1 INFO-VAX	Thu, 02 Sep 2004	Volume 2004 : Issue 487       Contents:! Re: A whopping 50 percent...  ??? ! Re: A whopping 50 percent...  ??? ! Re: A whopping 50 percent...  ??? ! RE: A whopping 50 percent...  ??? ! RE: A whopping 50 percent...  ??? ! RE: A whopping 50 percent...  ??? ! RE: A whopping 50 percent...  ??? ! RE: A whopping 50 percent...  ??? ! RE: A whopping 50 percent...  ??? ! RE: A whopping 50 percent...  ??? ! RE: A whopping 50 percent...  ??? ! Re: A whopping 50 percent...  ??? ! Re: A whopping 50 percent...  ??? / Re: Alphastation 200 storage controller options ( Re: Big IA64 test coming in a few months Re: charon vax emulator??? Re: Flat panel VT replacement? Re: Flat panel VT replacement? Re: Flat panel VT replacement? Re: Flat panel VT replacement? Re: foto Future of Decnet Re: Future of Decnet Re: Future of Decnet Re: Future of Decnet Re: Future of Decnet Re: Future of Decnet Re: Future of Decnet. Re: How do recover a missing member from RAID?" Re: Itanium: EFI... no fs0: device" Re: Itanium: EFI... no fs0: device" Re: Itanium: EFI... no fs0: device" Re: Itanium: EFI... no fs0: device" Re: Itanium: EFI... no fs0: device Look at this girl... Multinet v5 ssh login failure ! Re: Multinet v5 ssh login failure ! Re: Multinet v5 ssh login failure 
 Re: need help 
 Single Signon * Re: SMTP: stray messages left in directory# Re: strange decnet-plus problems... # Re: strange decnet-plus problems... # Re: strange decnet-plus problems... " Re: Two IPs on same NIC OVMS 7.2-1- [VMS V7.3-2] PCSI INSTALL /SAVE_RECOVERY_DATA   F ----------------------------------------------------------------------  $ Date: Thu, 2 Sep 2004 06:53:41 -0400* From: "Bill Todd" <billtodd@metrocast.net>* Subject: Re: A whopping 50 percent...  ???= Message-ID: <3audnWMoMsT0ZqvcRVn-tQ@metrocastcablevision.com>   . "Tom Linden" <tom@kednos.com> wrote in message3 news:NDEMLKKEBOIFBMJLCECIOEMGDLAA.tom@kednos.com... G > When you look at the list, you have to ask yourself what the hell was 2 > Digital thinking to embark on the alpha program.   Perhaps something like...   I Here we are, a dozen years or so into the VAX architecture, and we're not B only starting to have some problems keeping up in performance (andK especially price/performance) but 64-bit systems are beginning to appear on  the horizon.  G 1.  Can the VAX architecture as it stands now carry us through the next F decade or more?  Absolutely not:  if nothing else gets us, the lack ofE 64-bit support will start to before then.  So out goes the no-change,  low-cost, short-term option.  H 2.  Can we extend the VAX architecture compatibly to 64 bits to carry usF through the next 15 years or so?  Almost certainly not:  everything weI currently know about processor design suggests that CISC won't be able to K keep up with RISC (who would've thunk you might be able to *emulate* CISC - F especially one as complex as VAX - at near-RISC speeds back then?  youJ couldn't even have fitted the emulator on your chip, let alone a processorH as well), and the performance and price/performance divergence over thatG long a time would kill us.  So out goes the modest-change, medium-cost,  medium-term option.   I 3.  Can we build something completely new in 64 bits that's optimized for I speed and a long enough life to pay back the investment, steal a march on I the competition that's currently stolen a march on us, and emulate VAX in H *software* fast enough to satisfy the customers who can't/won't migrate?	 Hmmmmm...    - bill   ------------------------------   Date: 2 Sep 2004 07:06:20 -0500 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) * Subject: Re: A whopping 50 percent...  ???3 Message-ID: <lmpz8BS5laK+@eisner.encompasserve.org>   \ In article <41360E29.23023D31@teksavvy.com>, Jf Mezei <jfmezei.spamnot@teksavvy.com> writes: > Bob Koehler wrote:I >>    It takes a certina amount of money to build a computer.  If you put K >>    a slow chip in it you save a small amount.  RISC designs were running * >>    orders of magnitude faster than VAX. > O > I know that RISC chips had much higher clock rates. But did they actually run * > software "orders of magnitudes" faster ?  D    My own experience, including some rough CPU tests, indicated that8    they were indeed up to one order of magnitude faster.        ------------------------------   Date: 2 Sep 2004 07:08:32 -0500 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) * Subject: Re: A whopping 50 percent...  ???3 Message-ID: <AQwRiD9mCZ+y@eisner.encompasserve.org>   U In article <2pmplmFm9b76U1@uni-berlin.de>, bill@cs.uofs.edu (Bill Gunshannon) writes:  >   E > Oh pullezzze.....  One of the thngs that led to the rapid spread of  > Unix was free software.   C    Really?  I can remember getting tons of free software for my VMS ;    systems at a time when nobody was taking UNIX seriously.    ------------------------------  $ Date: Thu, 2 Sep 2004 07:01:45 -0700# From: "Tom Linden" <tom@kednos.com> * Subject: RE: A whopping 50 percent...  ???9 Message-ID: <NDEMLKKEBOIFBMJLCECIAEOEDLAA.tom@kednos.com>    < -----Original Message-----1 < From: Bill Todd [mailto:billtodd@metrocast.net] , < Sent: Thursday, September 02, 2004 3:54 AM < To: Info-VAX@Mvb.Saic.Com + < Subject: Re: A whopping 50 percent... ???  <  <  < 0 < "Tom Linden" <tom@kednos.com> wrote in message5 < news:NDEMLKKEBOIFBMJLCECIOEMGDLAA.tom@kednos.com... I < > When you look at the list, you have to ask yourself what the hell was 4 < > Digital thinking to embark on the alpha program. <  < Perhaps something like...  < K < Here we are, a dozen years or so into the VAX architecture, and we're not D < only starting to have some problems keeping up in performance (andC < especially price/performance) but 64-bit systems are beginning to  < appear on  < the horizon. < I < 1.  Can the VAX architecture as it stands now carry us through the next H < decade or more?  Absolutely not:  if nothing else gets us, the lack ofG < 64-bit support will start to before then.  So out goes the no-change,  < low-cost, short-term option.  I Wrong.  32 bit did just fine for the 10 years in question.  BTW, look how  many people still run VAXen.   < J < 2.  Can we extend the VAX architecture compatibly to 64 bits to carry usH < through the next 15 years or so?  Almost certainly not:  everything weK < currently know about processor design suggests that CISC won't be able to < < keep up with RISC (who would've thunk you might be able to < *emulate* CISC -  L Yes, it could have been extended look at Opteron.  It was based on extensionK of an ugly arhitecture.  Yes, we new we could emulate CISC much erlier than I you are aware.  In fact the 80386 used an 801 core which was based on the E work of Siemens/Intel on the 4/32.  AS400's use a Power PC core since  sometime in the 80's  H < especially one as complex as VAX - at near-RISC speeds back then?  youL < couldn't even have fitted the emulator on your chip, let alone a processorJ < as well), and the performance and price/performance divergence over thatI < long a time would kill us.  So out goes the modest-change, medium-cost,  < medium-term option. K As proof by counter example look at the x86's they are pushing 4GHz with an  ad hoc) architecture, VAX would have been easier.  < K < 3.  Can we build something completely new in 64 bits that's optimized for K < speed and a long enough life to pay back the investment, steal a march on K < the competition that's currently stolen a march on us, and emulate VAX in J < *software* fast enough to satisfy the customers who can't/won't migrate? < Hmmmmm...  < H It doesn't have to be entirely new, what kind of core you use depends on whatL you have.  The EV7 has 152 x 10**6 transistors, how many more would you needJ to emulate VAX,  I don't know, but I bet 10 x 10**6 would take care of it.  J RISC was a fashion that allow for simpler design, but the price paid was aJ meager instruction set, which is fine if all you need to do is run C code, and L that is where the Alpha designers fell short of their task as, say, comparedL to IBM.  The principle difference was management's failure to understand theI market and to ride herd on the engineers.  Example, unaligned access, one  tickK extra on Power PC.  Why is this important?  Because of the large code base.    < - bill <  <  <  < --- ( < Incoming mail is certified Virus Free.< < Checked by AVG anti-virus system (http://www.grisoft.com).A < Version: 6.0.735 / Virus Database: 489 - Release Date: 8/6/2004  <  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.735 / Virus Database: 489 - Release Date: 8/6/2004    ------------------------------  * Date: Thu, 2 Sep 2004 14:33:30 +0000 (UTC)( From: m.kraemer@gsi.de (Michael Kraemer)* Subject: RE: A whopping 50 percent...  ???5 Message-ID: <ch7avq$6qn$1@lnx107.hrz.tu-darmstadt.de>   _ In article <NDEMLKKEBOIFBMJLCECIKENFDLAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:  > C > < It was, however, enough for me (and others) to leave VMS (apart  > < from other reasons). >  > And those were ...  + Well, the buzzword of the time was "open".  G VMS was as much "closed" as MVS or other "propietary" OSs at that time, : UNIX were "open" systems with competition between vendors.  B It was foreseeable that in our part of the universe (edu/academia)6 new developments would be done primarily/only on UNIX.B Add to that that it is/was much easier to port between UNIces thanP between UNIces and proprietary OSs, let alone between different proprietary OSs.3 The suspection was also that DEC and their products E wouldn't make it through the 90's, a correct forecast as we all know.   > And, although this may be a local observation, I found a small0 collection of UNIX boxes (RS6k/AIX in our case) , easier to setup and maintain for a workgroup, than to deal with a centralized Vax cluster.O I remember not so infrequent cluster crashes with effective down times of local 7 workstations measured rather in hours than in minutes.  G The UNIX setup seems a lot more robust for our purposes. One could pull O SCSI and power plugs nearly on the fly to move workstations between offices and F counting rooms, thanks JFS I can't even spell "fsck" correctly. ThanksG ODM we had sort of plug and play when adding new devices and thanks LVM B creating, increasing, mirroring of filesystems was not a big deal.H Thanks smitty you need not be a long time guru to administer the boxes. 2 All this from day one of the platform's existence.A Other UNIces of that time had different merits, HP-UX for example + was often chosen due to its HP-VUE desktop.   H Given all that plus the price/performance ratio, there was no compellingE reason to stay with VMS, let alone to come back when the DEC universe # finally collapsed during the 90's.     ------------------------------  * Date: Thu, 2 Sep 2004 14:48:54 +0000 (UTC)( From: m.kraemer@gsi.de (Michael Kraemer)* Subject: RE: A whopping 50 percent...  ???5 Message-ID: <ch7bsm$72l$1@lnx107.hrz.tu-darmstadt.de>   _ In article <NDEMLKKEBOIFBMJLCECIAEOEDLAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:  > N > Yes, it could have been extended look at Opteron.  It was based on extensionM > of an ugly arhitecture.  Yes, we new we could emulate CISC much erlier than K > you are aware.  In fact the 80386 used an 801 core which was based on the G > work of Siemens/Intel on the 4/32.  AS400's use a Power PC core since  > sometime in the 80's  H make that the 90's and it is somehow correct. PPC appeared around 93/94.  M > As proof by counter example look at the x86's they are pushing 4GHz with an  > ad hoc+ > architecture, VAX would have been easier.   K one might be tempted to say that x86 is so primitive that it is nearly RISC  already.M And so it is easier to put RISC inside than with any other CISC architecture.    > L > RISC was a fashion that allow for simpler design, but the price paid was aL > meager instruction set, which is fine if all you need to do is run C code,  A which is what presumably most people do, apart from Cobol, maybe. M Anyway, who cares about the instruction set when everybody uses HL languages.  (OK, compiler writers ... ;-)    > and N > that is where the Alpha designers fell short of their task as, say, comparedN > to IBM.  The principle difference was management's failure to understand theK > market and to ride herd on the engineers.  Example, unaligned access, one  > tickM > extra on Power PC.  Why is this important?  Because of the large code base.   F when Power/PPC came around there wasn't such a large code base for it.   ------------------------------  $ Date: Thu, 2 Sep 2004 07:46:29 -0700# From: "Tom Linden" <tom@kednos.com> * Subject: RE: A whopping 50 percent...  ???9 Message-ID: <NDEMLKKEBOIFBMJLCECIEEOFDLAA.tom@kednos.com>    < -----Original Message-----C < From: Michael Kraemer [mailto:kraemer@biors6a.gsi.de]On Behalf Of  < Michael Kraemer , < Sent: Thursday, September 02, 2004 7:34 AM < To: Info-VAX@Mvb.Saic.Com + < Subject: RE: A whopping 50 percent... ???  <  < A < In article <NDEMLKKEBOIFBMJLCECIKENFDLAA.tom@kednos.com>, "Tom  " < Linden" <tom@kednos.com> writes: < > E < > < It was, however, enough for me (and others) to leave VMS (apart  < > < from other reasons). < >  < > And those were ... < - < Well, the buzzword of the time was "open".  I < VMS was as much "closed" as MVS or other "propietary" OSs at that time, < < UNIX were "open" systems with competition between vendors. < D < It was foreseeable that in our part of the universe (edu/academia)8 < new developments would be done primarily/only on UNIX.D < Add to that that it is/was much easier to port between UNIces thanB < between UNIces and proprietary OSs, let alone between different  < proprietary OSs.5 < The suspection was also that DEC and their products G < wouldn't make it through the 90's, a correct forecast as we all know.  < @ < And, although this may be a local observation, I found a small2 < collection of UNIX boxes (RS6k/AIX in our case) . < easier to setup and maintain for a workgroup. < than to deal with a centralized Vax cluster.C < I remember not so infrequent cluster crashes with effective down   < times of local9 < workstations measured rather in hours than in minutes.  I < The UNIX setup seems a lot more robust for our purposes. One could pull > < SCSI and power plugs nearly on the fly to move workstations  < between offices and H < counting rooms, thanks JFS I can't even spell "fsck" correctly. ThanksI < ODM we had sort of plug and play when adding new devices and thanks LVM D < creating, increasing, mirroring of filesystems was not a big deal.J < Thanks smitty you need not be a long time guru to administer the boxes. 4 < All this from day one of the platform's existence.C < Other UNIces of that time had different merits, HP-UX for example - < was often chosen due to its HP-VUE desktop.  < J < Given all that plus the price/performance ratio, there was no compellingG < reason to stay with VMS, let alone to come back when the DEC universe % < finally collapsed during the 90's.    H But as I recall, a lot of your code was written in PL/I for MVS and VAX,F which I presume you had to rewrite?  When you factored that in was the@ TCO (Total Cost of Ownership) still in favour of the Unix boxes?   <  < --- ( < Incoming mail is certified Virus Free.< < Checked by AVG anti-virus system (http://www.grisoft.com).A < Version: 6.0.735 / Virus Database: 489 - Release Date: 8/6/2004  <  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.735 / Virus Database: 489 - Release Date: 8/6/2004    ------------------------------  $ Date: Thu, 2 Sep 2004 08:01:18 -0700# From: "Tom Linden" <tom@kednos.com> * Subject: RE: A whopping 50 percent...  ???9 Message-ID: <NDEMLKKEBOIFBMJLCECICEOGDLAA.tom@kednos.com>    < -----Original Message-----C < From: Michael Kraemer [mailto:kraemer@biors6a.gsi.de]On Behalf Of  < Michael Kraemer , < Sent: Thursday, September 02, 2004 7:49 AM < To: Info-VAX@Mvb.Saic.Com + < Subject: RE: A whopping 50 percent... ???  <  < A < In article <NDEMLKKEBOIFBMJLCECIAEOEDLAA.tom@kednos.com>, "Tom  " < Linden" <tom@kednos.com> writes: < > D < > Yes, it could have been extended look at Opteron.  It was based  < on extensionD < > of an ugly arhitecture.  Yes, we new we could emulate CISC much 
 < erlier than A < > you are aware.  In fact the 80386 used an 801 core which was   < based on theI < > work of Siemens/Intel on the 4/32.  AS400's use a Power PC core since  < > sometime in the 80's < J < make that the 90's and it is somehow correct. PPC appeared around 93/94.  ! I have RS6000 manuals dated 1990.  < C < > As proof by counter example look at the x86's they are pushing   < 4GHz with an
 < > ad hoc- < > architecture, VAX would have been easier.  < B < one might be tempted to say that x86 is so primitive that it is 
 < nearly RISC 
 < already.C Actually, not.  It is an accumulator design, had address modes, all  standard features of CISC   B < And so it is easier to put RISC inside than with any other CISC  < architecture.  <  < > D < > RISC was a fashion that allow for simpler design, but the price  < paid was aC < > meager instruction set, which is fine if all you need to do is  
 < run C code,  < C < which is what presumably most people do, apart from Cobol, maybe. B < Anyway, who cares about the instruction set when everybody uses  < HL languages.  < (OK, compiler writers ... ;-)   B But it does make a difference, speaking as a compiler writer.  TheA 21064 took 17 ticks to compute A = B + C;  where these were 16bit ? integers, because there were no such instructions, on had to do C shifts and masks.  That is another example of management's failure. F That meant that existing C code would perform miserably.  It had to beB rewritten to conform to the new instruction set in order to perfom; well.  So, you can only do so much in writing the compiler.    <  < > and C < > that is where the Alpha designers fell short of their task as,   < say, compared B < > to IBM.  The principle difference was management's failure to  < understand theB < > market and to ride herd on the engineers.  Example, unaligned 
