0 INFO-VAX	Mon, 08 Jan 2001	Volume 2001 : Issue 15      Contents: "Hidden" local DCL symbols? % Re: 2 quetions disks, telnet printing ) Re: Almost new TK50s - anybody want them? 4 RE: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 RE: Compaq: A simple, affordable clustering solution4 RE: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 RE: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution Re: DEC 3000 model 300 Re: DEC 3000 model 300# RE: Ebay dead as a post - yet again # RE: Ebay dead as a post - yet again # Re: Ebay dead as a post - yet again ! Re: Flackey fibre channel support ! Re: Flackey fibre channel support 4 RE: Help for ORACLE v7.3 on VMS v7.1 and TCPIP v5.0A* Re: Need FTSS V4.2 or higher for VAXft 810 Online backups Re: Online backups Re: Online backups  The Old VMS Marketing Mares-nest7 vax basic - can someone remind me how this works again? ; Re: vax basic - can someone remind me how this works again? ; Re: vax basic - can someone remind me how this works again? ; Re: vax basic - can someone remind me how this works again?   F ----------------------------------------------------------------------   Date: 8 Jan 2001 04:24:38 GMT 5 From: ccburgess@uqstu.jdstory.uq.edu.au (Ian Burgess) $ Subject: "Hidden" local DCL symbols?. Message-ID: <93bfe6$88p$1@bunyip.cc.uq.edu.au>   A curious phenomenon...   ? There is a sample script supplied with Apache for OpenVMS which 1 displays certain symbols defined for the process.   L Playing with that example, I tried SHOW SYMBOL * with the following results,     $ SHOW SYMBOL *    $RESTART == "FALSE"    $SEVERITY == "1"   $STATUS == "%X00010001" K   APACHE$ENV == "$APACHE$ROOT:[000000]APACHE$ENVIRONMENT_FOR_DCL.EXE_ALPHA" 	   P1 = "" 	   P2 = "" 	   P3 = "" 	   P4 = "" 	   P5 = "" 	   P6 = "" 	   P7 = "" 	   P8 = ""   J But it did not show the 25 other (local) symbols, of which SERVER_SOFTWARE7 is one.  An explicit SHOW SYMBOL SERVER_SOFTWARE works,      $ SHOW SYMBOL SERVER_SOFTWARE -   SERVER_SOFTWARE = "Apache/1.3.12 (OpenVMS)"    but not with wild-cards!     $ SHOW SYMBOL SERVER_SOFTWAR* ?   %DCL-W-UNDSYM, undefined symbol - check validity and spelling   J Any idea what's going on here?   (OpenVMS 7.2-1 on Alpha, before you ask.)   Cheers, 
        IanG -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- G Ian Burgess                                     I.Burgess@its.uq.edu.au G Information Technology Services                 Phone:   61 7 336 54074 G Prentice Building                                          0401 993 458  The University of QueenslandG BRISBANE AUSTRALIA 4072                               www.its.uq.edu.au    ------------------------------  $ Date: Sun, 7 Jan 2001 22:26:56 -00001 From: "NewsReader" <NewsReader@NotOnYourLife.Com> . Subject: Re: 2 quetions disks, telnet printing3 Message-ID: <k4666.15$hE2.9365@news.enterprise.net>   L Don't forget to add /NOASSIST qualifier to MOUNT command in automated system startup proceduresG That way, if a disk is offline for any reason, the procedure won't hang ' indefinitely (waiting for OPCOM reply).     = "Richard B. Gilbert" <DRAGON@compuserve.com> wrote in message 1 news:200101070140_MC2-C0E2-899A@compuserve.com... 5 Message text written by INTERNET:unixguys@my-deja.com B >OK I have some spare time so I'm back to my 4000/600, and the vms learning process continues.   E I added 2 more dssi (total 3) disks and they are alive well and I can  mount and access them.  F I can't seem to glean from searches how to auto mount these two drivesF at boot, is there something similar to an fstab, filesystems or mnttab
 in VMS 7.2 ?<   K You include MOUNT commands in you system startup somewhere.  I usually have K a command procedure: MOUNT_DISKS.COM which I invoke from SYSTARTUP_VMS.COM. #  It would look something like this:  $! MOUNT_DISKS.COM $! $! Mount the rest of the disks.  $! Richard B. Gilbert  $! 07-JAN-2001 $!4 $ MOUNT /SYSTEM DIA1:  <volume label> [logical name]4 $ MOUNT /SYSTEM DIA2:  <volume label> [logical name] $ EXITH The system always assigns the logical name DISK$<volume label>.  You can6 specify an optional logical name of your own devising.   In SYSTARTUP_VMS.COM you put: = $ @SYS$MANAGER:MOUNT_DISKS      !Mount the rest of the disks.     I Or you could write something fancier.  It's possible to write a procedure G to auto discover all the disks, mount them privately to find the volume B label, dismount the disk and mount it again with /SYSTEM.  This isH complicated by the fact that most systems have a floppy disk drive and aD CDROM drive that the system would autodiscover and attempt to mount.  H Or you could write a procedure to open fstab.dat, parse the records, and mount all the disks listed.   K In most circumsances, most system managers just put the mount commands in a 4 file somewhere and ensure that the file is executed.   ------------------------------  # Date: Mon, 08 Jan 2001 04:49:45 GMT " From: unixguys <unixguys@home.com>2 Subject: Re: Almost new TK50s - anybody want them?( Message-ID: <3A59472C.E699B4BC@home.com>  F If you don't get any local takers, e-mail me and I'll see if I can pay= the exchange, shipping etc. to Canada as thats where I be ...    Thanks   Pat    David Spencer wrote:  G > I've got at least 70 TK50 tapes that are practically new, having been G > used three or four times at the most. They've been sitting on a shelf G > in my office for years collecting dust. I need that shelf space back!  > G > Any reasonable offer will not be refused. If I don't find some takers   > they're heading for the trash. > E > I'm located in Sacramento California. Presumably whomever gets them F > (or some portion thereof) must calculate shipping costs from here to > wherever they are. >  > TIA  >  > Dave Spencer
 > PageWeavers D > spencer@pageweavers.spaamfree.com (you know the spamming drill...)   ------------------------------  % Date: Sun, 07 Jan 2001 16:15:54 -0600 + From: "Main, Kerry" <Kerry.Main@compaq.com> = Subject: RE: Compaq: A simple, affordable clustering solution N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284B97@kaoexc3.kao.cpqcorp.net>   Terry,  K >>> If CPQ had intended to outswap VMS into hibernation mode, they wouldn't F have bothered with V7.3, or with the Oracle partnership. I don't thinkJ they'd be hiring developers, either. (If you're a Compaqtian and you bring1 in a qualified VMS hire, you get a $5K bounty)<<<   G Yep, and likely the openVMS Engineering group would not be aggressively L hiring college graduates either ie. focus on long term, new idea's and going
 after youth.    
