0 INFO-VAX	Fri, 16 Jan 2004	Volume 2004 : Issue 31      Contents: Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 500.000 AMD64's shipped... Re: 7.3-1 to 7.3-2 Alpha server 1000  Re: Detaching processes on VMS Re: Detaching processes on VMS Re: Detaching processes on VMS Re: Detaching processes on VMS Re: Detaching processes on VMS Re: Detaching processes on VMS Re: Detaching processes on VMS  Re: DHCP/BOOTP/PXE under OPENVMSH Re: Faced with flagging PC sales,  HPand others are pushing aggressivelyH Re: Faced with flagging PC sales,  HPand others are pushing aggressivelyH Re: Faced with flagging PC sales,  HPand others are pushing aggressivelyP Re: Faced with flagging PC sales,  HPand others are pushing aggressively into di* How many file header does indexf.sys have?. Re: How many file header does indexf.sys have?. Re: How many file header does indexf.sys have?. Re: How many file header does indexf.sys have?D Re: HP Integrity Superdome server sets world record for 10-terabyte,P Re: HP Integrity Superdome server sets world record for 10-terabyte, single-syst= Re: HP rockets to No. 5 on list of top U.S. patent recipients = Re: HP rockets to No. 5 on list of top U.S. patent recipients 1 Re: Humongus bug in DECTERM (postscript printing) 1 Re: Humongus bug in DECTERM (postscript printing) 1 Re: Humongus bug in DECTERM (postscript printing) < Re: IBM announces DB2 for HP Integrity Itanium-based servers Re: IP Address problems  Re: IP Address problems  Re: IP Address problems  Re: IP Address problems  Re: IP Address problems P Re: Judge Upholds Jury Decision In Microsoft Web-Browser Patent Case - U.S. Pate# Minimum Privs for Changing Password ' Re: Minimum Privs for Changing Password ' Re: Minimum Privs for Changing Password 0 Re: Mounting identically labelled disks (OVMS73)0 Re: Mounting identically labelled disks (OVMS73)0 Re: Mounting identically labelled disks (OVMS73)0 Re: Mounting identically labelled disks (OVMS73) Re: Mozilla on Alpha question  Re: Mozilla on Alpha question  Re: Mozilla on Alpha question 
 OPC DA client # Re: OT: AA pilot arrested in Brazil  Re: The Inquirer Re: The Inquirer Re: The Inquirer Re: The Inquirer Re: The Inquirer Re: The Inquirer Re: The Inquirer Re: The Inquirer@ Re: The Register: OpenVMS among most-secure of operating systems@ Re: The Register: OpenVMS among most-secure of operating systems@ Re: The Register: OpenVMS among most-secure of operating systems% Time synch between an Alpha and a Vax ) Re: Time synch between an Alpha and a Vax  Re: UTF-8 on VAX-VMS Re: UTF-8 on VAX-VMS Re: UTF-8 on VAX-VMS Re: UTF-8 on VAX-VMS' Re: VAXELN-to-VxWorks portability layer ' Re: VAXELN-to-VxWorks portability layer " Re: Will HP *ever* publicize OVMS?" RE: Will HP *ever* publicize OVMS?" Re: Will HP *ever* publicize OVMS?  F ----------------------------------------------------------------------   Date: 16 Jan 2004 07:25:23 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow>' Subject: Re: 500.000 AMD64's shipped... 5 Message-ID: <DTiotGxQ0bj6-pn2-QgMEJXQ7jEN1@localhost>   B On Thu, 15 Jan 2004 15:48:02 UTC, dennis@etinc.com (Dennis) wrote:  H > I'll take a 3.06Ghz Xeon over either. The stability and performance ofH > the peripheral chipsets, particularly the bus, are more important, andF > intel has a wide lead in that category. All of the AMD chipsets haveD > flaws. Additionally, most high end machines are I/O bound, not cpuG > bound. So for the vast majority of applications your little arguments , > and benchmarks are just plain meaningless. > F > Bravo to AMD however, for doing good marketing. Thank goodness thereF > is always a benchmark to make your case, no matter what it might be.  7 And Intels' never do? Not to mention infamous CPU bugs.    --   Cheers - Dave.   ------------------------------   Date: 16 Jan 2004 07:25:26 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow>' Subject: Re: 500.000 AMD64's shipped... 5 Message-ID: <DTiotGxQ0bj6-pn2-1UMx53F6JD9d@localhost>   E On Thu, 15 Jan 2004 18:20:42 UTC, David Froble <davef@tsoft-inc.com>   wrote:   > > Keith Parris wrote: K > >> At the moment, AMD's x86-64 is competing mostly against Intel's 32-bit 0 > >> chips, which soundly trounce it in SPECint:" > >>   Pentium 4EE (3.2 Ghz)  1620" > >>   AMD Opteron (2.2 Ghz)  1477 >  > F > The Pentium 4EE was announced to attempt to minimalize the AMD chip S > announcements.  Can you actually find one of these animals?  Do they exist?  How   > many have been sold? > R > At least Andy boy sticks mainly to reality.  Not that he wants to, but he knows K > we'll question any BS he posts.  The above doesn't help your credibility.  >  > Dave >   D Saw a price of circa 1200 Euro (IIRC) for one while looking last in 
 town week.   --   Cheers - Dave.   ------------------------------  % Date: Fri, 16 Jan 2004 04:42:43 -0500 * From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: 500.000 AMD64's shipped... 2 Message-ID: <pIydnaTyqL-bL5rdRVn-hg@metrocast.net>  D "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> wrote in message- news:I9fNb.12310$xU3.6621@news.cpqcorp.net...  > 9 > "David Froble" <davef@tsoft-inc.com> wrote in message >  > L > > Similar to the exporting of jobs from the US, how long can Intel support > their J > > expensive 'ego' CPU if they're not pulling in bundles of cash from the > 'breadF > > and butter' desktop market?  I don't care how much money they have > available I > > now.  If they start losing money, the well will run dry at some time.  Dry  > wells $ > > won't continue to produce IA-64. > >  > K > You are turning the world on it's head.  The R&D investments are done and  > paid for.   L Bummer!  You mean the entire Alpha team is now out looking for jobs?  Not toL mention the people who were bringing Montecito to market in 90 nm. such thatI it won't melt under load (I kind of assume that Madison II is pretty well 
 done by now)?   H Or perhaps they're all working gratis because they like Itanic *so* much, that they can't imagine doing anything else.   - bill   ------------------------------  % Date: Fri, 16 Jan 2004 05:23:54 -0500 * From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: 500.000 AMD64's shipped... 2 Message-ID: <1emdnaRb5s8xJprdRVn-vA@metrocast.net>  5 "David Froble" <davef@tsoft-inc.com> wrote in message & news:4006DAD1.8040006@tsoft-inc.com... > Keith Parris wrote:  > - > > Dirk Munk <munk@home.nl> wrote in message . news:<bu5gh4$k0i$1@news4.tilbu1.nb.home.nl>... > > L > >>However AMD has *one* 32/64 bit architecture for the laptop, the desktop and the K > >>server environment. Intel will have two very distinctive architectures,  64 bitD > >>for the servers and a rather different 32 bit for anything else. > >> > > B > > x86-64 may do well on the desktop, but the design is seriously@ > > compromised for the server space.  Maximum of 8-way (current& > > Superdomes are already at 64-way), >  > J > Lets see, we're comparing a chip to a system?  Credibility takes another dip.L > My understanding is that Opteron has on-chip glue for up to 8-way systems.K > IA-64 doesn't have any such, it's all done off-chip.  Can you explain why  such3 > couldn't also be done in an Opteron based system?   J Probably not, because it can be done and apparently will be done.  NewisysC has been working on it right along, and Sun has said it is as well.   K Since the vast majority of server requirements (read:  server chip volumes) J are met by 4-way (let alone 8-way) servers, and since AMD64 will be movingB to dual cores on the chip before 2006 (extending Hammer's range toH 16-processor servers), only higher-end niches are safe hiding places forK Itanic (well, from Hammers, anyway:  POWER4+/5/5+ will be more than capable G of giving Itanic fits there) - until such time as larger Hammer systems  become available.    > ' > > 40 bits or 1 TB of physical address H > > space (current Superdomes already support half that address space at > > 512 GB),  I Keith's excursions into marketeering seem to have caused a decline in his E competence and/or his honesty.  While today's Hammer *implementation* D supports only 40 bits (1 TB) of physical address space (an amount ofH physical memory which, as Keith notes, HP apparently considers more thanL sufficient as a limit for its current 64-processor Superdomes), the *design*J supports up to 52 bits (4 PB) of physical address space - likely more thanH enough for Hammer's lifetime (let alone Itanic's).  This should not be aK great surprise to Keith, since Alpha also had implementation limits in this 0 area that were far lower than its design limits.  < > maximum of 48-bit virtual address space instead of 64-bit,  F Wrong again, Keith:  this too is a limit of the current implementationG rather than of the design, which supports a full 64-bit virtual address  space.  " > > and only 16 integer registers.  H He got that one right.  Of course, this doesn't seem to have kept HammerL from surpassing Itanic in integer performance - including server benchmarks.I So if Hammer is 'seriously compromised' in this area, I guess that Itanic  must be even more so.    - bill   ------------------------------  % Date: Fri, 16 Jan 2004 10:36:02 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> ' Subject: Re: 500.000 AMD64's shipped... 0 Message-ID: <bu8eqi$f1o$2@new-usenet.uk.sun.com>   Rick Jones wrote: R > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote: > B >>SuperDome currently supports 1/2 TB of RAM and will this year or >>next year support 1 TB   >  > B > This. I suspect to be a bit more accurate about it, HP-UX on theF > Itanium Superdome will include support.  I believe the capability is > already in the hardware. >   4 I am sure it must be but then having the physical HW0 support but not having an OS to address it is in chocolate teapot territory.      Regards  Andrew Harrison    ------------------------------  % Date: Fri, 16 Jan 2004 10:37:31 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> ' Subject: Re: 500.000 AMD64's shipped... 0 Message-ID: <bu8etb$f1o$3@new-usenet.uk.sun.com>   David Svensson wrote:  > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bu3rct$qce$2@new-usenet.uk.sun.com>...  >  >>David Svensson wrote:  >> >>>Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bu1eh4$1ct$1@new-usenet.uk.sun.com>... >>>  >> >>   >>H >>>It is currently expensive and hotter, but in my real world experienceI >>>with testing customer database setups, nothing beats Itanium today and I >>>that matters. (this includes testing of all current CPU architectures)  >> >>I >>So you have tried an unspecified DBMS on a Power4+ P690, a USIII 1.2Ghz H >>SF Sun and a GS1280. You must have access to a great deal of expensive >>kit, congratulations.  >> >  > 	 > Thanks, H > but they are 4-way boxes, so there is no need to have P690 and GS1280.   Which 4 way Sun ?  Which 4 way IBM ?  Which 4 way AlphaServer ?    Regards  Andrew Harrison    ------------------------------  % Date: Fri, 16 Jan 2004 10:42:33 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> ' Subject: Re: 500.000 AMD64's shipped... 0 Message-ID: <bu8f6p$f1o$5@new-usenet.uk.sun.com>   JF Mezei wrote:  > Keith Parris wrote:  > H >>The reason you give (better IA32 pricing) is pure speculation.  But itF >>is clear that Dell and IBM think there's enough potential in ItaniumC >>that they don't want to miss out.  Dell even sat things out for a  >>while, and dove back in. >  > H > Has Dell designed its own motherboards or are they using off-the shelfM > "default" boards. If they are merely assemblyng boxes desiogned by Intel or N > someone else, they should be seen as a subcontractor of the org who designedN > those motherboards instead of a company who has bet their own money on IA64. >   B Dell don't generally design motherboards, they commision 3rd party suppliers to do that.   A So its unlikely that the Dell have done much more than design the  front bezel.   Regards  Andrew Harrison    ------------------------------  % Date: Fri, 16 Jan 2004 10:40:01 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> ' Subject: Re: 500.000 AMD64's shipped... 0 Message-ID: <bu8f22$f1o$4@new-usenet.uk.sun.com>   Keith Parris wrote: ] > JF Mezei <jfmezei.spamnot@istop.com> wrote in message news:<4006453E.51861944@istop.com>...  > N >>Neither have bet their business on IA64. They are only producing token IA64sP >>to please Intel in exchange for a better deal on other Intel hardware which is >>key to their business. >  > H > The reason you give (better IA32 pricing) is pure speculation.  But itF > is clear that Dell and IBM think there's enough potential in ItaniumC > that they don't want to miss out.  Dell even sat things out for a  > while, and dove back in. >   ? Really, then you will have missed the fact that IBM have chosen ; to go with Xeons for their 64 way Intel based server rather  than the Itanium.    Regards  Andrew Harrison    ------------------------------  % Date: Fri, 16 Jan 2004 06:08:07 -0500 * From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: 500.000 AMD64's shipped... 2 Message-ID: <4qCdnSW0k4qVW5rdRVn-hQ@metrocast.net>  > "Keith Parris" <keithparris_NOSPAM@yahoo.com> wrote in message7 news:cf15391e.0401141419.63b8626d@posting.google.com... K > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> = wrote in message news:<bu3gu1$mm2$1@new-usenet.uk.sun.com>... = > > But Opteron Integer performance is stubbornly faster than 9 > > Itanium despite being saddled with 1/8 the registers.  > >  > > 1477 SPECint vs 1322 > H > At the moment, AMD's x86-64 is competing mostly against Intel's 32-bit- > chips, which soundly trounce it in SPECint:   A Hmmm - do you always change the subject when someone refutes your G assertions, Keith?  Previously, you were comparing Hammer to Itanic and J claiming that the latter's lack of integer registers constituted some kindH of disadvantage.  The SPECint scores prove otherwise - and the fact thatI Pentium (also using only 8 integer registers) *also* leaves Itanic in the ! dust only underscores that point.    ...   E > Right now Itanium is artificially constrained in clock speed due to B > heat generation -- 130 watts is the power budget today's systems > allow.  K Really?  Last I knew, the 1.5 GHz Madison consumed only 107W, leaving about I 20% margin for increased clock speed if that were the *only* problem.  Do L you really think that Intel wouldn't be clocking Itanic at 1.8 GHz right nowI if it could do so?  Perhaps you've never looked at the 'schmoo' plot that H Intel released a year or so back showing that yields fell off a cliff atJ anything much above 1.6 GHz (and were bad enough even at 1.6 GHz that they! chose 1.5 GHz for their top end).   >   Intel just announced that after 5-6 years of hard work, theyH > now have the keys to producing low-k dielectric transistors, with 100XE > lower leakage current.  This can reduce the power consumption of an C > overall chip by 50%, allowing the clock rate to be cranked up and $ > still fit within the power budget.  L I wouldn't advise holding your breath until that process reaches production,K Keith.  *Maybe* by the time Tanglewood appears (if it doesn't quite make it % into 2006), but I wouldn't bet on it.    - bill   ------------------------------  % Date: Fri, 16 Jan 2004 06:52:53 -0500 * From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: 500.000 AMD64's shipped... 2 Message-ID: <w7-dnVmAuIMaTZrdRVn-uw@metrocast.net>  4 "David Svensson" <icerq4a@spray.se> wrote in message7 news:734da31c.0401150736.65c94cd6@posting.google.com... K > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> = wrote in message news:<bu3rct$qce$2@new-usenet.uk.sun.com>...  > > David Svensson wrote: ' > > > Andrew Harrison SUNUK Consultancy 8 <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message, news:<bu1eh4$1ct$1@new-usenet.uk.sun.com>... > > >  > > K > > > It is currently expensive and hotter, but in my real world experience L > > > with testing customer database setups, nothing beats Itanium today andL > > > that matters. (this includes testing of all current CPU architectures) > >  > > K > > So you have tried an unspecified DBMS on a Power4+ P690, a USIII 1.2Ghz J > > SF Sun and a GS1280. You must have access to a great deal of expensive > > kit, congratulations.  > >  > 	 > Thanks, H > but they are 4-way boxes, so there is no need to have P690 and GS1280.  H Ah - that explains why Itanic out-performed the competition.  But if youJ actually were able to obtain a top-of-the-line 4-processor 1.7 GHz POWER4+D system for comparison I'd be very interested to see the figures:  myL suspicion is that it would at the very least give Itanic a strong challenge,A since in larger system configurations the top-of-the-line POWER4+ L out-performs the top-of-the-line Itanic on a per-processor basis by close to
 50% on TPC-C.   K 4-processor EV7 scores would also be of interest, though Marvel likely only H really comes into its own at 32 processors and above (where it too mightC well beat Itanic on a per-processor basis if HP had any interest in K submitting a TPC-C result for it).  And for that matter if you were able to L test a 2.2 GHz 4-way Opteron box I'd expect its real-world performance to beI not much lower than Itanic's (even though its TPC-C score would likely be  about 25% lower).    - bill   ------------------------------  % Date: Fri, 16 Jan 2004 06:56:19 -0500 * From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: 500.000 AMD64's shipped... 2 Message-ID: <HvSdneWzTsbJTJrdRVn-jQ@metrocast.net>  - "CJT" <abujlehc@prodigy.net> wrote in message $ news:40047376.6030805@prodigy.net... > JF Mezei wrote:  >  > > David Svensson wrote:  > > J > >>Itanium and the mainstream AMD64 CPU's target different markets, so itH > >>is hardly surprising that there are more AMD64 shipped than Itanium.. > >>Anyone believing something else is stupid. > >  > > D > > The problem is that Carly/Curly killed Alpha in order to adopt a commodity , L > > industry standard high volume, low price chip that would also allow them toJ > > reduce the number of architectures That was the whole rationale behind the , > > publicly released excuse for the murder. > > J > > But if IA64 will be restricted to low volume, high cost, niche market, singleI > > source, and only a few customers (HP , SGI), then none of the reasons  thatL > > justified the killing of Alpha to move to IA64 are valid. Alpha was just asH > > viable as IA64 and already in the market, tested, and years ahead of IA64.  > A > Just as in Iraq, there's no way to turn back time even once the 2 > rationale for an action is revealed to be false.  J And just as in Iraq, there's a long-term price for arrogance and perfidy -' which HP is only just beginning to pay.    - bill   ------------------------------  % Date: Fri, 16 Jan 2004 07:07:39 -0500 * From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: 500.000 AMD64's shipped... 2 Message-ID: <W76dnW2UfKdhTprdRVn-vA@metrocast.net>  3 "jlsue" <jefflsxxxz@sbcglobal.net> wrote in message 2 news:82md00ppv9b62lha0ll4trv7hrdpihe1fj@4ax.com...J > On Wed, 14 Jan 2004 17:59:59 -0500, JF Mezei <jfmezei.spamnot@istop.com> > wrote: >  > >jlsue wrote: C > >> So you're saying that it's not valid for HP to reduce 4 server 	 platforms L > >> (PA-RISC, IA32/Proliant, Alpha, MIPS) down to two (and eventually one)? > > I > >No, I am saying it is not right for HP to make such a move under false E > >pretense.  IA64/epic was a pet project that should have never been 
