1 INFO-VAX	Mon, 13 May 2002	Volume 2002 : Issue 264       Contents: 32 bit and 64 bit references  Re: 32 bit and 64 bit references$ Re: Bob Palmer and the demise of DEC$ Re: Bob Palmer and the demise of DEC$ Re: Bob Palmer and the demise of DEC$ Re: Bob Palmer and the demise of DEC$ Re: Bob Palmer and the demise of DEC Re: Bug in ANALYZE/AUDIT ?3 Re: Capellas: Linux, Windows Will 'Eviscerate' Unix 3 Re: Capellas: Linux, Windows Will 'Eviscerate' Unix 3 Re: Capellas: Linux, Windows Will 'Eviscerate' Unix 3 Re: Capellas: Linux, Windows Will 'Eviscerate' Unix 3 Re: Capellas: Linux, Windows Will 'Eviscerate' Unix 3 Re: Capellas: Linux, Windows Will 'Eviscerate' Unix  Compaq auction- Re: cxxlink shared libraries on Alpha and VAX ' EV9 (Tarantula) paper. Also EV8 details  freeVMS 4 Re: Gartner rides again (was Re: HP Product Roadmap)& Re: High Water Concurrent Users Count?& Re: High Water Concurrent Users Count?& Re: High Water Concurrent Users Count?+ Re: howto create self-extracting zip files? + Re: howto create self-extracting zip files? + Re: howto create self-extracting zip files? + Re: howto create self-extracting zip files? + Re: howto create self-extracting zip files? + Re: howto create self-extracting zip files? + Re: howto create self-extracting zip files?  HP to selectively market VMS? G Re: MPE "clusters" (Was: Re: Capellas: Linux, Windows Will 'Eviscerate' % Network device for wide area clusters ) Re: Network device for wide area clusters B new academic programme (was: RE: Bob Palmer and the demise of DEC) Re: No new Alpha sales Re: No new Alpha sales% Re: Non-interactive TECO? (of course) ! Non-interactive TECO? (of course) ! Non-interactive TECO? (of course) % Re: Non-interactive TECO? (of course)  NULL device and print queues  Re: NULL device and print queues  Re: NULL device and print queues? Re: Old RAID 230 logical init very slow (63% done after 5 days)  Re: Powered by HP  Re: Powered by HP  Re: Powered by HP " Re: simple disk-shadowing question+ Re: Some more words of wisdom from Capellas + Re: Some more words of wisdom from Capellas + Re: Some more words of wisdom from Capellas 1 Re: SOT: Yo Andrew, you guys on a roll this week? 1 Re: SOT: Yo Andrew, you guys on a roll this week?  Re: Stallards smoking gun! Re: Stallards smoking gun! Re: Stallards smoking gun! Re: Stallards smoking gun! Re: Sun to use AMD chips ... Re: Switching the console mode Re: Switching the console mode Terminal Services on VMS client # Re: Terminal Services on VMS client  Re: USB on OpenVMS Re: USB on OpenVMS0 Re: What is good model for disk i/o w/shadowing?0 Re: What is good model for disk i/o w/shadowing?0 Re: What is good model for disk i/o w/shadowing?0 Re: What is good model for disk i/o w/shadowing?0 Re: What is good model for disk i/o w/shadowing? Re: Who cares about marketing! Re: Who cares about marketing! Re: Who cares about marketing! Re: Who cares about marketing! Re: Who cares about marketing! Re: [announce] FreeVMS 0.0.14  Re: [announce] FreeVMS 0.0.14 0 Re: [announce] FreeVMS 0.0.15 was FreeVMS 0.0.14  F ----------------------------------------------------------------------    Date: 13 May 2002 02:51:42 -0700 From: cbdeja@my-deja.com (-)% Subject: 32 bit and 64 bit references = Message-ID: <611952a3.0205130151.1b6bdd5f@posting.google.com>   9 I am a VMS novice, so this is probably a simple question!   B In the sys$getjpi() and sys$getjpiw() call the documentations saysF that you can supply an item list on the Alpha as a 32 bit reference or a 64 bit reference:   ;   OpenVMS usage: 32-bit item_list_3 or 64-bit item_list_64b F   type: longword (unsigned) for 32-bit; quadword (unsigned) for 64-bit   access: read only /   mechanism: by 32- or 64-bit reference (Alpha) &   mechanism: by 32-bit reference (VAX)  8 For a 32 bit reference the item descriptor structure is:	 item_code 
 buffer_length  buffer_addr  return_length_addr  F But for a 64 bit reference the item descriptor structure if different:	 item_code  MBO  MBMO
 buffer_length  buffer_addr  return_length_addr  = My question is. how does the compiler know which of these two A structures you have decided to use? Do you have to use a compiler  switch?   F I assume that if you get it wrong the system will read the wrong parts of the item structure.  : If it matters, I am developing on a 64 bit Alpha but usingC /POINTER_SIZE=32 , and the function I'm writing must also work on a  VAX.   Thanks   Colin    ------------------------------  % Date: Mon, 13 May 2002 11:29:59 +0100 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>) Subject: Re: 32 bit and 64 bit references + Message-ID: <abo4im$o98@newton.cc.rl.ac.uk>   a "-" <cbdeja@my-deja.com> wrote in message news:611952a3.0205130151.1b6bdd5f@posting.google.com...   ? > My question is. how does the compiler know which of these two % > structures you have decided to use?   D It doesn't need to; they are passed by reference. The system serviceJ can check the MBO fields and figure out which structure you have supplied.< All it needs to do is to sign extend the address, if 32-bit.   ------------------------------  % Date: Mon, 13 May 2002 12:27:38 +0100 U From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> - Subject: Re: Bob Palmer and the demise of DEC 0 Message-ID: <abo8dj$19b$2@new-usenet.uk.sun.com>   Nic Clews wrote:   > JF Mezei wrote:  > * >>Andrew Harrison SUNUK Consultancy wrote: >>7 >>>If is makes you feel any better Sun has a OpenVMS to $ >>>Solaris Affinity Program as well. >>> O >>Andrew, if one wanted to make some DECUS presentation to local folks, would a M >>local SUN office be able to provide technical information on such program ? L >>(not a sales pitch, but a real technical presentation of the program tools >>etc) ? >>E >>Or is there someone at Sun headquarters who could travel for this ?  >> > B > I certainly hope that the UNIX portability initiatives get equalD > 'airplay' from the relevant folks, where of course Sun is a target > source platform. >     7 That would be COE which Sun was involved in specifying.    Regards  Andrew Harrison    ------------------------------   Date: 13 May 2002 14:21:42 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) - Subject: Re: Bob Palmer and the demise of DEC , Message-ID: <aboi5m$167m$1@info.cs.uofs.edu>  = In article <d7791aa1.0205101548.113eadcc@posting.google.com>, +  bob@instantwhip.com (Bob Ceculski) writes: m |> "Peter Weaver" <peter.weaver@stelco.ca> wrote in message news:<abh22j$i23dt$1@ID-141708.news.dfncis.de>... C |> > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message * |> > news:abgv9p$ejl$4@info.cs.uofs.edu...	 |> > >... N |> > > It will happen no more than one year from the day HP decides it will noM |> > > longer issue License Paks for VMS.  Unless you figure all those really M |> > > big businesses that are still running VMS will be willing to run their A |> > > DP operations on a machine with the clock set back a year.  |> >  	 |> > >...  |> >  O |> > Bill, you're thinking about educational licenses and/or hobbyist licenses. ' |> > The licenses we buy do not expire.  |>  G |> maybe Bill calls in every 30 days and gets another trial license ...   J Bill doesn't ever call them.  All contact with Compaq has been through theJ University's datacenter.  All I knew was that every October my VAX stoppedJ working and I had to get a set of PAKS for it.  As people have said, it is7 a CSLG thing and does not apply to commercial licenses.   E And for those who are concerned about the future of VMS and how it is H affected by loss of educational exposure, I am pretty sure my VAXen haveG been dropped from the University's CSLG and that means I will no longer G be able to make them available to students or faculty.  Unless maybe HP K decides (real soon) to come out with a no cost, no restrictions educational I program that more or less mirrors the Hobbyist Program. I am nbot holding  my breath,  :-(    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  + Date: Mon, 13 May 2002 17:24:22 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> - Subject: Re: Bob Palmer and the demise of DEC ; Message-ID: <01KHOQTXOAFY95MSV4@sysdev.deutsche-boerse.com>   G > And for those who are concerned about the future of VMS and how it is J > affected by loss of educational exposure, I am pretty sure my VAXen haveI > been dropped from the University's CSLG and that means I will no longer I > be able to make them available to students or faculty.  Unless maybe HP M > decides (real soon) to come out with a no cost, no restrictions educational K > program that more or less mirrors the Hobbyist Program. I am nbot holding  > my breath,  :-(   E Actually, they have.  In fact, one of the main criticisms of the new  I educational programme is that it is modelled TOO CLOSELY on the hobbyist  I programme.  Depending on what you want to do, this might or might not be   a disadvantage.    ------------------------------   Date: 13 May 2002 15:34:01 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) - Subject: Re: Bob Palmer and the demise of DEC , Message-ID: <abomd9$18es$1@info.cs.uofs.edu>  ; In article <01KHOQTXOAFY95MSV4@sysdev.deutsche-boerse.com>, <  Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:J |> > And for those who are concerned about the future of VMS and how it isM |> > affected by loss of educational exposure, I am pretty sure my VAXen have L |> > been dropped from the University's CSLG and that means I will no longerL |> > be able to make them available to students or faculty.  Unless maybe HPP |> > decides (real soon) to come out with a no cost, no restrictions educationalN |> > program that more or less mirrors the Hobbyist Program. I am nbot holding |> > my breath,  :-( |>  H |> Actually, they have.  In fact, one of the main criticisms of the new L |> educational programme is that it is modelled TOO CLOSELY on the hobbyist L |> programme.  Depending on what you want to do, this might or might not be  |> a disadvantage.  : Which program is this??  The last one I saw discussed here: basicly made VMS a single user OS and put a whole bunch of; administrative responsibility on the head of professors who ; wanted to use them for class work (and anyone who knows any : professors will know how quickly that would kill an idea!)  9 If there is an education program now that issues the same 9 PAKS that you can get from the Hobbyist program that just 9 says "no commercial use" and has the same cost as the VMS / Hobbyist Program, how does one learn about it??    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Mon, 13 May 2002 09:50:04 -0700 ' From: David Mathog <mathog@caltech.edu> - Subject: Re: Bob Palmer and the demise of DEC + Message-ID: <3CDFEEBC.87A8835D@caltech.edu>    David Froble wrote:   N > With the exception of the hobbyist program, and the CSA program, All licenseK > PAKs that I've seen are perpetual.  Don't know about the education stuff.   6 The CSLG licenses expire yearly.  Not sure about those7 in the shame "new educational program."  Probably those ? do too since they seem to be the same as the Hobbyist licenses.   2 Should HP decide to stop issueing any of the above6 PAKS it would be nothing but a spoonful of dirt on the2 grave of VMS in academia.  DEC/Digital/Compaq have. already mismanaged VMS from being the dominant3 force in the academic market to virtual extinction. 5 Sort of like the dinosaurs but that's not quite right 4 since VMS isn't totally extinct.  VMS is more like a< Coelecanth I guess: where once it was smaller and widespread7 it is now large and very rare.  The debacle in academia 4 was a leading indicator of the similar but not quite3 so total slaughter in other markets.  It was also a = leading indicator of the demise of the companies in question.   4 It was clear to me at Bob Palmer's second reorg that8 he was going to drive Digital into the ground.  Pity the> BOD of Digital was populated by "hear no evil", "see no evil",9 "speak no evil", and "make no money".  Ditto for Compaq.  5 We'll see about HPQ but, well, so far they sure sound 2 like the same overpaid "can't do"  class of golden3 parachuted management we've come to know and abhor.    Regards,   David Mathog mathog@caltech.edu   ------------------------------  % Date: Mon, 13 May 2002 11:29:08 -0400 - From: "Peter Weaver" <peter.weaver@stelco.ca> # Subject: Re: Bug in ANALYZE/AUDIT ? 5 Message-ID: <abom4q$jrqjp$1@ID-141708.news.dfncis.de>   ; "Peter LANGSTOEGER" <peter@langstoeger.at> wrote in message - news:pP3D8.10842$5a4.154788@news.chello.at...  >...J > more than one value (maybe the code does already support it and only the1 > DCL definition is wrong - I haven't checked)...   I I checked, the code does not support it. If I modify the .CLD and allow a G list to be used, only the first value is used, the second value is not; % $! We have records with status of %X1 : $ ANAL/AUD/FUL/sel=(STATUS=CODE=(%X1)) sys$manager/out=x.x* $! We do not have records with status of 2K $ ANAL/AUD/FUL/sel=(STATUS=CODE=(%X2)) sys$manager/out=x.x ! We do not have  %X2 ' %AUDSRV-W-NOSELECT, no records selected & $! If we say "%X1,%X2", we get recordsG $ ANAL/AUD/FUL/sel=(STATUS=CODE=(%X1,%X2)) sys$manager/out=x.x ! We get  records here& $! If we say "$X2,%X1", we get nothingJ $ ANAL/AUD/FUL/sel=(STATUS=CODE=(%X2,%X1)) sys$manager/out=x.x ! We do get records here' %AUDSRV-W-NOSELECT, no records selected   H Better report it through standard methods so either the documentation is# updated, or the image and .CLD are.      -- Peter WeaverL Opinions are my own, and do not reflect the opinions of my employer, nor theK company that it sub-contracts to, nor the company that it sub-contracts to.    ------------------------------    Date: 13 May 2002 06:25:14 -0000= From: Doc.Cypher <Use-Author-Supplied-Address-Header@[127.1]> < Subject: Re: Capellas: Linux, Windows Will 'Eviscerate' Unix5 Message-ID: <20020513062514.4429.qmail@gacracker.org>   9 On 12 May 2002, bob@instantwhip.com (Bob Ceculski) wrote: @ >Doc.Cypher <Use-Author-Address-Header@[127.1]> wrote in message2 >news:<20020512112721.2778.qmail@gacracker.org>... >>  K >> I'd conclude from the way you post here that the best approach with your L >> systems would be social engineering. After all, if you're gullible enoughM >> to buy some of the BS coming from Compaq, and now HP, then I bet you'd buy  >> some BS from a hacker.  >>   >> Doc.  > > >I bought alot of this stuff from people like yourself on this= >board ... unless you are saying that Fred and Rob and Hunter = >and others who post here are idiots who don't know what they  >are talking about?   H No, and I suspect you are the only person on this **NEWSGROUP** who took what I wrote in that way.   ( >One thing I know is true about vms mail> >is hackers have posted on web sites wanting to know if anyone6 >knows a way to run dcl coms or exe's from it, because( >evidently they haven't been able to ...  J Can you point me to some of these websites where I can find such requests?   >from what I know of; >vms mail, you can't ... and now you are calling the people ( >in vms support at DEC/Q/HP idiots also,  H No I'm not. I said you were gullible. Why? Well, to give an example, youJ bought whatever B.S. was in this secret letter even though we had 8 monthsJ without one single mention of VMS by the powers that be at HP. You've also0 mistaken Bill Todd's rather wry humour for fact.   >because I have learned > >alot from them also ... so according to you, I should tear up= >our vms support contract, turn off comp.os.vms because these 2 >people are all idiots ... am I hearing you right?  = No, once again you are failing in your reading comprehension.      Doc. --  6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                             https://vmsbox.cjb.nett   ------------------------------  % Date: Mon, 13 May 2002 11:21:39 +0100dU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> < Subject: Re: Capellas: Linux, Windows Will 'Eviscerate' Unix/ Message-ID: <abo4ht$68$1@new-usenet.uk.sun.com>    Bob Ceculski wrote:n  s > Doc.Cypher <Use-Author-Address-Header@[127.1]> wrote in message news:<20020512112721.2778.qmail@gacracker.org>...E > J >>I'd conclude from the way you post here that the best approach with yourK >>systems would be social engineering. After all, if you're gullible enoughlL >>to buy some of the BS coming from Compaq, and now HP, then I bet you'd buy >>some BS from a hacker. >> >>Doc. >> > ? > I bought alot of this stuff from people like yourself on this)> > board ... unless you are saying that Fred and Rob and Hunter> > and others who post here are idiots who don't know what they= > are talking about?  One thing I know is true about vms mailx? > is hackers have posted on web sites wanting to know if anyone 7 > knows a way to run dcl coms or exe's from it, because = > evidently they haven't been able to ... from what I know ofc< > vms mail, you can't ... and now you are calling the people@ > in vms support at DEC/Q/HP idiots also, because I have learned? > alot from them also ... so according to you, I should tear up > > our vms support contract, turn off comp.os.vms because these3 > people are all idiots ... am I hearing you right?e >      Fred and Rob now lets see.  ; I wouldn't put Rob down on my list of people to get answerso> to technical questions about OpenVMS. Mainly because he mostly! posts about comparative religion.   ; Freddy is probably a good technical source but I would haven> to suspect any answer because he appears to have been strickenA with a the same bug that Rob has caught though he is seen postinge technical responses in between.H  @ I know he wouldn't thank me for this but Hoff would be my choice  he seems refreshingly untainted.    > Its interesting that you didn't include Hoff in your list, do   > you tend to gravitate towards the charismatic religious types.  = He has also said that OpenVMS is no immune to buffer overflown8 attacks perhaps thats why you omitted him from the list.   Regardst Andrew Harrisonu   ------------------------------  % Date: Mon, 13 May 2002 12:30:09 +0100hU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>i< Subject: Re: Capellas: Linux, Windows Will 'Eviscerate' Unix0 Message-ID: <abo8ib$19b$3@new-usenet.uk.sun.com>  4 This would seem slightly at odds with your Execs who4 think that everything will end up running on Windows1 or Linux. Perhaps part of HP and Compaqs intitiald4 investments in Linux could be spent on getting Linus cloned.a   Regardsa Andrew Harrison    Main, Kerry wrote:  % > Re: Linus and Linux kernel mod's ..  > A > Well, if the following articles are correct, while many may notFG > necessarily like the idea of one person at the healm, Linus does have-  > control of the kernel changes- > 8 > http://news.com.com/2100-1001-826093.html Jan 30, 2002+ > "Torvalds, developers at odds over Linux"3 > D > http://www.theregister.co.uk/content/4/24245.html (April 28, 2002)I > "Open source guru Eric Raymond says Linux kernel patching is in crisis,iD > and he has renewed calls for someone to assist Linus Torvalds as a > 'penguin patch lieutenant'.e > F > Speaking at a lecture organised by the UK Unix Users Group in LondonJ > last night, Raymond said that kernel patching was the one remaining part/ > of centralisation in open source development.t > I > He said Linus has "reached his stress limit" and that no one person canhC > deal with the number of kernel patches coming forward from kernele? > maintainers. Patches, many of which would help in the further F > development of Linux, are being dropped without good reason, Raymond > observed." > 	 > Regardsd >  > Kerry Main > Senior Consultanta > Global Services_ > HP Canada Ltd. > Voice: 613-592-4660i > Fax  :  819-772-7036 > Email: Kerry.Main@HP.com >  >  > -----Original Message-----0 > From: Peter da Silva [mailto:peter@abbnm.com]  > Sent: May 7, 2002 10:11 AM > To: Info-VAX@Mvb.Saic.Comt> > Subject: Re: Capellas: Linux, Windows Will 'Eviscerate' Unix >  > I > In article <ab704u$92n$2@aquila.mdx.ac.uk>,  <david20@alpha1.mdx.ac.uk>h > wrote: > H >>Are you sure I thought Redhat , Suse etc added their own installations >> > H >>routines, public domain applications etc but I thought the Kernel was E >>pretty strictly controlled. Linus doesn't want Linux to split into y3 >>umpteem thousands of incompatible Linux variants.1 >> > G > Linus controls the "standard" kernels pretty tightly, but I can't seeiJ > how he can prevent people from shipping modified kernels without keepingG > them from adding their own bugfixes. If he was doing that, then you'diJ > see lots of little "patchkit" fixes being distributed along, the way youD > did with 386BSD when Jolitz was in charge, or the way you get with > qmail. > C > It can't be that Linus and his minions are so fast at integratingaH > changes that nobody ever needs to do this. First, it would be insanelyI > difficult for them to integrate every possible kernel tweak that anyone I > might need into the kernel... second, the behaviour of Linux users whenl4 > a new release is coming up suggests otherwise. :-> >    ------------------------------    Date: 13 May 2002 07:17:02 -0600- From: koehler@encompasserve.org (Bob Koehler) < Subject: Re: Capellas: Linux, Windows Will 'Eviscerate' Unix3 Message-ID: <4Y3fAhA+Hm4y@eisner.encompasserve.org>s  V In article <abgoh6$1lj@web.eng.baileynm.com>, peter@abbnm.com (Peter da Silva) writes: > L > How does "see where the configuration information gets converted to actualM > changes in system state" translate to anything even vaguely resembling "allr6 > configuration files are written by startup scripts"?  8    Oh, so you expect me to disassemble to HPPARC binary?     I've got better things to do.   ------------------------------    Date: 13 May 2002 07:39:24 -0600- From: koehler@encompasserve.org (Bob Koehler)l< Subject: Re: Capellas: Linux, Windows Will 'Eviscerate' Unix3 Message-ID: <iCO+$LRCKLPs@eisner.encompasserve.org>a   In article <abgrtf$1ib$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> writes: >  > MPE has clusters >   <    Yeah, right.  MPE has clusters like Solaris has clusters.  G    Back to reality:  HP has clusters.  Back when DEC still existed, HP lE    said VMS clusters ARE the definitive implementation and that they o'    would like to have as good in HP-UX.o  F    Tru64 clusters, based on VMS technology, is one of the organs beingD    donated to HP-UX.  HP has largely fullfilled their wish by buying    the technology they praised.t  <    And some folks think HP doesn't know what they've bought.   ------------------------------    Date: 13 May 2002 09:04:51 -0700( From: bob@instantwhip.com (Bob Ceculski)< Subject: Re: Capellas: Linux, Windows Will 'Eviscerate' Unix= Message-ID: <d7791aa1.0205130804.47db09a8@posting.google.com>e   Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<abo4ht$68$1@new-usenet.uk.sun.com>... > Bob Ceculski wrote:o > u > > Doc.Cypher <Use-Author-Address-Header@[127.1]> wrote in message news:<20020512112721.2778.qmail@gacracker.org>...r >  > Fred and Rob now lets see. > = > I wouldn't put Rob down on my list of people to get answerse@ > to technical questions about OpenVMS. Mainly because he mostly# > posts about comparative religion.l > = > Freddy is probably a good technical source but I would have @ > to suspect any answer because he appears to have been strickenC > with a the same bug that Rob has caught though he is seen posting ! > technical responses in between.  > B > I know he wouldn't thank me for this but Hoff would be my choice" > he seems refreshingly untainted. >  > @ > Its interesting that you didn't include Hoff in your list, do  > @ > you tend to gravitate towards the charismatic religious types. > ? > He has also said that OpenVMS is no immune to buffer overflow4: > attacks perhaps thats why you omitted him from the list. > 	 > Regards  > Andrew HarrisonP  C only if the authors of the c code rewritten/added to vms didn't useu@ their descriptors properly ... bliss stuff which I assume is the' bulk of the code still will be fine ...b   ------------------------------  # Date: Mon, 13 May 2002 16:45:06 GMTo= From: peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson)  Subject: Compaq auctiona2 Message-ID: <3cdfeaec.16559859@news.cable.ntl.com>  	 Hi folks,a  < Compaq are having an auction tomorow 14/05/02 at dovebid.com      @ http://www.dovebid.com/Auctions/AuctionDetail.asp?auctionID=1260  E Not many ds20e's going mainly pentium stuff. Though you could pick upa$ a pallet of 45 Compaq 18gb scsi hdd.  C interesting other auctions coming up. Auction for excite@home - area they now kaput? hehe ;-)    cheers,     Peter Watkinson % peter.watkinson1@<nospam>ntlworld.com- remove <nospam> to reply :-)   ------------------------------    Date: 13 May 2002 02:27:26 -0700! From: soterro@yahoo.com (Soterro) 6 Subject: Re: cxxlink shared libraries on Alpha and VAX= Message-ID: <d5440555.0205130127.4adcff5a@posting.google.com>s   Hello,  F It would have worked with the solution of Duane, and it didn't becauseE moron as I am I copied the mangled symbol names WITH some extra charsf8 at the end :(( that's why they got longer than 31 chars.C Now it doesn't complain anymore about anything, it just squicks theeE disk already for 1 hour and will still work I guess. Apparently I hitrE that problem of CXXLINK where it re-creates again and again and again ? (and again) one by one the [.cxx_repository]*.obj files for thee> templates my library is using. Well, as long I deleted the oldC repository to have a fresh clean compile, I can do nothing but waittF for them to slooooooooooowly fill with .obj's the directory... then to9 use LINK with an OLB containing those files (am I right?)y  
 Thanks a lot,i Sorin    ------------------------------    Date: 14 May 2002 00:20:44 +0800, From: Paul Repacholi <prep@prep.synonet.com>0 Subject: EV9 (Tarantula) paper. Also EV8 details0 Message-ID: <878z6oj9yb.fsf@k9.prep.synonet.com>   Go tot( http://systems.cs.colorado.edu/ISCA2002/4 and try not to cry too much... The EV9 may have been4 an EV8 extended with Vectory units and instructions., 104 ops/clock peak! 32 DP FLOPS per cycle...  * But, hey, godotium will be so much better.   -- c< 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.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Mon, 13 May 2002 12:57:44 +0200-4 From: "Axel Haringa" <axel.haringa@nl.compuware.com> Subject: freeVMS< Message-ID: <newscache$8gq1wg$4g7$1@news.emea.compuware.com>   a free VMS clone!3  # http://freevms.free.fr/indexGB.html    ------------------------------    Date: 13 May 2002 07:23:07 -0600- From: koehler@encompasserve.org (Bob Koehler)e= Subject: Re: Gartner rides again (was Re: HP Product Roadmap) 3 Message-ID: <HU$+2UkvSo4V@eisner.encompasserve.org>n  h In article <d7791aa1.0205111039.690a118d@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:j > koehler@encompasserve.org (Bob Koehler) wrote in message news:<kY9w4g89UB0Y@eisner.encompasserve.org>...k >> In article <d7791aa1.0205091718.4e93061e@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:i >>  C >> > I would much rather be using vms, and when the mozilla browsere9 >> > becomes stable, that just may be the alternative to a >> > Internet Exploder!  >> a8 >>    So join Encompasserve, get on EISNER, and use VMS. > $ > what or who in the heck is eisner?  