1 INFO-VAX	Mon, 09 Jun 2003	Volume 2003 : Issue 318       Contents: access violation in SMTP Re: alphaServer diagnostics  Re: alphaServer diagnostics  Re: alphaServer diagnostics  Re: alphaServer diagnostics 1 Re: Another future of VMS article on the Inquirer ; Re: BLASTed directory locks, timing windows & endless loops # Re: CNN Story Not Favorable To Bush # Re: CNN Story Not Favorable To Bush  Re: Corporate takeovers " CSWB 1.2.1 fails to display images& Re: CSWB 1.2.1 fails to display images& Re: CSWB 1.2.1 fails to display images( Re: Decserver 700 software on CD or DDS?( Re: Decserver 700 software on CD or DDS? Dell to make Itanium 2 systemsN Re: ERRLOG.SYS (was Re: Your Own Installation Tool? (was: Install  Directory)) Re: Firewall for VMS / TRU64 Re: Firewall for VMS / TRU64 Re: Firewall for VMS / TRU64 Re: Firewall for VMS / TRU64 Re: Firewall for VMS / TRU64 Gnu make question  Re: Gnu make question * Re: Halt & Startup Issues with VS 4000 VLC* Re: Halt & Startup Issues with VS 4000 VLC* Re: Halt & Startup Issues with VS 4000 VLC/ Re: hanging UCX sessions, expert help required.  Help listing SYSGEN params RE: Help listing SYSGEN params RE: Help listing SYSGEN params Re: Help listing SYSGEN params Re: Help listing SYSGEN params Re: Help listing SYSGEN params/ Help with upgrade to VMS v7.3-1 from VMS v7.2-1 3 RE: Help with upgrade to VMS v7.3-1 from VMS v7.2-1 3 Re: Help with upgrade to VMS v7.3-1 from VMS v7.2-1 # Re: Here's a commercial for you all # Re: Here's a commercial for you all # Re: Here's a commercial for you all # Re: Here's a commercial for you all # Re: Here's a commercial for you all $ How get OS on totally blank machine?( Re: How get OS on totally blank machine?M Re: How to compile/link Fortran for 'portable' executable to down-level OVMS? " Re: HP #1 in Servers and Notebooks/ Re: Intel 64 bit Pentium seems to be on its way / Re: Intel 64 bit Pentium seems to be on its way / Re: Intel 64 bit Pentium seems to be on its way / Re: Intel 64 bit Pentium seems to be on its way / Re: Intel 64 bit Pentium seems to be on its way / Re: Intel 64 bit Pentium seems to be on its way E Re: JF Mezei "flamed, insulted, denigraded(sic)"!  Oh the humanity!!! E Re: JF Mezei "flamed, insulted, denigraded(sic)"!  Oh the humanity!!! C Re: Letter to HP Management from OpenVMS users around the world ... ( Re: Mezeing, and Mezei's severe problems) Re: Motif 1.3, Euro sign not standard ?!? ) Re: Motif 1.3, Euro sign not standard ?!?  Re: MSA1000 vs EVA3000 Re: MSA1000 vs EVA3000 Re: MSA1000 vs EVA3000 Re: MSA1000 vs EVA30008 Re: My IP was listed on SPEWS and I am not a spammer....' Re: OT: CNN Story Not Favorable To Bush ' Re: OT: CNN Story Not Favorable To Bush ' Re: OT: CNN Story Not Favorable To Bush  Portents of Itanium death  Re: Portents of Itanium death  Re: Portents of Itanium death  Re: Portents of Itanium death  Re: Portents of Itanium death  Re: Portents of Itanium death  Re: Portents of Itanium death  Re: Portents of VMS death ? Re: Printer recommendation, hobbyist/decwindows/mozilla and VMS ? Re: Printer recommendation, hobbyist/decwindows/mozilla and VMS ? Re: Printer recommendation, hobbyist/decwindows/mozilla and VMS  Re: Radeon 7500 on a PWS  Re: Running java programs in VMS  Re: Running java programs in VMS& Re: Speed of IDE controllers in DS10 ?& Re: Speed of IDE controllers in DS10 ? Re: timezone rule  Re: timezone rule  Re: timezone rule  Re: timezone rule  Re: VMS SQL client# Re: What do these Macro codes mean? # RE: What do these Macro codes mean? # RE: What do these Macro codes mean? # RE: What do these Macro codes mean? # RE: What do these Macro codes mean? J Re: [DCL] Is the max length of DCL lexical functions argument documented ?  Re: [OT] best HTML code cleaner?  F ----------------------------------------------------------------------  * Date: Mon, 9 Jun 2003 15:51:11 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)! Subject: access violation in SMTP $ Message-ID: <bc2adf$idr$3@online.de>  $ Message from user INTERnet on XXXXXXC INTERnet ACP SMTP Accept Request from Host: a.b.c.d     Port: 64855   & Message from user TCPIP$SMTP on XXXXXXD %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address= 000000006666& 6664, PC=0000000066666664, PS=0000001B  $ Message from user INTERnet on XXXXXXF INTERnet ACP AUXS error during process exit  Status = %SYSTEM-F-ACCVIO  G Has anyone seen the above?  Any idea what caused it?  Any idea where to H start looking?  (I have the SMTP logicals set to be verbose, so I hope IG can find something somewhere.  If anyone knows a quick answer---or, if  3 necessary, a quick fix---I'd appreciate it though.)    5.0A, soon to be upgraded.  :-|    ------------------------------   Date: 9 Jun 03 08:31:22 +0200 ) From: p_sture@elias.decus.ch (Paul Sture) $ Subject: Re: alphaServer diagnostics) Message-ID: <Qoy5v77nWWp+@elias.decus.ch>   f In article <3EE3E84F.85FF0547@firstdbasource.com>, Michael Austin <maustin@firstdbasource.com> writes: > Jim Strehlow wrote:  >>  ? >> What diagnostic tools are available for an AlphaServer DS10?  >>  - freeware download " >>  - on the software tools cd-rom$ >>  - for licensed purchase from HP? >>  A >> We have an OpenVMS AlphaServer DS10 on sofware support but not I >> hardware support which "locks up" the operator's console (keyboard and > >> mouse) running OpenVMS v7.3 patched up through a month ago. >>  + >> Thank you in advance for your responses.  >>   >> Jim Strehlow  >> Data911, Alameda, CA, USA >  > J > Can you give us a better definition of "locks up"  does just the consoleF > hang (serial or graphics?) while the rest of the system continues toJ > function or does the whole system hang (no communications, no processing	 > etc...)  > H > If the system completely freezes including the console (no <ctrl>P, noF > halt button) and the only thing that will clear it is a power cycle,. > then you may be looking at a memory problem. > G > Change the console to SERIAL, put a laptop using a serial Reflections H > session (or any other terminal emulator that will allow you to log theG > output to a file) and power cycle the system several times in a row.  < > See if it is kicking out any memory or CPU related errors. > ' > I don't recall which ones are free.    >     F The advice I got some time ago from the Charon-VAX people was that theD version of Hyperterm supplied with Winwoes is deliberately crippled,> but the good news is that you can get a proper working versionH (Private Edition) for free from http://www.hilgraeve.com/htpe/index.html   --  
 Paul Sture   ------------------------------  # Date: Mon, 09 Jun 2003 13:02:42 GMT 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)$ Subject: Re: alphaServer diagnostics5 Message-ID: <Sz%Ea.112941$lL2.1025583@news.chello.at>   l In article <4b6ec350.0306060859.2d7a24b7@posting.google.com>, JimStrehlow@data911.com (Jim Strehlow) writes:= >What diagnostic tools are available for an AlphaServer DS10?  > - freeware download   D DECevent is freely downloadable, but as the DS10 is EV6 (and up) and: DECevent doesn't support EV6 machines, this seems a No-Go.    > - on the software tools cd-rom  3 WEBES/CA V4.0 is on the 2001 Software Tools CD-Rom. L AFAIK CA V4.1 is on the 2003 Software Tools CD-Rom (from the Mar03 ConDist),L but current is CA V4.1.4 which is not downloadable and AFAIK not on ConDist.  " > - for licensed purchase from HP?  H If you have a support contract, you can ask HP exactly this questions...  ? >We have an OpenVMS AlphaServer DS10 on sofware support but not G >hardware support which "locks up" the operator's console (keyboard and < >mouse) running OpenVMS v7.3 patched up through a month ago.  I You know that V7.3-1 (with ECOs) is current and you most likely will want F to use the advantages V7.3-1 brought to us all. So consider upgrading.  G I have "lock-up" problems on my Alpha Workstation with the keyboard but L I think they are more related to DECwindows-MOTIF (V1.2-6) and the Right-AltL Key (or some key combination in the righthand corner of the german keyboard)L than to the hardware. Effect is that the keyboard can't be used for a coupleI of minutes (a timeout maybe) while (as opposed to your described problem)  the mouse keeps responding.   I But as long as you can't describe your lock-up problem in more detail, it K is hard to guess whether it is hardware (a memory problem ?) or software...    --   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   ------------------------------  $ Date: Mon, 9 Jun 2003 11:35:55 -0400, From: "warren sander" <warren.sander@hp.com>$ Subject: Re: alphaServer diagnostics, Message-ID: <3ee4ad94$1@usenet01.boi.hp.com>  $ you could go to the ds10 web site at+ http://h18003.www1.hp.com/alphaserver/ds10/ ( from there you can get firmware updates:  ? http://ftp.digital.com/pub/DEC/Alpha/firmware/readmes/ds10.html    documentation:9 http://h18003.www1.hp.com/alphaserver/ds10/ds10_tech.html    and/or you can get the  > service tools from http://h71000.www7.hp.com/serv_support.html5 http://h18000.www1.hp.com/support/svctools/index.html      --  K --------------------------------------------------------------------------- 7 Warren Sander                   WW E-Marketing (HP.COM) B Hewlett-Packard Company         Work:  warren.sander@remove.hp.comK 200 Forest Street MR01-3/K8     Personal: sander.ma.ultranet@remove.rcn.com . Marlboro, MA 01752              (508) 467-48755    My opinions are my own and I only speak for myself *          Read http://www.hp.com/go/openvmsK ---------------------------------------------------------------------------     9 "Jim Strehlow" <JimStrehlow@data911.com> wrote in message 7 news:4b6ec350.0306060859.2d7a24b7@posting.google.com... > > What diagnostic tools are available for an AlphaServer DS10? >  - freeware download! >  - on the software tools cd-rom # >  - for licensed purchase from HP?  > @ > We have an OpenVMS AlphaServer DS10 on sofware support but notH > hardware support which "locks up" the operator's console (keyboard and= > mouse) running OpenVMS v7.3 patched up through a month ago.  > * > Thank you in advance for your responses. >  > Jim Strehlow > Data911, Alameda, CA, USA    ------------------------------   Date: 9 Jun 2003 10:13:33 -0700 , From: JimStrehlow@data911.com (Jim Strehlow)$ Subject: Re: alphaServer diagnostics< Message-ID: <4b6ec350.0306090913.d9149c5@posting.google.com>  C The system hangs either while booting or at the DECwindows operator 3 console after logging in and opening a few windows.   F The particular system has been "on the road" for sales demonstrations;: so it gets banged around more than the normal AlphaServer.$ We are reseating all connectors etc.> The problem may be memory or motherboard ... or our travelling salespeople :-)   6 No errors appear in ERRLOG.SYS nor on the SHOW ERRORS.  D I have historically had tools for personal computer diagnostics that? do memory, monitor, disk tests, etc. The tools were either free % downloads or on Norton Utilities etc.   E I have had so few problems historically with VAX and Alpha chips that C I had not thought of wanting to use some sort of Alpha chip utility  program until now.   Jim Strehlow Data911, Alameda, CA, USA    ------------------------------   Date: 9 Jun 2003 08:19:08 -0700 1 From: keithparris_NOSPAM@yahoo.com (Keith Parris) : Subject: Re: Another future of VMS article on the Inquirer= Message-ID: <cf15391e.0306090719.5e933532@posting.google.com>   d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<ipucnQSyadqPCn-jXTWcqw@metrocast.net>...L > I did find Terry's recent Inquirer reference to $800 million in annual VMS > system profit rather a hoot  ... I > Since then, VMS revenues have dropped by at least half (from $4 billion I > annually as of Y2K/early 2001) to $2 billion annually just a few months G > later (as of the December, 2001 formal letter responding to Gartner's M > negative comments about VMS, so one might reasonably assume that the figure  > wasn't being low-balled).   B During HP Enterprise Users Week in Amsterdam a few weeks ago, MarkE Gorham told the attendees that VMS generates between $2.5 billion and D $3.0 billion in revenues, and $500 million in profits (in US dollars& and definition of "billion") annually.   ------------------------------   Date: 9 Jun 2003 08:40:36 -0700 1 From: keithparris_NOSPAM@yahoo.com (Keith Parris) D Subject: Re: BLASTed directory locks, timing windows & endless loops= Message-ID: <cf15391e.0306090740.2f71922e@posting.google.com>   j "Richard Maher" <maher_rj@hotspamnotmail.com> wrote in message news:<bbvlcl$e7i$1@titan.btinternet.com>...L > 1) I now get told when "something" happens on a given directory and I alsoG > check (if the second longword of the LVB has changed) if the file I'm K > looking for has arrived, but I need to know when the creation of the file N > has completed as opposed to started. So what I'm after is a run-of-the-mill,J > missionary-position file-lock. This seems not to be the arbitration lockJ > that Keith Parris has talked about previously, and is very hard to trackN > down. (SDA> sh proc/locks Doesn't work 'cos PID is zero. And SDA> show locksK > doesn't have any F11B$a locks)  What is the resource name and access mode  > for a simple *File* Lock?   F How about the RMS file lock? (Documented in the IDSM appendix on "LockF and resource use by OpenVMS Components")  Resource name is of the formE RMS$<FileID><sharing_flag><device_name> (where sharing_flag is a byte D with value 2 for a volume mounted /SYSTEM).  I expect that lock will/ be held until the file population is completed.    ------------------------------   Date: 9 Jun 2003 09:44:54 -0500 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) , Subject: Re: CNN Story Not Favorable To Bush3 Message-ID: <Oe$n0v0BNFOV@eisner.encompasserve.org>   c In article <MLeIgocQc+tm@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:  > / > When I was in the Boy Scouts, the saying was:  > 3 > 	Leave your campsite _cleaner_ than you found it.   8    Gee, I was never in the Boy Scouts.  Do they use VMS?   ------------------------------   Date: 9 Jun 2003 09:43:15 -0700 * From: morrisj@epsilon3.com (Jay E. Morris), Subject: Re: CNN Story Not Favorable To Bush= Message-ID: <8600a73e.0306090843.19113044@posting.google.com>   X VAXman-  @SendSpamHere.ORG wrote in message news:<00A21140.EA6CEF14@SendSpamHere.ORG>...W > In article <SFm39CaGhFJ$@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes:  .....  > I > Why kowtowing to his revered lordship the Emperor of Redmond of course.   1 Then today's Userfriendly cartoon seems relevant.   1 http://ars.userfriendly.org/cartoons/?id=20030609    ------------------------------   Date: 9 Jun 2003 05:46:17 -0700 6 From: dave.reynolds@gallatinsteel.com (David Reynolds)  Subject: Re: Corporate takeovers= Message-ID: <a813e962.0306090446.2099be5f@posting.google.com>   ` "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3EE220FB.C1618FF9@fsi.net>... > "Main, Kerry" wrote: > >   > > > -----Original Message-----7 > > > From: JF Mezei [mailto:jfmezei.spamnot@istop.com] ! > > > Sent: June 7, 2003 12:49 AM  > > > To: Info-VAX@Mvb.Saic.Com " > > > Subject: Corporate takeovers > > >  > > >  > > > Oracle buys Peoplesoft.  > > > ? > > > Corel (owner of Wordperfect) sells itself to a california . > > > venture capitalist for a very low price. > > >  > >  > > JF - > > - > > I thought you were joking .. Holy smoke !  > > # > > Reference Oracle press release: D > > http://www.oracle.com/corporate/investor_relations/FINALpsft.pdfH > > "ORACLE TO LAUNCH CASH TENDER OFFER FOR PEOPLESOFT FOR $16 PER SHARE > H > I heard that on the business news on the radio on the way to the trainH > station Friday a.m. and spent the trip in pondering what it might mean > for VMS:   > I > Oracle + PeopleSoft back-end on VMS instead of or in addition to HP-UX?  >  > Rather a tasty prospect, eh? > J > Could go the other way, too, I s'pose. Oracle + HP uses that as leverageG > to get VMS folks to move to HP-UX or to promote new sales of HP-UX on 	 > Itanic.   B Unfortunatly the following is from an InfromationWeek article (url	 wrapped?)   u http://www.informationweek.com/story/showArticle.jhtml;jsessionid=GVAD20IZK2BEKQSNDBGCKHSCJUMEKJVN?articleID=10300351   C "The acquisition of PeopleSoft will immediately make Oracle an even B more profitable and competitive company," said Oracle chairman and@ chief executive Larry Ellison. "Although we will not be activelyF selling PeopleSoft products to new customers, we will provide enhanced< support for all PeopleSoft products. Furthermore, we will beE incorporating the advanced features from the PeopleSoft products into / future versions of the Oracle eBusiness Suite."   E Also note that PeopleSoft was/is in progress of buying J.D. Edwards &  Co.    ------------------------------   Date: 9 Jun 03 13:39:29 +0200 ) From: p_sture@elias.decus.ch (Paul Sture) + Subject: CSWB 1.2.1 fails to display images ) Message-ID: <CJ5Ko3MwwfM7@elias.decus.ch>   H I have just installed CSWB (although it now calls itself SWB) 1.2.1, and it doesn't display any images.  C This could be a leftover from version 1.0. Has anyone seen anything  similar?   --  
 Paul Sture   ------------------------------  % Date: Mon, 09 Jun 2003 15:14:27 +0200 ) From: Bart Zorn <B.Zorn@xs4all.nospam.nl> / Subject: Re: CSWB 1.2.1 fails to display images 6 Message-ID: <3ee48834$0$49114$e4fe514c@news.xs4all.nl>   Paul Sture wrote: J > I have just installed CSWB (although it now calls itself SWB) 1.2.1, and  > it doesn't display any images. > E > This could be a leftover from version 1.0. Has anyone seen anything 
 > similar? >   G On a DS10 I have no problems with SWB V1.2.1. Before I installed it, I  5 had Mozilla 1.2.1 in use, which I first uninstalled.  D @SYS$COMMON:[MOZILLA]INSTALL REMOVE and then PRODUCT REMOVE MOZILLA.  I Next I installed SWB and apart from a few cosmetic differences, I see no  6 functional differences between the two 1.2.1 flavours.  	 Bart Zorn    ------------------------------  % Date: Mon, 09 Jun 2003 16:00:07 +0200  From: Dirk Munk <munk@home.nl>/ Subject: Re: CSWB 1.2.1 fails to display images 2 Message-ID: <bc246d$hme$1@news2.tilbu1.nb.home.nl>   Paul Sture wrote: J > I have just installed CSWB (although it now calls itself SWB) 1.2.1, and  > it doesn't display any images. > E > This could be a leftover from version 1.0. Has anyone seen anything 
