1 INFO-VAX	Tue, 18 Mar 2003	Volume 2003 : Issue 152       Contents:$ Re: 9gig 7200rpm drives in PWS500au?* Re: Alpha Station 200 4/233 Memory Upgrade* Re: Alpha Station 200 4/233 Memory Upgrade Re: alpha300 arrived, now... RE: alpha300 arrived, now... Re: Another missed opportunity9 Re: asking again- what was the fastest vax and alpha made 9 Re: asking again- what was the fastest vax and alpha made J Re: asking again- what was the fastest vax and alpha made (vms of  course)) Re: Changing CDE mouse pointer properties ) Re: Changing CDE mouse pointer properties  Re: ddachter@nmh.org Re: Deliver and UCX 6 Re: Editor wars - Was: Re: Upcoming VMS improvements ?6 Re: Editor wars - Was: Re: Upcoming VMS improvements ?= Re: EFI Specifications (was: I Have Seen The IA64 Future !!!) = Re: EFI Specifications (was: I Have Seen The IA64 Future !!!) - Re: Environment variables lost on AlphaPC 164 - Re: Environment variables lost on AlphaPC 164  Re: Firmaware update6 Re: How to connect Vaxstation 4000 VLC to LCD monitor?6 Re: How to connect Vaxstation 4000 VLC to LCD monitor?( Re: HP does operating system advertising# Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! # Re: I Have Seen The IA64 Future !!! . Re: Is http://www.montagar.com/hobbyist/ down?. Re: Is http://www.montagar.com/hobbyist/ down?. Re: Is http://www.montagar.com/hobbyist/ down?. Re: Is http://www.montagar.com/hobbyist/ down?. Re: Is http://www.montagar.com/hobbyist/ down?. Re: Is http://www.montagar.com/hobbyist/ down?. Re: Is http://www.montagar.com/hobbyist/ down?. Re: Is http://www.montagar.com/hobbyist/ down?0 Is the RCA jack on a SN-PBXGK-AB (4D10T) usable?7 Re: Issue building Perl after applying VMS73_ACRTL V3.0 7 Re: Issue building Perl after applying VMS73_ACRTL V3.0 # Layered Products for open VMS 7.2-1 ' Re: Layered Products for open VMS 7.2-1 * Re: National Moratorium to Appease Tyrants* Re: National Moratorium to Appease Tyrants* Re: National Moratorium to Appease Tyrants* Re: National Moratorium to Appease Tyrants  new releases of MySQL and Python$ Re: new releases of MySQL and PythonP Re: OpenVMS.org: Marvel article and HP's press release for   MarvelandAlpha RetaP Re: OpenVMS.org: Marvel article and HP's press release for   MarvelandAlpha RetaP Re: OpenVMS.org: Marvel article and HP's press release for   MarvelandAlpha RetaP Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha RetaiP Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha RetaiP Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha RetaiP Re: OpenVMS.org: Marvel article and HP's press release for MarvelandAlpha Retain Returning Mouse Position Re: send mail from an intranet?  Re: send mail from an intranet?  Re: send mail from an intranet?  Re: send mail from an intranet?  Re: send mail from an intranet?  Re: send mail from an intranet?  Re: send mail from an intranet?  Re: send mail from an intranet?  Re: send mail from an intranet?  Re: send mail from an intranet?  Re: send mail from an intranet? $ Shareable data image and VMS-cluster( Re: Shareable data image and VMS-cluster? RE: Somebody didn't get the message?  HP shows it doesn't care. ? Re: Somebody didn't get the message?  HP shows it doesn't care. ? Re: Somebody didn't get the message?  HP shows it doesn't care.  strange ENQ rates  Re: strange ENQ rates  SYS_CHECK equivelant on VMS.  Re: SYS_CHECK equivelant on VMS." TCPIP: how to change the MTU value Tuesday March 18 OpenVMS Pearl Re: unix Re: unix Re: unix Re: unix Re: unix Re: unix Re: unix Re: unix Re: unix  F ----------------------------------------------------------------------  # Date: Tue, 18 Mar 2003 07:22:52 GMT ! From: JBloggs@acme..spamless..com - Subject: Re: 9gig 7200rpm drives in PWS500au? 8 Message-ID: <udgd7v08j786g9rf9lnlht0it3323qcfkb@4ax.com>  < I had 3 10K drives in a Aspen Durango (Alpha PC164 500 Mhz) 4 but I did lose a drive last summer (stuck in WRTLCK)  @ After that I was more curious to see exactly what sort of temps @ the drives got to in the summer.  (at my location it rarely getsD warmer than 30 C, ~86 F), but the drives would still get quite warm,B often hot to the touch, and sometimes too hot to hold comfortably.  @ I rigged a battery-powered Radio indoor/outdoor LCD thermometer,< and with the sensor taped to the top of middle drive, I saw $ drive temps as high as 118 F, (~48C)  5 Since then, I rigged two Pabst 120mm fans front/back  > (one to pull air in, the other out), .and the drives nowadays,D are usually only 7-10 F (~5 C) degrees warmer than room temperature.  @ The CPU overclocker websites/newsgroups, can be decent resources@ for tracking down vendors selling good fans (ie, quiet, but with reasonable volume).   : The Pabst fans (made in Germany) are a bit hard to find in> the States, but so far, are the quietest fans I've run across,  ; Beware that some small (90mm) hi-volume fans can be really  : noisy (in fact, loud enough to hear in the next room over)    I >On Mon, 17 Mar 2003 16:50:13 -0800, "Tom Linden" <tom@kednos.com> wrote:  > K >have 3 IBM 10k dives in a PWS500, which probably run hotter and works fine  >  >>-----Original Message-----5 >>From: Bill McLaughlin [mailto:mcbill20@hotmail.com] & >>Sent: Monday, March 17, 2003 4:07 PM >>To: Info-VAX@Mvb.Saic.Com + >>Subject: 9gig 7200rpm drives in PWS500au?  >>H >>I have a PWS500au with three 7200rom 9 gig drives (Seagate ST39175LW).F >>The cover has been off so far as I just got the box a week or so agoG >>and I had to install a second SCSI card, etc. Now that it all works I G >>was going to put the cover on and put it in the rack. The side of the E >>case where the drives are feels pretty warm and I got to thinking-- D >>are these drives supported internally in this box? Anyone else out4 >>there with a PWS using these drives? Any problems?   ------------------------------  + Date: Tue, 18 Mar 2003 14:56:47 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> 3 Subject: Re: Alpha Station 200 4/233 Memory Upgrade ; Message-ID: <01KTO9PGB6DU9FRD0S@sysdev.deutsche-boerse.com>   H > While 384MB is the max 'supported' by an AlphaStation 200 4/233, it isD > possible to take it all the way up to 768MB.  I forget who in thisI > newsgroup took one up that far a few years ago.  I think the one that I 0 > use as a test system is now sitting at 512MB.   I IIRC, the user manual for my 255/233 says that 512 is supported BUT that  8 1 GB would be possible if 128 MB SIMMs become available.   ------------------------------  + Date: Tue, 18 Mar 2003 09:58:12 -0600 (CST)  From: sms@antinode.org3 Subject: Re: Alpha Station 200 4/233 Memory Upgrade ) Message-ID: <03031809581270@antinode.org>   9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> J > > While 384MB is the max 'supported' by an AlphaStation 200 4/233, it isF > > possible to take it all the way up to 768MB.  I forget who in thisK > > newsgroup took one up that far a few years ago.  I think the one that I 2 > > use as a test system is now sitting at 512MB.  > K > IIRC, the user manual for my 255/233 says that 512 is supported BUT that  : > 1 GB would be possible if 128 MB SIMMs become available.  D    The Digital AlphaStation 200 Series User Information manual says,H "You can increase your system's memory to 192 MB (384 MB when 64-MBSIMMs are available)."  * alp $ write sys$output f$getsyi("HW_NAME") AlphaStation 200 4/233   alp $ show memory /physical @               System Memory Resources on 18-MAR-2003 10:02:19.35  L Physical Memory Usage (pages):     Total        Free      In Use    ModifiedL   Main Memory (768.00Mb)           98304       60821       35465        2018  N Of the physical pages in use, 4749 pages are permanently allocated to OpenVMS.  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode.org     Saint Paul  MN  55105-2547    ------------------------------    Date: 18 Mar 2003 14:08:10 -00005 From: "Doc.Cypher" <doc_cypher@redneck.gacracker.org> % Subject: Re: alpha300 arrived, now... 6 Message-ID: <20030318140810.22247.qmail@gacracker.org>  5 NOTE: This message was sent thru a mail2news gateway. 8 No effort was made to verify the identity of the sender.8 --------------------------------------------------------  8 On 17 Mar 2003, timasmith@hotmail.com (Tim Smith) wrote:  G >Courtesy of EBay my alpha3000 has reached my basement, it boots and it B >is hungry for an O/S.  Now I could feed it Linux but it is reallyE >hungry for VMS so I have registered (again) on Encompass but I guess G >it will be a while before my hobbyist license is ready and then I need  >to order the media...  ' Congratulations on your new arrival. :)   K Unfortunately problems getting the required membership and that information K passed to Montagar to permit issuing of Hobbyist licenses seem to be rather 
 prevalent.  K There have been quite a few people grumbling in the Notes system running on H the Deathrow cluster. I've even had private mail indicating that one LUGJ gave the impression that attendance at local meetings was a requirement toB retain membership. That's not what these people are interested in.  G Although it was several years ago, I encountered similar issues getting I DECUS Belux membership information transferred to Montagar so I could get K my licenses. Since then I've been "practicing random acts of marketing", as K have others such as Beave who runs the Deathrow cluster. With the existence J of several VAX emulators it is technically easy for many of the people whoJ opened accounts on our free access systems to set up a system of their own$ and start learning how to manage it.  K On the off-chance that there are more VMS discussions on SlashDot, is there C any way things can be simplified for those wanting Hobbyist access?      Doc. --  : Time and money, the psychotropics of the business world...K ~ VAXman                                             https://vmsbox.cjb.net    ------------------------------  % Date: Tue, 18 Mar 2003 09:48:34 -0500 ! From: VAXVMS <bounce@notmail.com> % Subject: RE: alpha300 arrived, now... K Message-ID: <BA52530E3149734A9BAABDBBFA808E4903027BDB@rlghncst964.usps.gov>    Doc:    ( Let's discuss this outside of the group. Send me an email.   : BTW, as an ex-Georgian, the email address "cracked" me up. That's not real, is it?    WWWebb   Doc.Cypher spake thusly: > 6 >NOTE: This message was sent thru a mail2news gateway.9 >No effort was made to verify the identity of the sender. 9 >--------------------------------------------------------  > 9 >On 17 Mar 2003, timasmith@hotmail.com (Tim Smith) wrote:  > H >>Courtesy of EBay my alpha3000 has reached my basement, it boots and itC >>is hungry for an O/S.  Now I could feed it Linux but it is really F >>hungry for VMS so I have registered (again) on Encompass but I guessH >>it will be a while before my hobbyist license is ready and then I need >>to order the media...  > ( >Congratulations on your new arrival. :) > L >Unfortunately problems getting the required membership and that informationL >passed to Montagar to permit issuing of Hobbyist licenses seem to be rather >prevalent.  > L >There have been quite a few people grumbling in the Notes system running onI >the Deathrow cluster. I've even had private mail indicating that one LUG K >gave the impression that attendance at local meetings was a requirement to C >retain membership. That's not what these people are interested in.  > H >Although it was several years ago, I encountered similar issues gettingJ >DECUS Belux membership information transferred to Montagar so I could getL >my licenses. Since then I've been "practicing random acts of marketing", asL >have others such as Beave who runs the Deathrow cluster. With the existenceK >of several VAX emulators it is technically easy for many of the people who K >opened accounts on our free access systems to set up a system of their own % >and start learning how to manage it.  > L >On the off-chance that there are more VMS discussions on SlashDot, is thereD >any way things can be simplified for those wanting Hobbyist access? >  >  >Doc.  >-- ; >Time and money, the psychotropics of the business world... L >~ VAXman                                             https://vmsbox.cjb.net >======================== ! >William W. Webb - EMS Operations * >OpenVMS Systems Support - USPS DSSC Annex- >4730 Hargrove Road, Raleigh, NC 27616-2874   ? >919.325.7500 x4186  <FirstInitialLastNameAtEmailDotUSPSDotGov>    ------------------------------  + Date: Tue, 18 Mar 2003 11:24:58 +0000 (UTC) 5 From: "John Wallace" <johnwallace4@yahoo.dotco.dotuk> ' Subject: Re: Another missed opportunity 0 Message-ID: <b56vm9$edh$1@sparta.btinternet.com>  . "John Smith" <a@nonymous.com> wrote in messageB news:e3lda.58010$a41.25659@news01.bloor.is.net.cable.rogers.com... > 3 > "Michael Unger" <unger@decus.de> wrote in message ' > news:00A1CF3E.24B411D3.35@decus.de... 7 > > "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote:  > >  > > > [...]  > > > E > > > Coming back to the air traffic control system, was this upgrade 	 > (well I B > > > wouldn't consider a move to NT an "upgrade") prompted by the > mid-air collision C > > > last year ? If the media started to point fingers are the ATC  > system usingE > > > antiquated equipment, I can see the pressure to upgrade, and in  > such aB > > > context, they woudl be forced to buy something which "LOOKS"	 > modern.  > > > D > > > At least they should have gone to Sun. Somehow, NT isn't up to > snuff when it E > > > comes to real time, life critical systems. Announcinmg you will 
 > trust a toy E > > > operating system to control air traffic with thousands of lives  > depending on& > > > some wintel box is pretty scary. > > A > > I would consider a "_hard_ real-time" operating system (OS-9, 