 allowed to$ > >go beyond the pet project status. > ' > Lots of words, all devoid of meaning.   B Rather ironic words from one who's such a bullshit artist himself.  #   There is no such thing as a false  > pretense here.  D Perhaps your memory is failing.  Refresh it by revisiting the c.o.v.H archives for the period immediately following 6/25/01 to see the kind ofK lies Compaq was spinning in attempts to justify its decision on a technical  and financial basis.  7   Almost any chip can succeed given the right amount of J > dedication, commitment, and investment.  It appears that, perhaps, IntelF > had a much better combination of those characteristics for IA64 than Compaq > had for Alpha.  G Well, perhaps it had equal dedication (competence factors in as well) - J *after* it acquired the Alpha team.  But that's an ex post facto argument.J And the rationales offered by Compaq did not include lack of commitment orJ unwillingness to invest, so those factors are irrelevant to the contention= that the rationales that Compaq *did* advance were pure lies.   K Not everything else you said was pure drivel, but I've got better things to K do than respond to it, I'm afraid.  The bottom line is that axing Alpha was J an incompetent move (as well as an unethical one, given the commitments to! its future), merger or no merger.    - bill   ------------------------------  % Date: Fri, 16 Jan 2004 07:18:17 -0500 * From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: 500.000 AMD64's shipped... 2 Message-ID: <XYednbTQjeLnS5rdRVn-jw@metrocast.net>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:sB5lG0CN3lHt@eisner.encompasserve.org...    ...   '   Itanium server prices are going to be 1 > incredible a few years out and are sweet today.   H I don't have any more time to waste on you than I had to waste on jlsue,L Rob.  So I'll just note that Itanic server prices suck compared with OpteronI server prices, and show every indication of continuing to do so.  And one K main part of the reason is that Opteron's architecture makes low-end server L boxes (boards) *very* inexpensive (low chip count, easy to design, one basicK design spans the range from 1 to at least 4 processors) - an advantage that D EV7 shares, had HP any interest in exploiting it.  Combine that withI Opteron's ability to run x86 software at native speed, and Itanic for the # low-end and low-mid-range is toast.   I Axing Alpha was stupid for financial and technical reasons, as well as an J *incredibly* stupid slap in the face to a loyal customer base who had been' assured that Alpha's future was secure.    - bill   ------------------------------  % Date: Fri, 16 Jan 2004 07:28:35 -0500 * From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: 500.000 AMD64's shipped... 2 Message-ID: <0cOdndIkRMZ5RZrdRVn-hg@metrocast.net>  > "Keith Parris" <keithparris_NOSPAM@yahoo.com> wrote in message7 news:cf15391e.0401150652.672d26a3@posting.google.com...    ...   B > Intel has told us they plan to drive Itanium prices down to XeonD > levels, and move the market from 32 bits to 64 bits using Itanium.  L Hmmm.  Intel also told us that Itanic would debut in 1997-8.  And HP said itJ would offer 2x - 3x the performance of its RISC contemporaries.  And laterG that while Merced was really only a development platform (though it had D originally been what that 2x - 3x claim had applied to), sales would@ *really* start to ramp up within 6 months of McKinley's release.  J In other words, betting on Intel's (and HP's) Itanic statements has a long8 and distinguished history of being a losing proposition.   - bill   ------------------------------  % Date: Fri, 16 Jan 2004 07:42:15 -0500 * From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: 500.000 AMD64's shipped... 2 Message-ID: <-aidnZ_rdvOFQZrdRVn-sQ@metrocast.net>  3 "jlsue" <jefflsxxxz@sbcglobal.net> wrote in message 2 news:etmd009bshk583paeg2hrp220bitqkm0e5@4ax.com...   ...   K > You can't count, I guess.  Compaq had Alpha with VMS and Tru64.  They had J > MIPS for Tandem, Intel IA32 for Windows.  Linux perhaps ran on Alpha andK > Intel.  But Windows on Alpha would never have completely replace the need C > for Proliant servers, so you're stuck with 2-3 server businesses.   0 2 server platforms.  Tandem was moving to Alpha.   >  > Now HP will only have ONE.  H No, 2:  Itanic has no plans to offer full-performance x86 execution, and. that won't cease to be relevant any time soon.  .   Again, count it out yourself.  That's a damn# > sight better investment strategy.   L 2 vs. 2?  Your mathematics (or logic) seems a bit flakey, but that's nothing new.   - bill   ------------------------------  % Date: Fri, 16 Jan 2004 12:40:09 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> ' Subject: Re: 500.000 AMD64's shipped... 0 Message-ID: <bu8m3a$ho1$1@new-usenet.uk.sun.com>   Keith Parris wrote: Z > Dirk Munk <munk@home.nl> wrote in message news:<bu5gh4$k0i$1@news4.tilbu1.nb.home.nl>... > S >>However AMD has *one* 32/64 bit architecture for the laptop, the desktop and the  Q >>server environment. Intel will have two very distinctive architectures, 64 bit  B >>for the servers and a rather different 32 bit for anything else. >  > @ > x86-64 may do well on the desktop, but the design is seriously> > compromised for the server space.  Maximum of 8-way (currentH > Superdomes are already at 64-way), 40 bits or 1 TB of physical addressF > space (current Superdomes already support half that address space atE > 512 GB), maximum of 48-bit virtual address space instead of 64-bit,   > and only 16 integer registers.  2 The huge majority of servers sold today are XeonDP based. These support 1-2 CPU's.   1 I keep reminding you that the features you burble - on about have to have benefits. Having 64 way / support in a CPU that is only ever used for 1-2 - CPU's is feature that derives no benefits for 0 the systems buyers, the opposite in fact because it is likely to add cost.   3 Next comes the XeonMP, there is no standard support . in the XeonMP for more than 8 CPU's, hence the0 plethora of different systems architectures used0 to produce >8/4 way XeonMP based systems, but no1 one would say that this has seriously compromised  the XeonMP in the server space.   / And then we get to Opteron and Itanium, Opteron - in fact scores way better than Itanium in the + 1-8 way space because it has Hypertransport + and memory controllers built onto the chip, * its much easier to build a 1-8 way Opteron than a 1-8 way Itanium.   / Above 8 CPU's Opteron and Itanium are identical 0 something that you obviously didn't realise when) you started this particular strand of BS.   0 There is no standard >8 way support for Itanium,- SGI, Unisys, HP have all had to develop their  own platforms for that.   . The rest 40 vs 50 bit addressing is irrelevent/ while Itanium requires more registers than Xeon , or Opteron for a number of reasons including* speculative execution, register sacks etc.   Regards  Andrew Harrison    ------------------------------  % Date: Fri, 16 Jan 2004 14:48:42 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> ' Subject: Re: 500.000 AMD64's shipped... 0 Message-ID: <bu8tka$k7r$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:  >>D >>>When the 1-2 CPU i2000 workstation was announced, and the 1-4 CPUF >>>rx4610 was announced, there was also the oft-forgotten rx9610 which >>>went to 16 CPUs:  >>> < >>>http://www.hp.com/hpinfo/newsroom/press/2001/010529a.html >>>  >  > < >>So long forgotten that it didn't appear to be purchasable. >  >  > It was purchasable.  >    Did anyone buy one ?   Regards  Andrew Harrison  > rick jones   ------------------------------    Date: 16 Jan 2004 09:30:25 -0600+ From: young_r@encompasserve.org (Rob Young) ' Subject: Re: 500.000 AMD64's shipped... 3 Message-ID: <+scjWsm2laeX@eisner.encompasserve.org>   V In article <40075B61.F5FC6340@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes:   > L > One more reason for HP to choose the winning platform its going to bet itsO > business on. It had a good platform with PA-RISC that made HP-UX grow to that O > 20billion dollar business. It jeoperdized its whole business by going all hog G > on an unproven platform that was known to be bloated, problematic and P > expensive (when you factor on the compex compilers needed) instead of  keeping) > Pa-Risc, or joining forced with Alpha.   >   : 	You are either deliberately missing or overlooking where 8 	high-end (read: 64-bit Enterprise) computing is headed.  @ 	Intel has Spink and Bannon and Emer and crew.  It isn't much of@ 	a stretch to assume there is a future Itanium chip that has allB 	the features of EV7 and then some, i.e. a total system on a chip.B 	Dell will sell those SOC very reasonably.  As Fister alludes, you? 	are looking at Xeon pricing (i.e. $3000, $4000) per CPU costs. D 	Margins will be ripped from high-end CPUs.  HP *could* have ploddedB 	along , continuing to develop PA-RISC but they certainly wouldn'tC 	be able to sell PA-RISC for $5000 AND maintain the CPU devleopment 5 	infrastructure.  McNealy can't either and says this:   2 http://www.theregister.co.uk/content/61/34917.html  N McNealy plugged the upcoming Opteron line of servers due out from Sun near the# end of the first calendar quarter.    O "We believe we will be the premier brand for AMD," McNealy said, noting Sun was < late to the market for x86 processors by five to six years.   B While analysts continued to hammer away at when Sun will return toL profitability, during a conference call, McNealy deflected the discussion to near term success.      C 	Yeah - Sun late to the x86 market.  But they are there now as they E 	can't ignore the price differentials anymore - that and performance.   A 	How long before Sun knifes UltraSparc as more of their customers D 	leave for Windows or toys around with x86?  Point here is of course? 	is if Sun moves off UltraSparc in the next 2-5 years all this  A 	whining and crying about knifing Alpha becomes a total borefest,  	right?   D 	HP's business is transitioning to software and services.  As othersD 	have pointed out, having a common hardware base will cut costs. AndF 	yes for sometime it will require engineering to make 8-way and higherE 	systems (key differentiator).  But the risk there of course is Intel A 	sells a very nice OEMed 8-way, 16-way "board" as SOC allows them  	to.   					Rob   ------------------------------    Date: 16 Jan 2004 09:48:47 -0600+ From: young_r@encompasserve.org (Rob Young) ' Subject: Re: 500.000 AMD64's shipped... 3 Message-ID: <ABc++19mibJ9@eisner.encompasserve.org>    In article <bu8m3a$ho1$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > Keith Parris wrote: [ >> Dirk Munk <munk@home.nl> wrote in message news:<bu5gh4$k0i$1@news4.tilbu1.nb.home.nl>...  >>  T >>>However AMD has *one* 32/64 bit architecture for the laptop, the desktop and the R >>>server environment. Intel will have two very distinctive architectures, 64 bit C >>>for the servers and a rather different 32 bit for anything else.  >>   >>  A >> x86-64 may do well on the desktop, but the design is seriously ? >> compromised for the server space.  Maximum of 8-way (current I >> Superdomes are already at 64-way), 40 bits or 1 TB of physical address G >> space (current Superdomes already support half that address space at F >> 512 GB), maximum of 48-bit virtual address space instead of 64-bit,! >> and only 16 integer registers.  > 4 > The huge majority of servers sold today are XeonDP! > based. These support 1-2 CPU's.  > 3 > I keep reminding you that the features you burble / > on about have to have benefits. Having 64 wayJ1 > support in a CPU that is only ever used for 1-2 / > CPU's is feature that derives no benefits forh2 > the systems buyers, the opposite in fact because > it is likely to add cost.. > 5 > Next comes the XeonMP, there is no standard support 0 > in the XeonMP for more than 8 CPU's, hence the2 > plethora of different systems architectures used2 > to produce >8/4 way XeonMP based systems, but no3 > one would say that this has seriously compromised ! > the XeonMP in the server space.p > 1 > And then we get to Opteron and Itanium, Opteron6/ > in fact scores way better than Itanium in thee- > 1-8 way space because it has Hypertransport'- > and memory controllers built onto the chip,e, > its much easier to build a 1-8 way Opteron > than a 1-8 way Itanium.  > 1 > Above 8 CPU's Opteron and Itanium are identical 2 > something that you obviously didn't realise when+ > you started this particular strand of BS.0 >   > 	This is good and Sun's direction in a nutshell.  90% of their> 	future server sales will be pushed to Opteron with UltraSparc= 	occupying a niche - for a while.  Can't sustain that kind of + 	development for a very small niche, right?c   				RobS   ------------------------------    Date: 16 Jan 2004 09:45:48 -0600+ From: young_r@encompasserve.org (Rob Young)E' Subject: Re: 500.000 AMD64's shipped...C3 Message-ID: <kDgL1xodvq+d@eisner.encompasserve.org>a  _ In article <XYednbTQjeLnS5rdRVn-jw@metrocast.net>, "Bill Todd" <billtodd@metrocast.net> writes:s > : > "Rob Young" <young_r@encompasserve.org> wrote in message/ > news:sB5lG0CN3lHt@eisner.encompasserve.org...y >  > ...o > ) >   Itanium server prices are going to beo2 >> incredible a few years out and are sweet today. > J > I don't have any more time to waste on you than I had to waste on jlsue,N > Rob.  So I'll just note that Itanic server prices suck compared with Opteron > server pricesr  A 	But they are a whole lot better than 4-way Power boxes and 4-wayc> 	UltraSparc boxes andr 4-way PA-RISC boxes.  Opteron isn't forA 	everyone as Itanium has quite a few more OEMs committed to usingb* 	it than Opteron does in the server space.   				Robi   ------------------------------  % Date: Fri, 16 Jan 2004 15:35:33 +0000rO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>d' Subject: Re: 500.000 AMD64's shipped... 0 Message-ID: <bu90c5$l7c$1@new-usenet.uk.sun.com>   Rob Young wrote:  P > McNealy plugged the upcoming Opteron line of servers due out from Sun near the% > end of the first calendar quarter. r > Q > "We believe we will be the premier brand for AMD," McNealy said, noting Sun was > > late to the market for x86 processors by five to six years.  > D > While analysts continued to hammer away at when Sun will return toN > profitability, during a conference call, McNealy deflected the discussion to > near term success. l >  > E > 	Yeah - Sun late to the x86 market.  But they are there now as theyAG > 	can't ignore the price differentials anymore - that and performance.n > C > 	How long before Sun knifes UltraSparc as more of their customersuF > 	leave for Windows or toys around with x86?  Point here is of courseA > 	is if Sun moves off UltraSparc in the next 2-5 years all this eC > 	whining and crying about knifing Alpha becomes a total borefest,i	 > 	right?  > F > 	HP's business is transitioning to software and services.  As othersF > 	have pointed out, having a common hardware base will cut costs. AndH > 	yes for sometime it will require engineering to make 8-way and higherG > 	systems (key differentiator).  But the risk there of course is Intel C > 	sells a very nice OEMed 8-way, 16-way "board" as SOC allows them  > 	to. >   2 More attempts to divert attention, how predictable   Regardse Andrew Harrison2   ------------------------------    Date: 16 Jan 2004 09:59:50 -0800' From: icerq4a@spray.se (David Svensson)-' Subject: Re: 500.000 AMD64's shipped...o= Message-ID: <734da31c.0401160959.12dde163@posting.google.com>n  d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<w7-dnVmAuIMaTZrdRVn-uw@metrocast.net>...6 > "David Svensson" <icerq4a@spray.se> wrote in message9 > news:734da31c.0401150736.65c94cd6@posting.google.com...cM > > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>c@ >  wrote in message news:<bu3rct$qce$2@new-usenet.uk.sun.com>... > > > David Svensson wrote:a) > > > > Andrew Harrison SUNUK ConsultancyA: > <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message. > news:<bu1eh4$1ct$1@new-usenet.uk.sun.com>... > > > >  >  gM > > > > It is currently expensive and hotter, but in my real world experienceeN > > > > with testing customer database setups, nothing beats Itanium today andN > > > > that matters. (this includes testing of all current CPU architectures) > > >d > > >eM > > > So you have tried an unspecified DBMS on a Power4+ P690, a USIII 1.2Ghz2L > > > SF Sun and a GS1280. You must have access to a great deal of expensive > > > kit, congratulations.' > > >4 > >o > > Thanks,yJ > > but they are 4-way boxes, so there is no need to have P690 and GS1280. > J > Ah - that explains why Itanic out-performed the competition.  But if youL > actually were able to obtain a top-of-the-line 4-processor 1.7 GHz POWER4+F > system for comparison I'd be very interested to see the figures:  myN > suspicion is that it would at the very least give Itanic a strong challenge,C > since in larger system configurations the top-of-the-line POWER4+sN > out-performs the top-of-the-line Itanic on a per-processor basis by close to > 50% on TPC-C.(  C We got 1.5Ghz POWER4+. 1.7Ghz was not available in the 4-way boxes,.F and I don't think 1.7Ghz is availale now either. I agree though that aB 1.7GHz could be very close to a Itanium 1.5GHz. I do think howeverF that IBM's boxes get good database scores partly because of very large
 L3 caches.  M > 4-processor EV7 scores would also be of interest, though Marvel likely onlyxJ > really comes into its own at 32 processors and above (where it too mightE > well beat Itanic on a per-processor basis if HP had any interest in M > submitting a TPC-C result for it).  And for that matter if you were able tolN > test a 2.2 GHz 4-way Opteron box I'd expect its real-world performance to beK > not much lower than Itanic's (even though its TPC-C score would likely bei > about 25% lower).l   I agree about EV7.   ------------------------------  % Date: Fri, 16 Jan 2004 17:26:17 +0000rO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> ' Subject: Re: 500.000 AMD64's shipped...a0 Message-ID: <bu96ru$nf8$1@new-usenet.uk.sun.com>   Rob Young wrote:a > In article <XYednbTQjeLnS5rdRVn-jw@metrocast.net>, "Bill Todd" <billtodd@metrocast.net> writes:i > : >>"Rob Young" <young_r@encompasserve.org> wrote in message/ >>news:sB5lG0CN3lHt@eisner.encompasserve.org...h >> >>...  >>) >>  Itanium server prices are going to be  >>2 >>>incredible a few years out and are sweet today. >>J >>I don't have any more time to waste on you than I had to waste on jlsue,N >>Rob.  So I'll just note that Itanic server prices suck compared with Opteron >>server pricese >  > C > 	But they are a whole lot better than 4-way Power boxes and 4-way @ > 	UltraSparc boxes andr 4-way PA-RISC boxes.  Opteron isn't forC > 	everyone as Itanium has quite a few more OEMs committed to usinge, > 	it than Opteron does in the server space. >   ? Do you have any idea at all what a 4 way UltraSPARC box costs ?v  ' V440 with 4 CPU's 16GB of RAM = $25,995b  : How much do you think a 4 way Itanium or Power box costs ?  = The HP rx4640 with 4 x 1.3 Mhz Itanium II and 16 GB = $82,501t8 not anywhere close to a 4 way Sun or even a 4 way XeonMP which from HP come in at ~30K.  8 Its better not to just guess Rob, you always get caught.  ; And I liked your last point a lot Itanium may have a numberr9 of OEM's committed to it in the server space HP, SGI, etc     But quantity does not = quality.  < There are 4 main server vendors, Dell, HP, IBM, Sun the rest# are rather incedental to the piece.a  = So for Itanium you have HP, with very very lukewarm offeringsa? from IBM and Dell (IBM for example have chosen Xeon not Itaniumt( for their next big Intel based SMP box).  9 For Opteron you have Sun and IBM with IBM just announcinge! a supercomputer win with Opteron.(  6 eMachines now the 5th largest PC vendor have just made9 announcements Athlon-64. Remember it doesn't matter wherea7 Intels x86-64 processor first surfaces, mobile, desktop 6 or server because where ever it appears spells the end of the Itanium experiment.  : Of course the other problem is that while Itanium may have6 more OEM's than Opteron it has vanishingly tiny levels4 of support from the ISV community while Opteron does not have that problem.  7 Without software the only Itanium OEM's likely to do ok'< will be the ones like SGI focussing on HPC where availablity9 of development tools and libraries is more important thanr availability of applications.l   Regards1 Andrew Harrisona     Regards( Andrew Harrisons	 > 				Rob  >    ------------------------------  % Date: Fri, 16 Jan 2004 17:33:43 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> ' Subject: Re: 500.000 AMD64's shipped...r0 Message-ID: <bu979o$nkt$1@new-usenet.uk.sun.com>   Rob Young wrote: > In article <bu8m3a$ho1$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: >   1 >>And then we get to Opteron and Itanium, Opteron / >>in fact scores way better than Itanium in thel- >>1-8 way space because it has Hypertransport - >>and memory controllers built onto the chip,t, >>its much easier to build a 1-8 way Opteron >>than a 1-8 way Itanium.  >>1 >>Above 8 CPU's Opteron and Itanium are identicala2 >>something that you obviously didn't realise when+ >>you started this particular strand of BS.i >> >  > @ > 	This is good and Sun's direction in a nutshell.  90% of their@ > 	future server sales will be pushed to Opteron with UltraSparc? > 	occupying a niche - for a while.  Can't sustain that kind ofc- > 	development for a very small niche, right?j >   / What does this have to do with the discussion ?q   As an aside.3 Do I need to remind you that your predictions abouto0 Digital and Compaqs strategy and its impact were. so spectacularly wrong that any speculation on/ your part about Sun's strategy and its effects,y6 all be it aimed at distracting attention from Itaniums woes is rather ludicrous.P  5 You are a really terrible Fudster Rob, lots of it andb poor quality to boot.s   Regards  Andrew Harrison    ------------------------------    Date: 16 Jan 2004 09:50:12 -0800' From: icerq4a@spray.se (David Svensson)h' Subject: Re: 500.000 AMD64's shipped... = Message-ID: <734da31c.0401160950.10e597ee@posting.google.com>,   Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bu8etb$f1o$3@new-usenet.uk.sun.com>...s > David Svensson wrote:  > > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bu3rct$qce$2@new-usenet.uk.sun.com>...r > >  > >>David Svensson wrote:@ > >> > >>>Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bu1eh4$1ct$1@new-usenet.uk.sun.com>... > >>>  > >> > >> 1 > >>J > >>>It is currently expensive and hotter, but in my real world experienceK > >>>with testing customer database setups, nothing beats Itanium today and K > >>>that matters. (this includes testing of all current CPU architectures)t > >> > >>K > >>So you have tried an unspecified DBMS on a Power4+ P690, a USIII 1.2GhznJ > >>SF Sun and a GS1280. You must have access to a great deal of expensive > >>kit, congratulations.i > >> > >  > >  > > Thanks,>J > > but they are 4-way boxes, so there is no need to have P690 and GS1280. >  > Which 4 way Sun ?c > Which 4 way IBM ?  > Which 4 way AlphaServer ?h > 	 > Regardse > Andrew Harrisoni  M They were the best 4-way servers available from each vendor in November 2003.    ------------------------------  # Date: Fri, 16 Jan 2004 17:49:00 GMTn9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>D' Subject: Re: 500.000 AMD64's shipped... 3 Message-ID: <guVNb.12507$Ol6.6252@news.cpqcorp.net>e  # "Andrew Harrison SUNUK Consultancy"p > A > Do you have any idea at all what a 4 way UltraSPARC box costs ?o > ) > V440 with 4 CPU's 16GB of RAM = $25,995t >   L Yup.  1/3 the price at 1/3 the performance :)  Just add per-CPU license cost# for your database and stir well ;-)r   ------------------------------  # Date: Fri, 16 Jan 2004 17:49:03 GMTy& From: Rick Jones <foo@bar.baz.invalid>' Subject: Re: 500.000 AMD64's shipped...a3 Message-ID: <juVNb.12508$Ze6.9088@news.cpqcorp.net>r  P Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote: > Rick Jones wrote:r >> It was purchasable. > Did anyone buy one ?  F I've no idea.  I can say with confidence though, that at least as many> rx9610's were purchased as have 8-CPU Opteron systems thusfar.  