4    Now eisner.encompasserve.org, but the history is:    H         <<< EISNER::DRA1:[NOTES$LIBRARY]DECUSERVE_INFORMATION.NOTE;1 >>>6                            -< Decuserve Information >-P ================================================================================P Note 10.1                    The DECUServe node name                      1 of 2P EISNER::KILLEEN "Jeff Killeen DECUServe Chair"       18 lines  10-NOV-1988 22:357                           -< In memory of Dan Eisner >-tP --------------------------------------------------------------------------------K At 1987 Fall DECUS in Anaheim, the DECUServe Executive Committee decided to-J change the node name of this system from "DECUSV" to "EISNER" in memory of Dan Eisner.e  L Dan Eisner was the person most responsible for the existence of this system,K and the Executive Committee wishes this system to be a continuing memory ofa$ his devotion to DECUS and DECUServe.  O The nodename change was made today.  In addition, all occurrences of the stringrM "DECUSV" in all conferences and user notebooks have been changed to "EISNER". L The system logical name DECUSV has been given the value EISNER:: in order toM make the transition transparent in other areas, such as replying to mail sentW prior to the change.  O Please user EISNER:: instead of DECUSV:: for all future accesses to conferencesn and other data on this node.   /John Covert   ------------------------------  % Date: Mon, 13 May 2002 09:45:30 -0400D- From: "Richard D. Piccard" <piccard@ohio.edu>s/ Subject: Re: High Water Concurrent Users Count?(( Message-ID: <3CDFC377.99A702DE@ohio.edu>  H That sounds interesting, but I suspect some systems usage patterns wouldE involve brief residence in DCL during execution of LOGIN.COM and thendF immediate execution of a particular application.  Does DCL still countB as being concurrently shared, from INSTALL's point of view, duringE execution of an application that was initiated from DCL by any of thed< several methods (e.g., RUN, MCR, or foreign command symbol)?  	 						RDPt     Phil Howell wrote: > 6 > "Rick Dyson" <Rick-Dyson@UIowa.EDU> wrote in message% > news:3CDC41B6.6E9B65CD@UIowa.EDU...  > > Rick Dyson wrote:u > > >pN > > > Is there any counter that can be grabbed via lexical or program API that > wouldnG > > > hold the highest number of concurrent users on an OpenVMS node or0 > cluster sincei  > > > the last reboot?  Or ever? > > > L > > > Or do I have to write something to sniff the current value and compare	 > it withc > > > a saved max? > >IJ > > Thanks for all the pointers and solution suggestions.  I have a couple > differentdG > > ways to get the value now and have implemented a simple one to keepo
 > writing themK > > max value to a file.  But I guess there is not already this 'high water  > mark'>' > > value stored somewhere by VMS then?f > >e > > rickK > One way to get this figure is to use the output from $ install list /fulle= > this shows the maximum shared usage for any installed imagee > (since the last reboot) K > the usage of dcl or dcltables should indicate the maximum concurrent usert > processeshM > the usage of other shared images and rtls can give you clues on applicationn > usages > Phil   -- wB ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  # Date: Mon, 13 May 2002 14:14:24 GMT ' From: Rick Dyson <Rick-Dyson@UIowa.EDU>y/ Subject: Re: High Water Concurrent Users Count?r) Message-ID: <3CDFCA40.EC9FD0C4@UIowa.EDU>i  R > > Is there any counter that can be grabbed via lexical or program API that wouldS > > hold the highest number of concurrent users on an OpenVMS node or cluster sinceo > > the last reboot?  Or ever? > >vR > > Or do I have to write something to sniff the current value and compare it with > > a saved max? > R > Thanks for all the pointers and solution suggestions.  I have a couple differentQ > ways to get the value now and have implemented a simple one to keep writing the O > max value to a file.  But I guess there is not already this 'high water mark'-% > value stored somewhere by VMS then?0  M I have noticed the apparent discrepency between "Set Logins /Interactive" vs.@: "Show Users" vs. "Finger".  For instance, right now I get:  4 Set Logins/Inter ==> current interactive value = 177 Sys$GL_IJobCnt   ==> 177L Show Users       ==> Total number of users = 110,  number of processes = 1895 Finger (Penn St) ==> 120 interactive users in cluster' Show System      ==> 422  K This is in a cluster where all but 5 of the processes (from show users) arerN on the primary node.  I know there are loose interpretations of "users" withinP the O/S.  Do you count "humans" or "logins"?  What about sub/detatched processesN from a human.  Is it discreate "terminal" sessions that matter?  Etc.  OpenVMS* "system" processes are not counted, right?  L Any simple explanation for the wide variation of the above results?  In yourP opinion, is the first one the most accurate measure of discrete humans accessing the system?e   Thanks!o Rick   ------------------------------  % Date: Mon, 13 May 2002 11:28:42 -0400d1 From: "Mark D. Jilson" <jilly@clarityconnect.com>w/ Subject: Re: High Water Concurrent Users Count?a2 Message-ID: <3CDFDBAA.EE329B07@clarityconnect.com>  H There are problems in most all of the methods used to determine the # ofF 'users' on a system.  You have to watch out for subprocesses, detachedH processes, network processes and batch jobs.  What does your environmentH consider 'a user' ala the earlier question of if username A logs in fromF their PC w/ 4 sessions, is this 1 user or 4 users?  What if the detachG off a debugger, the calendar, the calculator, a news reader and a notesoF client along with those 4 sessions, is it still 1 user or 4 or 9?  The< Instal trick has the same issue for detached process runningA LOGINOUT.EXE.  Typically system performance data collectors use aHD combination of $GETJPI items to make a determination of 'a user' and= count multiple logins by the same username as multiple users.d   Phil Howell wrote: > 6 > "Rick Dyson" <Rick-Dyson@UIowa.EDU> wrote in message% > news:3CDC41B6.6E9B65CD@UIowa.EDU...d > > Rick Dyson wrote:u > > >uN > > > Is there any counter that can be grabbed via lexical or program API that > wouldrG > > > hold the highest number of concurrent users on an OpenVMS node or1 > cluster sincer  > > > the last reboot?  Or ever? > > >dL > > > Or do I have to write something to sniff the current value and compare	 > it withv > > > a saved max? > >DJ > > Thanks for all the pointers and solution suggestions.  I have a couple > differentsG > > ways to get the value now and have implemented a simple one to keeph
 > writing theaK > > max value to a file.  But I guess there is not already this 'high waterl > mark' ' > > value stored somewhere by VMS then?  > >: > > rickK > One way to get this figure is to use the output from $ install list /fullo= > this shows the maximum shared usage for any installed imagep > (since the last reboot)oK > the usage of dcl or dcltables should indicate the maximum concurrent user0 > processes M > the usage of other shared images and rtls can give you clues on applicatione > usagec > Phil   -- 2C Jilly	- Working from Home in the Chemung River Valley - Waverly, NYu0 	- jilly@clarityconnect.com			- Brett Bodine fan+ 	- Mark.Jilson@HP.com				- since 1975 or sod 	- http://www.jilly.baka.com   ------------------------------    Date: 13 May 2002 06:33:25 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) 4 Subject: Re: howto create self-extracting zip files?3 Message-ID: <9BtVqSadhJEE@eisner.encompasserve.org>   [ In article <3CDF3701.5F8BAB7F@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:n > Larry Kilgallen wrote: >> k^ >> In article <3CDF03AF.6FDC6787@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes: >> > Larry Kilgallen wrote:o >> >>oW >> >> In article <3CDE70E5.9AAC831D@nac.net>, "Wayne W. Scott" <wscott@nac.net> writes:l >> >>e9 >> >> > So I came here for guidance.  Any advice, please?- >> >>6F >> >> Use of self-extracting archives introduces a procedural security >> >> vulnerability. >> >8 >> > I guess that includes the downloadable DCXEXEs, eh? >> M: >> I don't know those, because I do not download software.	 >> [snip]  > 6 > How do you get patches? Do you wait for tapes/CD-Rs?  C The only patches I use come from CDROMs.  It has been 4 years sinced- I needed a patch, and that was for Macintosh.>   ------------------------------    Date: 13 May 2002 06:50:22 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)o4 Subject: Re: howto create self-extracting zip files?3 Message-ID: <iZxsCearMoxa@eisner.encompasserve.org>t  a In article <3cdfa475.305950193@news.process.com>, goathunter@goatley.com (Hunter Goatley) writes:nO > On 12 May 2002 21:49:47 -0600, Kilgallen@SpamCop.net (Larry Kilgallen) wrote:w > D >>For those who do, however, you should decode it with a decoder youC >>already possess, like VMSTAR, UNZIP, BACKUP, etc. rather than oneaC >>provided by the same person who provided you the data (who may ora, >>may not be the individual you hope it is). > G > The nice thing about self-extracting Zip .EXEs is that you can forgetfC > about running it and still use your own local UnZip to unpack it. 0 > You don't have to run it if you don't want to.   That is a good feature.   G With SEVMS one could probably _prevent_ unprivileged users from running- the self-extracting code.y   ------------------------------  # Date: Mon, 13 May 2002 11:34:21 GMTj- From: goathunter@goatley.com (Hunter Goatley)a4 Subject: Re: howto create self-extracting zip files?1 Message-ID: <3cdfa475.305950193@news.process.com>.  M On 12 May 2002 21:49:47 -0600, Kilgallen@SpamCop.net (Larry Kilgallen) wrote:f  C >For those who do, however, you should decode it with a decoder youtB >already possess, like VMSTAR, UNZIP, BACKUP, etc. rather than oneB >provided by the same person who provided you the data (who may or+ >may not be the individual you hope it is).o  E The nice thing about self-extracting Zip .EXEs is that you can forget2A about running it and still use your own local UnZip to unpack it. . You don't have to run it if you don't want to.   Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/c8 goathunter@goatley.com    http://www.goatley.com/hunter/< New Robert R. McCammon site: http://www.RobertRMcCammon.com/   ------------------------------   Date: 13 May 2002 15:18:44 GMT1 From: JONESD@er6.eng.ohio-state.edu (David Jones)i4 Subject: Re: howto create self-extracting zip files?: Message-ID: <abolgk$ss7$1@charm.magnus.acs.ohio-state.edu>  1 In message <3cdfa475.305950193@news.process.com>,X1   goathunter@goatley.com (Hunter Goatley) writes:sN >On 12 May 2002 21:49:47 -0600, Kilgallen@SpamCop.net (Larry Kilgallen) wrote: >hD >>For those who do, however, you should decode it with a decoder youC >>already possess, like VMSTAR, UNZIP, BACKUP, etc. rather than one-C >>provided by the same person who provided you the data (who may orb, >>may not be the individual you hope it is). > F >The nice thing about self-extracting Zip .EXEs is that you can forgetB >about running it and still use your own local UnZip to unpack it./ >You don't have to run it if you don't want to.w  K If the self-extracting archive has .EXEs that you are going to trust anywayyH (e.g. PCSI patch kits), you might as well trust also the extraction codeI embedded in that archive.  If it's only data such as graphics images, you 4 are better off using your own decoder when possible.        < David L. Jones               |      Phone:    (614) 292-6929- Ohio State University        |      Internet: L 140 W. 19th St. Rm. 231a     |               jonesd@er6s1.eng.ohio-state.edu: Columbus, OH 43210           |               vman+@osu.edu  1 Disclaimer: I'm looking for marbles all day long.t   ------------------------------  % Date: Mon, 13 May 2002 10:39:06 -0400 ( From: David Froble <davef@tsoft-inc.com>4 Subject: Re: howto create self-extracting zip files?, Message-ID: <3CDFD00A.6060205@tsoft-inc.com>   Larry Kilgallen wrote:  ] > In article <3CDF03AF.6FDC6787@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:e >  >>Larry Kilgallen wrote: >>T >>>In article <3CDE70E5.9AAC831D@nac.net>, "Wayne W. Scott" <wscott@nac.net> writes: >>>i >>> 5 >>>>So I came here for guidance.  Any advice, please?u >>>>C >>>Use of self-extracting archives introduces a procedural security  >>>vulnerability.  >>>n5 >>I guess that includes the downloadable DCXEXEs, eh?1 >> > 9 > I don't know those, because I do not download software.o > D > For those who do, however, you should decode it with a decoder youC > already possess, like VMSTAR, UNZIP, BACKUP, etc. rather than onenC > provided by the same person who provided you the data (who may orn, > may not be the individual you hope it is). >      An excellant suggestion.  N I would add, use a totally non-priv user account, and a scratch disk if one's M available.  For best results, the user account would not have write privs (a ;> directory) on the system disk.  The more paranoia, the better.  L However, should you believe in the invincibility of VMS security, the tooth N fairy, and such, the use of such a non-prived user account should be adequate 9 protection against any executable you run on your system.d  N Having a 'drop dead', restorable from backup, stand alond VMS system for just N such purposes is the best option.  That's my procedure.  (Now if only I had a L rather large and strong person with a club standing behind me at all times, G insuring that I didn't bypass proclaimed procedures at times ....)  :-)t   Dave   ------------------------------    Date: 13 May 2002 11:17:30 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)r4 Subject: Re: howto create self-extracting zip files?3 Message-ID: <lUXtXBw6mJIh@eisner.encompasserve.org>p  n In article <abolgk$ss7$1@charm.magnus.acs.ohio-state.edu>, JONESD@er6.eng.ohio-state.edu (David Jones) writes:3 > In message <3cdfa475.305950193@news.process.com>,h3 >   goathunter@goatley.com (Hunter Goatley) writes:BO >>On 12 May 2002 21:49:47 -0600, Kilgallen@SpamCop.net (Larry Kilgallen) wrote:- >>E >>>For those who do, however, you should decode it with a decoder you<D >>>already possess, like VMSTAR, UNZIP, BACKUP, etc. rather than oneD >>>provided by the same person who provided you the data (who may or- >>>may not be the individual you hope it is).  >>G >>The nice thing about self-extracting Zip .EXEs is that you can forgetaC >>about running it and still use your own local UnZip to unpack it.,0 >>You don't have to run it if you don't want to. > M > If the self-extracting archive has .EXEs that you are going to trust anyway J > (e.g. PCSI patch kits), you might as well trust also the extraction codeK > embedded in that archive.  If it's only data such as graphics images, you 6 > are better off using your own decoder when possible.  E The moment I unpack the archive is not the moment I want to trust the7E code from this other source.  Using an external extraction utility iscC a way to delay relying on what came in the mysterious bundle.  If IoD found a copy of BACKUP.EXE that was 14 blocks in size, for instance,D I might be suspicious.  There are other tests I might apply as well,C but certainly I am not going to announce them all here so everybodysG can make schemes for that day in the distant future when I might acceptc* software transferred over the Internet :-)   ------------------------------    Date: 13 May 2002 16:22:57 -0000= From: Doc.Cypher <Use-Author-Supplied-Address-Header@[127.1]>m4 Subject: Re: howto create self-extracting zip files?6 Message-ID: <20020513162257.21763.qmail@gacracker.org>  > On 13 May 2002, Kilgallen@SpamCop.net (Larry Kilgallen) wrote:   <snip>  F >The moment I unpack the archive is not the moment I want to trust theF >code from this other source.  Using an external extraction utility isD >a way to delay relying on what came in the mysterious bundle.  If IE >found a copy of BACKUP.EXE that was 14 blocks in size, for instance,tE >I might be suspicious.  There are other tests I might apply as well,ND >but certainly I am not going to announce them all here so everybodyH >can make schemes for that day in the distant future when I might accept+ >software transferred over the Internet :-)T   Larry, I'm a little curious...  K Since you trust media delivered (albeit indirectly) from VMS Engineering on K CD, would you take material that, for example, was digitally signed by Hoff K and transferred across the internet? This would rule out the possibility of- in-transit tampering.i     Doc. -- D6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                             https://vmsbox.cjb.net    ------------------------------    Date: 13 May 2002 05:54:31 -0700( From: bob@instantwhip.com (Bob Ceculski)& Subject: HP to selectively market VMS?= Message-ID: <d7791aa1.0205130454.2df5d5fb@posting.google.com>   0 what kind of hair brain market strategy is that?1 oh well, as long as I have my VMS, and the itanicb) port continues, who in the heck cares ...y     HP spells out big tin changes   # HP/UX, Itanic, AlphaServers, Linux e& By Mike Magee, 13/05/2002 09:38:32 BST  ? THIS IS AN IMPORTANT HP document, and vastly different from theuF storage document it sent to customers, in that it actually talks aboutE what the New Age P is doing about the Itanium, about Linux, about its-D Tandem nonstop platform, OpenVMS and other important products in the HPQ line up.  UpdateB Two corporate users of HPQ kit have written to give their views onD this document  they prefer to remain anonymous for obvious reasons:   1 C A decent article about Tru64, OpenVMS etc. Again The I-N-Q is doingx0 more marketing in this area than H-P-Q or C-P-Q.  C In (2nd hand) news from this morning at Lyon, HP have announced twonD large wins for VMS with a couple clients that want to move away from Unix.   F I see from the FAQ "big tin changes" that VMS will continue to be soldD to "selected new customers" (Question 4) and later it says something( about niche markets such as Health Care.  F Try asking HP how they plan to market the variety of platforms. CompaqD actually told their sales people what to sell the customer depending: on the situation (ie. the niche that the customer was in).  E Compaq's idea was Windows, Unix, VMS and Tandem as we work upwards in F reliability. If you want percentages, we are looking at something likeF Windows 60%, Unix 20%, VMS 8%, Tandem 12%.) The customer seems to have? had little choice because Compaq would rarely provide them with D options. In fact I have heard of a situation where Compaq proposed aC unix solution but IBM proposed a VMS solution and IBM got the sale!a  E Ask Marcello if they are going to continue the niche marketing and if C he says they are, ask why. I suspect it is only to make life easiert for their sales people.e   2uB I recommend you to look back at SGI's announcements of 1H01 - theyE were VERY similar. It is possible that HP's announcement has the samelB purpose - i.e. to pressure Intel into releasing the McKinley chip,? whether or not Intel and Microsoft are ready. See, for example:u  C Q37: There has been public speculation about potential slips in the=@ Intel roadmap - are you still confident in your Alpha-Itanium TM migration strategy?iF A: Yes. We are continuing to work closely with Intel and are confidentB that our long-term strategy of standardizing on the Itanium Family. Processor TM. architecture is the correct one.  A Note that the answer is to a different question, and is worthy ofs Blear at the dispatch box.   ------------------------------    Date: 13 May 2002 07:40:23 -0600- From: koehler@encompasserve.org (Bob Koehler) P Subject: Re: MPE "clusters" (Was: Re: Capellas: Linux, Windows Will 'Eviscerate'3 Message-ID: <sktp5$ZzeJHH@eisner.encompasserve.org>   c In article <3CDC0CC1.6CEDD5F8@aaa.com>, Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> writes:d* > Andrew Harrison SUNUK Consultancy wrote: >> f >> MPE has clustersn >> u2 >> http://www.hp.com/products1/mpeix/ha/index.html >> e >  > 7 > Well, there are clusters and then there are CLUSTERS.   H    Yeah, but MPE has all you need to convinve a eunich he was clustered.   ------------------------------  + Date: Mon, 13 May 2002 07:44:45 -0600 (MDT) " From: John Nebel <nebel@csdco.com>. Subject: Network device for wide area clustersG Message-ID: <Pine.OSF.4.21.0205130734270.30815-100000@athena.csdco.com>l  G Gigaswitches over a DS3 have been working fine for us for years.  SinceeF the DT site was moving I was wondering whether something else might beD better and was considering  Cabletron (Enterasys) Gigaswitch/RoutersC with DS3 cards. They support FDDI as well as 100baseT and 1000baseTa@ and are approximately 10x the backplane speed of the Gigaswitch.  7 DEC did market them as the successor to the Gigaswitch.s   Has anyone had luck with this?  
 John Nebel   ------------------------------  / Date: Mon, 13 May 2002 16:08:58 +0200 (MET DST)u& From: Rudolf Wingert <win@fom.fgan.de>2 Subject: Re: Network device for wide area clusters6 Message-ID: <200205131408.QAA01705@sinet1.fom.fgan.de>   Hello,  C we do use the SmartSwitchRouter from Cabletron within our Alpha/VAX-E Cluster, without any problem. But we do not have any FDDI or DS3 cardlC in it. This product was a commen developed by Digital/Cabletron andUE was bouhgt by Compaq as GigaSwitch Ethernet. AFAIK, the did implementrH on both lines the Digital trunk protokoll as one type of hunt protocols.   Regards Rudolf Wingert   ------------------------------  + Date: Mon, 13 May 2002 18:26:19 +0100 (MET)j9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> K Subject: new academic programme (was: RE: Bob Palmer and the demise of DEC)i; Message-ID: <01KHOSYDJQ3495MSV4@sysdev.deutsche-boerse.com>   J > |> Actually, they have.  In fact, one of the main criticisms of the new N > |> educational programme is that it is modelled TOO CLOSELY on the hobbyist N > |> programme.  Depending on what you want to do, this might or might not be  > |> a disadvantage. > < > Which program is this??  The last one I saw discussed here$ > basicly made VMS a single user OS   A Right, that's the problem.  Or does the hobbyist programme allow iE multi-user licenses now?  (I think that at least in the beginning it U didn't.)   > and put a whole bunch of= > administrative responsibility on the head of professors whog= > wanted to use them for class work (and anyone who knows anyr< > professors will know how quickly that would kill an idea!)  % OK, a different administrative issue.d   ------------------------------  % Date: Mon, 13 May 2002 07:56:03 -0400t' From: Glenn Everhart <Everhart@gce.com>f Subject: Re: No new Alpha saless' Message-ID: <3CDFA9D3.6E901266@gce.com>s  F At times I wonder whether the reason for the lack of follow through onM VMS marketing was a quiet phone call from someone at Microsoft warning Compaqi7 that they would be displeased with a VMS resurgence...?MJ   There have been admissions in court that this happened with Linux. WhileO VMS would appear to loom less large, because of its similarity to NT internallye) it could be viewed as a greater threat...   K It would be best if VMS were sold by a separate subsidiary, if that kind ofsL thing was going on, so that a threat by Microsoft would accomplish basicallyT zero. ("You'll raise the cost for Windows for our boxes? But you don't understand...I we only sell VMS for our boxes, and buy zero units of Windows per year.").  P This is of course complete speculation, and fear that there might be such a callL could conceivably result in the same effects without having had to leave any traces..  S Marketing VMS unfortunately takes some courage, in ways that should not in an ideala world be required..s       "David J. Dachtera" wrote: >  > JF Mezei wrote:v > >  > > Dave Gudewicz wrote:O > > > VMS? Today on Alpha and for the next few years and probably longer.  Long7P > > > term on IPF.  That's what I got out of the message.  And yes, I did noticeF > > > that VMS was mentioned under the UNIX section of the whitepaper. > >cN > > Yes, VMS will remain available for existing customers. However the messageN > > from HP is that Alpha based stuff will not be pitched to new customers who > > will be steered to HP-UX.r > H > O.k. This question has been kicking around in the back of my head long > enough, so I'm gonna ask it: > E > What obstacles would be faced by a reseller (current or future) whoyI > aggressively pushes VMS in spite of its unjustifiable pricing, hardwares: > support limitations and dearth of business applications? > D > No this is *NOT* a troll - I'm dead serious! If HPQ won't "run the > ball", why not a VAR/OEM?. > G > The thinking being: once the powers-that-be see what one savvy seller C > can do to push VMS in spite of all the strikes it has against it,c- > they've GOTTA understand the possibilities.g > E > Some folks won't see it till they believe it. Others, like DoubtingaD > Thomas, won't believe it until they see it. I figure this group isJ > fairly evenly divided on that score, maybe leaning toward the negative a< > bit. The replies, if any, should be interesting, at least. > O > > When the financials are worse than expected, Carly will announce additionaleL > > cutbacks. What is the first thing to go ? Non strategic products with no9 > > growth and revenus that go down. VMS is next in line.n > I > Sadly, this is what the business schools teach. Use the book instead of ' > your head. Rather explains a lot, eh?p >  > -- > David J. Dachterau > dba DJE Systems  > http://www.djesys.com/ > * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/o   ------------------------------    Date: 13 May 2002 23:49:30 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: No new Alpha salesn0 Message-ID: <87helcjbed.fsf@k9.prep.synonet.com>  3 "David J. Dachtera" <djesys.nospam@fsi.net> writes:r   > JF Mezei wrote:t   > > Dave Gudewicz wrote:A > > > VMS? Today on Alpha and for the next few years and probablyi> > > > longer.  Long term on IPF.  That's what I got out of theF > > > message.  And yes, I did notice that VMS was mentioned under the% > > > UNIX section of the whitepaper.r  F > > Yes, VMS will remain available for existing customers. However theD > > message from HP is that Alpha based stuff will not be pitched to/ > > new customers who will be steered to HP-UX.c  C > O.k. This question has been kicking around in the back of my heade# > long enough, so I'm gonna ask it:t  E > What obstacles would be faced by a reseller (current or future) who1@ > aggressively pushes VMS in spite of its unjustifiable pricing,C > hardware support limitations and dearth of business applications?$  D > No this is *NOT* a troll - I'm dead serious! If HPQ won't "run the > ball", why not a VAR/OEM?u   OK, but note the end bit...y  F You get the new prospect lined up, and are all set to go. If they have7 half a clue, they will be sanity checking what you say.o  A "Hello, hp, we want a machine to run 300 users at out health careI centre." ... > "VMS? Do they still make that. And Alphas are dead, they where canceled last year."  A So you are pushing it up hill in a BIG way... I will bet that anybD conversation with hp will include how you will have to port from theA dead end alpha, and how compatable HPPA and IA-64 are. Promise...   B That item referenced from TheInquirer is some what better than theF original.  