 > VxWorks,F > > ..) more appropriate than any Unix-based system. Winwoes of courseF > > tends to play with the hour-glass when it should perform important
 > actions. >  >  > Or in the old days VMSeln  >  >   C John surely meant o say VAXELN. I'm not personally aware of any ATC H applications of VAXELN but there may be some. VAXELN to my knowledge hasI been used for things like airport lighting control, luggage routing, etc.   J I'm not so sure Sun would be the obvious choice, even after recent changesL in CPQ/HP. European air traffic control in Belgium (Eurocontrol) used to runH on a collection of Tru64 boxes and had done so since the days of the DEC2 3000 and DEC OSF/1.  There is a brief reference atL http://www.hp.be/egov/en/references/index.asp?todo=all but the link is dead.K I can't find a Google-cached copy, but another marginally longer writeup of L the same project can be found from Barco (they did the graphics h/w and someK display s/w) at www.barco.com/corporate/en/pressreleases/show.asp?index=753 - I do not know what the system currently runs.   H I believe some of the newer local trains here in England use Windows forE supervision of non-critical stuff like signs, door management, aircon H management, etc. Of course Windows does fail occasionally, and when thatL happens the train has to stop while the "management system" reboots; variousH well-intentioned design interlocks mean it cannot be rebooted unless theG train is out of service, and obviously a delay of several minutes for a J Windows reboot leads to trains running late. We have many excuses for lateK trains in this country; slow Windows reboots is not one you'll hear, but it  is applicable.  K It has also been alleged that at least one multi-fatality UK train disaster K might have been less serious if a safety system had used a real embedded OS G instead of windows. There have been instances where ATP (Advanced Train F Protection?) could have prevented or lessened a collision, and ATP wasF available on the train and the route but not in use because (again) itJ needed the train to be out of service to start up the ATP system (ie whileE Windows reboots), and any in-service failure of ATP introduces a high H probability of late running, which can result in commercial penalties onL train operators. I have not been able to personally verify this information,) but it has a certain air of plausibility.   J There are proper PC-compatible RT kernels around, but they probably aren'tK programmable by kiddies writing in Visual Basic. QNX and VxWorks might have  been obvious ones to look at.   * Oh dear, I haven't managed to mention VMS.   regards  john   ------------------------------  % Date: Tue, 18 Mar 2003 10:25:42 +0000 ( From: Nic Clews <sendspamhere@127.0.0.1>B Subject: Re: asking again- what was the fastest vax and alpha made) Message-ID: <3E76F426.4D6D47AE@127.0.0.1>    dittman@dittman.net wrote: > + > Nic Clews <sendspamhere@127.0.0.1> wrote: J > > There is a VAX 7000 model 900 board, clocked at 733 (IIRC) rather thanI > > 600 MHz with 16 MB not 4 MB processor cache... Not actually a digital I > > product, even though the SYSLOA is present in more recent versions of 
 > > VMS... > " > Is this the Nemonix upgrade CPU?  > Yes. Actually it is a grey(ish) area, Digital (I mean Digital)H commissioned Nemonix with the upgrade to the 78xx series CPU, however itG got lost in the Compaqtion. e.g. the SCSI board did actually make it as C a real Compaq option and part number very briefly, but only Nemonix D supply the combination variants, even though they had *assigned* but never used Compaq part numbers.    http://www.nemonixinc.com/  , Website doesn't mention the 79xx CPU boards. --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------  % Date: Tue, 18 Mar 2003 10:31:24 +0000 ( From: Nic Clews <sendspamhere@127.0.0.1>B Subject: Re: asking again- what was the fastest vax and alpha made( Message-ID: <3E76F57C.B9BCAE9@127.0.0.1>   Tom Linden wrote:  > ? > Was always curious about overclocking VAXes.  I have run dual D > overclocked Celerons (from 360 to 550) for three years without any< > problems (of course had to put in extra fans and the like) >   B It isn't an 'overclock'. This was the final (?) VAX chip which wasG fabbed on the EV45 ALpha line as I understand it. They use the existing @ 7xxx series boards which are capable of clocking the chip at itsG intended rate. The upgrade involves returning your existing CPU boards, ( which in turn are re-engineered as 79xx.  3 I hope I'm not treading on any toes telling this...  --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------  % Date: Tue, 18 Mar 2003 10:48:59 -0500 ; From: "Brian Tillman" <Brian.Tillman@notnoone.notnohow.com> S Subject: Re: asking again- what was the fastest vax and alpha made (vms of  course) $ Message-ID: <3e773fe8$1@news.si.com>  - >>   NVAX+ is the fastest VAX microprocessor,  > J >You mean to tell me that the all mighty Microvax II doesn't rank up there ????9 >Well, I am very disapointed ;-) :-) :-) :-)_ :-) :-) :-)   A Perhaps a MicroVAX II doesn't, but a MicroVAX 3100 with a Nemonix ) accelerator in it is rated about 50 VUPS.  --  I Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot com 5 Smiths Aerospace  Addresses modified to prevent SPAM. @ 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991 8        This opinion doesn't represent that of my company   ------------------------------  % Date: Tue, 18 Mar 2003 10:13:32 +0200 - From: "Kari Keronen" <kari.keronen@digita.fi> 2 Subject: Re: Changing CDE mouse pointer properties: Message-ID: <ozAda.347$143.173674@reader1.news.jippii.net>  J This doesn't seem to work with CDE. Are you sure that CDE (not DECwindows)H even uses those files ? Both are of course Motif based, but config filesF differ. In DECwindows you could change the cursor color and shape from% session manager menu, but not in CDE.    -Kari-  9 "Christoph Gartmann" <gartmann@immunbio.mpg.de> kirjoitti / viestiss:b54ps4$csc$1@n.ruf.uni-freiburg.de... G > In article <52gda.6$143.3072@reader1.news.jippii.net>, "Kari Keronen"   <kari.keronen@digita.fi> writes:K > >I'm trying to change CDE mouse pointer color and perhaps size, but don't L > >seem to find information how to do that anywhere. Is this even possible ?$ > >Motif 1.2-5, OpenVMS Alpha 7.2-1. > E > I have a file DECW$SMB_POINTER_COLOR.DAT in my login directory. Its  contents is 
 > as follows:  >  > sm.pointer_shape:       0 ' > sm.pointer_foreground:  #F5F5EDED0000 ' > sm.pointer_background:  #FFFF05050000  > sm.versionid:   3.0  > , > Next, I have a file DECW$SMB_POINTER.DAT : > $ > sm.mouse_accel_denominator:     -1$ > sm.mouse_accel_numerator:       -1 > sm.versionid:   3.0 $ > sm.mouse_accel_threshold:       -1' > sm.pointer_button_order:        right  > *doubleClickDelay:      250  > *multiClickTime:        250  >  > K > My pointer is yellow as long as it is on the background. I hope the above  is, > sufficient, I don't remember how I did it. > 
 > Regards, >    Christoph Gartmann  > J > -- --------------------------------------------------------------------+J > | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |J > | Immunbiologie                                                        |J > | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |J > | D-79011  Freiburg, Germany                                           |J > +------------- http://www.immunbio.mpg.de/home/menue.html -------------+   ------------------------------   Date: 18 Mar 2003 09:33:00 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann) 2 Subject: Re: Changing CDE mouse pointer properties0 Message-ID: <b56p4c$5up$1@n.ruf.uni-freiburg.de>  j In article <ozAda.347$143.173674@reader1.news.jippii.net>, "Kari Keronen" <kari.keronen@digita.fi> writes:K >This doesn't seem to work with CDE. Are you sure that CDE (not DECwindows) I >even uses those files ? Both are of course Motif based, but config files G >differ. In DECwindows you could change the cursor color and shape from & >session manager menu, but not in CDE.  O I am using CDE here and I have a yellow pointer with a red border. I figured it 
 out: you need ! a) the DECW$SMB_POINTER_COLOR.DAT F b) in [.DT.SESSIONS] the file SESSIONETC.COM containing a single line:"       $ run sys$system:decw$wsinitN I just verified it. What I didn't figure out, though is how I could change theM colour when my pointer is over a terminal window. There it is still small and 	 black :-(    Regards,    Christoph Gartmann   H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, Germany                                           |H +------------- http://www.immunbio.mpg.de/home/menue.html -------------+   ------------------------------    Date: 18 Mar 2003 04:26:37 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)  Subject: Re: ddachter@nmh.org 3 Message-ID: <AyLzzw8YwUiR@eisner.encompasserve.org>   [ In article <3E7694D2.56AD569D@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:   E > Of course, if you look at the cup as half-full, you now know how to E > reach me during the work day if you have a question you think I can  > maybe answer...   B The other theory is that it is better to bother people only during5 the hours when they choose to receive such questions.    ------------------------------  % Date: Tue, 18 Mar 2003 09:35:53 -0600 % From: Larry Schudt <lschuldt@dls.net>  Subject: Re: Deliver and UCX8 Message-ID: <l3fe7vgc257refndmd29585psnu07q18t4@4ax.com>  , On Mon, 17 Mar 2003 20:38:49 -0400, JF Mezei( <jfmezei.spamnot@vl.videotron.ca> wrote:   >Larry Schudt wrote:( >> We just migrated from TCPWare to UCX, > M >UCX is dead. It is TCPIP Services now. Version 5.3 is available, 5.0 and 5.1  >are common. > C >> software that requires Deliver, which was apparently provided by  > F >Deliver is also available on the freeware CD if I remember correctly.  F I know UCX is dead, we're actually at version 5.4 of TCPIP services. IC tend to still use old names... Heck, I still say VMS, not OpenVMS!    ! BTW, I have it on the CD, thanks!    larry    ------------------------------  % Date: Tue, 18 Mar 2003 10:24:15 -0500 ! From: Jim Agnew <jpagnew@vcu.edu> ? Subject: Re: Editor wars - Was: Re: Upcoming VMS improvements ? ' Message-ID: <3E773A1F.D7EA5DD5@vcu.edu>   E chuckle!!!!  thanks, but the prob is that "they" don't buy it for all # systems...  they did for one, tho..    jim        David Pikcilingis wrote: > * > No need to abuse yourself like that Jim!C > Boston Business Computing makes EDT for UNIX and Windows systems.  >  > David Pikcilingis  > Boston Business Computing  > www.bosbc.com  > 0 > "Jim Agnew" <jpagnew@vcu.edu> wrote in message# > news:3E64BA13.9C0B0789@vcu.edu... L > > i almost went there, practiced vi on a dos laptop while reading o'reilly1 > > in a playhouse type place while babysitting..  > >  > > talk about self-torture... > >  > > and rotten coffee..  > >  > > jim  > >  > >  > > Shane Smith wrote: > > > L > > > NO!!! DON'T DO IT! >>>bullhorn-crackle<<< Mr. Dachtera, step away from > > > the machine! > > > G > > > If you do do it, don't expect us to clean up the drool or buy the  > > > straightjacket.  > > >  > > > Shane  > > >   > > > -----Original Message-----< > > > From: David J. Dachtera [mailto:djesys.nospam@fsi.net]* > > > Sent: Monday, March 03, 2003 6:19 PM > > > To: Info-VAX@Mvb.Saic.Com E > > > Subject: Re: Editor wars - Was: Re: Upcoming VMS improvements ?  > > >  > > > Shane Smith wrote: > > > > M > > > > VI was still a mess last time I was forced to use it (and I /do/ mean  > > > > forced). > > > L > > > True. I found myself on a UNIX box in 1986 when I'd been on IAS (TECO)I > > > for 18 months and RSTS/E and VMS (EDT) for just shy of three years.  > > > J > > > I managed to make vi as EDT-like as I could, but lost the .exrc fileH > > > when I changed jobs (to a VMS/All-in-1 shop). Could probably do it > > > again, I s'pose... > > >  > > > -- > > > David J. Dachtera  > > > dba DJE Systems  > > > http://www.djesys.com/ > > > . > > > Unofficial Affordable OpenVMS Home Page:% > > > http://www.djesys.com/vms/soho/  > >  > > --J > > "4,000 years ago I made a mistake."  Elrond Half-Elven, in "Fellowship > > of the Ring"   --  F "4,000 years ago I made a mistake."  Elrond Half-Elven, in "Fellowship of the Ring"   ------------------------------  # Date: Tue, 18 Mar 2003 17:29:57 GMT + From: LESLIE@JRLVAX.HOUSTON.RR.COM (leslie) ? Subject: Re: Editor wars - Was: Re: Upcoming VMS improvements ? : Message-ID: <pIIda.30221$Ke.1583471@twister.austin.rr.com>  " Jim Agnew (jpagnew@vcu.edu) wrote:G : chuckle!!!!  thanks, but the prob is that "they" don't buy it for all % : systems...  they did for one, tho..  :   $ There's the free EDT-like editor ED:  0     http://clio.rice.edu/EDstuff/ED_Overview.txt  J    "...ED is an EDT look-alike editor that is portable to many platforms. J     If you use EMACS, you'll probably hate it, but it does have some nice 