 Reference:A <http://www.compaq.com/inFORM/issues/issue32/human-int-29-b.html>    Regards,  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Services  Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----9 From: Terry C. Shannon [mailto:terryshannon@mediaone.net]  Sent: January 6, 2001 2:35 PM  To: Info-VAX@Mvb.Saic.Com = Subject: Re: Compaq: A simple, affordable clustering solution       : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3A576896.9BDBA8C2@videotron.ca... > "Terry C. Shannon" wrote: J > > Yeah, and there's the little matter of VMS being a multibillion-dollar; > > business characterized by very generous profit margins.  > L > I keep hearing this over and over again. But if VMS truly was such a greatJ > product that provided Compaq with much more otential profit than sellingH > wintel boxes, we wouldn't be begging Compaq to advertise and save VMS. > F > VMS built Digital into the second biggest computer maker back in the 1980s.K > But it has gone way down since then, and my feeling is that while "sales"  may I > be going back up, I am not sure that Compaq would be willing to release  the L > rate of loss of VMS sites is nor what the expectations in the coming years > are.  J If we are to believe CPQ's numbers, there are ~400K VMS systems out there.L This is down from a peak of perhaps ~550K systems in the early 90s, but downB marginally from 1997 (~450K systems). At the 1997 DECUS CopenhagenH symposium, Jesse Lipcon did a session wherein he showed a then-projectedH decline in the VMS installed base from ~450K to ~200-250K by CY2002. The@ Giga Group put out a research note predicting a similar decline.  I For reasons best known to CPQ, that decline failed to materialize. Again, I based on CPQ statements and figures, close to ~20 percent of the VMS base D wanders off the reservation each year. These emigres are replaced byI customers new to VMS. In 1999, newcomers represented 17 percent of sales.    > K > For instance,  SWIFT won't be supporting VMS anymore, so Compaq will have  lostH > all its SWIFT customers currently on VMS before the end of 2002. There goesD > Compaq's claim that VMS was a key platform for international funds	 tranfers. K > Will those few new sales compensate for the complete loss of a VMS market  ?  > E > So, if Compaq look at the projections of customers who know will be  dumping K > VMS soon (or are in the process of dumping it), it may conclude it is not ! > worth  any advertising efforts.  >  > > And a business that E > > drives the sale of plenty of high-margin, high-end Alpha systems.  > D > Funny, when Wildfire was launched, VMS was barely mentioned. If it	 generates K > so much profit, shouldn't Capellas have mentioned this to the Wall Street I > audience and outlined how the Wildfire, combined with galaxy, will make  VMS F > much more attractive and since VMS is a key product to Compaq due to higherD > yields and loyal customers, this will do a lot of good to Compaq's financials.   K I think so. I was at the announcement and relatively little was said in the J Rainbow Room about VMS. Heil and Capellas seemed to focus on Tru64 becauseH the primary competition for the GS-Series consists of enterprise serversG running Solaris, HP-UX, and AIX. There was, however, a surprisingly low 6 level of speeds-n-feeds content in the press briefing.  K If you wandered across the street to the CUSTOMER briefing that went on all J day long, you woulda learned all you wanted to know about Galaxy, OpenVMS,& Tru64, and the GS-Series architecture.   > J > NO ! They went on anout how Compaq had acheived some special status withJ > Oracle for the True64 and how Tru64 woudl benefit from Wildfire etc etc. >  > > Pulling L > > the plug on Alpha and/or VMS would ensure one thing: a mass migration of a J > > large installed base into the open arms of Scott, Lou, Carley, Mickie, orF > > some other competitor. And there's the little matter of DII-COE as well...  > I > There is a difference between pulling the plug and letting it go into a  coma. ? > DII-COE does't state that the system must remain competitive.   L If CPQ had intended to outswap VMS into hibernation mode, they wouldn't haveK bothered with V7.3, or with the Oracle partnership. I don't think they'd be E hiring developers, either. (If you're a Compaqtian and you bring in a ) qualified VMS hire, you get a $5K bounty)     ? > And DII-CEO is meaningless to the vast majority of customers.   L The Unix-like APIs mandated by DII-COE should render it much easier for ISVsH who are so inclined to port their apps to VMS. More apps would be a RealF Good Thing, and IMHO infinitely more valuable than a VMS television ad during the Super Bowl.   ------------------------------  $ Date: Sun, 7 Jan 2001 17:38:48 -0500' From: "Bill Todd" <billtodd@foo.mv.com> = Subject: Re: Compaq: A simple, affordable clustering solution ( Message-ID: <93aqv5$esj$1@pyrite.mv.net>  4 Main, Kerry <Kerry.Main@compaq.com> wrote in messageH news:910612C07BCAD1119AF40000F86AF0D805284B97@kaoexc3.kao.cpqcorp.net... > Terry, > D > >>> If CPQ had intended to outswap VMS into hibernation mode, they wouldn'tH > have bothered with V7.3, or with the Oracle partnership. I don't thinkL > they'd be hiring developers, either. (If you're a Compaqtian and you bring3 > in a qualified VMS hire, you get a $5K bounty)<<<  > I > Yep, and likely the openVMS Engineering group would not be aggressively H > hiring college graduates either ie. focus on long term, new idea's and going  > after youth.  L While some interesting ideas have indeed come out of academia, my impressionF is that most of them have originated above the student level, and thatH experienced engineers are considerably better-qualified both to evaluateJ those ideas and to implement products based on them than even the students- who may have cooperated in the academic work.   I Isn't the most common reason to hire new graduates the fact that they are K relatively inexpensive and relatively unpicky as to what they work on?  Can  you say 'maintenance'?   - bill   ------------------------------  % Date: Sun, 07 Jan 2001 17:18:59 -0600 + From: "Main, Kerry" <Kerry.Main@compaq.com> = Subject: RE: Compaq: A simple, affordable clustering solution N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284B98@kaoexc3.kao.cpqcorp.net>   Bill,   I >>> Isn't the most common reason to hire new graduates the fact that they  are K relatively inexpensive and relatively unpicky as to what they work on?  Can  you say 'maintenance'? <<<  * In your opinion. It is certainly not mine.  * Another view is as stated in this article:A <http://www.compaq.com/inFORM/issues/issue32/human-int-29-b.html> J "College graduates and students are very important to the continued growthF of the OpenVMS organization. They bring knowledge of current trends inJ computer science and a different perspective on the technologies that willL advance OpenVMS. Their knowledge and insights challenge and support the workH of the excellent, high-caliber engineering talent already in the OpenVMS organization."  G If OpenVMS Engineering had no plans to hire graduates, the fud would be = "see- they are not even going after new Engineering grad's !"   L Now that OpenVMS Engineering is hiring new grad's, the fud is "the grads are low cost, hence maintenance' ..    Geeezzz ..     Regards,  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Services  Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----, From: Bill Todd [mailto:billtodd@foo.mv.com] Sent: January 7, 2001 5:39 PM  To: Info-VAX@Mvb.Saic.Com = Subject: Re: Compaq: A simple, affordable clustering solution       4 Main, Kerry <Kerry.Main@compaq.com> wrote in messageH news:910612C07BCAD1119AF40000F86AF0D805284B97@kaoexc3.kao.cpqcorp.net... > Terry, > D > >>> If CPQ had intended to outswap VMS into hibernation mode, they wouldn'tH > have bothered with V7.3, or with the Oracle partnership. I don't thinkL > they'd be hiring developers, either. (If you're a Compaqtian and you bring3 > in a qualified VMS hire, you get a $5K bounty)<<<  > I > Yep, and likely the openVMS Engineering group would not be aggressively H > hiring college graduates either ie. focus on long term, new idea's and going  > after youth.  