1 INFO-VAX	Fri, 20 Apr 2001	Volume 2001 : Issue 220       Contents: /SUBSTITUTE_DOMAIN, Re: 1GHz Alpha EV68CB Microprocessor Exposed, RE: 1GHz Alpha EV68CB Microprocessor Exposed, Re: 1GHz Alpha EV68CB Microprocessor Exposed, Re: 1GHz Alpha EV68CB Microprocessor Exposed, Re: 1GHz Alpha EV68CB Microprocessor Exposed, Re: 1GHz Alpha EV68CB Microprocessor Exposed, Re: 1GHz Alpha EV68CB Microprocessor Exposed, Re: 1GHz Alpha EV68CB Microprocessor Exposed, Re: 1GHz Alpha EV68CB Microprocessor Exposed, Re: 1GHz Alpha EV68CB Microprocessor Exposed# BUY AND SELL ALMOST ANYTHING ONLINE $ C++ RTL pb at AST level execution...( Re: C++ RTL pb at AST level execution..., Re: Re: C++ RTL pb at AST level execution... Re: DECserver gurus... Re: DECserver gurus... Re: DECserver gurus..." Re: FreeVMS for AMD a pipe dream ?* Re: Hardcoding MTU under UCX (TCPIP V5.0a)* Re: Hardcoding MTU under UCX (TCPIP V5.0a)) Re: Listing all the mailboxes on a system ) Re: Listing all the mailboxes on a system ) Re: Listing all the mailboxes on a system ) Re: Listing all the mailboxes on a system $ Re: MAIL$SEND_ADD_ATTRIBUTE ("From")$ Re: MAIL$SEND_ADD_ATTRIBUTE ("From")$ Re: MAIL$SEND_ADD_ATTRIBUTE ("From")$ Re: MAIL$SEND_ADD_ATTRIBUTE ("From")$ Re: MAIL$SEND_ADD_ATTRIBUTE ("From")/ RE: OpenVMS article - please explain last line! 1 Re: OpenVMS Job seeking in Hong Kong, SAR, China. 1 Re: OpenVMS Job seeking in Hong Kong, SAR, China. 1 Re: OpenVMS Job seeking in Hong Kong, SAR, China.  Re: software ethernet sniffer  Spammed by I-S-E today Re: Spammed by I-S-E today Re: Spammed by I-S-E today Re: Spammed by I-S-E today Re: Spammed by I-S-E today. TCP/IP io$_acpcontrol inetacp$c_hostent_offset TFT and VMS  Re: TFT and VMS  Re: TFT and VMS P This is a fun one ... high interrupt rate after VMS upgrade to 7.2-1 ... but waiP Re: This is a fun one ... high interrupt rate after VMS upgrade to 7.2-1 ... butP Re: This is a fun one ... high interrupt rate after VMS upgrade to 7.2-1 ... but, Re: Using 3-phase power converters on VAXen?, Re: Using 3-phase power converters on VAXen?, Re: Using 3-phase power converters on VAXen? Re: VLCs0 Re: [BUG?] FREEWARE CD MENU: license expired :-(2 Re: [BUG] wrong APACHE file location after install2 Re: [BUG] wrong APACHE file location after install# [Q] default APACHE directory change ' Re: [Q] default APACHE directory change ' Re: [Q] default APACHE directory change ' Re: [Q] default APACHE directory change  Re: [Q] Free VMS WEB Server  [Q] Making Perl work with CSWS" Re: [Q] Making Perl work with CSWS  F ----------------------------------------------------------------------   Date: 19 Apr 2001 20:38:10 GMT' From: helbig@man.ac.uk (Phillip Helbig)  Subject: /SUBSTITUTE_DOMAIN - Message-ID: <9bnibi$1t2b$1@godfrey.mcc.ac.uk>   3 I can't see any difference whether or not HIDDEN is # specified with TCPIP SET CONF SMTP.   5 It looks like HIDDEN even when NOHIDDEN is specified.   : I know that with earlier versions of TCPIP, though perhaps. for different reasons, there WAS a difference.  : Perhaps someone can send me two example messages, with and without HIDDEN specified.    Thanks!    ------------------------------  % Date: Thu, 19 Apr 2001 16:16:44 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> 5 Subject: Re: 1GHz Alpha EV68CB Microprocessor Exposed , Message-ID: <3ADF47A7.F8D19156@videotron.ca>   Dirk Munk wrote:Q > We know that 1.6 GHz alpha's already were in existence more than a year ago. So R > producing a reliable 1 GHz system should be possible by know. AFAIK the 1.01 GHz- > ES45 should be out in a few weeks from now.   N Intel seems to have very little difficulty in increasing the speed of its 8086M at a rapid pace. is it correct to state that they get about 2 speed increases = per year ? (in Mhz measures, not necessarily in performance).   G Is there anything would would have prevented Compaq from producing 1ghz M versions of the existing alpha chip, just like Intel produced 1+ ghz versions  of its pentium III ?  M I realise that the upcoming chip incldes design improvements, but from a pure M mhz point of view, must the design be changed to allow greater speed, or just D the manufacturing process ? (with perhaps minor tewaking of design).   ------------------------------  % Date: Thu, 19 Apr 2001 15:34:35 -0500 + From: "Main, Kerry" <Kerry.Main@compaq.com> 5 Subject: RE: 1GHz Alpha EV68CB Microprocessor Exposed R Message-ID: <DC4745D1A85CA04180C83CDC706A9D180D9543@cthexc02.americas.cpqcorp.net>   JF,   H While I agree that the x86 chips have done well in increasing CPU speed,J keep in mind these greater than 1Ghz systems are primarily single cpu (WS)	 designs.    K Try and find a dual cpu (or more) server from any of the major players with L each cpu having greater than 1Ghz speed. The top x86 servers from Compaq and2 Dell are in the range of 700Mhz and 900Mhz models.  ? Also, as to what is happening with these P4 systems, check out: = http://www.inqst.com/articles/p4bandwidth/p4bandwidthmain.htm  "Can the Pentium 4 Recover?"   Regards,  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Services  Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----4 From: JF Mezei [mailto:jfmezei.spamnot@videotron.ca] Sent: April 19, 2001 4:17 PM To: Info-VAX@Mvb.Saic.Com 5 Subject: Re: 1GHz Alpha EV68CB Microprocessor Exposed      Dirk Munk wrote:I > We know that 1.6 GHz alpha's already were in existence more than a year  ago. So I > producing a reliable 1 GHz system should be possible by know. AFAIK the  1.01 GHz- > ES45 should be out in a few weeks from now.   I Intel seems to have very little difficulty in increasing the speed of its  8086C at a rapid pace. is it correct to state that they get about 2 speed 	 increases = per year ? (in Mhz measures, not necessarily in performance).   G Is there anything would would have prevented Compaq from producing 1ghz D versions of the existing alpha chip, just like Intel produced 1+ ghz versions of its pentium III ?  H I realise that the upcoming chip incldes design improvements, but from a pureH mhz point of view, must the design be changed to allow greater speed, or justD the manufacturing process ? (with perhaps minor tewaking of design).   ------------------------------    Date: 20 Apr 2001 05:15:21 +0800, From: Paul Repacholi <prep@prep.synonet.com>5 Subject: Re: 1GHz Alpha EV68CB Microprocessor Exposed - Message-ID: <87bsps7frq.fsf@prep.synonet.com>   / JF Mezei <jfmezei.spamnot@videotron.ca> writes:    > Dirk Munk wrote:  F > > We know that 1.6 GHz alpha's already were in existence more than aE > > year ago. So producing a reliable 1 GHz system should be possible F > > by know. AFAIK the 1.01 GHz ES45 should be out in a few weeks from > > now.   That is good news, if it is so.   D > Intel seems to have very little difficulty in increasing the speedC > of its 8086 at a rapid pace. is it correct to state that they get F > about 2 speed increases per year ? (in Mhz measures, not necessarily > in performance).  A I don't think 2x per year is accurate. This many be the case in a F single year period, a late ship, and an early follow on could do this.  D > Is there anything would would have prevented Compaq from producingD > 1ghz versions of the existing alpha chip, just like Intel produced& > 1+ ghz versions of its pentium III ?  C > I realise that the upcoming chip incldes design improvements, but D > from a pure mhz point of view, must the design be changed to allowB > greater speed, or just the manufacturing process ? (with perhaps > minor tewaking of design).  F In principle, just re-layout and go! Practise is a fair bit narkier...F First off, although we generally see a single number for feature size,C in reality there are several, and they may not scale 1:1 to the new @ process. So this adds to the re-layout, and, more important, can< change the circuit due to differing coupling between traces,E resitivity etc. So this may require a fair bit of work to get it back  on the line.  D But the new Alphas ar not just a shrink; they are also going to a CuC interconect, and to a slightly different CMOS. The Cu needs a total D re-layout to quite different rules to get the best from it, and thisB would probably need a second tape-out before production. Also, theF silicon is different to the Intel and Samsung lines, so that has to be factored in.  D Also, you probably have minor changes you want to do. Some parameterD could be making life miserable, so a small shift could be a big win.= Real production could give you a different skew to the normal C disribution, so a tweek can get more devises into the useable range  for you system design.  E Then, even if everything goes sweet in the fab, then you have to test D the little buggers. Can your, or your vendors testers cope? Have youE caused subtle timing changes that bite you, or cause race conditions? ! It the yield enough to live with?   3 Then you get to fight putting them into the system!    Then you get to do it again...  E BTW, anyone know the details of the package for the new 264? It seems 4 to be quite a change from what has been used so far.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  # Date: Thu, 19 Apr 2001 21:41:47 GMT - From: Terry C Shannon <shannon@world.std.com> 5 Subject: Re: 1GHz Alpha EV68CB Microprocessor Exposed D Message-ID: <Pine.SGI.4.21.0104191734510.21129-100000@world.std.com>  % On 20 Apr 2001, Paul Repacholi wrote:   1 > JF Mezei <jfmezei.spamnot@videotron.ca> writes:  >  > > Dirk Munk wrote: > H > > > We know that 1.6 GHz alpha's already were in existence more than aG > > > year ago. So producing a reliable 1 GHz system should be possible H > > > by know. AFAIK the 1.01 GHz ES45 should be out in a few weeks from