     features.        ED will:  B     o  Let you edit files on other hosts, if you are connected to         the Internet.7     o  Display many files on the screen simultaneously. A     o  Save key definitions and other editor settings on command. A     o  Let you mark your spot in a file, and return to it easily. 4     o  Let you put tab stops wherever you want them./     o  Let you use wildcards in search strings. 2     o  Let you redefine the keys on your terminal.M     o  Let you say things like: ED *.dat, if you want to edit all .dat files. @     o  Allow you to teach it how to talk to different terminals.A     o  Calculate the value of algebraic expressions that include          math functions.*     o  Sort a file or a portion of a file.%     o  Let you read the network news.   :     ED is free software, see the file COPYING for details.       ED is available from  #     ftp://clio.rice.edu/pub/   ..."     2 --Jerry Leslie   (my opinions are strictly my own)9   Note: leslie@jrlvax.houston.rr.com is invalid for email    ------------------------------  % Date: Tue, 18 Mar 2003 09:56:21 -0500 A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> F Subject: Re: EFI Specifications (was: I Have Seen The IA64 Future !!!). Message-ID: <3e773395$1_3@hpb10302.boi.hp.com>  L Te EFI is it's own animal from Intel.  I'm sure HP had input into it, as did
 Microsoft.  * Yes, there is such thing as a LAN console.  I An even more interesting thing is the management console on (for example) H the rx2600 - which has it's own microprocessor, and LAN device.  You canJ give this thing an IP address, and then connect to the console via Telnet, or from a web browser.  < We are also embracing OpenView, and the HP management tools.      ; "Fabio Cardoso" <fabiopenvms@yahoo.com.br> wrote in message : news:20030318045203.29518.qmail@web20202.mail.yahoo.com... > Dear Sirs and Ladies ! :-) > : > I didnt read the document yet ! Is the EFI specification3 > a merge of the SRM (Alpha) and GSP (PA-RISC) ???? 5 > So will it be acessible by the "lan console" ot the 7 > network interface  ??? I am asking it because I would 7 > like to have remote management of the firmware and it A > will be usefull for disaster recovery (remote sites management, ; > remote clusters troubleshooting) ! May I will have reason " > in this newsgroup someday  ? ;-) >  >  > 	 > Regards  >  > FC4 > --- Larry Kilgallen <Kilgallen@SpamCop.net> wrote:I > > In article <uktda.448$k72.294@news.cpqcorp.net>, hoff@hp.nospam (Hoff  > > Hoffman) writes: > > 4 > > >   Please get the EFI specs.  Please read them. > > > # > > >   http://developer.intel.com/  > > @ > > Easy for you to say -- you probably have JavaScript enabled. > > > > > http://www.intel.com/technology/efi/main_specification.htm > > H > > produces a lovely blank page other than the marketing fluff and html< > > garbage (using Netscape Communicator 4.75 on Macintosh). >  >  > =====  > ========================== > Fbio dos Santos Cardoso > OpenVMS System Manager > Rio de Janeiro - Brazil  > fabiopenvms@yahoo.com.br > ========================== > 4 > __________________________________________________ > Do you Yahoo!?H > Yahoo! Platinum - Watch CBS' NCAA March Madness, live on your desktop! > http://platinum.yahoo.com    ------------------------------  % Date: Tue, 18 Mar 2003 18:08:13 +0100 $ From: Michael Unger <unger@decus.de>F Subject: Re: EFI Specifications (was: I Have Seen The IA64 Future !!!)+ Message-ID: <00A1D101.CE3E9AC2.49@decus.de>   B "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote:  N > Te EFI is it's own animal from Intel.  I'm sure HP had input into it, as did > Microsoft. >  > [...]   E Obviously Micro$oft added the requirement for a FAT partition on the   system disk ... :-(    Michael    ------------------------------  % Date: Tue, 18 Mar 2003 08:29:49 +0100 6 From: Horst Drechsel <ai05@sternwarte.uni-erlangen.de>6 Subject: Re: Environment variables lost on AlphaPC 164< Message-ID: <00A1D0B1.00CC595B.5@sternwarte.uni-erlangen.de>  & >Date: Mon, 17 Mar 2003 09:18:04 +0100; >From: Karl Rohwedder <extern.karl.rohwedder@volkswagen.de> 7 >Subject: Re: environment variables lost on AlphaPC 164 0 >Message-ID: <b5403s$6ho1@doiweb4.volkswagen.de> >  >Horst Drechsel schrieb: >>  = >> on our AlphaPC 164, which doesn't boot any more after the  : >> exchange of a crashed power supply. We have removed all< >> PCI cards except the SCSI controller for the CD drive and= >> the graphics card, and powered off the floppy. When trying < >> to boot from an OpenVMS 7.1-1H1 distribution CD, it hangs* >> in boot after the first line of output: >>  > >>       OpenVMS (TM) Alpha Operating System, Version V7.1-1H1 >>  @ >> (after that it keeps on reading from the CD for a few seconds >> and then gets stuck)  >> eJ >> ----------------------------------------------------------------------- >> g >>>>>sh config >> o5 >>                        Digital AlphaPC 164 500 MHz. >>  J >> SRM Console V4.9-2             VMS PALcode V1.20-9, OSF PALcode V1.22-8 >> w >> SROM Revision: 2.2a >> DECchip (tm) 21164A-2% >> DECchip (tm) 21172 CIA ASIC Pass 3m >>  	 >> MEMORYa >> 256 Meg of system memoryx >> a >> m
 >> PCI Bus? >>      Bus 00  Slot 05: Digital S3 TRIO 64 Graphics Controllerg3 >>      Bus 00  Slot 07: NCR 53C810 Scsi Controller I >>                                    pka0.7.0.7.0          SCSI Bus ID 7?P >>                                    dka400.4.0.7.0         TEAC CD-ROM CD-532S8 >>      Bus 00  Slot 08: Intel 82378IB PCI to ISA Bridge3 >>      Bus 00  Slot 11: CMD PCI0646 IDE ControlleriD >>                                    dqa0.0.0.11.0         PCI EIDE >> l >> ISAN >> Slot    Device Name             Type         Enabled  BaseAddr  IRQ     DMA >> 0E >>         0      MOUSE            Embedded        Yes     60      12nD >>         1      KBD              Embedded        Yes     60      1D >>         2      COM1             Embedded        Yes     3f8     4D >>         3      COM2             Embedded        Yes     2f8     3D >>         4      LPT1             Embedded        Yes     3bc     7L >>         5      FLOPPY           Embedded        Yes     3f0     6       2 >>   >>  Q >> ------------------------------------------------------------------------------e >>  3 >>    When trying a converstional boot from CD with  >>   >>    >>>b -flags 0,1 dka400 >> /* >> we get the SYSBOOT> prompt. When saying >> d >>    SYSBOOT>continue   t >> c2 >> we end up with the same effect described above. >> When saying >> s >>    SYSBOOT>exit >> h >> we get the error message: >> oQ >> ------------------------------------------------------------------------------hD >>    %SYSBOOT-I-FILENOTLOC, Unable to locate SYS$PUBLIC_VECTORS.EXEQ >>    %SYSBOOT-F-LDFAIL, unable to load SYS$PUBLIC_VECTORS.EXE, status = 00000870s >> u >>    halted CPU 0 >> g >>    halt code = 5a  >>    HALT instructions executed >>    PC = 177f4	 >>    >>>eQ >> ------------------------------------------------------------------------------o    I >Error 870 is 'END-OF-FILE', may be your SYS$PUBLIC_VECTORS.EXEis defect.t+ >Try booting from VMS-CD and check|restore.Q >n >--  > , >mit freundlichen Gren | with best regards >.! >Karl Rohwedder		iT-Ingenieurteamr% >Ellernbruch 11		D-38112 Braunschweigp    A    Actually we get the above error message when using an originaleA OpenVMS distribution CD, we even tried two differnet original CDsr@ which had proven ok on other machines. I suspect the problem is > connected with the loss of information from the mainboard BIOS@ or ROM area due to the temporary power off (with removed onboardB 3V battery) and subsequent firmware update (from official firmware@ upgrade CD V5.0). What else could be tried to get that node back running again?      Horst     --M  ****************************************************************************e)   Horst Drechsel                         rL   Dr. Remeis Observatory                 drechsel@sternwarte.uni-erlangen.deL   Astronomical Institute                             Phone: +49-951-95222-15L   University Erlangen-Nuernberg                        Fax: +49-951-95222-22*   Sternwartstr.7, D-96049 Bamberg, GermanyM  ****************************************************************************e   ------------------------------  # Date: Tue, 18 Mar 2003 10:44:35 GMTe) From: Patrick Young <P.Young@unsw.EDU.AU>e6 Subject: Re: Environment variables lost on AlphaPC 164= Message-ID: <nMCda.4684$L57.42808@news-server.bigpond.net.au>a   Horst Drechsel wrote: = >>>on our AlphaPC 164, which doesn't boot any more after the  : >>>exchange of a crashed power supply. We have removed all< >>>PCI cards except the SCSI controller for the CD drive and= >>>the graphics card, and powered off the floppy. When trying-< >>>to boot from an OpenVMS 7.1-1H1 distribution CD, it hangs* >>>in boot after the first line of output: >>>i> >>>      OpenVMS (TM) Alpha Operating System, Version V7.1-1H1 >>>>@ >>>(after that it keeps on reading from the CD for a few seconds >>>and then gets stuck)s  ; Sorry, I didn't get back to your email however start of ours3 academic session was quite *evil* this time around.t  ? Please install SRM V4.9-2 and see how you go. 5.5-1 for example)F is quite broken in that it won't remember ISACFG settings (unless they( were made with a different SRM version).  B Failing that, remove everything, except memory, add a Trio64, plus< of course the SCSI controller for the CDROM. Then you get to; change memory. The answer to what is broken should start too show up appear at this point.   @ Off topic, we had a 2nd memory failure on our 8400 - the machineB would stop dead with the fault light on. I was in luck since I had, that spare module with only one failed SIMM.   ------------------------------  % Date: Tue, 18 Mar 2003 10:32:42 +0100 7 From: Robert Trawinski <robert.trawinski@softax.com.pl>  Subject: Re: Firmaware update / Message-ID: <b56n30$k1d$1@bozon2.softax.com.pl>    Robert Trawinski wroteD > I've just got Alpha. It's not machine from DEC/Compaq. Indeed its J > home-made computer. I'd like to install VMS 7.3-1 on it. SRM version is F >   5.8-1. When I enter show config I see that motherboard is Digital I > AlphaPC 164LX 599Mhz . Is it possible to get newer version of firmware?i >  > Robert >   A Thanks  for your help. I didn't describe all the story. Indeed I yD installed VMS 7.3-1 from CD and it works on my machine. I recognize E problem when I tried to install Perl 5.6-1. During PCSI installation rD process system hangs up. I thought that was a problem with firmware H version. The same problem appears when I tried to install TCPIP 5.3 ECO H 182 (installation of ECO 181 was succesfull) and some other VMS patches.   Does anybody know what happens?h   Robert   ------------------------------    Date: 17 Mar 2003 23:05:20 -0800) From: P.Young@unsw.EDU.AU (Patrick Young) ? Subject: Re: How to connect Vaxstation 4000 VLC to LCD monitor?p= Message-ID: <55f85d77.0303172305.2b5cef7f@posting.google.com>   X "pbc" <pbc@informchaos.com> wrote in message news:<b55cn1$7em$1@alpha2.radio-msu.net>...I > I wonder is there any way to connect my Vaxstation 4000 VLC (it has RGB9I > cable) to a regular LCD monitor or any other monitor without RGB input.b* > Any converters? Any other possibilities?  E My workstation at work is an OpenVMS PWS600AU (Alpha) connected to an 7 18 inch LCD monitor and it is *sweet*, crisp and clear.a  A These monitors however tend to want to run at the resolution theysC were designed for (in my case 1280x1024), at other resolutions theyw are plain ugly.i  D It is a long while since I've been involved with VAX systems, and myD memory is not good at the best of times, however systems of this ageB (Sun/DEC/SGI) used a range of scan frequencies with Sync on green.  H There might be some "supa dupa" LCD monitor out there to deal with this,B but I have not seen one - if there is it won't be cheap. The world? caters these days to the M$ Window(tm) PC users, unfortunately.6   ------------------------------  # Date: Tue, 18 Mar 2003 18:09:40 GMTsA From: "Colin Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk>t? Subject: Re: How to connect Vaxstation 4000 VLC to LCD monitor?h= Message-ID: <EhJda.5505$S%6.41116909@news-text.cableinet.net>   1 http://www.cadesigns.co.uk/mi1a.htm looks useful.t  G I found this by searching for +"VGA to RGB" +converter using AltaVista.'    K Conversely I also found the following (wrong way round for what you want tou do):  L Black Box do a XGA/VGA to RGB converter - AC9001-R2. Not cheap (265.00 list in UK), but probably works.r   You could roll your own -59 http://www.hut.fi/Misc/Electronics/circuits/vga2rgbs.html   G I found this by searching for +"RGB to VGA" +converter using AltaVista.    -- Hope this helps. Cheers, Colin.a' (colinDOT.butcherAT@xdeltaDOT.coDOT.uk)y    , "pbc" <pbc@informchaos.com> wrote in message) news:b55cn1$7em$1@alpha2.radio-msu.net...-I > I wonder is there any way to connect my Vaxstation 4000 VLC (it has RGBWI > cable) to a regular LCD monitor or any other monitor without RGB input.-* > Any converters? Any other possibilities? >0 > Peter0 >0 >    ------------------------------  # Date: Tue, 18 Mar 2003 14:48:39 GMTa# From: "John Smith" <a@nonymous.com>01 Subject: Re: HP does operating system advertisingsI Message-ID: <blGda.170075$em1.86436@news04.bloor.is.net.cable.rogers.com>t  8 "Steve Spires" <Steve.Spires@torex.com> wrote in message? news:91947A84607D9D48B8E674A5FAB54DA6854459@tahiti.tinuk.com... D We have a TV advert the same as this, except IIRC it doesn't mention anyt@ OS at all, and I think some of the sections were different, eg I& remember amazon.com being one of them.  = If they are going to mention operating systems at all, as Sue - suggested that HP does not, then why not VMS?   F Each double-page ad in my paper costs in the vicinity of $150,000 each time it is run.nE In this series of ads, HP mentioned 3 specific o/s'es by name, two ofa which they don't own.m   ------------------------------  % Date: Tue, 18 Mar 2003 09:32:06 -0500eA From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> , Subject: Re: I Have Seen The IA64 Future !!!. Message-ID: <3e772de6$1_1@hpb10302.boi.hp.com>  + "Dirk Munk" <munk@home.nl> wrote in messagem, news:b55jjl$adl$1@news2.tilbu1.nb.home.nl...  > Wow, you have been busy Fred ! >uK > There is indeed a misunderstanding somewhere. I received information fromu a veryI > reliable source stating that "If Intel has its way, it (= the EFI) wille replaceeJ > the BIOS in all PC's in the future". Now I assume that what you refer to as "thet; > console" is what is commonly known as the bios with PC's.a >s  J And here is the misunderstanding.  EFI *is* a console.  It lives in ROM orI Flash ROM.  It is extensible.  It is a small OS.  It uses a collection of D standards like ACPI (and AML byte code delivered by the platform) toL discover busses, built in devices, and program them.  It uses the PCI option+ ROMs to init and interact with PCI devices.   C The "BIOS" concept is an ad-hoc collection of firmware, with little I dicipline, but which has boxed in system makers to always provide 20 year-H old technology in their platforms so that devices can be initialized and used.m  H > So I assumed that in future (maybe not with the first IA64 generation) thisL > would be the case. Every boot would then be a kind of operation similar to thehK > failsave boot proces with an Alpha. So a very basic bootloader would loada the L > firmware/bios/console etc. from a FAT disk, just as the firmware is loaded from& > a floppy with a Alpha failsave load. >a  J Nope.  The burned in ROM contents has enough to boot.  The disk content isL extended, optional functionality.  A platform builder can also burn extendedF things into the ROM as well  - say someone wanted to add MOP as a boot	 protocol.b  H > In this case I assume the EFI would also contain all settings for thatJ > particular hardware, and not just some common diagnostics etc. So a veryC > different situation from what you are describing, but thinkable !a >a  J Intel, Microsoft and everyone is getting away from the ISA space, and intoJ things like PCI and beyond, where the code for the device in in the optionJ ROM.  EFI will also drive a new option ROM format that will be more useful5 than the current ROMs which are setup for BIOS calls.i  H But there is also platform-specifics that need to be embedded - like theI code to init and program the core logic chips, the interrupt controllers,iF find built-in devices that have no PCI config header (like some serialK ports, or IDE devices).  This is provided in ROM by the platform builder ins) a AML byte code, that is called via ACPI.s  I > Now maybe I misunderstood the information I got, and my source was onlyI referingI > to the human interface to the bios functions. If that is the case, then  > everything changes of course.o >n   ------------------------------  % Date: Tue, 18 Mar 2003 09:32:06 -0500 A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com>u, Subject: Re: I Have Seen The IA64 Future !!!. Message-ID: <3e772fe2$1_3@hpb10302.boi.hp.com>  + "Dirk Munk" <munk@home.nl> wrote in messageS, news:b55jjl$adl$1@news2.tilbu1.nb.home.nl...  > Wow, you have been busy Fred ! >tK > There is indeed a misunderstanding somewhere. I received information froms a veryI > reliable source stating that "If Intel has its way, it (= the EFI) willt replacebJ > the BIOS in all PC's in the future". Now I assume that what you refer to as "theS; > console" is what is commonly known as the bios with PC's.h >C  J And here is the misunderstanding.  EFI *is* a console.  It lives in ROM orI Flash ROM.  It is extensible.  It is a small OS.  It uses a collection of0D standards like ACPI (and AML byte code delivered by the platform) toL discover busses, built in devices, and program them.  It uses the PCI option+ ROMs to init and interact with PCI devices.t  C The "BIOS" concept is an ad-hoc collection of firmware, with littlerI dicipline, but which has boxed in system makers to always provide 20 year0H old technology in their platforms so that devices can be initialized and used.>  H > So I assumed that in future (maybe not with the first IA64 generation) thisL > would be the case. Every boot would then be a kind of operation similar to thegK > failsave boot proces with an Alpha. So a very basic bootloader would loads thedL > firmware/bios/console etc. from a FAT disk, just as the firmware is loaded from& > a floppy with a Alpha failsave load. >a  J Nope.  The burned in ROM contents has enough to boot.  The disk content isL extended, optional functionality.  A platform builder can also burn extendedF things into the ROM as well  - say someone wanted to add MOP as a boot	 protocol.y  H > In this case I assume the EFI would also contain all settings for thatJ > particular hardware, and not just some common diagnostics etc. So a veryC > different situation from what you are describing, but thinkable !S >   J Intel, Microsoft and everyone is getting away from the ISA space, and intoJ things like PCI and beyond, where the code for the device in in the optionJ ROM.  EFI will also drive a new option ROM format that will be more useful5 than the current ROMs which are setup for BIOS calls.   H But there is also platform-specifics that need to be embedded - like theI code to init and program the core logic chips, the interrupt controllers, F find built-in devices that have no PCI config header (like some serialK ports, or IDE devices).  This is provided in ROM by the platform builder ino) a AML byte code, that is called via ACPI.d  I > Now maybe I misunderstood the information I got, and my source was onlye referingI > to the human interface to the bios functions. If that is the case, then. > everything changes of course.p >b   ------------------------------  % Date: Tue, 18 Mar 2003 09:40:34 -0500mA From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com>c, Subject: Re: I Have Seen The IA64 Future !!!, Message-ID: <3e773111_3@hpb10302.boi.hp.com>  = "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in messagea) news:3E765FE0.61B7FFC8@vl.videotron.ca...i > Fred Kleinsorge wrote:J > > The EFI console in ROM can load the primary bootstrap (OS Loader) from theaL > > disk partition.  It can *also* use TFTP and DHCP to perform a BOOTP-like( > > network boot *NO LOCAL DISK NEEDED*. >cI > As far as the basic flash/rom EFI is concerned, is booting an "extendedo EFI"< > with all the SRM-style goodies the same as booting an OS ? >r   Yes.  I > If you can boot the "extended EFI" program from the network OR from theo FAT & > partition on a drive, then it is OK. >.  L Network boot is a file based boot.  So it contains whatever you put into theK executable that is downloaded.  It can be simple, or complex.  Consider ISL J (Infoserver) booting versus normal remote VMS booting.  It's just whatever we decide we need in it.  G > It has been said that diskette drives and serial ports are consideredeH > "legacy". is CD-ROM going to be the long term supported medium for the loadingsD > of EFI applications when there is none already loaded on a drive ? >i  K Serial ports might be considered legacy, but for servers they will probablyn& not dissapear.  Check the DIG64 specs.  K I'm not exactly clear on what you are asking here.  A bootable VMS disk forrE IPF will have the EFI utilities on it.  CDROM or magnetic media.  The K install CDROM/DVD for VMS will have them on it.  It will be a standard parta of any bootable media.  J > Or will serious servers designed to run VMS continue to be equipped withJ > different equipment such as serial ports that may provide other means of  > booting the EFI applications ? >   G You don't boot from a serial port.  You boot from disk or network.  ThetI console port can be anything from a serial line, to a graphics head, to aaH LAN connection - depending on the system and how you set it up.  The EFII interface is a simple interface that likes something that looks like a 24  line VT100.h   >BL > Any thoughts on supporting paper tape for the EFI to maintain the good old > Digital history ?t	 > :-) :-)c  K All you need to do is find one, figure out how to connect it, and write thet EFI driver for it.  J There *is* for example a USB floppy EFI boot driver that I've seen.  But I don't think we'll be using it.   ------------------------------  % Date: Tue, 18 Mar 2003 09:40:34 -0500 A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> , Subject: Re: I Have Seen The IA64 Future !!!. Message-ID: <3e773296$3_1@hpb10302.boi.hp.com>  = "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in message ) news:3E765FE0.61B7FFC8@vl.videotron.ca...e > Fred Kleinsorge wrote:J > > The EFI console in ROM can load the primary bootstrap (OS Loader) from the L > > disk partition.  It can *also* use TFTP and DHCP to perform a BOOTP-like( > > network boot *NO LOCAL DISK NEEDED*. > I > As far as the basic flash/rom EFI is concerned, is booting an "extended  EFI"< > with all the SRM-style goodies the same as booting an OS ? >a   Yes.  I > If you can boot the "extended EFI" program from the network OR from the  FATe& > partition on a drive, then it is OK. >t  L Network boot is a file based boot.  So it contains whatever you put into theK executable that is downloaded.  It can be simple, or complex.  Consider ISL J (Infoserver) booting versus normal remote VMS booting.  It's just whatever we decide we need in it.  G > It has been said that diskette drives and serial ports are consideredxH > "legacy". is CD-ROM going to be the long term supported medium for the loadingsD > of EFI applications when there is none already loaded on a drive ? >   K Serial ports might be considered legacy, but for servers they will probably & not dissapear.  Check the DIG64 specs.  K I'm not exactly clear on what you are asking here.  A bootable VMS disk foraE IPF will have the EFI utilities on it.  CDROM or magnetic media.  The+K install CDROM/DVD for VMS will have them on it.  It will be a standard parti of any bootable media.  J > Or will serious servers designed to run VMS continue to be equipped withJ > different equipment such as serial ports that may provide other means of  > booting the EFI applications ? >S  G You don't boot from a serial port.  You boot from disk or network.  The I console port can be anything from a serial line, to a graphics head, to aeH LAN connection - depending on the system and how you set it up.  The EFII interface is a simple interface that likes something that looks like a 24- line VT100.-   >-L > Any thoughts on supporting paper tape for the EFI to maintain the good old > Digital history ?u	 > :-) :-)C  K All you need to do is find one, figure out how to connect it, and write the. EFI driver for it.  J There *is* for example a USB floppy EFI boot driver that I've seen.  But I don't think we'll be using it.   ------------------------------  % Date: Tue, 18 Mar 2003 09:48:15 -0500nA From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com>n, Subject: Re: I Have Seen The IA64 Future !!!. Message-ID: <3e773297$1_1@hpb10302.boi.hp.com>   Yes/No.r  I They have seperated out those things that do not *need* to reside in ROM,tI and made it documented and extensible.  For example, someone wrote an FTPgL EFI application that allows you *from the console* to copy new images to theF EFI partition.  Veeery handy.  You could write an EFI application thatK understands ODS5 and can copy files or do other things.  You can write yourg: own diagnostic exersizer (in fact, HP provides just this).  K What is in ROM is the basic stuff that allows POST, booting, and bus/deviceb@ discovery/programming, plus a simple menu driven boot interface.  G You want a UNIX-like shell?  It's on the disk.  You want some debugging1K stuff, it's in the shell.  If a platform builder thinks that this interfacecI should be in the basic ROM, they can do that.  Linux/UNIX/VMS might thinkCH that should be standard, Windows people might never want it.  So you can pick and choose.  K Frankly, it nearly matches the outline for a new console to replace the SRMbI which a taskforce (of firmware and OS developers) I was on many years ago-; proposed - but which we could not justify in the end doing.+    = "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in messagex) news:3E76627A.620B1CD2@vl.videotron.ca...  > Fred Kleinsorge wrote:  > > The CONSOLE IS IN FLASH ROM. > L > Is it correct to state that the console program that resides in ROM is not asI > extensive/developped as VMS users are used to and that as a result, you. needK > an extended console program to match the capabilities that VMS users have  come
