0 INFO-VAX	Tue, 11 Feb 2003	Volume 2003 : Issue 84      Contents:5 Re: Any way to change shadow set name on system disk? 5 Re: Any way to change shadow set name on system disk? 5 Re: Any way to change shadow set name on system disk? 5 Re: Any way to change shadow set name on system disk?  Barcode Printing from VT510  Re: Barcode Printing from VT510  Re: Barcode Printing from VT510 P convert the decimal value of TERM_CHAR to a decimal ASCII text string in C with 7 Re: DECWindows Motif on a DS10L without a graphics card 7 Re: DECWindows Motif on a DS10L without a graphics card  DFO 2.7 and scripts G Re: Doc Format Gripes (was: Re: HP hideousness of the OPenVMS web pages G Re: Doc Format Gripes (was: Re: HP hideousness of the OPenVMS web pages G Re: Doc Format Gripes (was: Re: HP hideousness of the OPenVMS web pages 
 Help with CDs   Re: Hobbyist PAKs from Montagar?  RE: Hobbyist PAKs from Montagar?; Re: How to check - write cache enabled/disabled of a device + Re: HP hideousness of the OPenVMS web pages E HP's Web Pages (was: Doc Format Gripes (was: Re: HP hideousness ...)) $ Re: IBM says AMD dead in 5 years ...$ Re: IBM says AMD dead in 5 years ...$ Re: IBM says AMD dead in 5 years ...$ Re: IBM says AMD dead in 5 years ...$ Re: IBM says AMD dead in 5 years ...$ Re: IBM says AMD dead in 5 years ...$ Re: IBM says AMD dead in 5 years ...9 Kerberos T2.0 field test kit for hp OpenVMS now available 0 Marvel:  Even more marvelous that first thought! Re: OpenVMS Boot press release RE: OpenVMS Boot press release- Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 - Re: OpenVMS Boots on Itanium on Friday Jan 31 P Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha Retai  Read/Write from a global sectionA Re: RP06's and Field Service calls -- was Very large disks on VMS A Re: RP06's and Field Service calls -- was Very large disks on VMS A Re: RP06's and Field Service calls -- was Very large disks on VMS A Re: RP06's and Field Service calls -- was Very large disks on VMS A RE: RP06's and Field Service calls -- was Very large disks on VMS P Re: Smart Array 5300A, Storageworks 4300 Ultra Tower, OpenVMS 7.3-1, and an ES403 Re: Support for Gigabit Ethernet Clusters on Marvel P Re: Sysgen and UAF memory limit recommendations for Oracle on small machine (256P RE: Sysgen and UAF memory limit recommendations for Oracle on small machine (256 TCPIP: How to multihomed Re: TCPIP: How to multihomed Re: The importance of ISV's  Re: The importance of ISV's  Re: VMS source listings ?   F ----------------------------------------------------------------------    Date: 10 Feb 2003 23:28:44 -0800, From: mcbill20@hotmail.com (Bill McLaughlin)> Subject: Re: Any way to change shadow set name on system disk?< Message-ID: <e9cbc4f2.0302102328.3c45e46@posting.google.com>  ` "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3E4869E9.ED12ACB3@fsi.net>... > Robert Deininger wrote:  > > A > > In article <e9cbc4f2.0302101758.76a6b08a@posting.google.com>, 1 > > mcbill20@hotmail.com (Bill McLaughlin) wrote:  > > K > > >I have a two member shadowed system disk (DSA0) consisting of DKA0 and H > > >DKA100. After adding a second SCSI card (narrow) in a PCI slot, theE > > >built-in wide SCSI now shows up as PKB0 instead of PKA0.  So, of J > > >course the disks now show up as DKB0 and DKB100.  However, the shadowI > > >set still shows up as DSA0.  Is there any way to change this so that # > > >the shadow set would be DKB0:?  > > * > > Shadowed disks are always DSA devices. > C > Actually, one of the benefits of shadowing is that the underlying I > physicla device names can change and applications don't need to know or A > even be aware of that. They just address the shadow-set device, A > preferably by a logical name such DISK$volume_name or whatever.     F Sorry guys...I guess I wasn't clear on this. I understand about shadowE sets. What I meant was that having the shadow set be DSB0 rather than A the original DSA0 would seem more logical since the drives in the + shadow set are now DSBnn rather than DSAnn.    Bill   ------------------------------    Date: 11 Feb 2003 03:07:13 -0800- From: tessier-ashpool@usa.net (Chris Bardell) > Subject: Re: Any way to change shadow set name on system disk?= Message-ID: <9f261edc.0302110307.654aa5cd@posting.google.com>   p mcbill20@hotmail.com (Bill McLaughlin) wrote in message news:<e9cbc4f2.0302102328.3c45e46@posting.google.com>...b > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3E4869E9.ED12ACB3@fsi.net>... > > Robert Deininger wrote:  > > > C > > > In article <e9cbc4f2.0302101758.76a6b08a@posting.google.com>, 3 > > > mcbill20@hotmail.com (Bill McLaughlin) wrote:  > > > M > > > >I have a two member shadowed system disk (DSA0) consisting of DKA0 and J > > > >DKA100. After adding a second SCSI card (narrow) in a PCI slot, theG > > > >built-in wide SCSI now shows up as PKB0 instead of PKA0.  So, of L > > > >course the disks now show up as DKB0 and DKB100.  However, the shadowK > > > >set still shows up as DSA0.  Is there any way to change this so that % > > > >the shadow set would be DKB0:?  > > > , > > > Shadowed disks are always DSA devices. > > E > > Actually, one of the benefits of shadowing is that the underlying K > > physicla device names can change and applications don't need to know or C > > even be aware of that. They just address the shadow-set device, C > > preferably by a logical name such DISK$volume_name or whatever.  >  > H > Sorry guys...I guess I wasn't clear on this. I understand about shadowG > sets. What I meant was that having the shadow set be DSB0 rather than C > the original DSA0 would seem more logical since the drives in the - > shadow set are now DSBnn rather than DSAnn.  >  > Bill  C $DEFINE or $ASSIGN a logical? Experiment with /TRANLAT=CONCEAL etc.   T Forgive my asking this, but does it really matter? $SHO DEV will show the members...   --ChrisB3 (former VMS sysmgr, prior to Death By Beancounter).    ------------------------------    Date: 11 Feb 2003 08:28:36 -0800, From: mcbill20@hotmail.com (Bill McLaughlin)> Subject: Re: Any way to change shadow set name on system disk?= Message-ID: <e9cbc4f2.0302110828.293eb609@posting.google.com>   X Nic Clews <sendspamhere@127.0.0.1> wrote in message news:<3E48CEA7.E838408@127.0.0.1>... > Bill McLaughlin wrote: > > J > > I have a two member shadowed system disk (DSA0) consisting of DKA0 andG > > DKA100. After adding a second SCSI card (narrow) in a PCI slot, the D > > built-in wide SCSI now shows up as PKB0 instead of PKA0.  So, ofI > > course the disks now show up as DKB0 and DKB100.  However, the shadow H > > set still shows up as DSA0.  Is there any way to change this so that" > > the shadow set would be DKB0:? > F > Ah, you learn the value of DISK$volume_label logical names and other
 > logicals...  >   D I _do_ use logical names in all my system startup files. The problemB is, when you do things like "SHOW DEV D", you get shadow set DSA0,F comprised of DKB0 and DKB100. This is a very minor annoyance and won't3 cause any problems. However, do _you_ type the full A "DISK$volume_label" when looking for files, moving things around,  etc.?   C All I wanted was for the shadow name to be in the same format (same B controller ID) so that when I am typing quickly, or late at night,6 it's easier and more intuitive to use the short names.   A > I assume this is an alpha, you do have the use of the PAC (Port I > Allocation Class) where a SCSI port of A to D or whatever is taken to a 0 > $n$xxA.. See the documentation for it's usage. >  Thanks. I'll take a look.   J > However, that's a last resort, I would take the opportunity (for this isG > an opportunity, not a problem) to correct trying to refer to physical F > devices. You could define DKB0 to DSA0, if you've no need for a realI > DKB, but you are asking for a lot of trouble. Other responses correctly $ > pointed out it will always be DSA. > G > You could also try swapping the SCSI board slots. I did look into the F > details of this a long time ago, but basically depending on make andD > model, the slot probing order (and therefore the A, B designation)C > change be used to advantage, but YMMV, and you'll need to do some  > experimentation. >   > Uh...I'm not sure my soldering skills are that refined. Per myE original post, the new PKA controller is a PCI controller but the one 0 that moved to PKB is part of the mainboard.  ;-)  D > But, as I said at the top, try to lose your dependence on physical > device names.   E All my device names used in startup procedures, UAF records, etc. are % logical names. I agree-- good advice.    Thanks.  Bill   ------------------------------  % Date: Tue, 11 Feb 2003 16:54:06 +0000 ( From: Nic Clews <sendspamhere@127.0.0.1>> Subject: Re: Any way to change shadow set name on system disk?) Message-ID: <3E492AAE.32DBBA8A@127.0.0.1>    Bill McLaughlin wrote: > F > I _do_ use logical names in all my system startup files. The problemD > is, when you do things like "SHOW DEV D", you get shadow set DSA0,H > comprised of DKB0 and DKB100. This is a very minor annoyance and won't5 > cause any problems. However, do _you_ type the full C > "DISK$volume_label" when looking for files, moving things around,  > etc.?   ; Usually I define a process level logical, like X or Z ... ! A Unfortunately, because I deal with so many systems, I have almost B forgotten what it is like to have a LOGIN.COM and I pretty much do everything longhand :-(   E > All I wanted was for the shadow name to be in the same format (same D > controller ID) so that when I am typing quickly, or late at night,8 > it's easier and more intuitive to use the short names.  # I know what you mean, points taken.   I > > You could also try swapping the SCSI board slots. I did look into the H > > details of this a long time ago, but basically depending on make andF > > model, the slot probing order (and therefore the A, B designation)E > > change be used to advantage, but YMMV, and you'll need to do some  > > experimentation. > >  > @ > Uh...I'm not sure my soldering skills are that refined. Per myG > original post, the new PKA controller is a PCI controller but the one 2 > that moved to PKB is part of the mainboard.  ;-)  F No, this is the point. As the system probes each hose, including thoseD as part of the mainboard, it does a particular sequence. I fell overD exactly this problem some years ago, and merely choosing a differentB slot to install the card solved the issue. Try moving the board...  F (I believe it is true to say that physically adjacent slots may not be6 logically adjacent, but I stand to be corrected here). --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------    Date: 11 Feb 2003 03:10:53 -0800* From: jlanigan@esatclear.ie (John Lanigan)$ Subject: Barcode Printing from VT510= Message-ID: <bf2b1c04.0302110310.30980d35@posting.google.com>    Hi,   @ I need to print barcodes from a cobol application running on VMS through dec VT510s  D The printer is an OKI ML3320.  this works fine if the printer is setE up as a network printer on the alpha, or if using Reflections on a pc ! with a directly attached printer.   D However with the printer slaved off the VT510 no barcode appears, no" number, no gibberish, nothing.....  