 > > > now. > ! > That is good news, if it is so.  > G Alas, it does not appear to be the way things are gonna be. ES45s do in F fact exist (have seen same, back in January IIRC) but the original MayD launch date has been pushed out to 4CQ01. Apparently it's supply andG demand: all the early system orders have been placed by HPTC customers, ? and CPQ can't satisfy the existing orders until late this year.   H Rather than announce now and deliver in November, CPQ chose to delay theI announcement. There obviously is more to the matter than just speeding up D the assembly lines, else CPQ would be doing just that. Especially in. light of what's happening in the Wintel space.  G > BTW, anyone know the details of the package for the new 264? It seems 6 > to be quite a change from what has been used so far. >   I Well, it weighs more than its predecessor and it's larger. Also difficult 4 to de-lid or peel apart as was done for the photos.   H This is the second version of EV68B I've seen. The one that fell into my? hands back in October Y2K did not have pins stickin' out of it.    ------------------------------    Date: 20 Apr 2001 06:17:04 +0800, From: Paul Repacholi <prep@prep.synonet.com>5 Subject: Re: 1GHz Alpha EV68CB Microprocessor Exposed - Message-ID: <87y9sw5ycf.fsf@prep.synonet.com>   / Terry C Shannon <shannon@world.std.com> writes:   F > Rather than announce now and deliver in November, CPQ chose to delayC > the announcement. There obviously is more to the matter than just > > speeding up the assembly lines, else CPQ would be doing justD > that. Especially in light of what's happening in the Wintel space.  D The mind boggles to wonder what the situation would be like if times3 where a booming, instead of dropping into the pits.   D BTW, Netcraft has released their web report. Big winner, Apache. TheC downer is T64, gone from Alta Vista, Compaq and Amazon. Seems to be @ the case for all the comercial unixs, as BSD and Linux eat theirG shorts. Well done Q managment. You are Bills highlight in the report...    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Thu, 19 Apr 2001 20:03:33 -0400 # From: Paul DeMone <pdemone@igs.net> 5 Subject: Re: 1GHz Alpha EV68CB Microprocessor Exposed ' Message-ID: <3ADF7CD5.C6CF0990@igs.net>    Paul Repacholi wrote:  [...] H > Terry, do my eyes, or the photos, decieve me, or is the heatsink mount( > slug now the full size of the package?  > Low Theta Jc is where its at. I want to see the heat sink that	 bolts on.    > E > For those who have not noticed, Intel have suspended their new fab, G > so that leaves only IBM and TSMC with .13u fabs (does AMD have a .13u  > line in production?).   ; If they suspended one that leaves them with what? just four = 0.13 um fabs in the pipeline? I heard it was still full steam 1 ahead on the $7.5B capital expenditure this year.      --D Paul W. DeMone       The 801 experiment SPARCed an ARMs race of EPICE Kanata, Ontario      proportions to put more PRECISION and POWER into G demone@mosaid.com    architectures with MIPSed results but ALPHA's well $ pdemone@igs.net      that ends well.   ------------------------------  % Date: Thu, 19 Apr 2001 20:28:25 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> 5 Subject: Re: 1GHz Alpha EV68CB Microprocessor Exposed , Message-ID: <3ADF82A6.329295C0@videotron.ca>   "Main, Kerry" wrote:J > While I agree that the x86 chips have done well in increasing CPU speed,L > keep in mind these greater than 1Ghz systems are primarily single cpu (WS)
 > designs. > M > Try and find a dual cpu (or more) server from any of the major players with N > each cpu having greater than 1Ghz speed. The top x86 servers from Compaq and4 > Dell are in the range of 700Mhz and 900Mhz models.  M The reason I asked is that Alphe seems to get large increments with very long J periods of no news in between, so the chip tends to get forgotten and this< doesn't help those trying to quell rumours of Alpha's death.  K Which made me wonder just how difficult it would be for Compaq to release a K few improved alphas (just higher mhz, no new features) in between the major I revisions. That would give Alpha better visibility in the marketplace and 4 would help it keep up with the 8086 game controller.   ------------------------------  # Date: Fri, 20 Apr 2001 03:39:40 GMT - From: Terry C Shannon <shannon@world.std.com> 5 Subject: Re: 1GHz Alpha EV68CB Microprocessor Exposed C Message-ID: <Pine.SGI.4.21.0104192338460.5335-100000@world.std.com>   ' On Thu, 19 Apr 2001, Paul DeMone wrote:    >  >  > Paul Repacholi wrote:  > [...] J > > Terry, do my eyes, or the photos, decieve me, or is the heatsink mount* > > slug now the full size of the package? > @ > Low Theta Jc is where its at. I want to see the heat sink that > bolts on.  >   B I haven't seem the heat sink but I imagine it's rather prodigious!   ------------------------------  % Date: Fri, 20 Apr 2001 00:44:43 -0400 # From: Paul DeMone <pdemone@igs.net> 5 Subject: Re: 1GHz Alpha EV68CB Microprocessor Exposed ' Message-ID: <3ADFBEBB.7C4CCC92@igs.net>    JF Mezei wrote:  >  > Dirk Munk wrote:S > > We know that 1.6 GHz alpha's already were in existence more than a year ago. So T > > producing a reliable 1 GHz system should be possible by know. AFAIK the 1.01 GHz/ > > ES45 should be out in a few weeks from now.  > P > Intel seems to have very little difficulty in increasing the speed of its 8086O > at a rapid pace. is it correct to state that they get about 2 speed increases ? > per year ? (in Mhz measures, not necessarily in performance).  > I > Is there anything would would have prevented Compaq from producing 1ghz O > versions of the existing alpha chip, just like Intel produced 1+ ghz versions  > of its pentium III ?  G I think you have inadvertently answered your own question. What version F of PIII was released over 1 GHz? The infamous 1.133 GHz part. RememberE what happened to it? That is why PC MPU speed grades can be bumped in F 5 or 10% increments but server chips have fewer and bigger bumps - theJ server chips have to be exhausitively tested, verified, and characterized.I First in silicon and then in systems and that can take the better part ofrI a year (much more if issues turn up - the whole process starts over againC after the fix).E   > O > I realise that the upcoming chip incldes design improvements, but from a puresO > mhz point of view, must the design be changed to allow greater speed, or justiF > the manufacturing process ? (with perhaps minor tewaking of design).  E It is a process port. In a full custom design that is uses aggressiveeE physical and circuit design like the EV6x core there is a hell of loteH of fine tuning to do. There was a paper at ISSCC in February on DECpaq'sG port strategy to IBM's CU 0.18 um process including some of the circuituD changes that had to be made. At ISSCC 2000 there was a similar paper! for the port to 0.18 um aluminum.0  D Here's an example with numbers. The HP PA-8000 core is a complicatedB beast like EV6 but doesn't quite use as much funky circuitry. WhenA HP ported it from their 0.5 um process to Intel's 0.25 um processe@ (PA-8500) it originally simulated maxxing out at 275 MHz (it ranB up to 180 MHz in 0.5 um). It took a heck of a lot of work to bring> it up to the 400 MHz targeted release frequency. It eventuallyC topped out at 440 MHz. More work and a Leff process tweak lifted it  to 552 MHz (PA-8600).g   --D Paul W. DeMone       The 801 experiment SPARCed an ARMs race of EPICE Kanata, Ontario      proportions to put more PRECISION and POWER intoeG demone@mosaid.com    architectures with MIPSed results but ALPHA's wellS$ pdemone@igs.net      that ends well.   ------------------------------  % Date: Fri, 20 Apr 2001 00:31:22 -0500 * From: cjt & trefoil <cheljuba@prodigy.net>5 Subject: Re: 1GHz Alpha EV68CB Microprocessor ExposedM+ Message-ID: <3ADFC9AA.70BCB123@prodigy.net>   C To crossover threads slightly with another one here on comp.os.vms, B is it going to require a three phase power converter to drive one?   Terry C Shannon wrote: > ' > On 20 Apr 2001, Paul Repacholi wrote:s > 3 > > JF Mezei <jfmezei.spamnot@videotron.ca> writes:v > >s > > > Dirk Munk wrote: > >GJ > > > > We know that 1.6 GHz alpha's already were in existence more than aI > > > > year ago. So producing a reliable 1 GHz system should be possibleeJ > > > > by know. AFAIK the 1.01 GHz ES45 should be out in a few weeks from > > > > now. > >E# > > That is good news, if it is so.P > >dI > Alas, it does not appear to be the way things are gonna be. ES45s do inSH > fact exist (have seen same, back in January IIRC) but the original MayF > launch date has been pushed out to 4CQ01. Apparently it's supply andI > demand: all the early system orders have been placed by HPTC customers,eA > and CPQ can't satisfy the existing orders until late this year.f > J > Rather than announce now and deliver in November, CPQ chose to delay theK > announcement. There obviously is more to the matter than just speeding upvF > the assembly lines, else CPQ would be doing just that. Especially in0 > light of what's happening in the Wintel space. > I > > BTW, anyone know the details of the package for the new 264? It seems-8 > > to be quite a change from what has been used so far. > >  > K > Well, it weighs more than its predecessor and it's larger. Also difficultc5 > to de-lid or peel apart as was done for the photos.n > J > This is the second version of EV68B I've seen. The one that fell into myA > hands back in October Y2K did not have pins stickin' out of it.h   ------------------------------  # Date: Fri, 20 Apr 2001 04:30:35 GMTe From: gzhwfo@adsmontreal.com, Subject: BUY AND SELL ALMOST ANYTHING ONLINE8 Message-ID: <LXOD6.15905$g_.240026@wagner.videotron.net>   Visit the webs best free online classified ads web site where you can buy and sell almost anything right online! Its fast, easy and FREE!l   ------------------------------  + Date: Thu, 19 Apr 2001 19:20:17 +0000 (UTC),; From: Olivier.Hurez-Martin@ca.com ("Hurez-Martin, Olivier")I- Subject: C++ RTL pb at AST level execution...dF Message-ID: <21AAC260C0BCD411AE9B009027AA4DE702FB9EBF@usilms03.ca.com>  ( Who said AST programming is safe on VMS?   Everytime I do more than 2 or 3 operations in an AST (which I strive to keep as concise as possible, for that matter...), I run into problems in the C or C++ runtimes!r  MHere is an example on OpenVMS Alpha V7.2, with a C++ V6.2-035  ( CXXL$011_SHR.EXE image ident: "CXXL V2.0-0" ):  The AST (5 code lines; 2 to 5 too many, that is...) interrupt the main program while it was in the C++ runtime and reentered it... and of course, it is hanging in there...  I can't disable the ASTs as I am using threads!r   Grrrrr!a  t Would anyone know where the program is hanging (or how to find it out...) using the debug and SDA information below?   I tried to locate the relative addresses of those calls in SYS$SHARE:CXXL$011_SHR.EXE just above my program routines, e.g. 00000000000091B4 (using analyze/image to list the global symbol table) without any luck...n   Thanks for any help!   	-- Olivier.    