 > to expect ?3   ------------------------------  % Date: Tue, 18 Mar 2003 09:50:33 -0500:A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com>., Subject: Re: I Have Seen The IA64 Future !!!. Message-ID: <3e773297$3_1@hpb10302.boi.hp.com>  K Alpha VMS in general requires a CDROM for installation of new versions, andy6 initial boot if the system has not been pre-installed.  I You can satisfy that by using an Infoserver (if you still have one).  YousJ can also simply do a network boot with no local devices to another system.      8 "Andy Stoffel" <a.stoffel@adelphia.net> wrote in message8 news:Xns9341DAA6E4588acsfcgnetworksnet@216.166.71.230...1 > rob.buxton@wcc.spam.govt.nz enlightened us witht. > news:3e766812.948907917@news on 17 Mar 2003: > A > > On 17 Mar 2003 09:03:39 -0600, young_r@encompasserve.org (Rob  > > Young) wrote:i > >  > [snip]B > >>     Last I checked all Alphas have a CDROM.  Same should hold! > >>     true for all IA64 boxes.  > >>@ > > We have a DEC 3000 M600 here that definately does NOT have aD > > CDROM. Now it has been here substantially longer than I have but@ > > I don't believe a CDROM was removed. Our second DEC3000 M600A > > does have a CDROM and we also have an InfoServer. So, I'm not % > > convinced all Alphas have CDROMS.i >p> > Hmmm... I ran into an AlphaServer 1000 (back in '96 I think)@ > during a long stay at a customer site that was ordered SOMEHOWE > without a CD-ROM drive. Never could figure out how that happened... @ > but they didn't order it through us ( which our customers haveF > traditionally done) and the Digital salescritter who sold it to themD > was no longer "available" so I never did find out how it happened. >JF > Weird but not a crisis. There were other Alphas running VMS that DID2 > have CD-ROM drives. And the floppy drive worked. >  > -Andy-   ------------------------------  % Date: Tue, 18 Mar 2003 09:32:06 -0500 A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> , Subject: Re: I Have Seen The IA64 Future !!!. Message-ID: <3e773296$1_1@hpb10302.boi.hp.com>  + "Dirk Munk" <munk@home.nl> wrote in messager, news:b55jjl$adl$1@news2.tilbu1.nb.home.nl...  > Wow, you have been busy Fred ! >7K > There is indeed a misunderstanding somewhere. I received information fromz a veryI > reliable source stating that "If Intel has its way, it (= the EFI) willg replacemJ > the BIOS in all PC's in the future". Now I assume that what you refer to as "thee; > console" is what is commonly known as the bios with PC's.e >e  J And here is the misunderstanding.  EFI *is* a console.  It lives in ROM orI Flash ROM.  It is extensible.  It is a small OS.  It uses a collection of D standards like ACPI (and AML byte code delivered by the platform) toL discover busses, built in devices, and program them.  It uses the PCI option+ ROMs to init and interact with PCI devices.   C The "BIOS" concept is an ad-hoc collection of firmware, with littleuI dicipline, but which has boxed in system makers to always provide 20 year H old technology in their platforms so that devices can be initialized and used.d  H > So I assumed that in future (maybe not with the first IA64 generation) thisL > would be the case. Every boot would then be a kind of operation similar to the K > failsave boot proces with an Alpha. So a very basic bootloader would load  thetL > firmware/bios/console etc. from a FAT disk, just as the firmware is loaded from& > a floppy with a Alpha failsave load. >u  J Nope.  The burned in ROM contents has enough to boot.  The disk content isL extended, optional functionality.  A platform builder can also burn extendedF things into the ROM as well  - say someone wanted to add MOP as a boot	 protocol.h  H > In this case I assume the EFI would also contain all settings for thatJ > particular hardware, and not just some common diagnostics etc. So a veryC > different situation from what you are describing, but thinkable !a >t  J Intel, Microsoft and everyone is getting away from the ISA space, and intoJ things like PCI and beyond, where the code for the device in in the optionJ ROM.  EFI will also drive a new option ROM format that will be more useful5 than the current ROMs which are setup for BIOS calls.   H But there is also platform-specifics that need to be embedded - like theI code to init and program the core logic chips, the interrupt controllers,rF find built-in devices that have no PCI config header (like some serialK ports, or IDE devices).  This is provided in ROM by the platform builder in ) a AML byte code, that is called via ACPI.a  I > Now maybe I misunderstood the information I got, and my source was onlyn referingI > to the human interface to the bios functions. If that is the case, theni > everything changes of course.t >    ------------------------------  % Date: Tue, 18 Mar 2003 18:04:13 +0100 $ From: Michael Unger <unger@decus.de>, Subject: Re: I Have Seen The IA64 Future !!!+ Message-ID: <00A1D101.3F45AE55.39@decus.de>"  B "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote:   > [...]  > M > What is in ROM is the basic stuff that allows POST, booting, and bus/device B > discovery/programming, plus a simple menu driven boot interface. > I > You want a UNIX-like shell?  It's on the disk.  You want some debuggingsM > stuff, it's in the shell.  If a platform builder thinks that this interfaceuK > should be in the basic ROM, they can do that.  Linux/UNIX/VMS might think J > that should be standard, Windows people might never want it.  So you can > pick and choose. >  > [...]i  F What about making systems with a _removable_ Flash-ROM card? That way E (cheap) Winwoes systems with a small ROM as well as "fully featured" oH Unix or VMS systems can be produced without any change to the mainboard.  F (But I tend to digress -- that's marketing and not engineering stuff.)   Michaelt   ------------------------------  % Date: Tue, 18 Mar 2003 18:06:45 +0100v$ From: Michael Unger <unger@decus.de>, Subject: Re: I Have Seen The IA64 Future !!!+ Message-ID: <00A1D101.99D429B9.45@decus.de>   0 "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote:  T > In article <00A1D03A.6CAC46DD.42@decus.de>, Michael Unger <unger@decus.de> writes: > L > > But EFI, other than VAX console or Alpha SRM console, is located on the K > > system disk in a file system (FAT) different from the OS's file system  G > > (ODS), although it may be hidden in a "black box" (container file).m > > 9 > > That's a "mix-up" of specifications I really dislike.  > F > Then you should really dislike files that intermix block storage andE > record storage contrary to the RMS specification.  You can rid yourd, > system disk of such files with the command > % > 	DELETE SYS$SYSDEVICE:[*...]*.EXE;*l > D > On second thought, perhaps you should do that when booted from the > distribution CDROM :-)  D I dislike intermixed _file_ _systems_ for the _same_ OS on the same H (system) disk device; that's different from having files with different 0 internal structures within the same file system.   Michael    ------------------------------  % Date: Tue, 18 Mar 2003 18:05:31 +0100d$ From: Michael Unger <unger@decus.de>, Subject: Re: I Have Seen The IA64 Future !!!+ Message-ID: <00A1D101.6DCF9B5A.43@decus.de>a  B "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote:   > [...]  > L > Nope.  The burned in ROM contents has enough to boot.  The disk content isN > extended, optional functionality.  A platform builder can also burn extendedH > things into the ROM as well  - say someone wanted to add MOP as a boot > protocol.0 >  > [...]@  I So why isn't it done that way? (Code in a Flash-ROM as with Alpha; a few t7 MB of Flash-ROM should be affordable these days ... :-)t  E This would avoid the trouble with the FAT "partition" to be embedded " into the ODS file system.w  G Perhaps the code can even be compressed as has been done in the DECnis aD systems if it will be neccessary to to reduce the number or size of 
 Flash-ROM(s).e   Michaelt   ------------------------------  % Date: Tue, 18 Mar 2003 09:51:07 -0800 ' From: David Mathog <mathog@caltech.edu>e, Subject: Re: I Have Seen The IA64 Future !!!8 Message-ID: <20030318095107.7b98fa9f.mathog@caltech.edu>    On Mon, 17 Mar 2003 22:42:38 GMT$ hoff@hp.nospam (Hoff Hoffman) wrote:  c > In article <r8i23CQ5II$r@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes: e > :In article <3E74F036.AD283FDB@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:l > :> re: SRM console media.m > :>  N > :> I am concerned that in a cluster, not all nodes would be equipped with a  > :> CDROM.N > E >   Hardware cost-savings requirements overwrote the system manager's= >   sanity buffer, I see. :-)s >   @ Not really.  Or at least, not necessarily on all platforms.  OurE (Athlon) beowulf cluster has hard disks and floppies on all nodes but_I no CDROMs.  The BIOS is set on each to do a PXE boot and the dhcpd serveraH is configured to either respond to the boot request by either loading anI OS image or by triggering a normal boot from the internal disk.  We could D probably have lived without the floppies as well but it is worth theJ few dollars to be able to load a new BIOS without having to open the case.  M When the price of a (bootable) CDROM drive drops below that of a floppy driverE then you'll find this type of cluster having CD drives and no floppy.i  F > All Itanium systems will likely require direct or served access to aI > DVD+R or similarly capable device, though this assumption has certainly G > not yet been made official nor codified in the SPD in any way as yet.0  A Served _or_ direct would be ok - that would make it safe for mostR9 machines in a cluster to not have the local media device.0   Regards,   David Mathog mathog@caltech.edu> Manager, Sequence Analysis Facility, Biology Division, Caltech   ------------------------------    Date: 18 Mar 2003 11:51:43 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)s, Subject: Re: I Have Seen The IA64 Future !!!3 Message-ID: <p3eFBgMUXSbl@eisner.encompasserve.org>n  R In article <00A1D101.3F45AE55.39@decus.de>, Michael Unger <unger@decus.de> writes:  H > What about making systems with a _removable_ Flash-ROM card? That way G > (cheap) Winwoes systems with a small ROM as well as "fully featured" OJ > Unix or VMS systems can be produced without any change to the mainboard.  A I would prefer to be able to tell _anyone_ they can upgrade their - machine to VMS through software-only changes.i   ------------------------------    Date: 18 Mar 2003 11:53:04 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)-, Subject: Re: I Have Seen The IA64 Future !!!3 Message-ID: <eeus4BPmmgDg@eisner.encompasserve.org>0  R In article <00A1D101.6DCF9B5A.43@decus.de>, Michael Unger <unger@decus.de> writes:D > "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote: >  >> [...] >> CM >> Nope.  The burned in ROM contents has enough to boot.  The disk content issO >> extended, optional functionality.  A platform builder can also burn extended0I >> things into the ROM as well  - say someone wanted to add MOP as a boot  >> protocol. >>   >> [...] > K > So why isn't it done that way? (Code in a Flash-ROM as with Alpha; a few M9 > MB of Flash-ROM should be affordable these days ... :-)  >   L Presumably because of the desire to make VMS able to run on _any_ Itanium-2.   ------------------------------    Date: 18 Mar 2003 02:15:34 -0600/ From: guthrey@encompasserve.org (Chris Guthrey)B7 Subject: Re: Is http://www.montagar.com/hobbyist/ down? 3 Message-ID: <sElQJYL9qUU8@eisner.encompasserve.org>1  g In article <c5cf6e8.0303170601.46848244@posting.google.com>, baby_p_nut@yahoo.com (Baby Peanut) writes:  > F > Honey, I got the CD-ROM and installed OpenVMS and filled out the webH > site to get the PAKs and waited a month and filled out the web site toH > get the PAKs again and waited and waited.  If I went to Sun to buy theG > $20 copy of Solaris 9 on IA-32 they would have shipped me something I-E > could use.  Instead I'm out $30 and have no toy VMS to show for it.o    I When did you try this?  I went to the Montager site last night and issued O myself a PAK for my latest hobby machine (an AlphaServer 800!).  I recieved thep# PAK in the email within 30 seconds..  L So having established that the site is working, its time to back up abit and ask some simple questions:  F 1. Did you use a valid e-mail address, with no mail filter of any sort on your mailbox?  L 2. Did you enter a avalid DECUS membership # (I assume you did, cos you will$ get an error message if you don't!)?  @ 3. What was the response message after you submitted the form?    H There are alot of people here, myself included, who genuinely care aboutM OpenVMS and the encourage the growth of the VMS hobbyist community, somethingi@ that took many years of effor to establish. Help us to help you.   Cheers,f
