0 INFO-VAX	Sun, 28 Jan 2001	Volume 2001 : Issue 55      Contents:D ANNOUNCEMENT: Updates to WASD scripts - particularly for CSWS V1.0-1  Re: Filename wildcards under VMS Himalaya / Alpha / VMS Re: Himalaya / Alpha / VMS Re: Himalaya / Alpha / VMS Re: Himalaya / Alpha / VMS Re: Himalaya / Alpha / VMS Re: Himalaya / Alpha / VMS+ It's the end for VMS and other Wild Rumours $ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now$ Re: It's the end for VMS get off now2 Re: Malloc bug - anyone else got the same problem? Re: New Elsa Graphic patch- Re: Relative merits of LPD vx TELNET printing - Re: RMS bug, possible triggered by autoextend   Re: select() and getservbyname()  Re: select() and getservbyname()  Re: select() and getservbyname()  Re: select() and getservbyname()3 Re: Simple DECnet-IV Tasks Impossible under Phase-V  Source listings contents Re: Source listings contents TPU$WORK work files  VaxStation 4000/90A Question(s) # Re: VaxStation 4000/90A Question(s) 4 Re: Volume Shadowing vs. RA7000 (WAS: DS20 Slow I/O)4 Re: Volume Shadowing vs. RA7000 (WAS: DS20 Slow I/O)4 Re: Volume Shadowing vs. RA7000 (WAS: DS20 Slow I/O)9 What unit is  "Xmt Time" in SDA> SHO PORT/VC=VC_xxxxx   ? * Re: [VIM] VMS version doesn't do logicals?* Re: [VIM] VMS version doesn't do logicals?  F ----------------------------------------------------------------------  % Date: Sun, 28 Jan 2001 15:21:15 +0930 / From: Mark Daniel <mark.daniel@wasd.vsm.com.au> M Subject: ANNOUNCEMENT: Updates to WASD scripts - particularly for CSWS V1.0-1 / Message-ID: <3A73B353.3F5EF04D@wasd.vsm.com.au>   A With the release of CSWS V1.0-1 and it's revised CGI interface on E January 5th, a number of WASD and WASD-ported scripts should now work H again with VMS Apache.  There are also updates in some of these that mayF be relevant to OSU and Purveyor sites as well.  These kits should also9 be available on the OpenVMS Freeware CD V5 when released.   ;   HyperSPI++ v2.0.1 ... a VMS performance display interface "   WASD Scripts v1.4.0 ... Calendar8                           Conan the Librarian (VMS Help)<                           HyperReader/Shelf (BNU/Bookreader)2                           HyperSPI (basic version)3                           Query/Extract search tool D                           QDLogStats (elmentary access log analysis)4   VMSeti v1.4.1 ... VMS SETI@home processing monitor>   WWW Count v2.5.3 ... Muhammad Muquit's graphical hit counterB   yahMAIL v1.5.0 ... authenticated Web access to a user's VMS mail  / Available under OTHER on the WASD download page      http://wasd.vsm.com.au/wasd/  " Hope they're useful.  Mark Daniel.  > Caveat: testing all aspects of all scripts across three server@ environments is very time consuming (and not very interesting). C Therefore only basic testing is performed, meaning some aspects may ? unknowingly be broken in some environments.  It is freeware :^)    ------------------------------  % Date: Sat, 27 Jan 2001 16:09:43 -0600 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> ) Subject: Re: Filename wildcards under VMS - Message-ID: <3A734727.B6E96058@earthlink.net>    "Richard B. Gilbert" wrote:  > & > Message text written by Bruce Pagram > >Hi J >     I have got some utilities written in C that I'm porting to VMS and IF > am having trouble with wildcards.  Under Unix a command like "myprogJ > *.*" has the wildcard expanded by the shell and the filenames are passedF > to the utility via argv.  In VMS this does not happen.  Any thought? >  > Thanks in advance< > J >         Well, you could do it the VMS way and expand the wildcarded fileK > name(s) in the program or you could get a copy of Jym Dyer's "shell_mung" K > which will replace the original command line with one that looks as if it I > had been processed by the shell.  The wild card expansion process is, I G > believe, identical; the difference is how the expanded file names are L > handled.  In the one case you expand the wildcarded filespec once for eachM > matching file and process it, in the other you make a list, a la shell, and # > then process files from the list.  > - >         There are advantages to theVMS way.    To say the least.   F In another thread, someone was asking about listing directories of allF the files on all the disks MOUNTed to the system. Assuming the logicalG name ALL_DISKS is a search list of all the disks MOUNTed to the system, F a specification like ALL_DISKS:[000000...]*.*;* could return a list ofE file-specifications that could be 100's of Kbytes, even well into the  10's or 100's of megabytes.   B How would you fit THAT string length into a descriptor? The stringH length field would need to be a longword, at least. Good luck allocating7 that much memory ... even for a null-terminated string!   H Of course, you don't have that problem on UN*X - the shell can't recurse@ directories, AFAIK. I believe that's why "ls" has a "-r" option.  G Of course, I'm a VMS guy - so I may very well be (probably am) wrong...     --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------  % Date: Sat, 27 Jan 2001 23:50:17 +0000 ) From: Christof Brass <brass@infopuls.com>  Subject: Himalaya / Alpha / VMS , Message-ID: <3A735EB9.9BEE165D@infopuls.com>  > Two years ago I attended a Compaq technical presentation where= it was stated that a unique feature so called lockstep of the : CPUs used for the Himalaya systems, which is kind of an HW? transaction mechanism on CPU level as far as I understood, will @ be built into Alpha and future versions of NSK would be migrated0 to Alpha and the other CPU line will be dropped.  @ Does anybody know if this scenario is still true if it ever was?; Does anybody know what kind of CPU the Himalaya systems are + currently using and how powerful these are?   ; Could someone explain in a few words what the nature of the 9 Himalaya OS, the Tandem NSK as far as I understand, is as @ compared to VMS? I understood that a major concern within cov is* that VMS will be dropped in favour of NSK.> I like VMS most, not for its name but for its features (design? and philosophy are features). If NSK offers all these features, 
 fine with me.   @ And sorry if I'm confusing things. I think that Himalaya is some- kind of Tandem HW which only runs the NSK OS.    ------------------------------    Date: 27 Jan 2001 19:32:25 -0500* From: young_r@eisner.decus.org (Rob Young)# Subject: Re: Himalaya / Alpha / VMS + Message-ID: <ABo$349LoBev@eisner.decus.org>   X In article <3A735EB9.9BEE165D@infopuls.com>, Christof Brass <brass@infopuls.com> writes:@ > Two years ago I attended a Compaq technical presentation where? > it was stated that a unique feature so called lockstep of the < > CPUs used for the Himalaya systems, which is kind of an HWA > transaction mechanism on CPU level as far as I understood, will B > be built into Alpha and future versions of NSK would be migrated2 > to Alpha and the other CPU line will be dropped. > B > Does anybody know if this scenario is still true if it ever was?= > Does anybody know what kind of CPU the Himalaya systems are - > currently using and how powerful these are?  >   3 	Himalya uses MIPS.  Lock-step was added to EV7 and  	will be used by NSK.   = > Could someone explain in a few words what the nature of the ; > Himalaya OS, the Tandem NSK as far as I understand, is as B > compared to VMS? I understood that a major concern within cov is, > that VMS will be dropped in favour of NSK.   	Start here:   		http://www.google.com/    @ > I like VMS most, not for its name but for its features (designA > and philosophy are features). If NSK offers all these features,  > fine with me.    	But it won't.   > B > And sorry if I'm confusing things. I think that Himalaya is some/ > kind of Tandem HW which only runs the NSK OS.    	That would be about right.    			Rob   ------------------------------  % Date: Sat, 27 Jan 2001 19:41:37 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> # Subject: Re: Himalaya / Alpha / VMS , Message-ID: <3A736ABF.7AFE88BF@videotron.ca>   Christof Brass wrote: A > transaction mechanism on CPU level as far as I understood, will B > be built into Alpha and future versions of NSK would be migrated2 > to Alpha and the other CPU line will be dropped.  , http://himalaya.compaq.com/view.asp?IOID=565  = > Does anybody know what kind of CPU the Himalaya systems are - > currently using and how powerful these are?   N I beleive that their fastest processor is clocked at 300mhz, but remember that6 the OS is designed to scale with many many processors.  - http://himalaya.compaq.com/view.asp?IOID=4867     = > Could someone explain in a few words what the nature of the ; > Himalaya OS, the Tandem NSK as far as I understand, is as  > compared to VMS?  N NSK has a very poor user interface, more akin to TSO (no SPF). There is a fullL screen editor, but the rest is mostly very character cell, line/line. The OSG is not designed to support interactive computing, it si done solely for  transactional computing.  1 > I understood that a major concern within cov is , > that VMS will be dropped in favour of NSK.  H VMS used to compete against Tandem for some accounts (notably some stockN exchanges, lotteries etc). But with Compaq now betting the critical computinmgI on Tandem, you can bet that VMS won't be pitched against any account that  Tandem wants to have.   L So with VMS not allowed to compete against UNIX, NT and Tandem, what is left for it to compete against ?   A > and philosophy are features). If NSK offers all these features,  > fine with me.   J NSK is a very ackward machine to work on. It is also very expensive if youL wnat heavy duty performance. It really requires fault -tolerant thinking forM every action you take and requires that you carefull balance their equivalent N of a cluster so that when a process fails, the CPU that will take the failover will have enough capacity.  