 < access, one  < > tick? < > extra on Power PC.  Why is this important?  Because of the   < large code base. < H < when Power/PPC came around there wasn't such a large code base for it.  J From IBM's prespective there was,  it had to potentailly address the needsJ of their ENTIRE customer base.  Their compilers were written in a languageI they called PL8, which was essentially a modified PL/I, which facilitated 3 the other compilers like Cobol, Fortran and even C.    <  < --- ( < Incoming mail is certified Virus Free.< < Checked by AVG anti-virus system (http://www.grisoft.com).A < Version: 6.0.735 / Virus Database: 489 - Release Date: 8/6/2004  <  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.735 / Virus Database: 489 - Release Date: 8/6/2004    ------------------------------  * Date: Thu, 2 Sep 2004 15:23:50 +0000 (UTC)( From: m.kraemer@gsi.de (Michael Kraemer)* Subject: RE: A whopping 50 percent...  ???5 Message-ID: <ch7du6$7rt$1@lnx107.hrz.tu-darmstadt.de>   _ In article <NDEMLKKEBOIFBMJLCECIEEOFDLAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:  > J > But as I recall, a lot of your code was written in PL/I for MVS and VAX,& > which I presume you had to rewrite?   I Not necessarily. You refer to data analysis systems developed in the 80's H to support mainly small user groups, with experiment lifetimes of a few  years.  Y New (larger and more CPU power hungry) experiments coming up in the 90's immediately used E Fortran as their base language and hence could use the new UNIX boxes H at no extra cost. The "old" PL/I systems weren't suited for them anyway.K Now if you say that going from PL/I to F77 is a step back: you're certainly 7 right, however, it was the only common base available.    # > When you factored that in was the B > TCO (Total Cost of Ownership) still in favour of the Unix boxes?  M This is not the way it went. New apps/customers appeared simply on UNIX only, A so no porting involved. The old systems died as more and more of  % their customers finished their work.     ------------------------------  * Date: Thu, 2 Sep 2004 16:15:52 +0000 (UTC)( From: m.kraemer@gsi.de (Michael Kraemer)* Subject: RE: A whopping 50 percent...  ???5 Message-ID: <ch7gvo$8r4$1@lnx107.hrz.tu-darmstadt.de>   _ In article <NDEMLKKEBOIFBMJLCECICEOGDLAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:  >  >  > < -----Original Message-----E > < From: Michael Kraemer [mailto:kraemer@biors6a.gsi.de]On Behalf Of  > < Michael Kraemer . > < Sent: Thursday, September 02, 2004 7:49 AM > < To: Info-VAX@Mvb.Saic.Com - > < Subject: RE: A whopping 50 percent... ???  > <  > < C > < In article <NDEMLKKEBOIFBMJLCECIAEOEDLAA.tom@kednos.com>, "Tom  $ > < Linden" <tom@kednos.com> writes: > < > ' > < > AS400's use a Power PC core since  > < > sometime in the 80's > < L > < make that the 90's and it is somehow correct. PPC appeared around 93/94. > # > I have RS6000 manuals dated 1990.   E That's the original Power box. PPC came later. AS400 had its own CPU, E which was replaced by some Power/PPC variant in the mid 90's (or so).    > < E > < > As proof by counter example look at the x86's they are pushing   > < 4GHz with an > < > ad hoc/ > < > architecture, VAX would have been easier.  > < D > < one might be tempted to say that x86 is so primitive that it is  > < nearly RISC  > < already.E > Actually, not.  It is an accumulator design, had address modes, all  > standard features of CISC   " well, I forgot the smiley here ...   ! > < (OK, compiler writers ... ;-)  > D > But it does make a difference, speaking as a compiler writer.  TheC > 21064 took 17 ticks to compute A = B + C;  where these were 16bit A > integers, because there were no such instructions, on had to do E > shifts and masks.  That is another example of management's failure. H > That meant that existing C code would perform miserably.  It had to beD > rewritten to conform to the new instruction set in order to perfom= > well.  So, you can only do so much in writing the compiler.   J Seems more a problem of the CPU people not talking to the compiler people.C Or everybody at DEC thought that the future would be so "big" that  . nobody would need stinking 16bits anymore ...     > < J > < when Power/PPC came around there wasn't such a large code base for it. > L > From IBM's prespective there was,  it had to potentailly address the needsL > of their ENTIRE customer base.  Their compilers were written in a languageK > they called PL8, which was essentially a modified PL/I, which facilitated 5 > the other compilers like Cobol, Fortran and even C.   K Well, the best you can say is that IBM saw the need for "legacy" data types  even for fancy new CPUs.A As for the common base of their compilers, I slightly doubt that. @ The runtimes of their PL/I and Fortran were pretty incompatible,L at least on MVS. On AIX, it was better, since they were presumably developed from scratch.      ------------------------------  $ Date: Thu, 2 Sep 2004 09:40:32 -0700# From: "Tom Linden" <tom@kednos.com> * Subject: RE: A whopping 50 percent...  ???9 Message-ID: <NDEMLKKEBOIFBMJLCECIKEOIDLAA.tom@kednos.com>    < -----Original Message-----C < From: Michael Kraemer [mailto:kraemer@biors6a.gsi.de]On Behalf Of  < Michael Kraemer , < Sent: Thursday, September 02, 2004 9:16 AM < To: Info-VAX@Mvb.Saic.Com + < Subject: RE: A whopping 50 percent... ???  <  < @ < In article <NDEMLKKEBOIFBMJLCECICEOGDLAA.tom@kednos.com>, "Tom" < Linden" <tom@kednos.com> writes: < >  < >   < > < -----Original Message-----G < > < From: Michael Kraemer [mailto:kraemer@biors6a.gsi.de]On Behalf Of  < > < Michael Kraemer 0 < > < Sent: Thursday, September 02, 2004 7:49 AM < > < To: Info-VAX@Mvb.Saic.Com / < > < Subject: RE: A whopping 50 percent... ???  < > <  < > < D < > < In article <NDEMLKKEBOIFBMJLCECIAEOEDLAA.tom@kednos.com>, "Tom& < > < Linden" <tom@kednos.com> writes: < > < > ) < > < > AS400's use a Power PC core since  < > < > sometime in the 80's < > < @ < > < make that the 90's and it is somehow correct. PPC appeared < around 93/94.  < > % < > I have RS6000 manuals dated 1990.  < G < That's the original Power box. PPC came later. AS400 had its own CPU, G < which was replaced by some Power/PPC variant in the mid 90's (or so).    Same instruction set.    <  < > < F < > < > As proof by counter example look at the x86's they are pushing < > < 4GHz with an < > < > ad hoc1 < > < > architecture, VAX would have been easier.  < > < E < > < one might be tempted to say that x86 is so primitive that it is  < > < nearly RISC  < > < already.G < > Actually, not.  It is an accumulator design, had address modes, all  < > standard features of CISC  < $ < well, I forgot the smiley here ... < # < > < (OK, compiler writers ... ;-)  < > F < > But it does make a difference, speaking as a compiler writer.  TheE < > 21064 took 17 ticks to compute A = B + C;  where these were 16bit C < > integers, because there were no such instructions, on had to do G < > shifts and masks.  That is another example of management's failure. J < > That meant that existing C code would perform miserably.  It had to beF < > rewritten to conform to the new instruction set in order to perfom? < > well.  So, you can only do so much in writing the compiler.  < L < Seems more a problem of the CPU people not talking to the compiler people.D < Or everybody at DEC thought that the future would be so "big" that/ < nobody would need stinking 16bits anymore ...   K Precisely my point, why would you forgo an existing, sizable and very loyal > customer base in favour of a subjunctive vision of the future.   <  < > < L < > < when Power/PPC came around there wasn't such a large code base for it. < > < < > From IBM's prespective there was,  it had to potentailly < address the needs C < > of their ENTIRE customer base.  Their compilers were written in  < a languageA < > they called PL8, which was essentially a modified PL/I, which 