 DBG> sho callnI  module name     routine name      line           rel PC           abs PC:O                                               0000000000000000 0000000000000000mO ----- the above appears to be a null frame in the same scope as the frame belowoO  SHARE$PTHREAD$RTL                            0000000000012868 00000000001C6868 O  SHARE$PTHREAD$RTL                            0000000000043888 00000000001F7888rO  SHARE$PTHREAD$RTL                            000000000003F658 00000000001F3658bu  SHARE$CMA$TIS_SHR_DATA1                      0000000000001A54 FFFFFFFF80501A54     <<<<<<<<<<<< C++ RTL at AST levellO  SHARE$CXXL$011_SHR                           000000000001A7A0 00000000007247A0 O  SHARE$CXXL$011_SHR                           0000000000018920 0000000000722920aO  SHARE$CXXL$011_SHR                           0000000000000278 000000000070A278mO  SHARE$CXXL$011_SHR                           00000000000091B4 00000000007131B4+f *CLIB_LOCK       ClearEvent       70141       000000000000CD18 000000000007CE58  <<<<<<< AST triggeredO  SHARE$PTHREAD$RTL                            000000000005F4A0 00000000002134A0eO  SHARE$PTHREAD$RTL                            0000000000060114 0000000000214114.O                                               FFFFFFFF800EDD6C FFFFFFFF800EDD6ClO                                               FFFFFFFF800EDC44 FFFFFFFF800EDC44 O                                               FFFFFFFF800E2C58 FFFFFFFF800E2C58mO  SHARE$PTHREAD$RTL                            000000000003F62C 00000000001F362Cal  SHARE$CMA$TIS_SHR_DATA1                      0000000000001A54 FFFFFFFF80501A54  <<<<< C++ RTL at main levelO  SHARE$CXXL$011_SHR                           000000000001A7A0 00000000007247A0oO  SHARE$CXXL$011_SHR                           0000000000018920 0000000000722920cO  SHARE$CXXL$011_SHR                           0000000000000278 000000000070A278tO  SHARE$CXXL$011_SHR                           00000000000091B4 00000000007131B4sz *CLIB_LOCK       Lock             69475       0000000000003344 0000000000073484      <<<<< Current routine in main programO *CLIB_SEM        Unlock           68337       00000000000029E4 000000000009C924aO *CLIB_SEM        semop            73208       000000000000D3CC 00000000000A730CuO *TESTIPC         main             72967       00000000000029B4 000000000006B544tO *TESTIPC         __MAIN                       00000000000000A0 0000000000068C30vO                  POKE_MBX                     00000000000F9E2C 0000000000109E2CiO ----- the above appears to be a null frame in the same scope as the frame below-O                                               0000000000028A20 0000000000038A20yO  SHARE$PTHREAD$RTL                            000000000004EE6C 0000000000202E6CyO  SHARE$PTHREAD$RTL                            0000000000030B20 00000000001E4B20oO                                               FFFFFFFF82EA14DC FFFFFFFF82EA14DCi   -----------------------------------------------------------------------------------------------------------------------------------------t    x FWIW, under SDA, I can see the program is waiting for an event flag, which is not any of those I am using in my program: From the PCB: F Waiting EF cluster              0    Event flag wait mask     80C06338   SDA> eval 80C06338D Hex = FFFFFFFF.80C06338   Decimal = -2134875336         LNM$AQ_MUTEX     -- r+ Posted from mail1.cai.com [141.202.248.38] e1 via Mailgate.ORG Server - http://www.Mailgate.ORGa   ------------------------------  % Date: Thu, 19 Apr 2001 15:47:26 -0400e" From: Dan Sugalski <dan@sidhe.org>1 Subject: Re: C++ RTL pb at AST level execution... : Message-ID: <5.0.2.1.0.20010419154459.023a7008@24.8.96.48>  9 At 07:20 PM 4/19/2001 +0000, Hurez-Martin, Olivier wrote: ) >Who said AST programming is safe on VMS?u >aI >Everytime I do more than 2 or 3 operations in an AST (which I strive to eJ >keep as concise as possible, for that matter...), I run into problems in  >the C or C++ runtimes!t  L I wouldn't write any AST code that used calls in the C++ runtime, honestly. K I wouldn't trust that it was AST-safe or reentrant. (Personally I wouldn't lK write any AST code in C++ at all, but that's just me) If you must I'd make  K sure that exceptions were disabled for the C++ code, amongst other things.  G (And make really, really sure that you don't go allocating memory even   indirectly)-  B >Here is an example on OpenVMS Alpha V7.2, with a C++ V6.2-035  ( J >CXXL$011_SHR.EXE image ident: "CXXL V2.0-0" ):  The AST (5 code lines; 2 K >to 5 too many, that is...) interrupt the main program while it was in the .A >C++ runtime and reentered it... and of course, it is hanging in f: >there...  I can't disable the ASTs as I am using threads!  J You're not doing anything nasty like calling pthread routines from within F your AST code or anything, right? You can't do that since most of the ! pthread calls aren't thread-safe.    					Dan  I --------------------------------------"it's like this"------------------- 2 Dan Sugalski                          even samurai? dan@sidhe.org                         have teddy bears and eveni;                                       teddy bears get drunks   ------------------------------  + Date: Fri, 20 Apr 2001 02:57:18 +0000 (UTC) ; From: Olivier.Hurez-Martin@ca.com ("Hurez-Martin, Olivier")h5 Subject: Re: Re: C++ RTL pb at AST level execution... F Message-ID: <21AAC260C0BCD411AE9B009027AA4DE702FD8273@usilms03.ca.com>   Thanks for the quick answer :-)m    M > I wouldn't write any AST code that used calls in the C++ runtime, honestly..6 > I wouldn't trust that it was AST-safe or reentrant.    That is my feeling too...t  P > You're not doing anything nasty like calling pthread routines from within yourJ > AST code or anything, right? You can't do that since most of the pthread > calls aren't thread-safe.   H Nope... I know pthreads pretty well, the only calls that AST safe on VMSO are the ones with the _int_np() suffix, according to the doc. That is basically ' only one: pthread_cond_signal_int_np()        ! Unfortunately, I need the ASTs...0C I am striving to implement a signaling mechanism between processes,> using the lock manager:   W Process C (consumers, n instances possible...) acquire(s) a lock at Protected Read modeh\ with a blocking AST, and go wait on an event flag it(they) originally allocated and cleared.  N Process P (data producer...) queues a request for that lock at EXclusive mode,0 thus triggering the blocking AST in consumers...  D The blocking AST gets the event flag number as a parameter, sets it,; releasing C processes, which convert the lock back to NULL.u  X Process P's request for the lock gets granted.  It converts it back to NULL immediately.  Y The blocking AST is fairly simple... It does a trace (I see your eyebows shifting :-) pls V bare with me a few lines more...) I added to find out another problem in the same areaX (trace is conditioned on some flag previously set/unset by lnm)  and uses sys$setef()  )  R That thing works just fine for anywhere from a few thousands to 100's of thousandsR iterations. Then it gets stuck on that funny LNM$AQ_MUTEX - The funny part is thatD it hangs on that event flag, ** whether the trace is on or off ** ! M SDA shows no AST active in both cases, but when the trace is on, the debugger W points to the trace in the AST (which is the only thing that makes some sense to me)...eY When it is off, the consumer process hangs on the application event flag, as if it didn't Z get the signal.  Another funny thing at that point is that it does not hold the lock in PRL mode as it should, since the conversion back to null is one step downstream.     	-- Olivier.   -- h+ Posted from mail3.cai.com [141.202.248.42] d1 via Mailgate.ORG Server - http://www.Mailgate.ORGC   ------------------------------   Date: 19 Apr 2001 18:01:14 GMT# From: casinoop2@aol.com (CasinoOp2)m Subject: Re: DECserver gurus...a: Message-ID: <20010419140114.28269.00000286@ng-ch1.aol.com>   >>>>* As you may have DECnet Plus installed, ... Check you have MOP enabled.1; Next put your DS boot files in MOP$load instead of MOM$load  <<<<<t  K Yes, we have DECnet Plus and I THINK I have MOP enabled. How can I tell FORwM SURE? Also I mentioned that I am not receiving any console messages regarding-N load failures, it is possible I have these turned off by mistake? I know I canN write a message to OPA0: and see the console screen pop-up, but I noticed thatC login failures do not generate messages (which they do on the VAX).,   ... the saga continues.g      Clark   ------------------------------   Date: 19 Apr 2001 18:48:13 GMT# From: casinoop2@aol.com (CasinoOp2)i Subject: Re: DECserver gurus....: Message-ID: <20010419144813.03351.00000326@ng-fi1.aol.com>  H Finally success! I needed to define the logical MOP$NAMED_LOAD using the following command.  9 define/system/exec mop$named_load sys$sysroot:[decserver].  < Then away she went! Thanks everybody for all of your help!!!   Clarke   ------------------------------  % Date: Thu, 19 Apr 2001 23:34:44 +0200 , From: "Bart Zorn" <B.Zorn@TrueBit.n0spam.nl> Subject: Re: DECserver gurus... ; Message-ID: <3adf59c5$0$32946$e4fe514c@newszilla.xs4all.nl>e  J I'm glad that it works now. Generally, I would recommend to use LANACP and3 LANCP to provide download services in stead of MOP.i Have a look at it!  