K And of course, every change you make requires careful consideration because B you have to change all the nodes that are affected by your change.  B > And sorry if I'm confusing things. I think that Himalaya is some/ > kind of Tandem HW which only runs the NSK OS.   K Compaq seems to have dropped "TANDEM" from their vocabulary and uses Compaq  Himalaya instead.    ------------------------------  % Date: Sun, 28 Jan 2001 04:48:20 +0000 ) From: Christof Brass <brass@infopuls.com> # Subject: Re: Himalaya / Alpha / VMS , Message-ID: <3A73A494.6785DF37@infopuls.com>   Rob Young wrote: >  >         Start here:  > ( >                 http://www.google.com/ >    :-)    ------------------------------  % Date: Sun, 28 Jan 2001 05:00:18 +0000 ) From: Christof Brass <brass@infopuls.com> # Subject: Re: Himalaya / Alpha / VMS , Message-ID: <3A73A762.27F80ABD@infopuls.com>   JF Mezei wrote:  >  > Christof Brass wrote: C > > transaction mechanism on CPU level as far as I understood, will D > > be built into Alpha and future versions of NSK would be migrated4 > > to Alpha and the other CPU line will be dropped. > . > http://himalaya.compaq.com/view.asp?IOID=565 > ? > > Does anybody know what kind of CPU the Himalaya systems are / > > currently using and how powerful these are?  > P > I beleive that their fastest processor is clocked at 300mhz, but remember that8 > the OS is designed to scale with many many processors. > / > http://himalaya.compaq.com/view.asp?IOID=4867  > ? > > Could someone explain in a few words what the nature of the = > > Himalaya OS, the Tandem NSK as far as I understand, is as  > > compared to VMS? > P > NSK has a very poor user interface, more akin to TSO (no SPF). There is a fullN > screen editor, but the rest is mostly very character cell, line/line. The OSI > is not designed to support interactive computing, it si done solely for  > transactional computing. > 3 > > I understood that a major concern within cov is . > > that VMS will be dropped in favour of NSK. > J > VMS used to compete against Tandem for some accounts (notably some stockP > exchanges, lotteries etc). But with Compaq now betting the critical computinmgK > on Tandem, you can bet that VMS won't be pitched against any account that  > Tandem wants to have.  > N > So with VMS not allowed to compete against UNIX, NT and Tandem, what is left > for it to compete against ?  > C > > and philosophy are features). If NSK offers all these features,  > > fine with me.  > L > NSK is a very ackward machine to work on. It is also very expensive if youN > wnat heavy duty performance. It really requires fault -tolerant thinking forO > every action you take and requires that you carefull balance their equivalent P > of a cluster so that when a process fails, the CPU that will take the failover > will have enough capacity. > M > And of course, every change you make requires careful consideration because D > you have to change all the nodes that are affected by your change. > D > > And sorry if I'm confusing things. I think that Himalaya is some1 > > kind of Tandem HW which only runs the NSK OS.  > M > Compaq seems to have dropped "TANDEM" from their vocabulary and uses Compaq  > Himalaya instead.   / It looks like NSK is not a replacement for VMS.   ; Could VMS be a replacement for NSK? What needs to be done?    ; I seems that VMS is the OS which offers the widest range of 7 usage possibilities: realtime, transactional, failover, 8 interactive, multi-user, distributed - except running in< embedded systems (or are there even a military version which> exactly does that?). The problem could be that for each single@ dimension there is another OS which is more suitable. OTH VMS is; probably superior in more than one dimension if compared to 	 every OS. @ The conclusion is that VMS should be used where more than one of: these dimensions are important. If it could be made strong@ enough to compete on every single dimension than this is the way= to go. This is the Micro$oft strategy: beat the competitor on 5 its strongest dimension regardless what it will cost.   : Of course, political reasons could prevent any competition: between OSs. If VMS is going to fade away will this happen before OS/2 will disappear?    ------------------------------  % Date: Sun, 28 Jan 2001 00:39:03 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> # Subject: Re: Himalaya / Alpha / VMS , Message-ID: <3A73B068.8D3E98C8@videotron.ca>   Christof Brass wrote: < > Could VMS be a replacement for NSK? What needs to be done?  M VMS used to have it. Back in the days of VAX, Digital had added hooks for VMS & to operate on fault tolerant hardware.  G At one point of the Compaq assimilation, I was hoping that Tandem would S migrate to Alpha and eventually be merged with VMS. But that ain't going to happen.    ------------------------------  # Date: Sun, 28 Jan 2001 05:50:41 GMT - From: Terry C Shannon <shannon@world.std.com> 4 Subject: It's the end for VMS and other Wild RumoursD Message-ID: <Pine.SGI.4.21.0101280045360.24317-100000@world.std.com>  G This "Death of OpenVMS" rumour sounds specious. I missed the conference C webcast due to other obligations, but it sounds more like an act of J omission (no references to VMS) than a death warrant. Windoze is not readyH for prime time in the VMS space, nor will it be for quite some time. TheI business is actually growing again, the revenue is substantial--much more F than Himalaya NSK--and the margin is in excess of 50 percent. It wouldJ make no sense for CPQ to mess with this formula, especially with declining growth in the Wintel space.    cheers,   
 charlie matco    ------------------------------  % Date: Sat, 27 Jan 2001 14:12:07 -0500u- From: JF Mezei <jfmezei.spamnot@videotron.ca>T- Subject: Re: It's the end for VMS get off nowW, Message-ID: <3A731D86.5C2FC31B@videotron.ca>   Eric Ebinger wrote:e > F > Consider the audience.  Analysts are not, umm...the most technically > astute group known to man. S  L We're not talking about a lightbulb manufacturer here. We're talking about aD high tech computer manufacturer who wants to be an internet leader.   M Compaq broadcats its announcements on the internet for anyone and everyone to  listen in. f  L Compaq/Capellas was fully aware that VMS customers had been told to expect aK renaissance and that VMS would start to take of marketing space. Compaq wasSK fully aware that its TV campaign didn't do any of that, and was fully awaregK that VMS customers had noticed that and Compaq shoudl have been fully awareGG that in such situations, customers will look at all possible sources of,M information to try to see what Compaq has in store for VMS. Remember that VMStM customers have had over a decade of experience of being told not to worry andeD be happy while Digital would turn around and denigrate VMS publicly.  K Both news conferences were public relations events to show where Compaq hadaL been and where it was headed. And because it is harder, due to SEC rules, toM lie at such events, what Capellas said there has far more authority than what 2 any one Compaq employee says in a DECUS event etc.  M Because Capellas took great strides to mention his products, down to specific L major account wins, one can draw only one conclusion: VMS was not forgotten, it was omitted.-  N Furthermore, while our VMS ambassador extraordinaire Kerry has often mentionedL that VMS is making major inroads at companies that are tired of PCs and wantL to consolidate, Capellas went the exact other way and said something akin toH companies no longer trying to consolidate and prefering to have separate# servers which would help its sales.o    N If Capellas truly spoke against the Compaq culture, then he is incompetant andL should be fired. If he is not fired, then his gospel MUST be accepted as theK true Compaq gospel and any employee that speaks a different gospel is to behI considered a rogue employee who does not have the support of his company.e  M In other words, in a war of words between Capellas and Marcello (or any other " VMS person), Capellas always wins.   ------------------------------  % Date: Sat, 27 Jan 2001 14:52:06 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> - Subject: Re: It's the end for VMS get off now , Message-ID: <3A7326E2.EE422D63@videotron.ca>   Eric Ebinger wrote:)> > Yes, the analyst community (as a group) is pretty lame.  :^)  H Lame or not, they control the purse strings and can demand the head of aJ company, force a company to "retrsucture", force a reorganisation (musica= lWH chairs =E0 la Palmer), or force a company to sell parts or cut products.J Capellas must ask analysts their permission/blessing to do anything radic= al.r  J And unless you have a leader with vision and strong presence (aka: Steve = Jobs,sJ Scott McNealy, Larry Ellison and even, I have to admit, Billy Gates), the= n younJ must listen to the analysts tell you what to do, and they tell you what t= o do& based on how you can spin your company  D Capellas lacks charisma and vision. He justs utters keywords such asJ "enterprise" and "mobile computing" because he knows it tickes the analys= ts'iC fancy, however, I am not sure that Compaq understand enterprise. By J definition, enterpsise had meant anything-but-microsoft, and Compaq seems=  tov/ want to help Microsoft change this at any cost.     J Real leaders know that financial analyst meetings are great PR opportunit= ies,J and are very aware that what the analysts say is very important because t= hey J are quoted in the mass media and the customers watch the mass media. Real=  J leaders are very careful about saying things which could be quoted out of=	  context.h  J Capelllas must be held responsible for every word/sentence he uttered. At=  aH time where VMS was supposed to be given a second chance/renaissance, itsJ omission is either sign of sheer incompetence, or a clear signal of Compa= q's  intentions with regards to VMS.     @ > comparison that the bulk of the Compaq analysts have made.  AsA > a result you have analysts making recommendations assuming thate$ > Compaq is a PC company.  Period. =    J And Compaq is still a PC company. They just now use the word "enterprise"=  tom mean PCs sold to companies.   J And if I were a true64 fanatic, I would also be worried because Capellas = isJ doing to True64 what Palmer did to VMS: relegate it to some niches. Compa= qmJ considers Unix to be a "business critical" system (doesn't anyone see som= ee? irony in this ?) and targetted at high end technical computing.   J There is something very wrong with the structure at Compaq which defines = cleartJ turfs that each VP is allowed to play in and cannot play in anyone else's=  8 turf. The various parts of Compaq need to work together.  J There should not be a business-critical, enterprise and consumer division= s, there should be :AF wintel hardware, alpha hardware, Tandem, VMS, Tru64, big-server-sales,J medium-server-sales, and consumer-sales divisions, with each divisions al= lowed ! to sell anything built by compaq.y  J For instance, the consumer-sale division should be allowed to order an al= phaeJ based PC loaded with Linux or VMS and offer it for sale to consumers, and=  theJ medium range people should be allowed to pitch VMS solutions to young ups= tart companies.s  @ > So, you are claiming that Capellas is such an incompetent thatF > he doesn't pay attention to where the profits come from?  As long asC > VMS is profitable VMS will be supported and be part of the Compaqn	 > family.    Short memory eh ?   J Palmer though that NT would take over the world and that nothing could re= sistJ that tidal wave. He positioned the VMS ship, knowing it would sink, hopin= go+ that passengers would hop onto his NT ship.i  G If Capellas is convinced that NT will take over the world with its .NETaJ marketing, then he knows that even though VMS may still yield some profit= s,A those will dwindle soon so he might as well prepare the VMS ship.   J Look at the titanic: after it hit the iceberg, the generators continued t= ooJ function, the orchestras continued to play and they continued to serve dr= inksJ making it look like it wasn't a concern. But eventually, they had to admi= taJ that the ship was sinking to the passengers and that they had to evacuate=  the
 sinking ship.c  J So, just because VMS continues to generate profits doesn't mean that it i=
 sn't sinking.s      J More than ever, Compaq has positioned itself as a PC company due to its b= lind support of Microsoft.     @ > There is also exactly zero reason to believe that they are not= > interested in growing the market.   A much more appropriate-< > question is: Do they have a clue as to HOW to grow the VMS	 > market?i  J Wrong. Any company that prevents its product from competing against other=  J products is a company that does not want to grow that market. If Compaq w= ererJ interested in growing VMS, it would allow VMS to compete against Unix and=  NT.J However, I am sure that Unix and NT are allowed to compete against VMS. T= heJ fact that Unix is able to claim to have the world's best clustering witho= ut= VMS being allowed to defend itself is such a good indication.     3 > If Compaq doesn't want to grow the VMS market whyi > the COE effort?a  J Read Inform: the text is very clear that DII-COE was designed to meet cer= tainJ defence contractor requirements. One small niche. There used to be 6 mark= etJ niches for VMS. Now it seems we're down to one. Still think Compaq is try= ingI
 to grow VMS ?w  , >  Of course, once the COE modifications are> > made it should be possible to position VMS as an alternative7 > to Unix without a repitition of the Snake Oil fiasco.a  J It has always been possible to position VMS from the desktop to the large= stJ datacentre. But Digital anbd now Compaq chose not to do this because they=  did$ not want to hurt their wintel sales.  J If Wintel is your core product, will you let some other product steal sal= es from your core ? No.  J (And yes, even if VMS is more profitable than Wintel, Wintel is still Com= paq's  core product).  	 > That isa= > the time to try and grow VMS marketshare (at the expense ofg > the other Unix vendors).  J But that is where the problem lies. Compaq prefers to pitch True64 agains= tfJ other unix vendors and prefers to pitch NT to other NT vendors. And if VM= S J were to compete against UNIX, it would say bad things about Unix which wo= ulds reflect badly on True64.   ------------------------------  % Date: Sat, 27 Jan 2001 15:50:32 -0600b" From: No User <no.user@anon.xg.nu>- Subject: Re: It's the end for VMS get off nowo9 Message-ID: <98ac0ec94f96ff1f09b5cf14ed0dbdb3@anon.xg.nu>    X-No-archive: Yesy   27 Jan 2001, Bill Todd said:  I >This is clearly not something Compaq worries about:  either they realizeeK >they're not competent to capitalize on VMS and take a fatalistic attitude,oL >or they truly believe any need for VMS will disappear soon as Windows takesG >over the enterprise world, or they think existing VMS customers are soe< >locked in that no amount of abuse will cause them to leave.  C When Compaq bought up Digital, I was far from happy. I had concerns D they would not provide due care and attention for the greatest asset in the product portfolio.e  D Nothing has been done to dispel my concerns. Nor those of the people@ paying good money that has been used to prop up Compaq's fawning over Microsoft.   < Now things have only gotten worse. It is time to let VMS go.  C Not to the OS retirement home. To one where it will receive the duet care and attention it deserves.i  0 Hand out the source. No strings attached please.   ------------------------------  % Date: Sat, 27 Jan 2001 16:09:26 -0600M. From: "Dave Gudewicz" <dgudewicz@megsinet.net>- Subject: Re: It's the end for VMS get off now 9 Message-ID: <3a7347cc$0$70808$1dc6e903@news.corecomm.net>   # MC response to the license plate Q:a  ! Absolutely not!  Why did you ask?i  I I responded that w/no word of VMS during the meet the press sessions last I week, and you saying anything not mentioned is sorta dead meat, etc. etc.a  G Also got mail and a phone call at home from Peter Blackmore, Exec VP of,K Sales and Services and he assures me that VMS is an important ingredient touJ the success of Compaq.  I'm interested in hearing what else he has to say.K I wasn't around for the call at home, but impressed that he cared enough toI call.u  I This week had its ups and downs and I'd say mostly downs for fans of VMS.nJ But the flurry of contacts I had on Friday afternoon nad later made things look brighter.  - I'll keep you posted on further developments.-  L Best advice I can give at the moment.  Keep the faith.  I'm not ready to get off VMS just yet.c   Dave...l  < "Dave Gudewicz" <david.gudewicz@abbott.com> wrote in message2 news:94sj2k$duh$1@fizban.fizban.pprd.abbott.com...F > I sent email to Capellas this am basically asking if he's thrown his license- > plate in the scrap heap. >1I > You'll recall he wore the VMS license plate around his neck at the 1999HK > DECUS event in SanDiego, I believe it was.  When pressed on the continuedzF > VMS support issue by someone during the Q & A, he left the stage andL > presented his business card to the person that asked the question and said# > call me, I support VMS blah blah.o >p4 > The crowd reacted very positively to that gesture. >h! > Wonder how the crowd feels now.5 > 	 > Dave...a >o4 > "Alan Greig" <a.greig@virgin.net> wrote in message& > news:3A71C612.BC5E7F2F@virgin.net... > >  > >  > > Brian Tillman wrote: > >pG > > > I'm listening to the Analyst meeting now.  Compaq is expecting to 	 > capturemJ > > > market in web serving and Internet.  He's said they're after IBM and > Sun.F > > > Capellas also said they're going to be a software company again.	 > They'reTG > > > really wooing Oracle.  Enterprise systems are Tru64.  They're not 	 > gettingpF > > > into a price war on the low end just to gain market share there. > > >,' > > > Oops.  I just lost my connection.l > > >/ > > J > > I can't seem to get back in at the moment either. Windows Media Player > justH > > sits there. Didn't hear Capellas say they are going to be a software	 > company H > > again but I had to break off for a bit. What does he mean by "again"	 > anyway?d > >rJ > > Still I've now heard three speakers. No mention of VMS or if there was it > wastD > > so brief I missed it. Lots of mentions for Windows, a number for > Himalaya -I > > he makes out Himalaya is Compaq's entire presence in stock exchanges.o Unix > isJ > > usually only mentioned in terms of Compaq helping Microsoft to "attack the , > > soft underbelly of Unix" - direct quote. > > H > > Couple of mentions for Alpha supercomputers running Linux. Almost no	 > mention  > > of T64 by name by anyone.u > >n > > > --G > > > Brian Tillman                   Internet: tillman_brian at si.com G > > > Smiths Aerospace                          tillman at swdev.si.com C > > > 3290 Patterson Ave. SE, MS      Addresses modified to preventhB > > > Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"> > > >        This opinion doesn't represent that of my company > >t > > -- > > Alan Greig > >r > >e >d >    ------------------------------  % Date: Sat, 27 Jan 2001 16:21:08 -0600a7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> - Subject: Re: It's the end for VMS get off nowM- Message-ID: <3A7349D4.F1236CAE@earthlink.net>a   JF Mezei wrote:t > ( > "Brian Schenkenberger, VAXman-" wrote:8 > > How many times did the boy in the fable "cry wolf"?? >  > But if you have a bad flue,    ...a clogged chimney???!! :-)   2 > and your mother tucks you to bed every night andM > tells you "sleep tight, the flue will go away soon" and the flue never goess, > away, you will stop beleiving your mother.  