 > similar? > Q No, I have been using Mozilla for three years no (since V G0.5 , that was before  L V0.9-3), and I can't remember any such problem. At the moment I am on 1.4RC1L Did you read read the release notes on memory consumption, page files etc. ?   ------------------------------   Date: 9 Jun 2003 08:36:56 -0700 % From: tadamsmar@yahoo.com (Tom Adams) 1 Subject: Re: Decserver 700 software on CD or DDS? = Message-ID: <ea44f5a1.0306090736.251f4839@posting.google.com>   i eplan@langstoeger.at (Peter LANGSTOEGER) wrote in message news:<i45Ca.30498$_c6.219490@news.chello.at>... g > In article <ea44f5a1.0305310657.1c3705a9@posting.google.com>, tadamsmar@yahoo.com (Tom Adams) writes: L > >I installed DS7C011 from the distribution and I got WWENG1.SYS instead of > >WWENG2.SYS. > E > WWENG1.SYS is the original software for the DECserver700, while the J > WWENG2.SYS is the Network Access Software for the DECserver 700 and 900. > G > If you need features of the DNAS (like Appletalk, IPX, KERBEROS, PPP, I > RADIUS, RLOGIN, SECURID, TD/SMP, TN3270, ...) then WWENG1.SYS is surely N > not sufficient, otherwise it will work, while DNAS - requiring more memory -9 > might not fit into _your_ version of the DECserver 700.   6 How do I find WWENG2.SYS on the software distribution?   ------------------------------  % Date: Mon, 09 Jun 2003 11:34:41 -0500 , From: Bob Blunt <robert.blunt@hp.nospam.com>1 Subject: Re: Decserver 700 software on CD or DDS? , Message-ID: <3ee4bb5b$1@usenet01.boi.hp.com>   Tom Adams wrote:k > eplan@langstoeger.at (Peter LANGSTOEGER) wrote in message news:<i45Ca.30498$_c6.219490@news.chello.at>...  > g >>In article <ea44f5a1.0305310657.1c3705a9@posting.google.com>, tadamsmar@yahoo.com (Tom Adams) writes:  >>L >>>I installed DS7C011 from the distribution and I got WWENG1.SYS instead of >>>WWENG2.SYS. >>E >>WWENG1.SYS is the original software for the DECserver700, while the J >>WWENG2.SYS is the Network Access Software for the DECserver 700 and 900. >>G >>If you need features of the DNAS (like Appletalk, IPX, KERBEROS, PPP, I >>RADIUS, RLOGIN, SECURID, TD/SMP, TN3270, ...) then WWENG1.SYS is surely N >>not sufficient, otherwise it will work, while DNAS - requiring more memory -9 >>might not fit into _your_ version of the DECserver 700.  >  > 8 > How do I find WWENG2.SYS on the software distribution?  C As someone else mentioned, WWENG2.SYS is part of the DECserver NAS  D software/firmware package.  You'll probably have to get the current C versions from the Digital Networks people (http://www.dnpg.com/ or  H http://www.digitalnetworks.net/).  Older versions MAY be on the CONDIST G and that would depend on which ones you have available.  In MOST cases  G the NAS software WILL require at least 8MB memory on the DECserver 700. E This software hasn't been part of the CONDIST for several years IIRC.    bob    ------------------------------   Date: 9 Jun 2003 09:57:56 -0700 1 From: keithparris_NOSPAM@yahoo.com (Keith Parris) ' Subject: Dell to make Itanium 2 systems = Message-ID: <cf15391e.0306090857.38b57fe6@posting.google.com>    From OpenVMS.org:  --- # Dell does an about face on Itanium  E   Posted by Kenneth Farmer (Wednesday June 04 2003 @ 06:35PM EDT) [ ]   = Looks like Dell has done an about-face. They had announced an D Itanium-based server back in 2001, but quietly cancelled the productE shortly afterward. But they recently demonstrated an Itanium 2 server D at Microsoft's Windows Server 2003 launch, and say they plan to ship Itanium boxes later this year.   See:  & InfoWorld: Dell demos Itanium 2 server< http://www.infoworld.com/article/03/04/24/HNdelldemo_1.html   E ComputerWeekly: Dell demos Itanium 2 server, but keeps lid on details ' http://www.cw360.com/Article121223.htm    ( PC World: Dell Develops Itanium 2 Server8 http://www.pcworld.com/news/article/0,aid,110423,00.asp   ? The Inquirer: Dell will plump for cheapo Intel Itanium Madisons ( http://www.theinquirer.net/?article=9780 ---   D A quick Google search confirms that indeed, starting about SeptemberD 2002, Dell started telling customers that it plans to ship Itanium 2 systems in 2003.   From Information Week D (http://www.informationweek.com/story/IWK20020912S0002), way back in September 2002: E 'Says [Kevin] Rollins [Dell's president and chief operating officer]: F "We've never doubted that Itanium would be a winner. The question was,- when it would be ready" for Dell's strategy?'   D From CNET News (http://zdnet.com.com/2100-1103-966499.html) November	 20, 2002: D 'The Austin, Texas-based computer giant will come out with computersE containing Intel's latest 64-bit chip, called Itanium 2, according to @ Joe Marengi, senior vice president for Dell's Americas division.  D "We will support Intel all the way on this...We will have an IA-64 2F (Itanium 2) product on our roadmap," Marengi said in an interview withB CNET News.com at Comdex Fall 2002 here. "The technology is totally solid."' ... F 'While Marengi declined to discuss Opteron in depth, he noted that theC benchmark scores of Itanium are relatively strong and that the path 8 for adoption among corporate customers seems inevitable.  C "Over time, that (Itanium) is going to be the technology that takes F hold," he said. Last year [2001] he was negative on Itanium, but "this= year [2002] I am in a neutral-to-positive stance," he added.'  ... < 'SoundView Technology analyst Mark Speckter said that Dell'sA vacillation over the past year was largely for show. Adopting AMD D chips would have required Dell to design completely new computers asC well as stock additional parts. Servers running AMD chips have also 8 historically been shunned by corporate buyers, he added.  D Meanwhile, Microsoft has continued to pledge its support to Itanium.F Itanium has been "not a matter of if they (Dell) do it, it is a matter of when," Speckter said.  B Scott Randall, also an analyst at SoundView, said that Opteron wasA also hurt by delays.  If the chip had come out six months ago, it 8 would have posed more of a competitive threat, he said.'   ------------------------------  # Date: Mon, 09 Jun 2003 11:10:32 GMT 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)W Subject: Re: ERRLOG.SYS (was Re: Your Own Installation Tool? (was: Install  Directory)) 4 Message-ID: <IWZEa.109757$lL2.999688@news.chello.at>  _ In article <S4NFPvWVVkWR@cuebid.zko.dec.com>, brooks@cuebid.zko.dec.nospam (Rob Brooks) writes: E >clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) writes: b >> In article <+ENHMJHTsxvR@cuebid.zko.dec.com>, brooks@cuebid.zko.dec.nospam (Rob Brooks) writes:7 >>> keithparris_NOSPAM@yahoo.com (Keith Parris) writes:  >>>>  I >>>> For VMS on Itanium, bit-to-text translation (what ANALYZE/ERROR used B >>>> to do)  will be provided by the operating system folks again. >>> I >>> It isn't just for Itanium.  OpenVMS Alpha V7.3-2 will introduce a VMS A >>> Engineering-provided mechanism to do bit-to-text translation.  >>   >> That is excellent news.  G Indeed it is. And may I thank you (as the messenger) and the engineer - 5 and of course his/her boss - for making this happen ! > In the meantime, can some kind soul offer WEBES/CA V4.1.4 ? to& a (not so) humble OpenVMS Hobbyist ;-)  > >> Will patch kits be available for previous versions of VMS ? > F >No, but the engineer who did all the work to make this wonderful toolG >available says that older error logs (back to V7.2) can be analyzed on M >a V7.3-2 system.  I don't know if that function (analyzing older error logs) L >will be supported, but it should work.  He's actually looking to see if theM >new tool can be (easily) made to translate error logs from pre-V7.2 systems.   H Copying an ERRLOG.SYS file from one system to another is what we did forD years now and which seems not sufficient. Better would be to use theG new tool on older version systems (which means the tool should not only I be able to understand the format but also need to be backported to run on  the older systems).    --   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: Mon, 09 Jun 2003 11:11:31 +0100 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> % Subject: Re: Firewall for VMS / TRU64 0 Message-ID: <bc1mgt$otg$1@new-usenet.uk.sun.com>   Bob Ceculski wrote: b > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3EE14342.152CD897@fsi.net>... >  >>Bob Ceculski wrote:  >>> >>>look Andrew, the point is that with VMS you don't need nat, >>C >>Um, Bob? The vast majority of what NAT is all about has little or H >>nothing to do with protection from outside influence. It mostly has to? >>do with channeling bunches of unroutable addresses (10.x.x.x, I >>192.168.x.x, ...) through a single "registered" ("world-wide") address.  >> >> >>>you> >>>don't need filters, vpn, stealth or any other unix piece ofE >>>garbage because as defcon9 and certs prove and as you conveniently D >>>like to avoid them, VMS already has security! What don't you get? >>F >>I sincerely hope you are not in charge of network security where youI >>work. If such should be the case, however, InstantWhip is in whole heap  >>o'trouble! >  > E > you still fail to get the point!  The only thing a firewall is good B > for is DOS attacks ... if a vms box were put live without one onH > the internet, it would fare very well as defcon9 proved, "unhackable"!  / Ohhhhhhhhh Myyyyyyyyyyyyy Godddddddddddddddddd.   1 Followed by the sound of me falling off my chair.   7 Now I wonder if Bob is responsible for network security 1 at instantwhip.com who appear to be running their 5 corporate web site on Windows/IIS. Oh joy no firewall 0 and a Microsoft web infrastructure that would be like shooting fish in a barrel.    Regards  Andrew Harrison    ------------------------------   Date: 9 Jun 03 12:59:17 +0200 ) From: p_sture@elias.decus.ch (Paul Sture) % Subject: Re: Firewall for VMS / TRU64 ) Message-ID: <9xbk+qpcJ6QL@elias.decus.ch>    In article <bc1mgt$otg$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > Bob Ceculski wrote:    [snip]  F >> you still fail to get the point!  The only thing a firewall is goodC >> for is DOS attacks ... if a vms box were put live without one on I >> the internet, it would fare very well as defcon9 proved, "unhackable"!  > 1 > Ohhhhhhhhh Myyyyyyyyyyyyy Godddddddddddddddddd.  > 3 > Followed by the sound of me falling off my chair.   > And I have to see if these reports of LK keyboards surviving a dishwasher are true :-)    > 9 > Now I wonder if Bob is responsible for network security 3 > at instantwhip.com who appear to be running their 7 > corporate web site on Windows/IIS. Oh joy no firewall 2 > and a Microsoft web infrastructure that would be! > like shooting fish in a barrel.  >    www.netcraft.com reports this:  8 The site www.instantwhip.com is running Purveyor Encrypt! Domestic/v2.1 OpenVMS on OpenVMS.        --  
 Paul Sture   ------------------------------   Date: 9 Jun 2003 05:02:30 -0700 ( From: bob@instantwhip.com (Bob Ceculski)% Subject: Re: Firewall for VMS / TRU64 = Message-ID: <d7791aa1.0306090402.230b3704@posting.google.com>   ` "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3EE36259.6AA46C8F@fsi.net>... > Bob Ceculski wrote:  > > d > > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3EE14342.152CD897@fsi.net>... > > > Bob Ceculski wrote:  > > > > C > > > > look Andrew, the point is that with VMS you don't need nat,  > > > G > > > Um, Bob? The vast majority of what NAT is all about has little or L > > > nothing to do with protection from outside influence. It mostly has toC > > > do with channeling bunches of unroutable addresses (10.x.x.x, M > > > 192.168.x.x, ...) through a single "registered" ("world-wide") address.  > > >  > > > > you C > > > > don't need filters, vpn, stealth or any other unix piece of J > > > > garbage because as defcon9 and certs prove and as you convenientlyI > > > > like to avoid them, VMS already has security! What don't you get?  > > > J > > > I sincerely hope you are not in charge of network security where youM > > > work. If such should be the case, however, InstantWhip is in whole heap  > > > o'trouble! > > B > > wasn't there someone on this board a year or two ago that said> > > they had a vms box up on the web without any firewall, and@ > > challenged anyone to break in ... who was that and why don't > > we give you a crack at it? > J > Already been done. Proved that UN*X-land software ported to / running on/ > VMS still displays UN*X-land vulnerabilities.  > F > Wanna try it yourself? Be my guest. I'll be standing by with my cell; > phone to "call the paramedics" when you "crash and burn".  > ? > ...unless you feel that a VMS box running nothing but VMS (no G > applications that may have vulnerabilities of their own) is a useful, 9 > profitable appliance. If so, then again: be my guest...   5 so why do you run unix land software on your vms box? 7 and so now we all here go from stating that "VMS is the 8 most secure os on the planet" to "VMS is a piece of #%$&6 and you can't even run it as a webserver or ftp server4 or it will get hacked to pieces"!  So that means you7 can't put a VMS box running say apache on the internet, 4 lock it down, and put a simple firewall in front for- dos attacks?  Well then bring back rsts/e ...    ------------------------------   Date: 9 Jun 2003 09:17:34 -0500 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) % Subject: Re: Firewall for VMS / TRU64 3 Message-ID: <SqvpUwcaSKGu@eisner.encompasserve.org>   h In article <d7791aa1.0306071025.2ce6a66d@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:  C > and again I challenge you!  I'll bet a vms box without a firewallc? > is unhackable just like defcon9 found out ... I would love torA > see one put up on the web, and you just hack right into it, thegD > only restriction being a DOS attack (which your isp and a firewall? > can handle) ... so you think you are smarter then the defcon9r	 > people?a  A    I know lots of folks who put naked VMS systems directly on thee?    Internet with no firewall.  I've seen such systems attacked.t     I've never seen one breached.  @    I've seen Solaris, IRIX, and Windows systems behind firewalls:    breached without ever knowing they were being attacked.  C    But I put my home systems behind nat so my single ISP connectionn@    will work.  VMS doesn't include nat, and it doesn't include a    firewall.   ------------------------------  % Date: Mon, 09 Jun 2003 16:18:38 +0100rO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>e% Subject: Re: Firewall for VMS / TRU64I0 Message-ID: <bc28go$1er$1@new-usenet.uk.sun.com>   Bob Ceculski wrote:a  7 > so why do you run unix land software on your vms box?S9 > and so now we all here go from stating that "VMS is theg: > most secure os on the planet" to "VMS is a piece of #%$&8 > and you can't even run it as a webserver or ftp server6 > or it will get hacked to pieces"!  So that means you9 > can't put a VMS box running say apache on the internet,r6 > lock it down, and put a simple firewall in front for/ > dos attacks?  Well then bring back rsts/e ...   * No one has said that OpenVMS is a piece of crap.s  + A number of people have said that you don'tc* on the evidence of this thread have a clue what you are talking about.   - Don't confuse the two, you arn't an operatingo- system as far as I know and just because yourV, points about OpenVMS are demonstrably untrue) doesn't mean the OS is demonstrably crap.    Regards  Andrew Harrisone   ------------------------------  % Date: Mon, 09 Jun 2003 11:02:27 +0200 7 From: Robert Trawinski <robert.trawinski@softax.com.pl>4 Subject: Gnu make question/ Message-ID: <bc1if3$sn2$1@bozon2.softax.com.pl>   
 Hi everybody,M  H I try to put into makefile rule for compiling Oracle ProcC code. I need  use command like
 fallowing:  ? proc SQLCHECK=SEMANTICS CODE=CPP USERID=authorization source.pcV  I How can I get USERID value from process environment (from logical name)?   We use gnu make.   Robert   ------------------------------  * Date: Mon, 9 Jun 2003 15:51:16 +0000 (UTC), From: lewis@PROBE.mitre.org (Keith A. Lewis) Subject: Re: Gnu make question. Message-ID: <bc2adk$nis$1@newslocal.mitre.org>   Robert Trawinski <robert.trawinski@softax.com.pl> writes in article <bc1if3$sn2$1@bozon2.softax.com.pl> dated Mon, 09 Jun 2003 11:02:27 +0200: >Hi everybody, >tI >I try to put into makefile rule for compiling Oracle ProcC code. I need n >use command liket >fallowing:u >e@ >proc SQLCHECK=SEMANTICS CODE=CPP USERID=authorization source.pc > J >How can I get USERID value from process environment (from logical name)?  >We use gnu make.e  / I can tell you how to get something from DCL.  l  M     $ def/job USERID 'f$identifier(f$getjpi("","USERNAME"),"NAME_TO_NUMBER")'   K That assumes your username is the same as the identifier name (which is them, default in VMS but not strictly enforced).    + --Keith Lewis              klewis$mitre.org > The above may not (yet) represent the opinions of my employer.   ------------------------------   Date: 9 Jun 03 08:25:06 +0200 ) From: p_sture@elias.decus.ch (Paul Sture) 3 Subject: Re: Halt & Startup Issues with VS 4000 VLCa) Message-ID: <BSQ+M85++gBy@elias.decus.ch>o  m In article <t4QEa.100189$lL2.949127@news.chello.at>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes:ly > In article <01KWRWJWSEI4AOKN0V@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:TK >>> Yes exactly, I've seen this behaviour as well. Digital series VT's have ( >>> never, in my experience, done this.  >>D >>All of my hobbyist VAX machines have genuine |d|i|g|i|t|a|l| VT320F >>terminals as their consoles.  Most, if not all, show this behaviour. > N > Yes, mine too. And not only the original ones, but also the compatible ones. > Like the CIT 224, 324, 324+  > L > And opposed to the described behaviour, mine do send the break at Power-On > and not at Power-Off...X >   F ? As far as I recall, I've only witnessed a break at power on (variousJ DEC and third party clones). I can see the use in terms of a LAT connected9 terminal, but as a console, it can be a pain in the neck.d  F The first cluster I worked on was fully UPS protected, and it was withD some pleasure that we saw the lights go out one day, but could stillC carry on working. The lights came back on, and the cluster stopped.2I Doh! the 2 VT consoles were the only items on a non-UPS circuit, and spatv out a break on power up.     -- 7
 Paul Sture   ------------------------------  + Date: Mon, 09 Jun 2003 11:17:15 +0100 (MET)m9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> 3 Subject: Re: Halt & Startup Issues with VS 4000 VLCl; Message-ID: <01KWW087SK9EAM67GX@sysdev.deutsche-boerse.com>w  B > Easier (and less stress on the cables) than unplugging the cableB > every time you want to turn the terminal off, is to get an RS232? > A/B switch and connect the VAX to "A" and nothing to "B", and = > switch to "B" before turning it off.  (Make sure the switch.D > doesn't generate a break when you throw it.)  I bet these switchesG > show up on Ebay a lot, since it seems to me back in the terminal era,dC > almost everyone who had a terminal on their desk had an A/B(/C/D)pB > switch next to it.  They were probably all surplused when people> > switched to PC's or to VT420-vintage terminals with multiple > serial ports.e  H Interesting.  I once tried switching from the DEC 423 (console) to RS232C (don't know if anything was connected or not) in the terminal setupy. menu.  Apparently, this also produces a BREAK.   ------------------------------  $ Date: Mon, 9 Jun 2003 05:09:00 -0400  From: John Santos <JOHN@egh.com>3 Subject: Re: Halt & Startup Issues with VS 4000 VLCe6 Message-ID: <1030609045738.10662A-100000@Ives.egh.com>  , On Sat, 7 Jun 2003, David J. Dachtera wrote:   > John Santos wrote: > > , > > On Fri, 6 Jun 2003, Andrew Balaam wrote: > >  > > >  > > >o< > > > >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<< > > >eO > > > On 06/06/03, 09:45:51, Nic Clews <sendspamhere@127.0.0.1> wrote regarding 1 > > > Re: Halt & Startup Issues with VS 4000 VLC:f > > >p > > >h > > > > Andrew Balaam wrote:	 > > > > >l@ > > > > > >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<	 > > > > >MQ > > > > > On 06/06/03, 06:05:22, VLC Novice <The7car@comcast.net> wrote regardingr1 > > > > > Halt & Startup Issues with VS 4000 VLC:@ > > >eP > > > > > > Secondly, I'm using a Wyse 85 (in VT100 mode) as the console, but if9 > > > > > > the terminal is turned off, it halts the VLC.o	 > > > > >lP > > > > > The reason the vlc halts when the Wyse is switched off, is because theP > > > > > vlc sees the RS232 BREAK condition - the BREAK being a relatively longK > > > > > MARK or SPACE I can't remember exactly which one. If you pull thesQ > > > > > connector out out of the vlc before you turn off the Wyse, the vlc willDS > > > > > probably not HALT. If you want to use the Wyse again, without HALTing thedS > > > > > vlc, make sure you turn on the Wyse, let it go through all its power self - > > > > > test etc, and then plug it back in.. > > >WO > > > > Yes exactly, I've seen this behaviour as well. Digital series VT's have , > > > > never, in  my experience, done this. > > > 
 > > > > --G > > > > Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences1 > > > > nclews at csc dot com  > > >kO > > > Never having used a genuine DEC terminal, I can't comment on that. We haddM > > > a couple of CITOH VT100s and a couple of Wyse VT220 terminals, and they - > > > both sent BREAKs when turned on or off.O > > D > > Don't know about the VLC, but on many VAXes, you can disable theD > > "HALT ON BREAK" function by setting a switch correctly.  (If youC > > need to halt, e.g. for kernel debugging, then you have to reset)( > > the switch before pressing <BREAK>.) > > F > > Check the terminal settings to see if the "SEND BREAK ON POWEROFF"& > > is a feature that can be disabled. > > D > > Easier (and less stress on the cables) than unplugging the cableD > > every time you want to turn the terminal off, is to get an RS232A > > A/B switch and connect the VAX to "A" and nothing to "B", and ? > > switch to "B" before turning it off.  (Make sure the switch-F > > doesn't generate a break when you throw it.)  I bet these switchesI > > show up on Ebay a lot, since it seems to me back in the terminal era,aE > > almost everyone who had a terminal on their desk had an A/B(/C/D)pD > > switch next to it.  They were probably all surplused when people@ > > switched to PC's or to VT420-vintage terminals with multiple > > serial ports.o > G > If you can find the parts, and if you're into a little soldering, anyn@ > 6-pole, double- or more throw switch will do. Just cut up someI > DECconnect cables and solder directly to the lugs on the switch, unless  > you can find MMJ jacks.- > H > In the case of a rotary switch, make sure it's break-before-make (thatC > is, break one set of connections before making the next) to avoid I > certain issues. A 6PDT toggle switch would likely be break-before-make,s > but check the spec.'s.  
