.network>  
 Ernest wrote:h  H > Is the only way to get the list of *all* defined symbols (the closest L > thing DCL has to environment variables, AFAIK) to run an external program I > ("SHOW SYMBOL *") and parse the output?  I'm using Compaq C/C++ 6.4 on -M > Alpha VMS 7.2.  I noticed the "environ" global exists, but just contains a eH > few contrived variables that aren't really symbols or logicals at all. > I > Assuming I can get the full list captured into a C data structure, and fJ > add/modify/delete entries as I need, how do I then output these symbols  > for use by a child process?     H Unless you specify otherwise, all symbols and process logicals from the E parent process are copied to the child.  No special action is needed.   H You can use the getenv() call to look up both logical names and symbols.   See:  C http://www.openvms.compaq.com/commercial/c/5763p029.htm#index_x_820t    F If you need to look up the value of a logical name in a loop, I would 1 recommend using SYS$TRNLNM() instead of getenv().a  G The getenv() function is meant more for looking up values at the start  
 of a program.l  I Looking up a wild card list of all defined symbols is not a trivial task m from a user program.   -Johnt wb8tyw@qsl.network Personal Opinion Onlyt   ------------------------------  # Date: Tue, 29 Jan 2002 06:56:31 GMTr From: danco@pebble.org1 Subject: Re: How to get/put the symbol list in C?-- Message-ID: <slrna5chsu.non.danco@pebble.org>   B In article <3C562783.7000906@qsl.network>, John E. Malmberg wrote:  K > Looking up a wild card list of all defined symbols is not a trivial task h > from a user program.  I Is it even possible without looking at undocumented/unsupported locationsiG within the DCL CLI in P1 space (or wherever it may be mapped on Alpha)?    - Dans   ------------------------------    Date: 28 Jan 2002 20:12:47 -0800/ From: chris@applied-synergy.com (Chris Scheers)a( Subject: Re: HP 35470 DAT drive question= Message-ID: <754a27c1.0201282012.15012775@posting.google.com>   3 "Hans Vlems" <hvlems@zfree.co.nz> wrote in message mL > >>When the INIT command is issued the right LED flashes approx. four times >  a > >>second.gJ > >>I run VAX/VMS 7.2, no patches on a VAXstation 3100/GPX, hardware model >  typer > >>59.d > >>3 > >>So is this a Unix only device, or is it broken?  > >rC > The SCSI bus (the SCSI-B bus to be exact) is properly terminated.eJ > There is an RZ24L and a CD-ROM drive on the same bus and both work fine. > The HP drive has 5 jumpers:t > 3 for SCSI address > 1 for pwr term (not jumpered)fG > The fifth, not in use either (I'm not at home and forgot what it was)  > * > There are 8 dip switches on the bottom. # > 1,2: HP advises not to touch thems > 3 : Media recognition system	 > 4,5 : ?r > 6 : performance option > 7 : ? (parity ?)Q > 8 : DEC position, effectively switches to SCSI-1 handling of the space command.l > $ > I've tried the following settings: > all switches "ON", except:
 >  - 8 OFF >  - 3 and 8 OFF >  - 3, 6 and 8 OFFl >  - 6 and 8 OFF >  - 3, 5, 6, and 8 OFFu
 >  - 3 OFF > K > The last test showed a different result: the LEDs turned amber instead of  > green.F > Every dip switch change was done with the power off on the drive and > a reboot of VMS.    , The HP35470A WILL work on a VAXstation 3100.  C The firmware 11.09 is OK.  I think 12.09 is also available.  (There 3 were some problems back around the 9.09 timeframe.)-  E My 35480A (a 35470A with a compression chip) has its switches set to:t   Off: 1, 3, 6, 7, 8 On: 2, 4, 5h  @ 1 and 2 affect compression and aren't really used on the 35470A.  D (If I remember correctly, the setting I have for 1 & 2 says to writeE uncompressed tapes, but read both compressed and uncompressed tapes.)2  A If MRS is enabled and you use a tape that is not MRS labeled, the " drive will write protect the tape.  @ A flashing LED could indicate dirty heads.  Try a cleaning tape.  4 If it still doesn't work, the drive is probably bad.  
 Good luck!   ------------------------------  # Date: Mon, 28 Jan 2002 23:52:02 GMTt6 From: "Andy Bustamante" <a_c_bustamante@earthlink.net>* Subject: Re: I/O Performance on Alpha 4100D Message-ID: <Cul58.2675$ks5.285448@newsread1.prod.itd.earthlink.net>  C I have cluster with HG80's and they provide an amazing boost in i/oaG performance.  Yes they are expensive, but they also allow you configuren= reliable disk configurations in addition to the i/o benefits.n  H Another option may be to attempt to add enough "low cost" disk to spreadH your I/O out on multiple spindles.  The KZPAC controllers mentioned haveJ been very reliable for us, but don't provide the flexibility, particularly when a physical disk fails.r  I In either case, you have optional controller cache available.  There manybE options, but you need to look at what you can afford both in terms of. expenditure and downtime.e   --   Andy Bustamante-( remove the ascii-95's to reply by e-mail      . "BrianNFO" <briannfo@aol.com> wrote in message3 news:20020125133554.01384.00000540@mb-ba.aol.com... I > I haven't posted here in a while, but I'm looking to speed up I/O on an  AlphahD > 4100 5/400 running VMS 7.2-1.  Most of the devices are things like RZ40-VA, andH > I believe the VA indicates narrow scsi?  I don't know exactly what PCI > controllers are in use.  >iI > Applications on this system are I/O bound, and I would think I can make8 someL > big improvements through either a caching controller, or a combination new+ > controller, wide shelves and wide drives.e > ' > Any suggestions would be appreciated!o >o > Brianf >.   ------------------------------   Date: 28 Jan 2002 20:53:58 GMT' From: robinm@rpi.edu (Michael Robinson)e Subject: In Need of 9-track VMS , Message-ID: <a34dp6$kge$1@newsfeeds.rpi.edu>   Hello,?    I'm a proud owner of a VAX 8530 system.  Unfortunately, I ameD without boot media.  Specifically, while I have a working StandaloneE BACKUP set, I have nothing to BACKUP from!  The machine has ethernet, G and supposedly a TK50 interface, but I can't connect the tape that way.oD I seem to be lacking the cabling to do that.  (If anyone can tell meH how to make a cable that goes from the VAXBI headers of the DEBNA T1034,4 to a TK50 drive, that would probably help a lot too)D What I do have is a TU81+ and interface, cables, etc.  But I have noE tape...  Sooo, could anyone cut me a tape with VMS (any version above E v5.0 probably will do)?  I'm located in Troy, NY at RPI.  I'll gladlyo pay shipping...e   Michael Robinson robinm at rpi dot edu  RPI Electronics Club Vaxherd     ------------------------------  % Date: Mon, 28 Jan 2002 23:42:08 -0500t0 From: "Rob L Lyons" <rob.lyons@resilientsys.com># Subject: Re: In Need of 9-track VMSa+ Message-ID: <a35967$3cs$1@bob.news.rcn.net>e  A >    I'm a proud owner of a VAX 8530 system.  Unfortunately, I amtF > without boot media.  Specifically, while I have a working StandaloneG > BACKUP set, I have nothing to BACKUP from!  The machine has ethernet,rI > and supposedly a TK50 interface, but I can't connect the tape that way. F > I seem to be lacking the cabling to do that.  (If anyone can tell meJ > how to make a cable that goes from the VAXBI headers of the DEBNA T1034,6 > to a TK50 drive, that would probably help a lot too)  D The tape interface into the DEBNT/DEBNA was not supported although IF saw one working in a development lab once.  To avoid wasting time, you! can forget about finding a cable.t  	 Rob Lyonsd www.resilientsys.com   ------------------------------    Date: 29 Jan 2002 06:02:01 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Installing GNV - Message-ID: <87sn8qdtie.fsf@prep.synonet.com>,  % "Tom Linden" <tom@kednos.com> writes:   - > go to http://gnv.sourceforge.net/readme.htmn  o9 > Get the sources and install you will get the following:r ...n  ' Did that. Is there a INSTALL somewhere?n   -- h< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.o@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Mon, 28 Jan 2002 14:46:30 -0800 # From: "Tom Linden" <tom@kednos.com>c Subject: RE: Installing GNV,9 Message-ID: <CIEJLCMNHNNDLLOOGNJIIEIGEAAA.tom@kednos.com>f   > -----Original Message-----; > From: prep@k9 [mailto:prep@k9]On Behalf Of Paul Repacholia( > Sent: Monday, January 28, 2002 2:02 PM > To: Info-VAX@Mvb.Saic.Comi > Subject: Installing GNVs >  > ' > "Tom Linden" <tom@kednos.com> writes:n > / > > go to http://gnv.sourceforge.net/readme.htma >  p; > > Get the sources and install you will get the following:l > .... > ) > Did that. Is there a INSTALL somewhere?a  < No, setup.  Just follow the instructions on that page, viz.,       $ @[.gnv]setup.com