G Well, yeah - rust takes time ... unless it's masonry, and then it couldh last for centuries...    --   David J. Dachteras dba DJE Systemso http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.s   ------------------------------  % Date: Sat, 27 Jan 2001 23:14:02 +0000g% From: Alan Greig <a.greig@virgin.net>o- Subject: Re: It's the end for VMS get off now-* Message-ID: <3A73563A.247B8FC7@virgin.net>   Eric Ebinger wrote:i  F > Consider the audience.  Analysts are not, umm...the most technically> > astute group known to man.   Judging by recent statements onB > the viability of Compaq by one analyst they don't seem to have aA > clue that there are computers that don't run Windows.  It wouldoE > seem that pushing a platform other than Windows (or possibly Linux) ? > is a good way to get the analysts to recommend that investors = > drop Compaq stock.  Given the current state of Compaq stockg( > prices that would not be a good thing.  L But they took every opportunity to mention Himalaya. They did not take everyN opportunity to mention VMS. It may have got a very brief mention somewhere but9 it did not figure in their definition of Compaq's future.    ------------------------------  % Date: Sat, 27 Jan 2001 23:39:50 +0000e) From: Christof Brass <brass@infopuls.com>d- Subject: Re: It's the end for VMS get off now , Message-ID: <3A735C46.BB21A735@infopuls.com>   No User wrote: >  > X-No-archive: Yesw >  > 27 Jan 2001, Bill Todd said: > K > >This is clearly not something Compaq worries about:  either they realizeHM > >they're not competent to capitalize on VMS and take a fatalistic attitude, N > >or they truly believe any need for VMS will disappear soon as Windows takesI > >over the enterprise world, or they think existing VMS customers are so > > >locked in that no amount of abuse will cause them to leave. > E > When Compaq bought up Digital, I was far from happy. I had concernseF > they would not provide due care and attention for the greatest asset > in the product portfolio.l > F > Nothing has been done to dispel my concerns. Nor those of the peopleB > paying good money that has been used to prop up Compaq's fawning > over Microsoft.  > > > Now things have only gotten worse. It is time to let VMS go. > E > Not to the OS retirement home. To one where it will receive the duei! > care and attention it deserves.u > 2 > Hand out the source. No strings attached please.  8 From the FAQ VMS7: "OpenVMS VAX and OpenVMS Alpha source  listings CD-ROM sets include the< source listings of most of OpenVMS, ...". I also read in cov! that these sets are not complete.   6 *Why?* Is anybody out there who possesses the sources?  > From the FAQ VMS8: "In no particular order, OpenVMS components are implemented using Bliss, a@ Macro, Ada, PLI, VAX and DEC C, Fortran, UIL, VAX and Alpha SDL, Pascal, * MDL, DEC C++, DCL, Message, and Document."  = Is there anybody out there who studied important parts of thel> sources and can say something about the frequency distribution5 of the languages used and the quality of the sources?e  @ Did anybody ever try to rebuild a whole system from the sources?   ------------------------------  % Date: Sat, 27 Jan 2001 21:24:40 -0500e' From: "Bill Todd" <billtodd@foo.mv.com>a- Subject: Re: It's the end for VMS get off nowt( Message-ID: <94vvl3$8ur$1@pyrite.mv.net>  7 Dave Gudewicz <dgudewicz@megsinet.net> wrote in messaget3 news:3a7347cc$0$70808$1dc6e903@news.corecomm.net...o% > MC response to the license plate Q:n > # > Absolutely not!  Why did you ask?o >nK > I responded that w/no word of VMS during the meet the press sessions lasttK > week, and you saying anything not mentioned is sorta dead meat, etc. etc.  > I > Also got mail and a phone call at home from Peter Blackmore, Exec VP ofpJ > Sales and Services and he assures me that VMS is an important ingredient toL > the success of Compaq.  I'm interested in hearing what else he has to say.J > I wasn't around for the call at home, but impressed that he cared enough to > call.e >pK > This week had its ups and downs and I'd say mostly downs for fans of VMS.wL > But the flurry of contacts I had on Friday afternoon nad later made things > look brighter. >-/ > I'll keep you posted on further developments.0 >1J > Best advice I can give at the moment.  Keep the faith.  I'm not ready to get  > off VMS just yet.f  G Well, talk is cheap - but even so, the only positive talk VMS gets from): Compaq seems to be in direct response to customer queries.  J Personally, I'm surprised anyone has any faith left to keep, and can't seeJ any reason to advise them to do so.  But some people never learn, I guess.C And it pisses me off that Compaq is even now still so eager to taket advantage of them.  I If people wanted to do something that *might* be constructive, they couldnL band together and present Capellas with a list of what actions it would takeK to keep them from bailing out:  at least that would force Compaq to ante up=J or suffer immediate consequences, and help clear the air at the same time.J But the time for that kind of effort to be effective may be long gone now.   - bill   ------------------------------  % Date: Sun, 28 Jan 2001 00:35:20 -0500g- From: JF Mezei <jfmezei.spamnot@videotron.ca> - Subject: Re: It's the end for VMS get off nows, Message-ID: <3A73AF89.3586C2EE@videotron.ca>   Bill Todd wrote:K > If people wanted to do something that *might* be constructive, they couldnN > band together and present Capellas with a list of what actions it would takeM > to keep them from bailing out:  at least that would force Compaq to ante up $ > or suffer immediate consequences,     L Another possible action would be in fact to spead as much "Compaq is killingN VMS" messages around, as loun as possible to a point where Garner and CNET etcN write about it. Then, if Compaq wants to preserve what's left of VMS will haveF to take serious and public actions to show it means business with VMS.   ------------------------------  # Date: Sun, 28 Jan 2001 05:29:59 GMTe From: cstranslations@msn.com- Subject: Re: It's the end for VMS get off nowa) Message-ID: <950aok$2l2$1@nnrp1.deja.com>n  9 In article <3a7347cc$0$70808$1dc6e903@news.corecomm.net>,C1   "Dave Gudewicz" <dgudewicz@megsinet.net> wrote:c% > MC response to the license plate Q:e >i# > Absolutely not!  Why did you ask?s > F > I responded that w/no word of VMS during the meet the press sessions lastF > week, and you saying anything not mentioned is sorta dead meat, etc. etc. >mF > Also got mail and a phone call at home from Peter Blackmore, Exec VP of? > Sales and Services and he assures me that VMS is an importanti
 ingredient tosG > the success of Compaq.  I'm interested in hearing what else he has tot say.C > I wasn't around for the call at home, but impressed that he cared=	 enough toh > call.   F I've been doing OpenVMS since giving up on grad school almost 10 yearsG ago (so you can say it's the bread and butter). First exposure was backlF as an undergraduate (although I believe the alma mater dropped OpenVMSF in favor on Unix back around '92). For those entire 10 years it's beenF well known that VMS is a dying OS, VMS is a dead OS, VMS is a dead and9 buried OS. Who knows - many be it is (one of the 3) . . .   D I can walk into the nearest Barns anf Noble and there will be a wallC (rivaling the Great Wall of China) of books in the computer sectionoD covering Microsoft. It will start with the "dummy books." Windows 98? for Dummies, Windows 98 2nd edition for Dummies, Windows NT for D Dummies, Windows Me for Dummies, Windows 2000 for Dummies, Excel forD Dummies, Word for Dummies, Visual Basic for Dummies, (well - you getE the point). I can walk into the nearest McDonald's and ask the person=C that hands me a cheese burger if they've ever heard of Windows. I'm F sure that person will tell me yes (and give me a strange look). Then I@ can ask that person if he or she has ever heard of OpenVMS . . .  G In the face of this is there any big surprise in the fact that Capellas=G said he's re-doubling all efforts to help Microsoft take over the world # in a talk to a bunch of "analysts?"h  E By the same token what do you expect to hear (at this point) in a oneiE on one conversation other than some blab on just how committed Compaq  is to OpenVMS?   Joet     Sent via Deja.como http://www.deja.com/   ------------------------------  % Date: Sun, 28 Jan 2001 00:37:59 -0500k2 From: rdeininger@mindspring.com (Robert Deininger)- Subject: Re: It's the end for VMS get off nowiL Message-ID: <rdeininger-2801010038000001@user-2iveavb.dialup.mindspring.com>  W In article <3A735C46.BB21A735@infopuls.com>, Christof Brass <brass@infopuls.com> wrote:t  : > From the FAQ VMS7: "OpenVMS VAX and OpenVMS Alpha source" > listings CD-ROM sets include the> > source listings of most of OpenVMS, ...". I also read in cov# > that these sets are not complete.u > 8 > *Why?* Is anybody out there who possesses the sources?   I have access to an old set of VAX listing CDs.  Around VMS 5.5, IIRC.  I don't recall the mixture of languages, and in any case it is somewhat different for alpha.  I think newer stuff tends to be mostly in C.  2Note, the CDs contain _listing_ files, not source files.  Link maps are also included for many components.  If you wanted to compile (which would violate the terms of the license), you'd have to strip off all the page headers, line numbers, etc.  And you don't have the MMS files to help with the building.   --   Robert Deininger rdeininger@mindspring.comr   ------------------------------  % Date: Sat, 27 Jan 2001 22:43:30 +0100t/ From: sbl+news@dd.chalmers.se (Stefan Berglund) ; Subject: Re: Malloc bug - anyone else got the same problem?dA Message-ID: <slrn976g82.8f.sbl+news@ardamir.local.dd.chalmers.se>   5 On Fri, 26 Jan 2001 03:32:54 GMT, Hoff Hoffman wrote:nc > In article <3A709B03.ABB9CFA5@dial.pipex.com>, Bruce Pagram <bruce.pagram@dial.pipex.com> writes:  > I > :    There is a known bug in malloc() that is causing me problems and I K > :was wondering if anyone else had had the same problem, knew of the cause F > :etc.  In certain circumstances calls to malloc() generate a ROPRANDJ > :(reserved operand) exception and crash the application.  Compaq know ofH > :the bug the solution detailed in their support database is to rebuild > :the application using VAX C.h >  	<a lot of good things snipped>c  F >   In my experience, these bugs are almost found and resolved throughG >   non-trivial effort, effort usually involving centralizing all callsdH >   to memory managment into one module and implementing a debugging and: >   detection technique sometimes known as "fenceposts".    E I have not yet had any programming experience with VMS but perhaps my @ Unix experience can be of some help... (A lot of headshaking and& sounds of flamethrowers warming up) :)  E There is a tool called ElectricFence[1] that is an invaluable tool tou: find these kinds of memory and malloc/free related errors.B Unfortunatly it only runs on Unix, but if your program is somewhatF portable it maybe is less effort to port it to Unix and running efenceG to find the bug than trying to find it without such a tool[2]. It mighto* even be possible to port efence to VMS[3].  & 1. http://www.perens.com/FreeSoftware/' 2. I could be totally wrong of course, n> 3. Again I might be totally wrong, the code works by mapping aD    inaccessible page after the malloc:ed memoryfragment. The processC    gets a segmentation fault and dumps core when referencing beyonduD    the boundary. As this is quite Unixish it might require a rethinkD    and redesign to make it work with VMS, on the other hand it might	    not...w  C Which makes me wonder, how do the virtual memory model work on VMS?s   -- N
 			/Stefan 			sbl+news@dd.chalmers.se  " Life - the ultimate practical joke   ------------------------------    Date: 27 Jan 2001 21:53:39 +0100* From: eplan@kapsch.net (Peter LANGSTOEGER)# Subject: Re: New Elsa Graphic patch$* Message-ID: <3a733553$1@news.kapsch.co.at>  r In article <mLzeJ+u6y4vi@sable>, pmoreau@dev.ath.cena.fr (Patrick MOREAU, CENA Athis, Tel: 01.69.57.64.40) writes:L >A new graphics patch, DEC-AXPVMS-VMS712_GRAPHICS-V0400-4, has just rose itsO >head at the public patch server. This patch is referenced from the 7.2-1 page,rN >and I wonder if it can be safely installed on VMS 7.2-1 (the readme file only >reference VMS 7.1-2).  1 No. The drivers for V7.1-2 do not run on V7.2-1 !hA The patch area maintainers did again confuse V7.2-1 and V7.1-2...t  O >This patch hosts the new DDX server image for Elsa Gloria card on EV6, and I'md/ >rather imaptient to try it on my home DS10 ...s  % If you run V7.1-2, then no problem...    --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888 < <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  # Date: Sun, 28 Jan 2001 05:08:38 GMTc; From: Mark Garrett <Mark.Garrett@wedontwantyourspam.com.au>p6 Subject: Re: Relative merits of LPD vx TELNET printingC Message-ID: <B699F23B.10404%Mark.Garrett@wedontwantyourspam.com.au>i  > in article 94q6ig$ruh$1@bob.news.rcn.net, Carmine Castiglia at1 ccastiglia@engsint.com wrote on 26/01/2001 08:41:s  A > Sorry if this is a widely known subject but it's unclear to me:d > M > Given OpenVMS v 7.1-1H2 and DECNet Phase IV (UCX), which is the better/mosttI > efficient/most desirable/etc form of printer management: UCX$LPD_SMB or  > UCX$TELNETSYM? >  > C Just watch out the telnet comms is 7-bit so just don't try anything  containing 8bit data, eg fonts. D     lpd will deliver 8bit data but any client printing hands off jobH formatting to the receiver which if its a just about anything but a unixL intermediary then it will ignore your request for landscape etc etc and evenK if it is may still get it wrong. So your need to discover the concepts of a L bounce queue setup. One queue to format the job streams and drop it into theD other queue to deliver the preformatted data to the printer via lpd.  
     Cheers         Mark :)    ------------------------------  % Date: Sat, 27 Jan 2001 16:31:20 -0500r2 From: norm lastovica <norman.lastovica@oracle.com>6 Subject: Re: RMS bug, possible triggered by autoextend) Message-ID: <3A733E28.59AD46A@oracle.com>e  6 make certain that you've got all available vms patches4 installed.  there are a few cases of a corrupt erase7 patten buffer than can cause similar sounding problems.r   Bill Todd wrote: > / > John Nebel <nebel@csdco.com> wrote in message'C > news:Pine.OSF.4.21.0101260952540.17891-100000@athena.csdco.com...m > >n# > > Wonder if anyone has seen this:f > >"K > > a relative file becomes corrupted - this happened in several cases, oneoN > > was definitely after an autoextend - so that ana/rms reports the following > >b0 > > ***  VBN 69602:  Reserved flag bit 0 is set.0 > > ***  VBN 69602:  Reserved flag bit 4 is set.0 > > ***  VBN 69602:  Reserved flag bit 5 is set. > > J > > Successive executions of ana/rms sometimes report the error, sometimes4 > > don't even though the file hasn't been modified. > L > Not that I have a clue what I'm talking about, but doesn't that sound moreK > like a shadowing bug, where sometimes you read the bad copy and sometimes- > you don't? >  > - bill >  > >2L > > Mixed version cluster, VAX/VMS 7.2, AXP/VMS 7.2-1.  The disk is a 3-copy1 > > shadow set with one remote (over a DS3) copy.o > >e, > > The autoextend was performed by VAX/VMS. > >MI > > Haven't heard back from support and thought I'd ask here just in casem > > anyone has seen it.l > >y > > Thanks.  > >  > > John Nebel > >e   -- .> norman lastovica / oracle rdb engineering / usa / 610.696.4685   ------------------------------  # Date: Sat, 27 Jan 2001 19:41:25 GMT # From: marinto <marinto@my-deja.com> ) Subject: Re: select() and getservbyname()e) Message-ID: <94v895$819$1@nnrp1.deja.com>y  - In article <26JAN200109495932@gerg.tamu.edu>,-*   carl@gerg.tamu.edu (Carl Perkins) wrote:) > marinto <marinto@my-deja.com> writes...lF > }I've written some tcpip routines for unix and am about to port them toF > }vms using system services. After a quick look in the table "OpenVMS> > }System Service and Equivalent C Socket Call" in the OpenVMSF > }documentation I missed two functions that I need. The two functions areu  > }select() and getservbyname(). > }d > }Thanks in advance.- > }- > }/Marintoe >aD > The getservbyname() function is present in the C libraries, so you don'tc? > need to replace it with anything else. From the on-line help:4 >F > CC >. >   Socket_Routinesw >e >     getservbynamea >eC >            Gets information on the named service from the networkoG >            services database. It searches the services database untily@ >            a matching service name is found or until an EOF is encountered. >rH >            If a non-null protocol name is also supplied, searches must% >            also match the protocol.o >t >            Syntax: >d$ >                 #include <netdb.h> >i@ >                 struct servent *getservbyname(char *name, char *proto); >.  E Ok, I guess that solves my problem with getservbyname(). But I'm justeG curios...if one should use a language where one haven't got access to as< similar function, how could it be done with system services?  D > On the other hand, select() isn't (at least not in my somewhat old version0@ > of C). It isn't the "VMS way" of doing things. Others may have suggestions C > as to how to work around this, but it probably depends on exactlyU what you > are trying to do.   F In the unix version of my wait function I can wait until there is dataF to read from a client, data to read from a pipe (mailbox), a client isB ready to receive data, a client has disconnected or another client tries to connect.n  ? My wait function can already wait for data on several mailboxespE (IO$M_WRTATTN with an ast that sets an event flag and also sets whichtG mailbox it was set for, then I just wait for the event flag to be set).t   >i