 Good idea.  C Before terminal servers and multi-session terminals, I used to have G a little cottage industry building these for people at work.  Actually,vD you only need 3-pole switches (TD, RD and Ground) for most apps, butG having all 6 wires can't hurt.  I never built an MMJ version.  All mine G used DB25 connectors (for VT52's and VT100's) and twisted-pair telehones@ wire, but except for the finer wires, MMJ shouldn't be too hard.  E You don't even need to decode the colors/pins if you use all 6 wires.oF Just be sure to connect all the same color (or number) wire to each of
 the poles.  A Usually, the available switches had more positions than I needed, . so there was always at least one OFF position.   -- a John Santosh Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------   Date: 9 Jun 2003 03:33:55 -0700-+ From: freestyle_london@yahoo.co.uk (Andrew)i8 Subject: Re: hanging UCX sessions, expert help required.= Message-ID: <88f6adbc.0306090233.26f04f81@posting.google.com>i  - Just a quick thank you for those who replied.o  5 The server was pointing to a non-exitant DNS server. n  ^ "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3ECF8EC6.548C96@fsi.net>... > Andrew wrote:e > > H > > I am the network manager of our company. We have limited VMS supportI > > onsite, as the main admin work on the VMS boxes is oracle/applicationa
 > > based. > > / > > We have a problem with 2 servers currently.R > > C > > Both run VMS 7.1-2 with the most recent version of UCX with ECOtJ > > applied. We are planning a program to move to the new VMS and TCP/IP 5H > > soon ! before anyone asks, but I need a fix for our current systems.J > > We are in contact with compaq... sorry HP but so far we are being told# > > we are having network problems.i > >  > > Here is the Scenario > > J > > 2 telnet sessions running and loggged in. We type is a command that isD > > specific to our application so I'll not mention it here, and theJ > > response from that hangs. At the same time we cannot establish any newF > > telnet/FTP sessions or can the server print using Jetdirect to ourJ > > many printers onsite. But the any exiting FTP/Telnet sessions continue > > to work normally !!u > > J > > Help a network guy pass it back to our VMS people who keep blamming my
 > > network !  > E > Obviously - the failure to accept new incoming connections (is that E > *ALL* types? Telnet included? Can you PING the machine?) suggests aa2 > failure of the InternetACP process of some sort. >  > That's gonna be a toughie!   ------------------------------  $ Date: Mon, 9 Jun 2003 14:34:17 +01008 From: "Phil Sharpe" <fed_up_with_spam@nonexistant.co.uk># Subject: Help listing SYSGEN params 4 Message-ID: <bc22e6$ctj$1$8300dec7@news.demon.co.uk>  ? Anyone know how to get a complete listing of SYSGEN parameters.dG I've tried copying the contents of my scrollback buffer after SHOW/ALL,-> but there are gaps (presumably because of the screen heading).  G OpenVMS (VAX) v7.1 displaying in a "Reflection for the Web 3.0" window.g   TIA, Phil Sharpeo   ------------------------------  $ Date: Mon, 9 Jun 2003 10:00:44 -0400' From: "Main, Kerry" <Kerry.Main@hp.com> ' Subject: RE: Help listing SYSGEN params.R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB058929@tayexc19.americas.cpqcorp.net>   > -----Original Message-----B > From: Phil Sharpe [mailto:fed_up_with_spam@nonexistant.co.uk]=20 > Sent: June 9, 2003 9:34 AM > To: Info-VAX@Mvb.Saic.Comf% > Subject: Help listing SYSGEN paramsr >=20 >=208 > Anyone know how to get a complete listing of SYSGEN=20A > parameters. I've tried copying the contents of my scrollback=20rB > buffer after SHOW/ALL, but there are gaps (presumably because=20 > of the screen heading).  >=20? > OpenVMS (VAX) v7.1 displaying in a "Reflection for the Web=20o > 3.0" window. >=20 > TIA,
 > Phil Sharped >=20 >=20   Phil,i   Try the following: SYSGEN> set /output=3Dfile.txt SYSGEN> show /alli SYSGEN> show /speciale SYSGEN> exit $o   Regards   
 Kerry Main Senior Consultant$ Hewlett-Packard (Canada) Co.! Consulting & Integration Services  Voice: 613-592-4660v Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)  OpenVMS DCL - the original .COM    ------------------------------   Date: 9 Jun 03 07:30:54 PSTn From: mckinneyj@cpva.saic.com ' Subject: RE: Help listing SYSGEN paramsu( Message-ID: <mP+PZU7WYqdu@cpva.saic.com>   In article  G <FD827B33AB0D9C4E92EACEEFEE2BA2FB058929@tayexc19.americas.cpqcorp.net>,e*  "Main, Kerry" <Kerry.Main@hp.com> writes: >> -----Original Message-----gC >> From: Phil Sharpe [mailto:fed_up_with_spam@nonexistant.co.uk]=20  >> Sent: June 9, 2003 9:34 AMo >> To: Info-VAX@Mvb.Saic.Com& >> Subject: Help listing SYSGEN params >> >>6 >> Anyone know how to get a complete listing of SYSGEN? >> parameters. I've tried copying the contents of my scrollback0@ >> buffer after SHOW/ALL, but there are gaps (presumably because >> of the screen heading). >>= >> OpenVMS (VAX) v7.1 displaying in a "Reflection for the WebG >> 3.0" window.c >> >> TIA,l >> Phil Sharpe >> >> >  > Phil,. >  > Try the following: > SYSGEN> set /output=file.txt > SYSGEN> show /allr > SYSGEN> show /specialf > SYSGEN> exit > $  >   # Or maybe you'd prefer the followinga  * $ pipe mcr sysgen show/all | type sys$pipe   ------------------------------   Date: 9 Jun 2003 09:08:33 -0500l From: briggs@encompasserve.org' Subject: Re: Help listing SYSGEN params-3 Message-ID: <xSPd$jFfjYC+@eisner.encompasserve.org>   o In article <bc22e6$ctj$1$8300dec7@news.demon.co.uk>, "Phil Sharpe" <fed_up_with_spam@nonexistant.co.uk> writes:yA > Anyone know how to get a complete listing of SYSGEN parameters.zI > I've tried copying the contents of my scrollback buffer after SHOW/ALL, @ > but there are gaps (presumably because of the screen heading).  7 Annoying, isn't it?  Fancy behavior for no good reason.    The simple solution is to:   $ define sys$output foo.bari $ mcr sysgen SYSGEN>  SHOW /ALL SYSGEN>  SHOW /SPECIAL SYSGEN>   Exit   $ type foo.bar hA %TYPE-W-OPENIN, error opening DISK1420:[VAXS09]FOO.BAR;2 as inputs1 -RMS-E-FLK, file currently locked by another usere $ deassign sys$output	! *blush*  $ type foo.bar   Parameters in use: ActiveEI Parameter Name           Current    Default     Min.      Max.     Unit  o DynamicaI --------------           -------    -------    -------   -------   ----  i -------eN PFCDEFAULT                     64         64         0       2032 Pagelets   D ...     B Line wrap happens with the TYPE command, and is not present in theA file, foo.bar.  SYSGEN apparently assumes an adequate line lengthSA when generating output to disk file.  If you need to capture fromnD an emulator and you're using Reflection, the easy fix is (of course)   $ SET TERM /WIDTH=132/  . before typing the contents of the output file.   	John Briggs   ------------------------------   Date: 9 Jun 2003 09:47:56 -0500 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)a' Subject: Re: Help listing SYSGEN params 3 Message-ID: <Cq4BcIpp9WUR@eisner.encompasserve.org>h  o In article <bc22e6$ctj$1$8300dec7@news.demon.co.uk>, "Phil Sharpe" <fed_up_with_spam@nonexistant.co.uk> writes:eA > Anyone know how to get a complete listing of SYSGEN parameters.cI > I've tried copying the contents of my scrollback buffer after SHOW/ALL,l@ > but there are gaps (presumably because of the screen heading). > I > OpenVMS (VAX) v7.1 displaying in a "Reflection for the Web 3.0" window.  >  > TIA,
 > Phil Sharpem      $set host 0 /log=-sysgen.lise
    (login)    $set terminal/hards    $mcr sysgen sho /all 
    $logoff   ------------------------------  % Date: Mon, 09 Jun 2003 11:16:02 -0400e& From: Jilly <jilly@clarityconnect.com>' Subject: Re: Help listing SYSGEN paramso2 Message-ID: <3EE4A4B2.1B4D6529@clarityconnect.com>  
 MCR SYSGEN SET/OUTPUT {file.ext}- SHOW/ALL SHOW/SPECIAL EXIT   Phil Sharpe wrote: > A > Anyone know how to get a complete listing of SYSGEN parameters..I > I've tried copying the contents of my scrollback buffer after SHOW/ALL,a@ > but there are gaps (presumably because of the screen heading). > I > OpenVMS (VAX) v7.1 displaying in a "Reflection for the Web 3.0" window.  >  > TIA,
 > Phil Sharpeo   -- oC Jilly	- Working from Home in the Chemung River Valley - Waverly, NYn0 	- jilly@clarityconnect.com			- Brett Bodine fan+ 	- Mark.Jilson@hp.com				- since 1975 or so  	- http://www.jilly.baka.com   ------------------------------   Date: 9 Jun 2003 03:54:31 -0700a6 From: andrew.rycroft@intrinsitech.com (Andrew Rycroft)8 Subject: Help with upgrade to VMS v7.3-1 from VMS v7.2-1= Message-ID: <58ba0101.0306090254.5ffbdfdc@posting.google.com>o   Hi,u  C I am running a system with VMS v7.2-1 with DECnet Phase IV, and UCXW v4.2.   D When I attempt the upgrade off the CD, it proceeds to just after the> product version upgrade confirmation, and then aborts with the messages :-l  % DTSS-E-NOMSG, Message Number 0295897As% DTSS-E-NOMSG, Message Number 0295896A % DTSS-E-NOMSG, Message Number 1295896Ae  > There are no DTSS files on the system, as they do not use this	 facility.o  D I would be very mcuh obliged if someone can tell me what these mean, and what the remedy might be ?   Thanks Andrew   ------------------------------  $ Date: Mon, 9 Jun 2003 07:27:25 -0400' From: "Main, Kerry" <Kerry.Main@hp.com>u< Subject: RE: Help with upgrade to VMS v7.3-1 from VMS v7.2-1R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB058923@tayexc19.americas.cpqcorp.net>   > -----Original Message-----B > From: Andrew Rycroft [mailto:andrew.rycroft@intrinsitech.com]=20 > Sent: June 9, 2003 6:55 AM > To: Info-VAX@Mvb.Saic.Coma: > Subject: Help with upgrade to VMS v7.3-1 from VMS v7.2-1 >=20 >=20 > Hi,  >=20@ > I am running a system with VMS v7.2-1 with DECnet Phase IV,=20 > and UCX v4.2.  >=20? > When I attempt the upgrade off the CD, it proceeds to just=20e= > after the product version upgrade confirmation, and then=20o > aborts with the messages :-  >=20' > DTSS-E-NOMSG, Message Number 0295897Ae' > DTSS-E-NOMSG, Message Number 0295896AG' > DTSS-E-NOMSG, Message Number 1295896AE >=20> > There are no DTSS files on the system, as they do not use=20 > this facility. >=20= > I would be very mcuh obliged if someone can tell me what=20 , > these mean, and what the remedy might be ? >=20 > Thanks > Andrew >=20   Andrew,-  @ I found this problem documented in an internal problem database.F However, it stated it was a problem with V7.3 (GMT5 time zone) and not@ V7.3-1. Can you confirm the CD is a V7.3-1 version and not V7.3?  ? I would recommend contacting support if this is a V7.3-1 Cdrom.s   Regards9  
 Kerry Main Senior Consultant  Hewlett-Packard (Canada) Co.! Consulting & Integration Services> Voice: 613-592-4660  Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)  OpenVMS DCL - the original .COMS   ------------------------------  # Date: Mon, 09 Jun 2003 17:49:16 GMTS3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond) < Subject: Re: Help with upgrade to VMS v7.3-1 from VMS v7.2-11 Message-ID: <wM3Fa.2191$NF7.240@news.cpqcorp.net>,  v In article <58ba0101.0306090254.5ffbdfdc@posting.google.com>, andrew.rycroft@intrinsitech.com (Andrew Rycroft) writes: >Hi, >>D >I am running a system with VMS v7.2-1 with DECnet Phase IV, and UCX >v4.2. > E >When I attempt the upgrade off the CD, it proceeds to just after theW? >product version upgrade confirmation, and then aborts with theg >messages :- >l& >DTSS-E-NOMSG, Message Number 0295897A& >DTSS-E-NOMSG, Message Number 0295896A& >DTSS-E-NOMSG, Message Number 1295896A >s? >There are no DTSS files on the system, as they do not use thisp
 >facility.  A There are some DTSS files that are part of TCP/IP and/or OpenVMS.u  + What timezone was set prior to the upgrade.   F If you are using a timezone that generates an "invalid" timezone rule,E this could be the result.  Most of the GMTx, GMTPLUSx and GMT-x rules0B fall into this category.  They probably would not have cause this E problem on V7.2 systems, but on V7.3 and up things are more sensative@C to this -- because of the introduction of automatic changes to/fromf daylight saving time.t  . Here is something that could fix your problem:D *AFTER* you have a known good backup of the pre-upgrade system disk,C delete the following files on the copy of hte system disk that you t3 will be upgrading.  (i.e., not on the backup copy.)n  0 $!              SYS$SYSTEM:SYS$TIMEZONE_SRC.DAT.+ $!              SYS$SYSTEM:SYS$TIMEZONE.DATp0 $!              SYS$STARTUP:DTSS$UTC_STARTUP.COM/ $!              SYS$STARTUP:TDF$UTC_STARTUP.COMw  A Not all of these may exist so don't worry if you don't find them.dB BE SURE TO LOOK in commond (e.g. [VMS$COMMON.SYSEXE]) and specific? (e.g. [SYSx.SYSEXE]) directories.  *NONE* of these files may berC in specific directories!  OpenVMS Clusters do not support differentu time zones on different nodes.  E If these files are not found during the upgrade, you will be prompted  to re-set the timezone.o  I NOTE: Anyone using DTSS will have to re-set the DTSS timezone informationo if the files above are deleted!m               --  J       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------   Date: 9 Jun 03 09:20:19 +0200s) From: p_sture@elias.decus.ch (Paul Sture)t, Subject: Re: Here's a commercial for you all) Message-ID: <+mRSKmS4B+qr@elias.decus.ch>i  m In article <6HuEa.23146$3Sm.7291@news01.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes:e > < > "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message >>? >> Whereas _I_ would guess that for the 911 industry, uptime isi >> everything. >  > F > Alan and Larry are both correct.....And in earthquake-prone areas ofA > high population density, one might include the phrase "disasteri> > tolerance" in the equation as an absolute requirement in any4 > reasonable RFP for a system/application like this. >   F Just discussed this over coffee, and it was mentioned that HP is based@ in Cupertino and that is not a long way from an earthquake-prone region.d  8 Can anyone confirm this? If true, it does seem relevant.   -- 2
 Paul Sture   ------------------------------  # Date: Mon, 09 Jun 2003 10:17:15 GMTlL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr"), Subject: Re: Here's a commercial for you all6 Message-ID: <00A211BE.17263A7D@SSRL.SLAC.STANFORD.EDU>  U In article <+mRSKmS4B+qr@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes: n >In article <6HuEa.23146$3Sm.7291@news01.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes: >> e= >> "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in messagee >>>p@ >>> Whereas _I_ would guess that for the 911 industry, uptime is >>> everything.  >> h >> yG >> Alan and Larry are both correct.....And in earthquake-prone areas ofmB >> high population density, one might include the phrase "disaster? >> tolerance" in the equation as an absolute requirement in anye5 >> reasonable RFP for a system/application like this.. >> w >RG >Just discussed this over coffee, and it was mentioned that HP is basedIA >in Cupertino and that is not a long way from an earthquake-pronec >region. >.9 >Can anyone confirm this? If true, it does seem relevant.c  J HP is all over the San Francisco Peninsula.  HP Labs on Deer Creek road isI not very far from a big fault.  The SF Peninsula is definitely earthquakeBJ country, although it happened to be much more lightly hit than Santa Cruz,, San Francisco, or Oakland in the 1989 quake.   -- Alanh -- AO ===============================================================================i0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056tM  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025aO ===============================================================================-   ------------------------------  # Date: Mon, 09 Jun 2003 14:07:47 GMTE# From: "John Smith" <a@nonymous.com>p, Subject: Re: Here's a commercial for you allG Message-ID: <Tw0Fa.36080$3Sm.1781@news01.bloor.is.net.cable.rogers.com>   2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageF news:FD827B33AB0D9C4E92EACEEFEE2BA2FB05891C@tayexc19.americas.cpqcorp. net... > -----Original Message-----* > From: John Smith [mailto:a@nonymous.com] > Sent: June 7, 2003 7:38 PM > To: Info-VAX@Mvb.Saic.Comf. > Subject: Re: Here's a commercial for you all >r >  >s< > "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message/ > news:SpqJK7AkPQQG@eisner.encompasserve.org... : > > In article <00A20FDD.6AA74389@SSRL.SLAC.STANFORD.EDU>,B > winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")e	 > writes:d? > > > In article <3EE12288.6010404@tsoft-inc.com>, David Froble  > <davef@tsoft-inc.com> writes:E > > >>Lurker at Large wrote: > > >> > > >>: > > >>> The reason our product has the reputation it does?! > > >>> ** VMS ** and VMScluster!m > > >> > > >>> > > >>I figured I knew where this was going.  Then you totally	 surprised0
 > me.  TheB > > >>usual reason users want a particular product is features and > capability, notFB > > >>necessarily in that order.  I was thinking that your product > server the needs ofw< > > >>that industry much better than any other, not what you
 indicated. > > >>C > > >>So tell me, without a surperior product, and people, (I don'tr > know, you didn't? > > >>say), why would VMS and VMScluster make the users so mucho happier? > > >  > > >aF > > > I'm just guessing here that ease of management (and fewer people > devoted toD > > > system administration) and high availability over  roughly the
 > same set of.D > > > features and benefits are of great interest to taxpayer-funded 911>
 > centers,F > > > and that these are supplied by the choice of VMS and VMSclusters
 > rather thant > > > Windows. > >n@ > > Whereas _I_ would guess that for the 911 industry, uptime is