 Any ideas?   John   ------------------------------  % Date: Tue, 11 Feb 2003 12:09:55 +0000 ( From: Nic Clews <sendspamhere@127.0.0.1>( Subject: Re: Barcode Printing from VT510) Message-ID: <3E48E813.B330543D@127.0.0.1>    John Lanigan wrote:   B > I need to print barcodes from a cobol application running on VMS > through dec VT510s > F > The printer is an OKI ML3320.  this works fine if the printer is setG > up as a network printer on the alpha, or if using Reflections on a pc # > with a directly attached printer.  > F > However with the printer slaved off the VT510 no barcode appears, no$ > number, no gibberish, nothing.....  G We used to do barcode printing and data up/download through the printer  port on VT series terminals.  G Under program control, you send codes to enable and disable the printer B port, we did it under Cobol too. Pay attention to the SETUP of theE printer port, you need send data to host etc., and you have to set up H the comms like parity and word length. You may also want to use the dataE signalling. Some printers insist on having some of the modem controls = applied before they will print. You may be using the parallel @ functionality, I've no experience in this area, but it should be	 possible.    HTH  --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------  # Date: Tue, 11 Feb 2003 12:28:28 GMT 9 From: Alan Adams <alan.adams@orchard-way.freeserve.co.uk> ( Subject: Re: Barcode Printing from VT510? Message-ID: <89c415c34b.Alan.Adams@orchard-way.freeserve.co.uk>   < In message <bf2b1c04.0302110310.30980d35@posting.google.com>5           jlanigan@esatclear.ie (John Lanigan) wrote:    > Hi,  > B > I need to print barcodes from a cobol application running on VMS > through dec VT510s > F > The printer is an OKI ML3320.  this works fine if the printer is setG > up as a network printer on the alpha, or if using Reflections on a pc # > with a directly attached printer.  > F > However with the printer slaved off the VT510 no barcode appears, no$ > number, no gibberish, nothing..... >  > Any ideas?  K At a guess, the VT510 is swallowing the escape sequences the printer needs.   ' I can't think of a way to prevent that.    Alan   --  
 Alan Adams& alan.adams@orchard-way.freeserve.co.uk http://www.nckc.org.uk/    ------------------------------    Date: 11 Feb 2003 04:36:29 -0800# From: dsauvage@altern.org (psycose) Y Subject: convert the decimal value of TERM_CHAR to a decimal ASCII text string in C with  < Message-ID: <5b82c382.0302110436.93a5fb2@posting.google.com>   hi all, ? well i'm using smg$read_keystroke, so i got the term_char code, F now i want to convert it into an ascii character,i have read it's like that in fortran ...  >C+ A >C Use OTS$CVT_L_TI to convert the decimal value of TERM_CHAR to    >C a decimal ASCII text string.  >C-  > 3 >        STATUS = OTS$CVT_L_TI ( TERM_CHAR, TEXT )   >   ( but how can i do this in C programming ? thanks again  
 Best Regards,  David    ------------------------------  % Date: Mon, 10 Feb 2003 10:00:24 +0100 0 From: TRAWINSKI <robert.trawinski@softax.com.pl>@ Subject: Re: DECWindows Motif on a DS10L without a graphics card. Message-ID: <b27pme$jt0$1@bozon.softax.com.pl>   Billy Bobette2002 wrote: > Hi,  > J > I recently acquired a DS10L which I've configured to run OpenVMS v7.3-1.H > During the install I requested that DECWindows Motif be installed. TheM > machine does not have a graphics card at present (or a keyboard, monitor or J > mouse) and I access it from my PC using Reflections via TCP/IP (TELNET). > L > Given this hardware setup is it possible to install some software on my PCM > to act as the Motif Windows Manager and use my PC as the front-end to VMS ? M > I remember in the past is was pretty easy to redirect DECTerms etc from one I > workstation to another but I'm not sure about having the whole MWM on a  > remote machine.  > L > Incidentally I have Pathworks32 which I believe has a version of X11 for a > PC.  > 
 > Regards, > Steve  >  >   6 If you've got TCPIP v.5.3 you can turn on XDM services   $ TCPIP  TCPIP > ENABLE SERVICE XDM  E Than you have to configure your Xserver software on  PC to work with  H XDMCP protocol. You'll see login screen from Alpha/VMS on PC (depending + on Xserver configuration). It works for me.    Regards, Robert   ------------------------------  % Date: Mon, 10 Feb 2003 18:18:56 +0100 2 From: martin@radiogaga.harz.de (Martin Vorlaender)@ Subject: Re: DECWindows Motif on a DS10L without a graphics card; Message-ID: <3e47df00.524144494f47414741@radiogaga.harz.de>   1 TRAWINSKI (robert.trawinski@softax.com.pl) wrote:  > Billy Bobette2002 wrote:H > > Given this hardware setup is it possible to install some software onI > > my PC to act as the Motif Windows Manager and use my PC as the front-  > > end to VMS ? ... 8 > If you've got TCPIP v.5.3 you can turn on XDM services > 	 > $ TCPIP  > TCPIP > ENABLE SERVICE XDM  @ ...and have to read the appropriate chapter of the Compaq TCP/IPC Services manual. You have to set up some config file for it to work 7 (sorry to be fuzzy - no 5.3 manual in sight right now).    cu,    Martin --  D                     | Martin Vorlaender    |    VMS & WNT programmer-   Smiert Spamionem  | work: mv@pdv-systeme.de D                     |       http://www.pdv-systeme.de/users/martinv/4                     | home: martin@radiogaga.harz.de   ------------------------------    Date: 11 Feb 2003 09:12:14 -0800# From: pmartyn@csc.com (Phil Martyn)  Subject: DFO 2.7 and scripts= Message-ID: <23088ca1.0302110912.557198f4@posting.google.com>   C Anybody have any experience with DFO 2.7 and scripts. (3-node Alpha C cluster running OpenVMS 7.3 + all patches)I've just upgraded a test F system from DFO 2.6 and I now find that any scripts using predecessorsD won't run. The DFO scheduler process is running happily, the scripts> appear fine. And scripts with a specific run date/time run OK.D However, once the 'main' script completes the successor doesn't kickD in. (A manual start of the successor script executes OK). Any clues.E BTW - I've logged a call with HP/Compaq but nothing from them either, ( just yet. Any clues...Thanks in advance.   ------------------------------  % Date: Mon, 10 Feb 2003 16:16:56 -0500 * From: Paul Anderson <paul.anderson@hp.com>P Subject: Re: Doc Format Gripes (was: Re: HP hideousness of the OPenVMS web pages5 Message-ID: <100220031616562885%paul.anderson@hp.com>   E In article <00A1B482.0CE1DA9C@SendSpamHere.ORG>, < @SendSpamHere.ORG>  wrote:  E > It is not a portable standard until DEComHPaq provides a VMS reader 
 > for PDF!  B Why do you need an HP-supplied PDF reader?  (Not that I would mind having one.)  C > POSTSCRIPT???  HP stopped making POSTSCRIPT capable printers, eh?   @ Yeah, everybody just prints Word documents anyway, so we dropped PostScript and PCL.  ;-)  A PostScript is fine if you want to print the document (or portions 7 thereof), but it lacks a certain ease of searchability.   C DCPS provides documentation in five formats (Bookreader, HTML, PDF, D PostScript and text) with all but PDF produced by DECdocument.  MostB layered products don't supply all those formats, and even the DCPSD documents are available on the DCPS CD-ROM, not the quarterly Online Documentation Library.   Paul   --    Paul Anderson   OpenVMS Engineering    Hewlett-Packard Company    ------------------------------    Date: 10 Feb 2003 16:13:52 -0600+ From: kuhrt@encompasserve.org (Marty Kuhrt) P Subject: Re: Doc Format Gripes (was: Re: HP hideousness of the OPenVMS web pages3 Message-ID: <JTNHygWIxisG@eisner.encompasserve.org>   P In article <b28inc$dm0$1@web1.cup.hp.com>, hoff@hp.nospam (Hoff Hoffman) writes: > K >   Bookreader will be around as a viewer for quite a while, simply because I >   the older documentation hasn't and likely won't be converted forward  J >   into PDF or other newer formats.  Additionally, newer SGML-based toolsM >   used by the HP writers do not presently have the capability of generating J >   old DIGITAL proprietary-format file formats such as Bookreader, as (toJ >   my knowledge) no one has written a converter to translate PDF or otherL >   output into Bookreader-format files.  (This desire to move over to SGML-" >   based tools predates HP, too.)  C Isn't DEC Document SDML, which is an extended SGML?  Wouldn't that  @ mean that DD could be modified to handle SGML input?   Or is my   assumption way off the mark(up)?   ------------------------------   Date: 10 Feb 2003 22:19:30 GMT# From: hoff@hp.nospam (Hoff Hoffman) P Subject: Re: Doc Format Gripes (was: Re: HP hideousness of the OPenVMS web pages* Message-ID: <b298hi$ne7$1@web1.cup.hp.com>  S In article <00A1B482.0CE1DA9C@SendSpamHere.ORG>, VAXman-  @SendSpamHere.ORG writes:   H   At least some of you reading this reply -- particularly those that areI   running Alpha boxes more than five or six years old -- are probably not )   going to like this particular response.   Q :In article <b28inc$dm0$1@web1.cup.hp.com>, hoff@hp.nospam (Hoff Hoffman) writes:  : H :Mozilla, and I can get over it's pig-like appetite for system resourcesH :and it's generally sluggish take-a-coffee-break performance, cannot getH :me through most of the HP pages for one reason or another.  Until thereG :is a browser that can accomplish this feat, please have the web site's 4 :pages consider those of us using this "technology".  F   I knew I left something out.  Mozilla performance is bad on most anyH   Alpha system prior to EV56 (circa 1996), and on any memory-constrainedH   Alpha systems.  I tend to lack the patience for Mozilla use on an EV45D   AlphaStation 255 series box, but Mozilla works quite nicely on an    AlphaStation XP1000 box.    C   My rule of thumb: Java and Mozilla require Alpha EV56 or later.     *   I've mentioned my RoT before, of course.  E   Problems with performance of newer applications are not specific to G   older Alpha of course, as older SPARC, older Pentium and  other older C   microprocessors all encounter heavy loading when running Java and 3   Mozilla and other similar and newer applications.   I   One of the underlying concerns here and one that is specific to OpenVMS J   likely involves the sudden increase in system requirements combined withJ   a comparatively long history of gradually increasing system loads within(   OpenVMS and core OpenVMS applications.  D   Alpha systems that predate 1996 or so simply don't do so well withC   Java-based applications, and this jump in system requirements was C   comparatively sudden.  (The previously mentioned AlphaStation 255 E   serves just fine for most user interface tasks, but is underpowered <   for with Mozilla and Java and other related applications.)  D   As referenced earlier, local Pentium boxes (700MHz and 1GHz) don'tE   operate particularly well when running newer system loads (Mozilla, E   OpenOffice.org tools, Java, etc), either.  (And these Pentium boxes E   are newer than 1996, too.)  This increase in system requirements is #   not specific to any one platform.   G   I should also mention that our continued support for Alpha systems of F   generations prior to EV56 increases the difficulty accessing certainH   optimizations within OpenVMS -- compatability is not without some moreI   hidden costs.  (The compiler folks have implemented some clever schemes H   while adding EV56 code paths into images compiled from code emitted byI   the newer code generators, too.)  Tru64 UNIX decided to drop support in H   new releases for older Alpha systems, and I know of various major ISVsI   that are considering dropping support for pre-EV56 boxes for reasons of    performance.  J   As for upgrades, (used) EV6-based Alpha systems are available to OpenVMSI   hobbyists for under US$1K.  I expect nice (used) EV56 workstation boxes    can be easily had, as well.   I :>  Most new OpenVMS documentation is and will be in PDF format, as it is J :>  the de facto portable standard format for these materials -- the other : H :It is not a portable standard until DEComHPaq provides a VMS reader for :PDF!     H   So your central concern is around the integration and shipment and theH   support of the XPDF or other PDF reader in/into OpenVMS?  (Versions ofH   XPDF already ship with every OpenVMS release, as part of the Freeware,G   so this is clearly not one of simple availability of the reader.)  If I   you want XPDF support on OpenVMS from another vendor, there is (or was)    one offering this.    I :>  If there is a customer base that wishes to purchase Bookreader-format M :>  documentation materials, I would expect that a special distribution could K :>  be created.  (This would not be cheap, and I expect that would be a key + :>  factor in its probable low acceptance.)  : G :I can get over the Bookreader but how about then HTML that can be read  :without bastard Javascript.    D   I usually run with Java and JavaScript disabled, only occasionally9   enabling it when I encounter a poorly-designed webpage.   G   As has been remarked upon before, the golden age of web compatability G   has passed.  This is unfortunate, of course.  I and others here would I   hope to maintain compliance with applicable W3C standards -- that said, D   there is another high-profile project keeping us OpenVMS engineers   rather busy right now.     :POSTSCRIPT??? : 3 :HP stopped making POSTSCRIPT capable printers, eh?   H   Adobe apparently recalled most or all of the official Adobe PostscriptH   implementations, per the Adobe licensing agreements.  This contractualI   removal of Postscript is unfortunate and caused problems for DECwindows H   and particularly for the server-side Postscript support.  This removal!   has also been discussed before.   K   Ghostscript is available for OpenVMS and other platforms, but that option K   also violates your previous requirements for integration into and support    with the operating system.    J   AFAIK, HP provides various printers with emulations of Adobe Postscript,J   but does not offer new printers with Adobe Postscript support.  I do notI   know if these printers are based on Ghostscript, or another technology.     N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Tue, 11 Feb 2003 07:46:32 -0700 . From: "David D Miller" <ddmiller@raytheon.com> Subject: Help with CDsF Message-ID: <OFDBA79B99.A9E570B4-ON07256CCA.0050C0A9@rsc.raytheon.com>   Folks:  / This is an unusual request, please hear me out.   J Steve Hoffman and Digital Press have asked me to finish up the 2nd editionI of Baldwin's _OpenVMS System Management Guide_. As most of you know Steve I is deeply into OpenVMS/IA64 and he didn't have enough hours in the day to 0 finish his update. He made a great start though!  C Naturally I'd like to show examples using OpenVMS 7.3-1. However my D AlphaStation 200 4/233 is currently running 7.1 and it will take twoK updates to get to 7.3-1. I was planning to get a 7.3 hobby CD for the first I step and then purchase 7.3-1 from HP. But, as you know, the hobby disk is ! sold out so that plan is no good.   F I really hate to buy two CD sets from HP - each one is $400 and that'sK really expensive on my budget. I was hoping some kind soul would lend me an D old CD set (any CD set post 7.1) so I could make the transition more inexpensively.  K Baldwin/Hoffman/Miller will be my 4th book for Digital Press in the past 12 J years.  For those that don't know, authors don't get rich on book sales --J I'll probably get less than a dollar for each Baldwin/Hoffman/Miller sold.C We write for less tangible goals. In this case, as a service to the H community. My previous books were used as stepping stones in my teaching career.   G If you can help, please contact me off-line at this address. You'll get % your name in the Acknowledgements! :)   
 TIA, dave.   ------------------------------  # Date: Mon, 10 Feb 2003 20:16:03 GMT . From: peter@langstoeger.at (Peter LANGSTOEGER)) Subject: Re: Hobbyist PAKs from Montagar? 5 Message-ID: <7MT1a.255976$xv1.2999098@news.chello.at>   B In article <03021010234017@antinode.org>, sms@antinode.org writes:C >   Has anyone gotten a Hobbyist PAK e-mail message out of Montagar E >recently?  I've had no success for about a week, and expirations are 	 >looming.   D I've now an ES40 and can't get a license PAK either (before you ask,< yes, hobby; no, not bought, only borrowed for some weeks ;-)D I thought, it might be of the ES40 being a "unlisted AlphaServer"...   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Tue, 11 Feb 2003 13:52:57 -0500 ! From: VAXVMS <bounce@notmail.com> ) Subject: RE: Hobbyist PAKs from Montagar? K Message-ID: <BA52530E3149734A9BAABDBBFA808E4903027BC5@rlghncst964.usps.gov>    Forgive the top-posting.  > David and I exchanged emails yesterday (time to pay him for my2 web space) and I mentioned this thread in passing.  