So maybe hp are re-setting their product lines post merger.B I know if I was dependant on VMS income as an ISV, it would be 'SoA long, and thanks for SFA' if the all new customers to PHUX stays.f  @ Anyone got a PHUX system  they can post to VMS and see what the  reaction is? ;)h   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.s@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------   Date: 13 May 2002 06:33:35 GMT5 From: ccburgess@uqstu.jdstory.uq.edu.au (Ian Burgess)e. Subject: Re: Non-interactive TECO? (of course). Message-ID: <abnmnv$c8d$1@bunyip.cc.uq.edu.au>  \ In article <110520021550154228%elliott@yrl.co.uk>, Elliott Roper <elliott@yrl.co.uk> writes:A >Yesterday I offered a teco macro for fixing crlfs on a wild cardt/ >filspec from a .com. It had a bugette on exit.c2 >As promised, the improved macro is (fix_crlf.tec)$ >hxahken^eqa$<:en$;:eb^eq*$;ec$>ex$$ >uC >called as before from a command file (fix_crlf.com) which containsi >$mung = "$teco32 mung"D >$mung fix_crlf.tec,'p1' >' >bywD >@fix_crlf.com *.mem (or whatever it was that RNO used to screw up.) >eF >I forgot that en$ delivers a null filespec before failing on the nextC >pass. Adding a conditional to the eb and another break" ;" when itr >fails does the trick. >tD >I've been biting my tongue during the EVE v EDT debate about global >replace routines. So far... >lC >By simply inserting <fnstr1$str2;> just before the ec in the above'G >macro you get the whole wild carded global replace from a command filer@ >that consumed so many bytes of internet over the last few days. >e >...damn! I couldn't resist! > G >Teco takes a while to learn. After that, you don't type nearly as manye< >characters of commands to get the job done. Using the same G >"mung teco_macro_file,text" trick, it should be fairly easy to pass inaF >str1 and str2 from the command line. Not that you would bother. After@ >becoming a teco user, you would rather write teco than DCL. ;-)   Make that <fnstr1$str2$;>   ? If you want to avoid having to type <esc> into your procedures,  addtK  $ define/user TEC$INIT "96ee"  ! set escape character to be ` (96 decimal)- before the mung command.  I Of course your login.com would have defined TEC$INIT to be something likejO  $ define/nolog tec$init "96ee-1es2eh16ed256u0et&512""nq0#128#16u0-1ev'q0#4096"r for screen editing!   . Don't you just love this "clear as mud" stuff!     Ian Burgess  University of Queensland I.Burgess[at]its.uq.edu.au www.its.uq.edu.au5   ------------------------------  + Date: Mon, 13 May 2002 11:26:42 -0600 (MDT)>C From: "Ingemar Olson, Sperling (604)444-7367" <IOLSON@dairyland.ca>)* Subject: Non-interactive TECO? (of course)- Message-ID: <01KHOEFGHD5G8XANNL@dairyland.ca>r  ? I must be unusually dense this month, but I still don't get it.p   Eg:  test.com contains $ dir odsr45*.prth	 $ set ver  $ teco = "$teco32 teco"  $! teco = "$edit/teco"# $ define/user sys$command sys$input-$ $ teco ODSR4510_FINV_16_19_56_58.PRT EX   ------------------------------  + Date: Mon, 13 May 2002 11:30:18 -0600 (MDT)o* From: Ingemar Olson <IOLSON@dairyland.com>* Subject: Non-interactive TECO? (of course)- Message-ID: <01KHOEWLQ98K8XANNL@dairyland.ca>C   Thanks Larry, Elliott, Csaba.   > (Looks like that last I one I tried to send "took off" on me.)  ) Nonetheless I still can't get it to work.    If a have a file containing:	 $ set verp $ teco = "$teco32 teco"u# $ define/user sys$command sys$input $ $ teco ODSR4510_FINV_16_19_56_58.PRTC EX<esc><esc>    and those 2 <esc>'s really ARE the escape characterr $i $ set noverm   Then I get i $ @teste $ teco = "$teco32 teco" # $ define/user sys$command sys$input,$ $ teco ODSR4510_FINV_16_19_56_58.PRT' %SYSTEM-F-BADPARAM, bad parameter valuewB %DCL-W-SKPDAT, image data (records not beginning with "$") ignored $ set novera  G Same thing if I don't define the teco symbol and just $edit/teco <file>e  A Sorry, Elliott, the link to the doc you posted earlier is broken.   @ Can any of you guys actually do this (for sure) or are you just  shooting from the hip?   TIA     ...  Ingemar   ------------------------------  + Date: Mon, 13 May 2002 19:53:51 +0100 (MET)r9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>K. Subject: Re: Non-interactive TECO? (of course); Message-ID: <01KHOW2F4IOM95MSV4@sysdev.deutsche-boerse.com>r  % > $ define/user sys$command sys$inpute  $ $  DEFINE/USER SYS$INPUT SYS$COMMAND   might be better.  :-|    ------------------------------  % Date: Mon, 13 May 2002 07:32:59 -0400 * From: "Stewart, Bill" <wjs-corp@Kaman.com>% Subject: NULL device and print queuesa: Message-ID: <1E4B06029E11D211B47C0000F8207F4DB31D84@ESKC2>  % 	Sorry if this has been asked before.   # 	I am runing OPENVMS Alpha v 7.2-1.i  E 	Many years ago I think we had a way of pointing a print queue to thelJ null device NL:.  For the life of me I can't remember how, or if we really$ got it working.  Any help out there?   Bill Stewart   ------------------------------    Date: 13 May 2002 08:00:23 -0600 From: briggs@encompasserve.org) Subject: Re: NULL device and print queuesi3 Message-ID: <G4Lzk886qm20@eisner.encompasserve.org>e  g In article <1E4B06029E11D211B47C0000F8207F4DB31D84@ESKC2>, "Stewart, Bill" <wjs-corp@Kaman.com> writes:- > ' > 	Sorry if this has been asked before.5 > % > 	I am runing OPENVMS Alpha v 7.2-1.b > G > 	Many years ago I think we had a way of pointing a print queue to the5L > null device NL:.  For the life of me I can't remember how, or if we really& > got it working.  Any help out there?  F My recollection is that the standard print symbiont (PRTSMB.EXE) won't? work with the null device.  I don't remember if LATSYM.EXE can.x  G I know that one approach that I have taken is to use the print symbionte= modification routines to create a symbiont that does nothing.   , The code is reasonably short, so here it is.   	John Briggs    &         OPTIONS         /EXTEND_SOURCE  %         PROGRAM         NULL_SYMBIONTa           IMPLICIT        NONE  $         INCLUDE         'PSMDEF.INC'  .         EXTERNAL        PSM$REPLACE, PSM$PRINT.         INTEGER         PSM$REPLACE, PSM$PRINT           INTEGER         STATUS  +         EXTERNAL        NULL_OUTPUT_ROUTINEn  B         STATUS = PSM$REPLACE ( PSM$K_OUTPUT, NULL_OUTPUT_ROUTINE ):         IF ( .NOT. STATUS ) CALL SYS$EXIT ( %VAL(STATUS) )           STATUS = PSM$PRINT():         IF ( .NOT. STATUS ) CALL SYS$EXIT ( %VAL(STATUS) )           ENDe&         OPTIONS         /EXTEND_SOURCE  L         INTEGER FUNCTION NULL_OUTPUT_ROUTINE ( REQUEST_ID, WORK_AREA, FUNC,  	1	FUNCDESC, FUNCARG )           IMPLICIT        NONE  F         INTEGER         REQUEST_ID, WORK_AREA, FUNC, FUNCDESC, FUNCARG  $         INCLUDE         'PSMDEF.INC'"         INCLUDE         '($SSDEF)'  0         EXTERNAL        PSM$_EOF, PSM$_FUNNOTSUP  (         IF ( FUNC .EQ. PSM$K_OPEN ) THEN#             NULL_OUTPUT_ROUTINE = 1..         ELSE IF ( FUNC .EQ. PSM$K_CLOSE ) THEN#             NULL_OUTPUT_ROUTINE = 1S-         ELSE IF ( FUNC .EQ. PSM$K_WRITE  .OR.a7         1         FUNC .EQ. PSM$K_WRITE_NOFORMAT ) THENw#             NULL_OUTPUT_ROUTINE = 1f         ELSE6             NULL_OUTPUT_ROUTINE = %LOC(PSM$_FUNNOTSUP)         END IF           RETURN           ENDl   ------------------------------  # Date: Mon, 13 May 2002 13:43:45 GMTy From: system@SendSpamHere.ORGo) Subject: Re: NULL device and print queuese0 Message-ID: <00A0DDEB.18C68E44@SendSpamHere.ORG>  g In article <1E4B06029E11D211B47C0000F8207F4DB31D84@ESKC2>, "Stewart, Bill" <wjs-corp@Kaman.com> writes:p > & >	Sorry if this has been asked before. > $ >	I am runing OPENVMS Alpha v 7.2-1. >MF >	Many years ago I think we had a way of pointing a print queue to theK >null device NL:.  For the life of me I can't remember how, or if we reallyn% >got it working.  Any help out there?t >e
 >Bill Stewarte  I http://www.tmesis.com/downloads/NULLSMB.ZIP  contains a "null" symbiont'seI source.  It's very simple and easy to build and install.  I've sourced ituJ is 3 different lingos too -- Macro, Bliss and C -- so you have your choice. of languages to read and better understand it.   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMp            B5   "Well my son, life is like a beanstalk, isn't it?" e   ------------------------------  % Date: Mon, 13 May 2002 15:20:03 +0100c4 From: "Chris Sharman" <chris.sharman@ccagroup.co.uk>H Subject: Re: Old RAID 230 logical init very slow (63% done after 5 days)B Message-ID: <1021299626.12045.0.nnrp-07.9e989e7e@news.demon.co.uk>  3 Thanks to Chuck & everyone else for all the advice. J I'd actually got SRLMGR on the SWXCRMGR floppy, but not mentioned anywhere in the documentation.y It's V2.36 we're using.iK I've downloaded V2.70 from the website, but I can't verify anywhere whether0L it's the correct firmware for my piece of hardware (which I think might be aK kzpsc, but can't find out for certain). As it warns that the wrong firmware5H will make the kit unusable, I'm rather loathe to risk breaking it, sinceK 2.36 probably works well enough if I remember to use SRLMGR (& I've writtens it all over the manual now).H Unless anyone knows how to confirm what's in the machine/whether it will accept the firmware ?    Thanks all,a
 Chris Sharmane   ------------------------------    Date: 13 May 2002 07:52:22 -0600- From: koehler@encompasserve.org (Bob Koehler)i Subject: Re: Powered by HP3 Message-ID: <7J$Sb$5ZU2FI@eisner.encompasserve.org>-  a In article <3cdd0ad2$0$1416$1dc6e903@news.corecomm.net>, "Dave Gudewicz" <k9jdk@arrl.net> writes:>N > Or maybe ridding the world of the Open in OpenVMS.  Just another thought.... >   E    What, so they can change it to COE-VMS?  I'll just wait for COE tom.    hit mainstream and keep the O-P-E-N silent.   ------------------------------    Date: 13 May 2002 07:44:40 -0600- From: koehler@encompasserve.org (Bob Koehler)t Subject: Re: Powered by HP3 Message-ID: <bRtwlQhV8zIx@eisner.encompasserve.org>   n In article <YGYC8.36976$GLp1.3995@news01.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes: > I > Hell, maybe we can get them to port any useful features from HP-UX intoi > TRU64 on Alpha. :-)v >   G    What makes you think there are any?  I did HP-UX for a few years anddA    saw no advantages over other UNIX.  Did see some shortcomings.h   ------------------------------    Date: 13 May 2002 07:51:13 -0600- From: koehler@encompasserve.org (Bob Koehler)m Subject: Re: Powered by HP3 Message-ID: <onL+bdP0EEMt@eisner.encompasserve.org>g  o In article <ZV%C8.129090$q8.13390211@bin3.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes:  > > > "Dave Gudewicz" <david.gudewicz@abbott.com> wrote in message- > news:abhda6$koq$1@fizban.pprd.abbott.com...pK >> Some of the Rebel alliance lurk here and know this, the alliance can andi > hastM >> claimed victory despite overwhelming odds.  Let us not give in to the darkt >> side.  Never. > L > Let the Dark Side dangle some hope for VMS before the Rebels, and the manyM > with weak minds will be suddenly convinced that they can deal with it aftertD > all.  By the time they realize otherwise, all will have been lost. >   B    You fellows are wandering around the wrong universe.  Much as ID    appreciate next week's installment of that little fable, I'm muchF    more impressed by Reginald Barkley's DECnet Phase XVII hack for theC    PHA0: device.  I understand he expects to submit it for the 2403C    Freeware distribution.d   ------------------------------  + Date: Mon, 13 May 2002 15:33:16 +0100 (MET)t9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>s+ Subject: Re: simple disk-shadowing questionr; Message-ID: <01KHOMY86YIQ99DNNX@sysdev.deutsche-boerse.com>d  M > > IIRC, it is because the name of the disk in the Storage Control Block is o# > > limited to a particular length.  > D > The answer has already been posted (see quote below). Actually, itE > doesn't quite add up. The allocation class can be from 0 to 255. So E > $255$12char_label is actually 17 characters, which is one more thaneF > allowed according to the quoted message. So something is wrong here,< > unless the dollar signs are not included in the lock name.  D If the number is stored as a byte, then 0--255 would be allowed and  things would add up.   ------------------------------  % Date: Mon, 13 May 2002 11:30:42 +0100wU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>y4 Subject: Re: Some more words of wisdom from Capellas/ Message-ID: <abo52s$cb$1@new-usenet.uk.sun.com>    Steve Reece wrote:   > Interesting Andrew.t > F > Although it's a few months since I talked to a DBA, last I heard wasJ > that Oracle was developed on Sun OS, the precursor to Solaris.  As such,H > the Solaris version itself is a port of the source code.  Earlier than. > the VMS one, yes.  But nevertheless, a port. > 
 > The reason?k > D > Apparently it's so that the lowest common denominator is easier to