     $ bash >  > --  > > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda.iB >                                              West Australia 60760 > Raw, Cooked or Well-done, it's all half baked.H > EPIC, The Architecture of the future, always has been, always will be. >    ------------------------------  % Date: Mon, 28 Jan 2002 15:18:18 -0500l2 From: "Sue Skonetski" <susan.skonetski@compaq.com>- Subject: Intel Special issue of OpenVMS Timese1 Message-ID: <uki58.105$am1.1433@news.cpqcorp.net>e   Dear Newsgroup,r  K The Intel issue of OpenVMS Times will be posted and on line tomorrow thanksiF to Warren Sander http://www.openvms.compaq.com/openvmstimes/index.html  L Please note that this will be the last issue of OpenVMS Times where I am theL editor.  I have changed jobs in the VMS group.  Diane Gaudreau HPS Marketing will be the editor from now on.P  F Thank you for the support and feedback you have provided me around the newsletter.i  
 Warm Regards,y   Suea   ------------------------------  # Date: Mon, 28 Jan 2002 20:26:44 GMTh4 From: "Terry C. Shannon" <terryshannon@mediaone.net>1 Subject: Re: Intel Special issue of OpenVMS Times-+ Message-ID: <8ui58.1971$9U4.1564@rwcrnsc54>2  = "Sue Skonetski" <susan.skonetski@compaq.com> wrote in messagem+ news:uki58.105$am1.1433@news.cpqcorp.net...h > Dear Newsgroup,e >lF > The Intel issue of OpenVMS Times will be posted and on line tomorrow thanksH > to Warren Sander http://www.openvms.compaq.com/openvmstimes/index.html >,J > Please note that this will be the last issue of OpenVMS Times where I am the D > editor.  I have changed jobs in the VMS group.  Diane Gaudreau HPS	 Marketingi! > will be the editor from now on.  >e  F Sue, congratulations on your role in getting the OpenVMS Times off theI ground and up and running. I am certain that the comp.os.vms constituencyuD shares the (correct) opinion that you've done a superb job with this  downstream marketing collateral!   cheers,    terry sp   ------------------------------  # Date: Tue, 29 Jan 2002 03:56:49 GMT-1 From: "David J. Dachtera" <djesys.nospam@fsi.net> 1 Subject: Re: Intel Special issue of OpenVMS Times:' Message-ID: <3C561E94.88D1BF4D@fsi.net>4   "Terry C. Shannon" wrote:> > ? > "Sue Skonetski" <susan.skonetski@compaq.com> wrote in messagep- > news:uki58.105$am1.1433@news.cpqcorp.net...c > > Dear Newsgroup,  > >tH > > The Intel issue of OpenVMS Times will be posted and on line tomorrow > thanksJ > > to Warren Sander http://www.openvms.compaq.com/openvmstimes/index.html > >_L > > Please note that this will be the last issue of OpenVMS Times where I am > thedF > > editor.  I have changed jobs in the VMS group.  Diane Gaudreau HPS > Marketingo# > > will be the editor from now on.n > >i > H > Sue, congratulations on your role in getting the OpenVMS Times off theK > ground and up and running. I am certain that the comp.os.vms constituencynF > shares the (correct) opinion that you've done a superb job with this" > downstream marketing collateral!  , Terry's statements seconded whole-heartedly!   -- C David J. Dachtera: dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/r   ------------------------------   Date: 28 Jan 2002 13:24:58 CDT= From: wayne@tachysoft.xxx.120519.killspam.00c7 (Wayne Sewell)e; Subject: RE: Microsoft costs sucker windoze users billions!i. Message-ID: <l3fLwJMxdftU@tachxxsoftxxconsult>  z In article <3B55D7F383B0D31197D9009027541CBF1170E1D4@cmiexch1.cmi.itds.com>, Christopher Smith <csmith@amdocs.com> writes: >> -----Original Message-----h9 >> From: bob@instantwhip.com [mailto:bob@instantwhip.com]  > I >> With the Internet, however, security vulnerabilities became exposed tosH >> attack from any savvy programmer on the planet. If Gates' realignment > T > ...not to mention any idiot with Viral Basic, or possibly even a trained MONKEY.   >   G Doesn't have to be trained.  A brain-damaged street monkey could do it.r     -- aO ===============================================================================nM Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxa: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-) O =============================================================================== = Society Lady:  Are you familiar with the great wall of China?o5        Curly:  No, but I know a big fence in Chicago!t   ------------------------------  # Date: Mon, 28 Jan 2002 23:24:37 GMT ) From: Tom Hamilton <thamilton01@snet.net>L; Subject: Re: Microsoft costs sucker windoze users billions!mC Message-ID: <V4l58.11325$fG5.2973629563@newssvr10.news.prodigy.com>n   Bob Ceculski wrote:l  G > having fun losing money all you windoze users ... vms saved billions!n > ) > If Microsoft security fails, .Net fails  > By John Pescatoret > ZDNet News > January 23, 2002 > G > Microsoft has committed to making its products more secure and worthywH > of customers' trust. The philosophy outlined in a memo from Bill GatesF > last week lays out most of the imperatives that Gartner believes are8 > necessary for Microsoft to change the software maker's> > long-established product management and development culture. > @ > Microsoft became the world's most powerful software company byF > building software that gave the individual the control. Upgrades andG > new software always included more features to allow people to do more:D > things. Product managers got promoted by shipping software on timeF > with enough new features to compel customers to upgrade. Crashes andG > sloppy programming that left gaping holes for hackers became problemsd > of secondary importance. > H > With the Internet, however, security vulnerabilities became exposed toG > attack from any savvy programmer on the planet. If Gates' realignment H > of Microsoft to the Internet in 1996 had made security a prime concern@ > for new Internet features, enterprises would have avoided manyB > billions of dollars of cleanup costs because of the long list ofH > viruses and worms that have struck Internet-connected servers and PCs. > J Windows gives a user the control?  I find it the least capable OS when it  comes to doing what I want.    ------------------------------  % Date: Mon, 28 Jan 2002 11:52:25 -0800HC From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam>c& Subject: Re: mount problems on cluster0 Message-ID: <3C55ABF9.75887944@intel.com.nospam>   Rob Buxton wrote:n  8 > I'm having a similar problem, but it happens randomly.F > I'm only mounting a single disk in sylogicals and that disk contains/ > the UAF Files and command command procedures.e > G > I have a mixed Cluster, two VAX4000-700s and two Alpha DEC 3000s. The(F > VAX4000s have a DSSI bus that connects two HSDs. It is on these that" > the UAF and common files reside. > B > Every now and then when booting one of the Alphas the shadow set9 > containing the UAF and common files will fail to mount.s > D > My sylogicals.com file loops round and repeatedly does a sysman ioF > autoconfigure and attempts the mount again but to no avail. Gives up > after 5 attempts.n  B     At my previous job, I also mounted one disk in SYLOGICALS.COM,@ the one holding the cluster-common files (SYSUAF $ friends), but= I made sure I _could_ mount it before attempting to mount it. C Have you tested F$GetDvi("the-disk","EXISTS") before attempting ther? mount?  Looping on that, waiting for the disk to exist with saya2 10 seconds between queries, never failed for me...       -Ken --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfieldy! F20 Automation VMS System Support  kenneth.h.fairfield#intel.comt   ------------------------------    Date: 28 Jan 2002 15:05:35 -0600+ From: young_r@encompasserve.org (Rob Young) & Subject: Re: mount problems on cluster3 Message-ID: <59tkvk25ieLL@eisner.encompasserve.org>o  v In article <3C55ABF9.75887944@intel.com.nospam>, "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam> writes: > Rob Buxton wrote:a > 9 >> I'm having a similar problem, but it happens randomly.oG >> I'm only mounting a single disk in sylogicals and that disk containse0 >> the UAF Files and command command procedures. >>H >> I have a mixed Cluster, two VAX4000-700s and two Alpha DEC 3000s. TheG >> VAX4000s have a DSSI bus that connects two HSDs. It is on these thatl# >> the UAF and common files reside.  >>C >> Every now and then when booting one of the Alphas the shadow set : >> containing the UAF and common files will fail to mount. >>E >> My sylogicals.com file loops round and repeatedly does a sysman ioHG >> autoconfigure and attempts the mount again but to no avail. Gives upe >> after 5 attempts. > D >     At my previous job, I also mounted one disk in SYLOGICALS.COM,B > the one holding the cluster-common files (SYSUAF $ friends), but? > I made sure I _could_ mount it before attempting to mount it.dE > Have you tested F$GetDvi("the-disk","EXISTS") before attempting theeA > mount?  Looping on that, waiting for the disk to exist with say-4 > 10 seconds between queries, never failed for me... > 
 >     -Ken    B 	Very similar to what I have done.  At a previous gig, I came uponA 	a cluster with layers of crud... one annoying trait was that alln> 	disk mounts were in SYLOGICALS.COM with various commented out 	old mounts... painful.   9 	Cleaning that up, I did several things.  I wrote a mount>C 	procedure that loops through a config file, doing the right thing.lA 	Broke out most all product startups to batch /after="+:10", etc.sG 	Then flipped a logical when I was happy to let folks past SYLOGIN.COM.l  @ 	Not a problem as 7x24 squeezes your startups/shutdowns into offA 	hours.  If you want a speedier startup , you might want to stickI/ 	to tradtional inline "@" in SYSTARTUP_VMS.COM.   E 	Here is a snippet that helps with mounting disks that are available.      $       count = 0s $       max_count = 5: $ check_for_disk:t- $       exists = f$getdvi("''disk'","EXISTS"), $       if .not. exists  $       then! $               count = count + 1fP $               write sys$output "''disk' does not exist ... pausing 15 seconds  $               wait 00:00:15t' $               if count .eq. max_countd $               then  & $                  write sys$output ""P $                  write sys$output "       Problems... ''disk' does not exist. & $                  write sys$output ""! $                  goto read_next  $               endifa# $               goto check_for_disk7
 $       endif  $ bypass_disk_wait: G $       if (.not. cluster_mount) .and. f$getdvi("''disk'","MNT") then -e)         goto read_next  ! Already mountedm $ !  $ !3 $       set NoOn $       if cluster_mount $       thenG $               write sys$output "Mounting ''disk' Cluster-Wide . . . "n $       elseJ $               write sys$output "Mounting ''disk' on ''this_node' . . . "
 $       endifwP $       mount_command = "MOUNT/NOREBUILD/NOASSIST ''mount_type' ''disk' ''label  $!   	etc.e   				Robs   ------------------------------  # Date: Tue, 29 Jan 2002 00:05:38 GMT.) From: rob.buxton@wcc.govt.nz (Rob Buxton)d& Subject: Re: mount problems on cluster2 Message-ID: <3c55e19b.2507980084@news.wcc.govt.nz>   I'll expand. further...-  D The procedure I use is similar, but probably a bit more complex than Robs.bF The common disk containg UAF etc. is a shadow set, it's served by HSDs connected to a pair of VAXes.8F The VAXes are the voters of the cluster along with a quorum disk (also HSD served)aF When the Alphas boot the HSD disks are available - (the Alphas are too young to vote.)g  E Because it's a shadow set I need to check that each member Exists andnE is Available before mounting. The com file then mounts either both or ' the member it finds as the shadow set. Z  F BUT.... on random occasions it seems to pass the EXISTS tests but then does not mount the shadow set.C I've put in a bit more debugging to try and see exactly why it getso' upset but no Mount errors are reported.p  ; I do know it's accessed the HSD served disks because it hass/ established communication with the Quorum Disk.s  B It is odd because once it happens that it does not mount the disksD then it will  repeatedly not mount the disks (until my count of 5 is reached)* When it works it works first time through!  C The state of the disks it is trying to mount doesn't change betweenV success and failure.  B So, other posts suggesting the disks may not be on-line etc. don't apply. i  A I put in the sysman io autoconfigure because I was already havingpC problems. I don't quite see how this would cause the problem to get8 worse.   Rob.D On 28 Jan 2002 15:05:35 -0600, young_r@encompasserve.org (Rob Young) wrote:  w >In article <3C55ABF9.75887944@intel.com.nospam>, "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam> writes:- >> Rob Buxton wrote: >> 2: >>> I'm having a similar problem, but it happens randomly.H >>> I'm only mounting a single disk in sylogicals and that disk contains1 >>> the UAF Files and command command procedures.s >>>.I >>> I have a mixed Cluster, two VAX4000-700s and two Alpha DEC 3000s. TheoH >>> VAX4000s have a DSSI bus that connects two HSDs. It is on these that$ >>> the UAF and common files reside. >>>fD >>> Every now and then when booting one of the Alphas the shadow set; >>> containing the UAF and common files will fail to mount.a >>>hF >>> My sylogicals.com file loops round and repeatedly does a sysman ioH >>> autoconfigure and attempts the mount again but to no avail. Gives up >>> after 5 attempts.v >>  E >>     At my previous job, I also mounted one disk in SYLOGICALS.COM, C >> the one holding the cluster-common files (SYSUAF $ friends), buto@ >> I made sure I _could_ mount it before attempting to mount it.F >> Have you tested F$GetDvi("the-disk","EXISTS") before attempting theB >> mount?  Looping on that, waiting for the disk to exist with say5 >> 10 seconds between queries, never failed for me...e >> - >>     -Kenc >@ >.C >	Very similar to what I have done.  At a previous gig, I came upontB >	a cluster with layers of crud... one annoying trait was that all? >	disk mounts were in SYLOGICALS.COM with various commented out  >	old mounts... painful. > : >	Cleaning that up, I did several things.  I wrote a mountD >	procedure that loops through a config file, doing the right thing.B >	Broke out most all product startups to batch /after="+:10", etc.H >	Then flipped a logical when I was happy to let folks past SYLOGIN.COM. >)A >	Not a problem as 7x24 squeezes your startups/shutdowns into offuB >	hours.  If you want a speedier startup , you might want to stick0 >	to tradtional inline "@" in SYSTARTUP_VMS.COM. >yF >	Here is a snippet that helps with mounting disks that are available. >n >0 >$       count = 0 >$       max_count = 5 >$ check_for_disk:. >$       exists = f$getdvi("''disk'","EXISTS") >$       if .not. exists