 rick jones --  . a wide gulf separates "what if" from "if only"F these opinions are mine, all mine; HP might not want them anyway... :)A feel free to post, OR email to raj in cup.hp.com  but NOT BOTH...1   ------------------------------  % Date: Fri, 16 Jan 2004 18:01:47 +0000mO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>t' Subject: Re: 500.000 AMD64's shipped... 0 Message-ID: <bu98uc$o69$1@new-usenet.uk.sun.com>   David Svensson wrote:d > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bu8etb$f1o$3@new-usenet.uk.sun.com>...u >  >>David Svensson wrote:d >> >>>Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bu3rct$qce$2@new-usenet.uk.sun.com>... >>>  >>>l >>>>David Svensson wrote:e >>>> >>>> >>>>>Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bu1eh4$1ct$1@new-usenet.uk.sun.com>... >>>>>- >>>> >>>> >>>>J >>>>>It is currently expensive and hotter, but in my real world experienceK >>>>>with testing customer database setups, nothing beats Itanium today andWK >>>>>that matters. (this includes testing of all current CPU architectures)t >>>> >>>>K >>>>So you have tried an unspecified DBMS on a Power4+ P690, a USIII 1.2Ghz J >>>>SF Sun and a GS1280. You must have access to a great deal of expensive >>>>kit, congratulations.n >>>> >>>o >>> 
 >>>Thanks,I >>>but they are 4-way boxes, so there is no need to have P690 and GS1280.e >> >>Which 4 way Sun ?r >>Which 4 way IBM ?  >>Which 4 way AlphaServer ?t >>	 >>RegardsD >>Andrew Harrisoni >  > O > They were the best 4-way servers available from each vendor in November 2003.-   You didn't answer the question.h   Model numbers CPU's    Regardss Andrew Harrisons   ------------------------------  % Date: Fri, 16 Jan 2004 18:07:31 +0000cO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>s' Subject: Re: 500.000 AMD64's shipped...t0 Message-ID: <bu9994$o7u$1@new-usenet.uk.sun.com>   Fred Kleinsorge wrote:% > "Andrew Harrison SUNUK Consultancy"  > A >>Do you have any idea at all what a 4 way UltraSPARC box costs ?  >>) >>V440 with 4 CPU's 16GB of RAM = $25,995u >> >  > N > Yup.  1/3 the price at 1/3 the performance :)  Just add per-CPU license cost% > for your database and stir well ;-)e >  >  >   : Why would performance matter even if your point was true ?  = The 1-4 way market is decided pretty much on price, footprint  and application availability.v  = When most people have systems that average 10-20% utilisation B pissing competitions about relative performance thats not measured; in orders of magnitude is the last thing that customers are. interested.i  9 Why is it do you think that Sun has sold more V440's thanf( the entire Itanium market put together ?   Regardsw Andrew Harrisonn   ------------------------------  % Date: Fri, 16 Jan 2004 18:02:43 +0000sO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>n' Subject: Re: 500.000 AMD64's shipped... 0 Message-ID: <bu9903$o69$2@new-usenet.uk.sun.com>   David Svensson wrote:of > "Bill Todd" <billtodd@metrocast.net> wrote in message news:<w7-dnVmAuIMaTZrdRVn-uw@metrocast.net>... > 6 >>"David Svensson" <icerq4a@spray.se> wrote in message9 >>news:734da31c.0401150736.65c94cd6@posting.google.com...d >>L >>>Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> >>@ >> wrote in message news:<bu3rct$qce$2@new-usenet.uk.sun.com>... >> >>>>David Svensson wrote:7 >>>>& >>>>>Andrew Harrison SUNUK Consultancy >>: >><Andrew_No.Harrison_No@nospamn.sun.com> wrote in message. >>news:<bu1eh4$1ct$1@new-usenet.uk.sun.com>... >> >> e >>J >>>>>It is currently expensive and hotter, but in my real world experienceK >>>>>with testing customer database setups, nothing beats Itanium today andsK >>>>>that matters. (this includes testing of all current CPU architectures)  >>>> >>>>K >>>>So you have tried an unspecified DBMS on a Power4+ P690, a USIII 1.2GhzaJ >>>>SF Sun and a GS1280. You must have access to a great deal of expensive >>>>kit, congratulations.P >>>> >>>u
 >>>Thanks,I >>>but they are 4-way boxes, so there is no need to have P690 and GS1280.1 >>J >>Ah - that explains why Itanic out-performed the competition.  But if youL >>actually were able to obtain a top-of-the-line 4-processor 1.7 GHz POWER4+F >>system for comparison I'd be very interested to see the figures:  myN >>suspicion is that it would at the very least give Itanic a strong challenge,C >>since in larger system configurations the top-of-the-line POWER4+oN >>out-performs the top-of-the-line Itanic on a per-processor basis by close to >>50% on TPC-C.v >  > E > We got 1.5Ghz POWER4+. 1.7Ghz was not available in the 4-way boxes,tH > and I don't think 1.7Ghz is availale now either. I agree though that aD > 1.7GHz could be very close to a Itanium 1.5GHz. I do think howeverH > that IBM's boxes get good database scores partly because of very large > L3 caches. >   / Sort of like Itanium II's very large L2 caches.    regardso Andrew Harrison    ------------------------------   Date: 16 Jan 2004 14:51:53 GMT* From: Terry Aardema <taardema@nrcan.gc.ca> Subject: Re: 7.3-1 to 7.3-2t< Message-ID: <Xns94725000F2AECtaardemanrcangcca@132.156.36.9>  I LBohan@mail.esignal.com wrote in news:dnjd00t420v8en4qu2i6rod5cc7pqv3dqr@  4ax.com:   > if the following is true,  ,% > 1.  your 7.3-1 systems disk is ODS5l; > 2.  you plan to make a stand-alone 7.3-1 on a spare disk.s >     > > SYS$SYSTEM:AXPVMS$PCSI_INSTALL_MIN.COM,  may hang (forever?); > unless you rename the following files to all uppercase.  0 > A > [SYS0.SYSCOMMON]DEC-AXPVMS-OpenVMS-V0703-1-5.PCSI$DESCRIPTION;1 9 > [SYS0.SYSCOMMON]DEC-AXPVMS-OpenVMS-V0703-1-5.PCSI$TLB;1f > 7 > I believe I found this soln, only after searching on  = > HP's Technical Knowledge Base (login/registration required)g  @ I'm interested; did you try SET PROCESS/PARSE_STYLE=TRADITIONAL?  
 Terry Aardemae Systems and Network Manager 0 Natural Resources Canada/Canadian Forest Service #include <disclaimer.h>    ------------------------------    Date: 16 Jan 2004 05:29:30 -0800  From: aleurink@chello.nl (riedo) Subject: Alpha server 1000= Message-ID: <bec4f58c.0401160529.1d1dbe0e@posting.google.com>s  D Up for sale in Holland. HP/Compaq Alpha-Server 1000 4/233 with disks and 230 MB memory.D VT510 terminal and a VGA 15 en 17 inch monitor, with 2 keyboards forE Alpha serie LK46W.A2 en LK411-AA. Also a Decserver 90 M with adapter.t   ------------------------------  % Date: Fri, 16 Jan 2004 09:11:55 +0000  From: Roy Omond <Roy@Omond.net> ' Subject: Re: Detaching processes on VMSb( Message-ID: <4007AADB.7030900@Omond.net>   Brian Chase wrote:5 > In article <IKtItxwmx$IS@eisner.encompasserve.org>, $ >  <briggs@encompasserve.org> wrote: >  >  >>$ HELP DISCONNECT1 >>A >>But that's been mentioned before.  Virtual terminals work on my B >>system.  But I use Multinet.  It's not working for OP.  PossiblyA >>because you have to tweak TCP Services to turn it on for telnete >>sessions in that environment.h >  > H > Okay.  Using the examples in the OpenVMS FAQ, I just got this working G > for my rlogin and telnet sessions.  I guess I'll see about getting itsF > to work with SSH, LAT, and locally connected terminals when I get to  > the point of setting those up. > E > Does a disconnected session live indefinitely or are there timeoutsg > I'll need to work around?   . Yes, there's a timeout, but it's easily tamed:     $ mc sysgen  SYSGEN>  SHOW TTY_TIMEOUToC Parameter Name  Current    Default     Min.    Max.   Unit  Dynamici: --------------  -------    -------  ------- -------   ----@ TTY_TIMEOUT        900        900       0       -1   Seconds   D  ! SYSGEN>  HELP SYS_PAR TTY_TIMEOUTr   Sys_Parameters      TTY_TIMEOUT  ?         TTY_TIMEOUT sets the number of seconds before a process0G         associated with a disconnected terminal is deleted. The default G         value (900 seconds) is usually adequate. Note that using valuesuD         for TTY_TIMEOUT greater than one year (value %X01E13380) canH         cause overflow errors and result in a disconnected device timing         out immediately.  +         TTY_TIMEOUT is a DYNAMIC parameter.e    	 Roy Omondu Blue Bubble Ltd.   ------------------------------  % Date: Fri, 16 Jan 2004 12:50:53 +0100a* From: Paul Sture <nospam@sture.homeip.net>' Subject: Re: Detaching processes on VMSl0 Message-ID: <4007DE2D.4B9A0387@sture.homeip.net>   Brian Chase wrote: > 5 > In article <4ZoceGyrjDA5@eisner.encompasserve.org>,d0 > Larry Kilgallen <Kilgallen@SpamCop.net> wrote:, > > In article <bu6qj7$9js$1@pcls4.std.com>,+ > > bdc@world.std.com (Brian Chase) writes:i > C > > > At a fundamental level, does VMS have any provisions to allowlB > > > processes to let go of their terminals without, as Hoff saidI > > > elsewhere in the thread, cleaning up everything associated with the  > > > user's login session?m > >l > >       $ HELP DETACHs > ! > Ah, wishful thinking-- if only.i >   F Yes, this is something I have missed all along in VMS. I'm pretty sureD it was there in RSTS. I used to use a dial-in RSTS system to process> yearly price updates from a supplier, and seem to recall usingG DETACH/ATTACH in a very similar way to the one you described earlier inaD the thread, in particular to keep phone bills to a minimum when longB jobs were running, but also IIRC to recover from lost connections.  H DETACH was also there as a programming statement on the CTS300 system we< used at the time (Time Shared Dibol on top of RT-11). It wasH particularly useful in the case of reporting programs - use the terminalD to enter parameters, then detach and run the (often lengthy) report,? leaving the terminal free for other useful work. If the programnE subsequently did a terminal display (error message or requesting morerH disk space for a spoolfile), it would suspend itself until you issued an ATTACH command.   < Extremely useful, and yes, a shame it didn't make it to VMS.     -- l
 Paul Sture   ------------------------------  + Date: Fri, 16 Jan 2004 13:18:41 +0000 (UTC) % From: bdc@world.std.com (Brian Chase)-' Subject: Re: Detaching processes on VMS4( Message-ID: <bu8obh$lto$1@pcls4.std.com>  2 In article <40071F84.7C23AA8@applied-synergy.com>,1 Chris Scheers  <chris@applied-synergy.com> wrote:s  F > As has been mentioned, the VMS mechanism is CONNECT/DISCONNECT alongE > with virtual terminals (VTDRIVER).  These work for serial ports andnH > (with a bit of finagling) LAT ports.  I don't know whether or not they > work for TELNET ports. > J > The example you showed involved a FT device.  This is a DECterm window. J > I have never seen a virtual terminal associated with a DECterm windows.   > (Although it may be possible.) > I > Since you are using DECterm, why don't you just open as many windows asn! > you need and flip between them?a   Ewwww!  G Admittedly, the functionality I'm looking for is based more on personaleE preference than hard requirements.  I can live without it, but I findt@ noticable efficiency gains when I can corral all my command lineI interfaces into as few windows as possible.  On average, I've around 7-100J CLI sessions running when I'm logged into a system, generally grouped intoH only 2-3 xterms, depending on the underlying use I've designated for theI CLIs.  Given the general GUI/window clutter of modern desktops, I find itkG a lot easier to track down the the right CLI session by using something F like the screen util I mentioned in a significantly smaller number of F parent terminal windows.  My hands spend more time on the keyboard and$ less time fondling that $%&! rodent.  E I'm also not always using locally running xterms (or DECterms) for myrG work.  Quite often I'm remotely logged in over a slow dial-up line, and I it's not at all uncommon for me to be on a real VT320 or VT420.  Having a-F combined detachable, multi-term software terminal manager allows me toH move between all these different environments fluidly and without havingJ to readjust my internal context for different user-interface environments.  J At the moment, I'm not knowledgable enough of VMS for me to need more thanI two sessions open to keep me very occupied.  Being able to disconnect andaG reconnect to a login session using VMS's Virtual Terminals is more thansH enough for my present needs, at least if I can get them to work with SSH connections :-)i  2 Thanks to everyone for all the useful suggestions.   -brian.t -- .F --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ -----5                    Do not fold, mutilate, or spindle.    ------------------------------  % Date: Fri, 16 Jan 2004 14:57:29 +0100v* From: Paul Sture <nospam@sture.homeip.net>' Subject: Re: Detaching processes on VMS 0 Message-ID: <4007FBD9.1F10D2B4@sture.homeip.net>   Brian Chase wrote: > 4 > In article <40071F84.7C23AA8@applied-synergy.com>,3 > Chris Scheers  <chris@applied-synergy.com> wrote:p > H > > As has been mentioned, the VMS mechanism is CONNECT/DISCONNECT alongG > > with virtual terminals (VTDRIVER).  These work for serial ports and1J > > (with a bit of finagling) LAT ports.  I don't know whether or not they > > work for TELNET ports. > > K > > The example you showed involved a FT device.  This is a DECterm window.eK > > I have never seen a virtual terminal associated with a DECterm windows.'" > > (Although it may be possible.) > > K > > Since you are using DECterm, why don't you just open as many windows as # > > you need and flip between them?o >  > Ewwww! > I > Admittedly, the functionality I'm looking for is based more on personaltG > preference than hard requirements.  I can live without it, but I find B > noticable efficiency gains when I can corral all my command lineK > interfaces into as few windows as possible.  On average, I've around 7-10 L > CLI sessions running when I'm logged into a system, generally grouped intoJ > only 2-3 xterms, depending on the underlying use I've designated for theK > CLIs.  Given the general GUI/window clutter of modern desktops, I find ittI > a lot easier to track down the the right CLI session by using something G > like the screen util I mentioned in a significantly smaller number ofpH > parent terminal windows.  My hands spend more time on the keyboard and& > less time fondling that $%&! rodent. > G > I'm also not always using locally running xterms (or DECterms) for my I > work.  Quite often I'm remotely logged in over a slow dial-up line, andeK > it's not at all uncommon for me to be on a real VT320 or VT420.  Having a H > combined detachable, multi-term software terminal manager allows me toJ > move between all these different environments fluidly and without havingL > to readjust my internal context for different user-interface environments. > L > At the moment, I'm not knowledgable enough of VMS for me to need more thanK > two sessions open to keep me very occupied.  Being able to disconnect andoI > reconnect to a login session using VMS's Virtual Terminals is more thanhJ > enough for my present needs, at least if I can get them to work with SSH > connections :-)p > 4 > Thanks to everyone for all the useful suggestions.  + Are you aware of the MultiSessions product?   3 http://www.networkingdynamics.com/MultiSessions.htm   F It's a commercial product, and from the description it looks as thoughD it only works for sessions invoked from a single terminal - i.e. youG can't pick up a session started by another terminal. I have not used itW! myself, so can't comment further.    -- t