 > everything.  >  > = > Alan and Larry are both correct.....And in earthquake-proney9 > areas of high population density, one might include thep< > phrase "disaster tolerance" in the equation as an absoluteA > requirement in any reasonable RFP for a system/application liket this.. >  >m   John -  E I would add the word "proven" disaster tolerant solution to your lastoE statement. RFP responses should be able to provide testimonials wheredC the recommended solution has responded in a disaster scenario as ito wasi designed to.  ! As an example: (9/11 testimonial)aF http://h71000.www7.hp.com/openvms/brochures/commerzbank/commerzbank.pd fI   ...o   Kerry,  # Good choice of the additional word.e  C Now I have a suggestion for HP.....advertise the living s**t out ofr this.   , HP should ask itself the following question:8 Why do individuals and corporations call their insurance/ agants/brokers and purchase insurance policies?s  F The short answer is that those individuals and corporations deem it toF be financially prudent to spend a small amount of money for protection/ from potentially catastrophic financial losses.   B Why isn't VMS advertised this way? (I know that answer...VMS isn't advertised in any way)  A HP should team up with Prudential, or AIG, or Factory Mutual,  oriA anyone you can think of and do a joint marketing campaign...we'll E reduce your business insurance premiums on your business interruptionb= coverage if you run OpenVMS clusters. It's the same deal thatfF insurance companies offer residential customers a 10% discount if theyF have a monitored home alarm system. Why wouldn't they do the same withE business interruption coverage if it could save the insurance companyiA $1B annually overall (to pick a number out of the air) in claims?e  E Q. How many people in HP's Marketing Department inaction have come upm; with ideas like this? It may not be the best idea but it isv 'out-of-the-box' thinking.   ------------------------------  % Date: Mon, 09 Jun 2003 12:52:06 -0400-* From: JF Mezei <jfmezei.spamnot@istop.com>, Subject: Re: Here's a commercial for you all) Message-ID: <3EE4BB36.A676E57D@istop.com>    Paul Sture wrote:2H > Just discussed this over coffee, and it was mentioned that HP is basedB > in Cupertino and that is not a long way from an earthquake-prone	 > region.   4 It depends on where its true base of operations are.  L For instance, many banks in canada are officially based n Montreal, but theyQ only have a tokem regional presence here with all the real stuff done in Toronto.   C Also, I would suspect that a company the size of HP would be fairlyeK decentralised. So wiping out its head office in Kalifornia wouldn't troubleaN operations in Canada, Europe etc etc who are likely to have their own payroll,! accounting that meets local laws.e   ------------------------------  % Date: Mon, 09 Jun 2003 10:14:55 -0700 & From: Greg Cagle <gregc@gregcagle.com>, Subject: Re: Here's a commercial for you all/ Message-ID: <ve9g68iu06dmc2@corp.supernews.com>    JF Mezei wrote:h   > Paul Sture wrote:u > H >>Just discussed this over coffee, and it was mentioned that HP is basedB >>in Cupertino and that is not a long way from an earthquake-prone	 >>region.e >  > 6 > It depends on where its true base of operations are. > N > For instance, many banks in canada are officially based n Montreal, but theyS > only have a tokem regional presence here with all the real stuff done in Toronto.a > E > Also, I would suspect that a company the size of HP would be fairlynM > decentralised. So wiping out its head office in Kalifornia wouldn't troubleUP > operations in Canada, Europe etc etc who are likely to have their own payroll,# > accounting that meets local laws.l  F This is correct. Off the top of my head, major HP sites in the US and  Canada pre-merger were:n  G Palo Alto, CA (corporate HQ and HP labs) This *is* where the main data i center is, though.  
 Cupertino, CA  Mountain View, CAb
 Roseville, CA 
 Corvallis, OR 
 Vancouver, WAo Fort Collins, CO7 Atlanta, GA (where most of the IS functions are hosted)i Torontoo  A Add acquired Compaq sites as necessary, and plenty of other sites 0 worldwide and a few I've probably forgotten 8^).   - Greg -- o
 Greg Cagle gregc at gregcagle dot com   ------------------------------  # Date: Mon, 09 Jun 2003 15:58:19 GMTi( From: "Amy Lewis" <amylewis@pacbell.net>- Subject: How get OS on totally blank machine?-> Message-ID: <v82Fa.496$3m.52591807@newssvr13.news.prodigy.com>  I I have a VAXstation 4000, the system disk of which went bad, and I had tooL have it replaced, and it's totally blank, perhaps not even formatted. I haveJ a SCSI CD drive on it, and VMS 7.3 on CD, but don't know how to get the OS on the machine.i  I I know you have to go into SYSBOOT and point it at the CD drive, but it's K been so long since I did that I don't remember how. Any hints would be muchu appreciated.   Amyy   ------------------------------  % Date: Mon, 09 Jun 2003 11:14:45 -0500i, From: Bob Blunt <robert.blunt@hp.nospam.com>1 Subject: Re: How get OS on totally blank machine?e* Message-ID: <3ee4b6af@usenet01.boi.hp.com>   Amy Lewis wrote:K > I have a VAXstation 4000, the system disk of which went bad, and I had totN > have it replaced, and it's totally blank, perhaps not even formatted. I haveL > a SCSI CD drive on it, and VMS 7.3 on CD, but don't know how to get the OS > on the machine.t > K > I know you have to go into SYSBOOT and point it at the CD drive, but it'snM > been so long since I did that I don't remember how. Any hints would be much- > appreciated. >  > Amyu >  >   B Amy, the replacement disk "shows up" at the console when you do a:   SHOW DEVICEn  H at the console, correct?  If you have the OpenVMS/VAX V7.3 distribution F CD, and know the device name of your CD reader, you should be able to G boot from that CD with little trouble.  For example, IF your CD reader a  is DKA400, the command would be:   B DKA400  G The system SHOULD boot Standalone BACKUP and should eventually present eI you with a DCL prompt.  It MAY, during the device configuration process,  H present you a list of devices and ask if they're all configured.  Given G that both your CD reader and the destination disk show up in the list,  E you should be able to proceed with loading the core of the O/S.  For  E example, lets say that your new system disk will be DKA0, the BACKUP   commands would be:  ) $ BACKUP/IMAGE DKA400:VMS073.B/SAVE DKA0:   G When that command completes, you SHOULD be able to boot from DKA0 with y the following:   B DKA0  @ This SHOULD boot you into the start of the OpenVMS installation I procedure for the O/S.  From there, you'll answer a series of questions. oH   When those are completed the installation will proceed and move files $ around on the destination disk, etc.   As usual, YMMV...1   bob$   ------------------------------  # Date: Mon, 09 Jun 2003 10:28:46 GMTr0 From: "Ronald Milen" <milenronald@earthlink.net>V Subject: Re: How to compile/link Fortran for 'portable' executable to down-level OVMS?E Message-ID: <yjZEa.41636$Io.3643415@newsread2.prod.itd.earthlink.net>a   Hi,?  L     OpenVMS, like most OS, is not backward compatible.  There is a chance (IH haven't tried it) that a program compiled and linked on V7.2 will run onL V6.2, but most likely it won't.  Your best bet would be to adhere to compile & link it under V6.2.i  	 Ron Mileni milenronald@yahoo.comt  1 "rlfitch" <rlfitch@peakpeak.com> wrote in messagee) news:000d01c32e09$e6770910$0100a8c0@pj...rA > Need help in creating a 'portable' executable where the programrG > development is in Fortran on OVMS 7.2 with F90 compiler and then portSJ > (copy) the executable to OVMS 6.2 that has F77 compiler.  I believe thatF > the key is in the linking (rather than in the compile) but not sure. > 	 > Thanks,y > Ransom Fitch >h   ------------------------------  % Date: Mon, 09 Jun 2003 14:56:53 +0100EO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>r+ Subject: Re: HP #1 in Servers and Notebookss0 Message-ID: <bc23nf$t25$1@new-usenet.uk.sun.com>   Keith Parris wrote:  > From HP World News:r > ---h* > IDC: HP takes no. 1 server spot from IBM > P > HP took the top spot in worldwide server computer marketshare based on revenueO > during the first quarter from IBM, according to market research firm IDC.  HPMQ > had 27.9 percent of the market in the first quarter, 2.4 percentage points moret > than IBM.r6 > http://biz.yahoo.com/rc/030530/tech_computers_2.html > ---W  > HP still leads notebook market > / > http://zdnet.com.com/2100-1103_2-1012131.htmlc% Sadly they just shot back down again.n  = According to Gartner IBM is the largest Server vendor not HP.   > Also HP's numbers have declined faster than Gartner originally thought.  D Seems that the origional HP numbers were estimates (HP publish their> quarterly results a month later than Sun IBM etc). Gartner ran? with an initial figure that was close to the IDC one, HP didn't ? correct them, now that Gartner have taken HP's actual quarterly & numbers the story is rather different.   Regards. Andrew Harrisonr   ------------------------------  % Date: Mon, 09 Jun 2003 12:25:31 +0100rO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>a8 Subject: Re: Intel 64 bit Pentium seems to be on its way0 Message-ID: <bc1qrl$q9u$1@new-usenet.uk.sun.com>   Rick Jones wrote: R > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote: >  >>Rick Jones wrote:u >>C >>>Are we talking about a 64-bit Intel x86 compiler, or a differenti >>>compiler? >> > E >>There are plenty of other compilers, being done for x86-64 GCC, NAG- >>Portland, Microsoft. >  > e >>http://msdn.microsoft.com/library/default.asp?url=/library/en-us/kmarch/hh/kmarch/64bitamd_65nr.aspt >  > E > Thanks.  It will be interesting to see numbers from those compilers- > once they are published. >  > / >>>>20% would be a reasonable initial estimate.  >>>  >>>yB >>>Out of curiousity, how do you arrive at that estimate?  By whatE >>>quantity will cache pressure increase in the SPECfp2000 suite whenoH >>>going LP64 (Linux) or P64 (Windows) versus the gains from having more? >>>registers in 64-bit mode in the Opteron than in 32-bit mode?s >>>i >> > D >>Its not mine its AMD's for FP and I would cut them some slack if I@ >>was you, they were accurate for the 32bit numbers even if they* >>missed the initial launch speed of 2GHz. >  > H > All I was requesting was a reference to the methodology used to arriveE > at the estimate.  I arrived at the conclusion that the estimate wasnH > yours based on your wording.  Had you said instead "AMD estimates thatH > 64-bit compilation will increase their SPEC score by 20%" I would have? > instead asked for a pointer to a URL with that stuff in it :)g >   ; Reading the SPEC for the CPU would also help. When comparedg8 with x86, x86-64 has number of things that will make the< compiler writers task easier and should produce faster code.  C It has 4x more general purpose registers than x86 reducing register 4 pressure, it also has a large register file as well.   Regards  Andrew Harrison0   ------------------------------  % Date: Mon, 09 Jun 2003 16:12:41 +0100 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>l8 Subject: Re: Intel 64 bit Pentium seems to be on its way0 Message-ID: <bc285j$1br$1@new-usenet.uk.sun.com>   Rob Young wrote:X > In article <6JUN200303192648@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl Perkins) writes: > 0 >>"Jack Peacock" <peacock@simconv.com> writes...G >>}"Fred Kleinsorge" <my-last-name@stardotzko.dec.com> wrote in messagee >  > + >>There are also a variety of other points.  >>G >>One is that the IA64's EPIC architecure requires that the compiler dotH >>a lot of work that it doesn't have to do on the Opteron. Who, exactly,E >>produces an IA64 compiler that does a good job producing code for a  >>10,000 processor system? o >  > B > 	The Cray Opteron machine is a collection of nodes, not a 10000 > > 	processor system.  CPU counts that high, it won't be an SSI > 	regardless of vendor/CPU. > @ > 	IA64 compilers will be better.  Intel has many more resources? > 	at their disposal, especially with the addition of the Alpha  > 	compiler writers. >    Will they be better ?g  * Or are they just more difficult to write ?  6 Opteron compilers leverage x86 32 bit compilers and in5 fact make it easier for a compiler writer rather thane harder.   3 No one in their right mind would suggest that IA-64e0 compilers can either leverage x86 or are easy to write.  5 You are making the mistake of assuming that Intel has2+ a surplus of resources for the job in hand.s  ; Just consider one fact thats driving the amount of resourceu7 Intel needs to drive ISV acceptance for IA-64, Itaniums 8 complete incompatibility with any other Intel processor.  ; Intel have a large ISV portfolio on x86 mainly because each 8 new Intel processor was very largely compatible with the6 previous generation. If anything broke it was probably Microsofts fault.e  : Intel never had to go out and get ISV design wins' to help: people port, optimise and test their application, now they2 do with a vengence. That is one of the things they8 are spending money on and its also one of the areas that) isn't going well hence FX!32 for Itanium.f  7 AMD have the huge luxury of being able to say to ISV's,e6 recompile the app in your product suite that will make6 use of 64bit. Taking Oracle for example that would the3 RDBMS, but leave the rest, Oracle Forms etc etc etce+ as x86 apps don't bother porting to 64 bit.   7 How many Oracle apps are there 300-400, AMD need Oraclet5 to port the RDBMS, Intel the whole lot. AMD and their 7 partners have to support Oracle in this process so doeso5 Intel, it doesn't take a genius to realise that Intelr! will be doing a lot more support.t   Regards  Andrew Harrison 	 > 				Robd >    ------------------------------   Date: 9 Jun 2003 11:05:17 -0500 + From: young_r@encompasserve.org (Rob Young) 8 Subject: Re: Intel 64 bit Pentium seems to be on its way3 Message-ID: <w13OdyNAPfDh@eisner.encompasserve.org>s   In article <bc285j$1br$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > Rob Young wrote:   >> u >> oC >> 	The Cray Opteron machine is a collection of nodes, not a 10000  ? >> 	processor system.  CPU counts that high, it won't be an SSI  >> 	regardless of vendor/CPU.r >> nA >> 	IA64 compilers will be better.  Intel has many more resourcesc@ >> 	at their disposal, especially with the addition of the Alpha >> 	compiler writers.w >>   >  > Will they be better ?  >   A 	Being subjective, better - yes.  IA64 will run most applicationsu 	faster than Opteron.l  , > Or are they just more difficult to write ?  1 	Surely more difficult to write an IA64 compiler.p   > 8 > Opteron compilers leverage x86 32 bit compilers and in7 > fact make it easier for a compiler writer rather thann	 > harder.- >   ) 	True.  But compilers are a narrow field.   5 > No one in their right mind would suggest that IA-64y2 > compilers can either leverage x86 or are easy to > write.   	Probably not easier.-   > 7 > You are making the mistake of assuming that Intel hasa- > a surplus of resources for the job in hand.  >   7 	I didn't use the word surplus.  I said more resources.   = > Just consider one fact thats driving the amount of resource.9 > Intel needs to drive ISV acceptance for IA-64, Itaniumst: > complete incompatibility with any other Intel processor.  ; 	Money makes things happen.  Their porting program is fullyl? 	funded or it appears that way.  They have $12 billion in cash.cB 	Maybe they need to throw a few hundred million more to grease the 	wheels or keep them greased..   > 9 > AMD have the huge luxury of being able to say to ISV's, 8 > recompile the app in your product suite that will make8 > use of 64bit. Taking Oracle for example that would the5 > RDBMS, but leave the rest, Oracle Forms etc etc etcr- > as x86 apps don't bother porting to 64 bit.  >   ; 	Right.  But the portions of apps that are very x86 centrica= 	are small.  The issue is will IA64 compilers be able to take @ 	a Windows 2000 app and recompile and run with great performance: 	WITHOUT making the application writer jump through hoops.  9 > How many Oracle apps are there 300-400, AMD need Oraclei7 > to port the RDBMS, Intel the whole lot. AMD and their 9 > partners have to support Oracle in this process so doesb7 > Intel, it doesn't take a genius to realise that Intel # > will be doing a lot more support.c  = 	Oh, I wouldn't disagree.  Getting a new architecture off thel? 	ground isn't cheap.  Fortunately, Intel has very deep pockets.t   				Robo   ------------------------------  # Date: Mon, 09 Jun 2003 16:23:52 GMT-& From: jlsue <jefflsxxxz@sbcglobal.net>8 Subject: Re: Intel 64 bit Pentium seems to be on its way8 Message-ID: <12d9evs1l6kealmd31auqfrdcbbj84bfe9@4ax.com>  H On Fri, 06 Jun 2003 11:37:14 -0400, JF Mezei <jfmezei.spamnot@istop.com> wrote:   >Carl Perkins wrote:K >> Then there is the power and heat thing. What is the FP speed per watt ofL< >> power, and heat, rating of each of the IA64 and Opteron?  >.M >As I recall, Alpha was also pretty hot in its first iteration. Oh, I forgot, N >IA64 is at close to its 3rd iteration with Madison supposed to arrive "soon", >right ? >   I Er... wasn't the first version of Alpha that saw the light of day (from arH customer's purchasable system standpoint) EV3 or EV4?  And, as I recall,D those weren't all that blindingly fast for some apps compared to the fastest VAX chips at the time.   ------------------------------  % Date: Mon, 09 Jun 2003 17:41:26 +0100 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>l8 Subject: Re: Intel 64 bit Pentium seems to be on its way0 Message-ID: <bc2dc0$32g$1@new-usenet.uk.sun.com>   Rob Young wrote: > In article <bc285j$1br$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: >  >>Rob Young wrote: >  >  >>>SC >>>	The Cray Opteron machine is a collection of nodes, not a 10000 u? >>>	processor system.  CPU counts that high, it won't be an SSI  >>>	regardless of vendor/CPU.a >>>eA >>>	IA64 compilers will be better.  Intel has many more resourcesa@ >>>	at their disposal, especially with the addition of the Alpha >>>	compiler writers.  >>>r >> >>Will they be better ?i >> >  > C > 	Being subjective, better - yes.  IA64 will run most applications  > 	faster than Opteron.S  : When, this claim is currently not true. Itanium has a lead= over Opteron on FP but lags Opteron on Int which is what mosta customers care about.9     >  > , >>Or are they just more difficult to write ? >  > 3 > 	Surely more difficult to write an IA64 compiler.l >  > 8 >>Opteron compilers leverage x86 32 bit compilers and in7 >>fact make it easier for a compiler writer rather than:	 >>harder.v >> >  > + > 	True.  But compilers are a narrow field.> >  > 5 >>No one in their right mind would suggest that IA-64r2 >>compilers can either leverage x86 or are easy to >>write. >  >  > 	Probably not easier.  >  > 7 >>You are making the mistake of assuming that Intel hasO- >>a surplus of resources for the job in hand.o >> >  > 9 > 	I didn't use the word surplus.  I said more resources.  >   2 By implication you were suggesting that Intel have/ lots of engineers and AMD don't. My reply whichp3 you havn't answered points out that Intel need lotse0 of engineers and AMD don't, hence your point was/ irrelevant unless you can quantify the relative 