 > --- Carl >r   /Marinto     Sent via Deja.comy http://www.deja.com/   ------------------------------  # Date: Sat, 27 Jan 2001 19:49:34 GMT # From: marinto <marinto@my-deja.com> ) Subject: Re: select() and getservbyname()i) Message-ID: <94v8oc$8e6$1@nnrp1.deja.com>   , In article <94scba$b7o@gap.cco.caltech.edu>,&   mathog@seqaxp.bio.caltech.edu wrote:@ > In article <94s2lp$ofi$1@nnrp1.deja.com>, marinto <marinto@my- deja.com> writes:hF > >I've written some tcpip routines for unix and am about to port them toF > >vms using system services. After a quick look in the table "OpenVMS> > >System Service and Equivalent C Socket Call" in the OpenVMSF > >documentation I missed two functions that I need. The two functions are   > >select() and getservbyname(). > >s > >Thanks in advance.r > >e >Q > Have a look at:c >r; >   http://seqaxp.bio.caltech.edu/www/UNIX_TO_VMS_NOTES.TXTa7 >   http://seqaxp.bio.caltech.edu/www/X11_VMS_NOTES.TXT   8 Maybe I didn't look hard enough (I looked at gnuplot andE Xterm_vms_122_top sources) but I didn't find anything that could helpeG me with my current problem. But thanks for two links that might come ins handy in the future. :)t   >u > David Mathog > mathog@seqaxp.bio.caltech.edu @ > Manager, sequence analysis facility, biology division, Caltech >a   /Marinto     Sent via Deja.comm http://www.deja.com/   ------------------------------    Date: 27 Jan 2001 15:35:20 -05002 From: malmberg@eisner.decus.org (John E. Malmberg)) Subject: Re: select() and getservbyname() + Message-ID: <USopXfKc4z6n@eisner.decus.org>n  , >> In article <94s2lp$ofi$1@nnrp1.deja.com>,' > marinto <marinto@my-deja.com> writes: G >> >I've written some tcpip routines for unix and am about to port thems > toG >> >vms using system services. After a quick look in the table "OpenVMSe? >> >System Service and Equivalent C Socket Call" in the OpenVMScG >> >documentation I missed two functions that I need. The two functions  > aren! >> >select() and getservbyname().C >> >  6 Someone else has posted the getservbyname() reference.  ; select() is also implemented, On OpenVMS it only works with : sockets.  It is documented both in the Compaq TCPIP manual and in the Compaq C Manual.f  / The TCPIP documentation can be found online at: " http://www.openvms.compaq.com:8000  , The Compaq C documenation is also online at:  ) http://www.openvms.compaq.com/commercial/      -Johna wb8tyw@qsl.network Personal Opinion Only.   ------------------------------   Date: 27 Jan 2001 16:25 CSTn' From: carl@gerg.tamu.edu (Carl Perkins))) Subject: Re: select() and getservbyname()c- Message-ID: <27JAN200116253790@gerg.tamu.edu>n  ' marinto <marinto@my-deja.com> writes...s+ }  carl@gerg.tamu.edu (Carl Perkins) wrote:s* }> marinto <marinto@my-deja.com> writes...G }> }I've written some tcpip routines for unix and am about to port themt }to G }> }vms using system services. After a quick look in the table "OpenVMSl? }> }System Service and Equivalent C Socket Call" in the OpenVMSsG }> }documentation I missed two functions that I need. The two functions  }are! }> }select() and getservbyname().o }> } }> }Thanks in advance. }> } }> }/Marinto }>E }> The getservbyname() function is present in the C libraries, so youo }don't@ }> need to replace it with anything else. From the on-line help: }> }> CC0 }> }>   Socket_Routines }> }>     getservbyname }> [...]: } F }Ok, I guess that solves my problem with getservbyname(). But I'm justH }curios...if one should use a language where one haven't got access to a= }similar function, how could it be done with system services?l  E There is no such language (or, at least, there isn't supposed to be). D You can call the C library routines from any language (although someD will be easier than others - it depends on how much work you have toA do to use the returned data structure and pass it null terminateds+ strings by reference). Such is life on VMS.r  G You might be able to do it in other ways. If this call has equivalents,eJ the exact syntax probably depends on which IP stack you are using. ConsultN the documentation and/or online help (HELP MULTINET PROGRAMMING in the variousI parts doesn't indicate any equivalent). It would presumably be via a QIO.gG There should be some examples for your IP stack, but not necessary oneseF that do this specific thing, on your system somewhere. Multinet has anG examples subdirectory off the multinet root directory. Compaq's productd0 probably has a subdirectory off of SYS$EXAMPLES.  E }> On the other hand, select() isn't (at least not in my somewhat oldh }versionA }> of C). It isn't the "VMS way" of doing things. Others may haveh }suggestionsD }> as to how to work around this, but it probably depends on exactly	 }what you  }> are trying to do. } G }In the unix version of my wait function I can wait until there is data G }to read from a client, data to read from a pipe (mailbox), a client ishC }ready to receive data, a client has disconnected or another clienti }tries to connect. } @ }My wait function can already wait for data on several mailboxesF }(IO$M_WRTATTN with an ast that sets an event flag and also sets whichH }mailbox it was set for, then I just wait for the event flag to be set).	 }/Marintoi  E It turns out that there is a select() function available. I missed ite7 the first time (probaby had a typo in my HELP command):3   CC     Socket_Routines   
     select    A            Allows you to poll or check a group of sockets for I/OnF            activity. It can check what sockets are ready to be read or=            written, or what sockets have a pending exception.               Syntax:  !                 #include <time.h>o  A                 int select(int nfds, int *readfds, int *writefds,mG                               int *execptfds, struct timeval *timeout);t  B It doesn't say it here, but this routine is only valid for networkB connections (that would be the "n" in the "nfds", I expect) and is not useable for other things.n  C Doing it all with ASTs and/or event flags would be the typical "VMSeD way" - mostly ASTs for anything even moderatly complex (a thread perG connection is another way, these days). You set everything up such thateI it fires off an AST when it completes. This allows you to handle multiple G data streams in one process and one thread. Instead of using a bunch of G event flags, one common way to do it is that when the AST fires off, ittH puts the relevant data in the correct queue (for example, you might wantG a queue to put new connection info into, one to put completed read datarA into, and one to put completed write data into via functions likef? LIB$INSQTI and LIB$REMQHI) and then wakes up (via SYS$WAKE) thelB main process (which was hibernating via SYS$HIBER) then exits, the? main process then processes what is in the queue(s) and resumes:E hibernating when it finds them all empty. There's an example of this,tJ communicating via DECnet rather than TCP/IP, in SYS$EXAMPLES:DB_SERVER.C. C It is possible to translate this fairly directly into an equivalentoC that uses TCP/IP. In order to do this, you need to use QIO calls tosB communicate with your IP stack rather than the socket library (youC should be able to still use some socket library calls - such as thefC one above - mixed in if you are carefull and don't do processing atn/ AST level that could change the relevant data).   B For example, using Multinet you'd have a call that looks something7 like this in there somewhere late in the initial setup:   I /* Wait for a connection, fire accept AST routine when connection attempte  * is detected. */  ;     status= SYS$QIO(0, cp_tcp_chan, IO$_ACCEPT_WAIT, &iosb,-:                     &cp_tcp_accept, cm, 0, 0, 0, 0, 0, 0);  A This causes the "cp_tcp_accept" function to be run as an AST whenuC someone tries to connect. You never have to have anything elsewherelA in the program hang around specifically waiting for a connect, or4A polling to see if there has been a connection. When there is one,pC the specified function is executed as an AST. That routine then (in B my data queue paradigm) assigns a channel, accepts the connection,A inserts an entry into a queue with data about the new connection,o@ does a QIO like the one above to prepare for the next connectionA attempt, calls SYS$WAKE to wake up the main process, and returns.p@ The main process then wakes up, checks the queues, finds the new> connection data, and issues the first QIO to read from the newA connection (and perhaps does a write QIO with any initial "hello" 
 information).a   Fun with asynchronous IO.u   --- Carl   ------------------------------  % Date: Sat, 27 Jan 2001 16:16:43 -0600m7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>g< Subject: Re: Simple DECnet-IV Tasks Impossible under Phase-V- Message-ID: <3A7348CB.8341FF4A@earthlink.net>e   Robert Deininger wrote:  > h > In article <3A72168B.FC8C3128@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> wrote: > L > > Actually, this was installed as a clean system by people who didn't know8 > > better than to take all the defaults on the install. > | > Generally, when I have to clean up a mess made by someone who didn't read the manuals, I start by reading the manuals. :-) > L > > In the end, Arturo's suggestion accomplished in a few minutes what years? > > of poring over the documentation and HELP couldn't resolve.F > >f > > Amazing! > >bJ > > I still don't know what the appropriate NCL would be, but I'm not dead
 > > yet... >  > Um, NCL doesn't do this! t  H Guess that explains why *NO ONE* appears to know how to do it, or divineE the darkest magic NCL incantations from the great book of NCL spells.n   > DECNET_REGISTER is a separate tool.  I don't know why, probably some old history.  You are right, no amount of reading the NCL docs will teach you about DECNET_REGISTER.r > A> I know, you tried to read the docs, but you missed the forest for the trees of NCL.  It's easy to do, especially if you skip that thin little intro book.  (Like I skipped it at first.)  The decnet plus docs would be WAY easier to start with, if it did a better job of answering "just what the hell is all this stuff?!?",   AMEN to *THAT*, bro!  a> > The scary thing is, it starts to make sense after a while...  F Guess that explains assimilation into the Redmond collective, as well, eh?t   -- i David J. Dachteras dba DJE Systemsc http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/S  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------    Date: 27 Jan 2001 19:00:59 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) ! Subject: Source listings contents + Message-ID: <jazs4jZs$EpU@eisner.decus.org>   X In article <3A735C46.BB21A735@infopuls.com>, Christof Brass <brass@infopuls.com> writes:    : > From the FAQ VMS7: "OpenVMS VAX and OpenVMS Alpha source" > listings CD-ROM sets include the> > source listings of most of OpenVMS, ...". I also read in cov# > that these sets are not complete.g > 8 > *Why?* Is anybody out there who possesses the sources?  A They exclude material covering pending patents, material coveringi* unannounced features and the LMF listings.   ------------------------------  % Date: Sat, 27 Jan 2001 19:43:23 -0500t- From: JF Mezei <jfmezei.spamnot@videotron.ca>t% Subject: Re: Source listings contentss, Message-ID: <3A736B28.55CB2604@videotron.ca>   Larry Kilgallen wrote:C > They exclude material covering pending patents, material covering7, > unannounced features and the LMF listings.  L Weren't LMF listings published on the back of t-shirts at a DECUS conference some time ago ?    :-) :-)    ------------------------------  % Date: Sat, 27 Jan 2001 15:03:47 -0500t- From: JF Mezei <jfmezei.spamnot@videotron.ca>  Subject: TPU$WORK work files+ Message-ID: <3A73299E.6EA8865@videotron.ca>c  N If your sys$login has a version limit of 2, whenever you try to edit more thanN 2 files at the same time, you get a nasty error message to the effect that the) TPU$WORK.TPU$WORK file cannot be created.-  L There are, of course many ways to deal with this, so I am curious about what2 methods are most used , drawbacks advantages etc ?  ! Some of the way I could think of:   L create a directory with no version limit and define SYS$SCRATCH to go to it.  G create a TPU$WORK logical to point to a file that already exists with a  version limit of 0I (hoping that file never gets deleted otherwise next time you use TPU, the<$ directory's version limit kicks in).  K always use /NOWORK when starting TPU (unless you know you're editing a veryo large file).   ------------------------------  % Date: Sat, 27 Jan 2001 19:47:58 -0500e0 From: "Sharkonwheels" <tonym@compusourceDOT.net>( Subject: VaxStation 4000/90A Question(s)* Message-ID: <18Kc6.329$ze.55809@news4.mco>  B I've looked through the FAQ's, and hunted on the net, but have not found a definitive answer:  G On the VaxStation 4000/90A, which is the serial console port? The DB25,t  Or the MMJ connector?  C If it's the DB25, can I use a null-modem cable, connected to a PC'so4 serial port, and VT100 emulation for serial console?   Here's what I've done:  E Got a db9-to-db25 null modem cable (verified it works between 2 PC's) E Plugged both ends proper. Ran HyperTerminal on PC, 9600n81, Xon/Xoff.7- Next, I set the S3 switch to the UP position.h  G I boot up the Vax, I get unreadable garbage, twice. Same exact garbage,p printed twice.  # No response to keystrokes, nothing.   J The LED's all go through a nice, pretty (!) cycle, at which point they ALL go out, and ! the machine makes a cutesy sound..  K Still, I have no console output, or the ability to type and get a response.-    
 Any ideas?   Thanks in advance!   Tony tonymATcompusourceDOTnet   ------------------------------  % Date: Sat, 27 Jan 2001 18:40:17 -0700g% From: Dan O'Reilly <dano@process.com> , Subject: Re: VaxStation 4000/90A Question(s)B Message-ID: <5.0.2.1.2.20010127183934.00ae76a8@pop.clsp.qwest.net>  + At 05:47 PM 1/27/2001, Sharkonwheels wrote:pC >I've looked through the FAQ's, and hunted on the net, but have not  >found a definitive answer:i >hH >On the VaxStation 4000/90A, which is the serial console port? The DB25, >  Or the MMJ connector?   It's the MMJ connector.I     ------I +-------------------------------+---------------------------------------+vI | Dan O'Reilly                  |                                       |nI | Principal Engineer            |  "Those are my principles. If you     |FI | Process Software              |   don't like them I have others."     |rI | http://www.process.com        |                    -- Groucho Marx    |eI +-------------------------------+---------------------------------------+n   ------------------------------  # Date: Sat, 27 Jan 2001 20:01:34 GMTn" From: fooguy <jweisen@my-deja.com>= Subject: Re: Volume Shadowing vs. RA7000 (WAS: DS20 Slow I/O) ) Message-ID: <94v9es$905$1@nnrp1.deja.com>   G Can you just define what you mean when you say "hot table", "hot disk". G Is that a member of the shadow set that is constantly pretty far our ofsF sync? Are you recommending creating a software 10 disk RAID 10? (bound vol set + mirrored pair)  + In article <blP2q8lZgjEM@eisner.decus.org>,u-   young_r@eisner.decus.org (Rob Young) wrote:eH > In article <94usjg$vak$1@nnrp1.deja.com>, fooguy <jweisen@my-deja.com> writes:0 > F > > Ramdisk might not work know anyway, if we get the RA7000 I'm going toD > > want to spread Oracle out over like 10 mirrored pairs (more if I can...IuD > > think we quoted 18 disks, but if we attach both systems to the 1 cabinettD > > I can only access 12 disks on each machine). No way I can mirror that inc: > > Ram, although it still has to be cheaper than 10K$/MB. >n? > 	You may end up with a very hot table on one of your mirroredw; > 	pairs.  You can use RAID Software for OpenVMS and create ; > 	a host based 0+1 of maybe 5 stripes where each stripe iss= > 	a shadowset comprising two physical disks.  This gains you @ > 	the advantage of spreading a hot table across mucho spindles.; > 	That's why I was interested in just how hot your I/O is. A > 	Nice thing about the above , if you notice one of your stripesR> > 	is getting hot, you can add a third member to that stripe's
 > 	shadowset.  >a? > 	This is the software that folks with very hot I/O are using.t >.	 > 				Rob  >n >t   --- *********************************************m( "All I every wanted from life was to see, Larry Wall give Bill Gates a Perl Necklace."     Sent via Deja.comg http://www.deja.com/   ------------------------------    Date: 27 Jan 2001 19:53:08 -0500* From: young_r@eisner.decus.org (Rob Young)= Subject: Re: Volume Shadowing vs. RA7000 (WAS: DS20 Slow I/O) + Message-ID: <sCvYVzy$Oynp@eisner.decus.org>T  N In article <94v9es$905$1@nnrp1.deja.com>, fooguy <jweisen@my-deja.com> writes:I > Can you just define what you mean when you say "hot table", "hot disk"._I > Is that a member of the shadow set that is constantly pretty far our of H > sync? Are you recommending creating a software 10 disk RAID 10? (bound > vol set + mirrored pair) >   = 	When you speak about spreading Oracle across 10 mirror sets, ; 	you are talking about spreading Oracle's tables.  WARNING:a; 	I'm not an Oracle DBA nor do I pretend to mimic one.  I doa; 	know enough to be dangerous and I can do fulltext searches ; 	with the best of them.... here is what I am talking about:-  E http://cma.zdnet.com/book/oracle/chapters/0-672-31159-3/ch20/ch20.html  L Do not cluster tables if full-table scans are often performed on only one ofO the tables in the cluster. The additional space required by the cluster and the # additional I/O reduces performance.s  > 	That is somewhat related.  Again, if you had a very hot table+ 	you may end up with a saturated mirrorset.n  > 	Your DBA should be able to show you how many tables you have.? 	If you have 5 gigs of data, you may only have one table... butw@ 	that would be a poor design.  Maybe you have data broken out by= 	year and are scanning data by year, maybe you have ten yearst9 	worth of data .. 10 tables.  Again, someone should know.k    @ 	The software RAID 10 is not volume shadowing and bound members.@ 	It is stripes of single drives (RAID0) or stripes of shadowsetsA 	(which makes it RAID10) OR host-based RAID5 (steer clear of thats 	one).  ; http://www.openvms.compaq.com/openvms/storage/raidpage.htmlh   	It too has a loaner program.O  8 	Regarding recommending... using 10 disks and creating a< 	RAID10 would be one way of going about it.  But if you haveD 	5-6 gigs of production data, it seems like quite a waste.  However,? 	it doesn't seem like you have a whole lot of time to step back A 	and evaluate.  I am just suggesting the RAID10 would most likelyy= 	spread the I/O and limit your exposure to I/O bottlenecks ifu= 	one mirrorset is hotter than the others.  This may not be ani= 	issue.  That is why it is important to gather I/O statisticsi= 	as you go.  But if what you are looking at has never been int= 	production, that's not feasible.  If however it has been ande? 	now you are trying to improve it, the I/O stats would give you.; 	a very clear understanding of what is hot and what is not.p   				Robh  - > In article <blP2q8lZgjEM@eisner.decus.org>,e/ >   young_r@eisner.decus.org (Rob Young) wrote:>I >> In article <94usjg$vak$1@nnrp1.deja.com>, fooguy <jweisen@my-deja.com> 	 > writes:e >>G >> > Ramdisk might not work know anyway, if we get the RA7000 I'm goinge > toE >> > want to spread Oracle out over like 10 mirrored pairs (more if Ih	 > can...I E >> > think we quoted 18 disks, but if we attach both systems to the 1s	 > cabinetnE >> > I can only access 12 disks on each machine). No way I can mirroru	 > that ing; >> > Ram, although it still has to be cheaper than 10K$/MB.o >>@ >> 	You may end up with a very hot table on one of your mirrored< >> 	pairs.  You can use RAID Software for OpenVMS and create< >> 	a host based 0+1 of maybe 5 stripes where each stripe is> >> 	a shadowset comprising two physical disks.  This gains youA >> 	the advantage of spreading a hot table across mucho spindles.M< >> 	That's why I was interested in just how hot your I/O is.B >> 	Nice thing about the above , if you notice one of your stripes? >> 	is getting hot, you can add a third member to that stripe'sh >> 	shadowset. >>@ >> 	This is the software that folks with very hot I/O are using. >>