 < facilitated 7 < > the other compilers like Cobol, Fortran and even C.  < B < Well, the best you can say is that IBM saw the need for "legacy" < data types < even for fancy new CPUs.  4 In the business world, you still need fixed decimal.  C < As for the common base of their compilers, I slightly doubt that. B < The runtimes of their PL/I and Fortran were pretty incompatible,9 < at least on MVS. On AIX, it was better, since they were  < presumably developed < from scratch.   E Interesting point, IBM finally has a common language environment, LE, 	 something 7 vax had more than 20 years ago (thanks in part to PL/I)  <  < --- ( < Incoming mail is certified Virus Free.< < Checked by AVG anti-virus system (http://www.grisoft.com).A < Version: 6.0.735 / Virus Database: 489 - Release Date: 8/6/2004  <  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.735 / Virus Database: 489 - Release Date: 8/6/2004    ------------------------------  % Date: Thu, 02 Sep 2004 13:40:28 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> * Subject: Re: A whopping 50 percent...  ???, Message-ID: <41375B09.2A1AD7E3@teksavvy.com>   Michael Kraemer wrote:, > Well, the buzzword of the time was "open".I > VMS was as much "closed" as MVS or other "propietary" OSs at that time, < > UNIX were "open" systems with competition between vendors.  X In reality though, the "open" really didn't apply to the OS, but rather to the hardware.  M Remember Digital making sure nobody could build cards for the DEC proprietary N buses ? DEC not wanting free access to LAT ? DEC being very late in supportingK SCSI and before thatm insisiting we buy its more expensive proprietary disk  drives ?  N Unix was seen as open , not because one particular brand was open, but becauseK there was the image that all Unixes were compatible and that if one vendors J screwed you, you couls just swicth vendors and keep all your software. AndJ because the smaller companies couldn't really affors their own proprietaryM stuff, they were building more "industry standard" boxes and you could source & disk drives from different places etc.   ------------------------------  % Date: Thu, 02 Sep 2004 13:52:58 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> * Subject: Re: A whopping 50 percent...  ???, Message-ID: <41375DF7.D763B6B8@teksavvy.com>   Michael Kraemer wrote:C > which is what presumably most people do, apart from Cobol, maybe. O > Anyway, who cares about the instruction set when everybody uses HL languages.  > (OK, compiler writers ... ;-)   K RISC allowed for simpler design of chip which allowed for much higher clock K speeds. But the cost was many more instructions and bigger executables than D for CISC, so the performance was not necessarily proportional to the impressive Mhz rates.   N I'd like to know more about why early versions of Alpha had such terrible byteI manipulation. I still recall that during an early Alpha presentation at a N DECUS LUG meeting, I asked about that. So it isn't as if the Digital engineersM would have been ignorant about the need to compare bytes. (if *I* could see a K problem, then Shirley, the DEC engineers would have seen this problem too).   D However, byte manipulation limits didn't apply to all RISC machines.  H > when Power/PPC came around there wasn't such a large code base for it.  N Actually, there was. Because Apple did such a great job with its 68k emulator,N almost all of the old 86k software, including many "extensions" (aka: drivers)M were usable from day 1 on a PPC MAC. In fact, the early versions of MAC OS on J PPC had a great deal of 68k code still in it, and they had ported only theL most used portions portions (such as graphics) where programs spent the most time in.  N It is a real shame that Digital didn't imitate Apple when it moved from VAX toJ Alpha. That transition would have been far more succesful if all customersJ could have just  simply copied their executables, shareable images etc andN have the VMS operating system autonmatically invoke the translator dynamicallyN for each image file. This way, all software would have been available on AlphaM on day 1, with the natively compiled coming later for those software packages  that Palmer hadn't yet killed.   ------------------------------   Date: 2 Sep 2004 08:07:16 -0700 & From: jordan@ccs4vms.com (Rich Jordan)8 Subject: Re: Alphastation 200 storage controller options= Message-ID: <cc5619f2.0409020707.710bbc4e@posting.google.com>   k Keith Parris <keithparris_NOSPAM@yahoo.com> wrote in message news:<nHrZc.9395$RB4.1119@news.cpqcorp.net>... K > The Alphastation 200 is similar in most respects to the Alphaserver 300,  H > and I've had good luck so far in using the Supported Options List for  > the latter. See H > http://h18002.www1.hp.com/alphaserver/options/as300/as300_options.html  C Interesting, thanks.  The AS300 is actually more closely related to D the Alphastation 250 system (the AS400 seemed to have more in common@ with the AS200 based on specs) but if it works, I'm not going toF argue.  Time to pick up a KZPBA (which for some reason I don't have onD hand) and try it out.  Worst case (if it doesn't work in my AS200) IE can probably use it to replace a KZPDA in the AS600, or add a channel   to one of the PWS's around here.     Rich CCS    ------------------------------  $ Date: Thu, 2 Sep 2004 07:15:57 -0400* From: "Bill Todd" <billtodd@metrocast.net>1 Subject: Re: Big IA64 test coming in a few months = Message-ID: <qc2dnTCF94M9narcRVn-rw@metrocastcablevision.com>   8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:7uIH6wPRUXA6@eisner.encompasserve.org... 7 > In article <413505E0.7D672381@teksavvy.com>, JF Mezei & <jfmezei.spamnot@teksavvy.com> writes: > > K > > If the release of IA64 comes much after the 8686 for intel, it will add  fuel; > > to the argument that IA64 will always lag the industry.  > >  > ? > Not really.  You need to get out some more.  There was a good 3 > recent discussion about Montecito (2005) Itanium:  >  > http://tinyurl.com/3v2r5 > 4 > >Freq = 8720 GFLOPs/s/(218 nodes * 4 CPUs/node * 4 > >GFLOP/cycle) = 2.500 GHz. > > 2 > >A nice even number. Hmmm 20 GFLOP/s per socket, > >not bad at all! > H > It's been stated that Montecito would double 1'5 6M I2 performance. It appears - > clear now that it would be double per core.   L Except, of course that it *won't* be double per core, as further explorationK of that thread would reveal.  Not even if Montecito actually can run at 2.5eD GHz, which the difficulty that Intel is reportedly having bumping upJ Madison's speed (it's now being reported that they're cutting back MadisonK 9M to 1.6 GHz rather than the target 1.7 GHz, which is kind of ironic given.I the recent speculation at RWT by Itanic fans that they might hit 1.9 GHz) L calls into question even more than the problems they've had getting PrescottL (in 90 nm.) to run much of *any* faster than Northwood (in 130 nm.), despite> having extended the pipeline significantly in that transition.  I Not to mention the question of whether even if Montecito can clock at 2.5eL GHz they can *cool* it with both cores running at that speed.  There is someH talk of real miracles in this area, but then there's been enough of thatI kind of talk throughout Itanic's rather disappointing life that it's harde- not to take it with a grain or three of salt.a  B But even at 2.5 GHz, a Montecito core won't offer nearly twice theG performance of a current Madison core.  The clock rate will be only 67%SI faster, there'll only be twice the on-chip cache per core, and the memory ( subsystem won't be nearly twice as fast.  J The one thing that's always been consistent about your Itanic predictions,L Rob, is that the reality never comes anywhere near meeting them.  Until 2007J (when we'll see just what the Alpha people have come up with, and what IBMK has for POWER6, and what AMD has for K9), there's every indication that forpG high-end commercial workloads a POWER core will have close to twice theeJ throughput of an Itanic core, and that in the lower end (but with up to 32I cores per system in the Montecito time-frame, so it's not *that* low-end)hL Opteron will keep up with (and in many cases exceed) Itanic performance just' fine at a far lower system price point.   1 Perhaps you're the one who needs to get out more.h   - bill   ------------------------------  % Date: Thu, 02 Sep 2004 18:59:53 +0200c+ From: Wilm Boerhout <w3.boerhout@planet.nl>b# Subject: Re: charon vax emulator??? 5 Message-ID: <4137510f$0$2116$ba620dc5@nova.planet.nl>a   Larry Kilgallen wrote:   > 1 > How does that work on Alphas that have no USB ?s  H On Alphas, the USB license method is not used. Charon-VAX for Alphas is 2 an application that is licensed via a license PAK.   --  