L While some interesting ideas have indeed come out of academia, my impressionF is that most of them have originated above the student level, and thatH experienced engineers are considerably better-qualified both to evaluateJ those ideas and to implement products based on them than even the students- who may have cooperated in the academic work.   I Isn't the most common reason to hire new graduates the fact that they are K relatively inexpensive and relatively unpicky as to what they work on?  Can  you say 'maintenance'?   - bill   ------------------------------   Date: 8 Jan 2001 00:55:09 +0100 * From: eplan@kapsch.net (Peter LANGSTOEGER)= Subject: RE: Compaq: A simple, affordable clustering solution * Message-ID: <3a5901dd$1@news.kapsch.co.at>  | In article <910612C07BCAD1119AF40000F86AF0D805284B97@kaoexc3.kao.cpqcorp.net>, "Main, Kerry" <Kerry.Main@compaq.com> writes:L >>>> If CPQ had intended to outswap VMS into hibernation mode, they wouldn'tG >have bothered with V7.3, or with the Oracle partnership. I don't think5K >they'd be hiring developers, either. (If you're a Compaqtian and you bringp2 >in a qualified VMS hire, you get a $5K bounty)<<< >EH >Yep, and likely the openVMS Engineering group would not be aggressivelyM >hiring college graduates either ie. focus on long term, new idea's and goingu >after youth.  >  >Reference:dB ><http://www.compaq.com/inFORM/issues/issue32/human-int-29-b.html>   Come on, Kerry.p; We all know that OpenVMS engineering was never the problem.i> The marketing, ads and sales were and unfortunately still are.  B Yes, the Q top management might have changed their mind about VMS,: but what we still see here is, that VMS is on its way out.  F And decisions where VMS is in the game are usually long term decision.D We see VMS decline at customers which decided a couple of years ago.A Some of them you might convince to keep with VMS, a greater blockcD you already lost and the rest will continue to phase VMS out becauseF they see nothing (except a small "me too" for some products) where VMSJ brings advantages to them. ("Cluster ? Hah ! Everyone has clusters now !")  M We see hiring new engineers, we see new VMS versions (just like in the past),nK but what else ? eg. Does anyone tried to revert the EUREX (biggest european H stock exchange [system]) decision to drop VMS (and AIX; as a frontend) ?J Don't tell me the front-end has no meaning. It is mostly the core opsys ofH all attached banks and if VMS is dropped at EUREX, almost all banks willF start to drop VMS, too !! And this is the remaining key market where Q7 thinks VMS should reside. Still no alarm bell ringing ?n  E Remember AltaVista Firewall for OpenVMS ? It got cancelled many years L ago because of 1) no good TCPIP stack 2) too expensive 3) ergo no customers.E A firewall with cluster support built in is an advantage, but alas...   K Or the new edu program (I can't use it). I see no happiness here in USENET.PI I see some doubt (because of legal concerns). Without a broad acceptance, H it will not bring harvest in the short run. And for a long run, there isI not enough time. I see no big spendings of cheap Alphas to edu instituteswH (If I see something from Q it is WINTEL, and this one is still better byI other producers like DELL, HP and even IBM. I still wonder how Q got #1).2  J David wrote umpteen times 'RIP VMS'. I begin to fear, he is right, becauseJ the speed (of revert the VMS damage) is too slow and too much time passed.H And I hate this, because I know now about 15 different operating systemsH and VMS is by far the best. But is seems, not the best products survive.  D And now consider, if I as a VMS bigot have fears about VMS's future,F what do you think what the big mass of uninformed MGMs/PHMs chained to WINTEL boxes think ?  G Go on, a lot of work has to be done to motivate us again (and god knowsh how we want to see this happen)A   -- R< 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, 07 Jan 2001 18:56:29 -0600U7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>w= Subject: Re: Compaq: A simple, affordable clustering solutionn- Message-ID: <3A59103D.6C44E836@earthlink.net>    Peter LANGSTOEGER wrote: > ~ > In article <910612C07BCAD1119AF40000F86AF0D805284B97@kaoexc3.kao.cpqcorp.net>, "Main, Kerry" <Kerry.Main@compaq.com> writes:N > >>>> If CPQ had intended to outswap VMS into hibernation mode, they wouldn'tI > >have bothered with V7.3, or with the Oracle partnership. I don't thinkhM > >they'd be hiring developers, either. (If you're a Compaqtian and you bringr4 > >in a qualified VMS hire, you get a $5K bounty)<<< > > J > >Yep, and likely the openVMS Engineering group would not be aggressivelyO > >hiring college graduates either ie. focus on long term, new idea's and going  > >after youth.: > >T
 > >Reference:hD > ><http://www.compaq.com/inFORM/issues/issue32/human-int-29-b.html> >  > Come on, Kerry. = > We all know that OpenVMS engineering was never the problem.!@ > The marketing, ads and sales were and unfortunately still are. > D > Yes, the Q top management might have changed their mind about VMS,< > but what we still see here is, that VMS is on its way out.  C So, then - the answer to the perennial question, "what can *WE* do"AG would seem to be as Terry and others have suggested: every one who can,fE buy as many shares of Compaq as your finances will responsibly allow.tH Remember that the stock certificates must be in *YOUR* name, not that ofG your retirement account or its management firm. We will then exert whatYA leverage we can muster and force Capellas and the others to beginoG proactively marketing VMS or face ouster by demand of the stockholders.a  E This seems the only effective recourse available. I urge everyone whos+ can to participate actively in this effort.d   -- n David J. Dachtera  dba DJE Systemse 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.n   ------------------------------  % Date: Sun, 07 Jan 2001 20:30:04 -0500a- From: JF Mezei <jfmezei.spamnot@videotron.ca>5= Subject: Re: Compaq: A simple, affordable clustering solution , Message-ID: <3A59181A.F240B356@videotron.ca>   "Main, Kerry" wrote:I > Yep, and likely the openVMS Engineering group would not be aggressivelysN > hiring college graduates either ie. focus on long term, new idea's and going > after youth.  I Sorry, this does absolutely nothing to me. During the worse of the Palmer K "Digital is killing VMS" years, VMS got new versions, and got some promisesrJ from Digital that Digital would continue to support and improve VMS for at" least 5 years yada yada yada yada.  J So far, I have seen those arguments currently used by Compaq to defend itsH lack of marketing of VMS during the Palmer years. So, I am sorry to say,< Compaq has a long way to go to regain my confidence in them.   <steam venting mode>N For crying out loud, why is so goddam impossible for Compaq to take just a bitI of VMS's profits and make some public advertising so that at least we areiM given the impression that Compaq isn't affraid to publicly admit it has VMS ?pL Right now, I get the feeling that Compaq feels that VMS is a disease that is9 socially unacceptable so it won't talk about it publicly.i   <apologize mode> sorry for the language.i   ------------------------------  % Date: Sun, 07 Jan 2001 20:08:05 -0600f+ From: "Main, Kerry" <Kerry.Main@compaq.com>q= Subject: RE: Compaq: A simple, affordable clustering solution N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284B99@kaoexc3.kao.cpqcorp.net>   Peter,  B You are certainly entitled to your opinion, but I see the internalG announcements of Customer OpenVMS wins (both new Customers and existingeL ones) and they are significant - much more so than a year ago. And these are large Customers.  K At the last OpenVMS Ambassador meeting in Nashua there were 40 ISV partners.J in attendance - the largest number in quite awhile. Obviously, these folks( see the business is coming back as well.  $ Could the marketing be better? Sure.  K Does that mean that the sky is falling and the whole world is black because & you don't see OpenVMS in magazine ads?  J Two years ago I was doing a mix of NT and VMS, but due to business in lastK 12 months changing around, I am back 100% to OpenVMS. So, while business is F always different in various parts of the globe, it is certainly not as  gloomy as you make it out to be.  J [and please, no rat holes of Digital did this, Digital did that, or PalmerL did ..whatever. Every reader of this list has heard it all hundreds of timesK and no one disagrees that mistakes were made back then. All that is historyh and its time to move on ... ]   6 >>> ("Cluster ? Hah ! Everyone has clusters now !")<<<   ROTFL.  0 >>> Or the new edu program (I can't use it). <<<  > Sure you can. The legaleze wording just needs to be clarified.  J >>> Does anyone tried to revert the EUREX (biggest european stock exchange= [system]) decision to drop VMS (and AIX; as a frontend) ? <<<m  B So, based on this decision, are you saying AIX is doomed as well?   K >>> Go on, a lot of work has to be done to motivate us again (and god knowsa" how we want to see this happen)<<<  K I agree. And this work, albeit perhaps not that evident yet, is being done.f   Regards,  