 >$       then<" >$               count = count + 1Q >$               write sys$output "''disk' does not exist ... pausing 15 seconds g >$               wait 00:00:15( >$               if count .eq. max_count >$               then ' >$                  write sys$output "" Q >$                  write sys$output "       Problems... ''disk' does not exist.  ' >$                  write sys$output ""a" >$                  goto read_next >$               endif$ >$               goto check_for_disk >$       endif >$ bypass_disk_wait:H >$       if (.not. cluster_mount) .and. f$getdvi("''disk'","MNT") then -* >        goto read_next  ! Already mounted >$ ! >$ ! >$       set NoOnN >$       if cluster_mounts
 >$       then-H >$               write sys$output "Mounting ''disk' Cluster-Wide . . . "
 >$       elseeK >$               write sys$output "Mounting ''disk' on ''this_node' . . . ": >$       endifQ >$       mount_command = "MOUNT/NOREBUILD/NOASSIST ''mount_type' ''disk' ''label f >$!n >r >	etc. >  >				Rob >    ------------------------------  % Date: Mon, 28 Jan 2002 17:15:39 -0800tC From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam>u& Subject: Re: mount problems on cluster0 Message-ID: <3C55F7BB.24967308@intel.com.nospam>   Rob Buxton wrote:    > I'll expand. further...  >6F > The procedure I use is similar, but probably a bit more complex than > Robs. H > The common disk containg UAF etc. is a shadow set, it's served by HSDs > connected to a pair of VAXes.hH > The VAXes are the voters of the cluster along with a quorum disk (also
 > HSD served)-H > When the Alphas boot the HSD disks are available - (the Alphas are too > young to vote.). >>G > Because it's a shadow set I need to check that each member Exists andaG > is Available before mounting. The com file then mounts either both orV( > the member it finds as the shadow set.   [...]i  ?     Forgive me if I'm belaboring the obvious, but when you say,.D "[the procedure] then mounts either both or _the member it finds_ asE the shadow set", that last is not sufficient.  To mount a shadow set,aC _all_ members _must_ be visible to _all_ hosts which mount the set.oA If you have a 2-member shadow set, and only one member is visiblet> during boot, the mount will (or _should_) fail.  Have I missed
 something?       -Ken --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfieldc! F20 Automation VMS System SupportM kenneth.h.fairfield#intel.com7   ------------------------------    Date: 28 Jan 2002 10:23:00 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett)q1 Subject: Re: new partner also sent to openvms.orgs, Message-ID: <C0CxglB0O9rR@malvm6.mala.bc.ca>  , In article <3C5579F0.A4A8F45F@caltech.edu>, +   David Mathog <mathog@caltech.edu> writes:    > "Main, Kerry" wrote: >> lJ >> And to just add to Sues posting .. As mentioned before, this is a greatD >> example where an application written in Java becomes very easy to >> certify on OpenVMS. >>   > ; > ANSI C is just as portable as Java and a WHOLE LOT FASTERo > once it's actually running.n  F    I'm glad to see someone else voicing this concern about the scourgeC known as Java. I've yet to see a Java application that ran anywherecB near the speed of a native app - most of them I've seen are almostC unusable because they're so slow ( my experience is mainly confineddB to stuff from Oracle and the stuff Cisco gives you to manage their switches ).g  A    Of course I suppose as long as a vendors Java application runs @ no worse on VMS than on Unix/Windows ( ie wastes as much of the B machine in either case ) that's good enough - except the excessiveA machine cycles one needs to get decent performance will likely bel$ much more expensive on a VMS system.  = > different OS's.  At that point Java does have a portability_G > advantage.  However, it's becoming less and less clear how great that A > advantage is though since the advent of Cygwin and OS/X one caniG > typically build any gnuish application (ie, will compile with gcc andiC > build with gmake) for Unix/Windows/Mac with minimal work.  It maysB > not look native on one or more of these but it will run.  If the: > DII/COE stuff ever ships VMS should be in the same boat. >   @    The benefit of Java (from the vendor's point of view) is thatE they don't need to build multiple versions of the product. Just build E and test the image on their "reference platform", if it fails to workMB somewhere else that is the fault of the JVM implementation on that( machine, so therefore not their problem.   ------------------------------    Date: 28 Jan 2002 16:44:27 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)r1 Subject: Re: new partner also sent to openvms.orgx3 Message-ID: <6RIEwia8+aWM@eisner.encompasserve.org>C  ` In article <C0CxglB0O9rR@malvm6.mala.bc.ca>, nothome@spammers.are.scum (Malcolm Dunnett) writes:. > In article <3C5579F0.A4A8F45F@caltech.edu>, - >   David Mathog <mathog@caltech.edu> writes:n >  >> "Main, Kerry" wrote:b >>> K >>> And to just add to Sues posting .. As mentioned before, this is a greataE >>> example where an application written in Java becomes very easy tor >>> certify on OpenVMS.M >>>  >>  < >> ANSI C is just as portable as Java and a WHOLE LOT FASTER >> once it's actually running. > H >    I'm glad to see someone else voicing this concern about the scourgeE > known as Java. I've yet to see a Java application that ran anywhere.D > near the speed of a native app - most of them I've seen are almostE > unusable because they're so slow ( my experience is mainly confined D > to stuff from Oracle and the stuff Cisco gives you to manage their
 > switches ).o   I don't care for Java.   I don't care for C.    I don't care for Lotteries.b  D But I realize that not everyone shares my tastes, so I am happy thatG VMS also provides support for those other folks.  If they want to use atK reliable operating system, I will not campaign against their choice of VMS.-   ------------------------------    Date: 28 Jan 2002 20:23:54 -0000* From: "Mag Net News" <MAG@lb.bcentral.com>; Subject: News Alert:Military Resale Group, Inc.(OTCBB-MYRG)o( Message-ID: <1012249434.47848.qmail@ech>  L Military Resale Group, Inc=09=09=09=09Corporate Profile:=09     January 200= 2kE 2180 Executive Circle=09=09=09=09=09Ticker Symbol:       OTCBB - MYRGiH Colorado Springs, CO 80906=09=09=09=09Shares Outstanding:=09=096,630,004D Telephone:=09=09719-391-4564=09=09=09Estimated Float:=09=091,220,004I Facsimile:=09=09719-391-4565=09=09=0952 Week Range:=09      $2.27 - $0.03e- Company Contact:Mr. Ethan D. Hokit, PresidentrA For additional free information visit: http://xcelassociates.com/v' Email:=09Ehokit@MilitaryResaleGroup.comdL ---------------------------------------------------------------------------= -n ---r