 He said that    " 1) he was just back from vacation;" 2) he's going through his emails; C 3) something's munged in the hobbyist license PAK issuing process;   4) he's working on it.   WWWebb ============================  B In article <03021010234017@antinode.org>, sms@antinode.org writes:C >   Has anyone gotten a Hobbyist PAK e-mail message out of Montagar E >recently?  I've had no success for about a week, and expirations are 	 >looming.   D I've now an ES40 and can't get a license PAK either (before you ask,< yes, hobby; no, not bought, only borrowed for some weeks ;-)D I thought, it might be of the ES40 being a "unlisted AlphaServer"...   -- Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ========================  William W. Webb- EMS Operations  OpenVMS Systems Support - USPS' 4924 Green Road, Raleigh, NC 27616-2800 
 919.874.3043  * <FirstInitialLastNameAtEMAILDotUSPSDotGOV>   ------------------------------  % Date: Tue, 11 Feb 2003 00:19:51 +0100 ; From: "Jerome" <Jerome.Forissier@removethis.libertysurf.fr> D Subject: Re: How to check - write cache enabled/disabled of a device4 Message-ID: <newscache$yx84ah$57k$1@news.tiscali.fr>  0 "Nagaraj" <Nanu_Nagi@Yahoo.com> wrote in message7 news:d561ea2d.0302070814.286f6ade@posting.google.com... 	 > Hi All,  > D > I would like to get whether the device write-back cache enabled or > disabled.  > 8 > Say for example our device supports tape data caching. > E > Using SYS$GETDVI or some other system services I would like to know D > whether it is enabled/disabled. I'm able to get whether the deviceE > supports write-backup caching. (Here i'm checking for DEV$M_WBC, it G > gives me it supports the same but it won't tell whether it is enabled  > or disabled)   <snip>  E The OpenVMS I/O User's Reference Manual mentions the MT2$V_WBC_ENABLE E characteristic. That may be what you're looking for. Take a look at : L http://www.openvms.compaq.com/doc/731FINAL/6136/6136pro_010.html#mag_tape_ch ap6 5 (Short link: http://makeashorterlink.com/?R33F21A63).    -- Jerome   ------------------------------  # Date: Tue, 11 Feb 2003 12:59:10 GMT 9 From: Alan Adams <alan.adams@orchard-way.freeserve.co.uk> 4 Subject: Re: HP hideousness of the OPenVMS web pages? Message-ID: <c09318c34b.Alan.Adams@orchard-way.freeserve.co.uk>   * In message <b23e3b$dkm$1@aquila.mdx.ac.uk>6           david20@alpha2.mdx.ac.uk (David Webb) wrote:  e > In article <3E43FF74.21682DF7@aaa.com>, Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> writes: * > >And there, in nutshell, is the problem.- > >The c.o.v most-frequent-writers, does not  $ > >represent the VMS user community. > > - > >My *guess* is that >90% of those that have , > >to read VMS docs, acualy does this on (or, > >"from" or "through") a PC. They have a PC
 > >anyway....  > >  > >Jan-Erik Sderholm. > >  > K > And my guess is that >90% of those that have to read VMS docs may want to N > do it on a VMS system. They have a VMS system anyway - and most systems soldO > recently usually have a PC style monitor and graphics card on them - at least  > where I work.  > L > Whether my guess or your guess is representitive of the VMS user communityN > is impossible to gauge. The silent Majority don't speak hence the only guideG > we have to the VMS user community is those who post to forums such ass > comp.os.vms.  J When I was working with VMS (i.e. up to last April) we had 4 servers, withL VT510 consoles, and two workstations which were for user applications. I ranJ the whole system from a PC with Powerterm, having discovered it to be moreJ useful than a VT510, e.g. better screen sizes, nice cut and paste etc. TheL resize when changing the window size I particularly liked, as it changed theC font to keep the 80 or 132 columns displayed in full. No horizontale scrolling needed.   J This made Bookreader a pain, because the PC X display was slow and clunky.7 The bits of the DOCs in HTML were far better to access.d  L At home, when I eventually get the system running I will not have a PC or VTF available, so the options will be an emulator either serial or telnet,J running on a RISC OS system (the ones which used to be Acorn, and continue( after the demise of Acorn as a company).   >  >  > M > In my office - two people who manage VMS systems (myself and one other) and5N > two other people who act as Oracle DBAs on VMS systems each have a VMS AlphaM > workstation on their desks. All but one of the systems down in the computer.A > centre has a graphics monitor running X windows as its console.kP > Of those 4 people I'm probably the only one who could be classed as a frequent > poster to comp.os.vms. >  >  >  > David Webb > VMS and Unix team leader > CCSS > Middlesex University >    >  > >o > >n$ > >VAXman-, @SendSpamHere.ORG wrote: > >>   > >> Survey starts now...  > >> p > >> Me! > >> a > >> Survey ends now...e > >> M > >> Survey says... 100% > >>   -- e
 Alan Adams& alan.adams@orchard-way.freeserve.co.uk http://www.nckc.org.uk/    ------------------------------  % Date: Tue, 11 Feb 2003 18:12:35 +0100 $ From: Michael Unger <unger@decus.de>N Subject: HP's Web Pages (was: Doc Format Gripes (was: Re: HP hideousness ...))* Message-ID: <00A1B581.9E0719A3.8@decus.de>  & "Hoff Hoffman" <hoff@hp.nospam> wrote:   > [...]d >JF >   I usually run with Java and JavaScript disabled, only occasionally; >   enabling it when I encounter a poorly-designed webpage.t  @ To mention just one of these "poorly-designed" web pages: the HP! AlphaServer "Options" page, e.g.,   H http://h18002.www1.hp.com/alphaserver/options/asds10/asds10_options.html  C It is simply useless with JavaScript disabled. (Selection of options; classes -- category, subcategory, detail -- is impossible.)    > [...]F >d > :POSTSCRIPT??? > :e5 > :HP stopped making POSTSCRIPT capable printers, eh?o > J >   Adobe apparently recalled most or all of the official Adobe PostscriptJ >   implementations, per the Adobe licensing agreements.  This contractualK >   removal of Postscript is unfortunate and caused problems for DECwindows J >   and particularly for the server-side Postscript support.  This removal# >   has also been discussed before.S >  > [...]  >oL >   AFAIK, HP provides various printers with emulations of Adobe Postscript,L >   but does not offer new printers with Adobe Postscript support.  I do notK >   know if these printers are based on Ghostscript, or another technology.s  > Do you happen to know the reason why Adobe changed the licenseE agreements? As far as I know Adobe doesn't make printers itself so itt@ should be interested in "official" implementations of PostScript, interpreters. "Emulations" tend to be buggy.   Michael5   ------------------------------  # Date: Mon, 10 Feb 2003 06:53:41 GMTi  From: CJT <cheljuba@prodigy.net>- Subject: Re: IBM says AMD dead in 5 years ... * Message-ID: <3E474C81.9010502@prodigy.net>   John Smith wrote:w <snip> > G > IMHO, the current Administration is less likely to complain about IBMS- > buying AMD than the Clinton one would have.B >  >   9 The current administration probably wouldn't even notice.    ------------------------------    Date: 10 Feb 2003 13:11:22 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) - Subject: Re: IBM says AMD dead in 5 years ...i3 Message-ID: <OnWP0oW97Z5K@eisner.encompasserve.org>h  a In article <Ogn1rPEPpaY3@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:e' > 	Yes of course... and Eunice concurs:e > N > While Sun deserves credit for proving that powerful Unix servers can competeM > against Intel systems, it's clear computers using Sun processors eventually > > will be inundated, said Illuminata analyst Jonathan Eunice.   F    Oh yeah, great.  Except that a lot of powerfull Unix servers are on    Intel systems.-   ------------------------------    Date: 10 Feb 2003 12:25:33 -0600+ From: young_r@encompasserve.org (Rob Young):- Subject: Re: IBM says AMD dead in 5 years ...s3 Message-ID: <jXsJI5Ca+qsL@eisner.encompasserve.org>.  q In article <OPR1a.633646$F2h1.512899@news01.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes:g > : > "Rob Young" <young_r@encompasserve.org> wrote in message  = > "You can't fight it. It's like trying to fight the future."g" > As in "Resistance is futile."???  > 	Exactly.  Someone was probably very attached to wood burning,B 	steam powered transportation (whether train or auto).  Resistance6 	to oil powered vehicles/transportation proved futile.  = 	And someday resistance to hydrogen/water/pick_a_fuel powereda; 	transportation proves futile but we can join in the chorus0 	then:     			I strip away the old debris 			That hides a shining carn 			A brilliant red Barchetta 			From a better vanished time 			I fire up the willing enginet 			Responding with a roara 			Tires spitting gravel 			I commit my weekly crimea   			WindS
 			In my hair  			Shifting and drifting 			Mechanical musicg 			Adrenaline surge...   			Well-weathered leathera 			Hot metal and oil 			The scented country air 			Sunlight on chromea 			The blur of the landscape 			Every nerve aware    0 	And dream of the day of oil powered vehicles.		  G 	Likewise, pine for the days of our Commodores, Amigas and UltraSparcs.n   				Robd   ------------------------------  % Date: Mon, 10 Feb 2003 18:05:54 -0500 * From: "Bill Todd" <billtodd@metrocast.net>- Subject: Re: IBM says AMD dead in 5 years ...w2 Message-ID: <2aOcncj_5MzDrdWjXTWcoA@metrocast.net>  L "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote in message( news:3e47e487$1_1@hpb10302.boi.hp.com... >l7 > "Baby Peanut" <baby_p_nut@yahoo.com> wrote in messageo7 > news:c5cf6e8.0302100828.22bfaed@posting.google.com...-7 > > bob@instantwhip.com (Bob Ceculski) wrote in messageh: > news:<d7791aa1.0302061335.6341e20@posting.google.com>.... > > > http://www.theinquirer.net/?article=7641 > >fI > > Don't forget that the Hammer runs 32-bit code faster than the Itanic.. > >wB > > People will buy the Hammer as a cheap, fast Pentium substitue. > F > Don't forget that Intel will make faster and cheaper IA32 chips than Hammer.K  E Hammer will be as fast as or faster than Pentium according to the twotJ companies' published schedules (for as far out as they go).  They'll startI about equal in Q2 (Hammer at 2 GHz and P4 at 3.2 GHz), but Hammer goes to I 2.2 GHz in Q3 and 2.4 (possibly 2.6) GHz in Q4, while P4 stays at 3.2 GHz L until Q4 when the Prescott shrink brings it to 3.4 (possibly 3.6) GHz.  ThenK Hammer gets its own shrink in early 2004 and continues to match Prescott inm; clock-rate increases to stay somewhat ahead in performance.d   - bill   ------------------------------  % Date: Mon, 10 Feb 2003 18:19:17 -0500h* From: "Bill Todd" <billtodd@metrocast.net>- Subject: Re: IBM says AMD dead in 5 years ... 2 Message-ID: <9pCcnYnTfcYfrtWjXTWc3Q@metrocast.net>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:jXsJI5Ca+qsL@eisner.encompasserve.org...r > In articleF <OPR1a.633646$F2h1.512899@news01.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes:  > >u< > > "Rob Young" <young_r@encompasserve.org> wrote in message >o? > > "You can't fight it. It's like trying to fight the future." $ > > As in "Resistance is futile."??? >i? > Exactly.  Someone was probably very attached to wood burning, C > steam powered transportation (whether train or auto).  Resistancei7 > to oil powered vehicles/transportation proved futile.   H Er, that was because the newer technologies had at least some compellingH advantages over the old ones.  Itanic had *no* compelling advantage overH Alpha:  zilch, zero, nada.  If Compaq hadn't traded it the Alpha team toJ give it at least *some* prospects for becoming a worthwhile replacement inE the future, it would likely be dying already - and Alpha, with even aa= modicum of support from its owner, would be on the ascendant.g  H (By the way, your analogy would still be lousy even if Itanic *had* someH reason for existence:  it took decades for steam transportation's matureF advantages to be equaled by the newer technologies, during which steamL continued to make bundles of money for its vendors.  Alpha could easily haveG done the same for at least the remainder of this decade, even if Itanic % *eventually* could have been better.)r   - bill   ------------------------------  % Date: Mon, 10 Feb 2003 18:21:33 -0500-* From: "Bill Todd" <billtodd@metrocast.net>- Subject: Re: IBM says AMD dead in 5 years ...>2 Message-ID: <tKucnaETMYOUqdWjXTWcqg@metrocast.net>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:Ogn1rPEPpaY3@eisner.encompasserve.org...l   ...i  	   PerhapseB > someone gets a clue at the Sun BOD level and mentions:  "we sure= > are spending a lot of money on R&D, by the way ... have you 8 > heard how much money HP is saving ditching their CPU?"  I At which point anyone with half a clue responds "Yes - but have you heardrC how much more money they *lost* as a result of ditching their CPU?"s   - bill   ------------------------------  % Date: Mon, 10 Feb 2003 18:28:10 -0500s* From: "Bill Todd" <billtodd@metrocast.net>- Subject: Re: IBM says AMD dead in 5 years ...t2 Message-ID: <1qCcnaXrsfwLqNWjXTWcqQ@metrocast.net>  5 "Bill Todd" <billtodd@metrocast.net> wrote in message), news:9pCcnYnTfcYfrtWjXTWc3Q@metrocast.net...   ...   3   it took decades for steam transportation's mature.H > advantages to be equaled by the newer technologies, during which steam5 > continued to make bundles of money for its vendors.l  D At least for trains; steam-powered automobiles were replaced faster.G Another interesting parallel there, in that the heavy-duty applicationseG (like Alpha's) held out successfully while light-duty applications wereeG quicker to change.  For that matter, stationary applications still maken) major use of steam technology even today.n   - bill   ------------------------------  % Date: Mon, 10 Feb 2003 16:19:55 -0500d* From: "Leo Demers" <leo_dot_demers@HP.COM>B Subject: Kerberos T2.0 field test kit for hp OpenVMS now available* Message-ID: <b2955i$nbo$1@web1.cup.hp.com>   Dear OpenVMS Customer,  J HP is pleased to announce the availability of the Kerberos T2.0 field test kit for hp OpenVMS Alpha,.  E based on MIT Kerberos V5 Release 1.2.6. This new kit is available fore