 > achieve. >     / No its developed on Solaris. Of course your DBAs6 could be confusing SunOS 4.x.x with Solaris because if0 you run uname -a on a Solaris box you might get.  > SunOS sunburst 5.8 Generic_111919-03 sun4u sparc SUNW,Ultra-60  % This is otherwise known as Solaris 8.R  1 Of course the other reason why I know that Oraclet/ uses Solaris is the the boxes Oracle use do notV+ run SunOS 4.x.x. SunOS 4.x.x only nominally 1 supports SMP and does not have things like kernelg3 threads all of which would be a major dissadvantage  to the Oracle developers.h   Regards  Andrew Harrisont     > Steve. >  >  > * > Andrew Harrison SUNUK Consultancy wrote: >  >>Alan Greig wrote:  >> >>4 >>>http://www.theregister.co.uk/content/7/25142.html >>>a$ >>>Some quotes from Curly yesterday: >>>oO >>>HP in the past has been seen as a kind of amorphous blob, asked one analyst.s. >>>What can you do to clarify that perception? >>>rM >>>"We're driving the move to standards-based computing across the board," he  >>>replied.s >>>sN >>>In response to a question "What does Compaq stand for?", Capellas answered: >>> P >>>"We've got to be the first go-to-market partner Intel, the first go-to-marketQ >>>partner for Oracle and the first go-to-market partner for Microsoft," he said.- >>>- >>>-1 >>Mikey needs to get a bit closer to his partnersa >>before spouting this.y >>5 >>Oracle is devloped on Sun Solaris and given Compaqs85 >>market share with them no one could describe Compaqa0 >>as the first-go-to-market partner with Oracle. >>6 >>Being first to market with Intel and MS is a dubious5 >>accolade which most people would consider best keptr  >>quiet rather than broadcasted. >>	 >>Regardse >>Andrew Harrisonn >> >    ------------------------------    Date: 13 May 2002 08:13:55 -0600- From: koehler@encompasserve.org (Bob Koehler)l4 Subject: Re: Some more words of wisdom from Capellas3 Message-ID: <yBL7q5rjOeti@eisner.encompasserve.org>    In article <abo52s$cb$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> writes:,  1 > No its developed on Solaris. Of course your DBAa8 > could be confusing SunOS 4.x.x with Solaris because if2 > you run uname -a on a Solaris box you might get. > @ > SunOS sunburst 5.8 Generic_111919-03 sun4u sparc SUNW,Ultra-60 > ' > This is otherwise known as Solaris 8.u  H    Which of course, is why it claims to be SunOS 5.8 when running uname.   ------------------------------  % Date: Mon, 13 May 2002 15:43:23 +0100tU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>n4 Subject: Re: Some more words of wisdom from Capellas0 Message-ID: <abojsl$5eq$1@new-usenet.uk.sun.com>   Bob Koehler wrote:   > In article <abo52s$cb$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> writes:l >  > 1 >>No its developed on Solaris. Of course your DBAH8 >>could be confusing SunOS 4.x.x with Solaris because if2 >>you run uname -a on a Solaris box you might get. >>@ >>SunOS sunburst 5.8 Generic_111919-03 sun4u sparc SUNW,Ultra-60 >>' >>This is otherwise known as Solaris 8.i >> > J >    Which of course, is why it claims to be SunOS 5.8 when running uname.      7 Solaris refers to the bundle of SW which makes up an OSh7 shells, windowing, utilities etc. And then there is the  kernel.   # The Solaris kernel is called SunOS.0  ! Don't shoot me it wasn't my idea.a   Regards  Andrew Harrison    ------------------------------  % Date: Mon, 13 May 2002 12:26:07 +0100hU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>-: Subject: Re: SOT: Yo Andrew, you guys on a roll this week?0 Message-ID: <abo8ap$19b$1@new-usenet.uk.sun.com>   Bob Ceculski wrote:N   > Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<abeb3a$9s1$1@new-usenet.uk.sun.com>...r >  > D > So tell us Andrew, do you actually do anything for Sun, or are you- > their internet rendition of James Carville?d >   0 I post articles to this newsgroup from a desktop, system running the OS I advocate people use.   :):):):)   Regardss Andrew Harrisonm   ------------------------------    Date: 13 May 2002 10:50:38 -0700( From: bob@instantwhip.com (Bob Ceculski): Subject: Re: SOT: Yo Andrew, you guys on a roll this week?= Message-ID: <d7791aa1.0205130950.36accd5c@posting.google.com>g   Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<abo8ap$19b$1@new-usenet.uk.sun.com>...g > Bob Ceculski wrote:b >  > > Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<abeb3a$9s1$1@new-usenet.uk.sun.com>...d > >  > > F > > So tell us Andrew, do you actually do anything for Sun, or are you/ > > their internet rendition of James Carville?s > >  > 2 > I post articles to this newsgroup from a desktop. > system running the OS I advocate people use. > 
 > :):):):) > 	 > Regards  > Andrew Harrison_   so you run OpenVMS?o   ------------------------------  % Date: Mon, 13 May 2002 10:48:49 +0100hU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>.# Subject: Re: Stallards smoking gun! 0 Message-ID: <abo2kb$st6$1@new-usenet.uk.sun.com>   david20@alpha2.mdx.ac.uk wrote:r  ^ > In article <3CDC2FD3.633AB951@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > @ >>Just read the famous PDF containing the STallard damning text. >>N >>What I found interesting is the wording used to specify that VMS developmentG >>will continue ON ALPHASERVERS until 2006. And very separately, almost = >>deemphasized, it is stated that the port to IA64 continues.. >>T >>If HP had long term intentions with VMS, they would have worded something such as:K >>"We plan to continue to provide new versions on ALPHA servers until 2006,n8 >>after which new versions will appear only on Itanium". >>V >>It sesms to me that HP doesn't really have any intentions to productize VMS on IA64. >> > O > Of course another way of reading that might be that HP isn't quite so sure ofcK > IA64 anymore. Maybe none of their systems will appear only on Itanium. :)e >   > Have they explicitly stated :- > M > "We plan to continue to provide new versions on PA-RISC servers until 200x,rA > after which new versions of HP-UX will appear only on Itanium".T >     : If HP announce PA-9000 then we will all know that IA64 has< failed. At the moment HP are continuing to tweek the PA-8XXX# in the hope that IA64 will succeed.   ? Its worth noting that McKinley is about as unatractive to HP-PAaB customers as it is to Alpha customers since it will involve taking& a performance cut and paying to do it.   RegardsT Andrew Harrison    ------------------------------    Date: 13 May 2002 07:33:01 -0600- From: koehler@encompasserve.org (Bob Koehler)e# Subject: Re: Stallards smoking gun!n3 Message-ID: <ESWtlYJaMIWl@eisner.encompasserve.org>-  v In article <20020511125518.26826.qmail@gacracker.org>, Doc.Cypher <Use-Author-Supplied-Address-Header@[127.1]> writes: > J > Problem is, commitments have been made in the past and then broken. ManyM > people in the newsgroup (and doubtless many existing VMS customers too) are 3 > looking for actions to back up these commitments.g  F    Will posting the actual first boot date of "iVMS" (when it happens)G    be sufficient action?  How about of they bring an iVMS system to theB3    next national user meeting after the first boot?t   ------------------------------    Date: 13 May 2002 14:46:17 -0000= From: Doc.Cypher <Use-Author-Supplied-Address-Header@[127.1]> # Subject: Re: Stallards smoking gun! 6 Message-ID: <20020513144617.19452.qmail@gacracker.org>  > On 13 May 2002, koehler@encompasserve.org (Bob Koehler) wrote:B >In article <20020511125518.26826.qmail@gacracker.org>, Doc.Cypher5 ><Use-Author-Supplied-Address-Header@[127.1]> writes:  >> rK >> Problem is, commitments have been made in the past and then broken. Many N >> people in the newsgroup (and doubtless many existing VMS customers too) are5 >> looking for actions to back up these commitments. f > G >   Will posting the actual first boot date of "iVMS" (when it happens)-H >   be sufficient action?  How about of they bring an iVMS system to the4 >   next national user meeting after the first boot?  K Personally I was thinking more of effort to raise the public profile of the6I operating system. Continuing the efforts on the port isn't enough for me.mI Don't get me wrong, I'm glad it's happening and have every faith that VMSgD Engineering can deliver. However, I want to see noticable efforts toK promote awareness of the continued existence of VMS. I also want to see VMSnD allowed the opportunity to compete for new business instead of beingJ reserved for customers who know of its existence and are persistent enough to get offered a VMS solution.     Doc. -- i6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                             https://vmsbox.cjb.netb   ------------------------------  % Date: Mon, 13 May 2002 10:22:40 -0700 ' From: David Mathog <mathog@caltech.edu> # Subject: Re: Stallards smoking gun! + Message-ID: <3CDFF660.C2F595F8@caltech.edu>    John Reagan wrote:  J > If Compaq/HP didn't plan to productize OpenVMS on Itanium, they sure areI > paying me money for nothing then (as well as paying lots of other folksS > for nothing).N  : That's not necessarily true.  Assume that HPQ management's5 goal is merely to keep the current customers in theirO; existing service contracts and maybe sell a few new alphas.=8 If they announced a real EOL those customers would begin9 to bail immediately which would cut into this substantialT7 revenue stream quickly.  By putting some money into theE6 port (and for all we know Intel is paying them most of3 this, so it may be costing HPQ next to nothing) HPQ 5 makes it look like they are actually doing something, 2 when they may in fact just biding their time until, an already predetermined event or kill date.  = No, I'm not paranoid - we've seen this before.  It is exactly ; the same strategy Compaq employed with Alpha.  CPQ told its 8 customers one thing while management had already decided> that something else was going to happen, and were only waiting: for (apparently) a bad quarter where axing the Alpha could> be manipulated to make the books look better for that quarter.A Since HPQ seems to be following most of the CPQ roadmaps, and hasN: most of the same players in place, we have every reason to: expect exactly the same sort of crappy behavior from them.  A The flip side of the coin is that the same engineers are in place.> so at least we can also expect (in the mean time) a reasonably high level of support.     Regards,   David Mathog mathog@caltech.edu   ------------------------------  % Date: Mon, 13 May 2002 15:48:15 +0100 U From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>l% Subject: Re: Sun to use AMD chips ...'0 Message-ID: <abok5q$5jo$1@new-usenet.uk.sun.com>   Bob Ceculski wrote:    > Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<abeb3a$9s1$1@new-usenet.uk.sun.com>...n > D > So tell us Andrew, do you actually do anything for Sun, or are you- > their internet rendition of James Carville?d >    Humm  : No I have a day job and I also ammuse myself by responding to your posts.  5 Remember its your subject line "Sun to use AMD chips"i  7 If you don't want me to respond then perhaps you should-2 have chosen a different subject line and different content.   Regardsn Andrew Harrisonc   ------------------------------  % Date: Mon, 13 May 2002 16:43:12 -0000a- From: wspencer@ap.nospam.org (Warren Spencer)n' Subject: Re: Switching the console modes7 Message-ID: <920D8946Ewarrenspencer1977@209.249.90.100>l  6 martin@radiogaga.harz.de (Martin Vorlaender) wrote in 0 <3cdc8c5a.524144494f47414741@radiogaga.harz.de>:  C >Craig A. Berry (craig.berry@nospam.SignalTreeSolutions.com) wrote:h. >> bart@CTI01.COMVERSE.COM (Steve Bart) wrote:J >> >      Can someone tell me how to switch the console mode from one modeK >> > to the other when you do not have access to the console in the current H >> > mode? In other words, if I have a system (in this case a DS10) withI >> > the console set to graphics, but for whatever reason I cannot access C >> > the system via that connection, how can I change it to serial?i >...; >> try unplugging the keyboard and rebooting; some systems -F >> automatically default to serial if there is no keyboard plugged in : >> (though I don't know for sure the DS10 is one of them). >PD >I seem to remember that more recent Alphas switch to serial mode ifC >they sense no graphics device attached in the self-tests. So, alsorE >try to unplug the monitor. Like Craig, I'm not sure whether the DS10  >is one of them. >. >cu,	 >  Martin,  I On an Alpha 1000 OpenVMS 6.2-1H3 about 5 years ago, a techie showed me a rI trick where you logged into another terminal, issued a few DCL commands,  G and *poof*, the console left graphics mode and went into dump-terminal tI mode.  Sorry I don't recall the details, but I wanted to mention that it e6 can be done.  And yes, *poof* is a technical term <g>.   ws -- l   Warren Spencer' Senior Software Engineer (not a writer)  The Associated Press  < ** Time flies like an arrow.  Fruit flies like a bananna. **   ------------------------------  # Date: Mon, 13 May 2002 16:56:47 GMT 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> ' Subject: Re: Switching the console mode-9 Message-ID: <jfSD8.26$G_1.791620@cacnews.cac.cpqcorp.net>r  H Getting out of graphics mode, into VGA mode is easy.  Switching from theI graphics console to the serial port can't be done on-the-fly.  At minimum0E you need to force a cold start (that is, you need the equivalent of ao% console >>> INIT command to execute).z  F You should be able to find an old example C program (maybe it's on theK freeware CD) that allows you to change a console variable directly.  You'lltH need to change the CONSOLE variable.  Then you need to force the reboot.   Warren Spencer wrote in message . <920D8946Ewarrenspencer1977@209.249.90.100>...6 >martin@radiogaga.harz.de (Martin Vorlaender) wrote in1 ><3cdc8c5a.524144494f47414741@radiogaga.harz.de>:c >dD >>Craig A. Berry (craig.berry@nospam.SignalTreeSolutions.com) wrote:/ >>> bart@CTI01.COMVERSE.COM (Steve Bart) wrote:eK >>> >      Can someone tell me how to switch the console mode from one modemL >>> > to the other when you do not have access to the console in the currentI >>> > mode? In other words, if I have a system (in this case a DS10) withsJ >>> > the console set to graphics, but for whatever reason I cannot accessD >>> > the system via that connection, how can I change it to serial? >>...g; >>> try unplugging the keyboard and rebooting; some systemsiF >>> automatically default to serial if there is no keyboard plugged in; >>> (though I don't know for sure the DS10 is one of them).i >>E >>I seem to remember that more recent Alphas switch to serial mode if D >>they sense no graphics device attached in the self-tests. So, alsoF >>try to unplug the monitor. Like Craig, I'm not sure whether the DS10 >>is one of them.  >> >>cu, 