 Chris Guthreyc   ------------------------------    Date: 18 Mar 2003 04:23:13 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)E7 Subject: Re: Is http://www.montagar.com/hobbyist/ down?g3 Message-ID: <IHgyBQ1ymSY0@eisner.encompasserve.org>p  [ In article <3E768FD8.C3AF5B46@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:m > Baby Peanut wrote:	 >> [snip] G >> Honey, I got the CD-ROM and installed OpenVMS and filled out the webaI >> site to get the PAKs and waited a month and filled out the web site touI >> get the PAKs again and waited and waited.  If I went to Sun to buy thesH >> $20 copy of Solaris 9 on IA-32 they would have shipped me something IF >> could use.  Instead I'm out $30 and have no toy VMS to show for it. >  > Well, Dearie, ...l  H > Getting the PAKs depends on a lot of stuff that is beyond your control > and that of Montagar..  @ In particular, Baby Peanut may be thwarted forever by an onerousB Montagar/DECUS/Encompass/HP/Carly Fiorina/ meaningless bureacratic requirement.  @ I have never tried to get a hobbyist license, but I bet the form& has a few unanswerable questions like:   	Name:   ------------------------------  % Date: Tue, 18 Mar 2003 10:50:19 +00001' From: Andrew Harrison SUNUK Consultancyg7 Subject: Re: Is http://www.montagar.com/hobbyist/ down?n. Message-ID: <3E76F9EB.1040807@nospamn.sun.com>   JF Mezei wrote:. > Fred Kleinsorge wrote: > . >>Ah, Baby Peanut must be an Alias for Andrew. >  > M > Andrew may be your ennemy with whom you like to have catfights, but I thinkIO > that Andrew has more class than some silly peanut posts. He has not exhibited M > any motivation to blast decus/montagar in the past and likes to just insultbO > Digital with NUMA memory underperformance, Wildfire underperformance etc etc.e  5 I dislike anonymous posting, never trust a poster who , could be backing both sides of the argument.   Regardss Andrew Harrisone   ------------------------------  % Date: Tue, 18 Mar 2003 09:52:06 -0500uA From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com>i7 Subject: Re: Is http://www.montagar.com/hobbyist/ down? , Message-ID: <3e773298_1@hpb10302.boi.hp.com>  J Lighten up.  It was the reference to Sun that makes me wonder if this is a new or old Sun ringer.    = "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in messageo) news:3E7660D4.43CFBCB4@vl.videotron.ca.... > Fred Kleinsorge wrote: > >h0 > > Ah, Baby Peanut must be an Alias for Andrew. > G > Andrew may be your ennemy with whom you like to have catfights, but Id thinkhE > that Andrew has more class than some silly peanut posts. He has noto	 exhibitedlF > any motivation to blast decus/montagar in the past and likes to just insultJ > Digital with NUMA memory underperformance, Wildfire underperformance etc etc.   ------------------------------  % Date: Tue, 18 Mar 2003 15:07:20 +0000a' From: Andrew Harrison SUNUK Consultancy 7 Subject: Re: Is http://www.montagar.com/hobbyist/ down?p, Message-ID: <3E773628.60302@nospamn.sun.com>   Fred Kleinsorge wrote:L > Lighten up.  It was the reference to Sun that makes me wonder if this is a > new or old Sun ringer. >   5 Don't worry I wasn't taking your suggestion seriouslyo8 my point was more that I am not enamoured with anonymous posting in general.r   Regardst Andrew Harrisoni   > ? > "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> wrote in messageS+ > news:3E7660D4.43CFBCB4@vl.videotron.ca...t >  >>Fred Kleinsorge wrote: >>/ >>>Ah, Baby Peanut must be an Alias for Andrew.t >>G >>Andrew may be your ennemy with whom you like to have catfights, but Ia >  > think  > E >>that Andrew has more class than some silly peanut posts. He has not- >  > exhibited  > F >>any motivation to blast decus/montagar in the past and likes to just >  > insult > J >>Digital with NUMA memory underperformance, Wildfire underperformance etc >  > etc. >  >    ------------------------------  % Date: Tue, 18 Mar 2003 16:30:10 +00005 From: Roy Omond <Roy@Omond.net>.7 Subject: Re: Is http://www.montagar.com/hobbyist/ down?0) Message-ID: <3E774991.83ACE14E@Omond.net>C   Larry Kilgallen wrote:   > [...snip...] >tB > In particular, Baby Peanut may be thwarted forever by an onerousD > Montagar/DECUS/Encompass/HP/Carly Fiorina/ meaningless bureacratic > requirement. >tB > I have never tried to get a hobbyist license, but I bet the form( > has a few unanswerable questions like: >  >         Name:)  = I don't understand why people are assuming that "Baby Peanut"i? is *not* his/her real name :-)  I thought that would be quite ao common name in the USofA ...  	 Roy Omonds Blue Bubble Ltd.   ------------------------------    Date: 18 Mar 2003 18:08:09 -0000= From: Doc.Cypher <Use-Author-Supplied-Address-Header@[127.1]>t7 Subject: Re: Is http://www.montagar.com/hobbyist/ down?h6 Message-ID: <20030318180809.28456.qmail@gacracker.org>  > On 18 Mar 2003, Kilgallen@SpamCop.net (Larry Kilgallen) wrote:  A >I have never tried to get a hobbyist license, but I bet the formt' >has a few unanswerable questions like:R >H >	Name:n   Er, Yes. :)4  D But... I've no complaints about how they've looked after my privacy.     Doc.   ------------------------------    Date: 18 Mar 2003 11:48:00 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)r7 Subject: Re: Is http://www.montagar.com/hobbyist/ down?w3 Message-ID: <KfnrxAgIkBER@eisner.encompasserve.org>l  K In article <3E774991.83ACE14E@Omond.net>, Roy Omond <Roy@Omond.net> writes:U > Larry Kilgallen wrote: >  >> [...snip...]  >>C >> In particular, Baby Peanut may be thwarted forever by an onerousoE >> Montagar/DECUS/Encompass/HP/Carly Fiorina/ meaningless bureacratic  >> requirement.p >>C >> I have never tried to get a hobbyist license, but I bet the formp) >> has a few unanswerable questions like:b >> >>         Name: > ? > I don't understand why people are assuming that "Baby Peanut"4A > is *not* his/her real name :-)  I thought that would be quite ar > common name in the USofA ...  * Are we sure "Baby Peanut" is from the US ?; I was presuming it was from whatever your country is... :-)u   ------------------------------    Date: 18 Mar 2003 11:56:21 -0600+ From: kuhrt@encompasserve.org (Marty Kuhrt)n9 Subject: Is the RCA jack on a SN-PBXGK-AB (4D10T) usable?.3 Message-ID: <wC4NHNND3l9M@eisner.encompasserve.org>r  < I have a SN-PBXGK-AB, or 4D10T, card in my PWS500au.  On the? external part of the card, along with the standard VGA plug, is : an RCA video jack and a what looks to be a S-video plug.    @ Would it be possible to plug some video source into the RCA jack> and view or capture that video?  I have an X10 camera that I'd< like to use to capture snapshots and put on a webpage (being@ served by another node in the cluster running CSWS).  Naturally, I'd like an all VMS solution.n  < The box is running VMS V7.3-1 with all the relevant patches.   ------------------------------  % Date: Tue, 18 Mar 2003 12:05:26 -0000,* From: "Richard Brodie" <R.Brodie@rl.ac.uk>@ Subject: Re: Issue building Perl after applying VMS73_ACRTL V3.0+ Message-ID: <b57226$m4u@newton.cc.rl.ac.uk>   H "Charlie McCutcheon" <charlie.mccutcheonspam@nothp.com> wrote in message# news:3E7609EE.755DA211@nothp.com...1  A > >the problem arises. IMHO the CRTL ECOs are definitely "installeG > >with caution", and on a "better safe than sorry basis" I would leavee >bD > Sorry you feel that way.  This particular kit was designed to have) > minimal changes from previous releases.e  L It wasn't this particular kit that I got burnt by. I was just observing thatG small changes in the behaviour of a shared library, particularly one asEJ ubiquitous as the CTRL have a lot of potential for third-party applicationI bustage. More so than inner mode stuff which, by its nature, exposes lessoJ to the user . But the statistics are too poor to draw conclusions with any degree of confidence. YMMV.p   ------------------------------  # Date: Tue, 18 Mar 2003 17:26:35 GMTi< From: "Brad McCusker" <Brad.remove.McCuskerNOSP@MCompaq.com>@ Subject: Re: Issue building Perl after applying VMS73_ACRTL V3.00 Message-ID: <fFIda.487$TC2.137@news.cpqcorp.net>   --G "Craig A. Berry" <craig.berry@SignalTreeSolutions.com> wrote in messageh7 news:7f15589f.0303171947.2aae9de9@posting.google.com...y >oH > Having said that, there are a couple of small nits I'd like to mention > in this context. >vH > 1.)  The VMS73_ACRTL V3.0 does, as you note, document the new functionG > in its release notes and in its README.  The VMS731_ACRTL V1.0, which H > also adds poll() to the library, does not mention it in the README.  IG > think most folks will use primarily the README in deciding whether tol > apply the patch.  G Well how about that!  You are absolutely correct, it is not there.  TheoL strange thing is that it is there in the document in our archive.  SomewhereK between Engineering and Web, it got removed.  Thanks for pointing that out,./ looks like we got a process issue to deal with.r   >cE > 2.) The VMS731_ACRTL V1.0 release notes document the use of poll asgG > you show above, including the "#include <poll.h>" to get the function G > prototype and associated constants.  However, the ECO does not supplyiH > the poll.h header (and yes, I've looked in the text library as well asD > in the reference directory).  Using the function with a home-grownG > header is a good way to get in trouble (he said, as he goes off to do, > so). >u  I Blush... Yep, the poll.h header was inadvertantly ommitted from the finalsK .TLB that shipped with those ECOs.  Its in there now, and will be in futureb& ECOs and also the future VMS releases.  
 Brad McCusker  OpenVMS Engineeringr C RTL Project Leader Hewlett-Packard Companyo Nashua, NH USA   ------------------------------  % Date: Tue, 18 Mar 2003 10:49:24 +0100 3 From: "Negrita" <negrita.news.invalid@web2news.net>-, Subject: Layered Products for open VMS 7.2-1$ Message-ID: <19821N964@web2news.com>  G Can anyone tell me where I can find all the layered products on OpenVms6- 7.2-1. any help would be greatly appreciated.2 -- h4 Direct access to this group with http://web2news.com  http://web2news.com/?comp.os.vms   ------------------------------  % Date: Tue, 18 Mar 2003 11:12:09 +0000u( From: Nic Clews <sendspamhere@127.0.0.1>0 Subject: Re: Layered Products for open VMS 7.2-1) Message-ID: <3E76FF09.823C1E42@127.0.0.1>t   Negrita wrote: > I > Can anyone tell me where I can find all the layered products on OpenVms / > 7.2-1. any help would be greatly appreciated.w  	 Try this:C  : http://www.openvms.compaq.com/openvms/os/swroll/index.html   -- i? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencese nclews at csc dot comn   ------------------------------    Date: 18 Mar 2003 07:23:39 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 3 Subject: Re: National Moratorium to Appease Tyrantss3 Message-ID: <lsxwcdLIrf4K@eisner.encompasserve.org>i  q In article <4oWrZj+UPtI6@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:ha > In article <p9Gdncl-EP9cZuijXTWcpQ@metrocast.net>, "Bill Todd" <billtodd@metrocast.net> writes:e >> wO >> I was careful not to generalize excessively above.  But the preponderance ofbL >> reaction (admittedly not a full, balanced sample) reported in the article6 >> suggests that many, quite possibly a majority, are. > < >    The local station's polls show a clear majority.  Sigh. > I >    Worse, the next DJ described pulling the Chick's songs as "defendingc> >    the constituion".  I wonder who's constitution he's read?  E    The tide turned over the course of the day.  The majority wants totE    here the Chicks do Traveling Soldier.  Last night on the way home yE    when the song was scheduled to play "if yoiu want it", I was happy F    to hear it come on, made me smile.  This morning the DJ's are stillE    trying to stir controversy, but the song was playing when I turnedaH    the radio on, and now they're only critcizing their disagreement withG    what Natalie said and the timing of it.  It's OK with me if the DJ's A    feel that way and want to say so, they're free Americans, too.    ------------------------------    Date: 18 Mar 2003 07:25:26 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)t3 Subject: Re: National Moratorium to Appease Tyrantse3 Message-ID: <WbPT8KjBLp+J@eisner.encompasserve.org>b  p In article <ppqda.274238$UXa.247598@news02.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes:  & > Why the Republic of Texas of course.; > Texas IS the most important state in the Union, isn't it?hC > It certainly is the leading state....when it comes to executions.n@ > Guess that's why George finds it so easy to sleep knowing thatA > potentially hundreds of thousands of Iraqi's will be dead 'real C > soon' - he's been through execution watches before and it's nevere > seemed to bother him then.  F    I have family in Texas.  They are glad to be rid of him (OBTW, they.    are in be means liberal in their politics).  E    Texas' biggest worry is that some day Alaska will split in two andi-    Texas will become the third largest state.h   ------------------------------  % Date: Tue, 18 Mar 2003 10:21:41 -0500 ; From: "Brian Tillman" <Brian.Tillman@notnoone.notnohow.com>r3 Subject: Re: National Moratorium to Appease TyrantsD$ Message-ID: <3e773983$1@news.si.com>  I >I get the distinct impression that the media outlets are simply ignoringg allsJ >"peace" letters and count only the "lets roll" letters. And I since tuned outpJ >CNN as a reliable source of news. If CNN has lost "peaceful" watchers andL >retained only the "pro Bush ones", then it increases their justification toF >have a pro-war bias since it caters to their remaining viewers. (self selection).   D What???  The network that the conservatives call the "Communist NewsG Network" doesn't _really_ have a liberal bias??  What will they find toi criticize now? -- aI Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot come5 Smiths Aerospace  Addresses modified to prevent SPAM.e@ 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991r8        This opinion doesn't represent that of my company   ------------------------------  % Date: Tue, 18 Mar 2003 10:32:47 -0500S; From: "Brian Tillman" <Brian.Tillman@notnoone.notnohow.com>m3 Subject: Re: National Moratorium to Appease Tyrantsh$ Message-ID: <3e773c1b$1@news.si.com>  G >Again, if the US had any hard evidence that Iraq did in fact have WMD, F >what's the issue of giving the UN inspectors the intimate details andE >then letting the inspectors 'unearth' the truth?  Blix suspects that 6 >the UN is not getting the straight goods from the US.  K My local paper on Sunday revealed that some of the documents that the WhiterL House is using to "prove" that Iraq has a "nookyooler" program are, in fact,% forgeries.  My question is "by whom?"n -- cI Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot comu5 Smiths Aerospace  Addresses modified to prevent SPAM.h@ 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991t8        This opinion doesn't represent that of my company   ------------------------------  % Date: Tue, 18 Mar 2003 09:23:01 +0100 6 From: =?iso-8859-1?Q?Jean=2DFran=E7ois=20PI=C9RONNE?= ) Subject: new releases of MySQL and Python + Message-ID: <3E76D765.A46BD358@laposte.net>t  M I have updated the port of MySQL, the current version is now 4.0.11A which is  the lastest 4.0 version.  K A source kit can be download at http://www.pi-net.dyndns.org/anonymous/jfp/s  * C, C++ and MMS is needed to build the kit.  N I have received some patches for C 6.0 and C++ 6.2 which are not include but I can provide them separately. r  N I also suggest to try MySQL Control Center which appear to work nicely with an VMS MySQL server.DG It can be download from the official MySQL site (http://www.mysql.com/)     L The python kit is back, the licensing problem is solved (all the code of the  previous port has been removed).> Some new modules are present, for example the mySQL interface.G Building instruction and download pointer: http://vmspython.dyndns.org/o   Jean-Franois Pironne   ps.uL There is a bug into vmstar which don't handle correctly long pathname (> 100N characters), if someone want to fix this he is welcome, fortunately the Python! tarfile module work correctly :-)    ------------------------------    Date: 18 Mar 2003 11:10:34 +0100C From: vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann)f- Subject: Re: new releases of MySQL and Pythone- Message-ID: <3e76f09a$1@news.uni-konstanz.de>e  + In article <3E76D765.A46BD358@laposte.net>,sJ =?iso-8859-1?Q?Jean=2DFran=E7ois=20PI=C9RONNE?=  <jf.pieronne@laposte.net> writes: F |>I have updated the port of MySQL, the current version is now 4.0.11A