 requirements.0     > = >>Just consider one fact thats driving the amount of resourcei9 >>Intel needs to drive ISV acceptance for IA-64, Itaniumsm: >>complete incompatibility with any other Intel processor. >  > = > 	Money makes things happen.  Their porting program is fully A > 	funded or it appears that way.  They have $12 billion in cash. D > 	Maybe they need to throw a few hundred million more to grease the > 	wheels or keep them greased.m >   6 Does it, there is precious little sign of money having much of an impact yet.  4 4 Oracle products on Itanium 3 of which are the DBMS5 ??? and the smallest software portfolio outside linuxG on Zseries mainframes.   Hence FX!32 for Itanium. > 9 >>AMD have the huge luxury of being able to say to ISV's, 8 >>recompile the app in your product suite that will make8 >>use of 64bit. Taking Oracle for example that would the5 >>RDBMS, but leave the rest, Oracle Forms etc etc etct- >>as x86 apps don't bother porting to 64 bit.a >> >  > = > 	Right.  But the portions of apps that are very x86 centric-? > 	are small.  The issue is will IA64 compilers be able to take B > 	a Windows 2000 app and recompile and run with great performance< > 	WITHOUT making the application writer jump through hoops. >   6 You seem to have missed the point entirely. It doesn't4 matter if an app is specifically designed for x86 or3 not, the fact is if it runs on x86 now then it runss on AMD without re-compilation.  3 Only apps that benefit from 64bit need to be ported.1 and compiled for Opteron, all apps need to ported  and compiled to IA-64. > 9 >>How many Oracle apps are there 300-400, AMD need OracleK7 >>to port the RDBMS, Intel the whole lot. AMD and theirg9 >>partners have to support Oracle in this process so doese7 >>Intel, it doesn't take a genius to realise that Intel<# >>will be doing a lot more support.r >  > ? > 	Oh, I wouldn't disagree.  Getting a new architecture off thecA > 	ground isn't cheap.  Fortunately, Intel has very deep pockets.H >   3 And has never sucessfully done this before as a bigc0 bang approach like Itanium, in fact Intel have a6 track record of failing to introduce new architectures to rival their 32bit platform.   Regardss Andrew Harrisoni   ------------------------------   Date: 9 Jun 2003 12:14:38 -0500-+ From: young_r@encompasserve.org (Rob Young) 8 Subject: Re: Intel 64 bit Pentium seems to be on its way3 Message-ID: <MWWgUu7AQDoA@eisner.encompasserve.org>s   In article <bc2dc0$32g$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > Rob Young wrote: >> In article <bc285j$1br$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:u >> r >>>Rob Young wrote:y >> u >>   >>>>D >>>>	The Cray Opteron machine is a collection of nodes, not a 10000 @ >>>>	processor system.  CPU counts that high, it won't be an SSI >>>>	regardless of vendor/CPU. >>>>B >>>>	IA64 compilers will be better.  Intel has many more resourcesA >>>>	at their disposal, especially with the addition of the Alphaw >>>>	compiler writers. >>>> >>>- >>>Will they be better ? >>>0 >> : >>  D >> 	Being subjective, better - yes.  IA64 will run most applications >> 	faster than Opteron. > < > When, this claim is currently not true. Itanium has a lead? > over Opteron on FP but lags Opteron on Int which is what mostv > customers care about.V >   @ 	Sure it is.  Name ones on Opteron that are faster than Madison.  = 	Keep in mind, I'll be quoting 32 processor results also.  SonB 	a number count will be higher.  But maybe that isn't fair.  ShallD 	we just stick to 4-CPU count measurements?  Can you give me 3 weeksF 	too to add to my list?  They are no doubt saving a few for Madison's I 	debut (we see tpmC out and about but the obvious missing ones are Spec).   D 	Besides, 4 processor Opteron hardware availability is July 2003, soC 	it wouldn't be too unfair to wait until July to compare, would it?    >> M >> K8 >>>You are making the mistake of assuming that Intel has. >>>a surplus of resources for the job in hand. >>>A >> 4 >> C: >> 	I didn't use the word surplus.  I said more resources. >> a > 4 > By implication you were suggesting that Intel have1 > lots of engineers and AMD don't. My reply which 5 > you havn't answered points out that Intel need lotsu2 > of engineers and AMD don't, hence your point was1 > irrelevant unless you can quantify the relative  > requirements.  >   > 	That is your interpretation.  You are interpreting something	! 	I wasn't thinking or suggesting.    >  >>  > >>>Just consider one fact thats driving the amount of resource: >>>Intel needs to drive ISV acceptance for IA-64, Itaniums; >>>complete incompatibility with any other Intel processor.s >> u >> 1> >> 	Money makes things happen.  Their porting program is fullyB >> 	funded or it appears that way.  They have $12 billion in cash.E >> 	Maybe they need to throw a few hundred million more to grease thei  >> 	wheels or keep them greased. >> 0 > 8 > Does it, there is precious little sign of money having > much of an impact yet. >   ; 	Things are just getting started.  IA64 has had a very slow7! 	rollout and bumps along the way.7  : >>>AMD have the huge luxury of being able to say to ISV's,9 >>>recompile the app in your product suite that will make 9 >>>use of 64bit. Taking Oracle for example that would theg6 >>>RDBMS, but leave the rest, Oracle Forms etc etc etc. >>>as x86 apps don't bother porting to 64 bit. >>>v >>   >>  > >> 	Right.  But the portions of apps that are very x86 centric@ >> 	are small.  The issue is will IA64 compilers be able to takeC >> 	a Windows 2000 app and recompile and run with great performance:= >> 	WITHOUT making the application writer jump through hoops.w >> 2 > 8 > You seem to have missed the point entirely. It doesn't6 > matter if an app is specifically designed for x86 or5 > not, the fact is if it runs on x86 now then it runsn  > on AMD without re-compilation. >   9 	And it runs on Madison quite well without recompilation.   5 > Only apps that benefit from 64bit need to be portedi3 > and compiled for Opteron, all apps need to portedm > and compiled to IA-64.  > 	Not true.  As an example, surely they don't need to recompile< 	that calculator that ships with Windows.  It doesn't matterA 	how fast that runs.  There is an entire subclass that won't needhD 	to be recompiled.  I'm running VMS 7.3-1, 10+ years after the fact,? 	of VMS going from VAX to Alpha.  10+ years out there are stillt< 	a few VMS apps that run in "emulation" (those in the know, " 	apps that are VESTed VAX->Alpha):   Directory SYS$COMMON:[SYSEXE]t  ( DBLMSGMGR_TV.EXE;1  EDF_TV.EXE;1        $ MONITOR_TV.EXE;1    TECO32_TV.EXE;1   ? 	Secondly, Intel would be foolish if they think that IA64 couldeC 	be adopted without world-class compilers for Windows.  Surely, theiA 	compile/test cycle going from x86 Windows to Windows64 is fairly , 	clean.  It was that way going VAX to Alpha.   >> s: >>>How many Oracle apps are there 300-400, AMD need Oracle8 >>>to port the RDBMS, Intel the whole lot. AMD and their: >>>partners have to support Oracle in this process so does8 >>>Intel, it doesn't take a genius to realise that Intel$ >>>will be doing a lot more support. >>   >>  @ >> 	Oh, I wouldn't disagree.  Getting a new architecture off theB >> 	ground isn't cheap.  Fortunately, Intel has very deep pockets. >> . > 5 > And has never sucessfully done this before as a bigd2 > bang approach like Itanium, in fact Intel have a8 > track record of failing to introduce new architectures  > to rival their 32bit platform. >   = 	But many others have and your assumption is that Intel won'tp> 	get it right this time.  They have far too much riding on it,> 	far too much money in their pockets for it to fail.  However,> 	only time will tell how it plays out.  My money is on the big5 	money player Intel.  AMD is running out resources.     C 	Finally, - as a telling point - AMD can't even get Sun (an obvious D 	Intel rival) to adopt Opteron for their x86 servers.  If they can'tE 	get the best fit in the business "on board that Opteron juggernaut",m@ 	how could one presume Opteron is going to amount to little more 	than a niche product?   				RobR   ------------------------------  $ Date: Mon, 9 Jun 2003 19:52:14 +10002 From: "Alan Erskine" <alanerskine@optusnet.com.au>N Subject: Re: JF Mezei "flamed, insulted, denigraded(sic)"!  Oh the humanity!!!< Message-ID: <3ee45893$0$21555$afc38c87@news.optusnet.com.au>  : "James Oberg" <jamesoberg@houston.rr.com> wrote in message5 news:ICTEa.14728$lT4.1786771@twister.austin.rr.com...l >oJ > To harvest the full value of this forum, I don't want anybody flamed forJ > their opinions -- I want to see as wide and wild an array of opinions asJ > possible. But I have no issue with people getting criticized -- not justJ > corrected, but personally criticized -- for bad facts and logic, because? > that seems to me a betrayel of the standards that we need fore
 consideration J > of opinions that differ from our own. Now, I don't want to argue, I justB > want to register my opinion that Mezei's views are as subject toI > argumentation as anyone's, but the factual material he has presented issG > trustworthy enough -- and valuable enough -- to pass my filter, or at4 least,; > not set off any alarms, and I am grateful he shares them.I  8 I concur.  And I miss his information about the shuttle.   -- Alan Erskine alanerskine(at)optusnet.com.au Where's the WMD Mr Bush?   ------------------------------  # Date: Mon, 09 Jun 2003 10:31:59 GMTs> From: "Gregory Morrow" <gregoryDOTHEREmorrow@worldnet.att.net>N Subject: Re: JF Mezei "flamed, insulted, denigraded(sic)"!  Oh the humanity!!!J Message-ID: <zmZEa.195443$ja4.10361509@bgtnsc05-news.ops.worldnet.att.net>   Alan Erskine wrote:s  < > "James Oberg" <jamesoberg@houston.rr.com> wrote in message7 > news:ICTEa.14728$lT4.1786771@twister.austin.rr.com...o > >oL > > To harvest the full value of this forum, I don't want anybody flamed forL > > their opinions -- I want to see as wide and wild an array of opinions asL > > possible. But I have no issue with people getting criticized -- not justL > > corrected, but personally criticized -- for bad facts and logic, becauseA > > that seems to me a betrayel of the standards that we need foru > considerationiL > > of opinions that differ from our own. Now, I don't want to argue, I justD > > want to register my opinion that Mezei's views are as subject toK > > argumentation as anyone's, but the factual material he has presented iseI > > trustworthy enough -- and valuable enough -- to pass my filter, or atr > least,= > > not set off any alarms, and I am grateful he shares them.  >=: > I concur.  And I miss his information about the shuttle.    6 Anyone that "supports" JF Mezei is obviously a cretin.  C He is on the same low level as  those infamous trolling timewastersu8 atlsvo@webtv.com ", "Mr. Travel", "Mary/Roada", etc.....   So <PLONK> to you, too, Sir :-).   -- Best Greg   ------------------------------  # Date: Mon, 09 Jun 2003 17:09:46 GMTA9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>eL Subject: Re: Letter to HP Management from OpenVMS users around the world ...2 Message-ID: <ub3Fa.2185$hG7.1706@news.cpqcorp.net>  5 "Bill Gunshannon" <bill@cs.uofs.edu> wrote in message / news:balabg$124mh$1@ID-135708.news.dfncis.de...p5 > In article <Cgspal8ZeogX@eisner.encompasserve.org>,m? > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:N > >EL > >    McDonalds makes almost no profit on hamburgers, but sells them by the > >    billions. > J > That's a joke, right?  McDonalds makes a fortune selling hamburgers.  AsJ > a matter of fact, they make a fortune selling all the garbage they sell.J > Everything they sell has a pricetag several times the actual cost of the; > materials or the miinimum-wage labor it takes to make it.   J This may or may not be true.  In terms of margins, they probably make more9 on the soda, shakes and fries, than on the burger itself.   H Consider the fact that the *cup* for the drink costs more than the drink itself in most cases.h   ------------------------------   Date: 9 Jun 2003 18:50:52 +0200 # From: Tsk, Tsk <tsktsk@shameon.you>r1 Subject: Re: Mezeing, and Mezei's severe problemsn9 Message-ID: <L9QNCTVO37781.7848726852@Gilgamesh-frog.org>   2 Bob Myers <nospamplease@addressinvalid.com> wrote:   > 3 >"none" <none@null.dev> (JF Mezei) wrote in messageo" >news:3EE397CD.2CF7486@null.dev... >> re: navigationi >># >> Compaq is useless in the arctic.  >e9 >Assuming you meant "comass" and not a computer company -dF >sure, but how much of airline navigation do you think is primarily by >wet compass anyway? >hG >If it gets to the point where the pilot of a 777 is figuring out where9= >to go primarily by watching compass and wristwatch, I humblyy= >suggest that you have rather severe problems no matter where  >you are...:-)  H Well, you're certainly right about his severe problems, but why the hell are you Mezeing(*)???m  C * Mezeing - the practice of arguing with or engaging in any type ofmK communication with JF Mezei or any other idiot and retard on usenet.  It isr1 one of the worst possible breaches of Netiquette.p   ------------------------------  $ Date: Mon, 9 Jun 2003 11:42:37 -0400, From: "warren sander" <warren.sander@hp.com>2 Subject: Re: Motif 1.3, Euro sign not standard ?!?, Message-ID: <3ee4af26$1@usenet01.boi.hp.com>  E check out the euro site from 1999.. http://h71000.www7.hp.com/euro or - http://h71000.www7.hp.com/euro/decw_euro.htmle      7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messagei# news:3EE38E44.E4D9111A@istop.com...cK > Considering Motif is now at 2.1, does version 1.3 predate the creation ofi the E > Euro ? Perhaps this is why it isn't part of it, and must be enabledo
 separately > as an add-on.p   ------------------------------  % Date: Mon, 09 Jun 2003 19:02:45 +0200i From: Dirk Munk <munk@home.nl>2 Subject: Re: Motif 1.3, Euro sign not standard ?!?2 Message-ID: <bc2ess$m8k$1@news2.tilbu1.nb.home.nl>  P The problem seems to be that the ascii value of the  (euro if you can read it) K is not standard on all platforms. Perhaps we have to wait for 16 bit ascii   before everything gets normal.   warren sander wrote:G > check out the euro site from 1999.. http://h71000.www7.hp.com/euro ort/ > http://h71000.www7.hp.com/euro/decw_euro.htmlm >  >  > 9 > "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messageo% > news:3EE38E44.E4D9111A@istop.com...c > K >>Considering Motif is now at 2.1, does version 1.3 predate the creation of  >  > the  > E >>Euro ? Perhaps this is why it isn't part of it, and must be enabled  >  > separately >  >>as an add-on.t >  >  >    ------------------------------  # Date: Mon, 09 Jun 2003 12:41:20 GMTu6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) Subject: Re: MSA1000 vs EVA3000 5 Message-ID: <Qf%Ea.112280$lL2.1025583@news.chello.at>i  ] In article <3ee25592$1@news.wineasy.se>, "Lars Holmstrm" <lars.holmstrom@flysta.net> writes: K >Can anyone share ideas, pros, cons for going for an MSA1000 or the EVA3000  >platform ? G >We will try to build a multi site OVMS Cluster with GBE and FC as sitey
 >inteconnect.a> >I am specially concerned about the shadowing for the MSA1000.  H As I recently read, is official FC-AL Support for OpenVMS "still coming"H (for more than 18months now - it seems latent support is now in, but notG really supported) while FC-AL (which the MSA1000 - as I know it - uses)sI is on its way out, making place for switched FC (which VMS supported longnF before) cause of the (performance/stability) problems with FC-AL (alsoD in the HSV - they now seem to get all FC-Switches as an upgrade ;-).  6 Sorry, this is all hearsay, no real experience here...   -- 9 Peter "EPLAN" LANGSTOEGERc% Network and OpenVMS system specialiste E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  # Date: Mon, 09 Jun 2003 13:43:55 GMT < From: "John E. Malmberg" <Malmberg@dskwld.zko.dec.compaq.hp> Subject: Re: MSA1000 vs EVA3000a2 Message-ID: <va0Fa.2173$Vo7.1483@news.cpqcorp.net>    Peter 'EPLAN' LANGSTOEGER wrote:* > In article <3ee25592$1@news.wineasy.se>,7    "Lars Holmstrm" <lars.holmstrom@flysta.net> writes:o > L >>Can anyone share ideas, pros, cons for going for an MSA1000 or the EVA3000 >>platform ?  I The EVA series is more expandable, and has been tested in larger cluster l  configurations than the MSA1000.  C The MSA1000 would be more for your smaller configurations and SANS.e  H >>We will try to build a multi site OVMS Cluster with GBE and FC as site >>inteconnect.? >>I am specially concerned about the shadowing for the MSA1000.n  G There should be no problem with using Host Based Volume Shadowing with O2 the MSA1000 or an EVA series Fibre Channel system.  J > As I recently read, is official FC-AL Support for OpenVMS "still coming"J > (for more than 18months now - it seems latent support is now in, but notI > really supported) while FC-AL (which the MSA1000 - as I know it - uses) K > is on its way out, making place for switched FC (which VMS supported long>H > before) cause of the (performance/stability) problems with FC-AL (alsoF > in the HSV - they now seem to get all FC-Switches as an upgrade ;-).  H FC-AL has many issues that need to be addressed.  I can not say when it H will be ready for official support, but that support will be limited to  specific tested configurations.w  I Fibre Channel switches are coming down in price.  FC-AL still has a cost i) advantage for some low end configuration.o   -JohnI! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Onlyp   ------------------------------   Date: 9 Jun 2003 07:23:26 -0700 1 From: nothome@spammers.are.scum (Malcolm Dunnett)l Subject: Re: MSA1000 vs EVA3000H- Message-ID: <8elTRgmZyo$P@malvm7.mala.bc.ca.>D  3 In article <va0Fa.2173$Vo7.1483@news.cpqcorp.net>,  B     "John E. Malmberg" <Malmberg@dskwld.zko.dec.compaq.hp> writes: > K > Fibre Channel switches are coming down in price.  FC-AL still has a cost  + > advantage for some low end configuration.  > A    Is the DSGGB ( aka Silkworm ) the only FC switch that VMS willyB work with? Even used they still seem to be in the $10k price zone.D I often see the "Compaq Storageworks 16 port SAN Switch (127553-B21) on EBay for $1000 or so.   > -John # > malmberg@dskwld.zko.dec.compaq.hps > Personal Opinion Onlyh >    ------------------------------  # Date: Mon, 09 Jun 2003 17:30:06 GMTn& From: jlsue <jefflsxxxz@sbcglobal.net> Subject: Re: MSA1000 vs EVA3000r8 Message-ID: <dqg9evklslbbhafhjuau51icu7dko367in@4ax.com>  J On 7 Jun 2003 22:13:54 +0100, "Lars Holmstrm" <lars.holmstrom@flysta.net> wrote:  K >Can anyone share ideas, pros, cons for going for an MSA1000 or the EVA3000o >platform ?pG >We will try to build a multi site OVMS Cluster with GBE and FC as site 