 >>  Martin >wI >On an Alpha 1000 OpenVMS 6.2-1H3 about 5 years ago, a techie showed me ahI >trick where you logged into another terminal, issued a few DCL commands,nG >and *poof*, the console left graphics mode and went into dump-terminalnI >mode.  Sorry I don't recall the details, but I wanted to mention that it 7 >can be done.  And yes, *poof* is a technical term <g>.s >a >wsa >--_ >p >Warren Spencere( >Senior Software Engineer (not a writer) >The Associated Presse >s= >** Time flies like an arrow.  Fruit flies like a bananna. **c   ------------------------------  % Date: Mon, 13 May 2002 15:36:45 +0100I3 From: "Nick Glassock" <nicholas.glassock@amsjv.com> ( Subject: Terminal Services on VMS client& Message-ID: <3cdfcefd$1@pull.gecm.com>  J I am trying to find a solution to using Windows 2000 applications from VMS 6.1 / 6.2 clients.  L I am aware of the Citrix Metaframe 1.8 Windows 2000 product, but there is no ICA client for VMS.v  H I am aware of NCD's WinCenter "Unix Integration Services" X11 add on forL Citrix, but this runs against Metaframe 1.1 and has since been discontinued.  I I am aware that there is a Citrix Metaframe - Java ICA client, and a Java0K RDP client from www.hob.de but cannot find a browser / JVM that will run onI these versions of VMS.  - Any help or suggestions would be appreciated.  Nick   ------------------------------    Date: 13 May 2002 08:34:06 -0700% From: Alan Greig <a.greig@virgin.net>w, Subject: Re: Terminal Services on VMS client) Message-ID: <abomde01avc@drn.newsguy.com>   4 In article <3cdfcefd$1@pull.gecm.com>, "Nick says... >cK >I am trying to find a solution to using Windows 2000 applications from VMSo >6.1 / 6.2 clients.> > M >I am aware of the Citrix Metaframe 1.8 Windows 2000 product, but there is noa >ICA client for VMS.  0 The Citrix ICA Java client works under VMS/Java.  I >I am aware of NCD's WinCenter "Unix Integration Services" X11 add on fortM >Citrix, but this runs against Metaframe 1.1 and has since been discontinued.h  L I am not the local terminal services expert by any means but this comes as aO surprise to me. We have a very large number of Unix workstations which use X to L display Windows and I have never heard comments at department meetings about0 this going away. Has it maybe just changed name?  J >I am aware that there is a Citrix Metaframe - Java ICA client, and a JavaL >RDP client from www.hob.de but cannot find a browser / JVM that will run on >these versions of VMS.l  P Ah so you have heard of the Java client. First question. Are the 6.1 systems VAXM or Alpha? If VAX you are stuck. If Alpha then I *think* you should be able toa5 get the Java client running under Netscape 3 for VMS.-  . >Any help or suggestions would be appreciated. >Nickh >@ >o   ------------------------------  # Date: Mon, 13 May 2002 13:48:01 GMT 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>e Subject: Re: USB on OpenVMSc8 Message-ID: <luPD8.14$FH1.46126@cacnews.cac.cpqcorp.net>  G That doesn't help.  Point me to a spec for the PS2 command set and datal format for the mouse.r    8 Dirk Munk wrote in message <3CDCC03A.9060904@home.nl>...H >It seems this device is no longer sold by Logitech. However the present >trackball with wheel support:; >http://www.logitech.com/cf/products/productoverview.cfm/89lC >is sold with a USB -> PS2 adapter, so that it can be used with PS2 F >connections. The third button by the way, is activated by pushing the% >wheel. It doubles as a click-button.o >U >  >Fred Kleinsorge wrote:m >> Got the spec on it? >>; >> Dirk Munk wrote in message <3CDAE4FC.6070409@home.nl>...e >>K >>>I am using a Logitech TrackMan Marble+ PS2 mouse on my PC. 3 buttons andcC >>>a mousewheel. Works fine ! I'm using a similar Trackball withouta >>>mousewheel on my PWS. >>>o >>>Fred Kleinsorge wrote:  >>> K >>>>No.  I added it only for USB.  It was easy, it's just another axis on an >>>>pointing device. >>>>D >>>>I've never seen the spec for a PS2 thumbwheel mouse, or if it is possible >>>n >> to1 >>K >>>>detect it's presence (as opposed to any other type of mouse).  The PS-2> >>>>mouse design was abysmal.w >>>> >>>> >>>>< >>>>Dirk Munk wrote in message <3CDA47D9.8010403@home.nl>... >>>> >>>> >>>>>Fred Kleinsorge wrote:  >>>>>n >>>>>i0 >>>>>>Eh?  What's a two-cylinder VSXXX-AA mouse? >>>>>>L >>>>>>The mouse will be a standard 3-button USB mouse.  As an added bonus, I >>>>>r >>>>put  >>>> >>>>F >>>>>>support for the thumbwheel in for anyone that wants to plug in a >>>>>i >>>>thumbwheel >>>> >>>> >>>>>>mouse. >>>>>oJ >>>>>Does this mean I can connect a thumbweel mouse to the ps-2 port of my2 >>>>>PWS500 too ? From which version VMS / Motif ?I >>>>>That would be very nice, I could scroll through the messages in thisp >>>>>newsgroup.e >>>>>u >>>>>  >>>>>n >>>> >> >> >e >$   ------------------------------  # Date: Mon, 13 May 2002 13:48:57 GMTu5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>r Subject: Re: USB on OpenVMS 9 Message-ID: <dvPD8.15$1U1.476209@cacnews.cac.cpqcorp.net>t  6 Forrest is on vacation, he'll be back in a week or so.    8 Dirk Munk wrote in message <3CDCC2C0.9040100@home.nl>...H >So far I found one product with the Lucent chipset, and that is the ADS >USB Turbo Quad 4. >http://www.adstech.comrI >Nice thing with this chipset is that there are 4 independant USB busses,m9 >so it will get your four USB controllers instead of one.i >_H >Will DCPS be suported on USB printers as well? that would be very nice,H >because we could buy a cheap Postscipt printer and connect it. EthernetE >connections for these printers are usualy relatively very expensive.: >e >Forrest Kenney wrote:G >>     The history of USB and OpenVMS where it has been and where it isi. >> going.  There has been an ongoing effort toK >> get USB into OpenVMS.  The original plan was to support it with the DS10a* >> ES40 platform.  As part of this work weH >> built and for a limited time supplied a USB kit for a V7.2 code base.3 >> This code was also demonstrated at the San DecusaK >> in 1999.  At that time the plan was to ship full support for USB as part . >> of V7.3.  But plans changed for a number of( >> reasons and the work was put on hold. >>I >>     When it was obvious that some of the new  platforms would not haver any ) >> of the legacy ISA I/O devices work was,L >> restarted and the code was dug out of mothballs.  We recently finished up+ >> all the work that we know of to make USB C >> viable on these new platforms.  Officially only the embedded USBv3 >> controllers on these platforms are supported and J >> only for keyboards and mice.  All of the supported code is checked into/ >> the version that is presently in field test.t >>K >>     In addition support for keyboards, mice, we are shipping drivers fort/ >> printers and cables that conform the the USB6I >> printer standard.  We also have support for modems that conform to thes0 >> communications class standard.  These are notK >> going to be listed as supported because we did not have the resources toi4 >> test and qualify devices for this relase.  PleaseJ >> feel free to use them and report problems but don't be surprised if you0 >> get a these are not supported answer from the >> support organizations.o >>1 >>     What is not supported at the present time:o >>< >>               1) UHCI controllers, or USB 2.0 controllersJ >>               2) Any random OHCI based controllers.  We have tested and' >> had good luck with ones based on thepH >>                   Lucent USS344 chip and Symbios 6080.  We have tried some) >> other with varying degrees of success. , >>               3) Customer written drivers >> >> Forrest Kenneyt >> OpenVMS USB project leaderr >> >> >- >-   ------------------------------  % Date: Mon, 13 May 2002 12:35:36 +0100@U From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>W9 Subject: Re: What is good model for disk i/o w/shadowing?10 Message-ID: <abo8si$1ic$1@new-usenet.uk.sun.com>   Jan-Erik Sderholm wrote:e  # > Regarding disk I/O and shadowing.o >  >tI > On short, the VMS shadowing is very effective. Of course everything canuC > be measured, but in practical use, you'd be hard to see any major C > differense. Probably the better *read* performance will outweightf) > the slightly worse *write* performance.  >     : Typically 80% of disk activity is reads and 20% writes. So% on average you may well gain overall.e  ? At the end of the day disks are now not expensive on a cost pery? GB and generally volume shadowing pays for itself in time saveds- no having to recover data when a drive fails.i   Regardse Andrew Harrison1   ------------------------------  % Date: Mon, 13 May 2002 14:55:35 +0100.( From: Nic Clews <sendspamhere@127.0.0.1>9 Subject: Re: What is good model for disk i/o w/shadowing?u) Message-ID: <3CDFC5D7.5AC07FD1@127.0.0.1>s  ( Andrew Harrison SUNUK Consultancy wrote: > < > Typically 80% of disk activity is reads and 20% writes. So' > on average you may well gain overall.t  F This could be average typical but I have seen situations where systemsC are 80% write and 20% read. This is a totally different performanceeH model and the differences between RAID-5 and RAID-1 become very apparent? in this situation. Cache size and data rate have a bearing too.s  lA > At the end of the day disks are now not expensive on a cost per6A > GB and generally volume shadowing pays for itself in time saved / > no having to recover data when a drive fails.n  H Fully agreed, and picking the correct data protection mechanism may also= be a factor. While host based volume shadowing does seem verydF attractive, do consider what controllers can offer you. HBVS gives you@ system controlled volume splitting, controller gives potentiallyF scripted similar operations (more room for error) and RAID5 solutions,@ transparent failure but no room for data segregation for backup.  = Then there are the people that volume shadow RAID5 volumes...   H There a lot of valid points in this thread overall, you need to considerF it relevant to your situation. FWIW we set up a system with 3 shadowedB volumes (using 6 2 GB SCSI disks) on a single SCSI bus with no ill effects and no data lost.a -- t( Regards, Nic Clews CSC Computer Sciences nclews at csc dot com,   ------------------------------  / Date: Mon, 13 May 2002 16:15:34 +0200 (MET DST)o& From: Rudolf Wingert <win@fom.fgan.de>9 Subject: Re: What is good model for disk i/o w/shadowing?y6 Message-ID: <200205131415.QAA01740@sinet1.fom.fgan.de>   Hello,  G I think, that it is very important, that you use hardware RAIDs regard-sE less you use RAID1, RAID10 or RAID5. Then the write cache will not beaE disabled, as it is done by software shadowing or software RAID. I did(F messure the performance differences: 3.6MB/s for writes without cache,C 22.5MB/s for writes with disks onboard cache enabled (IBM 36GB diskd@ with 7200Upm). With the write cache enable, the read performanceC slowes down from 25MB/s to 22.5MB/s. If you disable the read cache,o you will see also only 3.6MB/s.i   Regards Rudolf Wingert   ------------------------------    Date: 13 May 2002 10:51:41 -0600 From: briggs@encompasserve.org9 Subject: Re: What is good model for disk i/o w/shadowing?w3 Message-ID: <f8vNjeXgEs0l@eisner.encompasserve.org>e  _ In article <200205131415.QAA01740@sinet1.fom.fgan.de>, Rudolf Wingert <win@fom.fgan.de> writes:  > Hello, > I > I think, that it is very important, that you use hardware RAIDs regard-eG > less you use RAID1, RAID10 or RAID5. Then the write cache will not be.G > disabled, as it is done by software shadowing or software RAID. I didoH > messure the performance differences: 3.6MB/s for writes without cache,E > 22.5MB/s for writes with disks onboard cache enabled (IBM 36GB disk-B > with 7200Upm). With the write cache enable, the read performanceE > slowes down from 25MB/s to 22.5MB/s. If you disable the read cache,i! > you will see also only 3.6MB/s.s  F I'm trying to rationalize those figures with an operational model that can explain them.   E The best I can figure is that you're benchmarking with an applicationuC that doesn't fill a pipeline with writes and instead waits for eachn. write to complete before queuing the next one.  = On the read side, one would guess that between read-ahead anda: cache-for-possible-reuse, you're managing an 83% hit rate.  9 Anyone want to present some other plausible explanations?u   	John Briggs   ------------------------------  % Date: Mon, 13 May 2002 16:09:26 +0100lU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>r9 Subject: Re: What is good model for disk i/o w/shadowing?a0 Message-ID: <aboldh$60k$1@new-usenet.uk.sun.com>   Nic Clews wrote:  * > Andrew Harrison SUNUK Consultancy wrote: > < >>Typically 80% of disk activity is reads and 20% writes. So' >>on average you may well gain overall.e >> > H > This could be average typical but I have seen situations where systemsE > are 80% write and 20% read. This is a totally different performanceMJ > model and the differences between RAID-5 and RAID-1 become very apparentA > in this situation. Cache size and data rate have a bearing too.  >  e    9 Well volume shadowing is a host based facility on OpenVMSa> host based RAID 5 would be a bad idea except for very limitted? circumstances and would be terrible in almost all circumstancesi7 with a failed drive in the RAID 5 set or under rebuild.   : But you are right you do see systems that are 20% read and= 80% write, what I was uggesting was an average, the origionalg= poster should be able to determine what % of reads and writesm their system has..       > A >>At the end of the day disks are now not expensive on a cost perhA >>GB and generally volume shadowing pays for itself in time savede/ >>no having to recover data when a drive fails.1 >> > J > Fully agreed, and picking the correct data protection mechanism may also? > be a factor. While host based volume shadowing does seem very H > attractive, do consider what controllers can offer you. HBVS gives youB > system controlled volume splitting, controller gives potentiallyH > scripted similar operations (more room for error) and RAID5 solutions,B > transparent failure but no room for data segregation for backup. > ? > Then there are the people that volume shadow RAID5 volumes...T >     < Controller based RAID 5 is a huge win over host based RAID 5; in terms of performance in normal operations and also under 8 drive failure. The drive failure/rebuild factor is often5 ignored, if a drive fails on a RAID 1 system then theS7 host based SW stops trying to queue I/O's to it and the 5 net effect to performance is low. If a drive fails inm: a RAID 5 set then the host based SW or controller based SW7 has to reconstruct the data from the remaining data andt parity.n  8 To rebuild the RAID 1 drive is simply a block copy, RAID: 5 involves reading all the drives and then re-creating the missing blocks.e   RegardsC Andrew Harrisonu   ------------------------------  % Date: Mon, 13 May 2002 11:33:09 +0100E( From: Nic Clews <sendspamhere@127.0.0.1>' Subject: Re: Who cares about marketing! ) Message-ID: <3CDF9665.15C4EA38@127.0.0.1>.   