 -------------g> Military Resale Group, Inc. (OTCBB-MYRG) recently became a newI publicly-owned company as a result of its acquisition in December 2001 by F Bactrol Technologies, Inc.  MYRG markets and distributes groceries andG related products to commissaries on U.S. military bases in the Coloradoo area, = where our operating roots go back five decades.  Previously al publicly-tradedcI corporate shell, Bactrol Technologies acquired Military Resale Group in a H reverse acquisition, leaving 6.6 million shares outstanding and only 1.2# million shares in the public float.o  L MYRG was incorporated in 1997 when it acquired Pittock Distributors, Inc., = a.L fifty-year old firm specializing in distribution of products to the Militar= y A Resale Market.  At the time of the acquisition of Pittock, it hadaH approximately $750,000 in revenue.  Revenue has increased over five-fold (5x)I in the past four years reaching $4.8 million in 2001.  MYRG believes that-H revenue can increase to a $12 million annualized rate by the end of 2002 withL the existing facilities.  President Ethan Hokit and GM Bob Hefner have each=  ; been in the local military resale market for over 25 years.n  H The Military Resale Market is a $15 billion niche market with 13,000,000L authorized patrons, which operates under the aegis of the Defense Commissar= ysL Agency (DeCA), a Department of Defense (DOD) agency. The DOD's objective, i= n I this activity, is to operate resale store systems that sell groceries (atiF commissaries) and general merchandise (at exchanges, i.e., PXs) at the lowestL possible prices. The resulting average cost savings last year of 30.4% make=  F the system a major fringe benefit (untaxed) to military personnel. TheL competitive structure of the Military Resale Market consists of a few large=  G distributors that do not dominate the industry, and hundreds of smallernJ distributors. MYRG's long-term strategy is to grow, primarily by acquiring6 other small distributors, and also by internal growth.   Positive Investment Themen  J 1. The Military Resale Market for groceries-merchandise is a $15 billion a year niche industry.H -13,000,000 authorized patrons benefit from lower prices (about 30%) and on-post shopping convenience.-I -Demand is cyclically stable due to the basic product nature and militaryp
 headcount.J -Distributors require some specialized knowledge - a barrier to entry - in< this industry regulated by DeCA (Defense Commissary Agency).  < 2. The shares possess several appealing investment features.L A.=09They are statistically attractive, as measured by the Market Value/Sal= es Ratio.I -MYRG=92s market value is currently about $2.0 million.  This is only .4xdL actual 2001 sales of $4.8 million and only .3x projected 2002 sales of abou= thL $6.5 million, which is well below the level of the Market Value/Sales ratio=   for most stocks.; B.=09The float is thin, containing only 1.2 million shares.mL C.=09A stock play on the Military Resale Market is a =93new=94 idea to inve= storso. and is defensive in a weaker economic setting.  L MYRG has set a target of reaching a trading range of $4.00 - $5.00 per shar= e A by year-end 2002 so that it can seek a national exchange listing.   < ------------------------------------------------------------ Disclaimer:=20L MAG publishes reports providing informationon selected companies that MAG b=L elieves has investment potential. MAG is not a registered investment adviso=L r or broker-dealer. This reportis provided as an information service only, =L and the statements and opinionsin this report should not be construed as an=L  offer or solicitation tobuy or sell any security. MAG accepts no liability=L  for any loss arisingfrom an investor's reliance on or use of this report. =L An investment inThe Above named company is considered to be highly speculat=L ive and shouldnot be considered unless a person can afford a complete loss =L of investment.MAG has been hired by a third party consultant, and is contra=L cted toreceive a cash advertising fee of $500-$5000 for the publication and=L circulation of this report. Subsequently MAG may buy or sell shares ofthe s=L tock of the above mentioned company in the open market. This reportcontains=L  forward-looking statements, which involve risks, and uncertaintiesthat may=L  cause actual results to differ materially from those set forthin the forwa=L rd-looking statements. For further details concerning theserisks and uncert=L ainties, see the SEC filings of the above mentioned companyincluding the co=1 mpany's most recent annual and quarterly reports.       G _______________________________________________________________________s Powered by List Builder  To unsubscribe follow the link::M http://lb.bcentral.com/ex/manage/subscriberprefs?customerid=3D11414&subid=3D=  4DC4C60E4AFF72E3&msgnum=3D16   ------------------------------  # Date: Tue, 29 Jan 2002 03:54:20 GMTn1 From: "David J. Dachtera" <djesys.nospam@fsi.net> 4 Subject: Re: ON ERROR when READ SYS$INPUT?!?!?!??!?!' Message-ID: <3C561DFE.1EB7B5F9@fsi.net>7   Phillip Helbig wrote:s > 
 > > $DISPLAY:  > > $  ON ERROR THEN CONTINUE.* > > $  READ/PROMPT="> " SYS$COMMAND ANSWER' > > $  ANSWER = F$EDIT(ANSWER,"UPCASE")  > > $GOTO DISPLAYo > > $  EXITf > D > Apologies for following up my own post here.  If ANSWER is defined/ > before the READ, then things work as desired.    Then again, you find thato  % $ READ/END=label file_lnm symbol_name    ...will trap CTRL+Z, while m   $ ON CONTROL_Y THEN action    H ...will catch CTRL+Y (should also catch CTRL+C). Be careful, though, and% remember how error traps work in DCL!2   -- 3 David J. Dachterau dba DJE Systems( http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/n   ------------------------------  % Date: Tue, 29 Jan 2002 07:02:28 +0100b, From: Didier Morandi <Didier.Morandi@gmx.ch>4 Subject: Re: ON ERROR when READ SYS$INPUT?!?!?!??!?!& Message-ID: <3C563AF5.73053A68@gmx.ch>   "David J. Dachtera" wrote: >  > Phillip Helbig wrote:  > >( > > > $DISPLAY:a > > > $  ON ERROR THEN CONTINUEs, > > > $  READ/PROMPT="> " SYS$COMMAND ANSWER) > > > $  ANSWER = F$EDIT(ANSWER,"UPCASE")e > > > $GOTO DISPLAY 
 > > > $  EXITn > >sF > > Apologies for following up my own post here.  If ANSWER is defined1 > > before the READ, then things work as desired.n >  > Then again, you find thats > ' > $ READ/END=label file_lnm symbol_namee >  > ...will trap CTRL+Z, while >  > $ ON CONTROL_Y THEN action > J > ...will catch CTRL+Y (should also catch CTRL+C). Be careful, though, and' > remember how error traps work in DCL!  >   Y and read/err=ERR will trap entries with more that 255 c. or with embedded ESC characters.s  I and read/end=EOF/timeout=10 will branch to the EOF label after 10 secondsc   and  and andc   Didier "I love DCL" M.   ------------------------------  % Date: Tue, 29 Jan 2002 07:03:31 +0100n, From: Didier Morandi <Didier.Morandi@gmx.ch> Subject: Re: Opens& Message-ID: <3C563B34.B5C4B46E@gmx.ch>   Fabio Cardoso wrote: >  > OPEN nothing   $ CLOSE nothingI   :-)r   ------------------------------  # Date: Mon, 28 Jan 2002 22:38:54 GMTm' From: Rick Dyson <Rick-Dyson@UIowa.EDU>1) Subject: PGPlot on OpenVMS and PNG Outputo) Message-ID: <3C55D2FE.AFB30914@UIowa.EDU>5  M Has anyone else who might be familiar with Tim Pearson's PGPlot package triedoN to build it with the PNG driver?  I have been trying to add this device and amI having problems getting the libraries to build correctly.  I keep gettingnP unresolved references to functions within the LibPNG library.  It may be due to J mixing the shareable and non-shared libraries.  I am pretty rusty on those
 LINK details.   G I was just hoping to find someone who has already solved the problem orm< maybe someone who I could work with to solve the problem. :)   Thanks!p  