 |>which is |>the lastest 4.0 version. |>! |>A source kit can be download at - |>http://www.pi-net.dyndns.org/anonymous/jfp/n |>, |>C, C++ and MMS is needed to build the kit. |>J |>I have received some patches for C 6.0 and C++ 6.2 which are not include |>but Ih |>can provide them separately.   |>H |>I also suggest to try MySQL Control Center which appear to work nicely	 |>with an. |>VMS MySQL server.oI |>It can be download from the official MySQL site (http://www.mysql.com/)p |> |>J |>The python kit is back, the licensing problem is solved (all the code of |>the " |>previous port has been removed).@ |>Some new modules are present, for example the mySQL interface.I |>Building instruction and download pointer: http://vmspython.dyndns.org/i |> |>Jean-Franois Pironne |> |>ps.eJ |>There is a bug into vmstar which don't handle correctly long pathname (> |>100iI |>characters), if someone want to fix this he is welcome, fortunately theB |>Python# |>tarfile module work correctly :-)- |>  $ You'll need ODS5-filesystem, right??   eberhard   ------------------------------  % Date: Tue, 18 Mar 2003 13:10:23 +0000n' From: Andrew Harrison SUNUK ConsultancyeY Subject: Re: OpenVMS.org: Marvel article and HP's press release for   MarvelandAlpha Reta . Message-ID: <3E771ABF.5070403@nospamn.sun.com>   Nic Clews wrote:* > Andrew Harrison SUNUK Consultancy wrote: >  >>Fred Kleinsorge wrote: >>N >>>"Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com>< >>>wrote in message news:3E759E72.7060909@nospamn.sun.com... >>>5 >> > C >>>>Most NUMA systems are also SMP systems as well. The GS160/320'se >>>>are SMP. >>>> >>>vO >>>No.  The S in SMP is for Symetric.  It isn't.  It is a MP system - a NUMA MPaM >>>system, more specific - a cache coherent, directory based, NUMA MP system..D >>>It many respects it may function (and can be treated) much like aH >>>traditional SMP system, but because of the asymetric nature of memoryM >>>access, there are real differences that the OS must take into account, and O >>>which *some* applications will benefit in performance if they also are aware O >>>that all CPUs may not be equal (in access to portions of memory).  It *is* ae0 >>>different beast from true classic SMP design. >>>e >>F >>SMP stands for symetric multi-processing, it refers access that each- >>CPU has to memory I/O kernel resources etc.p >>E >>You have a directory but all the CPU's from a scheduling standpointo7 >>have equal access to all the resources in the system.  >  > F > Symmetric is derived from the word "symmetry", "symmetrical" so I doF > understand what Fred is getting at. This is from a hardware point ofI > view. You can have symmetry in the quad itself, or about an even numberbJ > of quads, but there is no symmetry between what is in a quad and what is
 > outside it.e >   ? No you have no symmetry from a memory access latency standpoint@> but you do from a process scheduling standpoint. All processes< and the kernel can be equally scheduled on any of the CPU's.  : ASMP one of the alternatives is where there is no equality9 from a scheduling standpoint, for example the kernel runsr8 on one of the CPU's and the user level processes/threads on the others.  @ Of course most NUMA OS's have scheduler hints that try to ensure> that processes are scheduled on the quad where the memory that@ the process wants to access it located. How much benefit you get? from this depends on how NUMA the system is. VNUMA systems liket@ the GS160/320 and the Sequent NUMA-Q may well see a big benefits  if this hint works consistently.  @ But it is only a hint in and practice any process can run on any processor and may have to.  D The issue with something like the Oracle SGA is that you have commonC pool of memory which all the processes need to access so you end up A having to take the remote latency hit anyway. This applies to any C software that uses a large chunk of shared memory for inter process $ communication, caching, locking etc.  = You can of course turn these hints into a hard binding or use ; OPS/RAC in a box which does the same thing, it also has thehC big benefit on VNUMA systems of giving you an SGA per node reducingW: the inter node traffic to locking and recovery but you can' also do the former on most SMP systems.t  E > On the other hand, as far as the operating system is concerned, itsaJ > unadulterated view is of  a potentially symmetrical system, so Andrew isH > right too. You could have just a single processor in each quad, and in" > my eyes I'd call that symmetric. > H > It all greys considerably when thinking about what a RAD as a softwareJ > environment does in a NUMA hardware environment. It is a bit like coming@ > off the M6 southbound at junction 13, and heading down throughG > Wolverhampton (ring road) to rejoin the M5. Some of that road is dual.J > carriageway, and potentially may move faster than the motorway, but whenH > you're off the motorway, if it starts moving faster, you're limited to* > the urban speed limits of the ring road. > 2 > This analogy only works in the UK about 8.00 AM. >    ------------------------------  % Date: Tue, 18 Mar 2003 10:13:20 -0500CA From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> Y Subject: Re: OpenVMS.org: Marvel article and HP's press release for   MarvelandAlpha Retat, Message-ID: <3e773791_2@hpb10302.boi.hp.com>  J Andy, have you ever done low-level work?  Wanna tell us about local versusL far memory and lock starvation?  Even for a system with a "small" NUMA cost?J How about data replication because of the latency cost?  Even on a "small"( NUMA cost machine it can be significant.  G You can try to paper over the NUMA-ness of a system and pretend it is aiI classic SMP system.  And for the most part applications can do just that. I The OS is jumping through the hoops to try to present an environment thatdH functions much like that "classic" SMP box.  But for large applications,I with critical performance requirements *even* a "small" NUMA cost machine L will not function as they expect, and they *can* change their application to! know about NUMA aspects (or not).b      K "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com>w9 wrote in message news:3E771ABF.5070403@nospamn.sun.com...c >e >  > Nic Clews wrote:, > > Andrew Harrison SUNUK Consultancy wrote: > >i > >>Fred Kleinsorge wrote: > >>( > >>>"Andrew Harrison SUNUK Consultancy"' <Andrew_No.Harrison_No@nospamn.sun.com>t> > >>>wrote in message news:3E759E72.7060909@nospamn.sun.com... > >>>s > >> > > E > >>>>Most NUMA systems are also SMP systems as well. The GS160/320'sS > >>>>are SMP. > >>>> > >>> I > >>>No.  The S in SMP is for Symetric.  It isn't.  It is a MP system - ae NUMA MPsG > >>>system, more specific - a cache coherent, directory based, NUMA MPh system.oF > >>>It many respects it may function (and can be treated) much like aJ > >>>traditional SMP system, but because of the asymetric nature of memoryK > >>>access, there are real differences that the OS must take into account,b and K > >>>which *some* applications will benefit in performance if they also arer awarerJ > >>>that all CPUs may not be equal (in access to portions of memory).  It *is* a2 > >>>different beast from true classic SMP design. > >>>d > >>H > >>SMP stands for symetric multi-processing, it refers access that each/ > >>CPU has to memory I/O kernel resources etc.  > >>G > >>You have a directory but all the CPU's from a scheduling standpointt9 > >>have equal access to all the resources in the system.- > >- > >-H > > Symmetric is derived from the word "symmetry", "symmetrical" so I doH > > understand what Fred is getting at. This is from a hardware point ofK > > view. You can have symmetry in the quad itself, or about an even numbertL > > of quads, but there is no symmetry between what is in a quad and what is > > outside it.t > >a >bA > No you have no symmetry from a memory access latency standpoint @ > but you do from a process scheduling standpoint. All processes> > and the kernel can be equally scheduled on any of the CPU's. >m< > ASMP one of the alternatives is where there is no equality; > from a scheduling standpoint, for example the kernel runss: > on one of the CPU's and the user level processes/threads > on the others. >oB > Of course most NUMA OS's have scheduler hints that try to ensure@ > that processes are scheduled on the quad where the memory thatB > the process wants to access it located. How much benefit you getA > from this depends on how NUMA the system is. VNUMA systems likehB > the GS160/320 and the Sequent NUMA-Q may well see a big benefits" > if this hint works consistently. >aB > But it is only a hint in and practice any process can run on any > processor and may have to. >eF > The issue with something like the Oracle SGA is that you have commonE > pool of memory which all the processes need to access so you end upcC > having to take the remote latency hit anyway. This applies to anyrE > software that uses a large chunk of shared memory for inter processd& > communication, caching, locking etc. >h? > You can of course turn these hints into a hard binding or useb= > OPS/RAC in a box which does the same thing, it also has theeE > big benefit on VNUMA systems of giving you an SGA per node reducingl< > the inter node traffic to locking and recovery but you can) > also do the former on most SMP systems.d > G > > On the other hand, as far as the operating system is concerned, its*L > > unadulterated view is of  a potentially symmetrical system, so Andrew isJ > > right too. You could have just a single processor in each quad, and in$ > > my eyes I'd call that symmetric. > >sJ > > It all greys considerably when thinking about what a RAD as a softwareL > > environment does in a NUMA hardware environment. It is a bit like comingB > > off the M6 southbound at junction 13, and heading down throughI > > Wolverhampton (ring road) to rejoin the M5. Some of that road is dual0L > > carriageway, and potentially may move faster than the motorway, but whenJ > > you're off the motorway, if it starts moving faster, you're limited to, > > the urban speed limits of the ring road. > >d4 > > This analogy only works in the UK about 8.00 AM. > >  >e   ------------------------------  % Date: Tue, 18 Mar 2003 16:59:06 +0000o' From: Andrew Harrison SUNUK ConsultancyiY Subject: Re: OpenVMS.org: Marvel article and HP's press release for   MarvelandAlpha Retab, Message-ID: <3E77505A.20605@nospamn.sun.com>   Fred Kleinsorge wrote:L > Andy, have you ever done low-level work?  Wanna tell us about local versusN > far memory and lock starvation?  Even for a system with a "small" NUMA cost?L > How about data replication because of the latency cost?  Even on a "small"* > NUMA cost machine it can be significant. >    Yes   4 The problem that you are trying to put you finger on4 is of your own making. VNUMA (GS320, Sequent NUMA-Q)2 has its own special set of issues which are caused  by the very long remote latency.  2 SunFire servers are also NUMA systems but they are small NUMA not VNUMA.s  2 And we do have local/remote memory optimisation in2 Solaris its called MPO however the majority of our/ benchmark results for the F series pre-date thee introduction of MPO.  I > You can try to paper over the NUMA-ness of a system and pretend it is atK > classic SMP system.  And for the most part applications can do just that.iK > The OS is jumping through the hoops to try to present an environment that-J > functions much like that "classic" SMP box.  But for large applications,K > with critical performance requirements *even* a "small" NUMA cost machineRN > will not function as they expect, and they *can* change their application to# > know about NUMA aspects (or not).u >   ; As I said nearly all SunFire benchmarks have been publishedr? without MPO (it was only introduced for the F15K's in February) > late last year for F3800-F6800 and MPO does make a performance> difference but it isn't as critical as you imply since without> it we were getting perfectly respectable thoughput results for
 the machines.   = And none of the pre-MPO results have had to resort to RAC/OPSl; or any other internal partitioning scheme in a box in order # to hide the NUMAness of the system.v  < You may well have to jump though hoops but don't assume that$ this is the normal state of affairs.   Regardst Andrew Harrison    ------------------------------  % Date: Tue, 18 Mar 2003 10:39:33 +0000e' From: Andrew Harrison SUNUK Consultancy Y Subject: Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha RetaiD. Message-ID: <3E76F765.4030601@nospamn.sun.com>   Fred Kleinsorge wrote:M > "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com>v; > wrote in message news:3E759E72.7060909@nospamn.sun.com...e >  >> >>Main, Kerry wrote: >>