Bob Ceculski wrote:a > > > The more I think about this, why do I care so much about theC > mis-marketing of vms?  if hp like Q wants to keep quiet about thee  ; Bob, everything is about marketing and selling, everything.)  F When you talk to the poor user that has been without their systems forF several days, the way you keep their confidence in your services is by@ marketing. Tell them you're the best, let them know how much youD understand their situation and their pain, that you are representingH them to the hilt all the way with who ever's job it is to fix the thing.  E How the heck do you think microsoft get away with it? easy, "the nextnF thing" is going to be bigger and better, that's the line the poor userF keeps getting, and they've heard it that often they believe it. That's: the important thing (and there's that word again BELIEVE).  D Everything is selling. Even when I go to buy something, I _sell_ theD fact I am the right buyer (and therefore hopefully some privileges / discounts etc.)h  ? Olson was the idiot who didn't believe in marketing, he thought F 'technology sells'. Well it did for a little while, but now look where& that lack of attention has got things.  F The concept and idea of VMS has to be sold first, make it cool to haveG VMS, make it sexy, make it desirable, make it so you can't live withouti it, make it so you NEED it.h  G Most, if not all of this has zip to do with what VMS is good or bad at. @ All cars get you from A to B no matter who makes them. So do allH aircraft. It needs distinguishing with a concept, then sell the concept.- Why do you buy one car over another car, Bob?-  H I drive a Citroen (in English, LEMON!). My next car will of my choice beC a Citroen. Folks here can try to tell me that Audi /Mercs/ BMW Ford F whatever are better, but I won't hear you. I'm sold on the idea of theH Citroen. That is the point. Even though my car has a silly name, I still feel good about driving it.1  A Now, there is an additional bit to this. I may consider a RenaultmE (Laguna), it is also French, but that is a co-incidence. It certainlyiE looks pretty cool (later models) the only thing putting me off is theeG stupid woman driving a (Renault) Clio talking about 'size'. Do I reallyaG want to associate myself with a make who markets some of their vehiclesiH with that mentality? Also when they say "it goes", to me that encourages speeding, and again...  0 No, I'll stick with my lemon that you very much.  G Remember, everyone is saying that their operating system is more secureeE and has more uptime that the next, and is THE operating system to useaH for ebusiness. Simply advertising that VMS does just the same as claimed> by all the other operating systems isn't really going to help.    Something different is needed...  E So why do I drive Lemons? Well, there's no way I'd want to be a Merc*nD driver, they are all the same, indicators never work on them and theF lack of intelligence they are driven with, suggests their car choosingA capacity must be equally challenged. Audi** drivers are arrogant,cA Renaults I mentioned, and Ford, good grief, talk about mutton anda lamb...i  H *Volvo drivers often exhibit similar characteristics. Mind you, if I wasG selling something to a Merc or Volvo driver, if they have so much moneypF to throw around, I'll add a little to the price, they'll pay if I sell	 it right.w ** Hi Elliott! :-)  G When buying things, I _sell_ the fact I'm the best buyer, so much so, I H usually get a reduction *and* beat other buyers who'd probably have paid full whack.   G After that little look inside my head, here endeth your first lesson inA
 marketing.C (BTW, No offence to anyone driving vehicles of makes mentioned, I'mhE merely making a point. I know for a fact my BX makes people seasick!)t -- w( Regards, Nic Clews CSC Computer Sciences nclews at csc dot com-   ------------------------------  % Date: Mon, 13 May 2002 12:59:10 +0100nU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>@' Subject: Re: Who cares about marketing!h0 Message-ID: <aboa8p$20b$1@new-usenet.uk.sun.com>   Bob Ceculski wrote:e  > > The more I think about this, why do I care so much about theC > mis-marketing of vms?  if hp like Q wants to keep quiet about theeG > best os on the planet, then let them.  Why should I care.  As long asdD > I have my support, and modest product development, and the itaniumG > port, and alpha support thru 2011, why should I care?  If every otheriD > fool wants to run their enterprise on windoze/unix/linux, let themE > crash and burn.  I will survive the coming virus storm.  I will runkG > 99.9999 w/the best clustering, security, shadowing and reliability iniH > the business ... who cares about marketing?  Those who know the secretG > of vms will be very happy.  If HP is happy w/modest vms profits and a 6 > losing wintel business, then so be it ... who cares? >   - You do although you clearly don't realise it.e  7 You run software on your OpenVMS boxes, without SW theye+ are about as usefull as a chocolate teapot.i  7 At some time in the future the SW suppliers will review(8 their choice of HW/OS platforms or decide what platforms to deliver their new SW on.>  9 Marketing is one critical peice that helps them decide in3 your favour (or can).>  : Marketing isn't just adverts and other publicity materials9 its joint market development programmes with SW partners, > sponsorship, sales incentives, loaner machines and SW licenses> internal training etc all of which ultimately come out of some marketing pot.    People on this group claim that.  1 Compaq sales people had no clue what OpenVMS was.h- Compaq partners had no clue what OpenVMS was.p ISV's did not support OpenVMS.  5 All of these things can result from lack of marketingr
 resources.  6 Don't just think of marketing as adverts if you do you will miss the point entirely.    Regards  Andrew Harrison-   ------------------------------  % Date: Mon, 13 May 2002 08:55:55 -0500i1 From: "Dave Gudewicz" <david.gudewicz@abbott.com>:' Subject: Re: Who cares about marketing!U1 Message-ID: <abogq2$11b$1@fizban.pprd.abbott.com>    I reme   -- Dave...n  ) Adam and Eve had many advantages, but thee- principle one was that they escaped teething.  -----Mark Twaint  5 "Nic Clews" <sendspamhere@127.0.0.1> wrote in message.# news:3CDF9665.15C4EA38@127.0.0.1...n > Bob Ceculski wrote:t > >h@ > > The more I think about this, why do I care so much about theE > > mis-marketing of vms?  if hp like Q wants to keep quiet about thes > = > Bob, everything is about marketing and selling, everything.o >cH > When you talk to the poor user that has been without their systems forH > several days, the way you keep their confidence in your services is byB > marketing. Tell them you're the best, let them know how much youF > understand their situation and their pain, that you are representingJ > them to the hilt all the way with who ever's job it is to fix the thing. > G > How the heck do you think microsoft get away with it? easy, "the next.H > thing" is going to be bigger and better, that's the line the poor userH > keeps getting, and they've heard it that often they believe it. That's< > the important thing (and there's that word again BELIEVE). >wF > Everything is selling. Even when I go to buy something, I _sell_ theF > fact I am the right buyer (and therefore hopefully some privileges / > discounts etc.)d >dA > Olson was the idiot who didn't believe in marketing, he thoughtlH > 'technology sells'. Well it did for a little while, but now look where( > that lack of attention has got things. >lH > The concept and idea of VMS has to be sold first, make it cool to haveI > VMS, make it sexy, make it desirable, make it so you can't live withoutt > it, make it so you NEED it.I >dI > Most, if not all of this has zip to do with what VMS is good or bad at.rB > All cars get you from A to B no matter who makes them. So do allJ > aircraft. It needs distinguishing with a concept, then sell the concept./ > Why do you buy one car over another car, Bob?p >tJ > I drive a Citroen (in English, LEMON!). My next car will of my choice beE > a Citroen. Folks here can try to tell me that Audi /Mercs/ BMW FordtH > whatever are better, but I won't hear you. I'm sold on the idea of theJ > Citroen. That is the point. Even though my car has a silly name, I still > feel good about driving it.  > C > Now, there is an additional bit to this. I may consider a RenaultdG > (Laguna), it is also French, but that is a co-incidence. It certainlysG > looks pretty cool (later models) the only thing putting me off is the I > stupid woman driving a (Renault) Clio talking about 'size'. Do I really I > want to associate myself with a make who markets some of their vehiclespJ > with that mentality? Also when they say "it goes", to me that encourages > speeding, and again... >@2 > No, I'll stick with my lemon that you very much. >mI > Remember, everyone is saying that their operating system is more secureiG > and has more uptime that the next, and is THE operating system to use-J > for ebusiness. Simply advertising that VMS does just the same as claimed@ > by all the other operating systems isn't really going to help. >u" > Something different is needed... >yG > So why do I drive Lemons? Well, there's no way I'd want to be a Merc*pF > driver, they are all the same, indicators never work on them and theH > lack of intelligence they are driven with, suggests their car choosingC > capacity must be equally challenged. Audi** drivers are arrogant,oC > Renaults I mentioned, and Ford, good grief, talk about mutton ands	 > lamb...  >tJ > *Volvo drivers often exhibit similar characteristics. Mind you, if I wasI > selling something to a Merc or Volvo driver, if they have so much money:H > to throw around, I'll add a little to the price, they'll pay if I sell > it right.  > ** Hi Elliott! :-) >uI > When buying things, I _sell_ the fact I'm the best buyer, so much so, IlJ > usually get a reduction *and* beat other buyers who'd probably have paid
 > full whack.e >uI > After that little look inside my head, here endeth your first lesson ini > marketing.E > (BTW, No offence to anyone driving vehicles of makes mentioned, I'mrG > merely making a point. I know for a fact my BX makes people seasick!)p > --* > Regards, Nic Clews CSC Computer Sciences > nclews at csc dot coml   ------------------------------   Date: 13 May 2002 13:59:28 GMT1 From: "Dave Gudewicz" <david.gudewicz@abbott.com>l' Subject: Re: Who cares about marketing!d1 Message-ID: <abogs0$11f$1@fizban.pprd.abbott.com>n   ------------------------------   Date: 13 May 2002 14:44:54 GMT1 From: "Dave Gudewicz" <david.gudewicz@abbott.com>c' Subject: Re: Who cares about marketing!f1 Message-ID: <abojh6$1h1$1@fizban.pprd.abbott.com>l   ------------------------------   Date: 13 May 2002 07:17:51 GMT$ From: JKB <local@kepler.makalis.com>& Subject: Re: [announce] FreeVMS 0.0.144 Message-ID: <slrnaduq4v.aj.local@Kepler.makalis.com>  - Le Mon, 13 May 2002 at 00:25 GMT,  propos de  Re: [announce] FreeVMS 0.0.14,*  Rick Campbell crivait dans comp.os.vms :K > I have a copy of Wendin's PCVMS that includes the source code as well.  IgE > say's it has RMS calls.  The manual made some references to the OSTtM > (Operating System Toolkit, which I don't have and don't know if it's needed, > to recompile the OS).- > 5 > I can provide this to anyone that is interested....o  B 	Maybe you can propose this code on the FreeVMS mailing list ? Its 	address is freevms@ml.free.fr.e  	 	Regards,0   	JKB   ------------------------------   Date: 13 May 2002 13:57:24 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)e& Subject: Re: [announce] FreeVMS 0.0.14, Message-ID: <abogo4$15l0$1@info.cs.uofs.edu>  4 In article <slrnaduq4v.aj.local@Kepler.makalis.com>,'  JKB <local@kepler.makalis.com> writes:B0 |> Le Mon, 13 May 2002 at 00:25 GMT,  propos de! |> Re: [announce] FreeVMS 0.0.14,B- |>  Rick Campbell crivait dans comp.os.vms :hN |> > I have a copy of Wendin's PCVMS that includes the source code as well.  IH |> > say's it has RMS calls.  The manual made some references to the OSTP |> > (Operating System Toolkit, which I don't have and don't know if it's needed |> > to recompile the OS). |> > :8 |> > I can provide this to anyone that is interested.... |>  E |> 	Maybe you can propose this code on the FreeVMS mailing list ? ItsA" |> 	address is freevms@ml.free.fr.  > I would advise caution.  A quick (not extensive) search of the= INTERNET showed where a number of people tried to contact thes? author regarding the status of PCVMS.  There is nothing to show.? that any one was successful.  If PCVMS is still a validly copy-D@ righted program and any of the developers of FreeVMS look at the@ code of PCVMS it will polute the development and likely make the@ resultant product a derivative work under current copyright law.   bill   -- rJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   a   ------------------------------   Date: 13 May 2002 07:20:26 GMT$ From: JKB <local@kepler.makalis.com>9 Subject: Re: [announce] FreeVMS 0.0.15 was FreeVMS 0.0.14n4 Message-ID: <slrnaduq9q.aj.local@Kepler.makalis.com>  A 	New release avalaible at http://freevms.free.fr/indexGB.html ;-)   	 	Regards,o   	JKB   ------------------------------   End of INFO-VAX 2002.264 ************************Date: Mon, 13 May 2002 16:09:26 +0100lU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>r9 Subject: Re: What is good model for disk i/o w/shadowing?a0 Message-ID: <aboldh$60k$1@new-usenet.uk.sun.com>   Nic Clews wrote:  * > Andrew Harrison SUNUK Consultancy wrote: > < >>Typically 80% of disk activity is reads and 20% writes. So' >>on average you may w