 >inteconnect. > >I am specially concerned about the shadowing for the MSA1000. >/Lars >e   MSA1000n  
 Pros:	Cheapert 	Up to 3 TBo  & Cons:	Limited performance capabilities7 		(i.e., SCSI-bus drives limit spindle counts for LUNs)e   EVAm   Pros:	Virtual RAID+ 	I/O spread over more spindles for all LUNsA+ 	-  I/O Peformance capabilites much greatero3 	-  I/O Performance management simpler (can utilize - 		fewer luns, less need to partition database  		tablespaces, etc.)  	Snapshot/Snapclone capabilities 	Greater Expandability 	Fibre drives instead of SCSIs  A Cons:	Price, though the 3000 isn't that bad, it is more expensiveo
 	than the MSAi   ------------------------------  # Date: Mon, 09 Jun 2003 15:07:56 GMT 0 From: "brad.madison" <brad.madison@mail.tds.net>A Subject: Re: My IP was listed on SPEWS and I am not a spammer....., Message-ID: <3EE4A152.1F6A972B@mail.tds.net>   McWebber wrote:c > ; > "Spambo" <dj9mevb4gw0001@sneakemail.com> wrote in messagee) > news:3EE48BF2.53FD35B@sneakemail.com...  > >pA > > If you were to have a problem with your modem would you go to K > > comp.dcom.modems and call them "bastards" and tell them to "get off the E > > fucking planet" if you didn't like the answers they came up with?m > - > <Magoo>By George, I think he would.</Magoo>n  F Many people would doubtless have called the infamous Carl Lydick (RIP)D of comp.os.vms a bastard.  Carl gave sometimes marvelous answers (heG truly knew VMS extremely well) but the charge would have been, based on 0 his performance in comp.os.vms, exactly correct.  H Carl cruelly savaged a poor woman who asked how to get off "this mailingA list" for calling a newsgroup a mailing list.  The headers of theh6 woman's post showed she was on a linked  mailing list.  H Cross-posted so that if anyone in comp.os.vms discovers I'm lying he can reveal that fact here.   -- _> "For the first year, the total rcpts trapped was 281 Million."  B -- The Mushroom Guy (an operator outside the US, writing about his	 honeypot)r  . Stop spam, run Jackpot: http://jackpot.uk.net/   ------------------------------   Date: 9 Jun 2003 07:03:13 -0500c- From: Kilgallen@SpamCop.net (Larry Kilgallen)t0 Subject: Re: OT: CNN Story Not Favorable To Bush3 Message-ID: <BnqwgvMJZAdg@eisner.encompasserve.org>t  [ In article <3EE38006.9010905@bellatlantic.net>, bob smith <sfmc68@bellatlantic.net> writes:5  K > I guess we are in the same boat, I think we are obligated as citizens to /G > be critical of politicians, just as we are of company management, or a > Miss Management.  3 Fine, go do it in a newsgroup where it is relevant.    ------------------------------   Date: 9 Jun 2003 07:04:19 -0500h- From: Kilgallen@SpamCop.net (Larry Kilgallen)e0 Subject: Re: OT: CNN Story Not Favorable To Bush3 Message-ID: <pQQtvQQATZ7C@eisner.encompasserve.org>t  [ In article <3EE3B6AD.5060901@bellatlantic.net>, bob smith <sfmc68@bellatlantic.net> writes:   K > I did not and do not like clitton, he has lied to the public, lied under r > oath, and is a pervaricator.  I But he has not said anything about VMS, so the discussion does not belongo here.    ------------------------------   Date: 9 Jun 2003 09:42:35 -0500t; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)$0 Subject: Re: OT: CNN Story Not Favorable To Bush3 Message-ID: <jLNj7nmBFXuk@eisner.encompasserve.org>i  V In article <3EE39157.E54AFF93@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes: > bob smith wrote:A >> think every country and every citizen should have the right totJ >> criticize, complain aobut, or grumble about government or business.  In> >> the states, I believe it is both an obligation and a sport. >  > Think Dixie Chicks.i > N > (For those who do not know, many radio stations in the USA banned the payingM > of their songs after the Dixie Chicks became unpatriotic by criticising thenO > bush decision to invade another country). This type of behavious was repeatedi > over and over again.  H    And then others of us turned off the radio and put the Home album on.B    Still love Travelin' Soldier.  OBTW, what the Chicks did is notE    unpatriotic (as some have claimed) but they will have to live with .    thier fan's free expression of disagrement.  K > Interesting that outside the USA, it was well reported that the so callednM > evidence was bogus, so it isn't as if the information wasn't there. The USAsP > media chose not to show that side of the story. Would have been unpatriotic to# > criticise a president during war.t  H    This criticism was not hard to find before the war.  The inability ofE    US forces to find evidence of WMD in Iraq is an obvious difficultylB    that all can see.  But the need to remove Sadam by force is not:    generally held to rely soley on the possibility of WMD.   ------------------------------   Date: 9 Jun 2003 03:45:56 -0700e. From: mistdragon@zdnetonebox.com (mist dragon)" Subject: Portents of Itanium death= Message-ID: <7500353b.0306090245.7e521188@posting.google.com>t  B AMD is bring its x86-64 bit environment as often indicated in thisA forum. I've been looking its specs and porting issues lately. The4A issues on porting seem surprisingly small compared to issues withoE Itanium. Of course the whole architecture is different and porting ofnE anything at all to Itanium is a considrable effort because everything F from compilers up needs special attention as the optimization is movedE to compilers and in AMD64 64 bit mode is an extension to 32 bit mode.i  B It looks like that in the 2003H2 PC market will start moving to 64E bits as AMD agressively pushes it to mainstream. While to most of the>D common users its all the same if the architecture 32bit or 64bits, I@ believe that with good marketing and pricing 32bits will soon beB 'old-fashioned'. Intel will have difficult time proving that their+ 32bit is actually better than AMD's 64 bit.   F Rumour is that Intel has Yamhill-64 bit processor in their pocket. ButB putting that to market could kill Itanium or make it significantlyF less attractive people because why should they turn to Itanium if theyD can achieve the same 64 bits with x86-64. It looks like soon 64 bitsB cannot be the Itanium argument and Intel needs to find new ways to promote Itanium.  E Interesting situation. Which should carry more weight - technologicaloE superiority or mass market ? Knowing Intel, I would bet mass market - D remember i860, RIMM etc ? I have nagging feeling the users have beenE not as tolerant to changes as the market dominating position of InteltC would have been let to believe. Compatibility to existing x86-32 ise very important.   E Now, if this is the case, should Intel drop Itanium ? That would be a B loss of face. So any other ways ? Well, you can drop marketing andD promotion, target it to high-end or embedded only (knowing very wellC how hard it is to compete there). This is about the only forum I'vem? found in news that talks about Itanium. Has anyone seen ItaniumrE marketing in newspapers or magazines lately ? Anyone feeling de-ja-vui ?w  E Could this be a reason for not promoting OpenVMS ? If Itanium is fact D targeted EOL, does HP have enough money and willingness to switch to x32-64 ?   Interesting times...   Mist   ------------------------------   Date: 9 Jun 2003 09:10:05 -0500o+ From: young_r@encompasserve.org (Rob Young) & Subject: Re: Portents of Itanium death3 Message-ID: <xrjZXTJqY7Zp@eisner.encompasserve.org>t  n In article <7500353b.0306090245.7e521188@posting.google.com>, mistdragon@zdnetonebox.com (mist dragon) writes: > D > It looks like that in the 2003H2 PC market will start moving to 643 > bits as AMD agressively pushes it to mainstream. o  @ 	Wow!  Quick, get your resume out to the analyst firms.  Surely,@ 	you will be a rising star (tongue-in-cheek) for your prediction 	of 2H recovery for AMD.  5 	AMD was downgraded last week.  Appears that AMD will ? 	be doing GOOD if it doesn't lose anymore marketshare to Intel:u   http://www.marketwatch.com/news/yhoo/story.asp?source=blq/yhoo&siteid=yhoo&dist=yhoo&guid=%7B6FEEE9D5%2D803A%2D4BA4%2D9EB3%2D946C382EE6C9%7D  P "Shares of the semiconductor manufacturer (AMD: news, chart, profile) slumped 41O cents, or 5.6 percent, to $6.87 Monday after Dhanda sliced his rating to "sell"cO from "neutral." He said AMD would have trouble achieving its stated outlook forh2 "modest" sequential growth in the second quarter."  K Dhanda is projecting a 6 percent sequential decline in AMD's processor uniteH shipments in the second quarter, though the actual decline could be even9 greater if demand trends don't improve early this month.     	Conversely, there is Intel:  Y http://www.fool.com/news/mft/2003/mft03060603.htm?source=eptyholnk303100&logvisit=y&npu=yc  M Intel (Nasdaq: INTC) is a $144 billion company with $12.4 billion in cash ande equivalents    [snip]  L In contrast to troubled competitor Advanced Micro Devices (NYSE: AMD), IntelO said its PC processor business is trending to the high end of expectations, and M its Centrino wireless-chip launch was very satisfactory. Only flash memory is K slightly below sales hopes due to recent price increases, and communicatione products remain "soft."   K The company will report full results on July 15, when earnings of $0.13 per H share are expected -- up 44% from last year. For the year, a 20% rise inL earnings to $0.61 per share is currently the average estimate. The $22 stockL trades at 36 times the estimate. With $5 billion in trailing free cash flow,I Intel has a multiple to free cash flow of about 28, a premium to the S&P.i@ Investors are betting on much stronger cash flow in years ahead.     > While to most of theF > common users its all the same if the architecture 32bit or 64bits, IB > believe that with good marketing and pricing 32bits will soon beD > 'old-fashioned'. Intel will have difficult time proving that their- > 32bit is actually better than AMD's 64 bit.a    @ 	No.  Wall Street doesn't see it that way.  AMD's 64-bit desktopD 	PC may be mostly dead on arrival if Intel exceeds it in performance; 	and is cheaper.  That may happen.  After all, if AMD has a > 	cheaper and more powerful desktop part waiting in the wings, D 	Intel would be facing downgrade due to AMD's strong futures, right?? 	And AMD stock would be creeping up as Wall Street realizes AMDa= 	has an "Intel killer" waiting in the wings, right?  Or maybe > 	their is a killer 64-bit desktop application that will ensure% 	AMD takes off with their Clawhammer?e   				Rob    ------------------------------  % Date: Mon, 09 Jun 2003 17:14:09 +0200g$ From: Michael Unger <unger@decus.de>& Subject: Re: Portents of Itanium death5 Message-ID: <bc2b18$e62cr$1@ID-152801.news.dfncis.de>-  . On 09-Jun-2003 12:45, "mist dragon" (?) wrote:   > [...]j > G > Now, if this is the case, should Intel drop Itanium ? That would be asD > loss of face. So any other ways ? Well, you can drop marketing andF > promotion, target it to high-end or embedded only (knowing very wellE > how hard it is to compete there). This is about the only forum I've.A > found in news that talks about Itanium. Has anyone seen ItaniumaG > marketing in newspapers or magazines lately ? Anyone feeling de-ja-vu  > ?s >  > [...]i  4 Why exactly use Itanium for the *embedded* market???  F *Power-efficiency* is the main concern of embedded systems due to heatF dissipation and cooling issues. That apparently isn't Itanium's realm.   Michaele   --    @ Please do *not* send "Security Patch Notifications" or "SecurityA Updates"; this system isn't running a Micro$oft operating system.t= And don't annoy me <mailto:postmaster@[127.0.0.1]> please ;-)    ------------------------------  # Date: Mon, 09 Jun 2003 16:25:52 GMTH9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>t& Subject: Re: Portents of Itanium death1 Message-ID: <ky2Fa.2184$7u7.261@news.cpqcorp.net>r  L OK enough Misty.  What are you selling?  How about a name?  Who you shilling for?  E Itanium isn't "targeted EOL", and compatability with IA32 has no realtI bearing on the VMS discussion.  There are arguments pro and con about howtH well Opteron, and AMD will fare - and when (if ever) the general desktop- market will give a rats behind about 64-bits.A  L You've shown the axe you are grinding, so enough with your musings.  We will4 see how things shake out over the next couple years.      ; "mist dragon" <mistdragon@zdnetonebox.com> wrote in message 7 news:7500353b.0306090245.7e521188@posting.google.com...iD > AMD is bring its x86-64 bit environment as often indicated in thisC > forum. I've been looking its specs and porting issues lately. ThetC > issues on porting seem surprisingly small compared to issues witheG > Itanium. Of course the whole architecture is different and porting oftG > anything at all to Itanium is a considrable effort because everythingpH > from compilers up needs special attention as the optimization is movedG > to compilers and in AMD64 64 bit mode is an extension to 32 bit mode.  >>D > It looks like that in the 2003H2 PC market will start moving to 64G > bits as AMD agressively pushes it to mainstream. While to most of theiF > common users its all the same if the architecture 32bit or 64bits, IB > believe that with good marketing and pricing 32bits will soon beD > 'old-fashioned'. Intel will have difficult time proving that their- > 32bit is actually better than AMD's 64 bit.  >iH > Rumour is that Intel has Yamhill-64 bit processor in their pocket. ButD > putting that to market could kill Itanium or make it significantlyH > less attractive people because why should they turn to Itanium if theyF > can achieve the same 64 bits with x86-64. It looks like soon 64 bitsD > cannot be the Itanium argument and Intel needs to find new ways to > promote Itanium. >nG > Interesting situation. Which should carry more weight - technologicalrG > superiority or mass market ? Knowing Intel, I would bet mass market - F > remember i860, RIMM etc ? I have nagging feeling the users have beenG > not as tolerant to changes as the market dominating position of IntellE > would have been let to believe. Compatibility to existing x86-32 is  > very important.c >rG > Now, if this is the case, should Intel drop Itanium ? That would be aeD > loss of face. So any other ways ? Well, you can drop marketing andF > promotion, target it to high-end or embedded only (knowing very wellE > how hard it is to compete there). This is about the only forum I'vetA > found in news that talks about Itanium. Has anyone seen Itanium G > marketing in newspapers or magazines lately ? Anyone feeling de-ja-vut > ?p >iG > Could this be a reason for not promoting OpenVMS ? If Itanium is factIF > targeted EOL, does HP have enough money and willingness to switch to
 > x32-64 ? >p > Interesting times... >n > Mist   ------------------------------  % Date: Mon, 09 Jun 2003 13:05:11 -0400n* From: JF Mezei <jfmezei.spamnot@istop.com>& Subject: Re: Portents of Itanium death) Message-ID: <3EE4BE46.3583B432@istop.com>n   Rob Young wrote:> >         AMD was downgraded last week.  Appears that AMD willH >         be doing GOOD if it doesn't lose anymore marketshare to Intel:  M Have the analysts who are rating AMD provided any information on whether theyr( or their employer own a stake in Intel ?    J It is interesting that those analysts don't mention the fact that Intel isJ nowhere near recuperating the huge investment in IA64, and that there is aL risk that 64 bit on desktop would take off which woudl leave intel without a competing product.  J If one were to look at Intel's IA64 division only, would you agree that it. would definitely be a "sell" compared to AMD ?  K HP won't reveal how much Intel paid Compaq to retain the VMS engineers on acK "make work project" to port VMS to IA64. But the way things are going right-K now, I am not even sure Intel will get that money back from sales of VMS on_& IA64, at least not in he next 5 years.   ------------------------------  % Date: Mon, 09 Jun 2003 13:16:16 -0400o* From: JF Mezei <jfmezei.spamnot@istop.com>& Subject: Re: Portents of Itanium death) Message-ID: <3EE4C0DE.43974724@istop.com>a   Fred Kleinsorge wrote:G > Itanium isn't "targeted EOL", and compatability with IA32 has no realu! > bearing on the VMS discussion. l  I Yes it does, because of the reasons your bosses used to justify the Alphao% murder and moving to that IA64 thing.p  N Your bosses used keywords such as "commodity", "mass market" and "low cost" to$ justify moving to IA64 versus Alpha.  M The problem is that IA64 has so far achieved none of the qualities that CarlyhY and Curly had promised and there is nothing on the horizon showing such things happening.   K It is exactly because IA64 is not compatible with wintel platform that IA64 L will remain a low volume, proprietary and expensive chip, especially if/when8 Intel is forced to release a 64 bit version of the 8086.   ------------------------------  $ Date: Mon, 9 Jun 2003 13:33:00 -0400* From: "Bill Todd" <billtodd@metrocast.net>& Subject: Re: Portents of Itanium death2 Message-ID: <YYmdnV6cTqdJWXmjXTWcpw@metrocast.net>  D "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> wrote in message+ news:ky2Fa.2184$7u7.261@news.cpqcorp.net... E > OK enough Misty.  What are you selling?  How about a name?  Who yout shilling > for? >eG > Itanium isn't "targeted EOL", and compatability with IA32 has no realbK > bearing on the VMS discussion.  There are arguments pro and con about how J > well Opteron, and AMD will fare - and when (if ever) the general desktop/ > market will give a rats behind about 64-bits.> >uI > You've shown the axe you are grinding, so enough with your musings.  Wee will6 > see how things shake out over the next couple years.  J Hooray!  I can agree!  Just about everything, sensible and otherwise, thatL can be said on this subject already has been said, multiple times.  Now thatL Opteron has appeared it's time to sit back and see how things actually shake out.  K (Although it would be difficult to resist noting that we've been waiting toeG see how things actually shook out for over two years now in the case ofo