 Bart Zorn,
 True Bit B.V.p  0 "CasinoOp2" <casinoop2@aol.com> wrote in message4 news:20010419144813.03351.00000326@ng-fi1.aol.com...J > Finally success! I needed to define the logical MOP$NAMED_LOAD using the > following command. >o; > define/system/exec mop$named_load sys$sysroot:[decserver]r >s> > Then away she went! Thanks everybody for all of your help!!! >a > Clarko   ------------------------------  % Date: Thu, 19 Apr 2001 17:57:11 -0500i1 From: "David J. Dachtera" <djesys.nospam@fsi.net> + Subject: Re: FreeVMS for AMD a pipe dream ? ' Message-ID: <3ADF6D47.5A310616@fsi.net>o   Jan Vorbrueggen wrote: > * > Rudolf Wingert <win@fom.fgan.de> writes: > ? > > AFAIK the 64bit CPU of AMD will be incompatible to Intel's.0 > M > What Intel cpu are you talking about? If IA32, it will be fully compatible,BI > more compatible in fact than any AMD chip in existence (because it willzH > support all the instruction set extensions). If IA64, it will be fullyH > incompatible, because its design goal was to 1) support 64 bit addressO > spaces with minimal changes to the existing IA32 ISA, and 2) add instructionstN > (similar to Intel's SSE and SSE2) to finally enable the processor to compete; > in floating-point performance with the rest of the world.r > P > No way porting VMS to xHammer is going to be significantly easier than porting > it to any other ISA. > 
 >         Jant  G There have been countless threads on this topic. The problem is not thesH CPU, witness: Alpha, 64-bit RISC. VAX, 32-bit CISC. Hardly compatible at the outset.n  D The "problem" is the system-board "architecture", if you can call itG that. Only 16 interrupts, non-vectored. There are countless denizens ofaF this group more qualified than I to explain what VMS on VAX needed and/ Alpha "mobos" were designed to provide as well.s  : The CPU is not the problem - well, not the ENTIRE problem.  E No, what is needed is a mobo for (Intel, AMD, etc.) that provides thee8 same requirements. THEN the CPU issues can be addressed.   See, BT? I *DO* learn!   -- h David J. Dachterar dba DJE Systemsf http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/v  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.t   ------------------------------  % Date: Thu, 19 Apr 2001 14:00:29 -0400 % From: "Brian Tillman" <tillman_brian>I3 Subject: Re: Hardcoding MTU under UCX (TCPIP V5.0a)u" Message-ID: <3adf2819@news.si.com>  L >I have a need to set the MTU for the ethernet interface SE0 to be somethingK >less than the default of 1500.  I've found a way of doing this once UCX is6 >up and running: >e/ >    $ mcr TCPIP$IFCONFIG se0 down ipmtu 576 up  >iL >but the setting doesn't "stick".  Once I restart UCX, the old default valueI >of 1500 comes back.  How do I set this as a permanent setting under UCX?o   What about:   ; $ TCPIP SET CONFIGURATION PROTOCOL TCP/MTU_SEGMENT_SIZE=576v  " Stop, then restart TCPIP Services.   I've not tried this. --A Brian Tillman                   Internet: tillman_brian at si.comeA Smiths Aerospace                          tillman at swdev.si.comp= 3290 Patterson Ave. SE, MS      Addresses modified to prevent.< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Fri, 20 Apr 2001 10:14:44 +1000cB From: "Matt Muggeridge" <Matt.Muggeridge@compaq.ssppaammffree.com>3 Subject: Re: Hardcoding MTU under UCX (TCPIP V5.0a) 3 Message-ID: <OdLD6.1065$fB6.26585@news.cpqcorp.net>r  B These types of changes can be made to stick by adding them to your" TCPIP$SYSTARTUP command procedure.   Matt.   ? "John R. Lawson, Jr." <John.Lawson@TLMAGE.net> wrote in messagee. news:HumD6.37717$m8.414246@news.direcpc.com...C > I have a need to set the MTU for the ethernet interface SE0 to ber	 somethingnL > less than the default of 1500.  I've found a way of doing this once UCX is > up and running:  >y0 >     $ mcr TCPIP$IFCONFIG se0 down ipmtu 576 up > G > but the setting doesn't "stick".  Once I restart UCX, the old defaultd valuesJ > of 1500 comes back.  How do I set this as a permanent setting under UCX? >i >(   ------------------------------    Date: 20 Apr 2001 02:08:23 +0800, From: Paul Repacholi <prep@prep.synonet.com>2 Subject: Re: Listing all the mailboxes on a system- Message-ID: <87ofts7ofc.fsf@prep.synonet.com>P  6 "Michael D. Ober" <mdo.@.wakeassoc.com.nospam> writes:  A > Is there anyway to list all the VMS Mail mailboxes on a system?o   sho dev mban   -- e< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.a@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Thu, 19 Apr 2001 14:28:52 -0400e From: William_Bochnik@acml.com2 Subject: Re: Listing all the mailboxes on a system> Message-ID: <OFBA1E250C.A50BAEAA-ON85256A33.00656FAE@acml.com>  > He asked VMS MAIL - they're listed in sys$system:vmsmail.dat -9 you're on your own for dcl code to read that, though.....o      d                                                                                                     d                     Paul                                                                            d                     Repacholi                    To:  Info-VAX@Mvb.Saic.Com                         d                     <prep@prep.sy                cc:                                                d                     nonet.com>           Subject:     Re: Listing all the mailboxes on a system     d                                                                                                     d                     04/19/2001                                                                      d                     02:08 PM                                                                        d                                                                                                     d                                                                                                           6 "Michael D. Ober" <mdo.@.wakeassoc.com.nospam> writes:  A > Is there anyway to list all the VMS Mail mailboxes on a system?'   sho dev mbaS   --< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.          F ______________________________________________________________________  : The information contained in this transmission may contain< privileged and confidential information and is intended only< for the use of the person(s) named above. If you are not the< intended recipient,  or an employee or agent responsible for? delivering this message to the intended recipient,  any review,a@ dissemination, distribution or duplication of this communication> is strictly prohibited. If you are not the intended recipient,A please contact the sender immediately by reply e-mail and destroy1# all copies of the original message.    ------------------------------  % Date: Thu, 19 Apr 2001 14:37:43 -04001# From: Jim Agnew <Agnew@hsc.vcu.edu>42 Subject: Re: Listing all the mailboxes on a system* Message-ID: <3ADF3077.44E9C48@hsc.vcu.edu>  ) do you mean vms email mailboxes, or what?S  Q get mailuaf from goatley's fileserver, and that will do nicely.. it's what i usesE   "Michael D. Ober" wrote: > A > Is there anyway to list all the VMS Mail mailboxes on a system?  >  > --	 > Thanks,  > Mike Ober.   ------------------------------  % Date: Thu, 19 Apr 2001 18:03:07 -050001 From: "David J. Dachtera" <djesys.nospam@fsi.net>S2 Subject: Re: Listing all the mailboxes on a system' Message-ID: <3ADF6EAB.6A86D9B4@fsi.net>D   William_Bochnik@acml.com wrote:0 > @ > He asked VMS MAIL - they're listed in sys$system:vmsmail.dat -; > you're on your own for dcl code to read that, though.....    DUMP/RECORD should do nicely.F   -- C David J. Dachtera  dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/0  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------  % Date: Thu, 19 Apr 2001 19:05:35 -0000 - From: wspencer@ap.nospam.org (Warren Spencer) - Subject: Re: MAIL$SEND_ADD_ATTRIBUTE ("From")0/ Message-ID: <tdudnvbkoqms85@news.supernews.com>   / Steve.Lionel@compaq.com (Steve Lionel) wrote in0. <d89udtsj1kaaugtut13i60gv2vb915r68n@4ax.com>:   < >On Thu, 19 Apr 2001 16:56:15 -0000, wspencer@ap.nospam.org  (Warren0 >Spencer) wrote: >_ >>Hi,  >>A >>I'm attempting to use the MAIL$ Utility routines to create and S send a, >>mail message.  The destination address is  SMTP%"someone@somebox.org" -  
 -- snip --   > B >I don't know if Multinet has any bearing on this.  I know that I  usePB >this feature all the time.  Here is my Fortran code that does it: >h >! >! Add attributes  >!3 >        IN_ITMLST(1).ITMCOD = MAIL$_SEND_FROM_LINE - >        IN_ITMLST(1).BUFLEN = SIZEOF(SENDER)0+ >        IN_ITMLST(1).BUFADR = %LOC(SENDER)   >        IN_ITMLST(1).RETADR = 01 >        IN_ITMLST(2).ITMCOD = MAIL$_SEND_SUBJECT * >        IN_ITMLST(2).BUFLEN = SUBJECT_LEN, >        IN_ITMLST(2).BUFADR = %LOC(SUBJECT)  >        IN_ITMLST(2).RETADR = 0 >        IN_ITMLST(3) = EOLST-( >        ISTAT = MAIL$SEND_ADD_ATTRIBUTE (MSG_CTX,IN_ITMLST,OUT_ITMLST)6 >        IF (.NOT. ISTAT) CALL LIB$SIGNAL(%VAL(ISTAT)) >W? >The only thing I notice different about your code is that you w don't B >specify zero as the RETADR, not that I think this should make any >difference.  ; To test it, I put a zero in RETADR - no change in behavior.0  @ >You should get a SS$_NORPIV error if your process doesn't have  SYSPRVB >and you try to use this feature.  Can you verify in the debugger  that' >the itemlist has the correct contents?-  > I turned off SYSPRV and tested; you are (almost) correct - it  fails with SS$_NOSYSPRIV.@  A The contents of the item list appears to be correct; here's what g the debugger sees:   DBG> ex AttribIList. MAILTEST\AttribIList[0:1]9     [0]0         BufLength:      20         ItemCode:       8m         BufAdddress:    131104         RetLenAddress:  262148     [1]a         BufLength:      0          ItemCode:       0e         BufAdddress:    0n         RetLenAddress:  0o     DBG> ex/az 131104o. MAILTEST\SendFrom[0]:   "spock@enterprise.com"  - But here's a question:  should the string be tA "spock@enterprise.com", or perhaps "From: spock@enterprise.com",  ! or maybe something else entirely?  -- f1 << Marriage is Grand.  Divorce is Fifty Grand. >>a   Warren Spencer Senior Software Engineer The Associated Press  ? ** My employer does not necessarily agree with my statements **-   ------------------------------  % Date: Thu, 19 Apr 2001 20:10:25 -0000X- From: wspencer@ap.nospam.org (Warren Spencer)e- Subject: Re: MAIL$SEND_ADD_ATTRIBUTE ("From")./ Message-ID: <tduhhhl3t40he0@news.supernews.com>e  / Steve.Lionel@compaq.com (Steve Lionel) wrote in . <d89udtsj1kaaugtut13i60gv2vb915r68n@4ax.com>:   C >On Thu, 19 Apr 2001 16:56:15 -0000, wspencer@ap.nospam.org (Warrenc >Spencer) wrote: >e >>Hi,w >>G >>I'm attempting to use the MAIL$ Utility routines to create and send a H >>mail message.  The destination address is SMTP%"someone@somebox.org" -  