 Wilm BoerhoutA Zwolle, The Netherlands    wilmOLD@PAINTboerhout.nl2    (remove OLD PAINT from this address before use)   ------------------------------   Date: 2 Sep 2004 03:17:35 -0700I$ From: gspamtackett@yahoo.com (Galen)' Subject: Re: Flat panel VT replacement?o= Message-ID: <bdc65a53.0409020217.477c8220@posting.google.com>R   Bob,  B Using your search suggestions I couldn't spot anything that lookedD applicable. Mostly what I found were carts, racks, etc., for holding multiple terminals or PCs.  = Perhaps I overlooked the right search result. Any more clues?u   Galen     v koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote in message news:<vt7HMRjhg6aK@eisner.encompasserve.org>... > D >    If you can't find on on the web, I was searching via Google forK >    "work benches", "work stations", and "lan work stations".  It was one t >    of the first few vendors. > D >    I actually used a flat screen VT200 emulators for some racks weK >    assembled about 10 years ago.  Excelent emulation, but I don't recall . >    the vendor. > % >    Aviod anything that uses a dock.?   ------------------------------   Date: 2 Sep 2004 07:13:22 -0500<; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)k' Subject: Re: Flat panel VT replacement? 3 Message-ID: <d3Ddhy4fVKPN@eisner.encompasserve.org>o  U In article <00A373E0.D80DB058@SendSpamHere.ORG>,   VAXman-  @SendSpamHere.ORG writes:  >  > H > New iMac G5!  Run eXodus on it and you have the ultimate Xterminal! ;) >   A    You have a keyboard map that works with TPU when I bring it upeC    under the DECwindows interface on my Mac?  This is the one thinguC    I miss when I access VMS from MacOS.  I have to edit using a VT mD    emulator because that's the layer that maps they keyboard for me,;    but I find features of the DECwindows TPU quite usefull.e   ------------------------------  % Date: Thu, 02 Sep 2004 14:57:03 +0200-0 From: Keith Cayemberg <keith.cayemberg@arcor.de>' Subject: Re: Flat panel VT replacement?bB Message-ID: <413718a0$0$28474$9b4e6d93@newsread4.arcor-online.net>   healyzh@aracnet.com wrote:3 > Keith Cayemberg <keith.cayemberg@arcor.de> wrote:  >  >>Try the following link...g >  > " >>MIDCOM - Planar ELT320 Terminals/ >>http://www.midcomdata.com/terminal/planar.htmy >  > N > ISTR, that these are not only no longer made, but quite expensive.  What I'dM > like to do is get a VT525, an LK451 (I think that's the right keyboard) and- > a cheap 15" LCD. > K > Compaq made some "Thin Clients" that are supposed to be a VT replacement, N > you could hook a LCD up to these as well, but I don't know how good their VTC > emulation is.  I think they're called the Compaq EVO T10 and T20.  >  > 		Zane  1 And here are the Links to the EVO Thin Clients...    HP Evo Thin Client T203 http://h18004.www1.hp.com/products/thinclients/t20/e   HP Evo Thin Client T303 http://h18004.www1.hp.com/products/thinclients/t30/a   HP t5000 Series Thin Clients? http://h18004.www1.hp.com/products/thinclients/index_t5000.htmlo    F And here are some other WBT (Windows Based Terminal)Thin Clients with  Terminal Emulation...f  ; Sire Technology Ltd - Axel - Windows-CE-based LCD Terminalss% http://www.sire.co.uk/unix/axel.shtml   & Optoma - Windows Based Terminals (WBT)% http://216.122.244.150/public/wbt.aspe  L Sire Technology Ltd - Optoma - 15" / 17" TFT-integrated Thin Client Terminal( http://www.sire.co.uk/Optoma/Optoma.shtm  / VisionBank - TFT LCD Windows-CE-based Terminals-3 http://mars.optomanetworks.com/eng/products/lcd.htm1  & Cygnet Electra: Thin-Clients: EL-5000M( http://www.cygnetelectra.com/el5000m.php  @ I have, of course, no experience concerning the quality of these9 terminals or of the correctness of their emulation modes.e   Cheers!    Keith Cayembergo   ------------------------------  # Date: Thu, 02 Sep 2004 16:34:50 GMTR" From:   VAXman-  @SendSpamHere.ORG' Subject: Re: Flat panel VT replacement?o0 Message-ID: <00A37471.B8EEB644@SendSpamHere.ORG>  q In article <d3Ddhy4fVKPN@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:cV >In article <00A373E0.D80DB058@SendSpamHere.ORG>,   VAXman-  @SendSpamHere.ORG writes: >> 0 >> sI >> New iMac G5!  Run eXodus on it and you have the ultimate Xterminal! ;)e >> V >-B >   You have a keyboard map that works with TPU when I bring it upD >   under the DECwindows interface on my Mac?  This is the one thingD >   I miss when I access VMS from MacOS.  I have to edit using a VT E >   emulator because that's the layer that maps they keyboard for me,r< >   but I find features of the DECwindows TPU quite usefull.  H I have a POwerbook G4 17" laptop with has built-in bluetooth.  I orderedI Apple's bluetooth wireless keyboard.  It has the look, feel and layout ofi the LK keyboards.   C This is a picture of my Powerbook with wireless keyboard and mouse.h  0 http://www.tmesis.com/pix/POWERBOOK_G4-17IN.JPG;  D I purchased eXodus X11 and I can log in with a full CDE environment.D The top 4 keys of the alternate keypad are redefined with xmodmap toC send out the LK codes so that it can by used with EDT or TPU or anyE( other VMS alternate keypad away utility.    o -- o< http://www.ProvN.com  for the *best* OpenVMS system security=                       solutions that others only claim to be.e -- i, Cyber-Terrorism (si'-ber tayr'-or-iz-em) n.:M   The release of, the sale of, or the use of any Micro$oft software product!   -- lK VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COMp   ------------------------------  % Date: Thu, 02 Sep 2004 08:38:34 +0200e* From: Paul Sture <nospam@sture.homeip.net> Subject: Re: foto * Message-ID: <2pntfeFmrrqnU1@uni-berlin.de>   Keith A. Lewis wrote: ~ > "Kerry.Main" <Kerry.Main@hp.com> writes in article <pleudxbnbajxjvubyif@Mvb.Saic.Com> dated Tue, 31 Aug 2004 15:19:40 -0600: >   >>----------bsjorwsjybwetuizdlad9 >>Content-Type: application/octet-stream; name="foto.zip"g# >>Content-Transfer-Encoding: base64h6 >>Content-Disposition: attachment; filename="foto.zip" >  >  > PROBE$ unzip foto.zips, > Archive:  PROBE$SCRATCH0:[LEWIS]FOTO.ZIP;1 >    creating: [.foto] >   inflating: [.foto]foto.html0 >    creating: [.foto.foto].$ >   inflating: [.foto.foto]foto1.exe > L > Looks like a virus to me.  Even though the html file is full of Mickeysoft> > references, I shall not execute foto1.exe on my VMS systems. >   3 Undoubtedly the latest Bagle worm as reported here:e  9 http://www.theregister.co.uk/2004/09/01/bagle_downloader/o   ------------------------------   Date: 2 Sep 2004 06:08:07 -0700 . From: fabiopenvms@yahoo.com.br (Fabio Cardoso) Subject: Future of DecnetS= Message-ID: <f30679fb.0409020508.56b4c6bc@posting.google.com>w  	 Dear Sirs   ; As Decnet is becoming an almost extincted protocol,why not e+ HP opens or give it: to IEEE for example ? y? May be it can be improved as an industrial automation protocol.t   Regards    FC   ------------------------------  % Date: Thu, 02 Sep 2004 14:43:16 +0100 - From: Roy Omond <Roy.Omond@BlueBubble.UK.Com>r Subject: Re: Future of Decnetr* Message-ID: <2pomfsFmq5qnU1@uni-berlin.de>   Fabio Cardoso wrote:   > Dear Sirso   (rather presumptuous ?)s  = > As Decnet is becoming an almost extincted protocol,why not  - > HP opens or give it: to IEEE for example ? lA > May be it can be improved as an industrial automation protocol.t  9 The DECnet protocol is completely open, unhindered by anym0 "proprietariness", as you are evidently unaware.  B Several third-party companies have produced DECnet implementations on many different platforms.  < The protocol specifications have always been completely open/ and freely available, as far as I can remember.   	 Roy Omonde Blue Bubble Ltd.   ------------------------------  % Date: Thu, 02 Sep 2004 15:01:55 +0100r+ From: Rodrigo Ventura <yoda@isr.ist.utl.pt>n Subject: Re: Future of Decnetg) Message-ID: <m3isawdb1o.fsf@pixie.isrnet>h  > >>>>> "Roy" == Roy Omond <Roy.Omond@BlueBubble.UK.Com> writes:  E     Roy> The protocol specifications have always been completely opent8     Roy> and freely available, as far as I can remember.  F The same cannot be said about LAT, can it? I only recall reading about7 DECnet, and failing to find LAT protocol specification.i  > Regarding implementations of DECnet, the standard linux kernel* supports that protocol from some time now.   Cheers,i   Rodrigoe   -- j  : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboaa4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------  % Date: Thu, 02 Sep 2004 15:07:49 +0100s- From: Roy Omond <Roy.Omond@BlueBubble.UK.Com>4 Subject: Re: Future of Decnetr* Message-ID: <2pontuFnnqttU1@uni-berlin.de>   Rodrigo Ventura wrote:  : >>"Roy" == Roy Omond <Roy.Omond@BlueBubble.UK.Com> writes: >xG >     Roy> The protocol specifications have always been completely opent: >     Roy> and freely available, as far as I can remember. > , > The same cannot be said about LAT, can it?   No.  What's your point?    > I only recall reading about 9 > DECnet, and failing to find LAT protocol specification.i  = As expected.  LAT has absolutely *nothing* to do with DECnet.g  @ > Regarding implementations of DECnet, the standard linux kernel, > supports that protocol from some time now.  + Yep, I know.  Adds credence to what I said.b  	 Roy Omondu Blue Bubble Ltd.   ------------------------------   Date: 02 Sep 2004 14:48:27 GMT/ From: Thierry Dussuet <thierry@dussuet.lugs.ch>o Subject: Re: Future of Decnet 0 Message-ID: <slrncjecls.107.thierry@MARS.Family>   Hello!  ; On 2004-09-02, Rodrigo Ventura <yoda@isr.ist.utl.pt> wrote: ? >>>>>> "Roy" == Roy Omond <Roy.Omond@BlueBubble.UK.Com> writes:h >sF >    Roy> The protocol specifications have always been completely open9 >    Roy> and freely available, as far as I can remember.- >-H > The same cannot be said about LAT, can it? I only recall reading about9 > DECnet, and failing to find LAT protocol specification.e  L IIRC they're just hard to find.  There's a LAT implementation for Linux (for FreeBSD, too).  $ http://linux-decnet.sourceforge.net/   Thierryr   ------------------------------  % Date: Thu, 02 Sep 2004 15:49:27 +01002+ From: Rodrigo Ventura <yoda@isr.ist.utl.pt>  Subject: Re: Future of Decnetr) Message-ID: <m3eklkd8ug.fsf@pixie.isrnet>E  > >>>>> "Roy" == Roy Omond <Roy.Omond@BlueBubble.UK.Com> writes:  E     Roy> The protocol specifications have always been completely openo8     Roy> and freely available, as far as I can remember.1     >> The same cannot be said about LAT, can it?         Roy> No.  What's your point?  > Thats that was unfortunate. I head somewhere LAT was much moreB efficient than telnet in interactive sessions. Moreover, IIRC, LATF announces it services to the net, it is possible to browse LAT serversE in a network. Nothing of that is possible with telnet (unless you uses nmap or something like that...)U  ? If LAT specs were openly published, maybe it would be much more  widespread.   , But now we have ssh, and all is different...    B     >> I only recall reading about DECnet, and failing to find LAT     >> protocol specification.  F     Roy> As expected.  LAT has absolutely *nothing* to do with DECnet.  F LAT and DECnet have much in common, namely the fact of both have beingF developed by DEC, and are used in VMS, Ultrix, Digital UNIX, and maybe	 also Tru.s    E     >> Regarding implementations of DECnet, the standard linux kernelk1     >> supports that protocol from some time now.   4     Roy> Yep, I know.  Adds credence to what I said.  E Yes it does. It was a misconception from you to assume my post was torC remove credence from what you said! In fact, my post purpose was ton add, not to subtract.r   Cheers,n   Rodrigoe   -- r  : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboaf4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------  % Date: Thu, 02 Sep 2004 08:52:13 -0700a+ From: "Barry Treahy, Jr." <Treahy@MMaz.com>t Subject: Re: Future of Decneto' Message-ID: <413741AD.4070101@MMaz.com>    Roy Omond wrote:   > Fabio Cardoso wrote: >a >> Dear Sirs >  >  > (rather presumptuous ?)  >,G >> As Decnet is becoming an almost extincted protocol,why not HP opens  E >> or give it: to IEEE for example ? May be it can be improved as an  " >> industrial automation protocol. >r >F; > The DECnet protocol is completely open, unhindered by anyc2 > "proprietariness", as you are evidently unaware. >e DECnet yes, LAT no...    Barryi   --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                        -   ------------------------------   Date: 2 Sep 2004 01:30:39 -07000% From: Bart.Zorn@xs4all.nl (Bart Zorn)B7 Subject: Re: How do recover a missing member from RAID?e= Message-ID: <a98cd882.0409020030.5b8d7b24@posting.google.com>.  M You should use RAID INIT <disk> /SPARE and then RAID REPLACE <raidset> <disk>M   Check HELP RAID to be sure!    Regards,  	 Bart Zorn   ] "Yong Boon, Lim" <y0ngb00n@yahoo.com> wrote in message news:<ch603e$c4f$1@news6.jaring.my>... 3 > l've RAID5 with the following configuration, i.e.o > J > RAID Software for OpenVMS V2.5-2   Display Time:  2-SEP-2004 09:34:22.84> > Copyright 2002 Hewlett-Packard Company. All Rights Reserved. >  > RAID Array Parameters: > ( >         Current RAID Array ID:    DATA( >         Permanent RAID Array ID:  DATA% >         RAID Level:               5 * >         Current State:            NORMAL* >         Associated Spareset:      (none) >  > RAID Array Configuration:  >  >         Member' >     Index    Name               Statee' >     -----   ------              -----i( >       0     _PSS$DKA100:        NORMAL( >       1     _PSS$DKA200:        NORMAL( >       2     _PSS$DKA300:        NORMAL > , >  Virtual                         RecoveredN >    Unit        Size    Status   Operations      Reads      Writes     ErrorsN >  -------      ------  --------  ----------      -----      ------     ------N >  DPA0001:  142114552  ACCESS           141  411931221     8429095         20 > K > One of my SCSI harddisk DKA100 has been accidentally formatted, do anyones, > know how do recover from the remain member
 > of the set?  > 
 > Thank a lotm >  > Yong Boon, Lim   ------------------------------  % Date: Thu, 02 Sep 2004 12:41:10 +0200y3 From: Michael Unger <spam.to.unger@spamgourmet.com> + Subject: Re: Itanium: EFI... no fs0: devicee* Message-ID: <2pocm6Fndop5U1@uni-berlin.de>  - On 2004-09-01 22:11, "Fred Kleinsorge" wrote:    > [...]e > J > OK imagine it's a FC disk so a second system running Windows can see it.	 > "Could"oK > Windows hack the stuff around?  Maybe.  But it would mean that someone on  > the G > Windows system was deliberately repartitioning a disk with *existing*  > partitions --mN > NOT unassigned space.  In which case you will most likely have a corrupt VMS > disk.i > I > But hey.  Give physical access to a drive, and unlimited privs - and it  > doesn't matterL > what the OS is if someone wants to be malicious.  But I don't believe that
 > the Windows J > tools will "gratuitously" allow you to repartition a drive - or do it on > it's own.   E You may be correct with respect to Window's own tools; but what aboutPA worms or viruses targeted at "Windows 64bit for Itanic"? They are H running on an Itanic CPU, supported by an OS which knows of the relevantH disk structures (including EFI and partitions), and are able to (mis)use all of the APIs available.   > [...]l > M > The GUID is *not* a Microsoft structure.  It is a specification produced by  > Intel.  : But it has to be known to MS and its "world famous" OS ...   > [...]a  E And isn't it HP's concept having all OSes (Winwoes, Linux, HP-UX, andt9 even OpenVMS) run on the same system (on different CPUs)?i   Michaeli   -- s; Real names enhance the probability of getting real answers.c5 My e-mail account at DECUS Munich is no longer valid.D   ------------------------------   Date: 2 Sep 2004 07:14:56 -05002; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)e+ Subject: Re: Itanium: EFI... no fs0: device?3 Message-ID: <UehkDtTiG1Ew@eisner.encompasserve.org>.  \ In article <413603C0.D4BBBF5B@teksavvy.com>, Jf Mezei <jfmezei.spamnot@teksavvy.com> writes: > J > If/when Windows becomes available for IA64, and you have a system with 2O > physical drives, one loaded with Windows, will Windows see the "VMS" drive as N > 2 distinct drives (one with the EFI FAT file system, and one with a strange, > unknown file system on it ?)  9    Windows has been shipping for I64 for quite some time.r  E    Why add Windows to a system when what you want is VMS and what youg    need to boot is EFI?t   ------------------------------  # Date: Thu, 02 Sep 2004 14:05:45 GMTe3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond)/+ Subject: Re: Itanium: EFI... no fs0: devicei2 Message-ID: <ZMFZc.9447$y15.3196@news.cpqcorp.net>  , In article <413625BE.306D697@teksavvy.com>, / JF Mezei <jfmezei.spamnot@teksavvy.com> writes:e  L >Ok, this is where I am perplexed. If the drive is partitioned, wouldn't VMSM >just see its own partition on that logical drive ? Wouldn't it see the otherfO >partitions as different drives on the system ? If so, how come you wouldn't beaD >able to have multiple logical  drives on the same physical drive ?   I VMS does not concern itself with "partitions".  It sees a disk with files$J on it.  One of those files contains what the Itanium console software sees+ as a partition containing boot information.c   -- tJ       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------  # Date: Thu, 02 Sep 2004 16:47:48 GMTD" From:   VAXman-  @SendSpamHere.ORG+ Subject: Re: Itanium: EFI... no fs0: devicew0 Message-ID: <00A37473.88D6E0C0@SendSpamHere.ORG>   OK...   F After several days of dicking with this box, I now have a FS0: device.  ! I start to install VMS and I get:    Portion done: 0%...10% ...20%...30%...40%...50%...60%  T %PCSI-E-WRITEERR, error writing DISK$OPENVMSIV8_1:[VMS$COMMON.][SYSLIB]STARLET.OLB;2% -LBR-E-DUPKEY, duplicate key in index " %PCSI-E-OPFAILED, operation failedE Terminating is strongly recommended.  Do you want to terminate? [YES]h   :(   -- h< http://www.ProvN.com  for the *best* OpenVMS system security=                       solutions that others only claim to be.y --  , Cyber-Terrorism (si'-ber tayr'-or-iz-em) n.:M   The release of, the sale of, or the use of any Micro$oft software product!   -- gK VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COMd   ------------------------------  % Date: Thu, 02 Sep 2004 13:34:17 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>6+ Subject: Re: Itanium: EFI... no fs0: devicei, Message-ID: <41375996.C3A58C97@teksavvy.com>   Bob Koehler wrote:G >    Why add Windows to a system when what you want is VMS and what youe >    need to boot is EFI?   L Because I am trying to understand the use of the word "partition" along withK the notion that intel chose "industry standard structures for GUID and FAT.l  L So it seems that VMS bypasses the partition information and thinks the wholeN physical drive belongs to itself, and manages to allow others (namely the IA64I primitive ROM) to see partitions  by strategically placing the right dataa/ structures at the right physical disk adresses.c  M Consider the case of a disk being in some sort of array. For whatever reason,mL that disk is reassigned to another IA64 which runs Windows, Linux, or HP-UX.E Will windows see 2 drives (since the GUID would indicate the drive isl partitioned in 2 drives) ?  M Is it a given that in terms of the partition information in the GUID, the EFI F boot partition has blocks which do not straddle the other partition ?   K Is it correct to state that the second partition (the one that contains theyN rest of the disk) contains an imcomplete image of the ODS drive since it wouldL lack those disk blocks that contain the EFI partition and which the ODS file! system thinks belongs to itself ?T  M Consider cases where disk array software is being asked to backup drives. You-N would want to make sure that the array backup software uses the physical driveF as source instead of the 2 logical drives it woudl see due to the GUID partitining information.   ------------------------------   Date: 2 Sep 2004 17:07:31 GMTo" From: " E???k " <gzgimsun@163.net> Subject: Look at this girl...c3 Message-ID: <ch7k0j$4gk2227@imsp212.netvigator.com>m  > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD>N <META http-equiv=3DContent-Type content=3D"text/html; charset=3Dwindows-1252">9 <META content=3D"MSHTML 6.00.2800.1276" name=3DGENERATOR>0 <STYLE></STYLE> ) </HEAD><FONT face=3DArial><FONT size=3D2>R <BODY>( <DIV>The gal in China - Zhong Shan</DIV> <DIV>&nbsp;</DIV>D3 Call +86 13802741877 Law Yau for free chat.<BR><BR> K <DIV><IMG alt=3Dhttp://server5.uploadit.org/files/zhongman8888-MeiMei.jpg =cQ hspace=3D0=20 src=3D"http://server5.uploadit.org/files/zhongman8888-MeiMei.jpg" =sA align=3Dbaseline=20 border=3D0></DIV></BODY></HTML></FONT></FONT>R   ------------------------------  * Date: Thu, 2 Sep 2004 06:55:25 +0000 (UTC)% From: Dan Foster <usenet@evilphb.org>e& Subject: Multinet v5 ssh login failure6 Message-ID: <slrncjdh13.uo9.usenet@gaia.roc2.gblx.net>  F I'm seeing a behavior that I don't quite understand with Multinet v5.0! on an OpenVMS/Alpha 7.3-1 system.c  F I can't call this one in to support, and it isn't really that critical? since things work. :) So it's just more of a curiosity, really.   H If I do a ssh v2 login to SYSTEM on my VMS box, dubhe... it says there'sD a login failure since last login and at the same time, the following OPCOM message appears on OPA0:  8 %%%%%%%%%%%  OPCOM   1-SEP-2004 23:38:47.52  %%%%%%%%%%%' Message from user AUDIT$SERVER on DUBHEiQ Security alarm (SECURITY) and security audit (SECURITY) on DUBHE, system id: 10256/ Auditable event:          Network login failures1 Event time:                1-SEP-2004 23:38:47.52a* PID:                      0000021A        * Process name:             SSHD 0000       * Username:                 system          / Remote nodename:          <a remote IP address>d, Remote node id:           <a remote node ID>$ Remote username:          SSH:SYSTEMQ Status:                   %LOGIN-F-USERAUTH, error accessing authorization recorde  8 Yet I don't see either message when I do a ssh v1 login.  H I'm not sure I understand what exactly is special about ssh v2 vs v1 forC %LOGIN-F-USERAUTH such that ssh v1 logins don't trigger this? Is its9 something to do with password lifetime/expiration checks?   C Is there a way to pin down what exactly ssh v2 is trying to do witha/ respect to the user authentication information?e  F I know it's trying to access a record, but it's not clear which record" nor why the access attempt failed.  A Of note: the system is freshly installed, and right now, the onlya modifications to it are:  @ 	a) It has five key 7.3-1 ECOs applied (and rebooted after each)< 	   Latest version of each ECO: RMS, UPDATE, PCSI, XFC, LAN.  C 	b) Multinet 5.0 is the only third party layered product installed.s@ 	   No add'l LPs has been installed for the base system, either.  A 	c) No modifications has been made to the base install other than @ 	   for starting Multinet, DECnet, and SYS$BATCH... defining key? 	   systemwide logicals (eg SYSUAF), as well as adding a singler 	   user account.e  @ 	d) No DECwindows running, and system has already been AUTOGEN'd" 	   for proper Multinet operation.   -Dan   ------------------------------  % Date: Thu, 02 Sep 2004 06:39:44 -0600e% From: Dan O'Reilly <dano@process.com>(* Subject: Re: Multinet v5 ssh login failureA Message-ID: <6.1.2.0.2.20040902063643.026286f0@raptor.psccos.com>g  C If you're using the standard "vanilla" template for SSH2_CONFIG and/J SSHD2_CONFIG, then what's going on is that there are likely authenticationE methods being tried for which your system and/or user isn't correctly:G configured.  The default methodologies are to try publickey first, then:G password.  If the publickey mechanism hasn't been correctly configured,o0 then you'll see a login failure because of that.   You have a few options:   F 1. Correctly configure publickey (there's an example in the MN books).E 2. Add the "StrictIntrusionLogging No" line to SSH2_DIR:SSHD2_CONFIG.i> 3. Modify the AllowedAuthentications keyword to allow only the$     authentication methods you want.  ' At 12:55 AM 9/2/2004, Dan Foster wrote: G >I'm seeing a behavior that I don't quite understand with Multinet v5.0v" >on an OpenVMS/Alpha 7.3-1 system. > G >I can't call this one in to support, and it isn't really that critical @ >since things work. :) So it's just more of a curiosity, really. >tI >If I do a ssh v2 login to SYSTEM on my VMS box, dubhe... it says there'snE >a login failure since last login and at the same time, the followingr >OPCOM message appears on OPA0:e >o9 >%%%%%%%%%%%  OPCOM   1-SEP-2004 23:38:47.52  %%%%%%%%%%%,( >Message from user AUDIT$SERVER on DUBHEJ >Security alarm (SECURITY) and security audit (SECURITY) on DUBHE, system 	 >id: 1025-0 >Auditable event:          Network login failure2 >Event time:                1-SEP-2004 23:38:47.52# >PID:                      0000021Ae$ >Process name:             SSHD 0000! >Username:                 systeme0 >Remote nodename:          <a remote IP address>- >Remote node id:           <a remote node ID>e% >Remote username:          SSH:SYSTEMoL >Status:                   %LOGIN-F-USERAUTH, error accessing authorization  >records > 9 >Yet I don't see either message when I do a ssh v1 login.h >oI >I'm not sure I understand what exactly is special about ssh v2 vs v1 foreD >%LOGIN-F-USERAUTH such that ssh v1 logins don't trigger this? Is it: >something to do with password lifetime/expiration checks? >hD >Is there a way to pin down what exactly ssh v2 is trying to do with0 >respect to the user authentication information? >aG >I know it's trying to access a record, but it's not clear which record # >nor why the access attempt failed.t > B >Of note: the system is freshly installed, and right now, the only >modifications to it are:) > I >         a) It has five key 7.3-1 ECOs applied (and rebooted after each) E >            Latest version of each ECO: RMS, UPDATE, PCSI, XFC, LAN.  >cL >         b) Multinet 5.0 is the only third party layered product installed.I >            No add'l LPs has been installed for the base system, either.r > J >         c) No modifications has been made to the base install other thanI >            for starting Multinet, DECnet, and SYS$BATCH... defining keytH >            systemwide logicals (eg SYSUAF), as well as adding a single >            user account. >lI >         d) No DECwindows running, and system has already been AUTOGEN'dT+ >            for proper Multinet operation.s >  >-Danc   ------J +-------------------------------+----------------------------------------+J | Dan O'Reilly                  |  "There are 10 types of people in this |J | Principal Engineer            |   world: those who understand binary   |J | Process Software              |   and those who don't."                |J | http://www.process.com        |                                        |J +-------------------------------+----------------------------------------+   ------------------------------  * Date: Thu, 2 Sep 2004 15:13:44 +0000 (UTC)% From: Dan Foster <usenet@evilphb.org>m* Subject: Re: Multinet v5 ssh login failure6 Message-ID: <slrncjee7f.uo9.usenet@gaia.roc2.gblx.net>  h In article <6.1.2.0.2.20040902063643.026286f0@raptor.psccos.com>, Dan O'Reilly <dano@process.com> wrote:E > If you're using the standard "vanilla" template for SSH2_CONFIG andrL > SSHD2_CONFIG, then what's going on is that there are likely authenticationG > methods being tried for which your system and/or user isn't correctly I > configured.  