 Itanic...)   - bill   ------------------------------   Date: 9 Jun 2003 07:52:57 GMT / From: "Dave Weatherall" <djweath@attglobal.net> " Subject: Re: Portents of VMS death5 Message-ID: <DTiotGxQ0bj6-pn2-8EhQHesYVB9J@localhost>p  B On Tue, 3 Jun 2003 05:45:36 UTC, mistdragon@zdnetonebox.com (mist  dragon) wrote:  
 <Big Snip>  t+ > 1. VMS runs an application on platform n.c= > 2. A new VMS platform y with top-of-the-line GS is created.o- > 3. Application is tuned to top performance. E > 4. System is initialized and _only_ operating system is switched tou > Tru64.2 > 5. Same application is tuned to top performance. >   G > Conclusions: Tru64 outperforms VMS using the customers application ifhB > they buy that platform. No need to change applications, only OS. > C > Conspiracy theory: VMS staff wanted off VMS because they knew itseG > dying ? HP wanted them to move off VMS because they knew it was dyingcF > (you knew Tru64 is from HP too?) ? VMS staff was not adequate to itsG > task and knew no-one would want VMS in these days so they decided allmF > to switch to modern OS ? The staff wanted to retire after seeing too > much ? What is your theory ?  F Um. I thought Tru-64 was on the retirement list too now along with its< Alpha processor. Hasn't the port to Itanium been stopped to F 'encourage' the move to a consolidated 'HP Unix with Tru-64 features'  on Itanium?f   -- i Cheers - Dave.   ------------------------------   Date: 9 Jun 2003 07:52:59 GMT / From: "Dave Weatherall" <djweath@attglobal.net> H Subject: Re: Printer recommendation, hobbyist/decwindows/mozilla and VMS5 Message-ID: <DTiotGxQ0bj6-pn2-LPDNdCkz4WyC@localhost>t  B On Tue, 3 Jun 2003 20:29:37 UTC, jordan@ccs4vms.com (Rich Jordan)  wrote:  Z > VAXman-  @SendSpamHere.ORG wrote in message news:<00A20C93.B694FEE5@SendSpamHere.ORG>...j > > In article <cc5619f2.0306020759.4387df69@posting.google.com>, jordan@ccs4vms.com (Rich Jordan) writes: > > {...snip..}X) > > >Brian, I can't afford what you have!  > > . > > What?  My kids or the color laser printer? >  > E > Either of them, actually.   Need a cheap printer, and dogs are more  > cost effective than kids.E >  > Rich  F Not when it comes to maintaining pension schemes :-) Or maybe they are ?t   -- n Cheers - Dave.   ------------------------------  % Date: Mon, 09 Jun 2003 09:49:17 +0200c From: Dirk Munk <munk@home.nl>H Subject: Re: Printer recommendation, hobbyist/decwindows/mozilla and VMS2 Message-ID: <bc1ef3$lqu$1@news4.tilbu1.nb.home.nl>   Rich Jordan wrote:E > I have to get a printer, cheap, for home/hobby use, that can handleaH > printing rich pages from Mozilla and DECwrite, preferably also capableH > of handling plain text printouts like my current DEC ColorWriter 520icE > (which works fine from windoze for graphics).  I don't use DCPS but G > would be willing to if a Postscript-only printer is the best choice.  A > Color would be a great bonus, but photo-grade is not necessary.a > D > The printer can be parallel, through a Emulex NetQue, jetdirect orE > other network interface, or serial attached directly to the primaryt > Alpha workstation. > H > I know that the LJ5M at work is able to print Netscape 3.03 pages fromF > our weenie Alpha (which can't handle Mozilla), albeit without color,E > but I can't afford the power for a laser printer (given what has totG > run on the two available circuits, every warm up would end up poppingyE > a circuit breaker).  I've not been able to locate a Deskjet 1200 orvE > 1600 with Postscript around here to try, but thats what I'm leaningsE > towards (based on the LJ5M success), barring some feedback from theo > group.  Q I would like to have a affordable Postscript inkjet printer as well, however all -J printer manufacturers agree on a few things: Postscript Printers Shall Be C Expensive, and: Adding A Ethernet Interface Shall Double The Price.a  I Please keep in mind that no HP inkjet Postscript printers are officially $L supported in DCPS, and even stranger, are not on the list of printers to be B supported in future. I don't even know if DCPS supports color ????  M Another way of doing this is using a software Postscript RIP. Some HP inkjet mP printers can be ordered with such a program, but of course just for Windooz and  maybe Apple.  K Perhaps someone at HP would be so kind to think about their loyal hobbiest tI customers, and come up with a nice VMS driver for a HP inkjet printer ???        > @ > So, what can I look for in the <$500 range (and <$200 stronglyA > preferred, used is fine) that is usable for 'normal' DECwindows B > printing; no conversions through ghostscript, no jumping throughF > hoops?  Any brands other than HP?  Any handy interface box that fits> > between a printer and the system that converts postscript to2 > PCL-compliant graphics so I keep the DEC inkjet? > H > Personal experience with a working solution preferred!  Thanks for any > information. > & > Brian, I can't afford what you have! > 
 > Rich Jordan    ------------------------------   Date: 9 Jun 2003 10:32:38 -07001& From: jordan@ccs4vms.com (Rich Jordan)H Subject: Re: Printer recommendation, hobbyist/decwindows/mozilla and VMS< Message-ID: <cc5619f2.0306090932.5664e9e@posting.google.com>  X Dirk Munk <munk@home.nl> wrote in message news:<bc1ef3$lqu$1@news4.tilbu1.nb.home.nl>... > Rich Jordan wrote:G > > I have to get a printer, cheap, for home/hobby use, that can handlesJ > > printing rich pages from Mozilla and DECwrite, preferably also capableJ > > of handling plain text printouts like my current DEC ColorWriter 520ic3 > > (which works fine from windoze for graphics).    > S > I would like to have a affordable Postscript inkjet printer as well, however all dL > printer manufacturers agree on a few things: Postscript Printers Shall Be E > Expensive, and: Adding A Ethernet Interface Shall Double The Price.e >   = Looking at used 1200 and 1600 Deskjets on Ebay.  Depending ontD interfaces installed, I have an Emulex NetQue and several DECservers available for interfacing.  K > Please keep in mind that no HP inkjet Postscript printers are officially -N > supported in DCPS, and even stranger, are not on the list of printers to be D > supported in future. I don't even know if DCPS supports color ???? >   @ DCPS is nice, but if I can do a print to disk, then send the raw5 postscript file to the printer, I'll be ok with that.o  O > Another way of doing this is using a software Postscript RIP. Some HP inkjet <M > printers can be ordered with such a program, but of course just for Windooz- and  > maybe Apple. >   F Ghostscript.  Too much external effort; it'd make printing enough of aE chore that I'd use the wife's Mac instead.  Depressing (though not asr) much as having wintel on my desktop was!)w  M > Perhaps someone at HP would be so kind to think about their loyal hobbiest lK > customers, and come up with a nice VMS driver for a HP inkjet printer ???n >   B That was an active topic of discussion some years ago, but PCL wasD 'not invented here' so no go.  And postscript is fairly portable, atC least for monochrome (witness successful DECwindows printing to thetF LJ5 with HP Postscript), and even cheap on the used market.  Since PCLF is (still) a de-facto standard, and now it _is_ 'invented here', maybe1 there's a chance for DECwindows printing support.s    C How about it, Fred?  Any chance of additional printer support under  Motif/CDE/DECwindows/whatever?   Rich Jordann   ------------------------------  # Date: Mon, 09 Jun 2003 16:03:40 GMTr9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>i! Subject: Re: Radeon 7500 on a PWSf2 Message-ID: <wd2Fa.2180$RB7.1121@news.cpqcorp.net>  5 "John Travell" <john@travell.uk.net> wrote in messagek/ news:bbpibd$c45dm$1@ID-120847.news.dfncis.de... H > > "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> wrote in message > A > > > "Eduardo Anglada" <eduardo.anglada@uam.es> wrote in message 	 > > > Hi,d > > > F > > > I want to learn OpenVMS, but the PowerStorm 4D51T in my PWS-600aJ > > > is not supported by vms, so I wanted to know if the radeon 7500 will > work.aH > > > Any one has tried it? If it doesn't work what would you recommend? > > >i >iH > > Nope.  EV6 systems only - except the XP1000 (unsupported, but 1 head > *will*
 > > work). > >nE > > The old ELSA Gloria card will work in the PWS-600a.  I have a newm versiong > of/ > > the Oxygen VX1 code that "might" also work.r > >y >a > Fred, J > Is this by design, something in the code that only works on an EV6 cpu ?2 > Is it a firmware or physical environment issue ?J > Or is it simply that no-one has tested the Radeon7500 in an EV5 machine. >o   All 3 to an extent.n  & The code expects a flat address space.J The Radeon (as all new cards) does not have a VGA jumper, so firmware must disable the non-console VGA IO.w+ Nobody has tested this on a pre-EV6 system.:  K We can't jump through too many hoops to both provide code optimized for theiI current systems, and yet find ways to make it work on old systems that weuE don't sell anymore, and have no plans to qualify with the new option."   ------------------------------   Date: 9 Jun 2003 08:47:54 -0500A; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)o) Subject: Re: Running java programs in VMSo3 Message-ID: <iLKESeLsi9Vl@eisner.encompasserve.org>b  S In article <ve1euh7uo0cqf9@corp.supernews.com>, "Tim" <tim_a_b@hotmail.com> writes:,   J > But when I try to run it using java "MyClass" or java "MyClass.class" it* > complains that it cannot find the class.  @    Just the java/lang/Thread error, or a different one?  If it's,    different, please post the exact message.  G    Check to make sure that the version of the java you're using matches6B    the version of the setup.com file you're using.  Early releasesA    required the default classpath to include some of the built-inu!    classes, later versions don't./  G    Are you setting CLASSPATH or JAVA$CLASSPATH (which, not if) to find w7    MyClass?  What exact syntax are you using to set it?   H    What are CLASSPATH and JAVA$CLASSPATH set to before and after you add    the path to MyClass?u      Are the files all stream-lf?a  $    Did you put MyClass in a package?   ------------------------------  * Date: Mon, 9 Jun 2003 15:32:54 +0000 (UTC), From: lewis@PROBE.mitre.org (Keith A. Lewis)) Subject: Re: Running java programs in VMSE. Message-ID: <bc29b6$n5a$1@newslocal.mitre.org>  w "Tim" <tim_a_b@hotmail.com> writes in article <ve1euh7uo0cqf9@corp.supernews.com> dated Fri, 6 Jun 2003 18:05:30 +0200:lI >I do indeed have a very old JDK (1.1.8) and I had copied the .class fileA4 >from my Windows PC rather than compiling it on VMS.  K In general, that's allowed.  However, the java.lang.Thread class did get anpI upgrade at version 1.2, so it's better to check your work by compiling ond the old version.  J If you're serious about developing Java on VMS, you really should download the new JDK.  It's free.  9 >I did eventually get the javac compiler working by using 0 >@SYS$MANAGER:JAVA$SETUP.COM and then compiling. >oH >So I now have a class file that was compiled by the VMS javac compiler. >tJ >But now the problem that I am having is that 'java' cannot find the class >that I just created.   H Hmmmm, that's worse than when you started.  At least it was finding your3 class before (in order to know it needed Thread).  e  ? > - I have verified that the class is in the current directory. @ > - I have verified that the classpath contains a single period.I > - I compiled the program using javac "MyClass.java" as I know about the" >case sensitivity.  D If you post the results of the following sequence of commands, I bet% somebody here will spot your problem.   !     $ @SYS$MANAGER:JAVA$SETUP.COMc     $ show log *classpath.     $ javac "MyClass.java"     $ dir myclassa     $ java "MyClass"  > (If you don't know to save your results to a file in VMS, try:*     $ telnet localhost/log=myclasstest.log run the test, and then     $ logoutK at the end.  The commands and results will be in the file myclasstest.log.)m  + --Keith Lewis              klewis$mitre.org > The above may not (yet) represent the opinions of my employer.   ------------------------------  $ Date: Mon, 9 Jun 2003 09:31:08 -0700' From: David Mathog <mathog@caltech.edu>p/ Subject: Re: Speed of IDE controllers in DS10 ?h8 Message-ID: <20030609093108.5f679b08.mathog@caltech.edu>  " On Thu, 05 Jun 2003 22:33:11 +0200 Dirk Munk <munk@home.nl> wrote:n  K > Furthermore the DS10 has two IDE controllers, so with two disks I have a   > throughput of 66 MB/sec.  M I doubt that you'll see anything remotely close to that speed.  At least, not M if you're doing something typical like writing text through RMS to the disks./I Maybe with block IO or other low level access method you might be able tolO get nearer to that hardware limit.  It's been a long time since I've had accesseF to a DS10 with an IDE disk but I do distinctly recall that disk IO wasI SLOOOOOOOOOOOOOOOOOOOOOOOOOOOW.  The only reason our old beowulf with IDEnJ disks on DS10s was usable was because there was enough memory to cache the data in memory at all times.  T For comparison, a Tyan motherboard RH 7.2 machine can write binary data (through dd)K to an ATA100 disk at a sustained 40Mb/sec, and a different Tyan motherboardsC could push data to a single scsi disk (IBM IC35L036UWD via on boardgN SCSI 160 controller) at a sustained 48.8 M/s.  The test in both cases was "dd"P copying 8 Gb of data from /dev/zero to the disk. This was much, much larger thanO the amount of memory in either system, so it (intentionally) took the automatictV disk cache out of the equation.  The point being that these speeds on Linux are prettyQ much the upper limit of what you could ever hope to achieve in terms of hardware, Q and typically the VMS write speeds, due to RMS overhead, are at least 10x slower.2F Oh, and don't forget to take off 3X more for the ATA33 vs ATA100 disk.   Regards,   David Mathog mathog@caltech.edu> Manager, Sequence Analysis Facility, Biology Division, Caltech   ------------------------------  % Date: Mon, 09 Jun 2003 18:55:25 +0200e From: Dirk Munk <munk@home.nl>/ Subject: Re: Speed of IDE controllers in DS10 ?+2 Message-ID: <bc2ef4$j6h$1@news3.tilbu1.nb.home.nl>   Mmmmmmmmm.....  Q The original disk in this DS10 was a 5400rpm 10GB Seagate. Now, I'm sure that is WO a slow disk. I'm now using a 7200rpm 80GB Western Digital disk with 8MB cache, eO and I really can't complain about the speed. I can compare it with my previous  N PWS500 with SCSI, and the disk IO on this machine is certainly not slower, on 
 the contrary!S  Q Now I would like to install a nice 72GB 15000rpm SCSI-160 drive with controller, r but that is a bit expensive.  P I agree that my present disk is most likely slowed down by the speed of the IDE O controller, because it can read data faster then the IDE bus can transport it. aP The 8MB cache will help a lot in leveling out this problem. That is why I think N it would be nice if the IDE controller could be set to 66MB/sec (33 MHz). The F controller hardware seems to support it in principle from what I read.  P By the way, SCSI disks are slowed down as well by a (relatively) slow SCSI bus. Q I once tested a 15000rpm 18GB disk on the internal 80MB/sec SCSI controller of a .Q ES40. I got appr. 35MB/sec out of it. The same disk will do about half that much i3 on the internal 40MB/sec UW-SCSI busses of a HSG80.a           David Mathog wrote:u    O > I doubt that you'll see anything remotely close to that speed.  At least, notlO > if you're doing something typical like writing text through RMS to the disks.'K > Maybe with block IO or other low level access method you might be able touQ > get nearer to that hardware limit.  It's been a long time since I've had accessIH > to a DS10 with an IDE disk but I do distinctly recall that disk IO wasK > SLOOOOOOOOOOOOOOOOOOOOOOOOOOOW.  The only reason our old beowulf with IDE L > disks on DS10s was usable was because there was enough memory to cache the > data in memory at all times. > V > For comparison, a Tyan motherboard RH 7.2 machine can write binary data (through dd)M > to an ATA100 disk at a sustained 40Mb/sec, and a different Tyan motherboard E > could push data to a single scsi disk (IBM IC35L036UWD via on board P > SCSI 160 controller) at a sustained 48.8 M/s.  The test in both cases was "dd"R > copying 8 Gb of data from /dev/zero to the disk. This was much, much larger thanQ > the amount of memory in either system, so it (intentionally) took the automatic0X > disk cache out of the equation.  The point being that these speeds on Linux are prettyS > much the upper limit of what you could ever hope to achieve in terms of hardware,tS > and typically the VMS write speeds, due to RMS overhead, are at least 10x slower.5H > Oh, and don't forget to take off 3X more for the ATA33 vs ATA100 disk. > 
 > Regards, >  > David Mathog > mathog@caltech.edu@ > Manager, Sequence Analysis Facility, Biology Division, Caltech   ------------------------------  % Date: Mon, 09 Jun 2003 09:52:06 +0200 $ From: Michael Unger <unger@decus.de> Subject: Re: timezone rule5 Message-ID: <bc1ipq$ec457$1@ID-152801.news.dfncis.de>u  & On 08-Jun-2003 20:18, Bart Zorn wrote:   > karolina wrote:a >> Hi, >> 6  >> I got the timezone rules fromH >> http://h71000.www7.hp.com/doc/73final/6497/6497pro_007.html. I made aH >> little program that putenv(TZ=MET-1MET_DST-2,M3.5.0/2,M9.5.0/2). ThisH >> works fine, but when I start using time functions like mktime() using3 >> the tm_isdst flag in struct tm it tells me that:  >> Running Middle European Time  >>   >> [...] >> n< >> Looking a little closer at the rule it actually says thatH >> daylightsaving starts at Sun Sep 28 00:00:00 2003(M9.5.0/2 as above).G >> That's not right. As far as I know it starts in the late october eg.m >> M10,...?!  D That has been changed from "M9" to "M10" a few years ago; so "M9" is