 -- snip --  3 Well, problem solved.  Thanks for your reply Steve.   L Seems if you use MAIL$SEND_ADD_ATTRIBUTE to add a "from" line *after* using L MAIL$SEND_ADD_ADDRESS, it doesn't work.  Sadly, the example on page MAIL-25 A of the "OpenVMS Utility Routines" manual (v7.1) demonstrates the :J *incorrect* order for these calls, but since the example does not attempt . to change the "From" field, it probably works.   ws   --  1 << Marriage is Grand.  Divorce is Fifty Grand. >>7   Warren Spencer Senior Software Engineer The Associated Press  ? ** My employer does not necessarily agree with my statements **t   ------------------------------  % Date: Thu, 19 Apr 2001 16:10:30 -0400f- From: JF Mezei <jfmezei.spamnot@videotron.ca>a- Subject: Re: MAIL$SEND_ADD_ATTRIBUTE ("From")o, Message-ID: <3ADF4632.EBD9E607@videotron.ca>   Warren Spencer wrote:p >  > Hi,a > L > I'm attempting to use the MAIL$ Utility routines to create and send a mailJ > message.  The destination address is SMTP%"someone@somebox.org" - and it# > all seems to work fine - except -t > = > I need to alter the "From:" field on the outbound message. .  M Have you tried sending an email without a foreign transport ? (eg: sending toh another user on the same node).n  C When you are sending email thorugh a foreign transport, the foreigndN transport's code is invoked a few times during the process, the first of whichH is when you type the destination's name (to valaidate name, open networkI connections and return error messages right away) and once the message is = complete, the routines are called again (to confirm sending).e  J It is also quite possible that the "SMTP%" foreign transport is discarding6 that attribute to prevent someone from forging emails.   ------------------------------  % Date: Thu, 19 Apr 2001 16:12:17 -0400a- From: JF Mezei <jfmezei.spamnot@videotron.ca>a- Subject: Re: MAIL$SEND_ADD_ATTRIBUTE ("From") , Message-ID: <3ADF469D.8BB7CC55@videotron.ca>   Steve Lionel wrote:l= > >I need to alter the "From:" field on the outbound message.t  - >I know that I use this feature all the time.o  0 Forging your bosses' signature ? :-) :-) :-) :-)   ------------------------------  % Date: Thu, 19 Apr 2001 17:03:47 -0400h, From: Steve Lionel <Steve.Lionel@compaq.com>- Subject: Re: MAIL$SEND_ADD_ATTRIBUTE ("From"),8 Message-ID: <j8kudtsdb9r48p4h7pqdm4itl655se18b4@4ax.com>  , On Thu, 19 Apr 2001 16:12:17 -0400, JF Mezei% <jfmezei.spamnot@videotron.ca> wrote:f   >Steve Lionel wrote:> >> >I need to alter the "From:" field on the outbound message. >l. >>I know that I use this feature all the time. >s1 >Forging your bosses' signature ? :-) :-) :-) :-)t  C Hmm, interesting idea.  But no...  We have a Windows product called 5 Compaq Visual Fortran.  Support queries are sent to a A vf-support@compaq.com address - which ends up on an OpenVMS Alpha-B system.  A DCL procedure takes apart the mail message, decodes anyE attachments, and uses DEC NOTES to add a note with the user's message1F and the attachments are copied to a particular directory (available to< our PCs by Pathworks).  A Fortran program then mails back an9 auto-reply to the user with the return address made to be F vf-support@compaq.com - the subject line has a code that is recognized1 by the DCL procedure and tied to the note number.g  C To respond to a customer, the engineer adds a reply to the note andtA gives it the MAILIT keyword.  Another DCL procedure finds all theaF replies with MAILIT and sends it by e-mail back to the customer, again" on behalf of vf-support@compaq.com  E I suspect many of our Windows users would find it amusing that all of-/ the support processing is handled on OpenVMS!         - Steve Lionel (mailto:Steve.Lionel@compaq.com)C Fortran Engineerings& Compaq Computer Corporation, Nashua NH  6 Compaq Fortran web site: http://www.compaq.com/fortran   ------------------------------  % Date: Thu, 19 Apr 2001 15:03:35 -0500m+ From: "Main, Kerry" <Kerry.Main@compaq.com>n8 Subject: RE: OpenVMS article - please explain last line!R Message-ID: <DC4745D1A85CA04180C83CDC706A9D180D9538@cthexc02.americas.cpqcorp.net>  E >> My numbers were not false; they were the actual stream results.<<<r  C Regardless of where you got them, they do not reflect current Alpha 3 capabilities, so it is false to claim that they do.e  K >>> Just as I'm sure people understand the difference between the bandwidthe2 of a system, and the bandwidth of a processor. <<<  H Can you help me understand how having a fast CPU waiting for memory dataJ somehow will impress a user who is only interested in overall performance?% Do they care where the bottleneck is?m  L >>> It is correct that _current_ P4 systems (which also do 3.2 GB/s) do have= a lower peak bandwidth than the highest end Alpha systems.<<<   G Again, the current low end Alpha DS20 systems use a 5.2Gb/sec cross bar1D switch technology. No one thinks of the DS20 as a "highest end Alpha system".  I >>> However, the ServerWorks Grand Champion architecture will do 6.4 GB/soK with up to 4 P4 processors and has PCI-X. It blows the DS20 and ES40 out ofI the water to be sure. :-)<<e  J Sure, whenever its available, we will be glad to compare apples to apples.B Right now, comparing future designs to current HW is a bit silly.   E Anyway, here is a good report to consider when discussing P4 systems:e= http://www.inqst.com/articles/p4bandwidth/p4bandwidthmain.htms   :-)e   Regards,  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Services  Voice: 613-592-4660i Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----2 From: tsmurphy@addr.com [mailto:tsmurphy@addr.com] Sent: April 18, 2001 9:04 PM To: Info-VAX@Mvb.Saic.Coms8 Subject: Re: OpenVMS article - please explain last line!    4 Main, Kerry <Kerry.Main@compaq.com> wrote in messageL news:DC4745D1A85CA04180C83CDC706A9D180D9535@cthexc02.americas.cpqcorp.net...  @ > Yep, a real mystery how people can post such false numbers :-)  J My numbers were not false; they were the actual stream results. I got them$ directly off of the stream web site.  A > Current ES40 and DS20E systems use a 5.2Gb/sec cross bar switchh technology. L > I am sure folks here realize the difference between a cross bar switch and > standard bus technology.  L Just as I'm sure people understand the difference between the bandwidth of aK system, and the bandwidth of a processor. The bandwidth of a P4 _processor_BI is double the bandwidth of a Alpha _processor_. The P4 does 400 MHz x 64bdI (3.2 GB/s) and the EV6 does 200 MHz x 64b (only 1.6 GB/s). That is why P4t outperforms EV6 at UP stream.   J It is correct that _current_ P4 systems (which also do 3.2 GB/s) do have aG lower peak bandwidth than the highest end Alpha systems. Obviously, youpE would need to go above 2 CPU's to notice on the P4 end (and, for thatoC matter, you would need two EV6's to match the bandwidth of one P4).   J However, the ServerWorks Grand Champion architecture will do 6.4 GB/s withJ up to 4 P4 processors and has PCI-X. It blows the DS20 and ES40 out of the water to be sure. :-)h     Reference:*   http://biz.yahoo.com/bw/010227/0271.html  K > Also, keep in mind that the current x86 systems from ALL vendors that arenJ > greater than 1.x Ghz are primarily WS focussed ie. 2 cpus or less. Check out L > the server sections from all the major x86 vendors - try and find a server< > with greater than 2 cpu's that are each greater than 1Ghz.  H See the Serverworks link above. Four processors at a considerably higherD peak bandwidth than the DS20 and ES40. Not bad for an 8086, huh? :-)   ------------------------------    Date: 20 Apr 2001 02:03:42 +0800, From: Paul Repacholi <prep@prep.synonet.com>: Subject: Re: OpenVMS Job seeking in Hong Kong, SAR, China.- Message-ID: <87snj47on5.fsf@prep.synonet.com>n  + fabio_compaq@ep-bc.petrobras.com.br writes:    > What is BOFH ????p    ( Hum the Brasillion Operator from Hell...   -- -< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.a@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Thu, 19 Apr 2001 14:35:36 -0400r# From: Jim Agnew <Agnew@hsc.vcu.edu> : Subject: Re: OpenVMS Job seeking in Hong Kong, SAR, China.+ Message-ID: <3ADF2FF8.2617ACC9@hsc.vcu.edu>n   HEHEHEHEHEHAAAA    ...e   Paul Repacholi wrote:r > - > fabio_compaq@ep-bc.petrobras.com.br writes:e >  > > What is BOFH ????. > * > Hum the Brasillion Operator from Hell... >  > --> > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda.hB >                                              West Australia 60760 > Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Thu, 19 Apr 2001 15:55:34 -0300c) From: fabio_compaq@ep-bc.petrobras.com.brL: Subject: Re: OpenVMS Job seeking in Hong Kong, SAR, China.L Message-ID: <OF736DE199.3EA91960-ON03256A33.0067C97C@ep-bc.petrobras.com.br>   BOFH ?  K Hmm.. Brasilian Operator from the Hell ... Every System Manager is a Luxuryn Operator ! :-)  E Nice .. because in this Information Technology industry nowadays am I1 feeling in the Hell ....   Regardsa   FC        4 Jim Agnew <Agnew@hsc.vcu.edu> em 19/04/2001 10:32:13  / Favor responder a Jim Agnew <Agnew@hsc.vcu.edu>-             Info-VAX@Mvb.Saic.Come      : Assunto: Re: OpenVMS Job seeking in Hong Kong, SAR, China.    H here's a better one... i've GOT to get back to work.. whatever that is..    http://bofh.ntk.net/Bastard.html   Roy Omond wrote: > , > fabio_compaq@ep-bc.petrobras.com.br wrote: >' > > What is BOFH ????c > ? > Bastard Operator From Hell (see http://www.theregister.co.uk)  >  > Roy Omond  > Blue Bubble Ltd.   ------------------------------   Date: 19 Apr 2001 18:04:58 GMT# From: system@niuhep.physics.niu.edu & Subject: Re: software ethernet sniffer+ Message-ID: <9bn9ca$bve$1@husk.cso.niu.edu>o  Q In article <9bmq96$61o$1@husk.cso.niu.edu>, system@niuhep.physics.niu.edu writes:  >Hello, < >   I am in dire need of a ethernet sniffer for either a vaxG >or an alpha.  I had one but apparently it got lost somewhere along then >line.  < My thanks to all of the many folks who responded to my plea.   Robert   ------------------------------  % Date: Thu, 19 Apr 2001 11:26:05 -0700 2 From: "Randy Park" <rjpark@mindspring.com.nospaam> Subject: Spammed by I-S-E today02 Message-ID: <9bnakv$k9r$1@slb0.atl.mindspring.net>  9 Today I received a massive spam e-mail from I-S-E.  As an 9 occasional poster to comp.os.vms they obviously harvested > my munged e-mail address from this newsgroup and intelligently. removed the '.nospaam' from my e-mail address.   ------------------------------  % Date: Thu, 19 Apr 2001 16:10:19 -0400a7 From: David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com> # Subject: Re: Spammed by I-S-E today 2 Message-ID: <zUXfOk2uKAIQ6J1BHXy0YcCVP=oB@4ax.com>  2 I was Spammed by them as well.  It would take some+ intelligence on their list-bot to get mine.t   David R. Beattyo  0 On Thu, 19 Apr 2001 11:26:05 -0700, "Randy Park"& <rjpark@mindspring.com.nospaam> wrote:  : >Today I received a massive spam e-mail from I-S-E.  As an: >occasional poster to comp.os.vms they obviously harvested? >my munged e-mail address from this newsgroup and intelligentlyd/ >removed the '.nospaam' from my e-mail address.o >n >p >e   ------------------------------  % Date: Thu, 19 Apr 2001 16:44:33 -0400n# From: Jim Agnew <Agnew@hsc.vcu.edu>r# Subject: Re: Spammed by I-S-E todayi+ Message-ID: <3ADF4E31.8151C11A@hsc.vcu.edu>t  a not really.. looking at your email, the uppercase/lowercase difference is quite, er, LOUD!... ;-)i   j.   David Beatty wrote:  > 4 > I was Spammed by them as well.  It would take some- > intelligence on their list-bot to get mine.u >  > David R. Beattys > 2 > On Thu, 19 Apr 2001 11:26:05 -0700, "Randy Park"( > <rjpark@mindspring.com.nospaam> wrote: > < > >Today I received a massive spam e-mail from I-S-E.  As an< > >occasional poster to comp.os.vms they obviously harvestedA > >my munged e-mail address from this newsgroup and intelligentlyx1 > >removed the '.nospaam' from my e-mail address.. > >- > >- > >-   ------------------------------  # Date: Thu, 19 Apr 2001 21:39:34 GMTJ7 From: moroney@world.std.spaamtrap.com (Michael Moroney)g# Subject: Re: Spammed by I-S-E todayg& Message-ID: <GC26ty.Jzp@world.std.com>  4 "Randy Park" <rjpark@mindspring.com.nospaam> writes:  : >Today I received a massive spam e-mail from I-S-E.  As an: >occasional poster to comp.os.vms they obviously harvested? >my munged e-mail address from this newsgroup and intelligently / >removed the '.nospaam' from my e-mail address.p  E They spammed me as well, and their spambot was smart enough to removeh( the 'spaamtrap' from my address as well.  F If they only spammed comp.os.vms posters maybe they demunged addresses by hand?   -Mikep   ------------------------------  % Date: Thu, 19 Apr 2001 20:30:49 -0400u- From: JF Mezei <jfmezei.spamnot@videotron.ca>-# Subject: Re: Spammed by I-S-E today , Message-ID: <3ADF8336.17074CCF@videotron.ca>   Michael Moroney wrote:G > They spammed me as well, and their spambot was smart enough to removeu* > the 'spaamtrap' from my address as well.  L There are quite a few posters who manually reply to some posts via email andK the distribution list, and they manually remove the spam counter measures. r  < As a result, their reply makes it to the newsgroup with your0 manually-corrected email address in the headers.  