 Rick Dyson   ------------------------------  # Date: Tue, 29 Jan 2002 03:54:32 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network>a- Subject: Re: PGPlot on OpenVMS and PNG Outputt* Message-ID: <3C562119.3040702@qsl.network>   Rick Dyson wrote:e  O > Has anyone else who might be familiar with Tim Pearson's PGPlot package tried P > to build it with the PNG driver?  I have been trying to add this device and amK > having problems getting the libraries to build correctly.  I keep gettingGR > unresolved references to functions within the LibPNG library.  It may be due to L > mixing the shareable and non-shared libraries.  I am pretty rusty on those > LINK details.  > I > I was just hoping to find someone who has already solved the problem or > > maybe someone who I could work with to solve the problem. :)    I Please post the link command(s), and the exact error messages, and maybe   we can figure it out.u   -John	 wb8tyw@qsl.network Personal Opinion Onlyw   ------------------------------  # Date: Tue, 29 Jan 2002 05:53:25 GMTh- From: "Richard L. Dyson" <rickdyson@home.com> - Subject: Re: PGPlot on OpenVMS and PNG Outputl( Message-ID: <3C5638D4.F1C22492@home.com>   "John E. Malmberg" wrote:b >  > Rick Dyson wrote:  > Q > > Has anyone else who might be familiar with Tim Pearson's PGPlot package tried R > > to build it with the PNG driver?  I have been trying to add this device and amM > > having problems getting the libraries to build correctly.  I keep getting S > > unresolved references to functions within the LibPNG library.  It may be due toSN > > mixing the shareable and non-shared libraries.  I am pretty rusty on those > > LINK details.  > J > Please post the link command(s), and the exact error messages, and maybe > we can figure it out.   H 	OK :)  I am using LibZ v1.1.3 and LibPNG v1.2.0.  The same ones used inL several web analyzer packages that all run smoothly with them.  I am working with PGPlot v5.2.2 also..  ! $ LINK /NOUSER /SHARE=GRPSHR.EXE-p0       GRPCKG.OLB/INCLUDE=GROPEN,GRPCKG.OLB/LIB,-       SYS$INPUT/OPT,- (       SYS$SHARE:DECW$XLIBSHR.EXE/SHARE,-H       DISK$ICS:[DYSON.TOOLS.PGPLOT.PGPLOT-5_2_2.png]libpng.olb/library,-D       DISK$ICS:[DYSON.TOOLS.PGPLOT.PGPLOT-5_2_2.png]libz.olb/libraryB ! The following defines the version number of the shareable image. GSMATCH = LEQUAL,1,50 = ! The following PSECT list must include all the common-blocks   ! declared in GRPCKG and PGPLOT. PSECT = GRCM00 ,LCL,NOSHRk PSECT = GRCM01 ,LCL,NOSHR  PSECT = GRCS02 ,LCL,NOSHRu  J ... several PSECT for each PGPlot library routine in shareable library ...  F ! Transfer vectors: Only user-callable routines have transfer vectors.F ! If you change this file and want it to be compatible with a previousE ! version, you must add transfer vectors only at the end of the list.eB ! If you remove a transfer vector, replace it with a pointer to an ! error processing routine. ! SYMBOL_VECTOR=(GRAREA=PROCEDURE,-l GRCHAR=PROCEDURE,- GRCHSZ=PROCEDURE,- GRCLOS=PROCEDURE,-  + ... lots of individual routine listings ...'   then unresolved references:o  ' %LINK-W-NUDFSYMS, 11 undefined symbols:e/ %LINK-I-UDFSYM,         PNG_CREATE_INFO_STRUCT  0 %LINK-I-UDFSYM,         PNG_CREATE_WRITE_STRUCT 1 %LINK-I-UDFSYM,         PNG_DESTROY_WRITE_STRUCT v$ %LINK-I-UDFSYM,         PNG_INIT_IO % %LINK-I-UDFSYM,         PNG_SET_IHDR t% %LINK-I-UDFSYM,         PNG_SET_PLTE  % %LINK-I-UDFSYM,         PNG_SET_TEXT V% %LINK-I-UDFSYM,         PNG_SET_TRNS e& %LINK-I-UDFSYM,         PNG_WRITE_END ' %LINK-I-UDFSYM,         PNG_WRITE_INFO  & %LINK-I-UDFSYM,         PNG_WRITE_ROW : %LINK-W-USEUNDEF, undefined symbol PNG_SET_IHDR referenced)         in psect $LINK$ offset %X00000070a   ... and more ...  L All of these PNG routines are in the LibPNG.olb library, mostly in the png.c file.o   Thanks for any help!  
 Best Regards,t Rick   ------------------------------    Date: 28 Jan 2002 15:17:37 -0600+ From: young_r@encompasserve.org (Rob Young)k Subject: Re: regarding HPc3 Message-ID: <T$fdaMiCWf5K@eisner.encompasserve.org>i  e In article <iri58.469499$oj3.89426819@typhoon.tampabay.rr.com>, "John N." <JNixon@cfl.rr.com> writes:e >a? > *************************************************************  >  > From a news article: > K > People Mr. Hewlett has talked to say he worries that the chip maker Intel K > and particularly the software giant Microsoft will reap the rewards while>C > systems makers like Hewlett and Compaq struggle to make a profit.e > @ > ************************************************************** > * > Mabye he is not quite so dumb after all. >   < 	Course not.  But he did leave out Dell... gee, I guess Dell@ 	doesn't exactly struggle to make a profit on that Wintel kit...
 	Hmmmmm...  = 	Point here, not to be overlooked, is that industry change istD 	a constant.  HP and Compaq realize this.  If the industry continues= 	to adopt Wintel kit and Wintel provides big $$$ incentive toiA 	adopt their kit and you can find the point-and-click guys prettyy1 	readily, etc... it sort of has its own momentum.s  E 	I don't agree that the age of proprietary systems is over, but thereo? 	is a whole bunch more NT in datacenters today then there was 3e8 	years ago.  And why isn't NT proprietary?  Marketing.     		Wintel 		= Industry Standard 		All others      = Proprietaryx   	Except of course NSK!  HA!i   				Robs   ------------------------------  # Date: Mon, 28 Jan 2002 20:23:42 GMT'# From: "John N." <JNixon@cfl.rr.com>e Subject: regarding HPs? Message-ID: <iri58.469499$oj3.89426819@typhoon.tampabay.rr.com>   H When I first heard about the merger, I thought it was great.  I had highL (naive) opinions of HP management, and not quite so high (educated) opinionsI of Compaq management.  But the more I hear the more I realize we are in aS lose-lose situation.  K Also at first, I didn't like Walter Hewlett.  I thought he was just another J inbred whiner.  However, he made a statement that proves he is not as dumb as some people think he is.r= *************************************************************e   From a news article:  I People Mr. Hewlett has talked to say he worries that the chip maker Intel-I and particularly the software giant Microsoft will reap the rewards while.A systems makers like Hewlett and Compaq struggle to make a profit.m  > **************************************************************  ( Mabye he is not quite so dumb after all.   ------------------------------  % Date: Mon, 28 Jan 2002 14:17:02 -0500A+ From: Steve Lionel <Steve.Lionel@intel.com>h Subject: Re: STR$lowercase ?8 Message-ID: <di8b5ugddjlve6416a7fd5lvfrfk1esqgu@4ax.com>  C On Mon, 28 Jan 2002 17:32:49 GMT, Rick Dyson <Rick-Dyson@UIowa.EDU>n wrote:  P >There is a STR$UPCASE system service in the STR$ system, but I have a user thatO >would like to force lowercase in his Fortran program and I can't quickly find  Q >an answer to his question.  Is there such a pre-existing VMS service or functionr2 >that can be easily called from a Fortran program?  F One can call LIB$MOVTC with LIB$AB_LOWERCASE as the translation table.C The catch is that you have to hand-craft a descriptor for this, bute$ it's easy to do.  Here's an example:   	character* 10 old,new 	external lib$ab_lowercase 	integer*4 tabledesc(2)  	old = 'ABCDEFGHIJ'o 	tabledesc(1) = 256 % 	tabledesc(2) = loc(lib$ab_lowercase)l* 	istat = LIB$MOVTC (old,' ',tabledesc,new) 	write (*,*) new 	end       Steve Lionel Compaq Fortran Engineering Intel Corporatione