 Kerry Main Senior Consultante Compaq Canada Inc. Professional Services, Voice: 613-592-4660d Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----0 From: eplan@kapsch.net [mailto:eplan@kapsch.net] Sent: January 7, 2001 6:55 PMh To: Info-VAX@Mvb.Saic.Com = Subject: RE: Compaq: A simple, affordable clustering solutionf    
 In articleJ <910612C07BCAD1119AF40000F86AF0D805284B97@kaoexc3.kao.cpqcorp.net>, "Main,& Kerry" <Kerry.Main@compaq.com> writes:L >>>> If CPQ had intended to outswap VMS into hibernation mode, they wouldn'tG >have bothered with V7.3, or with the Oracle partnership. I don't thinkiK >they'd be hiring developers, either. (If you're a Compaqtian and you bringf2 >in a qualified VMS hire, you get a $5K bounty)<<< >eH >Yep, and likely the openVMS Engineering group would not be aggressivelyG >hiring college graduates either ie. focus on long term, new idea's and  goingn >after youth.  >  >Reference: B ><http://www.compaq.com/inFORM/issues/issue32/human-int-29-b.html>   Come on, Kerry.r; We all know that OpenVMS engineering was never the problem. > The marketing, ads and sales were and unfortunately still are.  B Yes, the Q top management might have changed their mind about VMS,: but what we still see here is, that VMS is on its way out.  F And decisions where VMS is in the game are usually long term decision.D We see VMS decline at customers which decided a couple of years ago.A Some of them you might convince to keep with VMS, a greater blockiD you already lost and the rest will continue to phase VMS out becauseF they see nothing (except a small "me too" for some products) where VMSJ brings advantages to them. ("Cluster ? Hah ! Everyone has clusters now !")  F We see hiring new engineers, we see new VMS versions (just like in the past),K but what else ? eg. Does anyone tried to revert the EUREX (biggest european H stock exchange [system]) decision to drop VMS (and AIX; as a frontend) ?J Don't tell me the front-end has no meaning. It is mostly the core opsys ofH all attached banks and if VMS is dropped at EUREX, almost all banks willF start to drop VMS, too !! And this is the remaining key market where Q7 thinks VMS should reside. Still no alarm bell ringing ?a  E Remember AltaVista Firewall for OpenVMS ? It got cancelled many yearstL ago because of 1) no good TCPIP stack 2) too expensive 3) ergo no customers.E A firewall with cluster support built in is an advantage, but alas...e  K Or the new edu program (I can't use it). I see no happiness here in USENET. I I see some doubt (because of legal concerns). Without a broad acceptance,sH it will not bring harvest in the short run. And for a long run, there isI not enough time. I see no big spendings of cheap Alphas to edu institutesuH (If I see something from Q it is WINTEL, and this one is still better byI other producers like DELL, HP and even IBM. I still wonder how Q got #1).e  J David wrote umpteen times 'RIP VMS'. I begin to fear, he is right, becauseJ the speed (of revert the VMS damage) is too slow and too much time passed.H And I hate this, because I know now about 15 different operating systemsH and VMS is by far the best. But is seems, not the best products survive.  D And now consider, if I as a VMS bigot have fears about VMS's future,F what do you think what the big mass of uninformed MGMs/PHMs chained to WINTEL boxes think ?  G Go on, a lot of work has to be done to motivate us again (and god knowsa how we want to see this happen)S   -- y< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888g< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Sun, 07 Jan 2001 21:23:17 -0500e- From: JF Mezei <jfmezei.spamnot@videotron.ca>o= Subject: Re: Compaq: A simple, affordable clustering solutiono, Message-ID: <3A592490.33FBA313@videotron.ca>   "Main, Kerry" wrote:& > Could the marketing be better? Sure.  L No, it is not a question of "better" is is a question of "Could there be ANY marketing ?"  L Sending balls to existing customers was cool, but unles sthis isn't followed) up by something concrete, it was in vain.-  M > Does that mean that the sky is falling and the whole world is black becauseq( > you don't see OpenVMS in magazine ads?  L Expectations were set high with the balls and posters, and now, we are again7 let down with Compaq going back to its own wintel self.o  L > [and please, no rat holes of Digital did this, Digital did that, or Palmer  M Ahh, but it is just making the point that to some of us, Compaq's handling ofnK VMS is starting to look a lot like Palmer's handling of VMS (except perhapss' with a less active "kill VMS" portion).q  I If VMS were so profitable how come they can't see that marketing it wouldeL increase sales significantly because it would signify that Compaq is seriousK with VMS ? Compaq's refusal to showcase VMS is what is killing VMS, and therK way to solve this is to showcase VMS. That will send a strong signal to thei/ world. (both existing and potential customers).c  H You have to tell the world you have a product named "VMS" if you want to eventually sell that product.d   ------------------------------   Date: 8 Jan 2001 00:34:23 -0500e* From: young_r@eisner.decus.org (Rob Young)= Subject: Re: Compaq: A simple, affordable clustering solutiont+ Message-ID: <jQxEcRTm9lE8@eisner.decus.org>   R In article <93aqv5$esj$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes: > 6 > Main, Kerry <Kerry.Main@compaq.com> wrote in messageJ > news:910612C07BCAD1119AF40000F86AF0D805284B97@kaoexc3.kao.cpqcorp.net...	 >> Terry,i >>E >> >>> If CPQ had intended to outswap VMS into hibernation mode, theyS
 > wouldn'tI >> have bothered with V7.3, or with the Oracle partnership. I don't thinkeM >> they'd be hiring developers, either. (If you're a Compaqtian and you bringF4 >> in a qualified VMS hire, you get a $5K bounty)<<< >>J >> Yep, and likely the openVMS Engineering group would not be aggressivelyI >> hiring college graduates either ie. focus on long term, new idea's and  > going  >> after youth., > N > While some interesting ideas have indeed come out of academia, my impressionH > is that most of them have originated above the student level, and thatJ > experienced engineers are considerably better-qualified both to evaluateL > those ideas and to implement products based on them than even the students/ > who may have cooperated in the academic work.s > K > Isn't the most common reason to hire new graduates the fact that they are M > relatively inexpensive and relatively unpicky as to what they work on?  Cano > you say 'maintenance'? >   ? 	I don't know about maintenance.  Young and aggressive, 80 hour C 	weeks.  Lots of those in start-ups and whatnot.  At the D.C. DECUS B 	in 1995 I met this Cornell grad.. 22-24 years old working the VMSA 	booth.  A VMS engineer.  He had done this, was going to do that.-B 	Sounded real ambitious.  I am sure they pick those folks up every? 	now and then.  By the way, when I was a new grad I insisted onl@ 	getting a job using VMS.  That is what I knew.. but for someoneE 	to come to work for VMS engineering it probably means an opportunityoA 	to help develop one of finest OSes out there.  There aren't that:D 	many opportunities.  Only so many world-class OSes.  Now if someoneA 	were to try to lure you to Orem Utah to work on Netware, I would  	indeed be a bit nervous. . . .      				