 Paul Sture   ------------------------------   Date: 16 Jan 2004 14:58:51 GMT1 From: JONESD@er6.eng.ohio-state.edu (David Jones)i' Subject: Re: Detaching processes on VMSe: Message-ID: <bu8u7b$9v8$1@charm.magnus.acs.ohio-state.edu>  P In message <bu8obh$lto$1@pcls4.std.com>, bdc@world.std.com (Brian Chase) writes:! >... Being able to disconnect andrH >reconnect to a login session using VMS's Virtual Terminals is more thanI >enough for my present needs, at least if I can get them to work with SSHn >connections :-)  E Getting virtual terminal support for the SSH server I wrote was a bit O problematic.  By default, the server creates the interactive processes with theBM same flag used when creating DECterm sessions so you don't get prompted for auO username/password.  Unfortunately, skipping that dialog also makes loginout.exesH skip all the virtual terminal code (binding the PTY to a VT, looking forM disconnected processes, etc).  I was never able to find a supported interfacerJ that lets you do all the magic that loginout does in switching a process's terminal to a VT device.  M The workaround I used was to add a configuration option that makes the serverhL 'log in' the PTY session more conventionally, but hiding the dialog from theL user.  Loginout will then do all the virtual terminal setup (if VT's loaded)D and you can disconnect/reconnect.  The downside is that you must useA username/password authenication and can't use RSA authentication.f  + I can't speak for the SSH EAK for Multinet.       < David L. Jones               |      Phone:    (614) 292-6929- Ohio State University        |      Internet:tL 140 W. 19th St. Rm. 231a     |               jonesd@er6s1.eng.ohio-state.edu: Columbus, OH 43210           |               vman+@osu.edu  1 Disclaimer: I'm looking for marbles all day long.n   ------------------------------    Date: 16 Jan 2004 07:52:00 -0800% From: Bart.Zorn@xs4all.nl (Bart Zorn)e' Subject: Re: Detaching processes on VMSo= Message-ID: <a98cd882.0401160752.24758b7d@posting.google.com>v  b Paul Sture <nospam@sture.homeip.net> wrote in message news:<4007DE2D.4B9A0387@sture.homeip.net>... > H > Yes, this is something I have missed all along in VMS. I'm pretty sureF > it was there in RSTS. I used to use a dial-in RSTS system to process@ > yearly price updates from a supplier, and seem to recall usingI > DETACH/ATTACH in a very similar way to the one you described earlier ineF > the thread, in particular to keep phone bills to a minimum when longD > jobs were running, but also IIRC to recover from lost connections. > J > DETACH was also there as a programming statement on the CTS300 system we> > used at the time (Time Shared Dibol on top of RT-11). It wasJ > particularly useful in the case of reporting programs - use the terminalF > to enter parameters, then detach and run the (often lengthy) report,A > leaving the terminal free for other useful work. If the program.G > subsequently did a terminal display (error message or requesting more J > disk space for a spoolfile), it would suspend itself until you issued an > ATTACH command.e > > > Extremely useful, and yes, a shame it didn't make it to VMS.  E As far as I can remember, RSTS does not have a DETACH command, but iteD has a detach SYS() call which is used by many system programs. ThereF also was a hack available for the terminal driver which allowed you to: type CTRL/D while a program was running to detach the job.   Regards,  	 Bart Zorn>   ------------------------------  % Date: Fri, 16 Jan 2004 10:19:26 -060052 From: "-Andy-" <see2go4me@spamdelicious.yahoo.com>' Subject: Re: Detaching processes on VMSg6 Message-ID: <Xns947273319B22Esee2go4me@216.196.97.132>  = Bart.Zorn@xs4all.nl (Bart Zorn) enlightened us on 16 Jan 2004n with:   7 > As far as I can remember, RSTS does not have a DETACHe; > command, but it has a detach SYS() call which is used by e > many system programs.   = Which 'appears to be' undocumented but is available in BASIC o> in VMS (At least on VAX and Alpha :-) and hopefully Itanium). > [It may be documented somewhere but I had to dig up some RSTS < documentation the last time I had to wrangle a program that 8 uses it (I think Chapter 8 in the RSTS/E V8 programmers 4 manual ?) to get any idea of what is was doing... ].  = Assuming of course this is the same thing. And like with all h undocumented things YMMV.a   -Andy-   ------------------------------  # Date: Fri, 16 Jan 2004 14:35:24 GMTe9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>t) Subject: Re: DHCP/BOOTP/PXE under OPENVMSd3 Message-ID: <MESNb.12491$o16.4940@news.cpqcorp.net>   < "Chris Scheers" <chris@applied-synergy.com> wrote in message- news:4006E644.2BDCFBA6@applied-synergy.com...r > Fred Kleinsorge wrote: > >SL > > A PXE server is planned for Alpha/Itanium OpenVMS in the next release of VMSeC > > (later this year).  Itanium booting will only use PXE (no MOP).c > >f8 > > PXE is a superset of BOOTP, it uses MTFTP, and DHCP. >v2 > Will Itanium OpenVMS still provide a MOP server? >-  H IIRC - you will be able to _serve_ MOP and PXE requests from Itanium andL Alpha.  You will only be able to make PXE _load_ requests from Itanium.  YouI will only be able to make MOP load requests from Alpha (well, technicallywE the SRM console does support BOOTP - but VMS doesn't use it as a loadR method).   ------------------------------  % Date: Fri, 16 Jan 2004 10:43:43 +0000cO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>>Q Subject: Re: Faced with flagging PC sales,  HPand others are pushing aggressivelyt0 Message-ID: <bu8f8v$f1o$6@new-usenet.uk.sun.com>   Bill Gunshannon wrote:2 > In article <100bhi3tik2po18@corp.supernews.com>,4 > 	Greg Cagle <news@removethisgregcagle.com> writes: > 6 >>Bob Koehler said the following on 1/14/2004 1:48 PM: >> >>e >>>In article <100b2n1lp6kdua9@corp.supernews.com>, Greg Cagle <news@removethisgregcagle.com> writes:n >>>r >>>e/ >>>>I assume you're talking about PA-RISC here.- >>>- >>>-K >>>   HP used to talk about HP-PARC and PA-RISC, but always made them sound E >>>   the same.  I always assumed they only had one RISC architecture0C >>>   back then, but two different names for two slightly differento >>>   aspects. >>>:I >>>   Does anyone know what the real meaning/difference is of HP-PARC and  >>>   PA-RISC? >>I >>I've worked with and for HP since the late 80s, and have never heard ofd? >>"HP-PARC". A Google search brings up a few hits, but they arerB >>all obvious typos or errors. An internal search of HP's intranet >>reveals no hits. >  > I > I always thought that HP-PARC stood for tht Hewlett Packard - Palo AltotC > Research Center.  In a report on Venture Capital I find the line:aE >     "Few anchor companies or organizations (eg. HP-PARC, Stanford)"  >    Wasn't that Xeroxn   Regardst Andrew Harrisonn" > But I could be wrong.......  :-) >  > bill >    ------------------------------  # Date: Fri, 16 Jan 2004 11:04:27 GMTiL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")Q Subject: Re: Faced with flagging PC sales,  HPand others are pushing aggressively 6 Message-ID: <00A2BF65.EA5AE14D@SSRL.SLAC.STANFORD.EDU>   In article <bu8f8v$f1o$6@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: >Bill Gunshannon wrote:- >> nJ >> I always thought that HP-PARC stood for tht Hewlett Packard - Palo AltoD >> Research Center.  In a report on Venture Capital I find the line:F >>     "Few anchor companies or organizations (eg. HP-PARC, Stanford)" >> e >d >Wasn't that Xerox  J There's more than one Palo Alto research center.  Xerox PARC is the famousH one.  Digital called theirs WRL; HP's got HP Labs up on Deer Creek Road.   -- Alan-   -- 4O ===============================================================================.0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056eM  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025UO ===============================================================================<   ------------------------------   Date: 16 Jan 2004 18:30:17 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)Q Subject: Re: Faced with flagging PC sales,  HPand others are pushing aggressively19 Message-ID: <bu9ajp$f67mm$1@ID-135708.news.uni-berlin.de>u  0 In article <bu8f8v$f1o$6@new-usenet.uk.sun.com>,R 	Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > Bill Gunshannon wrote: >> tJ >> I always thought that HP-PARC stood for tht Hewlett Packard - Palo AltoD >> Research Center.  In a report on Venture Capital I find the line:F >>     "Few anchor companies or organizations (eg. HP-PARC, Stanford)" >> n >  > Wasn't that Xeroxl >   D Xerox had a "PARC" as well, and consdering that Palo Alto was calledE Silicon Valley I would imagine most of the major players at that timei had research facilities there.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   r   ------------------------------  % Date: Fri, 16 Jan 2004 01:48:50 -0500t* From: JF Mezei <jfmezei.spamnot@istop.com>Y Subject: Re: Faced with flagging PC sales,  HPand others are pushing aggressively into di4) Message-ID: <4007893C.56C3FECD@istop.com>v   Dave Weatherall wrote:H > >    Yes.  PowerPC is bi-endian.  IBM had OS/2 running on it in little% > >    endian mode, possibly WNT too.a > G > Hm! Excellent. Then I second Brian's motion nominating it as the nextr > VMS port target :-)e    J Tough call. Power does offer the best of both worlds, but having the PowerK chips provide the big enterprise features, and PowerPC provide the low costH  desktop and small server market.  D Could the VMS engineers make VMS run on Apple hardware (PowerPC) forK workstation and low end systems , while having it run on Power systems from  IBM for the large scale stuff ?v   ------------------------------    Date: 16 Jan 2004 04:39:17 -0800 From: mb301@hotmail.com (MB)3 Subject: How many file header does indexf.sys have?1= Message-ID: <1d08b916.0401160439.5619bdc7@posting.google.com>r  J How do you find out how many file headers you have in the indexf.sys file?  
 INITIALIZE  
   /HEADERS  "         /HEADERS=number-of-headers  B      Specifies, for disk volumes, the number of file headers to be@      allocated for the index file. The minimum and default value@      is 16. The maximum is the value set with the /MAXIMUM_FILES      qualifier.                           Directory $1$DGA3:[000000]  > INDEXF.SYS;1           18280/74375     21-AUG-2001 09:41:47.93   ------------------------------  % Date: Fri, 16 Jan 2004 08:23:47 -0600o( From: brandon@dalsemi.com (John Brandon)7 Subject: Re: How many file header does indexf.sys have?B1 Message-ID: <04011608234713@dscis6-0.dalsemi.com>e  L > How do you find out how many file headers you have in the indexf.sys file?  L I have a disk drives that was just initialized with /HEADERS=400000 and this is the result:   Directory $1$DGA5:[000000]  P INDEXF.SYS;1                               77193/402183  15-JAN-2004 15:24:45.17       J*o*h*n B*r*a*n*d*o*nu VMS Systems Administratorn* firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Fri, 16 Jan 2004 09:34:28 -0500a From: norm.raphael@metso.com7 Subject: Re: How many file header does indexf.sys have?gQ Message-ID: <OF617730D8.79049914-ON85256E1D.004F8A9D-85256E1D.00507829@metso.com>a   OpenVMS V7.3-1 Alpha, ODS2   From DFU Report MA101a  F       ***** Volume info for ODS2 volume MA101: (from HOME block) ******  Volume name                      :  MA101#  Volume owner                     : #  Volume set name                  :r,  Highwater mark. / Erase on del.  :  No / No'  Cluster size                     :  10n+  Maximum # files                  :  900000.)  Header count                     :  2010B(  First header VBN                 :  261(  Free headers                     :  845   Also  , $ DUMP/HEAD/BL=EN=0 MA101:[000000]INDEXF.SYS    B Dump of file MA101:[000000]INDEXF.SYS;1 on 16-JAN-2004 08:35:43.83; File ID (1,1,0)   End of file block 2270 / Allocated 450260p  (                              File Header   [snip]   Map area     Retrieval pointers0         Count:         20        LBN:          00         Count:         10        LBN:       10300         Count:         10        LBN:   182292900         Count:     450220        LBN:   17779070  / Checksum:                                 38409w  F This tells me I have 450,000 preallocated headers and a max of 900,000 files.  C brandon@dalsemi.com (John Brandon) wrote on 01/16/2004 09:23:47 AM:   H > > How do you find out how many file headers you have in the indexf.sys file?  > I > I have a disk drives that was just initialized with /HEADERS=400000 andr this > is the result: >  > Directory $1$DGA5:[000000] >pF > INDEXF.SYS;1                               77193/402183  15-JAN-2004
 > 15:24:45.17c >< >m >h > J*o*h*n B*r*a*n*d*o*nm > VMS Systems Administratoru, > firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Fri, 16 Jan 2004 12:44:47 -0500c& From: David M Smith <dsmit115@csc.com>7 Subject: Re: How many file header does indexf.sys have?e8 Message-ID: <bl8g009d660dp7rvldckmptj4cvfjr7ib1@4ax.com>  < On 16 Jan 2004 04:39:17 -0800, mb301@hotmail.com (MB) wrote:  K >How do you find out how many file headers you have in the indexf.sys file?d >'  N Do you mean how many are actually there right now, and/or in use right now, orG the maximum you could (theoretically) reach? All these numbers could ber
 different!  P Why do you care, in other words what problem are you trying to solve? I might beO able to give better feedback. This is an interesting topic, but there are a lote% of "wrinkles" of which to be aware...qI -------------------------------------------------------------------------iI David M. Smith 302.391.8533                       dsmit115 at csc dot comnI Computer Sciences Corporation     (Opinions are those of the writer only)iI -------------------------------------------------------------------------    ------------------------------  % Date: Fri, 16 Jan 2004 10:10:27 -0600o( From: Wayne Sewell <wayne@tachysoft.com>M Subject: Re: HP Integrity Superdome server sets world record for 10-terabyte,p0 Message-ID: <00A2BFA1.6D9CA42A.13@tachysoft.com>  - >From: Paul Repacholi <prep@prep.synonet.com>0 >X-Newsgroups: comp.os.vmsN >Subject: Re: HP Integrity Superdome server sets world record for 10-terabyte,H >         single-system TPC-H benchmark 10-terabyte, single-system TPC-H> >         benchmark 10-terabyte, single-system TPC-H benchmark& >Date: Fri, 16 Jan 2004 22:39:44 +0800  ? >"Peter Weaver" <WeaverConsultingServices@sympatico.ca> writes:i >vE >> HP is simply the current owner of VMS. I do not buy VMS because it.F >> has the HP logo on it, I buy VMS because it is VMS! Today I am a HPG >> customer because HP owns VMS, if I woke up tomorrow to the news thatn< >> CA bought out VMS then I would be a CA customer. Period." >nG >Thanks for reminding us all that in fact it could be worse than it is.a >r    O Yeah, that crossed my mind too.  Never dealt with CA myself, but I've heard allh0 the horror stories about good products gone bad.  M True, I would also prefer vms be bought by somebody not also mired in eunuchseN and the billyworld, and who would allow it to compete, but somebody other than CA.e   WaynexO ===============================================================================cN Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html   oO ===============================================================================1B Jed Clampett, checking into hotel: "This place got a cement pond?", 	Ellie May: "And do yuh let critters in it?"   ------------------------------  % Date: Fri, 16 Jan 2004 22:39:44 +0800S, From: Paul Repacholi <prep@prep.synonet.com>Y Subject: Re: HP Integrity Superdome server sets world record for 10-terabyte, single-systt- Message-ID: <87k73sufdb.fsf@prep.synonet.com>n  > "Peter Weaver" <WeaverConsultingServices@sympatico.ca> writes:  D > HP is simply the current owner of VMS. I do not buy VMS because itE > has the HP logo on it, I buy VMS because it is VMS! Today I am a HP0F > customer because HP owns VMS, if I woke up tomorrow to the news that; > CA bought out VMS then I would be a CA customer. Period."o  F Thanks for reminding us all that in fact it could be worse than it is.   -- o< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 16 Jan 2004 02:34:00 -0800. From: fabiopenvms@yahoo.com.br (Fabio Cardoso)F Subject: Re: HP rockets to No. 5 on list of top U.S. patent recipients= Message-ID: <f30679fb.0401160234.7f772931@posting.google.com>.  v keithparris_NOSPAM@yahoo.com (Keith Parris) wrote in message news:<cf15391e.0401141439.7721b9e3@posting.google.com>...; > http://www.hp.com/hpinfo/newsroom/press/2004/040112c.html  > H > HP Announces Intellectual Property Licensing Group to Leverage Growing > Patent Portfolio > ; > HP rockets to No. 5 on list of top U.S. patent recipients> > " > PALO ALTO, Calif., Jan. 12, 2004 > > > Seeking to leverage its rapidly growing patent portfolio, HPF > (NYSE:HPQ) - the company with "Invent" in its logo - today announcedD > it has formed an intellectual property (IP) licensing organizationH > designed to increase revenue and improve technical collaborations with > partners.  > C > The company moved up sharply in the annual list of top 10 privatenG > sector organizations receiving the most U.S. patents announced by thenF > U.S. Patent and Trademark Office today, ranking No. 5, up from No. 9D > the previous year. HP was awarded 1,759 U.S. patents in 2003, a 270 > percent increase over the 2002 level of 1,385. > A > "We're proud of the fact that our focused innovation strategy - D > investing in technologies where we can lead and partnering for theE > rest - is paying off," said Shane Robison, executive vice presidenttG > and chief strategy and technology officer, HP. "Over time, we believetH > that the new IP we are generating will help contribute to top-line and? > bottom-line growth in both established and emerging markets.": > C > Worldwide, HP increased its patent portfolio from about 17,000 to_F > 21,000 in 2003, according to Steve Fox, HP vice president and deputy@ > general counsel, intellectual property. With 4,000 new patents@ > worldwide last year, HP earned an average of 11 patents a day. > ...l    P How many Alphaservers, OpenVMS (related products) and Itanium-servers patents ?    Regards    FC   ------------------------------  # Date: Fri, 16 Jan 2004 15:04:39 GMTt9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> F Subject: Re: HP rockets to No. 5 on list of top U.S. patent recipients4 Message-ID: <b4TNb.12493$p76.12052@news.cpqcorp.net>  ; "Fabio Cardoso" <fabiopenvms@yahoo.com.br> wrote in messaget7 news:f30679fb.0401160234.7f772931@posting.google.com...r> > keithparris_NOSPAM@yahoo.com (Keith Parris) wrote in message9 news:<cf15391e.0401141439.7721b9e3@posting.google.com>...s= > > http://www.hp.com/hpinfo/newsroom/press/2004/040112c.htmle > >iJ > > HP Announces Intellectual Property Licensing Group to Leverage Growing > > Patent Portfolio > >1= > > HP rockets to No. 5 on list of top U.S. patent recipientsa > >A$ > > PALO ALTO, Calif., Jan. 12, 2004 > >j@ > > Seeking to leverage its rapidly growing patent portfolio, HPH > > (NYSE:HPQ) - the company with "Invent" in its logo - today announcedF > > it has formed an intellectual property (IP) licensing organizationJ > > designed to increase revenue and improve technical collaborations with