 download froma  9 http://www.openvms.compaq.com/openvms/products/kerberos/.C  J MIT has issued a CRITICAL Security Advisory for KDC vulnerabilities in all releases of-  (  MIT Kerberos V5 prior to Release 1.2.5.   For more information, seecL http://web.mit.edu/kerberos/www/advisories/MITKRB5-SA-2003-001-multiple.txt.  K Kerberos Version 1.0 for OpenVMS is based on MIT Kerberos V5 Release 1.0.5,d  2 and is affected by these security vulnerabilities.  H HP strongly advises all of our Kerberos customers to upgrade to Kerberos T2.0 for hp OpenVMS.  L  The Kerberos T2.0 field test kit is available for your testing now, and the final Version 2.0 will bet  # made available as soon as possible.o  L Please note that the T2.0 kit is experiencing a problem using the Kerberized Telnet server in hp-  E TCP/IP Services V5.3 for hp OpenVMS. We are working on resolving thish problem and will  9  provide updates on the Kerberos for hp OpenVMS web page..  : Thank you for your continued interest in OpenVMS Security.   Regards,  
 Leo Demers    OpenVMS Security Product Manager   OpenVMSSecurity@hp.com   ------------------------------    Date: 11 Feb 2003 08:04:40 -0800( From: bob@instantwhip.com (Bob Ceculski)9 Subject: Marvel:  Even more marvelous that first thought!r< Message-ID: <d7791aa1.0302110804.275523d@posting.google.com>  9 http://www.openvms.org/stories.php?story=03/02/10/4408466u   ------------------------------  % Date: Tue, 11 Feb 2003 10:14:29 -0500o+ From: "Martin O'Connor" <moconnor@dvfs.com>o' Subject: Re: OpenVMS Boot press releaseh6 Message-ID: <b2b40m$1aerjl$1@ID-118202.news.dfncis.de>   Sue,X Most of us who "read" teh news group but only speak occasionally greatly appriciate your9 information. Don't let a few spoil it for the rest of us.s   Marty O'Connor  @ "Sue Skonetski" <susan.skonetski@hp.nospam.com> wrote in message$ news:b2atvh$n9l$1@web1.cup.hp.com... Folks,  1 We have come to an impasse, the newsgroup and me.t  I I can either wait until stuff shows up on the official HP site's and thenaG post it or I can post it as soon as I get it.  I thought that the choir- would want to know first. G But if I post stuff early and just get negative response it seems a bitdL pointless to me.  And while folks may consider it preaching to the "choir" IK prefer to think of it as "family" and you always take care of your "family"0E first and "family" does not mind if there are a few uncrossed T's and 
 undotted i's.w  & The choice is yours, just let me know.   suef    @ "Sue Skonetski" <susan.skonetski@hp.nospam.com> wrote in message$ news:b28hg8$dne$1@web1.cup.hp.com...@ > http://h18003.www1.hp.com/hps/ipf-enterprise/openvms_boot.html >  >i   ------------------------------  % Date: Tue, 11 Feb 2003 10:23:31 -0500o/ From: "Hank Vander Waal" <hvanderw@mansply.com>r' Subject: RE: OpenVMS Boot press release ; Message-ID: <000a01c2d1e1$89af0270$cb96a8c6@manufact5l8vs8>a   Sue,  ? Let us know as soon as you know.   Family ALWAYS comes first !!t? Ignore the "black sheep" of the family (every family has 'em!!)w   Hank Vander Waal  p/ VMS - is there anything that DARES to compare??-! Unix  - for hackers - by hackers!- You get what you pay for !   ------------------------------  % Date: Mon, 10 Feb 2003 13:47:09 +0000l' From: Andrew Harrison SUNUK ConsultancyD6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 31, Message-ID: <3E47AD5D.30905@nospamn.sun.com>   Fred Kleinsorge wrote:0 > "John Smith" <a@nonymous.com> wrote in messageF > news:d3z%9.253525$pDv.240534@news04.bloor.is.net.cable.rogers.com... >  >>G >>Come on Fred...with a Sparc port pretty soon everyone will be calling D >>OpenVMS the 'Portable Linux of Real Operating Systems' (tm).   :-) >>H >>So if x86-64 is good enough for Sun to move their high-end servers to,6 >>HP would also port OpenVMS to x86-64 and dump IA-64? >> >  > K > I don't think I've ever said that x86-64 can't work.  But it doesn't haveC > any J > compelling reason for HP to dump IA64 in favor of it.  It is our PA-RISC > replacement. >   : Of course it is, but what HP does or thinks doesn't matter. its Intel that is having to carry the IPF can.  : The fact that IPF is the replacement for PA-RISC is a huge< part of the problem. Intel got to where they are now largelyB by making very small incremental changes to x86, people complained> about the architecture but Intel kept its ISV's on board. TheyA had abortive programs to get out of x86 i960 for example but they B all failed because of lack of ISV support and lack of design wins.  A Now Intel has done exactly what it avoided doing all those years,f> completely broken compatibility with their existing volume ISV< platform and it shows, of the 16000 or so x86 apps there are; currently ~100 available and supported for IPF notable gaps = include Oracle. And this is with the full weight of Intel and 1 HP's marketing thrown into getting ISV's to port.I      H > x86-64 has plenty of merit when viewed as a 64-bit upgrade to the IA32K > market.  That is really it's strength.  It only has any meaning if AMD isC > ableG > to A) compete in price/performance with IA32;  and B) they can find a K > server systems vendor who will build the high margin and high end systemsm > out of it. >   F There is every indication that they will be able to do both. Althon-64@ is a cost reduced CPU at around 100 million transistors and in a> 130 nanometer process it should be competitive price wise with IA32 CPU's from Intel.  ; They already have one high end systems design win from Crayt? and I have little doubt that they will appeal to their existinge OEM's such as Dell, Sony etc.   < In addition their proposition is now the one that made Intel9 so dominant, gradual and compatible changes that preserveX9 compatibility for ISV's. If you want good x86 performancem8 for existing apps well fine you can have it, if you want7 64 bit support for games, imaging, etc etc then you canl? have it without and x86 compatibility issues. Not a proposition  available to Intel.   < Hammer offers AMD the opportunity to establish themselves as@ the processor that people want in their desktops for performance< reasons, rather than their current propostion which is based around low cost.  = IPF is red ink for Intel their profits come from IA32, losingm9 market share to a processor that they cannot compete withp> from a functionality standpoint should be a big fear for them.    N > Sun is saddled with an aging, toothless Sparc architecture, and will need toH > dump millions into it to make it competetive.  At a time when they are
 > bleedingI > red ink and can't really afford the R&D.  Plus, since they are FABless,rM > they will always have a problem getting onto the bleeding edge processes infJ > a timely way.  IA64 can't hold much interest for them, because they lose > someJ > of their ability to differentiate themselves from HP and others.  x86-64 > would N > be a way of placing the bet on who will win, and differentiating themselves. >   A Wonderfull a zero sum content posting from Freddy Boy, you reallyoH earned it this time for displaying complete lack of knowledge in public.  > USIII is currently being fabricated in a 130 nanometer process hence its 53 watt power budget.0  @ Itanium is currently being fabricated in a 150 nanometer process@ hence its 130 watt power budget well that and having 220 million transistors.  - Guess which is the more leading edge FAB ????   @ And of course going FABless something that you may remember yourC old employer doing (how short your memory is) gives Sun the abilityRA to choose the best available FAB. Currently we use 2 FAB partnersm< but we could easily add another if their process was better.  C As for aging, SPARC is no more ageing than IPF, they both date back > to the 80's or don't you know you own processor history. Heard= of HP-WW or Cydrome ????? It just took HP a bit longer to geti IPF to market.  > It really would be better for all involved but most of all you= is you restricted your comments to subjects that you know andv= can articulate, USB support in OpenVMS for example, stay wellt; away from IPF, SPARC, Power, performance, UNIX, Linux, DBMSe= performance. Stick to what it says in your job spec its going- to be less embarassing.      Regards0 Andrew Harrisono   ------------------------------  % Date: Mon, 10 Feb 2003 18:08:05 +0000i' From: Andrew Harrison SUNUK Consultancyg6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 31. Message-ID: <3E47EA85.8030008@nospamn.sun.com>   Fred Kleinsorge wrote:? > "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in message,+ > news:3E46C9B1.80D61FA3@vl.videotron.ca...  >  >>Keith Parris wrote:c >>C >>>computing paradigms with new CPU architectures sweep through theeE >>>industry about every 10 years or so, so VMS will need to be portedtH >>>again in another 10 years, and again 20 years from now, 30 years from >>>now, and so on. >>L >>The problem with the port to IA64 is the possibility that a port to a more@ >>viable platform will be needed before EV79 runs out of breath. >> >  > L > Huh?  IA64 is in the top tier of performance for 64-bit architectures.  DoJ > you really expect some breakthrough that will propel Power or Opteron toI > some higher plateau which IA64 will not be able to compete at?  Hardly.- >  > ? >>The confidence level of IA64's long term success is not high.( >  > L > Amongst who?  You?  The less than 6 vocal critics in this group?  Among CS& > types who just hate it for it's ISA? >     Well how about the ISV's Freddy.   IA32 has 16000+ apps., HP-PA has 5000+ apps.. IA-64 has ????  apps.a  1 IA-64 has been available to developers since 2001s/ Oracle just released support for HP-UX on IA-64o. and no other Oracle products, hardly a ringing endorsement.  - The SGI box runs Linux, no Oracle products ate all supported for that.o  . People who want to use Microsoft OE's are also going to be out of luck.    N > It would seem that HP, SGI, Fujitsu, and Intel among others would seem to be > confident enough.   ; Sure but what are people expected to run on these systems ?t Compilers !!  4 Can we rule HP and Intel out, you have to seem to be4 confident, anything else and you will be the turkeys who voted for christmas.   Regardsh Andrew Harrison    ------------------------------  % Date: Mon, 10 Feb 2003 18:11:59 +0000a' From: Andrew Harrison SUNUK Consultancyt6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 31. Message-ID: <3E47EB6F.4050902@nospamn.sun.com>   Fred Kleinsorge wrote:0 > "John Smith" <a@nonymous.com> wrote in messageG > news:qxP1a.633524$F2h1.557501@news01.bloor.is.net.cable.rogers.com...m > < >>"Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message/ >>news:xDl4AbfLe0Mu@eisner.encompasserve.org...f >>> >>>In article <3E472B53.3E25D057@fsi.net>, "David J. Dachtera" >>! >><djesys.nospam@fsi.net> writes:n >>F >>>>...additionally, Alpha's announced EOL falls within that "maybe 3" >>>  >>year >>E >>>>span, which still leaves VMS racing full throttle toward a bridgel >>>h >>that >>A >>>>does not yet exist. Continuing that metaphor, "the AlphaCide"h >>>y
 >>could be >>@ >>>>likened to cutting the brake lines on that speeding vehicle. >>> B >>>Alpha's announced EOL included the statement that Alpha systemsC >>>would still be made as long as there is a significant market foraF >>>them.  That is no different from systems using any other processor. >> >>E >>That's like asking in 2003, "Who wants to purchase a brand new 1998 H >>Mecedes for the same price as the new 2003 model?" Any takers? I think >>not. >> >  > M > Not true.  For instance, I prefer the beauty of the BMW Z3's lines over therN > Z4, or over the Z8's price ;-).  So, a *brand new* Z3 despite being designed& > in the mid-90's is still attractive. >   F Great, this explains a lot. It must be Ok because it looks Ok, doesn't8 perform that well but hey it looks alright so why worry.  2 And you thought I was talking about the Z3 !!! ;-)   Regards  Andrew Harrisonc   ------------------------------  % Date: Mon, 10 Feb 2003 16:40:45 -0400o0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 31/ Message-ID: <3E480E4C.7199A1DF@vl.videotron.ca>t   Larry Kilgallen wrote:A > Alpha's announced EOL included the statement that Alpha systemseB > would still be made as long as there is a significant market forE > them.  That is no different from systems using any other processor.     M Who defines what a "significant market" is ? Do you trust those who make thato