Robi    P ********************************************************************************P *                                RIP Netware                                   *P ********************************************************************************   ------------------------------  % Date: Mon, 08 Jan 2001 01:37:09 -0500e2 From: rdeininger@mindspring.com (Robert Deininger)= Subject: Re: Compaq: A simple, affordable clustering solutionhL Message-ID: <rdeininger-0801010137090001@user-2iveal5.dialup.mindspring.com>  { In article <910612C07BCAD1119AF40000F86AF0D805284B99@kaoexc3.kao.cpqcorp.net>, "Main, Kerry" <Kerry.Main@compaq.com> wrote:n    2 > >>> Or the new edu program (I can't use it). <<< > @ > Sure you can. The legaleze wording just needs to be clarified.   Kerry, Mr. Marcello told us nearly a month ago that this problem "will be fixed".  Have you seen any signs inside the Q that something is happening?   -- ( Robert Deininger rdeininger@mindspring.come   ------------------------------  # Date: Mon, 08 Jan 2001 01:47:59 GMTs From: m_stanifer@my-deja.com Subject: Re: DEC 3000 model 300s) Message-ID: <93b68d$grq$1@nnrp1.deja.com>   D I wonder if I am having the same type problem? I have a DEC 2000 300C AXP (jensen) I have tried loading Open VMS on it and when it starts F booting it gets to the same point everytime and the screen goes black.G I had thought it was about when the x-window server was taking over thet: display ... but I don't know vms much so I can only guess.C Recently I recieved a copy of OSF v3.2 I loaded that and all seemed @ well till it gets to the login screen then .... black screen ...F There are two 1 gig disks on this machine I want to run OSF on one andB VMS on the other ... if anyone has any help at all I would be most	 greatful.. Thanks
 Mike Stanifero       Sent via Deja.coms http://www.deja.com/   ------------------------------  % Date: Mon, 08 Jan 2001 01:49:50 -0500c2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: DEC 3000 model 300 L Message-ID: <rdeininger-0801010149500001@user-2iveal5.dialup.mindspring.com>   In article <rdeininger-0601011808380001@user-2iveb5h.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) wrote:  H > In article <9365d9$ro3$1@nnrp1.deja.com>, townleycc@my-deja.com wrote: >  > > Finally got back here. > > 1 > > Just obtained DEC 3000 model 300 Workstation,e4 > > with (I think) Dunix on it. Only problem is that2 > > I cannot get a display on my VR297-V3 monitor.2 > > This I think will handle 60 Hz at 1024X864 but3 > > the WS video outpu is probably higher. I cannotS5 > > identify the video card as it seems to integratedf3 > > on he M/B, and no obvious jumpers. I have had ao4 > > look at the dec3000.pdf from the compaq site,but  > > this refers to later models.. > > Elseways, can I enable some sort of serial2 > > console (without initial console access!) so I3 > > can stick a serial cable onto the printer port?  > L > You can hook a terminal to it, if you get the right cable.  I don't recallP > the details for the model 300.  We have a couple of them at work, and a manual > somewhere.  > Filling in some details, now that I have the manual to hand...   The alternate console port on the DEC 3000 300 family systems is a DB-25 connector RS-232 port.  9600 Baud, no parity, 1 stop bit, 8 data bits.T  G If you turn off the system and remove the keyboard/mouse cable, it willoD switch to the alternate console (terminal) when you turn it back on.  J Or you can use the jumper to force it to use the alternate console even ifG the keyboard/mouse cable is still connected.  The jumper is labeled W2.n> Connect pins 1 and 2 to use the graphics port for the console.F Connect pins 2 and 3 to use the terminal/printer port for the console. Pins 3 and 4 are reserved.  L You are supposed to use an anti-static wristband when you change the jumper.( And the system should be off, obviously.   This is from the DEC 3000 Model 300/300L AXP Harware Reference Guide, part number EK-PELCN-OG.B01.  Chances are, the same applies to the models 300X and 300LX.  But I make no promises.   -- v Robert Deininger rdeininger@mindspring.coml   ------------------------------  % Date: Sun, 07 Jan 2001 14:04:46 -0600 + From: "Main, Kerry" <Kerry.Main@compaq.com>t, Subject: RE: Ebay dead as a post - yet againN Message-ID: <910612C07BCAD1119AF40000F86AF0D805284B91@kaoexc3.kao.cpqcorp.net>   Paul,t  F >>> Why mirrored ( I assume controller shadowing here ) and shadowing,( instead of say a 3 member shadow set? <<   A few thoughts on this -  I For large, high profile sites is that it makes it easier to implement andnD support multi-site clusters and provides a good backup mechanism forA applications / databases that do not yet support online backups. l  K Inevitably, most large Customer sites now recognize that it is no longer aneH acceptable business risk to be out of business if they have a datacenter? fire (rest of building ok - pick favourite disaster scenario as F alternative). Many are now looking at how best to implement multi-site business continuity plans.  J Based on these reasons, I usually recommend a combination of hardware raid1 and host software raid for the following reasons:l  K - like network costs in most large area's, disks are relatively cheap todays! and getting cheaper all the time.u  J - For multi-site clusters, the loss of a single disk means that when it isK fixed, a mix of HW/HBS shadowing means there is no multi-site link or localeK system / IO adapter impact to rebuild the disk. The disk is rebuilt locallye  by the disk hardware controller.  G - for applications / databases that do not support online backups, this B approach can be very useful from an availability point of view. AnK application can be shutdown (or better - quiesced) for a couple of minutes,oF single members dismounted from the dual member shadowset, applicationsK restarted and the single member disks backed up to tape. When the backup istJ complete, the single members are remounted online as part of the shadowset8 with no availability impact on the current applications.  H By mixing HBS dual member volumes with HW raid, this process can be doneD without the worry of a single disk failure impacting the applicationC availability while the single member is being backed up. If plannedgF correctly (placing of drives/controllers), these single members can beD backed up with minimal impact on the currently running applications.  J ** Note: After the backup is done, this requires the new member to catchupL to the current drives (additional IO's), the application availability is notH impacted. Coming in OpenVMS V7.3 (now in FT2) will be the ability to useK this procedure and then when the member is remounted, only the changes willeI need to be applied to the re-entering member, this significantly reducingc' the amount of IO's required to re-sync.r   Regards,  
 Kerry Main Senior Consultant1 Compaq Canada Inc. Professional Services. Voice: 613-592-4660e Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----3 From: Paul Repacholi [mailto:prep@prep.synonet.com]d Sent: January 6, 2001 6:38 AMs To: Info-VAX@Mvb.Saic.Comh, Subject: Re: Ebay dead as a post - yet again    , young_r@eisner.decus.org (Rob Young) writes:   >.  But when I fliprF > 	open my "Preferred Way To Set Up High Availability Clusters" manual< > 	I see I want my QMAN files on mirrored disks, preferrably> > 	mirrored and THEN shadowed just in case a controller hangs.. > 	Disks are cheaper and cheaper all the time.  B Why mirrored ( I assume controller shadowing here ) and shadowing,% instead of say a 3 member shadow set?i   Just curious...h   -- a< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.l@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Sun, 07 Jan 2001 14:17:12 -0600 + From: "Main, Kerry" <Kerry.Main@compaq.com>w, Subject: RE: Ebay dead as a post - yet againN Message-ID: <910612C07BCAD1119AF40000F86AF0D805284B92@kaoexc3.kao.cpqcorp.net>   G'day Terry,  L >>> eBay claims its site has demonstrated greater than 99% uptime and backupI systems have functioned without incident during the last four quarters.<<   G As many vendors (including Compaq) are now finding out, the age old "we J don't count scheduled system downtime" in our high availability numbers isI rapidly becoming UNACCEPTABLE - if it impacts application availability in  any way.  J This practice is a throw back to the old mainframe environments where theyJ used to (and many still do) go down for weekly/monthly/quarterly scheduledK maintenance. In those days, Customers agreed that this downtime did not get3, counted in the vendors availability numbers.  K In todays eWorld and mission critical environments, the reality is that theoA high availability bar is being raised a number of notches higher.b  K Imho, Customers should ask all vendors: Does your high availability numbers L include the time for scheduled system related downtime like HW upgrades / HW< replace, OS upgrades, OS patches, firmware upgrades etc. ???   If not, why not?   :-)    Regards,  