 > > partners.6 > >4E > > The company moved up sharply in the annual list of top 10 privatenI > > sector organizations receiving the most U.S. patents announced by theqH > > U.S. Patent and Trademark Office today, ranking No. 5, up from No. 9F > > the previous year. HP was awarded 1,759 U.S. patents in 2003, a 272 > > percent increase over the 2002 level of 1,385. > >eC > > "We're proud of the fact that our focused innovation strategy -tF > > investing in technologies where we can lead and partnering for theG > > rest - is paying off," said Shane Robison, executive vice presidentuI > > and chief strategy and technology officer, HP. "Over time, we believe J > > that the new IP we are generating will help contribute to top-line andA > > bottom-line growth in both established and emerging markets."i > >nE > > Worldwide, HP increased its patent portfolio from about 17,000 to H > > 21,000 in 2003, according to Steve Fox, HP vice president and deputyB > > general counsel, intellectual property. With 4,000 new patentsB > > worldwide last year, HP earned an average of 11 patents a day. > > ...a >o >'G > How many Alphaservers, OpenVMS (related products) and Itanium-serversd	 patents ?y >a  H Alpha, and AlphaServers generated a lot of Patents.  I remember a PatentI award dinner where one of the AlphaServer designers kept having to get upaH for endless awards - he had something like 32 patents that year.  I felt good to have 2.   J Software patents are a strange duck - was what you wrote just a clever wayL to do something?  Or is it something that should be patented (along with theH resultant paperwork, searches for prior art, and patent lawyers)?  We goK through cycles where an organization gets serious about them, but "usually"yF we aren't as focused on them as part of the job as hardware engineers.  J Galaxy recieved at least a good half dozen, including the first one I knowF of for Soft Partitioning.  Over the years, there have been a number ofL interesting ones in VMS, usually by engineers who had a hardware engineeringJ background and knew the process.  We have several on memory error handlingI that was done on the VAX 9000 for instance.  We even have one on a methodeI for colormap/window layering in X11 on VAX (now *that's* an odd one).  In H general, the patent count is pretty low for the current engineers in the9 organization, with the high being someone with around 10.5  I HP is a bit serious about them, and with some upcomming work - who knows,;1 maybe there will be a new spurt of them from VMS.e   ------------------------------  % Date: Fri, 16 Jan 2004 01:45:35 -0500g* From: JF Mezei <jfmezei.spamnot@istop.com>: Subject: Re: Humongus bug in DECTERM (postscript printing)) Message-ID: <4007887A.38FFF38E@istop.com>d   Mike Naime wrote:iN > Are you positive that your printer supported postscript, and that it was not8 > the postscript output that was causing all the errors?  Q Yes, it is a declaser 5100, thorugh DCPS, printer set to support postscript only.p  G The problem is that DECterm doesn't have code that generates postscript N output. But the printer dialogue allows you to choose postscript output, which< is disregarded by DECTERm which still generates ansi output.   A good way to test this:  M select text on DECTerm, then choose to print selected text. Choose postscriptOJ format, and send it to a queue that is defined to be postscript input (andN output). Job *should* print fine, but generates errors on the first word since( what you sent to printer was ascii text.  N Select same text, then choose postscript format and send it to a queue definedM to be ANSI by default. If DECTERM generated postscript output, the postscriotiN source should print since DCPS would treat it as a text file and convert it to5 poscript code that would print the postscript source.k  T Alternatively, you can simply choose to print to file and look at the file contents.   ------------------------------    Date: 16 Jan 2004 06:27:42 -0800. From: martinkirby12@yahoo.co.uk (Martin Kirby): Subject: Re: Humongus bug in DECTERM (postscript printing)< Message-ID: <224291b.0401160627.3bf9c6bf@posting.google.com>  C The queued printer options dialog is a standard DXmPrintWgt widget.aE That widget has resources to control the print formats. Therefore therF DECterm code, or UIL, should be setting the list to just those DECterm can produce output for.-  B The help on the Options button talks about setting Line printer or? Terminal formats. So I wonder if there used not to be all these D formats supported by the widget and they were added later. A look atF the cms libraries suggests that if that is the case it happened before 1990.e   Martin Kirby  [ JF Mezei <jfmezei.spamnot@istop.com> wrote in message news:<40066411.15900218@istop.com>...e< > Digital/Compaq/HP are guilty of destroying a whole forest. > N > On DECTERM (VAX VMS 7.2), you can select text and ask to print it. The printO > dialogue asks you what format you want to print, and this include Postscript.- > N > So, figuring I was printing to a postscript printer, I chose "postcript" and5 > queued the print job to the postscript print queue.o > O > Low and behold, paper started spewing out of the printer indicating all sorts: > of errors.... :-)Z > P > Seems that even if you specify postscript output, DECTERM still generates just
 > ASCII text.c > M > perhaps DECTERM should override this and add a /DATA=ANSI flag to the print U > job. (or remove the postscript option if it is incapable of generating postscript).o   ------------------------------  + Date: Fri, 16 Jan 2004 17:53:29 +0000 (UTC)cP From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply): Subject: Re: Humongus bug in DECTERM (postscript printing)$ Message-ID: <bu98ep$qai$1@online.de>  F In article <BQKNb.44442$VV4.13220@twister.rdc-kc.rr.com>, "Mike Naime" <mnaime@kc.rr.com> writes: o   > 7 > JF Mezei <jfmezei.spamnot@istop.com> wrote in messagea% > news:40066411.15900218@istop.com...a> > > Digital/Compaq/HP are guilty of destroying a whole forest. > A > If you really want to see destroying a forest, look at what thetN > Government/Military does.  What you are describing is nothing in comparison.M > I had several printers that printed 24 hours a day for 7 days strait.  TheynM > only time that they stopped was when you changed out the paper, or ribbons. N > In the early incarnation, there was a blank page printed between each singleJ > page report so that it could be read easier.  Now that is wasting paper!  C I remember some guys testing a new A0 (i.e. one-square-meter pages)iF printer connected to a unix system.  They forgot the "-h" to suppress  the header page....x   ------------------------------    Date: 16 Jan 2004 02:09:46 -0800. From: fabiopenvms@yahoo.com.br (Fabio Cardoso)E Subject: Re: IBM announces DB2 for HP Integrity Itanium-based serversB= Message-ID: <f30679fb.0401160209.3d9b6aa9@posting.google.com>-  a Paul Repacholi <prep@prep.synonet.com> wrote in message news:<87y8s9udq8.fsf@prep.synonet.com>... 5 > keithparris_NOSPAM@yahoo.com (Keith Parris) writes:e > @ > > It will be interesting to see, as the UNIX/Linux PortabilityC > > Initiative makes such tasks easier, if IBM eventually announcesi* > > support for DB2 on OpenVMS on Itanium. > 3 > They had both DB2 and CICS for Vax way back when.m  ; Hmmm ! I am imaginig a SuperDome running OpenVMS/RDB in one 0 partition and MVS/DB2 in another partition ! :-)   RegardsC   FC   ------------------------------    Date: 16 Jan 2004 01:02:25 -0800$ From: chrisgaze@msn.com (Chris Gaze)  Subject: Re: IP Address problems= Message-ID: <57512a77.0401160102.4afe92d4@posting.google.com>    Michael Austin <maustin@no-more-spam.firstdbasource.com> wrote in message news:<T%INb.8163$Ea6.738@newssvr22.news.prodigy.com>...g > Chris Gaze wrote:l >  > > Hi,n > > @ > > I have a problem with my Alpha server and a router I have.   > > = > > I have set up two ip addresses on the alpha (on one nic).a0 > > One address, the main one, is 192.168.10.102% > > The alias address is 172.16.26.64e- > > The address of my router is 172.16.26.126f > > E > > When I first boot up the alpha everything works fine. After aboutsH > > 15-20 mins I am unable to contact the router, all my pings time out. > > H > > If i ping an address that is not in use on the 172.16.26.0 network II > > get, as expected, time outs.  However after pinging an unused address ) > > I am able to contact my router again.f > > 3 > > I have no problems on the 192.168.10.0 network.o > > # > > Any ideas would be appriciated.m > > 
 > > Thanks >  > : > What is your default gateway and give me the output from >  > $TCPIP SHOW ROUTE  > $TCPIP SHOW INTERFACE/CLUSTERe >  > Which router?e > $ > if these routes do not exist, try: > D > TCPIP> set route 172.16.26.0/net/mask=24/network/gate=172.16.26.64I > TCPIP> set route 172.16.26.0/net/mask=24/network/gate=172.16.26.64/perme >  > Michael Austin   Hi Michael,    Thanks for your help.   / The output of $TCPIP SHOW ROUTE is as follows :u  9 AH    10.100.100.100                        172.16.26.126e5 AH    127.0.0.1                             127.0.0.1i8 AN    172.16.26.0/25                        172.16.26.658 AH    172.16.26.65                          172.16.26.659 AN    172.21.0.0/16                         172.16.26.126d9 AH    192.168.2.42                          172.16.26.126 : AN    192.168.10.0/24                       192.168.10.102: AH    192.168.10.102                        192.168.10.1029 AH    194.202.236.23                        172.16.26.126 : AH    217.199.170.114                       192.168.10.254: AH    217.199.170.191                       192.168.10.254  ; The output of $TCPIP SHOW INTERFACE/CLUSTER is as follows :   O Interface   IP_Addr         Network mask          Receive          Send     MTUr  O  LO0        127.0.0.1       255.0.0.0               39515         39515    4096oO  WE0        192.168.10.102  255.255.255.0         2412292       1915754    1500TO  WEA0       172.16.26.65    255.255.255.128       2412292       1915754    1500s     Thanks again for your help   Christ   ------------------------------  % Date: Fri, 16 Jan 2004 05:09:36 -0500t* From: JF Mezei <jfmezei.spamnot@istop.com>  Subject: Re: IP Address problems) Message-ID: <4007B83B.1C031815@istop.com>t   Chris Gaze wrote:b; > AH    10.100.100.100                        172.16.26.126 7 > AH    127.0.0.1                             127.0.0.1y: > AN    172.16.26.0/25                        172.16.26.65: > AH    172.16.26.65                          172.16.26.65; > AN    172.21.0.0/16                         172.16.26.126e; > AH    192.168.2.42                          172.16.26.126l< > AN    192.168.10.0/24                       192.168.10.102< > AH    192.168.10.102                        192.168.10.102; > AH    194.202.236.23                        172.16.26.126 < > AH    217.199.170.114                       192.168.10.254< > AH    217.199.170.191                       192.168.10.254  M Is there a reason you do not have a default route ? Default route shows up asnJ 0.0.0.0 on the first (left) IP address in the above display and would haveK "AN" line type. With a default route, you'd be able to remove all the routelO entried pointing to the router sicn they woudl be handled by the default route.l   Also, doesn't:: > AN    172.16.26.0/25                        172.16.26.65  D direct any connection attempt to 172.16.26.126 (your router) through 172.16.26.25 ?   ------------------------------    Date: 16 Jan 2004 06:51:06 -0600 From: briggs@encompasserve.org  Subject: Re: IP Address problems3 Message-ID: <7w3ZkuiMe+bV@eisner.encompasserve.org>o  d In article <57512a77.0401151412.39db3c07@posting.google.com>, chrisgaze@msn.com (Chris Gaze) writes: > Hi,  > > > I have a problem with my Alpha server and a router I have.   > ; > I have set up two ip addresses on the alpha (on one nic).r. > One address, the main one, is 192.168.10.102# > The alias address is 172.16.26.64S+ > The address of my router is 172.16.26.126d > C > When I first boot up the alpha everything works fine. After aboutkF > 15-20 mins I am unable to contact the router, all my pings time out. > F > If i ping an address that is not in use on the 172.16.26.0 network IG > get, as expected, time outs.  However after pinging an unused address ' > I am able to contact my router again.t > 1 > I have no problems on the 192.168.10.0 network.s  C Definitely on the strange side.  But it sounds like your NIC is not 6 responding to ARP requests for its 172.16.26.64 alias.  E When you first boot, you send out a "gratuitous ARP" on both aliases.tC So all the other equipment on the 172.16 net caches your ARP entry.   D 15-20 minutes later, those entries are timing out.  You can get yourD echo requests outbound because your ARP cache is still solid (you'llA accept ARP's in the 172.16 space, even though you won't respond).iC But replies to you can't make it because the ARP cache on the othersD end is empty entry and because your machine isn't responding to ARP.  B If you ping an unused IP address, that forces your machine to sendB out an ARP request.  All the other equipment on the segment caches5 that entry and you're good for another 15-20 minutes.e  A You could verify that diagnosis with a sniffer.  But I don't havei" any idea what you can do about it.   	John Briggs   ------------------------------  % Date: Fri, 16 Jan 2004 14:28:53 +0100 - From: "Winfried Bergmann" <dummy@empuron.com>u  Subject: Re: IP Address problems9 Message-ID: <bu8oqv$ehblk$1@ID-170759.news.uni-berlin.de>   1 <briggs@encompasserve.org> schrieb im Newsbeitragd- news:7w3ZkuiMe+bV@eisner.encompasserve.org... ? > In article <57512a77.0401151412.39db3c07@posting.google.com>, & chrisgaze@msn.com (Chris Gaze) writes: > > Hi,n > > > > > I have a problem with my Alpha server and a router I have. > > = > > I have set up two ip addresses on the alpha (on one nic). 0 > > One address, the main one, is 192.168.10.102% > > The alias address is 172.16.26.64u- > > The address of my router is 172.16.26.126  > >oE > > When I first boot up the alpha everything works fine. After about8H > > 15-20 mins I am unable to contact the router, all my pings time out. > > H > > If i ping an address that is not in use on the 172.16.26.0 network II > > get, as expected, time outs.  However after pinging an unused address@) > > I am able to contact my router again.m > >g3 > > I have no problems on the 192.168.10.0 network.t >sE > Definitely on the strange side.  But it sounds like your NIC is noty8 > responding to ARP requests for its 172.16.26.64 alias. >EG > When you first boot, you send out a "gratuitous ARP" on both aliases. E > So all the other equipment on the 172.16 net caches your ARP entry.  >sF > 15-20 minutes later, those entries are timing out.  You can get yourF > echo requests outbound because your ARP cache is still solid (you'llC > accept ARP's in the 172.16 space, even though you won't respond).tE > But replies to you can't make it because the ARP cache on the other F > end is empty entry and because your machine isn't responding to ARP. >vD > If you ping an unused IP address, that forces your machine to sendD > out an ARP request.  All the other equipment on the segment caches7 > that entry and you're good for another 15-20 minutes.y >iC > You could verify that diagnosis with a sniffer.  But I don't havee$ > any idea what you can do about it.  L If that's the problem, it might be possible to enter the ethernet address ofL the affected NIC manually in the ARP cache of the router and other machines,I directly connected in this subnet. I wouldn't assume, manual entries time  out in the ARP cache.h   ------------------------------    Date: 16 Jan 2004 07:50:43 -0600 From: briggs@encompasserve.org  Subject: Re: IP Address problems3 Message-ID: <ZOqBjeqgC82N@eisner.encompasserve.org>e  i In article <bu8oqv$ehblk$1@ID-170759.news.uni-berlin.de>, "Winfried Bergmann" <dummy@empuron.com> writes:h3 > <briggs@encompasserve.org> schrieb im Newsbeitrag / > news:7w3ZkuiMe+bV@eisner.encompasserve.org...,F >> Definitely on the strange side.  But it sounds like your NIC is not9 >> responding to ARP requests for its 172.16.26.64 alias.  ...SN > If that's the problem, it might be possible to enter the ethernet address ofN > the affected NIC manually in the ARP cache of the router and other machines,K > directly connected in this subnet. I wouldn't assume, manual entries timeV > out in the ARP cache.   E Yes.  A static ARP on the router would work.  You could even have thet router do proxy ARP for you.   Example (Cisco router):M  ( arp 10.40.96.2 00a0.8e31.980e ARPA alias    A The "alias" keyword means that the router will not only stick themB entry in its own ARP cache but it will also respond to ARP queriesB for it, thus advertising the entry to the rest of the LAN segment.# (Which is one flavor of proxy ARP).c   	John Briggs   ------------------------------  # Date: Fri, 16 Jan 2004 14:43:00 GMTr9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>pY Subject: Re: Judge Upholds Jury Decision In Microsoft Web-Browser Patent Case - U.S. PateX3 Message-ID: <ULSNb.12492$U16.5528@news.cpqcorp.net>t  + "Dirk Munk" <munk@home.nl> wrote in messageV, news:bu74ho$oaj$1@news2.tilbu1.nb.home.nl... >w > >e > > I > > Meanwhile, the U.S. Patent and Trademark Office announced in Novembera thatK > > it is reconsidering the patent. The decision followed warnings that anyoF > > Microsoft redesign of its software would render many Web pages and products3 > > of independent software designers incompatible.- > >-C > Wow, that would really be inconceivable! A M$ product that is notr
 compatibleD > with a previous version of the same M$ product. Or even worse, not