L It only takes one person doing this to you for your name to become available to the spammers.   ------------------------------  % Date: Thu, 19 Apr 2001 21:21:10 +0100t, From: "Richard Maher" <maher_rj@hotmail.c*m>7 Subject: TCP/IP io$_acpcontrol inetacp$c_hostent_offset 3 Message-ID: <9bnha5$q5s$1@plutonium.btinternet.com>    Hi,T  " TCP/IP Services for VMS V4.2 ECO 3 Alpha VMS 7.1-2.  G Command = inetacp$c_hostent_offset * 256 + inetacp_func$c_gethostbynameT  I Is it just me or does this call only look at the ucx$host file and ignore  any name service?2  I The name servers are doing their thing as inetacp$c_trans will testify toT) but hostent_offset returns ss$_endoffile.D   While I'm here:-  L Does someone have a non-C record definition describing the hostent structureJ returned from this call? There are a couple of longwords that I'm not sure about.  B How big should the output parameter string (P4?) be to gurantee noJ ss$_badparam? I recall that bookreader had examples of 256 and 512 what is that based on?  2 Why aren't dynamic string descriptors checked for?   Regards Richard Maherf   ------------------------------  # Date: Thu, 19 Apr 2001 18:57:29 GMTr, From: peterw@u.genie.co.uk (Peter Watkinson) Subject: TFT and VMS< Message-ID: <3adf3451.2281945@newshost.netscapeonline.co.uk>   Hi,l  F I'm thinking of buying a Flatscreen monitor to use with my systems. IsF there any problem with using it with VMS? For that matter is there anyC problem with using any monitor with VMS I haven't got my vms system  either yet!r  	  regards,      Peter WatkinsonL Email: peterw@u.genie.co.uko( Internet: http://you.genie.co.uk/peterw/A Windsurf International.com http://www.windsurf-international.com/o* PW Navigate.com http://www.pwnavigate.com/   ------------------------------  % Date: Thu, 19 Apr 2001 16:40:17 -0300r) From: fabio_compaq@ep-bc.petrobras.com.brt Subject: Re: TFT and VMSL Message-ID: <OF462B5584.CD89AEA0-ON03256A33.006BD977@ep-bc.petrobras.com.br>  ; You can use a flat monitor  if it has a SVGA connection.... K In case of using an old VAX / Alpha with RGB connectors you can test to use- a conversion kit from RGB to SVGA ....X   Regardse   FC        = peterw@u.genie.co.uk (Peter Watkinson) em 19/04/2001 15:57:29F  8 Favor responder a peterw@u.genie.co.uk (Peter Watkinson)             Info-VAX@Mvb.Saic.Come       Assunto: TFT and VMS         Hi,b  F I'm thinking of buying a Flatscreen monitor to use with my systems. IsF there any problem with using it with VMS? For that matter is there anyC problem with using any monitor with VMS I haven't got my vms system  either yet!i  	  regards,      Peter Watkinson. Email: peterw@u.genie.co.uk ( Internet: http://you.genie.co.uk/peterw/A Windsurf International.com http://www.windsurf-international.com/n* PW Navigate.com http://www.pwnavigate.com/   ------------------------------  % Date: Thu, 19 Apr 2001 20:50:17 -0400 2 From: "John Gemignani, Jr." <john@ossc.DELETE.net> Subject: Re: TFT and VMS+ Message-ID: <3adf88a6$1@newsfeed.vitts.com>c  K I have a Viewsonic VP151 (15" with lotsa options, including S-Video input). K The video card is a ZLXP-L2, with Open3D.  The problem is the resolution is G 1240x1024 (I think) and the refresh is too high for it, so II use it ont other systems.  I Some video cards will allow you to change these parameters, some may not. ) So be sure that your panel can handle it.u   -John:  9 "Peter Watkinson" <peterw@u.genie.co.uk> wrote in message96 news:3adf3451.2281945@newshost.netscapeonline.co.uk... >  >  > Hi,e >iH > I'm thinking of buying a Flatscreen monitor to use with my systems. IsH > there any problem with using it with VMS? For that matter is there anyE > problem with using any monitor with VMS I haven't got my vms systemn
 > either yet!e >g >  regards,y >t >d > Peter Watkinsoni > Email: peterw@u.genie.co.uk:* > Internet: http://you.genie.co.uk/peterw/C > Windsurf International.com http://www.windsurf-international.com/e, > PW Navigate.com http://www.pwnavigate.com/ >i >t   ------------------------------  % Date: Thu, 19 Apr 2001 21:42:14 -0700 + From: "Don Whitlow" <dwhitlow@telocity.com> Y Subject: This is a fun one ... high interrupt rate after VMS upgrade to 7.2-1 ... but waie; Message-ID: <IrND6.662$0d.177992@newsrump.sjc.telocity.net>h   Hi All,P  J I've got a good one for everyone. I just upgraded my 4 member SCSI clusterG to VMS 7.2-1 from 7.1-2. Cluster consists of 2 Alphaserver 4100's and 21J Alphaserver 2100's. All boot from the same mirrored system disk located inI an ESA10000 shared by all 4 systems via DWZZH-05 SCSI hub. Currently onlynJ running a single SCSI path. (Was upgrading to 7.2 for better multiple path( SCSI options, but that's another story.)  B So, here's the problem. After the upgrade, the 2 AS2100s display aK ridiculously high interrupt rate on CPU 00. In fact, the percentage of time0I that cpu spends processing interrupts after system boot up jumps right toAJ 30% and eventually will climb to 99-100% and never come back down. NothingI on the system is running away, MP mode isn't especially high on that cpu,oI and there don't seem to be any hardware errors that I can see. Here's theiK kicker. I saved my old 7-1.2 system drive and rebooted back onto that disk.mK THE PROBLEM IS STILL THERE! The pre-7.2 system drive had no changes made togH it. Nothing changed on the hardware side. Firmware was not upgraded. I'm
 going mad! :)r  I The two 4100's do not display this behavior, and run like champs, on bothMK 7.1-2 and 7.2-1. Does anybody have any suggestions on what to look at next? K We have a call into Compaq Gold support, but they're pretty much as lost ashL we are on this one. I know this sounds a lot like a hardware problem at thisI point, but nothing is logging any errors. Besides, it's happening on bothtK 2100 boxes. I also thought maybe it was the shared SCSI bus for the ESA10K,s2 but wouldn't the 4100's see the problem also then?  J Any thoughts, ideas, or sacrifices on our behalf to the OpenVMS gods would be much appreciated. ;)e   Thanks in advance, Donn   ------------------------------  % Date: Fri, 20 Apr 2001 08:29:56 +0300a& From: Guy Peleg <guy.peleg@compaq.com>Y Subject: Re: This is a fun one ... high interrupt rate after VMS upgrade to 7.2-1 ... butt* Message-ID: <3ADFC954.8C0CBF03@compaq.com>   DonV  - It does sound like hardware to me as well....oK It is known that a bad Ethernet card (located anywhere on the network) thato sends trashy packets' might cause high interrupt rate on VMS.-K Try booting each 2100 separate and see if problem persist. Then try booting, minimum (STARTUP_P1="min")J and see what happens (might want to disconnect machines from the network).  M If you want to reach "real" conclusion try and activate SYSTEM_PC option fromu DECPS. It willP collect info about time spent in interrupt state. This info could be analyzed by your local compaq supportnO and analyze in which routine VMS is spending its time and most important - WHY!    Hope that help,a  	 Guy Pelege OpenVMS Engineeringi   Don Whitlow wrote:  	 > Hi All,h > L > I've got a good one for everyone. I just upgraded my 4 member SCSI clusterI > to VMS 7.2-1 from 7.1-2. Cluster consists of 2 Alphaserver 4100's and 2lL > Alphaserver 2100's. All boot from the same mirrored system disk located inK > an ESA10000 shared by all 4 systems via DWZZH-05 SCSI hub. Currently onlydL > running a single SCSI path. (Was upgrading to 7.2 for better multiple path* > SCSI options, but that's another story.) >hD > So, here's the problem. After the upgrade, the 2 AS2100s display aM > ridiculously high interrupt rate on CPU 00. In fact, the percentage of timeoK > that cpu spends processing interrupts after system boot up jumps right toIL > 30% and eventually will climb to 99-100% and never come back down. NothingK > on the system is running away, MP mode isn't especially high on that cpu,tK > and there don't seem to be any hardware errors that I can see. Here's thesM > kicker. I saved my old 7-1.2 system drive and rebooted back onto that disk.sM > THE PROBLEM IS STILL THERE! The pre-7.2 system drive had no changes made tofJ > it. Nothing changed on the hardware side. Firmware was not upgraded. I'm > going mad! :)  > K > The two 4100's do not display this behavior, and run like champs, on both]M > 7.1-2 and 7.2-1. Does anybody have any suggestions on what to look at next?SM > We have a call into Compaq Gold support, but they're pretty much as lost asrN > we are on this one. I know this sounds a lot like a hardware problem at thisK > point, but nothing is logging any errors. Besides, it's happening on both:M > 2100 boxes. I also thought maybe it was the shared SCSI bus for the ESA10K,m4 > but wouldn't the 4100's see the problem also then? >aL > Any thoughts, ideas, or sacrifices on our behalf to the OpenVMS gods would > be much appreciated. ;)e >c > Thanks in advance, > Dona   ------------------------------  # Date: Fri, 20 Apr 2001 05:48:12 GMTd From: Dirk Munk <munk@home.nl>Y Subject: Re: This is a fun one ... high interrupt rate after VMS upgrade to 7.2-1 ... butl' Message-ID: <3ADFCD9B.48855D0F@home.nl>h   Just a few questions:   L Did you check the firmware version on all of your systems (upgrades may also& upgrade firmware for SCSI controllers)> did you roll in all patches for 7.2-1 that can be applicable ?/ How about the software of the HSZ controllers ?      Don Whitlow wrote:  	 > Hi All,t >oL > I've got a good one for everyone. I just upgraded my 4 member SCSI clusterI > to VMS 7.2-1 from 7.1-2. Cluster consists of 2 Alphaserver 4100's and 2GL > Alphaserver 2100's. All boot from the same mirrored system disk located inK > an ESA10000 shared by all 4 systems via DWZZH-05 SCSI hub. Currently onlyeL > running a single SCSI path. (Was upgrading to 7.2 for better multiple path* > SCSI options, but that's another story.) >aD > So, here's the problem. After the upgrade, the 2 AS2100s display aM > ridiculously high interrupt rate on CPU 00. In fact, the percentage of timeaK > that cpu spends processing interrupts after system boot up jumps right to.L > 30% and eventually will climb to 99-100% and never come back down. NothingK > on the system is running away, MP mode isn't especially high on that cpu,-K > and there don't seem to be any hardware errors that I can see. Here's thecM > kicker. I saved my old 7-1.2 system drive and rebooted back onto that disk.RM > THE PROBLEM IS STILL THERE! The pre-7.2 system drive had no changes made topJ > it. Nothing changed on the hardware side. Firmware was not upgraded. I'm > going mad! :). > K > The two 4100's do not display this behavior, and run like champs, on both9M > 7.1-2 and 7.2-1. Does anybody have any suggestions on what to look at next? M > We have a call into Compaq Gold support, but they're pretty much as lost as.N > we are on this one. I know this sounds a lot like a hardware problem at thisK > point, but nothing is logging any errors. Besides, it's happening on bothKM > 2100 boxes. I also thought maybe it was the shared SCSI bus for the ESA10K,A4 > but wouldn't the 4100's see the problem also then? >eL > Any thoughts, ideas, or sacrifices on our behalf to the OpenVMS gods would > be much appreciated. ;). >  > Thanks in advance, > Don    ------------------------------  # Date: Thu, 19 Apr 2001 18:14:39 GMT  From: Dirk Munk <munk@home.nl>5 Subject: Re: Using 3-phase power converters on VAXen?i' Message-ID: <3ADF2B0B.8971290E@home.nl>a   Larry Kilgallen wrote:  k > In article <s+3pOvgtD$rM@eisner.encompasserve.org>, malmberg@encompasserve.org (John E. Malmberg) writes:u >cL > > It is my understanding that the standard U.S. household feed is a singleI > > phase nominal 220 to 240 Volts.  