 Kerry Main Senior Consultanto Compaq Canada Inc. Professional Servicesr Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----9 From: Terry C. Shannon [mailto:terryshannon@mediaone.net]  Sent: January 4, 2001 10:30 PM To: Info-VAX@Mvb.Saic.Come, Subject: Re: Ebay dead as a post - yet again    D According to Information Week, eBay had a big-time visit from Mister
 Murphy....  E Auction site eBay Inc. was plagued by crashes and failures Wednesday,rJ causing the site to go offline for almost 11 hours... the company said theL outage was the result of a series of failures, compounded by the decision toH put off needed hardware repairs until after the busy holiday season, and' that it intends to make rapid upgrades.iJ The outage began about 11:34 a.m. PST Wednesday when hardware in a back-upD system malfunctioned. When that system was rebooted, another problemK developed in storage shared by both the primary and back-up systems, takingeG down the entire site. A third backup was activated for approximately 40sF minutes before crashing as well. The site wasn't back online and fully operational until 9:56 p.m.   H eBay claims its site has demonstrated greater than 99% uptime and backupG systems have functioned without incident during the last four quarters.h   ------------------------------   Date: 8 Jan 2001 00:13:01 -0500 * From: young_r@eisner.decus.org (Rob Young), Subject: Re: Ebay dead as a post - yet again+ Message-ID: <vUZhGeIMtPKM@eisner.decus.org>M  _ In article <87ae943o80.fsf@k9.prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:e. > young_r@eisner.decus.org (Rob Young) writes: >  >>.  But when I flipG >> 	open my "Preferred Way To Set Up High Availability Clusters" manualt= >> 	I see I want my QMAN files on mirrored disks, preferrablyo? >> 	mirrored and THEN shadowed just in case a controller hangs.v/ >> 	Disks are cheaper and cheaper all the time.  > D > Why mirrored ( I assume controller shadowing here ) and shadowing,' > instead of say a 3 member shadow set?c >  > Just curious...  >    	Because I can ;-)  = 	No... but seriously.  You may stick QMAN files on a database < 	volume that isn't that hot.  Something you are breaking out= 	to backup. While the shadowset is broken out, you don't want C 	to have a hardware hit of the master take you down.  Small window? = 	Maybe 3-4 hours per day.  Murphy would get you.  I was beingiA 	bothersome.  You could most likely get by with a RAID1 mirrorsett. 	or a shadowset but certainly don't need both.   				Robb   ------------------------------  # Date: Mon, 08 Jan 2001 02:13:31 GMTe From: kparris@my-deja.comu* Subject: Re: Flackey fibre channel support) Message-ID: <93b7o7$i41$1@nnrp1.deja.com>   A It sounds like the basic issue here is that there is presently noLG support in VMS for failover from a direct (Fibre Channel) connection tonE an MSCP-served connection.  That would allow continued access to bothNF shadowset members from both sites despite the failure of an inter-siteF FC link.  That functionality is slated for a future release of VMS, we were told at CETS2000.  D In the interim, providing an MSCP-served path alone (i.e. using MSCPF serving to access FC disks at the opposite site _instead_ of using theE intuitively-obvious inter-site FC link) is probably the only safe way:F to go for now if you need very high availability or disaster toleranceC in a multi-site cluster under VMS.  You'll want redundancy for thatn inter-site SCS link, of course.o  C Fibre Channel's appeal is that it has higher bandwidth than CI, andtD frankly, because it is seen as newer, sexier, "open", and "industry-E standard" technology.  Bandwidth and other performance considerationstE are but one factor in an overall solution targetted at providing very A high availability or disaster tolerance (i.e. if the system can'tiE perform as well as needed, in many cases it may well be considered topB be "down" or unavailable).  But acceptable performance is only one factor.t  E From what I've observed lately, [probably due to the lack of a viableeA CI replacement strategy] Compaq seems to be pushing Fibre ChanneloA technology hard in the VMS installed base, and as a result, it islE getting into some VMS production sites that it is not really ready tot handle yet.   C I'm in wait-and-see mode right now.  At least until I see Fast_Path D support for Fibre Channel (slated for 7.3), mini-merge support (with@ longer-term prospects than the HSG80 alone), and especially moreG stability/maturity in the hardware/switches/drivers, etc., I personally C won't advise using Fibre Channel in a mission-critical or disaster-r tolerant VMS cluster.uG ------------------------------------------------------------------------D Keith Parris|Integrity Computing,Inc.| parris(at)decuserve.decus.orgF VMS Consulting: Clusters, Perf., Alpha porting, Storage&I/O, Internals     Sent via Deja.comT http://www.deja.com/   ------------------------------   Date: 8 Jan 2001 01:09:55 -0500f* From: young_r@eisner.decus.org (Rob Young)* Subject: Re: Flackey fibre channel support+ Message-ID: <eo7bY11sxAqx@eisner.decus.org>c  E In article <93b7o7$i41$1@nnrp1.deja.com>, kparris@my-deja.com writes:aC > It sounds like the basic issue here is that there is presently no I > support in VMS for failover from a direct (Fibre Channel) connection toeG > an MSCP-served connection.  That would allow continued access to botheH > shadowset members from both sites despite the failure of an inter-siteH > FC link.  That functionality is slated for a future release of VMS, we > were told at CETS2000. >   : 	But couldn't you just as easily build out your fibre such< 	that loss of a single link doesn't "bite" you?  I'm looking= 	at that now and multipath failover would save the day there.v  F > In the interim, providing an MSCP-served path alone (i.e. using MSCPH > serving to access FC disks at the opposite site _instead_ of using theG > intuitively-obvious inter-site FC link) is probably the only safe way H > to go for now if you need very high availability or disaster toleranceE > in a multi-site cluster under VMS.  You'll want redundancy for thatk! > inter-site SCS link, of course.? >   @ 	Memory channel and Ethernet.  Shorter distances (10K and under)9 	are easier than maybe for some customer's situations ;-)D  E > Fibre Channel's appeal is that it has higher bandwidth than CI, andeF > frankly, because it is seen as newer, sexier, "open", and "industry-G > standard" technology.  Bandwidth and other performance considerationswG > are but one factor in an overall solution targetted at providing veryjC > high availability or disaster tolerance (i.e. if the system can'tkG > perform as well as needed, in many cases it may well be considered tooD > be "down" or unavailable).  But acceptable performance is only one	 > factor.a >   > 	Some of this bandwidth stuff would admittedly be a wash.  ButA 	I think it would help alot with backup speeds.  If you are using @ 	newer drives -- Seagates at 15K can burst to 160 MB/s (Ultra 3)= 	and sustain 55 MB/s (IIRC).  Couple that with fibre switches ? 	and fibre controllers and multiple tape drives you can see how*? 	you can shrink those backup windows back down.  With 7.3 , the*9 	very painful merges after backup go away with mini-copy.0    G > From what I've observed lately, [probably due to the lack of a viablecC > CI replacement strategy] Compaq seems to be pushing Fibre Channel C > technology hard in the VMS installed base, and as a result, it isnG > getting into some VMS production sites that it is not really ready tot