The default methodologies are to try publickey first, thenhI > password.  If the publickey mechanism hasn't been correctly configured,t2 > then you'll see a login failure because of that.  4 Ahhh! That was indeed it; thanks! (Works great now.)   -Dan   ------------------------------  # Date: Thu, 02 Sep 2004 14:31:53 GMT 1 From: Michael Austin <maustin@firstdbasource.com>h Subject: Re: need help= Message-ID: <t9GZc.16046$dE6.2680@newssvr22.news.prodigy.com>    Larry Kilgallen wrote:a > In article <c4fc6627.0409010614.31e9f5e9@posting.google.com>, supe_ulhas@rediffmail.com writes:@ > ! >>how do i access SETCIM records,g. >>what is the exact structure of setcim record >  >  > What is a SETCIM record ?e >  > What is a SETCIM ?  K SETCIM is a tool for extracting and displaying data from a Digital Control :O System.  You can access data points of various valves and meters etc... from a dN DCS.  Used in manufacturing sites - I used it at a DuPont site 7-8 years ago. ; For a complete description, go to the aspentech.com site...g   --   Michael Austin. F Donations still welcomed. Http://www.firstdbasource.com/donations.html :)   ------------------------------  % Date: Thu, 02 Sep 2004 08:55:36 -0700.+ From: "Barry Treahy, Jr." <Treahy@MMaz.com>s Subject: Single Signon' Message-ID: <41374278.1020802@MMaz.com>t  F Just curious how many folks here have been successful in implementing C single signon in a hetro environment with VMS, Windows Server, and d@ Linux/Unix systems, and at what type of cost for the supporting G framework...  If discussing off-list is preferred, please feel free to u contact me directly...     Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                        h   ------------------------------  % Date: Thu, 02 Sep 2004 13:18:26 -0400e- From: JF Mezei <jfmezei.spamnot@teksavvy.com>*3 Subject: Re: SMTP: stray messages left in directorye, Message-ID: <413755E1.6256E9A5@teksavvy.com>   John Santos wrote:E > What's your uplink to the Internet?  If you are going through a NAThF > router and PPPOE on a DSL circuit, you might be having MTU problems.  K But wouldn't such problems happen for every message that contains more thanoN 1492 bytes ? (eg: any message where the SMTP server would send a packet largerL than 1492 bytes and where the router would be assumed to fragment the packet automatically ?y  M I can try to tell my vax to limit its MTU to 1492 (which would affect the LAN7H side as well), but I woudl have to wait weeks to find out if the problem reoccurs or not.   ------------------------------   Date: 2 Sep 2004 07:21:09 -0500i; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)y, Subject: Re: strange decnet-plus problems...3 Message-ID: <hL9qgpiNwKYr@eisner.encompasserve.org>v  f In article <4093a3af.0409010949.5d858ebc@posting.google.com>, magalettac@massdor.com (Carmine) writes:F > I can only set host to my new system via decnet ,if I first set hostA > from this box to others , has anyone seen anything like this ??t >  > Node1o > $ set hos Node2h? > %SYSTEM-F-UNREACHABLE, remote node is not currently reachablel >   F    Something's wrong with your DECnet routing.  What are you using for    a router?   ------------------------------   Date: 2 Sep 2004 06:56:11 -0700 & From: magalettac@massdor.com (Carmine), Subject: Re: strange decnet-plus problems...= Message-ID: <4093a3af.0409020556.52bce379@posting.google.com>o  C Yes , the node infomartion is correct on both nodes , each have thelE others correct information . Is there a process that stays alive when 5 an initial session gets created that then goes away ?n    X Dirk Munk <munk@home.nl> wrote in message news:<ch59t5$5sg$1@news5.tilbu1.nb.home.nl>...B > Did you add your new node to the name space of the other nodes ? >  > (MC DECNET_REGISTER) >  > Carmine wrote:H > > I can only set host to my new system via decnet ,if I first set hostC > > from this box to others , has anyone seen anything like this ??p > > 	 > > Node1g > > $ set hos Node22A > > %SYSTEM-F-UNREACHABLE, remote node is not currently reachableg > > 	 > > Node1t > > $ telnet node2 > > (works fine...)o	 > > Node2  > > $ set hos node1k > > (works fine...)  > > logout from both nodes > > 	 > > Node1o > > $ set hos node2  > > (works for a while...) > >  > > Later in day r	 > > Node1o > > $ set hos node2eA > > %SYSTEM-F-UNREACHABLE, remote node is not currently reachable. > > 	 > > Node1o. > > $ set hos node1(to itself , always works ) > > ? > > This is a brand new system is there a patch I am missing ??s
 > > VMS 7.3-2t > > decnet-osi for VMS   ------------------------------  * Date: Thu, 2 Sep 2004 15:32:29 +0000 (UTC)6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER), Subject: Re: strange decnet-plus problems...1 Message-ID: <newscache$zv6f3i$pxu1$1@news.sil.at>-  f In article <4093a3af.0409020556.52bce379@posting.google.com>, magalettac@massdor.com (Carmine) writes:D >Yes , the node infomartion is correct on both nodes , each have theF >others correct information . Is there a process that stays alive when6 >an initial session gets created that then goes away ?  & No. It looks like a cache times out...   -- y Peter "EPLAN" LANGSTOEGERn% Network and OpenVMS system specialistp E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Wed, 01 Sep 2004 21:48:31 +0200T. From: "Hans M. Aus" <aus@vim.uni-wuerzburg.de>+ Subject: Re: Two IPs on same NIC OVMS 7.2-1r7 Message-ID: <aus-72DBA1.21483101092004@news.cis.dfn.de>-   An observation:   # After adding the second IP address:   H 1) The printers that we always leave ON continued to work fine with the  raw IP 141.- QUE address.   H 2) With the printers which we turn off in the evening, the QUEs stalled D in the morning. After $stop/next/que and $start/que printing worked H again. The problem went away after we changed the QUE  address IP 10.- .   ------------------------------  * Date: Thu, 2 Sep 2004 15:49:05 +0000 (UTC)6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)6 Subject: [VMS V7.3-2] PCSI INSTALL /SAVE_RECOVERY_DATA1 Message-ID: <newscache$on7f3i$y0v1$1@news.sil.at>o  L I recently installed the last ECOs for VMS V7.3-2 and noticed that something7 changed in the way the ECOs behave during installation.w  J Until last month (most if not) all VMS732_* ECOs asked for saving recoveryF information if one hasn't specified /SAVE_RECOVERY_DATA during INSTALL which looked like this:T  % 	               *********************i  @ 	You have the option of saving directories, files, and libraries? 	that are replaced, modified, or deleted by the installation of/A 	this remedial kit.  These files along with a copy of the product 9 	database will be stored on the system disk in a separates@ 	directory tree.  Choosing to save this recovery data will allow> 	you to uninstall this patch at a later time using the PRODUCT@ 	UNDO PATCH command.  If recovery data is not saved you will not6 	be able to remove this remedial kit from your system.  ! 	               **** WARNING ****e  : 	If you choose to install this kit without saving recovery= 	data, you will lose your ability to uninstall any previouslyr@ 	installed remedial kits for which recovery data has been saved. 	M0 	Do you wish to save this recovery data [YES] ?:  G As the Default was YES and the question was asked I got lazy and didn'toG specify /SAVE_RECOVERY_DATA during ECO install (though the HELP clearlyr3 states that /NO_SAVE_RECOVERY_DATA is the default).     G With the last ECOs I installed (VMS732_HBMM, VMS732_TRACE, VMS732_F11X) E I noticed that this has changed now and no such question is asked but> a different warning appears:  O  * If you continue, recovery data for the patches listed above will be deleted.tK  * The deletion of recovery data does not affect the installation status ofeL  * patches applied to products that are not participating in this operation.I  * However, continuing with this operation prevents you from uninstallingwK  * these patches at a future time by use of the PRODUCT UNDO PATCH command.i    Do you want to continue? [NO]    G So, I now have to specify /SAVE_RECOVERY_MODE with my ECO installation.tD (and yes, I already lost recovery data by not reading carefully ;-).  $ Charlie, could you comment, please ?  G Is there a logical to make /SAVE_RECOVERY_MODE default (like I did withd PCSI$LOG and PCSI$TRACE) ?   TIAr   -- h Peter "EPLAN" LANGSTOEGERt% Network and OpenVMS system specialisti E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------   End of INFO-VAX 2004.487 ************************