 compatible9 > with the official standards! No one has ever seen that.t >e  G What I like it that MS is big enough to threaten the Patent Office.  MS & could simply *license* the technology.   ------------------------------  # Date: Fri, 16 Jan 2004 16:05:53 GMT ' From: Rick Dyson <rick-dyson@uiowa.edu>:, Subject: Minimum Privs for Changing Password1 Message-ID: <HrLBDu.DGq@sysadm.physics.uiowa.edu>e  H This is a fundamental question and I am embarassed to have to ask, but I am in a hurry. :)   H What minimum privs would a user need to be able to change someone else's password via Authorize?y  E Does anyone have any quick suggestions on a method for this that doesdC not need to elevate a low user to SYSPRV (or such)?  Something that # could be up and running right away?i   Thanks!c rick   ------------------------------  % Date: Fri, 16 Jan 2004 12:46:04 -0500 & From: David M Smith <dsmit115@csc.com>0 Subject: Re: Minimum Privs for Changing Password8 Message-ID: <op8g00hdg8elqipms3s0avh0h4gnjtt2tc@4ax.com>  J On Fri, 16 Jan 2004 16:05:53 GMT, Rick Dyson <rick-dyson@uiowa.edu> wrote:  I >This is a fundamental question and I am embarassed to have to ask, but Ir >am in a hurry. :) >fI >What minimum privs would a user need to be able to change someone else'sS >password via Authorize? > F >Does anyone have any quick suggestions on a method for this that doesD >not need to elevate a low user to SYSPRV (or such)?  Something that$ >could be up and running right away? >S >Thanks! >rick(  C SYSPRV or BYPASS. There is no reasonable (secure) method otherwise.gI -------------------------------------------------------------------------AI David M. Smith 302.391.8533                       dsmit115 at csc dot comnI Computer Sciences Corporation     (Opinions are those of the writer only).I -------------------------------------------------------------------------A   ------------------------------   Date: 16 Jan 04 13:49:05 EST From: grant@rigel.cc.wmich.edu0 Subject: Re: Minimum Privs for Changing Password/ Message-ID: <lDvjaEDRDHht@mcduck.acs.wmich.edu>   [ In article <HrLBDu.DGq@sysadm.physics.uiowa.edu>, Rick Dyson <rick-dyson@uiowa.edu> writes:sJ > This is a fundamental question and I am embarassed to have to ask, but I > am in a hurry. :)= > J > What minimum privs would a user need to be able to change someone else's > password via Authorize?S > G > Does anyone have any quick suggestions on a method for this that does E > not need to elevate a low user to SYSPRV (or such)?  Something thatH% > could be up and running right away?= > 	 > Thanks!= > rick >  You need SYSPRV or equivalent.  C If you want someone to be able to do this without giving them other3> abilities which would be granted by SYSPRV, there are options:G 1) Write a privileged application, which will do the necessary updates,eD    and control access to the application, either with ACLs, internal    checks or both.B 2) Provide a captive account which restricted group of people haveA    password for. Let it prompt for account and password, and then ?    do AUTHORIZE MODIFY account/PASSWORD=newpassword [/NOPWDEXP]P  B Option 1 has the advantage that you can actually make the requiredE tests of HISTORY, DICTIONARY, and any password policy in a reasonablet way.  E Option 2 can be done very quickly -- but make as many validity checks  as possible, anyway.   ------------------------------  % Date: Fri, 16 Jan 2004 13:30:08 +0000 - From: John Laird <nospam@laird-towers.org.uk>i9 Subject: Re: Mounting identically labelled disks (OVMS73)k8 Message-ID: <a5pf00lasluemm9aikdqnnh9uvhrufms4l@4ax.com>  5 On Thu, 15 Jan 2004 21:30:35 +0000 (UTC), News Readera# <read.news@hostname.invalid> wrote:   I >Now there really is no issue about confusing identically labelled disks!aF >No, really.  On VMS I can distinguish them easily - they're differentG >devices, and on Unix I can distinguish them easily - they've differents >mount points. >nA >So, how to stop VMS's nannying; letting me nfs share identicallyl >labelled disks?  K If you are prepared to forego the file and record locking provided for free I by VMS, then sure there is no issue with identically labelled disks.  But E most of us like having our data protected (which only comes across aseI nannying to the terminally insane), and afaik you cannot disable locking.hH So, shared devices must have unique volume labels within their "domain".J Domain is process, group or system, so you can have three instances of the "same" disk.  J I think there are sound technical reasons why the device name is not used,E but can't offhand think of what they are.  Better minds than mine can 	 explain. J   --  7 For people who like peace and quiet: a phoneless cord. w   Mail john rather than nospam...a   ------------------------------  + Date: Fri, 16 Jan 2004 13:43:33 +0000 (UTC)u0 From: News Reader <news.reader@hostname.invalid>9 Subject: Re: Mounting identically labelled disks (OVMS73)E/ Message-ID: <bu8pq5$vfq$1@grandcanyon.binc.net>v  - On Thu, 15 Jan 2004 23:11:45 -0500, JF wrote:l  E >mount one disk without the /SYSTEM or /CLUSTER parameter. It will be F >mounted privately. You can then SET VOLUME/LABEL=newlabel to make theF >drive look different. Unmount and you can remount the drive/SYSTEM at( >which point it should become available.  ? THANKS JF, appreciate the help and the set volume /label worked4A however looks like more is needed to stop VMS's nannying; lettingw; me share nfs share (originally) identically labelled disks.0  4 More solutions please! - The story so far follows...   $ ! the original, still on-lineh $ show dev dad5?   < Device           Device 	  Error    Volume	  Free  Trans Mnt<  Name            Status 	  Count     Label	 Blocks Count Cnt> DAD5:            Mounted	      0  REGNUM_DISK2	 831432   124	1   $ ! the backup $ show dev dad20  < Device           Device 	  Error    Volume	  Free  Trans Mnt<  Name            Status 	  Count     Label	 Blocks Count Cnt> DAD20:           Mounted	      0  REGNUM_DISK2	 861205     1	1  J $ ! change the label of the privately mounted backup disk and remount /sys' $ set volume /label:REGNUM_BACK2 dad20:1 $ umount DAD$REGNUM_BACKUP2n- $ bind /system /write /connect REGNUM_BACKUP2aC %LADCP-I-CONNAVAIL, successful connection to service REGNUM_BACKUP2*I %LADCP-I-BIND, service bound to logical unit DAD$REGNUM_BACKUP2 (_DAD22:) @ $ mount /noassist /system /write DAD$REGNUM_BACKUP2 REGNUM_BACK21 %MOUNT-I-MOUNTED, REGNUM_BACK2 mounted on _DAD22:  $ ! C $ ! the use of /write is to avoid the possibility of problems we'reoC $ ! seeing being caused by attempts to NFS share a read-only volume  $ ! D $ ! the disk is an NFS Unix container disk; the logicals and symbolsD $ ! are there to get (volatile) physical device and unit numbers outB $ ! of the tcpip commands so that this whole thing can be scripted $ !AE $ define /system /executive /translation:concealed disk_backup_home -  _$ disk$regnum_back2:[000000.] $ show log disk_backup_home H    "DISK_BACKUP_HOME" = "DISK$REGNUM_BACK2:[000000.]" (LNM$SYSTEM_TABLE) $ E $ translation = f$string(f$trnlnm("disk_backup_home") - ":[000000.]")  $ show symb translation #   TRANSLATION = "DISK$REGNUM_BACK2"  $ , $ tcpip map "/disk_backup_home" 'translation $ ! G $ ! succeeded; underlying volume has been mapped.  Now the container...) $ !eH $ translation = f$string(f$trnlnm("disk_backup_home") - ":[000000.]" + - _$ ":[export]")h $ show symb translation ,   TRANSLATION = "DISK$REGNUM_BACK2:[export]" $w" $ tcpip map "/backup" 'translation8 %TCPIP-E-MAPERROR, error processing MAP or UNMAP request- -SYSTEM-F-INCVOLLABEL, incorrect volume label2 $ !1A $ ! failed; the container DISK$REGNUM_BACK2:[export] has not been ) $ ! mapped to a Unix share called /backupt $ !s> $ ! Section 10.10.6 of the TCPIP V5.1 Management manual states? $ ! "If you are making copies or restoring files from a backup.@A $ ! The container file records the volume label and the Files--11 B $ ! file identifiers of the actual files on the disk.  If you copyD $ ! the file system or change the volume label, you must run ANALYZE> $ ! CONTAINER/REPAIR after you copy the files so that the fileF $ ! identifiers and volume label are corrected for the new location of $ ! the files."  $ !uA $ ! Right, so since we've changed the label and the container hasfB $ ! recorded the old label we need to fix this up with the /repair $ ! option.  Here goes...e $ !i. $ tcpip analyze container /repair 'translation8 %TCPIP-I-ANA_STARTSUPER, starting super block validation8 %%%%%%%%%%%  OPCOM  16-JAN-2004 12:50:27.15  %%%%%%%%%%%" Message from user SYSTEM on VMSBOXI %TCPIP-E-UFS_BAD_ID, container file system volume identification does noti match OpenVMS disk volumei $ !r@ $ ! Yeah, that's why I've asked you to fix it up, so what gives? $ !-   ------------------------------    Date: 16 Jan 2004 08:42:12 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)a9 Subject: Re: Mounting identically labelled disks (OVMS73)v3 Message-ID: <LJAD2gXi6v5E@eisner.encompasserve.org>0  ` In article <bu70pq$grd$2@grandcanyon.binc.net>, News Reader <read.news@hostname.invalid> writes: > B > So, how to stop VMS's nannying; letting me nfs share identically > labelled disks?,  C    We had a patch for that developed under VMS 2.5.  I did the portC8    the VMS 4.0.  Not something you want to do routinely.  C    VMS uses the volume name in it's locks to maintain the integritya6    of the file system.  You can't readily change that.  C    Why not mount the disk the way you had it and copy the files outs-    of the container using the provided tools?!   ------------------------------  + Date: Fri, 16 Jan 2004 17:04:43 +0000 (UTC)e. From: News Reader <read.news@hostname.invalid>9 Subject: Re: Mounting identically labelled disks (OVMS73)a/ Message-ID: <bu95jb$2no$1@grandcanyon.binc.net>   ) On 16 Jan 2004 08:42:12 -0600, Bob wrote:wC >> So, how to stop VMS's nannying; letting me nfs share identicallyd >> labelled disks? > D >   We had a patch for that developed under VMS 2.5.  I did the port >   the VMS 4.0. >dD >   Why not mount the disk the way you had it and copy the files out. >   of the container using the provided tools?  ; JF posted the solution to making two originally identically 9 labelled disks available system-wide -- mount one of themt> privately, relabel it as something else with set volume /label then remount system-wide.=  > The problem is that the old label is recorded in the container@ and tcpip analyze /repair just baulks at the discrepancy without> actually fixing it.  You had a patch for _that_ under VMS 4.0?  = As you say, I can undo the relabelling (so that they are both<; identical again) and use the provided tools -- specificallye= tcpip import and tcpip export -- as SYSTEM in the VMS processD@ owning the privately mounted backup and, one by one, iterating a? series of lengthy commands, export out of the backup and importt into the online disk...u  8 That procedure requires an intermediate file - you can't= pipe them together, and the command lines wrap to "two" linesf; because everything requires absolute pathnames and a wealthn@ of mandatory `options' cum parameters.  One of those is the fileA mode.  So there's an extra step of doing a tcpip dir /full on thei; backup disk in order to find the mode (permission) and edit- that into each import command.  ? Sure its a workaround, but its not a solution to the originallyaC posed problem.  I'd actually like to nfs mount the backup disk and,W@ on the Unix client, as a regular user open a second file browser= pane, look through and just drag one, or a selected group, of-9 the files across.  Does it have to be any more difficult?m  = I'm standing-by to hear how you do this on a VMS file server.o   ------------------------------  + Date: Fri, 16 Jan 2004 11:56:14 +0000 (UTC)o% From: bdc@world.std.com (Brian Chase)M& Subject: Re: Mozilla on Alpha question( Message-ID: <bu8jgu$tej$1@pcls4.std.com>  3 In article <f+VP$k2PG9by@eisner.encompasserve.org>,/. Larry Kilgallen <Kilgallen@SpamCop.net> wrote:3 > In article <OmHNb.19272$Eq.14143@clgrps12>, Alderg& > <PGDEHMKOKIMD@spammotel.com> writes: > > dittmer@attg.net wrote::  9 > >> I am having difficulty getting Mozilla v 1.4 to run. P > >> It apprears to have installed properly, but when I invoke @mozilla, it saysD > >> running mozilla.bin, and then comes back to the command prompt. > > @ > > You're invoking it by running the COM file provided, right ? > > & > > $ @SYS$COMMON:[MOZILLA]MOZILLA.COM > * > That sounds like a poor engineering job. > + > It ought to run as a regular DCL command.   A If you mean to say that it shouldn't be wrapped in a DCL script, hB I'd disagree.  Though I haven't actually seen the contents of the @ MOZILLA.COM script itself, I'm guessing it echoes the purpose ofE mozilla Bourne shell wrapper script used on Unix systems.  The script'A does housekeeping in terms of setting up the environment to tell eB the mozilla binary where to find its plugins, fonts, and the JVM. C It also figures out whether you're currently running an instance of A mozilla, so if you invoke the command a second time, it knows to v< open a new browser window using the already running process.  > Given all the subtle and not-so-subtle variations between the @ supported platforms (non-Unix included), it's simpler to do this: housekeeping in small platform specific wrapper scripts.     -brian.c -- sF --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ -----5                    Do not fold, mutilate, or spindle.    ------------------------------  % Date: Fri, 16 Jan 2004 07:51:44 -0600r6 From: "Craig A. Berry" <craigberry@mac.com.spamfooler>& Subject: Re: Mozilla on Alpha questionT Message-ID: <craigberry-6C7673.07514416012004@dsl081-159-101.chi1.dsl.speakeasy.net>  C In article <40073172_4@news1.prserv.net>, <dittmer@attg.net> wrote:e    6 > I am having difficulty getting Mozilla v 1.4 to run.M > It apprears to have installed properly, but when I invoke @mozilla, it saysfA > running mozilla.bin, and then comes back to the command prompt.y > N > I was expecting a new window with the pre-installed Netscape browser to come > up..  E Read the release notes, paying particular attention to everything it  C says about account quotas, system parameters, and file protections.a  D http://h71000.www7.hp.com/openvms/products/ips/mozilla_relnotes.html   ------------------------------    Date: 16 Jan 2004 08:46:56 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)t& Subject: Re: Mozilla on Alpha question3 Message-ID: <Vt$zP+XJyGbG@eisner.encompasserve.org>   D In article <40073172_4@news1.prserv.net>, <dittmer@attg.net> writes: > 6 > I am having difficulty getting Mozilla v 1.4 to run.M > It apprears to have installed properly, but when I invoke @mozilla, it sayslA > running mozilla.bin, and then comes back to the command prompt.s  , > Any ideas on how to diagnose this problem? > Any more info needed?d      "$set verify" comes to mind.    ------------------------------  % Date: Fri, 16 Jan 2004 16:57:34 +0100e) From: "Jean Mertens" <jmertens@skynet.be>. Subject: OPC DA client4 Message-ID: <40080996$0$309$ba620e4c@news.skynet.be>  C Any OPC DA client available for VMS : references, source code , ...  Thanks   ------------------------------    Date: 16 Jan 2004 08:35:57 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) , Subject: Re: OT: AA pilot arrested in Brazil3 Message-ID: <bKExyrAHtTLo@eisner.encompasserve.org>n  n In article <f30679fb.0401150744.5b477602@posting.google.com>, fabiopenvms@yahoo.com.br (Fabio Cardoso) writes:O > As americans are too puritane to show it over  the Net I am putting the link:n >   E    It was on the news the other day already.  Our media can hardly be     described as puritan.   ------------------------------   Date: 16 Jan 2004 15:06:04 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: The Inquirere9 Message-ID: <bu8ukr$feq0t$1@ID-135708.news.uni-berlin.de>t  6 In article <4006feaf.3276015@news.cable.ntlworld.com>,@ 	peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson) writes: >  > A > If the inquirer is so pro AMD64 and Linux why does it run on M$o
 > servers? > G > Accordingly if it changed to GNU/Linux serers and used PHP can anyone?$ > tell me why PHP isn't Open source?  E You must be one of those people who thinks anything not infected withpF the GNU Public Virus isn't OpenSource.  All of the sources for PHP are freely available.u   bill   -- eJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   u   ------------------------------  # Date: Fri, 16 Jan 2004 14:51:27 GMTh= From: peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson)d Subject: Re: The Inquirert5 Message-ID: <4007fa06.294203@news.cable.ntlworld.com>o  4 On Thu, 15 Jan 2004 18:58:07 -0600, "Craig A. Berry"& <craigberry@mac.com.spamfooler> wrote:  7 >In article <4007082f.5708625@news.cable.ntlworld.com>,h@ > peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson) wrote: > 7 >> On Thu, 15 Jan 2004 15:31:54 -0600, "Craig A. Berry"i) >> <craigberry@mac.com.spamfooler> wrote:- >> -: >> >In article <4006feaf.3276015@news.cable.ntlworld.com>,C >> > peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson) wrote:  >> >L >> >>> Accordingly if it changed to GNU/Linux serers and used PHP can anyone( >> >> tell me why PHP isn't Open source? >> > >> >No, because it is: >> > >> >http://www.php.net/license/c >> o >>  $ >> Tell me this is HTML Open source? >iG >Um, *what's* your question?  You asked why the software package known yE >as PHP is not open source.  I pointed out that it is, in fact, open sD >source and gave a reference to the license that dictates the terms E >under which the source code may be used.  That's what "open source" oH >generally means: the source code may be used, redistributed, modified, H >or all of the above depending on the terms of the particular license.  C >In your new question, I don't know what "this" refers to (the PHP r@ >package, the PHP license, the web page on which the license is E >described?), and I have no idea what you mean by "HTML Open source".i      E If you go to any web page written in HTML then using your browser youe  can view the source of the HTML.  E Now go to a page written in PHP can you view the source? Can you F**Kd  , Not very open source is this PHP then is it?   cheers,p     Peter Watkinsont% peter.watkinson1<nospam>@ntlworld.como remove <nospam> to reply :-)   ------------------------------  % Date: Fri, 16 Jan 2004 16:15:09 +01000- From: "Winfried Bergmann" <dummy@empuron.com>) Subject: Re: The Inquirer 9 Message-ID: <bu8v26$fdaee$1@ID-170759.news.uni-berlin.de>1  F "peter.watkinson1@ ntlworld.com (Peter Watkinson)" <nospam> schrieb im; Newsbeitrag news:4007fa06.294203@news.cable.ntlworld.com...o6 > On Thu, 15 Jan 2004 18:58:07 -0600, "Craig A. Berry"( > <craigberry@mac.com.spamfooler> wrote: >t9 > >In article <4007082f.5708625@news.cable.ntlworld.com>,pB > > peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson) wrote: > >w9 > >> On Thu, 15 Jan 2004 15:31:54 -0600, "Craig A. Berry" + > >> <craigberry@mac.com.spamfooler> wrote:o > >>< > >> >In article <4006feaf.3276015@news.cable.ntlworld.com>,E > >> > peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson) wrote:> > >> >G > >> >>> Accordingly if it changed to GNU/Linux serers and used PHP can  anyone* > >> >> tell me why PHP isn't Open source? > >> > > >> >No, because it is: > >> >! > >> >http://www.php.net/license/g > >> > >>& > >> Tell me this is HTML Open source? > >dH > >Um, *what's* your question?  