 > handle yet." > E > I'm in wait-and-see mode right now.  At least until I see Fast_Path F > support for Fibre Channel (slated for 7.3), mini-merge support (withB > longer-term prospects than the HSG80 alone), and especially moreI > stability/maturity in the hardware/switches/drivers, etc., I personallytE > won't advise using Fibre Channel in a mission-critical or disaster-e > tolerant VMS cluster.-  F 	All the work for HSG80 mini-merge would somehow not make it into the  	follow-on?  Why?D   				Robe   ------------------------------  % Date: Sun, 07 Jan 2001 20:16:03 -0600n+ From: "Main, Kerry" <Kerry.Main@compaq.com>0= Subject: RE: Help for ORACLE v7.3 on VMS v7.1 and TCPIP v5.0AsN Message-ID: <910612C07BCAD1119AF40000F86AF0D805284B9A@kaoexc3.kao.cpqcorp.net>   Jeremy,h  H I am certainly no Oracle expert, but I seem to remember an issue wherebyL Oracle Server defaults to listening on one of two possible ports ... Are the= Oracle config files on the client pointing to the right port?l   Regards,  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Servicest Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----2 From: JBUTLER@utmem.edu [mailto:JBUTLER@utmem.edu] Sent: January 1, 2001 1:04 PM- To: Info-VAX@Mvb.Saic.Como9 Subject: Help for ORACLE v7.3 on VMS v7.1 and TCPIP v5.0A.    C 	I have OpenVMS v7.1 Alpha and ORACLE v7.3.2 and was using UCX v4.2eL with ECOs.  I upgraded to TCPIP v5.0A ECO1 and the ORACLE Listener no longerH works.  It seems to start up and you can successfully telnet to the port2 being used, but the SQLNET connection gives error.> 	Does anyone have experience with this?  Do you have any hints( for why the version of TCP would matter? 			Thanks, Jenny Butlerm 			JBUTLER@UTMEM.EDU   ------------------------------  $ Date: Mon, 8 Jan 2001 13:05:41 +13009 From: "Antony Wardle" <antony.wardle@nnnoospam.met.co.nz>d3 Subject: Re: Need FTSS V4.2 or higher for VAXft 810r/ Message-ID: <gq766.109$Ll4.4280@ozemail.com.au>c   Mine is the same.t   I have got all the vax cds.    Was it renamed at some point?    antony   sear it:[cd]*vax*98.txt FTSS   ****************************** IT:[CD]VAX_BIN_MAR98.TXT;1   "VAXft System Services"l YEAAA  2* .0   SSB  NCH  Y  N  N  FTSS020          3      , <LBohan@dbc.spam_less..com> wrote in message2 news:agaf5t04ufqg5s2ji4g7ov79b8fr2v7pkm@4ax.com...C > On Fri, 05 Jan 2001 20:26:35 -0600 (CST), sms@antinode.org wrote:h > J > >   Sadly, March 1996 is the latest SPL I have for VAX.  Perhaps someoneH > >with a newer one has an [FTSSxxx] directory where xxx > 020.  It's inI > >the Master Index as "VAXft System Services".  (Or some one could offeraJ > >to sell me a newer SPL for almost no money, and I'd look through it for	 > >free.)e > >dK > >------------------------------------------------------------------------u > > F > >   Steven M. Schweda               (+1) 651-699-9818  (voice, home)F > >   382 South Warwick Street        (+1) 763-781-0308  (voice, work)J > >   Saint Paul  MN  55105-2547      (+1) 763-781-0309  (facsimile, work)< > >   sms@antinode.org                sms@provis.com  (work) >hF > and I took a look at a Sep 1997 VAX SPL, and FTSS was still at v2.0. >t# > $ search dka600:[cdrom]*.dat FTSSm >.( > DKA600:[000000.CDROM]CD_CONTENTS.DAT;1 > B > "VAXft System Services" YEAAA  2.0  SSB  NCH  Y  N  N  FTSS020 3 >a   ------------------------------  % Date: Sun, 07 Jan 2001 15:34:08 -0500X- From: JF Mezei <jfmezei.spamnot@videotron.ca>o Subject: Online backupsr, Message-ID: <3A58D2BD.24CE8CD9@videotron.ca>  K Ok, Kerry raised the question of applications that support on-line backups.@  H What exactly is involved/expected for an application to support a backup1 operation without the application shutting down ?p   ------------------------------   Date: 7 Jan 2001 17:17:43 -0500i9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)s Subject: Re: Online backupsh+ Message-ID: <EsEyDmiHZen$@eisner.decus.org>   \ In article <3A58D2BD.24CE8CD9@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:M > Ok, Kerry raised the question of applications that support on-line backups.  > J > What exactly is involved/expected for an application to support a backup3 > operation without the application shutting down ?y  C The application must ensure the data currently on disk represents a C coherent point in the history of the data -- not having some updatem3 halfway to disk with the other half just in memory.n  < This is trivial for an application where RMS records bear no? relationship to each other -- seldom the way applications work.n   ------------------------------  $ Date: Sun, 7 Jan 2001 17:44:21 -0500% From: "John Vottero" <John@mvpsi.com>  Subject: Re: Online backups / Message-ID: <t5hsbfru99ah04@news.supernews.com>m  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3A58D2BD.24CE8CD9@videotron.ca...D > Ok, Kerry raised the question of applications that support on-line backups. >oJ > What exactly is involved/expected for an application to support a backup3 > operation without the application shutting down ?i  B 1. Use a database which supports online backups (like Oracle Rdb).  C 2. Make sure that a "business transaction" doesn't span a "databaset
 transaction".    ------------------------------  # Date: Mon, 08 Jan 2001 04:08:24 GMTi4 From: "Terry C. Shannon" <terryshannon@mediaone.net>) Subject: The Old VMS Marketing Mares-nesti< Message-ID: <Y2b66.42817$1t.2516952@typhoon.ne.mediaone.net>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3A59181A.F240B356@videotron.ca... > "Main, Kerry" wrote:K > > Yep, and likely the openVMS Engineering group would not be aggressively J > > hiring college graduates either ie. focus on long term, new idea's and goingl > > after youth. >tK > Sorry, this does absolutely nothing to me. During the worse of the PalmertD > "Digital is killing VMS" years, VMS got new versions, and got some promisesL > from Digital that Digital would continue to support and improve VMS for at$ > least 5 years yada yada yada yada. >eL > So far, I have seen those arguments currently used by Compaq to defend itsJ > lack of marketing of VMS during the Palmer years. So, I am sorry to say,> > Compaq has a long way to go to regain my confidence in them. >o > <steam venting mode>L > For crying out loud, why is so goddam impossible for Compaq to take just a bitpK > of VMS's profits and make some public advertising so that at least we arepI > given the impression that Compaq isn't affraid to publicly admit it has  VMS ?:K > Right now, I get the feeling that Compaq feels that VMS is a disease that) is; > socially unacceptable so it won't talk about it publicly.t  ! VMSilis? VMSorrhea? VMSchancroid?l  K On at least three occasions in the past six months I've seen and heard Rich H Marcello explicitly state that the early 90's were a bad time indeed forF OpenVMS (several interesting quotes to that effect in my newsletter; aI couple of pertinent issues are on the Acersoft site for all and sundry to.I peruse). Rich most certainly was not defending the behavior, tactics, and6> strategy of his predecessor or the senior managers at Digital.  H There are very few high-ranking Digital people still employed at Compaq.K Ironically, one of these people is Don Harbert, who ran the Unix group back0J in the bad old days. Don is now in charge of all things Alpha-related, andL thus VMS falls under his purview. Since VMS represents a significant portionE of Don's revenue stream, he certainly isn't going to malign the OS ory subject it benign neglect.  K In what venues should Compaq advertise VMS? It's easy to suggest televisioneK advertising, but how does one focus the message on prospective buyers? ThateH said, Compaq COULD raise the profile of the OS at a very low incrementalK cost via due diligence in the marketing department: ensure that the OS getsmI mentioned whenever and wherever it's appropriate to do so (annual report,c5 earnings announcements, benchmark results, etc, etc).   J CPQ now has the OpenVMS Times, which is more than the firm had a year ago.K Problem is, this publication targets the installed base (which is not a badnK thing to do since it costs ~6X as much to acquire a customer as to maintainnH one) but not does not go beyond the base. The same can be said for whiteF papers, technical data, presentations, etc... these are available, butK require those interested to initiate action (e.g. visit the Web site, which < is easier said than done from the www.compaq.com home page).  