 >> 				Rob >> >> >  > --/ > *********************************************f* > "All I every wanted from life was to see. > Larry Wall give Bill Gates a Perl Necklace." >  >  > Sent via Deja.come > http://www.deja.com/   ------------------------------    Date: 27 Jan 2001 22:49:45 -0500* From: young_r@eisner.decus.org (Rob Young)= Subject: Re: Volume Shadowing vs. RA7000 (WAS: DS20 Slow I/O)u+ Message-ID: <tZ8xOPRhRzKe@eisner.decus.org>e  X In article <sCvYVzy$Oynp@eisner.decus.org>, young_r@eisner.decus.org (Rob Young) writes:P > In article <94v9es$905$1@nnrp1.deja.com>, fooguy <jweisen@my-deja.com> writes:J >> Can you just define what you mean when you say "hot table", "hot disk".J >> Is that a member of the shadow set that is constantly pretty far our ofI >> sync? Are you recommending creating a software 10 disk RAID 10? (boundt >> vol set + mirrored pair)i >> a   	Few other considerations...  6 	With database tuning you can give Oracle a nice chunkD 	of SGA for cache buffers, maybe a Gig (more?)  That - with your 5-6= 	Gigs of production data - might minimize the possiblities ofh= 	hot spots.  Got to watch what I say... I've had to follow myr= 	own posts and others correct me.  Hot spots may not be a big G 	deal given your ratio of cache to data.  Second, if the controller(s) dF 	(can't recall and am not familiar with the one you are using but I do9 	remember it doesn't support controller based RAID) don'te= 	contain write-back cache using Shadowsets with modest writes ? 	will bottleneck the shadowsets.  Meaning:  in HBVS (host based ; 	volume shadowing) both members must acknowledge the write o> 	before I/O completion is acknowledge.  So a typical technique= 	to bypass that "gotcha" is to use controllers with writebackeE 	caching turned on the physical disks that are members of shadowsets.o< 	To guard against catastrophic cache failure (it is rare butA 	has happened), typically shadowsets are split across controllersg! 	or better yet, controller pairs.w  B 	In your case, if you had modest writes to one of your shadowsets,D 	that could buckle performance to that shadowset.  I had a situationB 	whereby I had controllers with shadowsets and the controllers did@ 	not support writeback cache.  The performance was bad for a hot? 	database with modest writes.  Would be nice to know your ration< 	of reads to writes.  RAID10 will help with controllers that< 	don't support writeback as you are striping across a number
 	of disks.  : 	You can see why your vendor was/is trying to steer you to5 	HSZ technology.  The HSZs support writeback caching.t  A 	Our problem out here is we are seeing only a portion of what youh< 	see.  That isn't a complaint, but the very real risk is our> 	advice would be imperfect if for no other reason than lack of> 	information.  Information exchange is often quite slow and at' 	times difficult in a text based forum.o   				Robu   > ? > 	When you speak about spreading Oracle across 10 mirror sets,t= > 	you are talking about spreading Oracle's tables.  WARNING:l= > 	I'm not an Oracle DBA nor do I pretend to mimic one.  I dol= > 	know enough to be dangerous and I can do fulltext searchesh= > 	with the best of them.... here is what I am talking about:I > G > http://cma.zdnet.com/book/oracle/chapters/0-672-31159-3/ch20/ch20.htmt > N > Do not cluster tables if full-table scans are often performed on only one ofQ > the tables in the cluster. The additional space required by the cluster and then% > additional I/O reduces performance.a > @ > 	That is somewhat related.  Again, if you had a very hot table- > 	you may end up with a saturated mirrorset.i > @ > 	Your DBA should be able to show you how many tables you have.A > 	If you have 5 gigs of data, you may only have one table... but B > 	that would be a poor design.  Maybe you have data broken out by? > 	year and are scanning data by year, maybe you have ten years ; > 	worth of data .. 10 tables.  Again, someone should know.t >  > B > 	The software RAID 10 is not volume shadowing and bound members.B > 	It is stripes of single drives (RAID0) or stripes of shadowsetsC > 	(which makes it RAID10) OR host-based RAID5 (steer clear of thatr > 	one). > = > http://www.openvms.compaq.com/openvms/storage/raidpage.html  >  > 	It too has a loaner program.  > : > 	Regarding recommending... using 10 disks and creating a> > 	RAID10 would be one way of going about it.  But if you haveF > 	5-6 gigs of production data, it seems like quite a waste.  However,A > 	it doesn't seem like you have a whole lot of time to step backeC > 	and evaluate.  I am just suggesting the RAID10 would most likely ? > 	spread the I/O and limit your exposure to I/O bottlenecks if ? > 	one mirrorset is hotter than the others.  This may not be anp? > 	issue.  That is why it is important to gather I/O statisticse? > 	as you go.  But if what you are looking at has never been inc? > 	production, that's not feasible.  If however it has been andcA > 	now you are trying to improve it, the I/O stats would give you = > 	a very clear understanding of what is hot and what is not.  > 	 > 				Robe > . >> In article <blP2q8lZgjEM@eisner.decus.org>,0 >>   young_r@eisner.decus.org (Rob Young) wrote:J >>> In article <94usjg$vak$1@nnrp1.deja.com>, fooguy <jweisen@my-deja.com>
 >> writes: >>>NH >>> > Ramdisk might not work know anyway, if we get the RA7000 I'm going >> touF >>> > want to spread Oracle out over like 10 mirrored pairs (more if I
 >> can...IF >>> > think we quoted 18 disks, but if we attach both systems to the 1
 >> cabinetF >>> > I can only access 12 disks on each machine). No way I can mirror
 >> that in< >>> > Ram, although it still has to be cheaper than 10K$/MB. >>>tA >>> 	You may end up with a very hot table on one of your mirroredl= >>> 	pairs.  You can use RAID Software for OpenVMS and creater= >>> 	a host based 0+1 of maybe 5 stripes where each stripe ish? >>> 	a shadowset comprising two physical disks.  This gains youhB >>> 	the advantage of spreading a hot table across mucho spindles.= >>> 	That's why I was interested in just how hot your I/O is.rC >>> 	Nice thing about the above , if you notice one of your stripesn@ >>> 	is getting hot, you can add a third member to that stripe's >>> 	shadowset.s >>>rA >>> 	This is the software that folks with very hot I/O are using.i >>>d >>> 				Robo >>>u >>>  >> r >> -- 0 >> *********************************************+ >> "All I every wanted from life was to seer/ >> Larry Wall give Bill Gates a Perl Necklace."a >>   >> j >> Sent via Deja.com >> http://www.deja.com/z   ------------------------------  # Date: Sat, 27 Jan 2001 18:59:41 GMTm4 From: "Peter Ljungberg" <peter.p.ljungberg@telia.se>B Subject: What unit is  "Xmt Time" in SDA> SHO PORT/VC=VC_xxxxx   ?4 Message-ID: <xUEc6.9520$Qb7.1396343@newsb.telia.net>   SDA>   VMScluster data structures   --------------------------  ; -- Channel Summary for Virtual Circuit (xxxxxx) 80D22F00 --e  ; Address Type Xmt Time Size Preferred Best Last State Changec  K -------- --------- -------- ---- --------- -------- ------------------------  C 80D23080 Preferred FF9BAA46 1412 15659 7673 26-JAN-2001 21:45:08.85e  ;                    ^^^^^^^^  <---   in what unit is this ??s      ? 80D23180 Active FF9BAABF 1412 9110 5158 26-JAN-2001 20:01:00.70:  ; 80DE1F00 Dead 00006E67 1412 671 718 26-JAN-2001 21:45:10.63v  ; 80D6DC80 Dead 0000556D 1412 1226 93 26-JAN-2001 21:45:04.60    SDA>       ----- Channel Selection ----  9 Msg Rcv 11362493 No Xmt Chan 0 Preferred Channel 80D23080-  I Unsequence 5 Rcv Short Msg 0 Delay Time FF9BBF4B   <---   and the unit ofF
 Delay Time  ?.  3 Sequence 8407819 Illegal Seq Msg 0 Buffer Size 1412e  ) ReRcv 2179 Bad Checksum 0 Channel Count 2   8 Lone ACK 2945092 TR DFQ Empty 0 Channel Selections 26664  ( Cache 7400 TR MFQ Empty 0 Protocol 1.4.0  5 Ill ACK 0 CC MFQ Empty 0 Open 26-JAN-2001 20:01:00.70P  = Bytes Rcv 1532965643 Cache Miss 0 Cls 26-JAN-2001 20:00:59.69m               /P.Lj    ------------------------------  # Date: Sat, 27 Jan 2001 18:51:42 GMTt* From: Uncle Jeff <jeff.howie@federated.ca>3 Subject: Re: [VIM] VMS version doesn't do logicals?d) Message-ID: <94v5bs$5kg$1@nnrp1.deja.com>T  > Hi all. Trailer - OpenVMS 7.1 Alpha, vim 5.7 binary d/l'd from www.polarfox.com.h  D I was using version 4.5 for quite a while (where I got the binary, IC can't remember), and all was a-ok. Now after I've upgraded it seemsrF that vim scrambles the logical values I pass to it, so I can only open files in my working directory.  E For example, we have the following system logical defined to point tor part of our source tree:  $    DEVAXP:[SRC]> sh log system$dvsrcD       "SYSTEM$DVSRC" = "DISK3:[FI.SRC.DV.SYSTEM]" (LNM$SYSTEM_TABLE)  A Now, if I either pass a file name as a command line argument _or_V0 try to open a file after vim is opened, like so:       vim system$dvsrc:start.p  + The file name gets mangled exactly like so:d  "     systemDISK3:[FI.SRC.DV]start.p  E After vim is opened, I _can_ open the file if I use unix-like syntax:   '     :vi /disk3/fi/src/dv/system/start.pi  C But that is a pain. Like I said, it worked in the 4.5 version I wasa using before. Any ideas?   --  ( >- THKS  %^>s+  /~\  Uncle Jeff (get rid of JUNK to email)a% |  \) Jeff.HowieJUNK@JUNKfederated.cae+  \_|_ Linux, the choice of a GNU generations     Sent via Deja.comg http://www.deja.com/   ------------------------------  % Date: Sat, 27 Jan 2001 15:39:33 -0500h- From: JF Mezei <jfmezei.spamnot@videotron.ca>v3 Subject: Re: [VIM] VMS version doesn't do logicals? , Message-ID: <3A7331FE.5D61F708@videotron.ca>   Uncle Jeff wrote:m& >    DEVAXP:[SRC]> sh log system$dvsrcF >       "SYSTEM$DVSRC" = "DISK3:[FI.SRC.DV.SYSTEM]" (LNM$SYSTEM_TABLE)2 > try to open a file after vim is opened, like so: >     vim system$dvsrc:start.p- > The file name gets mangled exactly like so:>$ >     systemDISK3:[FI.SRC.DV]start.p  ' Prior to calling vim, do the following: , SHOW DEFAULT.  (the unix equivalent of pwd).  @ You might want to SET DEF SYSTEM$DVSRC  prior to invoking "vim".  7 Also, you might want to find our exactly what "vim" is. L $SHOW SYMBOL vim might reveal it is a command procedure which you could listQ to see what it does. Maybe it is the one which puts you in a different directory.   N if vim is not a symbol, then it is a installed as a command and thus you can't do much about it.F   ------------------------------   End of INFO-VAX 2001.055 ************************