 decision ?  N What if nobody is interested in IA64 and stays with Alpha, and HP finds itselfN with a failing IA64 project and needs to push more people onto it to save it ?K It can then easily state that the Alpha demand isn't strong enough and bam,e production ends within 2 days.  J Also, look at ALL-IN-1. Just to please Microsoft, Palmer sabotaged its ownS veryt profitable All-in-1 in order to shove exchange onto itself and its customers.f  J I would not be surprised one bit if HP accelerated the retirement of Alpha! just to help its struggling IA64.p   ------------------------------  % Date: Mon, 10 Feb 2003 16:50:20 -0400f0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 31/ Message-ID: <3E48108A.624F12E8@vl.videotron.ca>h  * Alan Winston - SSRL Admin Cmptg Mgr wrote:M > Seriously, how many people have "must have a really kick-ass character-cell B > oriented mail client" on their server OS evaluation checklists?   P > this driving any serious purchasing decisions, especially when TCP/IP ServicesL > bundles IMAP and POP so you can use a variety of PC-based mailers, some of7 > which don't suck as bad as Outlook - or so I'm told.]     L It isn't so much the MAIL user interface that needs a rework, it is the mail architecture that does.   L For instance, ability to store content size in the message envelope (so that@ IMAP would be MUCH MUCH faster and consume far fewer resources).  H For instance, ability to store copies of messages you send in a specific5 folder instead of being sent to you as a new message.i  N Ability to create a message in a specific folder and not send it. (ok, perhapsL this is the same as ability to store a copy of outbound messages in "OUTBOX" or some other folder).  K Ability to store the message date other than date VMSmail received it. (eg:s6 the date the message was sent, not the date received).  M Field for message read receipt flag (no read receipt, read receipt requested,e read-receipt sent)  ' Better integration of RFC mail headers.a   ------------------------------  % Date: Mon, 10 Feb 2003 17:02:25 -0400m0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 31/ Message-ID: <3E48135E.6E19DA89@vl.videotron.ca>    Fred Kleinsorge wrote:N > VMS ran for the last decade on the fastest, most poweerful systems on earth.I > It was not a magic bullet.  It wasn't enough ON IT'S OWN to reverse thei > fortunes of VMS. -    H You can thank Alpha if you still work on VMS. Alpha saved VMS' butt fromI Palmer and company.  It is because of Alpha that VMS survived despite all- efforts by Palmer and company.  N Put VMS on a vanilla chip nobody wants, and it loses any edge it used to have,K at which point the lack of marketing, or Scoptt Stallard type of statementsM$ will bring what is left of VMS down.  K And with Unix systems now "claiming" to have high end clustering, VMS loseshK its edge because its lack of marketing prevents it from stating very loudlyv< that UNIX' clustering is nowhere near what VMS has to offer.  K Alpha gave VMS a good differentiator. Out it on a vanilla chip that doesn'ttL impress anyone, and VMS will have to stand on its own, without marketing and, without any serious push from its new owner.   ------------------------------  % Date: Mon, 10 Feb 2003 19:07:19 -0500m* From: "Bill Todd" <billtodd@metrocast.net>6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 312 Message-ID: <AA-dnZRidf9eo9WjXTWcqg@metrocast.net>  L "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote in message& news:3e47dcc5_2@hpb10302.boi.hp.com... >h0 > "John Smith" <a@nonymous.com> wrote in messageG > news:qxP1a.633524$F2h1.557501@news01.bloor.is.net.cable.rogers.com...u   ...   E > > If HP went to all remaining VMS & Tru64 customers today and askedoJ > > "Assuming HP put the same amount of effort into developing them, if weH > > gave you a choice of EV8x or IA-64 based systems to purchase 4 yearsJ > > hence, which would you prefer to purchase?", I bet the answer would be > > overwhelmingly EV8x. >wL > What would the answer be if the question was only access to very large EV8K > SMP systems at 10x the price of an IA-64 system, with no low-end offeringy at > all?  I It would probably be "Gee, if my vendor's that stupid, over the long termkK maybe I'd better shop elsewhere even if I don't like their systems as much:oK a competent vendor has better longevity prospects."  You know, kind of like J the way a lot of customers already reacted to the Alphacide (throw anotherJ bone-headed move at the customer base and you'd be able to pick up some of& the remainder who reluctantly stayed).  J IIRC EV68 systems are planned to hold down the low end for EV7.  Since theL EV6 core remains competitive with today's best competition it should be goodJ enough for the low end for quite a few years, though before then EV7 couldH likely be shrunk enough to qualify for low-end pricing (it'll be cheaperJ than the extra parts count for off-chip cache and memory controller - thatK might even be true today, but it'll be a cold day in Hell before any EV7 isSL sold at a price that reflects its marginal production cost:  that would tend to *expand* its market).   - bill   ------------------------------  % Date: Mon, 10 Feb 2003 19:17:16 -0500A* From: "Bill Todd" <billtodd@metrocast.net>6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 312 Message-ID: <MtOcnUpQApWE3NWjXTWc2Q@metrocast.net>  < "Andrew Harrison" <andrew.harrison@sun.com> wrote in message  news:3E47C7AE.9090309@sun.com...   ...t  8 > There are currently 2 reasons why it has a much larger > die size.s > . > 1 Its on a 150 nanometer process other CPU's+ > like SPARC are on 130 nanometer processes   G Since that's the second time you've said that, I guess is isn't a typo.dK Itanic2's process is 180 nm, and Madison's will be 130 nm (as is typical in ; the industry:  only SPARC made a 150 to 130 nm transition).    - bill   ------------------------------  % Date: Mon, 10 Feb 2003 19:32:23 -0500c* From: "Bill Todd" <billtodd@metrocast.net>6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 312 Message-ID: <cdmdnYP-hts52dWjXTWcrg@metrocast.net>  L "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote in message( news:3e47db5e$1_1@hpb10302.boi.hp.com... >V? > "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in message + > news:3E46C9B1.80D61FA3@vl.videotron.ca...S   ...i  I > > The problem with the port to IA64 is the possibility that a port to a  moreB > > viable platform will be needed before EV79 runs out of breath. > >- >-H > Huh?  IA64 is in the top tier of performance for 64-bit architectures.  I And Alpha was at the absolute top of 64-bit performance - but that didn'ta make it *viable*.d  K Of course, 'performance' has multiple dimensions and Itanic2 falls short in  many, possibly most of them.     DoJ > you really expect some breakthrough that will propel Power or Opteron to@ > some higher plateau which IA64 will not be able to compete at?  L POWER4/4+ is *already* at a level where Itanic can't compete:  it drives twoI cores (each comparable to Itanic2 in raw performance) with the same power J Itanic2 uses for one, and unlike Itanic2 (at least to the degree that I'veL seen benchmarks for the NEC and Unisys large-system Itanic2 offerings) has aL well-integrated multi-processor architecture (not quite like EV7's, but withG somewhat similar advantages in MP interaction and memory scaling).  IBMCG currently prefers to exact a premium price for such superiority, but iskL reportedly rapidly bringing to market versions that will compete down toward the low end.  F And Opteron will be all over Itanic in smaller configurations (up to 8J processors now, up to 16 when the dual-core Opterons appear) in both priceI and performance (its EV7-like bandwidth and inter-processor communicationuI won't be available in Itanics for at least 3 more years, and its raw corewF integer performance should equal Itanic's for the foreseeable future).  J As I've said from the start, Itanic's problem is that it's neither best atG *anything* (though that was wrong:  if you ignore power consumption, ittL really *is* best at FP-style performance, though even there not by too much)K nor good enough at enough things to be attractive.  If it succeeds, it will = be due to Intel's muscle rather than due to any innate merit.o   - bill   ------------------------------  # Date: Tue, 11 Feb 2003 00:38:34 GMT L From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 316 Message-ID: <00A1B4AB.512C602B@SSRL.SLAC.STANFORD.EDU>  b In article <3E48108A.624F12E8@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:+ >Alan Winston - SSRL Admin Cmptg Mgr wrote:sN >> Seriously, how many people have "must have a really kick-ass character-cellC >> oriented mail client" on their server OS evaluation checklists? t >sQ >> this driving any serious purchasing decisions, especially when TCP/IP ServicesiM >> bundles IMAP and POP so you can use a variety of PC-based mailers, some ofo8 >> which don't suck as bad as Outlook - or so I'm told.] >f > M >It isn't so much the MAIL user interface that needs a rework, it is the mail  >architecture that does. u >nM >For instance, ability to store content size in the message envelope (so that A >IMAP would be MUCH MUCH faster and consume far fewer resources).l >oI >For instance, ability to store copies of messages you send in a specifice6 >folder instead of being sent to you as a new message. > O >Ability to create a message in a specific folder and not send it. (ok, perhapsnM >this is the same as ability to store a copy of outbound messages in "OUTBOX"i >or some other folder).u >nL >Ability to store the message date other than date VMSmail received it. (eg:7 >the date the message was sent, not the date received).f >eN >Field for message read receipt flag (no read receipt, read receipt requested, >read-receipt sent)e >A( >Better integration of RFC mail headers.    Thanks for that clarification.     -- Alans    O ===============================================================================s0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056 M  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025 O ===============================================================================s   ------------------------------  % Date: Tue, 11 Feb 2003 10:56:16 +1100S- From: "Paul Nankervis" <paulnank@hotmail.com>l6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 311 Message-ID: <b29e4e$mqq$1@ausnews.austin.ibm.com>u  H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:18aQYkTO$2CB@eisner.encompasserve.org... ? > In article <cf15391e.0302090938.524c3a46@posting.google.com>,d3 keithparris_NOSPAM@yahoo.com (Keith Parris) writes:. > >JJ > > During the last port, the goal was to get VMS to run on Alpha.  In theI > > present porting effort, the team started with the assumption that VMSsG > > will be around for decades more, and that this port will NOT be the + > > last port of VMS to a new architecture.h >rJ >    Which means that HP (or maybe Compaq) learned what K.O. never figured: >    out.  VAXes sold because they had damn good software. >eE >    If K.O. had known he was running a software company instead of aeH >    computer manufacturing firm maybe he would have done the 80386 port3 >    and DEC would still be number 2 in the market.o  L Yes, I agree completely.  Around ten years ago I exchanged emails with a DECI employee who still actively contributes to this forum. He argued that thei officialF line was that DEC ONLY supported VMS because it sold VAXes and Alphas.G The thinking was that DEC was a hardware company which sometimes had torL write software in order to move product. Basically this was used as the main: argument against doing a VMS port to *ANY* other platform.   I wonder what he thinks now?  F My position was, and still is, that the product I want is VMS. I don't actuallyG care what hardware is underneath provided it has good price/performancem7 (oh - and something about application availability!!!).cH I would still like to see an I386 port as that is the most common systemJ I have access to. But I doubt it will ever happen. (What about I386/VMS as& an unsupported hobbyist release? :-) )  G I know that many people in this group see Alpha as a differentiator and4
 lament itsI demise. I think that it was a good product but after many years it didn'ti manageI to take off even for the more popular Windows or Unix platforms. It badlyo neededL to be picked up by other vendors (eg Apple) to get adequate volumes to match, the development effort put into other chips.  L I saw an article the other day which suggested Pixar are going to move theirH movie rendering from SUN to Intel based systems. This demonstrated to meF that even SUN are having trouble keeping up with the price/performanceK curve being set by Intel. Looks pretty certain that the number of competing.H architectures is going to reduce over the next few years. We will see...   Paul N.x   ------------------------------  % Date: Mon, 10 Feb 2003 20:07:08 -0500u* From: "Bill Todd" <billtodd@metrocast.net>6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 312 Message-ID: <bb-dndrMsNtV0dWjXTWcqg@metrocast.net>  L "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote in message( news:3e47e1e2$1_2@hpb10302.boi.hp.com...   ....  G > VMS ran for the last decade on the fastest, most poweerful systems onD earth.I > It was not a magic bullet.  It wasn't enough ON IT'S OWN to reverse then > fortunes of VMS.  K Exactly.  With zero support from its owner, but running on industry-leadingu1 hardware, VMS just about managed to hold its own.c  ;   HP-classic *isn't* responsible for any of the things that 3 > might have been done a decade ago to change that.p  J HP-classic isn't the issue:  remember, they're HP*Q* now, and the 'Q' partE retains a *great deal* of responsibility for what's happened (and notrJ happened) since Pfeiffer (and his efforts to promote Alpha and the systems on it) got the boot.     They have inherited VMS.I > What might suprise you is that HP-classic actually has a lot of respecth for  > VMS.  : That respect, plus about $1, will get you a cup of coffee.   >  They want it to succeed.   K So what?  *You* want it to succeed too, but, frankly, what you're providingmI to it, important as it is, won't be sufficient, any more than it has beeni for the past decade.  D VMS won't succeed unless HP as a corporation is willing to *make* itE succeed.  That takes more development effort, and *dramatically* moreE