I I would argue--and by no means for the first time--that Compaq marketeers,H ought to be proactively and aggressively engaging the trade and businessL press when newsworthy VMS happenings happen. The Fourth Estate won't come toF you unless you're the Next New Thing (Linux, dot coms, whatever). As IJ learned in Viet Nam, "winning hearts and minds" is most effective when youK grab 'em by the hair (short or otherwise)... the hearts and minds will theni follow.e   ------------------------------  % Date: Sun, 07 Jan 2001 16:26:07 -0500w! From: Beyonder <beyonder@vrx.net>W@ Subject: vax basic - can someone remind me how this works again?8 Message-ID: <0inh5t4cdu6mih9mhlm321o7hfsnhiakgh@4ax.com>  ' Seems I've forgotten the simple things.u% to read in a single byte from a file,e open "test.txt" for input as #1h input #1,a$   $ but no gaurantee it's a single char?* how to output a single char without cr/lf? print #1,chr(32);y  ) and I remember linput for whole lines ando) standard print, but for whatever reason Is* can't remember how to do file manipulation at the char/byte level...   $ it's been waaay tooo long I think...  ( any memory jogging would be appreciated.   B.   ------------------------------  $ Date: Sun, 7 Jan 2001 15:43:28 -08001 From: "Randy Park" <rjpark@mindspring.com.nospam>%D Subject: Re: vax basic - can someone remind me how this works again?2 Message-ID: <93auuv$nok$1@slb6.atl.mindspring.net>  , Beyonder <beyonder@vrx.net> wrote in message2 news:0inh5t4cdu6mih9mhlm321o7hfsnhiakgh@4ax.com...) > Seems I've forgotten the simple things. ' > to read in a single byte from a file,t! > open "test.txt" for input as #1t
 > input #1,a$a >l& > but no gaurantee it's a single char?, > how to output a single char without cr/lf? > print #1,chr(32);p >>+ > and I remember linput for whole lines anda+ > standard print, but for whatever reason I , > can't remember how to do file manipulation > at the char/byte level...j > & > it's been waaay tooo long I think... > * > any memory jogging would be appreciated. >i > B.  + You didn't forget.  There is no function ind* VAX/DEC BASIC to read only one byte from a- file.  There is a function, INKEY$, that willc/ read one character from a keyboard, but I don'tA- think that is what you want. (I think it also  uses the SMG routines.)r  - After 25 years of programming in DEC flavoredo. BASIC, I've never encountered the need to read0 only one byte from a file, unless perhaps I knew- the record was only one byte.  BASIC likes to - process records rather than individual bytes.l  2 There are two ways to accomplish what I'm guessing you want to do:e  3 1.  Use the QIO system service.  Not easy if you've  never used it.  0 2.  Write your own "Read a Byte" routine that on0 the first call reads a record from the file, but0 returns only the first byte.  Then on the second3 call, returns the second byte.  When a call is maden/ and there are no more bytes in the record, thenb read the next record.t  . Of these solutions, I would personally use the0 second if I needed to process only one byte at a time.e  ) Also, be sure to use HELP/LIBR=BASICHELP.n   ------------------------------  % Date: Mon, 08 Jan 2001 00:19:06 -0500 ! From: Beyonder <beyonder@vrx.net>vD Subject: Re: vax basic - can someone remind me how this works again?8 Message-ID: <k4ji5tc8dqtuj54pvlui2m1i102pa008p0@4ax.com>  / On Sun, 7 Jan 2001 15:43:28 -0800, "Randy Park"n% <rjpark@mindspring.com.nospam> wrote: , >You didn't forget.  There is no function in+ >VAX/DEC BASIC to read only one byte from an. >file.  There is a function, INKEY$, that will0 >read one character from a keyboard, but I don't. >think that is what you want. (I think it also >uses the SMG routines.)  ( yeah, keyboard routines don't help much.  . >After 25 years of programming in DEC flavored/ >BASIC, I've never encountered the need to reada1 >only one byte from a file, unless perhaps I knews. >the record was only one byte.  BASIC likes to. >process records rather than individual bytes.  2 never? that's amazing. what if you need to process7 a "binary" file? what about file encryption/decryption?': what about a million other uses where the file isn't text?9 I even play with manipulating EXE files on a daily basis.h5 then there's graphic images, raw files, and so forth.a  3 >There are two ways to accomplish what I'm guessingt >you want to do: >l4 >1.  Use the QIO system service.  Not easy if you've >never used it.e  2 well if I've never heard of it, probably not easy.  1 >2.  Write your own "Read a Byte" routine that onr1 >the first call reads a record from the file, buto1 >returns only the first byte.  Then on the secondv4 >call, returns the second byte.  When a call is made0 >and there are no more bytes in the record, then >read the next record.  6 that won't work at all. there are no "records" in pure4 binary files. ie: EXEs, or encrypted files, or other; types, like graphic images, and so forth. the list goes on.a  / >Of these solutions, I would personally use the.1 >second if I needed to process only one byte at a- >time.  6 won't work. for obvious reasons. will mangle the file.  * >Also, be sure to use HELP/LIBR=BASICHELP.   will do!  ; BTW: I thought there was something I could do in the "OPEN"tA statement that might help. but I'll have to check the HELP again.    B.   ------------------------------  $ Date: Sun, 7 Jan 2001 22:45:51 -08001 From: "Randy Park" <rjpark@mindspring.com.nospam>nD Subject: Re: vax basic - can someone remind me how this works again?3 Message-ID: <93bnus$gqj$1@nntp9.atl.mindspring.net>e  , Beyonder <beyonder@vrx.net> wrote in message2 news:k4ji5tc8dqtuj54pvlui2m1i102pa008p0@4ax.com...1 > On Sun, 7 Jan 2001 15:43:28 -0800, "Randy Park"d' > <rjpark@mindspring.com.nospam> wrote:r. > >You didn't forget.  There is no function in- > >VAX/DEC BASIC to read only one byte from ah0 > >file.  There is a function, INKEY$, that will2 > >read one character from a keyboard, but I don't0 > >think that is what you want. (I think it also > >uses the SMG routines.) >e* > yeah, keyboard routines don't help much. >p0 > >After 25 years of programming in DEC flavored1 > >BASIC, I've never encountered the need to read 3 > >only one byte from a file, unless perhaps I knewu0 > >the record was only one byte.  BASIC likes to0 > >process records rather than individual bytes. >a4 > never? that's amazing. what if you need to process9 > a "binary" file? what about file encryption/decryption?h< > what about a million other uses where the file isn't text?; > I even play with manipulating EXE files on a daily basis.d7 > then there's graphic images, raw files, and so forth.i >w  6 The only file I've seen on VMS that were not organized6 as records are the page/swap files.  EXEs are 512 byte4 fixed length records.  I've modified EXEs by reading4 the block (record in this case) I want, and changing1 the target byte(s) within the block.  The graphicn4 image files I've seen have also been sequential with4 512 byte fixed length records.  If you do a DIR/FULL  command, you'll see what I mean.  5 > >There are two ways to accomplish what I'm guessingc > >you want to do: > >r6 > >1.  Use the QIO system service.  Not easy if you've > >never used it.f >e4 > well if I've never heard of it, probably not easy. >y3 > >2.  Write your own "Read a Byte" routine that on.3 > >the first call reads a record from the file, buti3 > >returns only the first byte.  Then on the second 6 > >call, returns the second byte.  When a call is made2 > >and there are no more bytes in the record, then > >read the next record. >S8 > that won't work at all. there are no "records" in pure6 > binary files. ie: EXEs, or encrypted files, or other= > types, like graphic images, and so forth. the list goes on.  >s  3 Sure it will.  Read the target block/record, modifys3 the byte(s), write the block/record back out.  Realr4 simple, just as if you were using assembly language.  1 > >Of these solutions, I would personally use ther3 > >second if I needed to process only one byte at aa > >time. >s8 > won't work. for obvious reasons. will mangle the file. >t, > >Also, be sure to use HELP/LIBR=BASICHELP. >k
 > will do! > = > BTW: I thought there was something I could do in the "OPEN"-C > statement that might help. but I'll have to check the HELP again.y >i  3 Nope.  BASIC is not like C when it comes to I/O, itS6 wants to I/O using records, which is really ok because6 practically every file on VMS is organized in records.1 Really...  Do a DIR/FULL command on your file. Ito3 probably has a sequential organization.  The record + format might be fixed, variable, or stream.e   ------------------------------   End of INFO-VAX 2001.015 ************************