You asked why the software package knownF > >as PHP is not open source.  I pointed out that it is, in fact, openE > >source and gave a reference to the license that dictates the termsrF > >under which the source code may be used.  That's what "open source"I > >generally means: the source code may be used, redistributed, modified, H > >or all of the above depending on the terms of the particular license.D > >In your new question, I don't know what "this" refers to (the PHPA > >package, the PHP license, the web page on which the license iseG > >described?), and I have no idea what you mean by "HTML Open source".v >d >P >lG > If you go to any web page written in HTML then using your browser youM" > can view the source of the HTML. >9G > Now go to a page written in PHP can you view the source? Can you F**Kr >o. > Not very open source is this PHP then is it?   Is that a joke? If not ...G I could write a compiler for an arbitrary language and put this to open L source. If you look at the binary for the target machine, compiled with thatF "open source" tool, could you see the source of your code? I guess youL didn't get the idea behind the term "open source". Check www.opensource.org.   ------------------------------  % Date: Fri, 16 Jan 2004 16:55:49 +0100t- From: "Winfried Bergmann" <dummy@empuron.com>t Subject: Re: The Inquirern9 Message-ID: <bu91ee$em1f4$1@ID-170759.news.uni-berlin.de>e  F "peter.watkinson1@ ntlworld.com (Peter Watkinson)" <nospam> schrieb im< Newsbeitrag news:4008031a.2618703@news.cable.ntlworld.com...  > Maybe I should make it simpler >oH > write a web page in ASP (Active Server Pages) put it on a Microsoft OSE > web server using IIS (Internet Information Server) 4/5 or whatever.6 >8F > Now go to the url with whatever web browser your using. Now use view? > source. You can see the HTML but can you see the embedded ASPfD > (Vbscript, javascript, perlscript or whatever) No.  This is closedE > source. Likewise the Microsoft Operating system, web server etc are. > closed source. >iF > Now accordingly why if PHP is "Open Source" why can't I view the WebF > page source code with my web browser - not even Mozilla on what everB > flavour OS you like. Still doesn't sound very open source to me. >oD > It's tough enough being a programmer at the moment anyway I'd much9 > rather get paid for just copying someone elses code ;-)n  C I'm sorry, but english is not my native language, so maybe I didn't  understand your statement:K Do you claim PHP not being open source because you can't see the PHP source)- code, when visiting a PHP generated web site?    ------------------------------  # Date: Fri, 16 Jan 2004 15:36:49 GMTs= From: peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson)N Subject: Re: The Inquirerw6 Message-ID: <4008031a.2618703@news.cable.ntlworld.com>  7 On Fri, 16 Jan 2004 16:15:09 +0100, "Winfried Bergmann"r <dummy@empuron.com> wrote:   >uG >"peter.watkinson1@ ntlworld.com (Peter Watkinson)" <nospam> schrieb im < >Newsbeitrag news:4007fa06.294203@news.cable.ntlworld.com...7 >> On Thu, 15 Jan 2004 18:58:07 -0600, "Craig A. Berry" ) >> <craigberry@mac.com.spamfooler> wrote:  >>: >> >In article <4007082f.5708625@news.cable.ntlworld.com>,C >> > peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson) wrote:a >> >: >> >> On Thu, 15 Jan 2004 15:31:54 -0600, "Craig A. Berry", >> >> <craigberry@mac.com.spamfooler> wrote: >> >> = >> >> >In article <4006feaf.3276015@news.cable.ntlworld.com>,nF >> >> > peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson) wrote: >> >> >dH >> >> >>> Accordingly if it changed to GNU/Linux serers and used PHP can >anyone + >> >> >> tell me why PHP isn't Open source?  >> >> >/ >> >> >No, because it is:r >> >> > " >> >> >http://www.php.net/license/ >> >>C >> >>t' >> >> Tell me this is HTML Open source?t >> >I >> >Um, *what's* your question?  You asked why the software package known.G >> >as PHP is not open source.  I pointed out that it is, in fact, open F >> >source and gave a reference to the license that dictates the termsG >> >under which the source code may be used.  That's what "open source" J >> >generally means: the source code may be used, redistributed, modified,I >> >or all of the above depending on the terms of the particular license.1E >> >In your new question, I don't know what "this" refers to (the PHP B >> >package, the PHP license, the web page on which the license isH >> >described?), and I have no idea what you mean by "HTML Open source". >> >> >>H >> If you go to any web page written in HTML then using your browser you# >> can view the source of the HTML.6 >>H >> Now go to a page written in PHP can you view the source? Can you F**K >>/ >> Not very open source is this PHP then is it?  >  >Is that a joke? If not ...eH >I could write a compiler for an arbitrary language and put this to openM >source. If you look at the binary for the target machine, compiled with that6G >"open source" tool, could you see the source of your code? I guess you M >didn't get the idea behind the term "open source". Check www.opensource.org.n >a >n       Maybe I should make it simpler  F write a web page in ASP (Active Server Pages) put it on a Microsoft OSC web server using IIS (Internet Information Server) 4/5 or whatever.e  D Now go to the url with whatever web browser your using. Now use view= source. You can see the HTML but can you see the embedded ASP2B (Vbscript, javascript, perlscript or whatever) No.  This is closedC source. Likewise the Microsoft Operating system, web server etc are  closed source.  D Now accordingly why if PHP is "Open Source" why can't I view the WebD page source code with my web browser - not even Mozilla on what ever@ flavour OS you like. Still doesn't sound very open source to me.  B It's tough enough being a programmer at the moment anyway I'd much7 rather get paid for just copying someone elses code ;-).   cheers,      Peter Watkinson % peter.watkinson1<nospam>@ntlworld.com  remove <nospam> to reply :-)   ------------------------------  % Date: Fri, 16 Jan 2004 09:53:09 -0600 2 From: "-Andy-" <see2go4me@spamdelicious.yahoo.com> Subject: Re: The Inquirero6 Message-ID: <Xns94726EBD1D0EEsee2go4me@216.196.97.132>  7 peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson)t$ enlightened us on 16 Jan 2004 with:     t  u; > If you go to any web page written in HTML then using your . > browser you can view the source of the HTML.  < Sure. But that has nothing to do with "open source". That's 8 just how most html display programs ( aka web browsers)  work.0: Someone could very easily write a program that displays a 9 document written using/in html that doesn't allow you to o access the source of the page.  < If you are really feeling masochistic you can also view the 8 source for a MS Word or pdf document. Doesn't make THEM  open source.  ; > Now go to a page written in PHP can you view the source?    3 Sure you can. You can view the source of the OUTPUT.; of the PHP. To view the source of the PHP that created the f< html you'll have to have access to the system it is running  on.   5 But that's true of any server based CGI. You see the b3 results, not the engine that generates the results.   . > Not very open source is this PHP then is it?  8 None of this is related to open source. Or VMS for that 6 matter. Unless the PHP license requires that anything 9 written in PHP be licensed as open source (I don't know, 69 haven't checked it's license), if PHP is open source all  2 you'll be able to do is see the source to the PHP 8 interpreter/compiler/whatever. Not to things written in  PHP.   -Andy-   ------------------------------  % Date: Fri, 16 Jan 2004 10:16:49 -0600n( From: Wayne Sewell <wayne@tachysoft.com> Subject: Re: The Inquirert0 Message-ID: <00A2BFA2.5165192B.21@tachysoft.com>  > >From: peter.watkinson1@<nospam>ntlworld.com (Peter Watkinson) >X-Newsgroups: comp.os.vms >Subject: Re: The Inquirer6 >Message-ID: <4007fa06.294203@news.cable.ntlworld.com>     >sF >If you go to any web page written in HTML then using your browser you! >can view the source of the HTML.  >t    N Can you see the source code of the web server that served the html to you?  OfC course not, but some web servers *are* open source, such as Apache.   B As someone else said, you don't understand what open source means.   WaynewO ===============================================================================eN Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html   lO ===============================================================================eB Jed Clampett, checking into hotel: "This place got a cement pond?", 	Ellie May: "And do yuh let critters in it?"   ------------------------------  # Date: Fri, 16 Jan 2004 16:11:42 GMTe2 From: "Ken Farmer" <KFarmer@NOSPAM.SpyderByte.com> Subject: Re: The Inquirerw@ Message-ID: <23UNb.231957$Vu5.17173639@twister.southeast.rr.com>  L "peter.watkinson1@ ntlworld.com (Peter Watkinson)" <nospam> wrote in message0 news:4008031a.2618703@news.cable.ntlworld.com...9 > On Fri, 16 Jan 2004 16:15:09 +0100, "Winfried Bergmann"u > <dummy@empuron.com> wrote: >   > Maybe I should make it simpler > H > write a web page in ASP (Active Server Pages) put it on a Microsoft OSE > web server using IIS (Internet Information Server) 4/5 or whatever.u >eF > Now go to the url with whatever web browser your using. Now use view? > source. You can see the HTML but can you see the embedded ASPhD > (Vbscript, javascript, perlscript or whatever) No.  This is closedE > source. Likewise the Microsoft Operating system, web server etc areu > closed source. >uF > Now accordingly why if PHP is "Open Source" why can't I view the WebF > page source code with my web browser - not even Mozilla on what everB > flavour OS you like. Still doesn't sound very open source to me. > D > It's tough enough being a programmer at the moment anyway I'd much9 > rather get paid for just copying someone elses code ;-)k    F PHP is used primarily to dynamically generate web pages pulling from a	 database.   + Want code, snippets, tutorials or articles?u   PHP.net - Home of PHPr http://www.php.net   weberdev, http://www.weberdev.com/maincat.php3/106/PHP   PHPBuilder.com http://www.phpbuilder.comd  $ O'Reilly's OnLamp.com: PHP DevCenter http://www.onlamp.com/php/   HotScripts.com: PHPd( http://www.hotscripts.com/PHP/index.html   Why PHP?9 http://www.webdevelopersjournal.com/articles/why_php.htmlr  % 7  Reasons Why PHP is Better than ASP ( http://php.weblogs.com/php_asp_7_reasons   Site Navigation with PHPj http://hotwired.lycos.com/webmonkey/templates/print_template.htmlt?meta=/webmonkey/99/25/index2a_meta.html  L I use php to generate OpenVMS.org and I must say I love it. Being similar toB perl it was easy for me to just start coding it without a problem.   Hope that helps.   Kene   -- Kenneth Farmer  <><n OpenVMS.org  |  dcl.OpenVMS.orgg   ------------------------------  % Date: Fri, 16 Jan 2004 10:29:18 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>tI Subject: Re: The Register: OpenVMS among most-secure of operating systemsr0 Message-ID: <bu8edv$f1o$1@new-usenet.uk.sun.com>   Mark Berryman wrote:* > Andrew Harrison SUNUK Consultancy wrote: >  >> Mark Berryman wrote:a >>, >>> Andrew Harrison SUNUK Consultancy wrote: >>>  >>>> Bob Koehler wrote:h >>>>G >>>>> In article <bu12iq$qi9$1@new-usenet.uk.sun.com>, Andrew Harrison  G >>>>> SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:o >>>>>  >>>>>R >>>>>> Ohhh yessssss yooou are.e >>>>>  >>>>>4 >>>>>0 >>>>>1 >>>>>R >>>>>sI >>>>>    Care to speculate as to which one of us has access to the sourcen >>>>>    listings and NDA info?i >>>>> I >>>>>    I have no reason to BS you, Andrew.  When I say I'm not guessingh* >>>>>    its a fact.  Not a boast, a fact. >>>>>c >>>>F >>>> Well lets just examine the external evidence shall we rather than4 >>>> your rahter laughable secret squirrel response. >>>>C >>>> Even CERT which is a highly unreliable source puts OpenVMS waye? >>>> ahead of VMS for vunerabilities. There was the DECnet Worm5 >>>> and thats just about it.  >>>>C >>>> On the other had the current OS has SSH, Bind, and a number ofo$ >>>> other advisories posted for it. >>>>A >>>> Security through obscurity cuts both ways because it removes ; >>>> your ability to prove your point while I at least haves  >>>> collateral to support mine. >>>>< >>>> So Ohhh Yesss You Arrr untill that is you can prove you
 >>>> arn't >>>r >>>i >>>  >>>pH >>> But I have proven it.  Unfortunately, for whatever reason, you have C >>> not been able to accept (or perhaps understand) the proof.  It u1 >>> probably isn't worth it but I will try again.c >>>tG >>> I have SSH on my VMS system.  It is NOT openssh.  Are you aware of  I >>> any advisories against it?  I can state that my VMS system has never n> >>> been in any sort of security or DoS danger because of SSH. >>>:F >>> I have BIND on my system.  None of the vulnerabilities posted for I >>> BIND have ever been able to impact my VMS version of BIND and I test uJ >>> every one.  So, how does the fact that I run BIND impact the level of  >>> security of my VMS system? >>>m >>7 >> There is only 2 reasons why you claim could be true.  >>A >> 1.    You arn't running a commecial version of bind on OpenVMS > >>     all the commercial versions either have CERT advisoriesA >>     for them or patches for cert advisories (if you understandy> >>     the sorry state of OpenVMS CERT reporting you will know >>     what this means). >  > " > No.  I run a commercial version. >  >>E >> 2.    You have installed a version more recent than the advisoriesq6 >>     anything else and you are at odds with reality. >  > H > No, again not true.  I apply the vendor-supplied patches after I test - > for whether my system is vulnerable or not.r > D > Ah, I think I see the crux of the matter.  You seem to think that I > anything that has been patched for security reasons must mean that any  E > VMS system running that piece of software must have had a security aJ > issue.  If so, then this is a proven false belief.  I will try to state  > things a little clearer.  >K > Vendors that distribute opensource products (e.g. BIND) as part of their -K >  product apply the patches released by the maintainers of the opensource iH > product in order to keep their code in sync.  This does NOT mean that K > the issue being addressed by the patch necessarily impacted the vendor's -
 > product. >   @ So you are now claiming that HP and the layered IP Stack vendors? have made a mistake in providing a patch that references a CERT@	 advisory.i  " Now you are really grubbing about.   Keep it up.o  H > So, name one security flaw in BIND that caused a security issue for a  > VMS system running BIND. >   < I don't have to the fact that a mandatory patch exists means* that the security vunerability does exist.  ? Do you honestly expect us to believe that the engineering teamse? producing the patches would make their customers go through theiB expense of applying a mandatory patch if it wan't in fact required to fix a security hole.d  < If your claim is true it doesn't say much for the engineers.  H >>> You have tried to claim the certain patches to VMS layered products I >>> should have been reported to CERT (if CERT was to be used as a valid iE >>> metric) because they addressed security issues.  However, when I  E >>> followed up on the ones you claimed, that claim turned out to be yI >>> specious (the patches were not, in fact, fixing any kind of security e >>> issue).k >>>b >>G >> Rubbish, Teardrop, Bind, SSH all require patches and the informationpE >> for this is either in the CERT advisory or in the case of TeardropU >> available from HP.P >  > K > They required patches, yes.  But in all three of the cases you cite, the mI > problem that was patched was not exploitable on any of the hundreds of aH > VMS systems I've been responsible for.  I've tried to tell you before H > that I have explicitly tested each of these cases.  You have no first K > hand knowledge but are simply drawing assumptions from reading something.w >   @ The fact is that many of the CERT advisories for other OS's also@ do not apply to many of the customers using the OS's. The client@ I work for uses a hardened Solaris distribution, this strips outA all the services and interfaces they don't need. This is a commons= practice, Sun provides documentation on the services commonlys
 not required.a  B So if the CERT vunerability is in a removed service/interface then0 they are not vunerable and do not need to patch.  @ This doesn't in any way diminish Sun's responsibility to fix theA holes and report accurately if there are holes because there willr7 be customers somewhere who could be hit by the exploit.n  = Exactly the same applies to OpenVMS, however with OpenVMS youa7 have no assurances that OpenVMS vunerabilites have beenn reported to CERT.l     >>I >>> So, let's get specific.  You say the CERT advisories aren't valid as iH >>> a metric because too many security issues in VMS don't get reported H >>> to CERT.  Let's start there.  Name some.  I'll follow up and verify F >>> your accuracy.  Remember, it has to be something that impacts the ! >>> security or stability of VMS.t >> >> >> >> >>< >> VMS722_DW_MOT_MUP-V0100 DECWindows MUP, No CERT Advisory. >  >  Bulletin number ESB-2001.460h > 4 >> ACMS_U2_043, ACMS security hole No CERT Advisory. > ( >  Quoting from the patch release notes:& >                       **** NOTE **** > E >      This problem does not compromise the security of  the  OpenVMS- >      operating  system.o >   ; Repeat after me it a security hole, lots of CERT advisoriesm5 are theoretical attacks, you may have noted that some.3 of the advisories state there there have not be anyL6 security compromises based on these theoretical holes.  : If you want to remove all these types of advisory from the0 CERT listing then just say otherwise stop BSing.  6 >> ALPSMUP01_070 No CERT Advisory (Know what a SMUPis) >  >  From the Release Notes:M >      Digital has received reports of possible unintended disclosure of usersI >      access information on Alpha systems running OpenVMS Alpha or SEVMSg