 marketing.  L Not only are such requirements lacking at present, but there's absolutely no sign that they'll appear.s  +   They are *not* going to revive Alpha just-H > for VMS however, and they aren't going to sponsor 2 propreitary UNIXes (andJ > which one has the big market share and profitability?).  And*if* VMS can? > deliver on it's commitments, then success will drive success.l  L Oh, bullshit.  You've already admitted above that nothing *VMS itself* couldI do over the past decade was sufficient for it to succeed:  why in Hell dow) you think that will magically change now?.     Money and.J > attention flow to the groups that are making money, and delivering their > promises.1  L Yeah, right - just like they did in Compaq (I distinctly remember remarkably/ similar words from Rich Marcello 2+ years ago).r   > L > Despite doom and gloom predictions, Itanium2 is a respectable chip, and weE > are building powerful, mission critical server systems around them.k  G Which you did around Alphas as well, and, as you noted above, it wasn't  sufficient.n   ...m  4 > Frankly, the glass *is* half full, not half empty.  I Wrong.  Before the Alphacide, it was half full (or half empty - whicheverzI you prefer).  The Alphacide both sloshed a lot of water out and created a.% continuing leak threatening the rest.      You can help to continue7 > to fill the glass, or throw rocks and hope it breaks.g  H *Nobody's* filling the glass, and the leakage caused by the Alphacide isL continuing.  VMS is dying, Fred:  that may or may not have been the case twoK years ago, but now it's crystal-clear.  Even as of a year ago revenues wereoH reported to be down by 50% (and don't blame the general down-turn:  theyJ were holding up very well compared with the rest of Compaq until Alpha gotE the axe), which suggests that sales are down a great deal more (sinceaC revenues include on-going service agreements for existing systems).i  G VMS isn't going to survive as anything but a stagnant, 'legacy' product K unless it gets a major infusion of corporate commitment soon.  Appearing onfH Itanic - even if Itanic is a success - isn't going to do it.  That's whyL people are throwing rocks:  they're aiming at HP, not at VMS, because unlessL they can get HP to do something it appears to have absolutely no inclinationG to do on its own, it really won't matter whether VMS might get hit by ac stray shot.r  L Itanic's intrinsic worth or lack thereof is mostly a side-show, but it's oneJ aspect of the betrayal that VMS and its customers suffered at the hands ofH the 'Q' portion of HPQ and thus is fair game.  If you don't care for theJ continuing criticism (in this and other areas), get HP to provide a *real*G 'magic bullet', in the form of aggressive development and marketing, todK compensate for the very real bullet that Compaq put into VMS when Alpha was  killed.    - bill   ------------------------------  % Date: Tue, 11 Feb 2003 08:35:17 -0500 A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com>e6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 31. Message-ID: <3e48fc16$1_1@hpb10302.boi.hp.com>  < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3E486FE5.BAA2098A@fsi.net...l > "David J. Dachtera" wrote: > >  > > Fred Kleinsorge wrote:
 > > [snip] > > > You can help to continue > > > to fill the glass, > >sC > > Really? To whom should I send my air-ready commercials for bothrJ > > television and radio? ... my camera-ready ad copy for the print media? >nE > Better still: I'll do the production and book the ad-space/air-time I > myself. That way, at least I can be certain that it will happen and notl- > get lost in the VMS advertising black-hole.  >rH > Just tell me where to have the invoices sent! If I don't hear from youI > by this time next week (17-Feb-2003), I'll just have 'em send the bills  > to Mark Gorham.s > 7 > (Lesson: If you want something done, do it yourself!)- >-E > DJE Systems just appointed itself the marketing arm of HP's OpenVMSl > division!s >e  5 I think your currently being paid what you are worth.g   ------------------------------  % Date: Mon, 10 Feb 2003 20:32:42 -0600c1 From: "David J. Dachtera" <djesys.nospam@fsi.net>W6 Subject: Re: OpenVMS Boots on Itanium on Friday Jan 31' Message-ID: <3E4860CA.BC103498@fsi.net>s   Fred Kleinsorge wrote: > > > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message# > news:3E4550BE.8B241137@fsi.net...r > > David Svensson wrote:m > > > [snip]M > > > My overall point is that you could at least be a bit positive once in ai > while./ > > > Is really everything negative these days?d > >pL > > No, but sometimes the glass *IS* half-empty, and insisting that it isn't5 > > leads one into the realms of denial and delusion.n > >  > M > Sometimes it is.  Nonetheless, urinating in the glass to fill it up doesn'tr > help.D  & You're advising yourself, there, Fred.   > Reality check: > N > Alpha development ends with EV79.  NOTHING will stop that inevitable action.H > No amount of complaining and wishful thinking.  There is NO fincancial, > incentive for anyone to do anything else.   G Well, actually there is, just no one at HP/Q has the balls to pursue itl? since you'd have to go up against Godfather Gates and his boys.r    > Tru64 functionality will mergeN > with HP-UX, VMS will port to IA64, and the HP-classic portion of the companyA > was and remains fully committed to Itanium as the future serverc > architecture.e  D I'll be sure that finds its way onto HP/Q's headstone when(if?) IA64 bites the dust.O  N > VMS ran for the last decade on the fastest, most poweerful systems on earth.I > It was not a magic bullet.  It wasn't enough ON IT'S OWN to reverse the  > fortunes of VMS. s  H Very true. It took years of neglect and intentional suppression (the VMSF world knows it as "stealth marketing") to stuff VMS down the toilet so royally and properly.   F My hat is off to the perpetrators - BRAVO!!! Couldn't have done betterH myself (especially since I know better than to do anything so ..., well, I'd better shutup about it).  ; > HP-classic *isn't* responsible for any of the things thatMM > might have been done a decade ago to change that.  They have inherited VMS.e  F ...and now faces the choice of doing the right thing or continuing theE sins of the past. I think the jury's decision as to what path they've0 chosen is pretty unequivocal.o  M > What might suprise you is that HP-classic actually has a lot of respect fors! > VMS.  They want it to succeed. d  A ...but aren't about to lift a finger to make that happen. They're2E waiting for the "faerie dust", whatever the hell that might happen to3F be. (Yes, my blood pressure is pushing the high-end of the scale right now!)e  + > They are *not* going to revive Alpha justa > for VMS however,  D (an indefinite stay of execution would suffice, at least until IntelF finally comes up with a workable 64-bit chip (my grand-kids *MAY* liveE to see that day, I certainly won't) or someone else manages to do so)s  < > and they aren't going to sponsor 2 propreitary UNIXes (andJ > which one has the big market share and profitability?).  And*if* VMS can@ > deliver on it's commitments, then success will drive success.   ! Still hoping for "faerie dust"...t   > Money andeJ > attention flow to the groups that are making money, and delivering their > promises.n  E Experience seems to contradict that. VMS makes LOTSA bux, and they do  jack-SHIT with it!  G Prove me wrong: quote me a publication name, date and page number whereaC a VMS ad has appeared in the last six months (which is being prettyhC damned generous, considering); quote me the call sign and city of arG station (radio or TV) that has run a VMS commercial anytime in the lastp six months.   9 Talk is cheap! SHOW ME (what) THE MONEY (was spent on)!!!8  E > Despite doom and gloom predictions, Itanium2 is a respectable chip,-   Time and the market will tell.   > and weL > are building powerful, mission critical server systems around them.  Intel% > has not lessened it's committment. -   ...or improved its results.a  ' > HP hasn't.  Other server vendors like 3 > SGI and Fujitsu haven't.  Windows-64 runs on it.    < Size of the current user base? (Where can I find the data to
 corroborate?)    > HP-UX.  Linux.  And VMSh4 > (which will roll out over the course of the year). > I > x86-64 offers nothing to HP-UX, Tru64, Linux, or VMS users that Itaniume > doesn't offer and more.   F Here again, time and the market will tell. That's not for any of us to say.  0 > It's appeal is to the shrink-wrap IA32/WindowsA > market.  It's odds of success are still fairly slim even there,y  H ...unless, of course, they get a workable, competitive product out thereB before Intel does. Then, it's, "Hello AMD", and "Bye-bye Intel and HP/Q"!!!  
 > they areI > chasing a small segment of the market with a product that doesn't offer.K > system makers a significant advantage over continuing to use IA32 for thee# > volume and low-end server market,t  H ...except, of course, the ability to upgrade to 64-bit when the software becomes available, ...  + > and non-x86 systems for the mid range and ! > mission critical server market.   5 > Frankly, the glass *is* half full, not half empty. t  0 Rather depends on your paradigm, now doesn't it?   > You can help to continue > to fill the glass,  ? Really? To whom should I send my air-ready commercials for bothIF television and radio? ... my camera-ready ad copy for the print media?  D Hurry please! Post that info before another 24-hours goes by! Time's a-wastin'!!!  $ > or throw rocks and hope it breaks.  E No, thanks! HP/Q are doing an incomparable and unbeatable job of thatS themselves, already.   -- w David J. Dachteraa dba DJE Systemsa http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Mon, 10 Feb 2003 10:09:33 -05000& From: jlsue <jlsuexxxz@screaminet.com>Y Subject: Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha Retaiw8 Message-ID: <4udf4vgeitf0q62eanviqq33srb8n6seu0@4ax.com>  E On Mon, 10 Feb 2003 13:10:28 +0000, Andrew Harrison SUNUK Consultancyr. <Andrew_No.Harrison_No@nospamn.sun.com> wrote:   >i > 
 >jlsue wrote: H >> On Tue, 04 Feb 2003 11:05:56 +0000, Andrew Harrison SUNUK Consultancy1 >> <Andrew_No.Harrison_No@nospamn.sun.com> wrote:h >> n >> t >>>  >>>jlsue wrote:  >>>rI >>>>On Mon, 03 Feb 2003 15:23:21 +0000, Andrew Harrison SUNUK Consultancys2 >>>><Andrew_No.Harrison_No@nospamn.sun.com> wrote:= >>>You havn't looked hard enough, there were a number on this 7 >>>group. My favourite response being Rob's who pitchedh9 >>>Marvel as the solution to the problem, this was a year 
 >>>or so ago.a >>   >> eE >> That's not true.  I looked all over google, and even put a note insG >> here about it challenging you before.  There was only ONE entry that H >> even came  close and, as I said earlier - now for the third time - it< >> was not clear at all why this person's system was slower. >>   >b@ >Well to start off there is Compaqs own benchmarks. The Kingston< >apps benchmark you published showed that a 16 way GS160 was= >27% faster than a 12 way GS140, you don't need to be a mathst+ >grad to work out that this was unexpected.i >h
 >Or how aboutr >r >http://groups.google.com/groups?q=GS160+group:comp.os.vms&start=10&hl=en&lr=&ie=UTF-8&selm=3D35AD137AAAD411A6BA0008C7B1B12D6DBDDE%40MBCALBEXC03.BENDER.COM&rnum=11a >t< >Note how they solved their performance problems on a GS160.  F Thank you.  It's only been about 2 years worth of hounding to get this3 kind of info out of you to support your statements.    > ; >This was the 11th URL returned just searching for GS160 in ; >the comp.os.vms archives. I can provide others it you want&< >but I would sugges that you stop BSing about having trawled >through google.  C Well, to be fair, in previous discussions I told you what my searchcA included and what I was able to find; and you failed to respond -tD well, actually, your response was always something like "it's there, find it yourself".  F Also, for clarification, I did the search for "GS160" and came up withC 353 returned messages - though many of the messages are part of thesF same threads, and strangely, many of them have you as the author.  ButA am I to assume that you expect the rest of us to read through 353w& notes to find support for your claims?  F Your implication above that it is easy to find one making the claim of8 performance slowdowns going to GS160 is still incorrect.  E Oh, and the example you did pull up ends with the author stating thatsE they were very pleased with their system.  In fact, in another threadoB from Nov 2000, he relates this somewhat, and mentions the Sun woesF that their "corporate office" has - though and NDA prevented them from discussing those woes.   >> sF >> Now this is typical.  You make some attack, and then leave it up toC >> the rest of us to find the proof.  If you can't back up your ownoE >> claims, then stop making them.  It is not our job to validate your  >> statements. >bB >I will do you a deal, the guy who founded the company is a friendD >and I don't want to embarass him, the GS320's were never a sensible= >choice. So I will email you the name of the company, you canp= >check with your local account rep and then once you have youa> >can eat dirt in public without mentioning name of the company >who had to make the switch. >- >Is that a deal ?-  E Not really.  I don't doubt it's possible, but it's also possible thatUF the GS320 could perform very well if configured/managed correctly - we5 have nothing to base a conclusion on in this example.K  D Oh, and as has been stated many times: one or two instances does notD prove the rule (and the one from c.o-v is not necessarily the winnerF you claim it is).  I know *I've* said this to you many times (e.g., in= VMScluster discussions).  We can all pick-and-choose our dataS> collection points to prove what we want.  But real, meaningfulB conclusions can only be drawn from a full analysis.  All your dataC points prove is that *some* applications may not benefit - so what? " Did anyone ever claim differently?  F So to claim that performance is worse on the GS160 than the GS140/84002 (i.e., stating it as a general rule) is still FUD.   ------------------------------  % Date: Mon, 10 Feb 2003 18:33:01 -0000e/ From: "Matt Tyler" <matt_tyler@cableinet.co.uk>m) Subject: Read/Write from a global sectione9 Message-ID: <BdS1a.23$uV5.6@news-binary.blueyonder.co.uk>   K I am having problems writing /reading from a VAX VMS global section. I haveyL posted two small Pascal test programs to write/read to/from element one of aL ten element integer array. I have checked previous posts but cannot work outF why I am only reading zeros from the global section. Any help would be appreciated.   Cheers   Matt  " {****************WRITE***********}   inherit ('SYS$LIBRARY:STARLET',s-           'SYS$LIBRARY:PASCAL$LIB_ROUTINES')]o  $ PROGRAM SHARED_WRITE(INPUT, OUTPUT); TYPE    int_pointer = ^INTEGER; VARl>   GS_ARRAY : [VOLATILE, ALIGNED(13)] ARRAY [1..10] OF INTEGER;)   IN_ADDR  : ARRAY [1..2] OF int_pointer;h   err      : INTEGER;    i_count  : INTEGER;o  " [ASYNCHRONOUS,EXTERNAL (LIB$WAIT)]& FUNCTION lib$wait(seconds    : SINGLE;4                   flags      : UNSIGNED := %IMMED 0;I                   float_type : UNSIGNED := %IMMED 0) : INTEGER; EXTERNAL;-   BEGIN-&   IN_ADDR[1] := ADDRESS (GS_ARRAY[1]);'   IN_ADDR[2] := ADDRESS (GS_ARRAY[10]);U  #   err := $CRMPSC(inadr  := IN_ADDR,BB                  flags  := SEC$M_GBL + SEC$M_DZRO + SEC$M_EXPREG +