 Nashua, NH  . Compaq Fortran - http://www.compaq.com/fortranK Intel Fortran - http://developer.intel.com/software/products/compilers/f50/-   ------------------------------  % Date: Mon, 28 Jan 2002 12:02:12 -0800aC From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam>i Subject: Re: STR$lowercase ?0 Message-ID: <3C55AE44.D4F23E30@intel.com.nospam>   Rick Dyson wrote:u  Q > There is a STR$UPCASE system service in the STR$ system, but I have a user that9O > would like to force lowercase in his Fortran program and I can't quickly find=R > an answer to his question.  Is there such a pre-existing VMS service or function3 > that can be easily called from a Fortran program?a  D     In the past I've used calls to BAS$EDIT, IIRC.  There was an oldF INFO-VAX post that gave the calling arguments, but I seem to have lostC in during my job move. :-(  ISTR that BAS$EDIT gives you everythingnC that F$EDIT gives, plus perhaps one or two others.  Perhaps someonen5 else can dig up that old post (or provide a new one)?    --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfielda! F20 Automation VMS System Supporte kenneth.h.fairfield#intel.com.   ------------------------------  # Date: Mon, 28 Jan 2002 20:25:57 GMTb' From: Rick Dyson <Rick-Dyson@UIowa.EDU>y Subject: Re: STR$lowercase ?) Message-ID: <3C55B3D5.5F431CAB@UIowa.EDU>G   "Kenneth H. Fairfield" wrote:l >  > Rick Dyson wrote:  > S > > There is a STR$UPCASE system service in the STR$ system, but I have a user thatnQ > > would like to force lowercase in his Fortran program and I can't quickly findaT > > an answer to his question.  Is there such a pre-existing VMS service or function5 > > that can be easily called from a Fortran program?0 > F >     In the past I've used calls to BAS$EDIT, IIRC.  There was an oldH > INFO-VAX post that gave the calling arguments, but I seem to have lostE > in during my job move. :-(  ISTR that BAS$EDIT gives you everything E > that F$EDIT gives, plus perhaps one or two others.  Perhaps someone=7 > else can dig up that old post (or provide a new one)?y  M 	Is there a way I can call F$Edit (String, "lowercase") from within a FORTRAN0I program easily?  I think that command will do what I want but it is a DCLa
 Lexical...   Rick   ------------------------------  # Date: Mon, 28 Jan 2002 22:16:12 GMTs' From: Rick Dyson <Rick-Dyson@UIowa.EDU>t Subject: Re: STR$lowercase ?) Message-ID: <3C55CDAC.D139D938@UIowa.EDU>x   Steve Lionel wrote:t > E > On Mon, 28 Jan 2002 17:32:49 GMT, Rick Dyson <Rick-Dyson@UIowa.EDU>f > wrote: > R > >There is a STR$UPCASE system service in the STR$ system, but I have a user thatP > >would like to force lowercase in his Fortran program and I can't quickly findS > >an answer to his question.  Is there such a pre-existing VMS service or functionn4 > >that can be easily called from a Fortran program? > H > One can call LIB$MOVTC with LIB$AB_LOWERCASE as the translation table.E > The catch is that you have to hand-craft a descriptor for this, but5& > it's easy to do.  Here's an example: >  >         character* 10 old,new # >         external lib$ab_lowercase   >         integer*4 tabledesc(2) >         old = 'ABCDEFGHIJ' >         tabledesc(1) = 256. >         tabledesc(2) = loc(lib$ab_lowercase)3 >         istat = LIB$MOVTC (old,' ',tabledesc,new)d >         write (*,*) newf
 >         end   D 	Excellent Suggestion Steve!  I was unaware of that Library service.  A 	With your help, I have written this which appears to work.  DoesaE anyone see any problems with it?  It is not documented nor tested forl robustness,sG but it appears to work with a simple test program I through together...   6         Integer Function Str$LowerCase (Input, Output)  %         Character*(*)   Input, Output %         Integer         TableDesc (2)n(         External        Lib$AB_LowerCase           Str$LowerCase = -1         TableDesc (1) = 2564.         TableDesc (2) = Loc (Lib$AB_LowerCase)A         Str$LowerCase = Lib$MovTC (Input, ' ', TableDesc, Output)n           Return         Enda  ( THANKS AGAIN!  That was pretty fast too!   Rick   ------------------------------  % Date: Mon, 28 Jan 2002 14:27:39 -0800-C From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam>o Subject: Re: STR$lowercase ?0 Message-ID: <3C55D05B.ED92A1A0@intel.com.nospam>   Small addendum:t  C     To call BAS$EDIT from Fortran, pass the two character variablesi< in the "usual" way, but the "edit mask" bits by value, e.g.,       INTEGER    BAS$M_UPCASEa!     PARAMETER (BAS$M_UPCASE = 32)      ...s>     CALL BAS$EDIT (NEW_STRING, OLD_STRING, %VAL(BAS$M_UPCASE))   to convert to upper case.e --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfieldb! F20 Automation VMS System Supporto kenneth.h.fairfield#intel.com    ------------------------------  % Date: Mon, 28 Jan 2002 14:34:35 -0800aC From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam>  Subject: Re: STR$lowercase ?0 Message-ID: <3C55D1FB.7EF9A5C1@intel.com.nospam>  J >         With your help, I have written this which appears to work.  DoesG > anyone see any problems with it?  It is not documented nor tested fori
 > robustness,nI > but it appears to work with a simple test program I through together...h > 8 >         Integer Function Str$LowerCase (Input, Output) >t' >         Character*(*)   Input, Outputi' >         Integer         TableDesc (2)e* >         External        Lib$AB_LowerCase >r >         Str$LowerCase = -1 >         TableDesc (1) = 256_0 >         TableDesc (2) = Loc (Lib$AB_LowerCase)C >         Str$LowerCase = Lib$MovTC (Input, ' ', TableDesc, Output)p >l >         Return
 >         Ende  B     I'd just note that "-1" is not the default return status valueC I'd choose since that value tests TRUE in Fortran (in fact, settingnE IVAL = .TRUE. sets the integer value to -1 (as opposed to the logicalRE value).  Better to initialize to .FALSE. or "0", unless the caller isnD expected to test for specific return values.  This would at least be@ consistent with the VMS condition value returned by LIB$MOVCT...   --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfieldg! F20 Automation VMS System Supportr kenneth.h.fairfield#intel.como   ------------------------------  % Date: Mon, 28 Jan 2002 14:20:32 -0800 C From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@intel.com.nospam>m Subject: Re: STR$lowercase ?0 Message-ID: <3C55CEB0.2A76FC42@intel.com.nospam>   Rick Dyson wrote:g  L >         Is there a way I can call F$Edit (String, "lowercase") from within   > a FORTRANtK > program easily?  I think that command will do what I want but it is a DCLd > Lexical...  I     Short answer, "no".  Longer answer, if you were willing to play games0H with logical names and spawned subprocesses, yes, but that seems like anF awfully large hammer to use for tightening the "temple" screws on your eyeglasses... 8^O :-)r  I     If I were to consider such a thing, I'd at least avoid the subprocess E by using callable TPU with a small (custom) section file and a singleAC procedure that used TPU's TRANSLATE function.  At that point, Steven: Lionel's suggestion to use LIB$MOVTC would be even easier.  C     FWIW, I pulled my personal archive of saved c.o.v articles from G BACKUP and found the BAS$EDIT post.  I include it below in its entiretynF (note that the most useful information comes from Bart Z. Lederman whoE is still active in this news group).  The bad news is, I don't see anoD item code for LOWERCASE! :-(  OTOH, there are still more unused bitsB that could be defined...whether or not BAS$EDIT will _do_ anything. with them...  (Alas, the answer it "nothing".)       Kene --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfield-! F20 Automation VMS System Supporte kenneth.h.fairfield#intel.com0  M ============================================================================= + From: webosk@itwol.bhp.com.au (Ken Bosward)r Newsgroups: comp.os.vms ! Subject: Summary: Callable F$EDIT-: Message-ID: <1993Jun22.084752.890@iwsd01.itwol.bhp.com.au> Date: 22 Jun 93 08:47:52 +1000( Organization: BHP Information Technology	 Lines: 76t  F Thanks to all those who responded re a "callable F$EDIT" routine. Most? people drew my attention to the BASIC built-in EDIT. Thanks to:d  .   