It is center tapped to ground at thesK > > local distribution transformer, allowing two 110 to 120 volt legs to be!" > > distributed through the house. > >oJ > > The maximum potential between a single wire and ground is about 120 V. > > K > > Beyond the transformer may vary by region, but since in the rural areaseK > > where I can see the wires, only two lines are going to the distribution L > > transformers on most residential streets.  This leads me to believe thatI > > only a single phase is shared by that neighborhood from some upstreamt > > feed point.l > >oH > > Getting 3 phases to a house can mean requiring three wires from thatL > > upstream feed point.  Sharing the existing phase could present potentialK > > balancing problems for the distribution point, as the expectation would-? > > be that you have a significant load to justify the expense.c > >. > > H > > Just following this thread has shown how much confusion there is out > > there on electrical wiring.6 > >hE > > And do not take my word on how things are wired.  Look it up from  > > an authorative source. > C > Your 120/240 description is correct, and you wisely avoided DeltagD > vs. Wye descriptions for three-phase.  But, I did look up a sourceC > authoritative for my neighborhood, the pole out front :-).  On my-A > street there is a three phase (four wire) 13,800 volt main feed<C > across the top of the poles (I thought 13,900 volts, but a repairsC > crew corrected my assumption one day.  Every block or so there ismE > a transformer (pole pig) taking one of those phases to reduce it tos? > the 120/240 single phase (sometimes called "two phase" at 240wB > volts) feed for the next group of houses.  When I was growing upC > the 120/240 volt feed was always three wires one above each other0C > from pole to pole, while the three phase wires were separate from 7 > each other between crossbars on the successive poles.2 >I@ > These days, in Massachusetts anyway, replacement 120/240 feeds= > tend to have three wires twisted together while replacement < > three phase 13KV feeds tend to be separated on the corners: > of rectangular insulators about 8 inches diagonally.  My9 > assumption is that this change is to resist damage froma@ > ice storms, so your experience in Florida may be different :-)  m This message shows an important difference between the US and Europe. In most of Western Europe we don't have k power distribution-on-poles, at least not to houses.. The power cables are usually dug in the ground, hencerm the transformers can be bigger and fewer (and 3 phase). In the US power distribution cables hang freely along k the streets, at least in more rural areas. During a short visit to the US I went to Key West. It is amazing f to see how the beautiful look of a very nice town can be destroyed by these power and telephone cables dangling along the streets.n   ------------------------------    Date: 20 Apr 2001 02:12:02 +0800, From: Paul Repacholi <prep@prep.synonet.com>5 Subject: Re: Using 3-phase power converters on VAXen?o- Message-ID: <87k84g7o99.fsf@prep.synonet.com>   # Shane.F.Smith@Healthnet.com writes:s  B > In England or out here in the Colonies, I've never seen one that > used 3 phase.-   What colony is that?  B Here near every one is 3 phase. A 'single phase' stove is one thatE needs only a single 440V feed. One Delta leg. There is a move to call-? them two phase... (yes, I know it's not really 440, but 240 was"+ 256. That was a WA oddity as far as I know)T   Small one are 240V stoves.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.s@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Thu, 19 Apr 2001 12:32:15 -0600.% From: Dan O'Reilly <dano@process.com>s5 Subject: Re: Using 3-phase power converters on VAXen? A Message-ID: <5.0.2.1.2.20010419123132.00a72b40@ntbsod.psccos.com>   , At 12:12 PM 4/19/2001, Paul Repacholi wrote:$ >Shane.F.Smith@Healthnet.com writes: >aD > > In England or out here in the Colonies, I've never seen one that > > used 3 phase.d >  >What colony is that?b >w  >Here near every one is 3 phase.  8 ..so, what you're saying is, you never forget a phase...                 sorry, couldn't resist...v     ------I +-------------------------------+---------------------------------------+.I | Dan O'Reilly                  |                                       |yI | Principal Engineer            |  "Why should I care about posterity?  |tI | Process Software              |   What's posterity ever done for me?" |-I | http://www.process.com        |                    -- Groucho Marx    |aI +-------------------------------+---------------------------------------+h   ------------------------------  # Date: Fri, 20 Apr 2001 05:00:24 GMTi+ From: rjordan@mars.mcs.net (Richard Jordan)h Subject: Re: VLCsm4 Message-ID: <InPD6.9016$JN5.222623@news.goodnet.com>  H VAXstation 4000 VLC models cannot reliably handle the higher heat output8 of the 7200 RPM and faster drives; the low profile ones H will physically fit and probably run (assuming power draw) but there is E not sufficient ventilation to remove the heat; drive runs too hot andoI you get reduced life.  I've heard from a few folks who had other problemscG when running hot fast drives (power supply toasting, system reliability 
 issues, etc).u  G I'm pretty sure the specs for the VLC only had 5400RPM drives listed askF compatible.  If you can find a 2 or 4GB 5400RPM drive you're probably 4 fine using it internally (low profile, of course).     Rich Jordan   rjordan@mcs.net   ------------------------------  % Date: Thu, 19 Apr 2001 14:07:49 -0400c% From: "Brian Tillman" <tillman_brian>d9 Subject: Re: [BUG?] FREEWARE CD MENU: license expired :-(.$ Message-ID: <3adf29d0$1@news.si.com>  = >The contents of a newer version of the CD than yout have areb >available via the web at: > 2 >http://www.openvms.compaq.com/freeware/index.html   The requested URL was not founds HTTP status code: 404  -- rA Brian Tillman                   Internet: tillman_brian at si.com A Smiths Aerospace                          tillman at swdev.si.comw= 3290 Patterson Ave. SE, MS      Addresses modified to preventv< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Thu, 19 Apr 2001 16:21:49 -04002) From: "Kevin" <IHateMSOutlook@compaq.com>o; Subject: Re: [BUG] wrong APACHE file location after installn3 Message-ID: <nNHD6.1057$fB6.26237@news.cpqcorp.net>   F It appears that APACHE$CONFIG cannot determine where you installed the	 software. ( Please post some additional information:L     1. Where did you install the software.  In particular, did you include a /DESTINATION6         qualifier in the PRODUCT INSTALLATION command?0     2. Please post the configuration data files:,             $ type sys$startup:apache$*.dat;   Thanks.   . "Didier Morandi" <DMo@ims.ch> wrote in message  news:3ADF0AC5.AF92DF01@ims.ch...+ > I'm trying to install APACHE for OpenVMS:r >t > $ set def sys$managere > $ dir apa* >i > Directory SYS$COMMON:[SYSMGR]t >t > APACHE$CONFIG.COM;1t >/ > Total of 1 file. > $ @apache$config >d< >                 Compaq HTTP Server Configuration Procedure >o@ >         This procedure helps you define the parameters and theF >         operating environment required to run the Compaq HTTP Server% >         for OpenVMS on this system.a; > ERROR: Required OpenVMS APACHE$WWW account does not existh, > ERROR: Missing required command procedure:% SYS$SYSROOT:[SYSMGR]APACHE$ADDUSER.CO  > Mp > %SYSTEM-F-ABORT, abort >  >0B > Actually, the APACHE$ADDUSER.COM is not in sys$manager, it is in# > SYS$SYSDEVICE:[VMS$COMMON.APACHE]w >  > (also sent to nobody)  >a > D.   ------------------------------  % Date: Fri, 20 Apr 2001 07:28:06 +0200P! From: Didier Morandi <DMo@ims.ch>u; Subject: Re: [BUG] wrong APACHE file location after installo& Message-ID: <3ADFC8E6.EA0C370F@ims.ch>  L I am an ol'timmer, I answered <RETURN> to all installation questions, so theN installation took place in [vms$common]. The doc says p2.3 of the InstallationO and Configuration Guide v1.3.12 OpenVMS Alpha, 2: "Make sure you are logged ong+ as a privileged user (for example SYSTEM)".p   I was.D p 2.5 it says "To configure the server, enter the following command:   @sys$manager:apache$config"u  P As the apache$config.com is in the installed directory instead of sys$manager, IL imagine that the .PCSI$DESC file doesn't "file" the file in the right place.  ) Please post the PCSI description file :-):   D. s  ' SYS$SYSROOT:[SYSMGR]APACHE$CONFIG.DAT;1S  g ##, ##  WARNING: Do not edit this file yourself.7 ##  Configuration data file is built and maintained by e* ##  SYS$COMMON:[SYSMGR]APACHE$CONFIG.COM;1 ### ##  Written 19-APR-2001 18:32:26.22f ## VERSION                 1.0A+ PATH_PCSI               SYS$COMMON:[APACHE] + PATH_COMMON             SYS$COMMON:[APACHE]o# PATH_HTDOCS             DATA2:[WEB] < PATH_SPECIFIC           SYS$COMMON:[APACHE.SPECIFIC.IMS0202] TCPIP_SERVICE           NO     Kevin wrote: > H > It appears that APACHE$CONFIG cannot determine where you installed the > software.h* > Please post some additional information:N >     1. Where did you install the software.  In particular, did you include a > /DESTINATION8 >         qualifier in the PRODUCT INSTALLATION command?2 >     2. Please post the configuration data files:. >             $ type sys$startup:apache$*.dat; >  >i   ------------------------------  % Date: Thu, 19 Apr 2001 19:55:50 +0200 ! From: Didier Morandi <DMo@ims.ch> , Subject: [Q] default APACHE directory change& Message-ID: <3ADF26A6.6CF5349C@ims.ch>  N The default location for the index.html file is SYS$COMMON:[APACHE.HTDOCS] and the file is index.html_en.  O I used the apache$config procedure to change the default location to my OpenVMShO doc directory located at DATA2:[WEB], where the index.html and [.721FINAL] (andy others) are located.  * The apache$config.dat now looks like this:   IMS0202> ty APACHE$CONFIG.DAT;1h ##, ##  WARNING: Do not edit this file yourself.7 ##  Configuration data file is built and maintained by i* ##  SYS$COMMON:[SYSMGR]APACHE$CONFIG.COM;1 ### ##  Written 19-APR-2001 18:32:26.22t ## VERSION                 1.0 + PATH_PCSI               SYS$COMMON:[APACHE] + PATH_COMMON             SYS$COMMON:[APACHE],# PATH_HTDOCS             DATA2:[WEB]p< PATH_SPECIFIC           SYS$COMMON:[APACHE.SPECIFIC.IMS0202] TCPIP_SERVICE           NO  I The problem is that, after having restarted APACHE, it still displays theyD default installation page, even after a reload, when I try to access http://ims0202.IMS.CH.   Any idea, anyone?s Thanks,3   D. . (RTFM did not help)    ------------------------------  % Date: Thu, 19 Apr 2001 11:43:41 -0700a0 From: Mark Berryman <Mark.Berryman@Mvb.Saic.Com>0 Subject: Re: [Q] default APACHE directory change, Message-ID: <3ADECF6D.44EA7D71@Mvb.Saic.Com>   Didier Morandi wrote:e > P > The default location for the index.html file is SYS$COMMON:[APACHE.HTDOCS] and > the file is index.html_en. > Q > I used the apache$config procedure to change the default location to my OpenVMSpQ > doc directory located at DATA2:[WEB], where the index.html and [.721FINAL] (andi > others) are located. > , > The apache$config.dat now looks like this: > ! > IMS0202> ty APACHE$CONFIG.DAT;1o > ##. > ##  WARNING: Do not edit this file yourself.8 > ##  Configuration data file is built and maintained by, > ##  SYS$COMMON:[SYSMGR]APACHE$CONFIG.COM;1 > ##% > ##  Written 19-APR-2001 18:32:26.22S > ## > VERSION                 1.0f- > PATH_PCSI               SYS$COMMON:[APACHE]0- > PATH_COMMON             SYS$COMMON:[APACHE] % > PATH_HTDOCS             DATA2:[WEB] > > PATH_SPECIFIC           SYS$COMMON:[APACHE.SPECIFIC.IMS0202] > TCPIP_SERVICE           NO > K > The problem is that, after having restarted APACHE, it still displays the F > default installation page, even after a reload, when I try to access > http://ims0202.IMS.CH. >  > Any idea, anyone?,	 > Thanks,t >  > D. > (RTFM did not help)r  E You edited the wrong file (and there is a reason this file says don'tu; edit it yourself).  The file you want to edit is located at @ APACHE$ROOT:[CONF]HTTPD.CONF; look for the DocumentRoot keyword.  