 SEC$M_PAGFIL, &                  gsdnam := 'SHARED_X',                  chan   := 0,e                  pagcnt := 1);  6   IF (err <> SS$_CREATED) AND (err <> SS$_NORMAL) THEN   BEGIN %      WRITELN(' Unable to map to GS');e      LIB$STOP(err);@   ENDe   ELSE   BEGIN.      FOR i_count := 1 to 60 DO
      BEGIN7         WRITELN('GS_ARRAY now equals : ', GS_ARRAY[1]);          lib$wait(5.0);	      END;A   END; END.     {**************READ***********}>    [inherit ('SYS$LIBRARY:STARLET',-           'SYS$LIBRARY:PASCAL$LIB_ROUTINES')]g  $ PROGRAM SHARED_WRITE(INPUT, OUTPUT); TYPE    int_pointer = ^INTEGER; VARp>   GS_ARRAY : [VOLATILE, ALIGNED(13)] ARRAY [1..10] OF INTEGER;)   IN_ADDR  : ARRAY [1..2] OF int_pointer;A   err      : INTEGER;c   i_count  : INTEGER;t  " [ASYNCHRONOUS,EXTERNAL (LIB$WAIT)]& FUNCTION lib$wait(seconds    : SINGLE;4                   flags      : UNSIGNED := %IMMED 0;I                   float_type : UNSIGNED := %IMMED 0) : INTEGER; EXTERNAL;t   BEGINe&   IN_ADDR[1] := ADDRESS (GS_ARRAY[1]);'   IN_ADDR[2] := ADDRESS (GS_ARRAY[10]);r  #   err := $CRMPSC(inadr  := IN_ADDR,r2                  flags  := SEC$M_GBL + SEC$M_WRT +D                            SEC$M_DZRO + SEC$M_EXPREG + SEC$M_PAGFIL,&                  gsdnam := 'SHARED_X',                  chan   := 0,                   pagcnt := 1);  6   IF (err <> SS$_CREATED) AND (err <> SS$_NORMAL) THEN   BEGINe/      WRITELN(' Error creating /mapping to GS');h      LIB$STOP(err);    END!   ELSE   BEGIN       FOR i_count := 1 to 10 DO
      BEGIN         GS_ARRAY[1] := i_count;3:         WRITELN('GS_ARRAY[1] now equals : ', GS_ARRAY[1]);         lib$wait(30.0);V	      END;    END; END.   ------------------------------  # Date: Tue, 11 Feb 2003 07:41:54 GMTg4 From: Brian Inglis <Brian.Inglis@SystematicSw.ab.ca>J Subject: Re: RP06's and Field Service calls -- was Very large disks on VMS8 Message-ID: <e39h4vg8cjpnpcdkdavlnaofdufrf0giqg@4ax.com>  
 In articleI <BE56C50EA024184DAF48F0B9A47F5CF402660D54@kaoexc01.americas.cpqcorp.net>, & Main, Kerry <Kerry.Main@hp.com> wrote:  H >Heartbreak is being a FE on call on Friday night and hoping it is goingG >to be a quiet weekend as you have lots of stuff to do and then gettingoC >the following call from some mission critical customers operator -nH >"well, there was a strange noise coming out of this one RP0x, drive, soI >I moved the pack to each of the other drives we had and now they are alll >making noises .. Any idea's?" >tG >Weekend is now shot and around the clock job starts to replace all thecI >heads on all the Customers drives where that Operator moved the bad pack-A >to. First task is to stop and pick up baseball bat as it will bef1 >involved in the first task when you get on site.   ; Heartbreak is being a couple of sys progs who get RM05 pack-? errors, can smell a head crashed on one of the drives, and calld Field Circus. : Guy comes in, we describe the problem, he says he'll run aA diagnostic, goes and gets the MAINDEC pack, and walks over to theS crashed drive. lA Major stereo protests, but arrogant a**hole knows better, ignoresl= imprecations of "don't think you should put that pack in that @ drive" (and doesn't pick up on our reasonable tones, a sure sign? that a sys prog is waiting to see a spectacular failure), loadsi  up the pack, and can't boot it. A To our shock, horror, amazement, and amusement, he then tries theo> pack in four more drives (our nostrils are just quivering with; the heady smell of aluminum and ferric oxide vapours), thent7 finally he gives up, and calls in the disk specialist. p  ? Assessment: need five drives worth of twenty heads. Nearest setb= of heads: three hours away with the RM05 kit in another town. < Next nearest set of heads: country level stores. More heads:? somewhere in Massachusetts. Took three days, many plane flightso9 and cab trips, and many hours of FS guys time to deliver,.. replace, and align the hundred heads toasted.   > Can't even remember who the guy was now, but IIRC, our manager= told the FS manager he did not ever want that guy on our sitet( again, so probably never saw him again.   9 Thanks. Take care, Brian Inglis 	Calgary, Alberta, Canada0 -- rF Brian.Inglis@CSi.com 	(Brian dot Inglis at SystematicSw dot ab dot ca),     fake address		use address above to reply@ abuse@aol.com tosspam@aol.com abuse@att.com abuse@earthlink.com ? abuse@hotmail.com abuse@mci.com abuse@msn.com abuse@sprint.com oB abuse@yahoo.com abuse@cadvision.com abuse@shaw.ca abuse@telus.com - abuse@ibsystems.com uce@ftc.gov				spam traps    ------------------------------  # Date: Tue, 11 Feb 2003 13:52:57 GMTmF From: lederman@star.enet.dec.DISABLE-JUNK-EMAIL.com (Bart Z. Lederman)J Subject: Re: RP06's and Field Service calls -- was Very large disks on VMS/ Message-ID: <Ze72a.855$%W.408@news.cpqcorp.net>i  g In article <b29obk$jhq$1@shell.monmouth.com>, pechter@shell.monmouth.com (Bill/Carolyn Pechter) writes:n  F >They were refurbing the air conditioning and painting again (one timeE >they painted the ceiling by standing on the spinning RP06 drive topsl? >(which probably would've caused loss of limbs if the glass hade >broken...).  > I ran into something like that at my first job.  The PDP-11/709 with an RP04 as the main disk that served the engineeringw' department suddenly stopped responding.   9 When I got to the computer room a new maintenance man was 9 standing on the drive to reach the ceiling tiles.  He hadb8 pressed the button on the front of the lid, so it opened7 about an inch (there was some sort of safety latch that@4 prevented it from opening all the way when the drive was still spinning).  7 When I explained to him how much a disk drive cost, and 1 that he would be responsible for it, his attitudee changed quickly.  7 We were lucky, the drive was o.k. and after closing the 5 drive cover the system picked up where it left off (It always liked RSX-11M).  ; This was rather an unusual occurrence.  The people who weret9 allowed into our operating area were normally much betters: trained and supervised.  In all of the time I worked there: I can't recall anything even remotely resembling this ever: happening.  And this person would never have been let near/ any of our on-line systems without supervision.x   -- i(  B. Z. Lederman   Personal Opinions Only  8  Posting to a News group does NOT give anyone permission8  to send me advertising by E-mail or put me on a mailing  list of any kind.  5  Please remove the "DISABLE-JUNK-EMAIL" if you have ah5  legitimate reason to E-mail a response to this post.    ------------------------------  % Date: Tue, 11 Feb 2003 07:09:29 -0700r% From: Dan O'Reilly <dano@process.com> J Subject: Re: RP06's and Field Service calls -- was Very large disks on VMSB Message-ID: <5.1.0.14.2.20030211070651.01c33a18@raptor.psccos.com>  . At 06:52 AM 2/11/2003, Bart Z. Lederman wrote:J >In article <b29obk$jhq$1@shell.monmouth.com>, pechter@shell.monmouth.com  >(Bill/Carolyn Pechter) writes:c >dH > >They were refurbing the air conditioning and painting again (one timeG > >they painted the ceiling by standing on the spinning RP06 drive topshA > >(which probably would've caused loss of limbs if the glass hade > >broken...). >a? >I ran into something like that at my first job.  The PDP-11/70 : >with an RP04 as the main disk that served the engineering( >department suddenly stopped responding. >7: >When I got to the computer room a new maintenance man was: >standing on the drive to reach the ceiling tiles.  He had9 >pressed the button on the front of the lid, so it openeds8 >about an inch (there was some sort of safety latch that5 >prevented it from opening all the way when the drive  >was still spinning).m > 8 >When I explained to him how much a disk drive cost, and2 >that he would be responsible for it, his attitude >changed quickly.h  E I has the same sort of experience with some RM03 drives when I worked'E for DEC.  A system I was running suddenly took a whole slew of errorshE on 2 of 3 RM03's.  Quickly making my way down to the computer room, I E discovered a workman tugging cables thru an overhead cable tray, withcG 1 foot on a ladder and the other on a disk drive.  Every time he made aoA sharp jerk on the cable bundle, the drive took a burst of errors.r  @ Why the things never crashed, I don't know, but they didn't...~~   ------J +-------------------------------+----------------------------------------+J | Dan O'Reilly                  |  "There are 10 types of people in this |J | Principal Engineer            |   world: those who understand binary   |J | Process Software              |   and those who don't."                |J | http://www.process.com        |                                        |J +-------------------------------+----------------------------------------+   ------------------------------   Date: 11 Feb 2003 17:07:51 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)J Subject: Re: RP06's and Field Service calls -- was Very large disks on VMS6 Message-ID: <b2bal7$18v6m5$1@ID-135708.news.dfncis.de>  T In article <BE56C50EA024184DAF48F0B9A47F5CF4023D9C95@kaoexc01.americas.cpqcorp.net>,* 	"Main, Kerry" <Kerry.Main@hp.com> writes: >  > ? >>>> "Son, you're going to learn how to replace heads on a diskn > drive..."<<< > H > Mmmm.. I wonder how many FE's today would even know what a scope was - > let alone know how to use it.E >   A A bit ago I picked up some real nice PDP-11's from a school in NY-> that was moving to DELL's.  Among the spares was a pretty nice@ scope.  I really wanted it but at the last minute they said "No,> we're going to keep that."  I couldn't imagine they had anyone@ left there who would know what to do with it and anyway, what do= you use a scope for when your servers are made by DELL? I wasvC disappointed, but after all, it was their stuff so I reigned myself6A to just be happy with what I got.  Imagine my surprise when I got B back to the office, started unpacking and found all the probes for@ the scope.  They have never contacted me to ask if I had them soA I can only imagine the scope is still sitting onthe shelf where Iu left it.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   s   ------------------------------  % Date: Tue, 11 Feb 2003 10:50:30 -0800e$ From: Shane Smith <ssmith@icius.com>J Subject: RE: RP06's and Field Service calls -- was Very large disks on VMS0 Message-ID: <01C2D1BB.668C8590@sulfer.icius.com>  E Back in my student days the IT department got some new shelves put inFH the computer room. Soon as time came for backups, all the removable-packE drives went boom. After a while someone thought to ask the workman he@H did. Pointing to the drives, he said "Well, I used those cabinets to saw the wood on...."   Shane    ------------------------------  % Date: Tue, 11 Feb 2003 19:29:32 +0100o> From: "Gerhard Schweinschwaller" <gerhard@schweinschwaller.at>Y Subject: Re: Smart Array 5300A, Storageworks 4300 Ultra Tower, OpenVMS 7.3-1, and an ES40 . Message-ID: <1044987545.451021@news.liwest.at>  , i use a sa5300 on a ds25 since december 2002  7 works perfect .... high performance .... no problem....n  I only the documentation about installing the management agents is a little ! bit bugy ... some wrong filenamest   Gerhard Schweinschwaller, OpenVMS System Manager since 1985 in Austria    : "DigiDemon" <digidemon@hotmail.com> schrieb im Newsbeitrag2 news:pan.2003.02.10.18.54.02.213161@hotmail.com...A > Going to redo this beast next month I think....any surprises ori7 > experiences that anyone would like to share?  Thanks!i >( > Jamesp   ------------------------------  % Date: Tue, 11 Feb 2003 06:28:03 -0800U, From: "James Gessling" <jgessling@yahoo.com>< Subject: Re: Support for Gigabit Ethernet Clusters on Marvel5 Message-ID: <b2b19l$1b44vc$1@ID-46415.news.dfncis.de>c  > Does this mean LAN fastpath support too?  That would be great.   Jim   > "Sue Skonetski" <susan_skonetski@hotmail.com> wrote in message7 news:857e9e41.0302101426.4a3d3709@posting.google.com...  > Dear Newsgroup,= >=H > I just sent this out, please be aware that it may take a little bit of4 > time for all the local offices to hear about this. >i# > The following WEB SITE is public.C > E >  Part number: 3X-DEGXA-TA (Copper)    - SA (Fiber)   Orderable now.  >dF >  Full Support statements in a month or so.  It'll required 7.3-1 and > the LAN ECO kit..= >= > Warm regards,  > Sue-   ------------------------------    Date: 10 Feb 2003 23:37:19 -0800, From: mcbill20@hotmail.com (Bill McLaughlin)Y Subject: Re: Sysgen and UAF memory limit recommendations for Oracle on small machine (256u= Message-ID: <e9cbc4f2.0302102337.1499a840@posting.google.com>s   "Main, Kerry" <Kerry.Main@hp.com> wrote in message news:<BE56C50EA024184DAF48F0B9A47F5CF402660D5D@kaoexc01.americas.cpqcorp.net>...o > Bill,t > I > As a fyi, I would suggest upgrading to VMS V7.3-1 as I suspect you willt5 > find the overall performance much better than V7.3.   E When I can get a later CD I'll do that. The 7.3 was the latest that I  could dig up right now.d   > J > Having stated that 256MB is not a lot of memory for a server these days. >  > I > Heck, most desktop PC's require 256MB to run latest Windows XP softwareg2 > with a number of standard desktop applications.   E True, but thanks to Digital's usurious pricing for things like memoryyC (5-10x going market rate), upgrading is not an option. This machinedD started with 128MB and I paid $300 for used memory to get it to 256.B And, the $300 that I paid was a little more than 1/2 of the normal# price for used memory for this box.t   > C > However, Oracle 9.2.0.2 (9i R2) for OpenVMS is also available foraC > download off the standard Oracle download OTN site and from a fewiD > reports, is much better than the original 9I release that came outF > (9.0.1.x) awhile ago (on all platforms). Oracle 9.2.0.2 is certified > with OpenVMS V7.3-1 as well. >   = Actually, 9.2.0.2 is the version I am working on installing. t  I > There are also some very recent articles in Oracle's Metalink (databasee> > of known issues and problems and user forums) online supportH > environment. A number of these outline recommended process quota's for > Oracle 9i R@ on OpenVMS. > I > In the troubleshooting side, there are a number of options available tohJ > monitor process quota's in real time to see what quota is being used up.F > One is Availability Manager (free download off OpenVMS site) and theD > others include freeware process monitoring dcl procedures that are	 > around.d >   B Yeah, maybe I should take a look at that. I actually considered itA before but figured the last thing I needed to do was take up more  memory. I'll check it out.   >  > 	 > Regards  >  > Kerry Main > Senior Consultantn > Hewlett-Packard (Canada) Co.# > Consulting & Integration Services  > Voice: 613-592-4660N > Fax   : 613-591-4477 > Email: kerryDOTmain@hpDOTcom/ >     (remove the DOT's and replace with "."'s)i! > OpenVMS DCL - the original .COMI >   D As always, thanks for your help. I definitely appreciate this forum.   Bill   >  > -----Original Message-----6 > From: Bill McLaughlin [mailto:mcbill20@hotmail.com] ! > Sent: February 10, 2003 1:46 PM  > To: Info-VAX@Mvb.Saic.ComlJ > Subject: Sysgen and UAF memory limit recommendations for Oracle on small > machine (256MB)n >  > H > Is anyone out there running Oracle 9i on an Alpha with only 256MB? TheH > release notes state that this is the absolute minimum for a machine toB > run 9i. Now that I have Java updated and the appropriate patchesI > installed, the Java installation procedure runs OK. However, it gets toTH > about 80% and then hangs the machine (after about 5 hours). The secondJ > time I was able to catch it before the point of no return and discoveredI > the three Oracle install processes were in RWMPB state. I quickly added J > a new paging file and was able to keep the system from hanging. However,A > the Oracle installation immediately died with no error message.t > D > I have been trying to adjust the working set parameters and sysgemG > parameters but this is very time consuming as the failures occur fives& > hours or more into the installation. > F > If anyone out there has successfully installed 9i on a small machineD > like this, I'd appreciate hearing how you did it (page file quota, > wsmax, etc.) > G > The machine has two paging files of approximately 500,000 blocks each  > and is running VMS 7.3.r > 	 > Thanks., > Bill McLaughlin,   ------------------------------  % Date: Tue, 11 Feb 2003 10:42:39 -0800M$ From: Shane Smith <ssmith@icius.com>Y Subject: RE: Sysgen and UAF memory limit recommendations for Oracle on small machine (256l0 Message-ID: <01C2D1BA.4E65C590@sulfer.icius.com>  E Go have a chat with Dave Turner at Island. He has a good supplier fore memory, he'll sort you out.g   Shanee   -----Original Message-----8 From: mcbill20@hotmail.com [mailto:mcbill20@hotmail.com]( Sent: Monday, February 10, 2003 11:37 PM To: Info-VAX@Mvb.Saic.ComiF Subject: Re: Sysgen and UAF memory limit recommendations for Oracle on small machine (256MB)Y    2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageH news:<BE56C50EA024184DAF48F0B9A47F5CF402660D5D@kaoexc01.americas.cpqcorp net>...o > Bill,h > I > As a fyi, I would suggest upgrading to VMS V7.3-1 as I suspect you willt5 > find the overall performance much better than V7.3.1  E When I can get a later CD I'll do that. The 7.3 was the latest that Ia could dig up right now.m   > J > Having stated that 256MB is not a lot of memory for a server these days. >  > I > Heck, most desktop PC's require 256MB to run latest Windows XP softwares2 > with a number of standard desktop applications.   E True, but thanks to Digital's usurious pricing for things like memoryrC (5-10x going market rate), upgrading is not an option. This machineeD started with 128MB and I paid $300 for used memory to get it to 256.B And, the $300 that I paid was a little more than 1/2 of the normal# price for used memory for this box.  <snip>   ------------------------------  % Date: Tue, 11 Feb 2003 06:27:45 -0400g0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>! Subject: TCPIP: How to multihomedi. Message-ID: <3E48D021.BA96528@vl.videotron.ca>  G Ok, I would like to have one VAX (TCPIP Services 5.3) appear to be 2 IPy( adresses on the same ethernet interface.  5 I have the primary address defined and it works fine.e   SE0 is 10.0.0.10 l  I How the ?%&?% do I define a second interface and/or second IP address ?  t  M All the doc says is to look at LIST COMMUNICATIONS_CONTROLLER , but what do Ig do with that output ?y  E Must I DEFINE COMMUNICATION_CONTROLLER to enable some sort of virtualy8 interface that maps to the same ethernet device ( ESA0 )    J My need is as follows: when a web request comes from the outside of my NATJ lan, I want to route it to the second IP address on my vax so that the web+ server will treat that request differently.o   ------------------------------  % Date: Tue, 11 Feb 2003 13:03:10 +0100n$ From: "Peter Flunger" <p-i-b@gmx.at>% Subject: Re: TCPIP: How to multihomedn0 Message-ID: <b2aopv$3sl$1@newsreader1.netway.at>  2 "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wroteI > Ok, I would like to have one VAX (TCPIP Services 5.3) appear to be 2 IPu* > adresses on the same ethernet interface. >t7 > I have the primary address defined and it works fine.v >  > SE0 is 10.0.0.10 >'I > How the ?%&?% do I define a second interface and/or second IP address ?t >vJ > All the doc says is to look at LIST COMMUNICATIONS_CONTROLLER , but what do I > do with that output ?-  > I am not sure about TCP/IP on VAXes, but on an Alpha you would> enter a permanent secondary TCP/IP Address for the same NIC as1 $TCPIP SET CONFIG INTERFACE xxA0: /host=a.b.c.d - .   /net=<your mask>/broad=<your-broadcast-mask>* If your primary interface is called SE0 in $TCPIP SHOW INTERFACEA5 the name for the seondary would be SEA0: in the above  example.: To enable the Interface on te running system, simply enter  $TCPIP SET INTERFACE xxA0: .....   hth , Peterw   ------------------------------  # Date: Mon, 10 Feb 2003 19:46:14 GMTr# From: "John Smith" <a@nonymous.com> $ Subject: Re: The importance of ISV'sG Message-ID: <akT1a.12615$Qf1.9644@news02.bloor.is.net.cable.rogers.com>y  D "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote in0 message news:3e47e7e5$1_2@hpb10302.boi.hp.com... > John,t >aA > Quite often, all that is needed is for the ISV to see the sales  opportunity.  E If I might remind you, it has been approximately 10 years of AffinityoD and other ill-conceived programs and neglect that brought most ISV'sE to the point of not even considering a port of their products to VMS,eD much less continuing support for their existing products on VMS. TheF ones that remain have decent products, but all too often the ones thatC remain are not the ones with 'mind share' with the street/customer.o  F And of those remaining, are they 100% current with VMS and/or are they< at 100% functional parity with their other releases on other
 platforms?  F Most *new* or former ISV's will want to see some solid commitment fromE HP with some other high profile ISV's who have recently been 'broughteE on-board' before they get involved. And that commitment from HP wouldn: also include visible advertising and marketing of the o/s.      E > I'm not going to follow up on this, because I suspect that there isf noB > specific business here.  Your opinion of the 3rd party reporting	 tool also  > seems low.  F Believe what you want Fred. It's true that the customer doesn't have aC blank check made out in the name of Fred Kleinsorge or to HP at thetE moment, but they do have a pressing business problem. In the end, forkF a variety of reasons, it may come to pass that there is no opportunity; here, but in the interim we will continue to suggest to ourl< customer/ISV what we believe are good strategies to pursue -@ including, but not limited to, a possible port to Tru64/Linux orE OpenVMS on Alpha for this aspect of their business. Fortunately I can D insulate the customer and ISV from comments like yours, but even oldD jaded Fred would be surprised at the size of the customer's business' and the sophistication of the ISV tool.M   ------------------------------  # Date: Mon, 10 Feb 2003 21:12:27 GMT.# From: "John Smith" <a@nonymous.com>s$ Subject: Re: The importance of ISV'sJ Message-ID: <%AU1a.314136$pDv.312611@news04.bloor.is.net.cable.rogers.com>  > "Malcolm Dunnett" <nothome@spammers.are.scum> wrote in message' news:xArFg99TmQoF@malvm7.mala.bc.ca....  > In article> <HCX0a.294684$pDv.93953@news04.bloor.is.net.cable.rogers.com>,* >    "John Smith" <a@nonymous.com> writes: >t > >oD > > I appreciate the offer Fred, but there's really nothing that can beC > > done. The tool is a widely used enterprise-class reporting toole from a > > company on the west coast. >hB >     Was any thought given to the idea of running the database onC > VMS or Tru64 and running the reporting tool on the other systems?t >hE >     Without knowing a lot about the application it seems to me thate ifE > you had a high performance database at the back end ( assuming heren9 > for arguments sake that VMS/Tru64 on a Marvel fits thatt requirement ), you= > could then have any number of "front end" boxes running the 	 reportingn@ > tool on Windows, Solaris, Linux or whatever and making queries againstt > that database.    F We are looking at all possible solutions that don't break the bank and' permit the customer to keep up with thetC daily/weekly/monthly/quarterly/annual peak processing requirements.kE Currently running Oracle 8i and the reporting data has been partiallya2 pre-summarized to minimize the reporting overhead.   ------------------------------   Date: 11 Feb 2003 00:44 CDTa' From: carl@gerg.tamu.edu (Carl Perkins)e" Subject: Re: VMS source listings ?- Message-ID: <11FEB200300445992@gerg.tamu.edu>a  Z In article <3E47E39A.6095D37F@NelsonUSA.com>, Alan Frisbie <Abuse@NelsonUSA.com> writes...+ }Alan Winston - SSRL Admin Cmptg Mgr wrote:s } - }> >   OBTW:  is it Pee-El-One or Pee-El-Aye?d }> aP }> Pee-El-One.  (The I is Roman numeral 1.  IBM had a great idea about combiningR }> the best of Fortran, Cobol, and Algol to make the first really good Programming- }> Language; thus, Programming Language One.)s } ( }You mean the One-Bee-Em company?    :-) }  }AlanV  & You mean the One-Bee-Thousand company?   --- Carl   ------------------------------   End of INFO-VAX 2003.084 ************************ctions, Itanium2 is a respectable chip,-   Time and the market will tell.   > and weL > are building powerful, mission critical server systems around them.  Intel% > has not lessened it's committment. -   ...or improved its results.a  ' > HP hasn't.  Other server vendors like 3 > SGI and Fujitsu haven't.  Windows-64 runs on it.    < Size of the current user base? (Where can I find the data to
 corroborate?)    > HP-U\kZ"(z	r;eߒKrÙt'샖oN\&m,Iex/RΧǐyw
ԵFJ+_TJ=Jy`GוZ
atT#
J([v?/GӬFZ)oUE:&vjI( ~&lpسoq1[bGAԪo=D3m5UwƘ/l,
Tu-@gF̯>SV-ȪA~Zxtgb,9:R	=iؒ}}ԬJ%]d7&q<5CtQ0ghWYz2
Sdo0%/J~MU~c< I6ٞI?}Xv)NiDM]ɿPK   6(W#.! G!    u-lp0100.exeXM0
v9d$dIKQ@(9(+`BdAD Id	$Hɒâw?uSՕgvu,`0   Ȁv?рD`]8f.^A~~N".9-49\rr^
ഷ
tt$˗33	rwIp9B-Z.H,/8>+PE $}Kdr.OW  ' )IGh1/B;!,>[6$> z-Ah@yPER8
7>i9u~аD$11R9w?	@	õT
uULT
IT()h Ѵ
pAU=,(e/_g8KI]uԱu= sjxڟthyJΓr	@5V᧑51iaŢ,di֋jm9R߮R3O^~k\Z$ܖ,5(Y_'DEa[*2q36[,:ub~|v8ӂjJFzzirMO('	%ف$.'|wEMnj4]C>~EKr}UL<SQæpOzkrpNi޳睧>
A7jUR<GEl7UStX3Z}\F˽D?E얪&r&N0ϴZ9ELiGRuMT2s|IS.,2TEKqnfӎ*f\7[q4VvjBss	O?/}^v
ϯI1ǃ㧞wsi[ޢvǞ<R e:E!Bz]|0{-ܟRGʺj{&J0ißXRXbMwKO7/|iI?/Q,0E*|eÂG:,	l]p|y2Lͭq]ÂO4һO\>".]ɿ]|Ft/GРs9zd>(>8Ў;+YۥK".KgSn:NUͥSZrigOpq 2{|9Z2!?0B?7>o[C<*z<V?D2,tѯFN$<q\!6&jͅX>ܭ̀]w_@{HR3Ӭy0:WHUgvHLbXr7_qTk}%_]kSF%K:Z1i3',Ӧ1owT՜W*	Q%yKIqs	ܹai	o1.N_5~@KFZ).$rٚA.{`[tEKu'/9]:,	Sh'i4&w +T?mK+7_ё.:H!26º]Uab7׆