frisch@zams.kapsch.co.at (Martin Frischherz)'   BRENNAN@CCHS.SU.EDU.AU (Luke Brennan)t.   "B. Z. Lederman" <LEDERMAN@Eisner.DECUS.Org>)   GVROD@ISUVAX.IASTATE.EDU (rod eldridge)wE   simons/G=Colin/S=Sewell/O=H.A.Simons.Ltd/OU=PACIFIC@mhs.attmail.comn  = I have enclosed mail from Bart Lederman for your information.   J --------------------------------------------------------------------------A     I have only read-only access to NEWS, so you may want to posts!  this as a followup for everyone.-  A     While there is no STR$ routine to do what you want, there is,.;  fortunately, such a routine built in to the BASIC Run TimeV:  Library.  Since this is part of the VMS operating system,B  everyone has it.  It does everything that F$EDIT does, plus more.B  I've published instructions on how to use it in the (now defunct)@  DECUS Newsletter, and it's also on various DECUS Library tapes.    B     But it's fairly easy to use.  The following is the header file<  I wrote to define the function for C programs, which I call  BAS$EDIT.Ht   #ifndef __BAS$EDIT_LOADEDd! #define __BAS$EDIT_LOADED       1  #endif   /*B    Define a prototype and values for the VAX/BASIC EDIT$ function.  D    The EDIT$ function performs one or more string editing functions,2    depending on the value of its integer argument.      Example (in BASIC)r  (    New_string$ = EDIT$(Old_string$, 48%)      Example (in C)w  8    (void) BAS$EDIT (&New_String, &Old_String, Edit-Exp);  E    New_String and Old_String are string descriptors passed by addresso    (reference).e  C    Edit-Exp is an integer value specifying one or more edit actions5:    from the set listed below, which can be ANDed together.  8    There does not appear to be a return status to check.  <    All of the definitions here are my own and NOT Digital's.  #    Bart Z. Lederman     15-Jun-1992i */   void BAS$EDIT ();   J /*                      Values          Effect                          */  J #define BAS$M_NOPARITY  1       /* Trim parity bits                     */J #define BAS$M_COLLAPSE  2       /* Discard all spaces and tabs          */J #define BAS$M_DISFORM   4       /* Discard characters: CR, LF, FF, ESC, */J                                 /* RUBOUT, and NULL                     */J #define BAS$M_DISLEAD   8       /* Discard leading spaces and tabs      */J #define BAS$M_COMPRESS  16      /* Reduce spaces and tabs to one space  */J #define BAS$M_UPCASE    32      /* Convert lowercase to uppercase       */J #define BAS$M_BRACKET   64      /* Convert [ to ( and ] to )            */J #define BAS$M_DISTRAIL  128     /* Discard trailing spaces and tabs     */J #define BAS$M_KEEPQUOTE 256     /* Do not alter characters inside       */J                                 /* quotes                               */J --------------------------------------------------------------------------  ? Ken Bosward, BHP Information Technology, Wollongong. Australia.0 E-mail: webosk@itwol.bhp.com.au,   ------------------------------  # Date: Tue, 29 Jan 2002 00:47:19 GMTi From: danco@pebble.org Subject: Re: STR$lowercase ?- Message-ID: <slrna5bs8l.ngh.danco@pebble.org>e  ; In article <3C55B3D5.5F431CAB@UIowa.EDU>, Rick Dyson wrote:b  G >>     In the past I've used calls to BAS$EDIT, IIRC.  There was an oldd    F Umm, but BAS$EDIT doesn't have "uppercase to lowercase" functionality,* does it?  I don't see it in any case.  :-)     EDIT$A     Syntax      ,            str-var = EDIT$(str-exp, int-exp) 	    	        Values   Effect 		 		1%       Trim parity bitsn& 		2%       Discard all spaces and tabs? 		4%       Discard characters: CR, LF,FF, ESC, RUBOUT, and NULLI* 		8%       Discard leading spaces and tabs. 		16%      Reduce spaces and tabs to one space) 		32%      Convert lowercase to uppercase $ 		64%      Convert [ to ( and ] to )+ 		128%     Discard trailing spaces and tabsN0 		256%     Do not alter characters inside quotes 			                      - DanE   ------------------------------  # Date: Mon, 28 Jan 2002 19:52:03 GMT ) From: "Alan Jones" <atj@blueyonder.co.uk>ND Subject: Re: Younger recruits versus experienced veterans  ( was The& Message-ID: <DZh58.606$Fh.56560@news2>  ; "Dr Robin Bignall" <docrobin@red.sylvania> wrote in messaget2 news:gnd25ug0p1u5tivjk15305vt6ap6v3ucpa@4ax.com...$ [...]["DOes "scone" rhyme wtih "cone  5 ...... The pronunciation depends on which part of thenH > country you are in and how genteel you can fool yourself into thinking > the other pronunciation is.D >TH > In the Midlands, working people from backgrounds similar to those AlanC > quoted rhymed them with cones when buying them from the baker. IneC > tea-shops which considered themselves posh (had doileys, cups and F > saucers instead of mugs, and fabric napkins) they were called skons.G > We always thought it was the fault of affected southerners from southiG > of the river Trent. Events of the past 50 years have proved us right: B > everything is the fault of those southerners, but that's another > story.  C There must be more to this. We are told that "skon" is the ScottisheK pronunciation for the little Scottish cake; so where did the "cone" versiondJ come from? I still think it's a "spelling pronunciation" by people who had) seen the word but never been to Scotland.t  : Alan Jones [whom Robin may have mistaken for someone else]   ------------------------------  # Date: Tue, 29 Jan 2002 01:12:47 GMTg$ From: phil@dawks.demon.co.uk (dawks)D Subject: Re: Younger recruits versus experienced veterans  ( was The0 Message-ID: <3c55f5c4.24461073@news.demon.co.uk>  E On Mon, 28 Jan 2002 19:52:03 GMT, "Alan Jones" <atj@blueyonder.co.uk>  wrote:   >t< >"Dr Robin Bignall" <docrobin@red.sylvania> wrote in message3 >news:gnd25ug0p1u5tivjk15305vt6ap6v3ucpa@4ax.com... % >[...]["DOes "scone" rhyme wtih "coner > 6 >...... The pronunciation depends on which part of theI >> country you are in and how genteel you can fool yourself into thinking. >> the other pronunciation is. >>I >> In the Midlands, working people from backgrounds similar to those Alan D >> quoted rhymed them with cones when buying them from the baker. InD >> tea-shops which considered themselves posh (had doileys, cups andG >> saucers instead of mugs, and fabric napkins) they were called skons.lH >> We always thought it was the fault of affected southerners from southH >> of the river Trent. Events of the past 50 years have proved us right:C >> everything is the fault of those southerners, but that's anotherh	 >> story.* >*D >There must be more to this. We are told that "skon" is the ScottishL >pronunciation for the little Scottish cake; so where did the "cone" versionK >come from? I still think it's a "spelling pronunciation" by people who had * >seen the word but never been to Scotland. >sE i suspect that the majority of South Yorkshire miners didn't vacation*E in Scotland so it's hardly surprising that it's pronounced the way it-C was spelt. Same where i grew up (South Wales), a scone was what you , ate at home, a skon was ate at a restaurant.  ; >Alan Jones [whom Robin may have mistaken for someone else]d   Batman, is that you?   phil.    >n >0   dawks. -- The world is divided into two sorts of people: those that believe that the world is divided into two sorts of people and those that don't.   ------------------------------   End of INFO-VAX 2002.055 ************************