 Mark Berryman    ------------------------------  % Date: Thu, 19 Apr 2001 14:51:08 -0400i7 From: David Beatty <David.Beatty@sasSPAMITUPTHEZZZ.com> 0 Subject: Re: [Q] default APACHE directory change2 Message-ID: <DDPfOuNalWQr3bAl9AozDzA4FKeL@4ax.com>  A The file you need to change is APACHE$ROOT:[CONF]HTTPD.CONF.  ThemD entry is DocumentRoot.  There is some directory security in there as6 well.  Make sure you read through the documentation at http://www.apache.org/.o   David R. Beattya  F On Thu, 19 Apr 2001 19:55:50 +0200, Didier Morandi <DMo@ims.ch> wrote:  O >The default location for the index.html file is SYS$COMMON:[APACHE.HTDOCS] andl >the file is index.html_en.i >iP >I used the apache$config procedure to change the default location to my OpenVMSP >doc directory located at DATA2:[WEB], where the index.html and [.721FINAL] (and >others) are located.? >a+ >The apache$config.dat now looks like this:  >a  >IMS0202> ty APACHE$CONFIG.DAT;1 >## - >##  WARNING: Do not edit this file yourself.o8 >##  Configuration data file is built and maintained by + >##  SYS$COMMON:[SYSMGR]APACHE$CONFIG.COM;1e >## $ >##  Written 19-APR-2001 18:32:26.22 >##e >VERSION                 1.0, >PATH_PCSI               SYS$COMMON:[APACHE], >PATH_COMMON             SYS$COMMON:[APACHE]$ >PATH_HTDOCS             DATA2:[WEB]= >PATH_SPECIFIC           SYS$COMMON:[APACHE.SPECIFIC.IMS0202]o >TCPIP_SERVICE           NOl >aJ >The problem is that, after having restarted APACHE, it still displays theE >default installation page, even after a reload, when I try to accessg >http://ims0202.IMS.CH.u >u >Any idea, anyone? >Thanks, >  >D.  >(RTFM did not help)   ------------------------------  % Date: Fri, 20 Apr 2001 07:30:26 +0200o! From: Didier Morandi <DMo@ims.ch>s0 Subject: Re: [Q] default APACHE directory change& Message-ID: <3ADFC972.EF00CF1A@ims.ch>  P Mark, I wrote "I used the apache$config procedure to change the default location4 to my OpenVMS doc directory located at DATA2:[WEB]".   Anyway, thank you for the info.t   D.   Mark Berryman wrote: > G > You edited the wrong file (and there is a reason this file says don'th= > edit it yourself).  The file you want to edit is located atrB > APACHE$ROOT:[CONF]HTTPD.CONF; look for the DocumentRoot keyword.   ------------------------------  % Date: Thu, 19 Apr 2001 11:03:54 -0700l! From: Don Sykes <don@alphase.com>o$ Subject: Re: [Q] Free VMS WEB Server+ Message-ID: <3ADF288A.829D4FFC@alphase.com>y  , This is a multi-part message in MIME format.& --------------D3DF7C5EF6FC3B5DF9F33DC8) Content-Type: text/html; charset=us-asciia Content-Transfer-Encoding: 7bitt  > <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html>H I've been using Netscape's FastTrack on VMS 7.2 w/ TCP 5.0, but it seemsG to have some serious problems: a) VMS dies every few weeks and&nbsp; b) I the WebServer's I/O &amp; CPU rates are astronomical. The following (showa@ sys) is after 10 days of up time (and I get &lt; 100 hits a day)` <br><font size=-1>0000009E WEBSERVER_1&nbsp;&nbsp;&nbsp;&nbsp; HIB&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;S 3 10112792&nbsp;&nbsp; 0 03:04:09.81&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 3266&nbsp;&nbsp;  1086 MS</font>H <br>That's > 10 million i/o's &amp; > 3 hrs CPU. It's also almost alwaysE in COM and cranking ~ 100 i/o's per min when no one is accessing it !.I <br>Bottom line is, I don't recommend it and I guess I'd like to hear howe Apache works out for you.  <br>Donm <p>Didier Morandi wrote:H <blockquote TYPE=CITE>Where can I find a free WEB server for OpenVMS any version, please? I found aH <br>DECthread-based product but the URL for the distrib location goes to nl: 
 <p>Thanks, <p>D.rJ <p>ps: Thank you, Hoff, for having taught me that I can do CGI with DCL... :-)</blockquote> </html>d  & --------------D3DF7C5EF6FC3B5DF9F33DC8- Content-Type: text/x-vcard; charset=us-ascii;   name="don.vcf"u Content-Transfer-Encoding: 7bit0' Content-Description: Card for Don Sykes   Content-Disposition: attachment;  filename="don.vcf"w   begin:vcard  n:Sykes;Doni$ tel;cell:Available to customers only tel;fax:415-485-6895 tel;work:415-457-8532h x-mozilla-html:TRUEh url:www.alphase.come org:Alpha Software Express, LLCe8 adr:;;1380 Lincoln Ave - Suite 5;San Rafael;CA;94901;USA version:2.1d email;internet:don@alphase.com! title:Principal Software Engineerg note;quoted-printable:Website:  http://alphase.com=0D=0AResume:  http://alphase.com/DonsCV.html=0D=0AAlternate Email Address:  alphase@pacbell.net x-mozilla-cpt:;-19520i fn:Don Sykes	 end:vcardt  ( --------------D3DF7C5EF6FC3B5DF9F33DC8--   ------------------------------  # Date: Thu, 19 Apr 2001 23:01:47 GMTa' From: "Mr. Generic" <someone@microsoft> ' Subject: [Q] Making Perl work with CSWSo@ Message-ID: <v7KD6.67$hH1.5137@newsread3.news.pas.earthlink.net>  L I have been trying to get mod_perl to work with CSWS (Apache 1.3.14) withoutJ much luck. I believe my sample code works (below), as I have tried it on aH Tru64 server and a Solaris box with success. It runs from DCL (below). II believe the problem may lie in the ShBang ("#!") line, but I've not foundr5 the magic path. I've included the PERL logical below.   E This OS is OVMS V7.2-1 for Alpha. Any help would be much appreciated.,  
 Dave Lloyd   Bechtel SAIC  
 Las Vegas, NVs  < ------------------------------------------------------------   YMA211> type sample.pl   #!/perl    # Executed with:  % # http://localhost/cgi-bin/sample.cgid   use CGI;  
 $q = new CGI;)   print $q->header();p  2 print $q->start_html(-title=>'Sample CGI Script');  & print $q->h1("Environment Variables");   print $q->p(),"Remote Host: ";   print $q->remote_host();   print $q->p(),"Remote IP: ";   print $q->remote_addr();   print $q->end_html();s   exit(0);   YMA211>s   YMA211> perl sample.pl  8 (offline mode: enter name=value pairs on standard input)   Exit   Content-Type: text/htmle  . <!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN">  , <HTML><HEAD><TITLE>Sample CGI Script</TITLE>  : </HEAD><BODY><H1>Environment Variables</H1><P>Remote Host: localhost<P>Remote IPx   : 127.0.0.1</BODY></HTML>f   YMA211>    YMA211> sho symb/glob perl  % PERL == "$PERL_ROOT:[000000]PERL.EXE"    YMA211>W   YMA211> sho log PERL_ROOT   : "PERL_ROOT" = "YMA211$DKA0:[SYS0.SYSCOMMON.PERL5_005_03.]" (LNM$SYSTEM_TABLE)   YMA211>    ------------------------------  % Date: Thu, 19 Apr 2001 23:28:13 -0500tC From: "Craig A. Berry" <craig.berry@nospam.SignalTreeSolutions.com>e+ Subject: Re: [Q] Making Perl work with CSWS I Message-ID: <craig.berry-84612F.23281319042001@newsrump.sjc.telocity.net>   @ In article <v7KD6.67$hH1.5137@newsread3.news.pas.earthlink.net>,)  "Mr. Generic" <someone@microsoft> wrote:-  N > I have been trying to get mod_perl to work with CSWS (Apache 1.3.14) withoutL > much luck. I believe my sample code works (below), as I have tried it on aJ > Tru64 server and a Solaris box with success. It runs from DCL (below). IK > believe the problem may lie in the ShBang ("#!") line, but I've not foundn7 > the magic path. I've included the PERL logical below.n  C What errors are you seeing, if any?  What is the definition of the yE PERLSHR logical name?  Are you running the PCSI-installed version of p1 Perl from Compaq or did you acquire it elsewhere?a   ------------------------------   End of INFO-VAX 2001.220 ************************