 >      Alpha.l > " > I did not find this one at CERT. > ; >> DCE_030_SSRT3608-V0100 COM/DCE Denial of Service no CERT  > I > This was a "potential" error that impacted only DCE.  It had no impact  J > on VMS or the data maintained on it.  According to the info I have this H > was a case of "we don't know if anyone could exploit this, we haven't J > heard that anyone has, but let's fix it anyway" which is typical of the B > proactive stance people buy VMS for.  This does not qualify for  > submission to CERT.n >   : I guess you just proved my point, thanks and at least from8 your standpoint the discussion is over unless you really0 don't understand why you just lost the argument.  < Lets just remove all the UNIX CERTs that might not be loaded9 because they arn't using that interface/service or can bes5 removed as part of a hardening process/protected by aT5 FireWall etc etc because that is exactly the argument 4 you have just advanced and sorry but it is total BS.   More salt I am afraid. >>5 >> I can provide more if you are interested but to be 1 >> honest your just lost the argument so lets noto >> rub salt in your wounds.h >  > J > Considering your low batting average, you might want to consider trying  > again. >     9 >> And just to illustrate that 3rd party layered products 7 >> are just as bad or put another way more OpenVMS CERTm >> BS gets trashed.h >>G >> http://www.multinet.process.com/scripts/eco/eco_tlb.com?FTP-053_A044] >>& >> Look for mandatory security update.@ >> Now wouldn't it be nice to know what that was, again no CERT. >>I >> http://www.multinet.process.com/scripts/eco/eco_tlb.com?NAMED-011_A044u? >> No reference to which CERT advisories they are fixing and ofbC >> course no response in the Vendor section of the CERT advisories.t >>H >> http://www.multinet.process.com/scripts/eco/eco_tlb.com?SMTP-030_A044 >> Mandatory Security Update >> Again no CERT advisory. >  > K > I've never looked for Multinet issues in CERT so I'll have to table this M* > until I have a little time for research. >   ? Why ist one of the tenets of Bob CERT BS campaign that Multinetr) etc are wonderfull compared with UCX etc.e   >>> H >>> Step two.  Name any opensource product that is now available on VMS I >>> for which an exploit was discovered that could be used to compromise  G >>> a VMS system (or even provide a means to make a DoS attack against iC >>> it).  I am aware of one, but that was in a product that no one eI >>> concerned about security would ever run.  Remember, the fact that an AE >>> exploit was discovered does NOT necesarily mean that the exploit e0 >>> could actually be used against a VMS system. >>>i >>< >> Why should I bother with OpenSource closed source is good
 >> enough. >  > H > Um, *you* are the one who brought up opensource (you know, BIND, SSH,  > etc.). > I >>> I'm calling you on your bluff.  You claim the security concerns with TA >>> VMS tend to get obfuscated.  Trot 'em out and show 'em to us.2 >>>26 >> I guess that you now regret even posting but thanks% >> for being a very usefull fall guy.  >  >  > Nope, no regrets at all. >  Really.n   regardsl Andrew Harrisone   ------------------------------  % Date: Fri, 16 Jan 2004 10:55:56 +0000-O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>nI Subject: Re: The Register: OpenVMS among most-secure of operating systemsa0 Message-ID: <bu8fvs$fle$1@new-usenet.uk.sun.com>   jlsue wrote:J > On Thu, 15 Jan 2004 02:23:45 -0500, JF Mezei <jfmezei.spamnot@istop.com> > wrote: >  > I >>When they discovered the flaw in the POP server that would allow one toeO >>overwrite any file on the system because the impage was installed with SYSPRVSP >>and could be called interactively with a log file specification, did that make >>it to CERT ? >  > L > Whose POP server?  Do I run that on my systems?  If not, then now could it4 > affect me and why would it actually be a VMS CERT? >   0 In fact its the HP POP server but that is rather irrelevant.   1 It sould be a VMS CERT because its a VMS specific-* hole in an a service that runs on OpenVMS.  5 Why not just obey the same set of rules that everyonej else does for CERT advisories.     > L >>I suspect that there are many issues specific to VMS that don't make it toL >>CERT.  CERT is prebably much more concerned about widely used systems, not, >>niche systems with very vew installations. >  > M > Define many.  By one defiintion of that term I could infer that you believe.I > there are many security-related fixes in VMS that are not reported.  Ist > that what you're implying? >   5 Well given your earlier point its obvious that you asi4 a HP employee see no need to report vunerabilites to5 CERT if they arn't part of base OpenVMS distribution.m  5 Assuming that your view reflects culture and practicei6 in HP then the answer to your question has to be many.   Regardsn Andrew Harrisonl   ------------------------------  % Date: Fri, 16 Jan 2004 14:33:45 +0000uO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>(I Subject: Re: The Register: OpenVMS among most-secure of operating systemsi0 Message-ID: <bu8soa$jvc$1@new-usenet.uk.sun.com>   Bob Koehler wrote: > In article <bu1dtd$168$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: >  >>Bob Koehler wrote: >> >>>In article <bu12iq$qi9$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:i >>@ >>Even CERT which is a highly unreliable source puts OpenVMS way< >>ahead of VMS for vunerabilities. There was the DECnet Worm >>and thats just about it. >  > ? >    That's hard to beleive since OpenVMS is VMS.  How can A bed >    ahead of A? >   ! Err no.  This is what you posted.o  < "ROTFLOL.  Back in the 80's, if you had a computer need, you< went out and bought a VAX.  VMS was the easiest thing to get@ a hold of.  It gots lots of scrutiny.  It was often connected to? world-wide networks.  I wasn't full of security holes then, and E it's not now.  Solaris has always been, but gradually gotten better."h  < OpenVMS 1.0 was introduced by Digital in 1992 when the Alpha9 was intoduced, it may well have basically been a port ando< a re-branding but what you run now as far as HP is concerned on Alpha boxes is OpenVMS.  @ The last VMS release of the 80's was 5.3 at estimated ~6 million lines of code.  = Currently OpenVMS 7.x has more like 25 million lines of code.e  * Both numbers exclude UCX, Pathwork/AS etc.  7 Are you claiming that you have examined all the currento7 source code, that you examined the 5.3 or ealier source 8 code and that you know that OpenVMS now is as secure and VMS was in the 80's.   > 9 >>So Ohhh Yesss You Arrr untill that is you can prove youc >>arn'to >  > F >    Your inability to follow logic undermines your reputation as someI >    kind of Sun engineer.  Next time we deal with Sun, we'll be stearingn >    clear of you. >   2 Ohh dear yet again comprehension problems dog you.   Regardsa Andrew Harrison    ------------------------------  % Date: Fri, 16 Jan 2004 17:00:09 +0100h) From: "Jean Mertens" <jmertens@skynet.be>l. Subject: Time synch between an Alpha and a Vax4 Message-ID: <40080a31$0$296$ba620e4c@news.skynet.be>  F Any DCL commands procedure example to synchronize system date and time/ between an Alpha and a Vax both under OpneVms ?s   Thanks   ------------------------------  % Date: Fri, 16 Jan 2004 10:57:56 -0600n6 From: "Craig A. Berry" <craigberry@mac.com.spamfooler>2 Subject: Re: Time synch between an Alpha and a VaxT Message-ID: <craigberry-57B61C.10575616012004@dsl081-159-101.chi1.dsl.speakeasy.net>  4 In article <40080a31$0$296$ba620e4c@news.skynet.be>,+  "Jean Mertens" <jmertens@skynet.be> wrote:t  H > Any DCL commands procedure example to synchronize system date and time1 > between an Alpha and a Vax both under OpneVms ?   H Don't use DCL.  Use the established protocols (TCP/IP's NTP or DECNet's ' DTSS).  And read the FAQ, specifically:4  : http://h71000.www7.hp.com/wizard/faq/vmsfaq_004.html#time6   ------------------------------    Date: 16 Jan 2004 04:39:28 -0800. From: fabiopenvms@yahoo.com.br (Fabio Cardoso) Subject: Re: UTF-8 on VAX-VMS1= Message-ID: <f30679fb.0401160439.72e3633a@posting.google.com>h  [ JF Mezei <jfmezei.spamnot@istop.com> wrote in message news:<40076461.EEC6CC4C@istop.com>...hM > Wanting to serve WAP content,  I need to generate UTF-8 encoded text files.h > M > The utilities documented on VMS's help do not work on VAX-VMS because UTF-8a3 > encodings are absent. (reference: ICONV utility).  > P > Does anyone know where one can obtain UTF-8 encodings that can be compiled andK > used by ICONV, or some standalone utility that can convert VMS files fromi > LATIN1/DEC-MCS to UTF-8 ?- > M > Would there be some easy way to declaresome file semantic or whatever othernN > mechanism that would allow one to edit such a file with TPU, and upon saving= > the file, it would automatically be stored in UTF-8 again ?a  1 Try CKERMIT ! We used it to convert some files ! e   Regards    FC   ------------------------------    Date: 16 Jan 2004 04:41:12 -0800. From: fabiopenvms@yahoo.com.br (Fabio Cardoso) Subject: Re: UTF-8 on VAX-VMSl< Message-ID: <f30679fb.0401160441.9fc8e8c@posting.google.com>  [ JF Mezei <jfmezei.spamnot@istop.com> wrote in message news:<40076461.EEC6CC4C@istop.com>...dM > Wanting to serve WAP content,  I need to generate UTF-8 encoded text files.  > M > The utilities documented on VMS's help do not work on VAX-VMS because UTF-803 > encodings are absent. (reference: ICONV utility).i > P > Does anyone know where one can obtain UTF-8 encodings that can be compiled andK > used by ICONV, or some standalone utility that can convert VMS files fromn > LATIN1/DEC-MCS to UTF-8 ?  > M > Would there be some easy way to declaresome file semantic or whatever othercN > mechanism that would allow one to edit such a file with TPU, and upon saving= > the file, it would automatically be stored in UTF-8 again ?a  ( We used the following syntax for CKERMIT  * translate file.XML;1  latin1 utf8 test.txt   Regards    FC   ------------------------------  % Date: Fri, 16 Jan 2004 17:58:42 +0400-& From: Valentin Likoum <me@privacy.net> Subject: Re: UTF-8 on VAX-VMS.9 Message-ID: <bu8qmk$ed9m7$1@ID-184585.news.uni-berlin.de>s   JF Mezei wrote:0N > 	Wanting to serve WAP content,  I need to generate UTF-8 encoded text files. > M > The utilities documented on VMS's help do not work on VAX-VMS because UTF-8 3 > encodings are absent. (reference: ICONV utility).o > P > Does anyone know where one can obtain UTF-8 encodings that can be compiled andK > used by ICONV, or some standalone utility that can convert VMS files from  > LATIN1/DEC-MCS to UTF-8 ?y > M > Would there be some easy way to declaresome file semantic or whatever othersN > mechanism that would allow one to edit such a file with TPU, and upon saving= > the file, it would automatically be stored in UTF-8 again ?e  :   As ICONV came from unix land so I'd look for conversion < table sources (including UTF8) on some flavour of unix, say 5 FreeBSD. This approach has allowed me to set correct -& timezone info what was missing in VMS.   --  
 Best regards,1
   Valentin)   valentin.likoum at ncc dot volga dot ruT   ------------------------------  % Date: Fri, 16 Jan 2004 10:49:12 -0600T6 From: "Craig A. Berry" <craigberry@mac.com.spamfooler> Subject: Re: UTF-8 on VAX-VMS:T Message-ID: <craigberry-1CDAAE.10491216012004@dsl081-159-101.chi1.dsl.speakeasy.net>  ) In article <40076461.EEC6CC4C@istop.com>, ,  JF Mezei <jfmezei.spamnot@istop.com> wrote:  N > 	Wanting to serve WAP content,  I need to generate UTF-8 encoded text files. > M > The utilities documented on VMS's help do not work on VAX-VMS because UTF-8(3 > encodings are absent. (reference: ICONV utility).   A It's not a VAX issue.  iconv doesn't do UTF-8 conversions on any  F version of VMS out of the box (well, at least not versions shipped in @ English-speaking countries).  You have to install the optional  E internationalization kit in order to get the necessary conversions.  p! There are some instructions here:   4 http://h71000.www7.hp.com/commercial/c/5763p016.html  D When it says the kit is "on the same media as the OpenVMS operating H system" it's only semi-true; it's on the second CD ("layered products", H IIRC) that ships with the system CD, not the system CD itself.  This is G obviously a bummer for hobbyists using the Montagar CDs, since I don't r believe they include the kit.t  D Note that Perl 5.8 and later includes Unicode support that does not < depend on the CRTL's iconv or the inadequate out-of-the-box  conversions.  Seep   $ perldoc perluniintro  > You can do conversions with the bundled Encode module like so:           use Encode 'from_to'; .         from_to($data, "iso-8859-1", "utf-8");   ------------------------------   Date: 16 Jan 2004 07:25:24 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow>0 Subject: Re: VAXELN-to-VxWorks portability layer5 Message-ID: <DTiotGxQ0bj6-pn2-djWpl4zlxU92@localhost>1  0 On Thu, 15 Jan 2004 22:26:48 UTC, Jerry Hudgins # <jerry@hudgins.delete-me.us> wrote:T  I > I recall that Digital offered a VAXELN portability layer for VxWorks atpJ > one point.  Presumably, that disappeared as a product long ago.  Was theF > source for it ever released?  If so, would one someone who has it beI > willing to provide me with a copy?  This isn't for commercial use; it'sr& > really just to satisfy my curiosity. >  > -jch >   D ISTR that I found a ref. to it somewhere out on the WWW. I've never ? been able to find it agin tho'. For me,   it was to assess its .3 viability as a route to another (VxWorks) platform.o    --   Cheers - Dave.   ------------------------------  % Date: Fri, 16 Jan 2004 08:13:09 -0500>2 From: "Stanley F. Quayle" <squayle@insight.rr.com>0 Subject: Re: VAXELN-to-VxWorks portability layer- Message-ID: <40079D15.3365.2A35301@localhost>   H > > I recall that Digital offered a VAXELN portability layer for VxWorksG > > at one point.  Presumably, that disappeared as a product long ago. cF > > Was the source for it ever released?  If so, would one someone who@ > > has it be willing to provide me with a copy?  This isn't for= > > commercial use; it's really just to satisfy my curiosity. E > ISTR that I found a ref. to it somewhere out on the WWW. I've neverT@ > been able to find it agin tho'. For me,   it was to assess its5 > viability as a route to another (VxWorks) platform.o  1 You could always use CHARON-VAX [Shameless Plug]!   
 --Stan Quaylee Quayle Consulting Inc.  
 ----------C Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363  Fax: +1 614 868-1671l1 8572 North Spring Ct. NW, Pickerington, OH  43147 = Preferred address:  stan@stanq.com       http://www.stanq.comy   ------------------------------  + Date: Fri, 16 Jan 2004 12:01:58 +0000 (UTC)t% From: bdc@world.std.com (Brian Chase)n+ Subject: Re: Will HP *ever* publicize OVMS?r( Message-ID: <bu8jrm$tcp$1@pcls4.std.com>  K In article <T9zNb.164079$AAe1.151059@news01.bloor.is.net.cable.rogers.com>,r" John Smith <a@nonymous.com> wrote:  _ > http://story.news.yahoo.com/news?tmpl=story&cid=1738&ncid=738&e=9&u=/zd/20040115/tc_zd/116601o > @ > Hewlett-Packard Co. on Thursday will announce that it earned aH > record-breaking $2.5 billion Linux (news - web sites)-based revenue inJ > fiscal 2003, with its Linux services and solutions business posting a 40  > percent rise over fiscal 2002. > K > While the revenue was derived from the sale of Linux-related products andeK > services, the Palo Alto, Calif., company did not specify exactly what was . > included and counted as Linux-based revenue. >  > * > Where's the HP press release about OVMS?  F Yeah, and so much for Kerry's generic "HP branding" explanation about   why OpenVMS isn't being plugged.   -brian.S -- )F --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ -----5                    Do not fold, mutilate, or spindle.b   ------------------------------  % Date: Fri, 16 Jan 2004 08:07:40 -0500a' From: "Main, Kerry" <kerry.main@hp.com>e+ Subject: RE: Will HP *ever* publicize OVMS?eR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB237D52@tayexc19.americas.cpqcorp.net>   > >=20 > >=20, > > Where's the HP press release about OVMS? >=208 > Yeah, and so much for Kerry's generic "HP branding"=204 > explanation about why OpenVMS isn't being plugged. >=20	 > -brian.a > --H > --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ -----7 >                    Do not fold, mutilate, or spindle.  >=20 Brian,  C Minor correction, while the statement about focus on HP branding isoE likely true (not that I agree with it), but that was not me that made ( the recent statements about HP branding.   :-)-   Regards   
 Kerry Main Senior ConsultantM HP Services Canada Voice: 613-592-4660l Fax: 613-591-4477s Email: kerryDOTmainAThpDOTcomw. (remove the DOT's and AT for email address)=20   ------------------------------  + Date: Fri, 16 Jan 2004 13:28:34 +0000 (UTC)r% From: bdc@world.std.com (Brian Chase)1+ Subject: Re: Will HP *ever* publicize OVMS?h( Message-ID: <bu8ou2$rji$1@pcls4.std.com>  R In article <FD827B33AB0D9C4E92EACEEFEE2BA2FB237D52@tayexc19.americas.cpqcorp.net>,& Main, Kerry <kerry.main@hp.com> wrote:  8 > > Yeah, and so much for Kerry's generic "HP branding" 6 > > explanation about why OpenVMS isn't being plugged.   > Brian, > E > Minor correction, while the statement about focus on HP branding is"G > likely true (not that I agree with it), but that was not me that made * > the recent statements about HP branding. >  > :-)e  2 Oops.  My bad.  It was Keith Parris who said that.   -brian.> --  F --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ -----5                    Do not fold, mutilate, or spindle.    ------------------------------   End of INFO-VAX 2004.031 ************************