 obsolete now.r   >> t >> [...] >>   >> Is this a known problem?o& >> What should I do to get this right?D >> Is there any page that is upto date for the right timezone rules?4 >> Where can I find the standard for timezone rules?   See (in English):U  C <URL: http://www.bipm.fr/enus/5_Scientific/c_time/time_server.html>p+ (Bureau International des Poids et Mesures)   E The following listing may be of some help too, but it is available ina German only:  1 <URL: http://www.ptb.de/de/org/4/43/432/zeit.htm>  (German Metrologic Institute)l   >> r >> [...] >  > [...]  > G > Assuming that you are in the MET timezone (ST GMT+1, DST GMT+2), the l> > correct timezone rule is MET-1MET DST-2,M3.5.0/02,M10.5.0/03  A The switching from ST to DST and vice versa is done at 1:00 *UTC* H (formerly called GMT) in both cases; so from ST to DST it is at 2:00 andJ from DST to ST it is at 3:00 *local* *time* -- at least in Central Europe.   >  > [...]a   Michael    -- c  @ Please do *not* send "Security Patch Notifications" or "SecurityA Updates"; this system isn't running a Micro$oft operating system. = And don't annoy me <mailto:postmaster@[127.0.0.1]> please ;-)    ------------------------------  % Date: Mon, 09 Jun 2003 13:10:03 +0200e) From: Bart Zorn <B.Zorn@xs4all.nospam.nl>- Subject: Re: timezone rule6 Message-ID: <3ee46b0c$0$49113$e4fe514c@news.xs4all.nl>   Michael Unger wrote:( > On 08-Jun-2003 20:18, Bart Zorn wrote: >  >  >>karolina wrote:1 >> >>>Hi, >>>e  >>>I got the timezone rules fromH >>>http://h71000.www7.hp.com/doc/73final/6497/6497pro_007.html. I made aH >>>little program that putenv(TZ=MET-1MET_DST-2,M3.5.0/2,M9.5.0/2). ThisH >>>works fine, but when I start using time functions like mktime() using3 >>>the tm_isdst flag in struct tm it tells me that:i >>>Running Middle European Timee >>>y >>>[...] >>>m< >>>Looking a little closer at the rule it actually says thatH >>>daylightsaving starts at Sun Sep 28 00:00:00 2003(M9.5.0/2 as above).G >>>That's not right. As far as I know it starts in the late october eg.s >>>M10,...?! >  > F > That has been changed from "M9" to "M10" a few years ago; so "M9" is > obsolete now.e >  >  >>>[...] >>>I >>>Is this a known problem?e& >>>What should I do to get this right?D >>>Is there any page that is upto date for the right timezone rules?4 >>>Where can I find the standard for timezone rules? >  >  > See (in English):  > E > <URL: http://www.bipm.fr/enus/5_Scientific/c_time/time_server.html>r- > (Bureau International des Poids et Mesures)b > G > The following listing may be of some help too, but it is available inw > German only: > 3 > <URL: http://www.ptb.de/de/org/4/43/432/zeit.htm>d > (German Metrologic Institute)  >  >  >>>[...] >> >>[...]  >>G >>Assuming that you are in the MET timezone (ST GMT+1, DST GMT+2), the s> >>correct timezone rule is MET-1MET DST-2,M3.5.0/02,M10.5.0/03 >  > C > The switching from ST to DST and vice versa is done at 1:00 *UTC*oJ > (formerly called GMT) in both cases; so from ST to DST it is at 2:00 andL > from DST to ST it is at 3:00 *local* *time* -- at least in Central Europe.  ' Correct. And that is also what the rule . MET-1MET DST-2,M3.5.0/02,M10.5.0/03 specifies.   >  >  >>[...]r >  > 	 > Michaelr >    Bart   ------------------------------  % Date: Mon, 09 Jun 2003 14:18:41 +0200o6 From: Adolf Sonderegger <adolf.sonderegger@bluewin.ch> Subject: Re: timezone rule* Message-ID: <3EE47B21.382DD05A@bluewin.ch>   karolina wrote:z   > Hi,l >  > I got the timezone rules fromoG > http://h71000.www7.hp.com/doc/73final/6497/6497pro_007.html. I made agG > little program that putenv(TZ=MET-1MET_DST-2,M3.5.0/2,M9.5.0/2). This?G > works fine, but when I start using time functions like mktime() usinge2 > the tm_isdst flag in struct tm it tells me that: > Running Middle European Time > Sun Mar 30 00:00:00 2003 > Sun Sep 28 00:00:00 2003 >i; > Looking a little closer at the rule it actually says thateG > daylightsaving starts at Sun Sep 28 00:00:00 2003(M9.5.0/2 as above).eF > That's not right. As far as I know it starts in the late october eg. > M10,...?!a >MC > Looking at other timezone rules at the hp.com page I noticed thats3 > there are more timezone strings that's not right?aC > That it ends at Sun Mar 30 00:00:00 2003 is right as shown above.o >M > Is this a known problem?% > What should I do to get this right?oC > Is there any page that is upto date for the right timezone rules?w3 > Where can I find the standard for timezone rules?o >t > please help! >g > //karolina   Hello2  & This is a known bug in OpenVMS V7.3-1!( HP will deliver a Patch! Date not known.   regardsp Adolf Sonderegger    ------------------------------  % Date: Mon, 09 Jun 2003 15:09:05 +0200e) From: Bart Zorn <B.Zorn@xs4all.nospam.nl>e Subject: Re: timezone rule6 Message-ID: <3ee486f1$0$49110$e4fe514c@news.xs4all.nl>   Adolf Sonderegger wrote: >  > karolina wrote:e >  >  >>Hi,n >> >>I got the timezone rules fromtG >>http://h71000.www7.hp.com/doc/73final/6497/6497pro_007.html. I made amG >>little program that putenv(TZ=MET-1MET_DST-2,M3.5.0/2,M9.5.0/2). ThiseG >>works fine, but when I start using time functions like mktime() using02 >>the tm_isdst flag in struct tm it tells me that: >>Running Middle European Time >>Sun Mar 30 00:00:00 2003 >>Sun Sep 28 00:00:00 2003 >>; >>Looking a little closer at the rule it actually says thattG >>daylightsaving starts at Sun Sep 28 00:00:00 2003(M9.5.0/2 as above).tF >>That's not right. As far as I know it starts in the late october eg. >>M10,...?!y >>C >>Looking at other timezone rules at the hp.com page I noticed thatn3 >>there are more timezone strings that's not right?lC >>That it ends at Sun Mar 30 00:00:00 2003 is right as shown above.r >> >>Is this a known problem?% >>What should I do to get this right?nC >>Is there any page that is upto date for the right timezone rules?k3 >>Where can I find the standard for timezone rules?  >> >>please help! >> >>//karolina >  >  > Helloc > ( > This is a known bug in OpenVMS V7.3-1!* > HP will deliver a Patch! Date not known. > 	 > regardsh > Adolf Sondereggerm >   C Can you supply more information? As far as I know, after they have bI struggled with the European Time Zones for a number of OpenVMS versions,  G I believe HP now finally got it right. If they deliver a patch, I hope 2$ that they do not introduce new bugs!  	 Bart Zorn.   ------------------------------  % Date: Mon, 09 Jun 2003 15:28:57 +0100s0 From: Chris Sharman <chris.sharman@sorry.nospam> Subject: Re: VMS SQL clientm4 Message-ID: <bc26o2$jhf$1$8300dec7@news.demon.co.uk>   Tried various things - no luck.tG I should make it clear that I've never used SQL, so I'm something of a  E newbie. We're installing the Navision MIS system, and moving various aD data into it from VMS. The VMS apps (which used to look up customer I details, stock records, etc via RMS) now need to look up (and update) on w Navision's SQL database.  F Installed the bundled Attunity stuff - could find no documentation on I how to use it - not sure whether it does what I want (use MS SQL data in I my VMS application).  A Tried freedts (both from freedts.org, and Ruslan Laishev's work).5  G freedts.org: got freetds-0.6.1 and the vms bits out of freetds-current.u? Got a lot of MMS warnings (1 for most modules), which I didn't o0 understand, followed by linker warnings (below).H Laishev: collected all the individual modules: not sure whether they're 7 meant to be standalone, or inserted into the other kit mL somewhere/somehow. When I built (with MMS) it complained odbc.c was missing.  H SQL requests via http seems to be another option, but I'm unsure of the & security/performance of this approach.  G Can anyone shed any light on where I'm going wrong, and/or whether I'm  C likely to get reasonable performance from any of these approaches ?p   Many thanks, Chris      $ mms/mac=tdsver=tds80   CC/DECC c /DEFINE=(TDS80,HAVE_CONFIG_H=1)/PREFIX=(ALL)/FLOAT=IEEE/NAMES=AS_IS/OBJECT=[.SRC.TDS]CHALLENGE.OBJ f /INCLUDE=([],[.include]) [.SRC.TDS]CHALLENGE.Cp1 %LIBRAR-S-INSERTED, module CHALLENGE inserted in -8 DISK$ADEV:[CHRIS.AWORK.FREETDS.FREETDS-0_61]LIBTDS.OLB;1* %MMS-W-GWKACTNOUPD, Actions didn't update % LIBTDS.OLB([=[.SRC.TDS]CHALLENGE.OBJ)f   ...m  2 link/NOTRACE/exe=[.SRC.TDS.UNITTESTS]DYNAMIC1.EXE " [.SRC.TDS.UNITTESTS]DYNAMIC1.OBJ, 0 [.SRC.TDS.UNITTESTS]COMMON.OBJ,[]libtds.OLB/libr ary ' %LINK-W-NUDFSYMS, 11 undefined symbols:t( %LINK-I-UDFSYM,         tds_close_socket- %LINK-I-UDFSYM,         tds_read_conf_sectionn, %LINK-I-UDFSYM,         tds_read_config_info& %LINK-I-UDFSYM,         tds_set_parent& %LINK-I-UDFSYM,         tds_swap_bytes# %LINK-I-UDFSYM,         tds_versiony( %LINK-I-UDFSYM,         tdsdump_dump_buf# %LINK-I-UDFSYM,         tdsdump_logt# %LINK-I-UDFSYM,         tdsdump_offm" %LINK-I-UDFSYM,         tdsdump_on$ %LINK-I-UDFSYM,         tdsdump_open   ------------------------------   Date: 9 Jun 2003 08:51:26 -0500 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)n, Subject: Re: What do these Macro codes mean?3 Message-ID: <QHkxqyufd3dq@eisner.encompasserve.org>p  h In article <ZTKDa.3467$c6.3274@bos-service2.ext.raytheon.com>, Chris Olive <nospam@raytheon.com> writes: > I > Indeed.  I always found myself feeling that VAX MACRO-32 wasn't really uH > "assembly language" at all.  Of course on the Alpha, it isn't, but in G > VAX days, I always thought it was between assembly and 3GL, such was k > (is) the instruction set.t  E    Well I always thought it was "assembly language", but I was reallya>    impressed that the instruction set did the kinds of things H    programmers need (CALLx subroutines, etc.) instead of just the kinds 9    of things hardware designers need (move, add, branch).a   ------------------------------   Date: 9 Jun 2003 08:55:26 -0500 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) , Subject: RE: What do these Macro codes mean?3 Message-ID: <40CyOv3Kfaez@eisner.encompasserve.org>y  T In article <QBRTqPlshJm+@eisner.encompasserve.org>, briggs@encompasserve.org writes: > ? > It was an eye opening claim.  Since most of us would take theb@ > time to lay out data structures carefully to take advantage of@ > natural alignment but I, for one, would not have thought aboutE > inserting NOOPs in my VAX MACRO code sequences as a way to increase  > performance.  C    Sounds like you haven't looked at a Fortran machine code listingsB    since the VAX 6000 series came out.  At about that time FortranA    started generating lots of NOPs.  I was left wondering if thiso3    was a performance enhancment or hit on my MV II.i   ------------------------------   Date: 9 Jun 2003 09:20:19 -0500o From: briggs@encompasserve.org, Subject: RE: What do these Macro codes mean?3 Message-ID: <8hjJSZubFe3$@eisner.encompasserve.org>   q In article <40CyOv3Kfaez@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:mV > In article <QBRTqPlshJm+@eisner.encompasserve.org>, briggs@encompasserve.org writes: >> -@ >> It was an eye opening claim.  Since most of us would take theA >> time to lay out data structures carefully to take advantage of A >> natural alignment but I, for one, would not have thought about.F >> inserting NOOPs in my VAX MACRO code sequences as a way to increase >> performance.  > E >    Sounds like you haven't looked at a Fortran machine code listingmD >    since the VAX 6000 series came out.  At about that time FortranC >    started generating lots of NOPs.  I was left wondering if this 5 >    was a performance enhancment or hit on my MV II.a  H Perhaps selective vision.  The only time I look at machine code listingsG is when I'm trying to troubleshoot a "this can't possibly be happening"n@ type error.  And when that happens, I'm usually focused in a bitC close.  NOOPs disappear real easy when you're looking for somethingr else.   @ I'm a little curious about the nature of the performance benefit> for aligning branch labels.  Do you pick up the boost by beingC reasonably sure of picking up a useful amount of instruction streaml= on the first memory access?  Or are you trying to improve thep? odds that you can fit a tight loop entirely within the I cache? 6 Or are there subtler issues that I'm missing entirely?   	John Briggs   ------------------------------  # Date: Mon, 09 Jun 2003 15:49:01 GMTtF From: lederman@star.enet.dec.DISABLE-JUNK-EMAIL.com (Bart Z. Lederman), Subject: RE: What do these Macro codes mean?2 Message-ID: <N%1Fa.2178$Zu7.1149@news.cpqcorp.net>   Regarding alignment:  D On the VAX (which is where the original example probably came from),@ alignment really isn't an issue.  I've never seen anyone measure
 a difference.d  @ On the Alpha, alignment is significant.  The performance penalty can be easily measured.u  B On the Itanium, it appears so far that alignment is essential. TheB performance penalty for un-aligned access is significantly greaterD than it is for the Alpha (I've heard an order of magnitude, but keepD in mind that this news group is not an official or supported channel of information).   -- i(  B. Z. Lederman   Personal Opinions Only  8  Posting to a News group does NOT give anyone permission8  to send me advertising by E-mail or put me on a mailing  list of any kind.  5  Please remove the "DISABLE-JUNK-EMAIL" if you have a85  legitimate reason to E-mail a response to this post.J   ------------------------------  $ Date: Mon, 9 Jun 2003 09:51:44 -0700# From: "Tom Linden" <tom@kednos.com>s, Subject: RE: What do these Macro codes mean?9 Message-ID: <CIEJLCMNHNNDLLOOGNJIGENIHFAA.tom@kednos.com>t  E BTW, one tick penalty for unaligned access on Power PC.  Why you ask?i@ Because they have an existing code base they don't wish to lose.   >-----Original Message-----t >From: Bart Z. Ledermanb7 >[mailto:lederman@star.enet.dec.DISABLE-JUNK-EMAIL.com]s$ >Sent: Monday, June 09, 2003 8:49 AM >To: Info-VAX@Mvb.Saic.Com- >Subject: RE: What do these Macro codes mean?r >d >t >Regarding alignment:a >tE >On the VAX (which is where the original example probably came from),sA >alignment really isn't an issue.  I've never seen anyone measureo >a difference. > A >On the Alpha, alignment is significant.  The performance penaltyd >can be easily measured. > C >On the Itanium, it appears so far that alignment is essential. TheeC >performance penalty for un-aligned access is significantly greateraE >than it is for the Alpha (I've heard an order of magnitude, but keep E >in mind that this news group is not an official or supported channeln >of information).  >n >-- ) > B. Z. Lederman   Personal Opinions Only- > 9 > Posting to a News group does NOT give anyone permissionw9 > to send me advertising by E-mail or put me on a mailinge > list of any kind.j >n6 > Please remove the "DISABLE-JUNK-EMAIL" if you have a6 > legitimate reason to E-mail a response to this post. >  >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com).GA >Version: 6.0.484 / Virus Database: 282 - Release Date: 5/27/2003. >s ---s& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.484 / Virus Database: 282 - Release Date: 5/27/2003   ------------------------------  # Date: Mon, 09 Jun 2003 12:05:19 GMTa6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)S Subject: Re: [DCL] Is the max length of DCL lexical functions argument documented ?S5 Message-ID: <3K_Ea.111135$lL2.1025583@news.chello.at>f  q In article <cf15391e.0306060901.110f8afc@posting.google.com>, keithparris_NOSPAM@yahoo.com (Keith Parris) writes: r >peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) wrote in message news:<s_PCa.83132$_c6.706880@news.chello.at>...N >> I just made a litte test and found that the max lengths of DCL entities are >> H >> READ			2048? >> WRITE			up to 1024 (where every expression can't exceed 256)d8 >> WRITE/SYMBOL		2048 (but only symbols, no expressions)8 >> F$LENGTH		1024 (but including whole DCL command line)8 >> F$LOCATE		1024 (but including whole DCL command line) >l? >There are significant improvements slated for 7.3-2, includingoF >increases in the length of DCL command lines, symbols, and the length, >of records you can read and write with DCL.  J Wow. That's the 2nd good news in a week. Congratulations, you made my day.N Can I also assume, that this limits will then be documented at the appropriate9 places (and not only in the New Features Guide) as well ?i   -- a Peter "EPLAN" LANGSTOEGERX% Network and OpenVMS system specialista E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Mon, 09 Jun 2003 10:35:09 +0200i2 From: martin@radiogaga.harz.de (Martin Vorlaender)) Subject: Re: [OT] best HTML code cleaner?s; Message-ID: <3ee446bd.524144494f47414741@radiogaga.harz.de>u  # Didier Morandi (no@spam.com) wrote:  > Martin Vorlaender wrote:E > > My guess is that the installation failed to set up the perl PATH.DC > > Try opening a DOS box and start perl there using the full path.aD > > On the other hand, IIRC Win32 perl needs some DLLs, so without a& > > proper PATH setting it won't work. >,  > No, its here: C:\Perl\bin\;... >nK > I did not understand that perl was a "foreign command" so I tried to run f > the PPM program. >a > Typing the command% > C:\perl -i_bak demoronize fred.htmls > givesn@ > Can't open perl script "demoronize": No such file or directory- > as there is no demoronize.* in C:\Perl\bin\r  F "demoronize" is meant to be the file holding the perl script I posted.H Of course you need to be in the directory the script file resides in forG the above command to work, or else supply the full path to its location  to perl.   cu,    Martin -- hD                        |  Martin Vorlaender  |  VMS & WNT programmer1   OpenVMS: When you    |  work: mv@pdv-systeme.de E   KNOW where you want  |     http://www.pdv-systeme.de/users/martinv/y8   to go today.         |  home: martin@radiogaga.harz.de   ------------------------------   End of INFO-VAX 2003.318 ************************