 >>>Andrew, >>>.E >>><<< In almost all real scenarios rather than TPC-C  and the otherskF >>>mentioned below. Clustering for scalability will cost you more thanI >>>buying a larger SMP server and will give you more variable results.<<<y >>> J >>>Please .. SMP servers have scaling issues due to cpu cache invalidationK >>>issues and NUMA servers have issues if memory locality is not taken intor >>>consideration.  >>>- >>6 >>Please try not to be too condescending, it makes the0 >>fact that you are getting rather confused over8 >>terminology more rather than less embarassing for you. >>A >>Most NUMA systems are also SMP systems as well. The GS160/320's5
 >>are SMP. >> >  > N > No.  The S in SMP is for Symetric.  It isn't.  It is a MP system - a NUMA MPL > system, more specific - a cache coherent, directory based, NUMA MP system.C > It many respects it may function (and can be treated) much like acG > traditional SMP system, but because of the asymetric nature of memoryeL > access, there are real differences that the OS must take into account, andN > which *some* applications will benefit in performance if they also are awareN > that all CPUs may not be equal (in access to portions of memory).  It *is* a/ > different beast from true classic SMP design.o > D SMP stands for symetric multi-processing, it refers access that each+ CPU has to memory I/O kernel resources etc.e  C You have a directory but all the CPU's from a scheduling standpointi5 have equal access to all the resources in the system.r   Regards  Andrew Harrisone   ------------------------------  % Date: Tue, 18 Mar 2003 10:07:13 -0500tA From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> Y Subject: Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha Retaio, Message-ID: <3e773622_1@hpb10302.boi.hp.com>  K "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com>s9 wrote in message news:3E76F765.4030601@nospamn.sun.com...n >a >n > Fred Kleinsorge wrote:' > > "Andrew Harrison SUNUK Consultancy"h' <Andrew_No.Harrison_No@nospamn.sun.com>y= > > wrote in message news:3E759E72.7060909@nospamn.sun.com...I > >e > >> > >>Main, Kerry wrote: > >> > >>>Andrew, > >>>nG > >>><<< In almost all real scenarios rather than TPC-C  and the others H > >>>mentioned below. Clustering for scalability will cost you more thanK > >>>buying a larger SMP server and will give you more variable results.<<<e > >>>pL > >>>Please .. SMP servers have scaling issues due to cpu cache invalidationH > >>>issues and NUMA servers have issues if memory locality is not taken into > >>>consideration.l > >>>d > >>8 > >>Please try not to be too condescending, it makes the2 > >>fact that you are getting rather confused over: > >>terminology more rather than less embarassing for you. > >>C > >>Most NUMA systems are also SMP systems as well. The GS160/320'sr > >>are SMP. > >> > >m > >gH > > No.  The S in SMP is for Symetric.  It isn't.  It is a MP system - a NUMA MPiF > > system, more specific - a cache coherent, directory based, NUMA MP system.EE > > It many respects it may function (and can be treated) much like arI > > traditional SMP system, but because of the asymetric nature of memorysJ > > access, there are real differences that the OS must take into account, andhJ > > which *some* applications will benefit in performance if they also are awareII > > that all CPUs may not be equal (in access to portions of memory).  It  *is* a1 > > different beast from true classic SMP design.a > >vF > SMP stands for symetric multi-processing, it refers access that each- > CPU has to memory I/O kernel resources etc.o >cE > You have a directory but all the CPU's from a scheduling standpointh7 > have equal access to all the resources in the system./ >l  F Nonesense.  It is not symetric.  If it were, all CPUs would have equalK access to all memory, and they don't.  It is not something they can ignore.lK From a scheduling standpoint, you now want to consider near and far memory.:L From a lock standpoint you need to consider locality, and the potential needJ for "fair" locks to prevent starvation.  From a performance standpoint youL need to look at page replication - at least of kernel code segments - if not> read only data, or even read/write data (with fault on write).  F The OS might mitigate the effects, and some applications can choose toE ignore the NUMA aspects of the platform.  But for large multithreadedaA applications - it can be a bad performance decision to ignore it.   I A NUMA system is a distinct design from the traditional SMP design.  MostcF people will, and do draw a distinction between NUMA and an unqualified- statement that the system is an "SMP" design.    ------------------------------  % Date: Tue, 18 Mar 2003 17:06:00 +00003' From: Andrew Harrison SUNUK Consultancy Y Subject: Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha Retait. Message-ID: <3E7751F8.2040704@nospamn.sun.com>   Fred Kleinsorge wrote:M > "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com>e; > wrote in message news:3E76F765.4030601@nospamn.sun.com...  >> >  > H > Nonesense.  It is not symetric.  If it were, all CPUs would have equalM > access to all memory, and they don't.  It is not something they can ignore.mM > From a scheduling standpoint, you now want to consider near and far memory.cN > From a lock standpoint you need to consider locality, and the potential needL > for "fair" locks to prevent starvation.  From a performance standpoint youN > need to look at page replication - at least of kernel code segments - if not@ > read only data, or even read/write data (with fault on write). >   @ Rubbish this is just a scheduler hint, it isn't a hard boundary.? ASMP is a hard boundary. Some processors run some processes and>= some cannot, the OS tuning in OpenVMS and Tru64 does not make  this kind of restriction.   = The E10K is a UMA SMP machine but the scheduler has processort> affinity, this is where the scheduler tries to re-schedule theA process on the CPU that last ran it.  In your book this would notr2 be an SMP machine but everyone excepts that it is.  H > The OS might mitigate the effects, and some applications can choose toG > ignore the NUMA aspects of the platform.  But for large multithreaded C > applications - it can be a bad performance decision to ignore it.a > K > A NUMA system is a distinct design from the traditional SMP design.  MostlH > people will, and do draw a distinction between NUMA and an unqualified/ > statement that the system is an "SMP" design.  >   ? No it isn't because we are talking about two entirely different ; things. Your distinction is really between what you need to-< do from an OS tuning standpoint for a UMA machine vs an NUMA< machine, they are still SMP systems unless you can never run< a process on an arbitrary processor and you can only run the kernel on selected processors.   Regardsp Andrew Harrisonr   ------------------------------  % Date: Tue, 18 Mar 2003 12:45:05 +0000S( From: Nic Clews <sendspamhere@127.0.0.1>Y Subject: Re: OpenVMS.org: Marvel article and HP's press release for MarvelandAlpha Retainh) Message-ID: <3E7714D1.FFEC0452@127.0.0.1>l  ( Andrew Harrison SUNUK Consultancy wrote: >  > Fred Kleinsorge wrote:O > > "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com> = > > wrote in message news:3E759E72.7060909@nospamn.sun.com.... > >-  C > >>Most NUMA systems are also SMP systems as well. The GS160/320's  > >>are SMP. > >> > >pP > > No.  The S in SMP is for Symetric.  It isn't.  It is a MP system - a NUMA MPN > > system, more specific - a cache coherent, directory based, NUMA MP system.E > > It many respects it may function (and can be treated) much like a I > > traditional SMP system, but because of the asymetric nature of memorysN > > access, there are real differences that the OS must take into account, andP > > which *some* applications will benefit in performance if they also are awareP > > that all CPUs may not be equal (in access to portions of memory).  It *is* a1 > > different beast from true classic SMP design.o > > F > SMP stands for symetric multi-processing, it refers access that each- > CPU has to memory I/O kernel resources etc.n > E > You have a directory but all the CPU's from a scheduling standpointt7 > have equal access to all the resources in the system.o  D Symmetric is derived from the word "symmetry", "symmetrical" so I doD understand what Fred is getting at. This is from a hardware point ofG view. You can have symmetry in the quad itself, or about an even number>H of quads, but there is no symmetry between what is in a quad and what is outside it.   C On the other hand, as far as the operating system is concerned, itsmH unadulterated view is of  a potentially symmetrical system, so Andrew isF right too. You could have just a single processor in each quad, and in  my eyes I'd call that symmetric.  F It all greys considerably when thinking about what a RAD as a softwareH environment does in a NUMA hardware environment. It is a bit like coming> off the M6 southbound at junction 13, and heading down throughE Wolverhampton (ring road) to rejoin the M5. Some of that road is dualtH carriageway, and potentially may move faster than the motorway, but whenF you're off the motorway, if it starts moving faster, you're limited to( the urban speed limits of the ring road.  0 This analogy only works in the UK about 8.00 AM.   -- a? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------    Date: 18 Mar 2003 05:29:54 -08007 From: nigel.guest@partner.commerzbank.com (Nigel Guest)-! Subject: Returning Mouse Position0= Message-ID: <3c577827.0303180529.10d374e5@posting.google.com>    Hi,:  ? I am trying to return the mouse position and mouse key pressed.D  C To be more precise, i am using SMG and Basic (though most languagesu9 will do for this) on VMS 7.2-1, running under Reflection.2  E I can set up a VB routine running under Reflection, and set the mouseeE keys from inside reflection to run this routine, which sends a string F containing x and y co-ords.  The software can then pick up this stringC and work out what should be run, but this only works if a mouse keys (left or right) is clicked.I  D I would prefer to be able to find out the mouse position, regardlessD of whether a mouse button has been clicked, and i would prefer to doA this from within VMS if possible, with no changes required to thet reflection environment.h  ? If it is not possible to find out the position without clickinge% buttons, then I could live with that.    Am I asking for to much ?    ------------------------------  % Date: Tue, 18 Mar 2003 08:42:41 +0100r$ From: "Peter Flunger" <p-i-b@gmx.at>( Subject: Re: send mail from an intranet?0 Message-ID: <b56ili$obt$1@newsreader1.netway.at>  6 "Didier Morandi" <Didier.Morandi.nospam@Free.fr> wroteJ > ok, it works, but only if the destination is @nerim.net which is my ADSLD > ISP. Now I need to figure out why it does not accept to reroute my' > messages to other Internet addresses.l  , How does your general SMTP config look like." Do you use a specific SMTP gateway6 ( tcpip set conf smtp/gate=.. ) or do you let the SMTP? server decide where to send your messages to using MX queries ?)> ISPs more often than not are very specific about routing email via their SMTP gateways.< They either let you send email to <user>@isp-domain from any< other domain OR let you send emial from <user>@isp-domain to any other domain. < OpenVMS TCP/IP SMTP is capable of sending mail to the target SMTP gateway directly.   Petere   ------------------------------  % Date: Tue, 18 Mar 2003 09:54:21 +0100c4 From: Didier Morandi <Didier.Morandi.nospam@Free.fr>( Subject: Re: send mail from an intranet?& Message-ID: <3E76DEBD.8090104@Free.fr>   Peter Flunger wrote:  - >How does your general SMTP config look like.t# >Do you use a specific SMTP gatewayl7 >( tcpip set conf smtp/gate=.. ) or do you let the SMTP @ >server decide where to send your messages to using MX queries ?? >ISPs more often than not are very specific about routing email( >via their SMTP gateways.d= >They either let you send email to <user>@isp-domain from anyn= >other domain OR let you send emial from <user>@isp-domain to. >any other domain.= >OpenVMS TCP/IP SMTP is capable of sending mail to the target  >SMTP gateway directly.  >i1 DTL02> tcpip sh conf smtp                        p  # SMTP Configuration                 -J                                                                    OptionsL Initial interval:   0 00:30:00.00       Address_max:    16       NOEIGHT_BITH Retry interval:     0 01:00:00.00       Hop_count_max:  16       NORELAYL Maximum interval:   3 00:00:00.00                                TOP_HEADERS  G Timeout             Initial       Mail    Receipt       Data  TerminatetG   Send:                   5          5          5          3         10    Receive:                50  " Alternate gateway:  SMTP.NERIM.NET General gateway:    not definedu  # Substitute domain:  HIDDEN, FREE.FR7 Zone:               NERIM.NETe   Postmaster:         TCPIP$SMTPC Log file:           SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOGa  0 Generic queue       Queues   Participating nodes  " TCPIP$SMTP_DTL02_00    1     DTL02   ------------------------------  % Date: Tue, 18 Mar 2003 10:10:10 +0100 $ From: "Peter Flunger" <p-i-b@gmx.at>( Subject: Re: send mail from an intranet?0 Message-ID: <b56npk$qq8$1@newsreader1.netway.at>  6 "Didier Morandi" <Didier.Morandi.nospam@Free.fr> wrote$ > Alternate gateway:  SMTP.NERIM.NET! > General gateway:    not definedi There we go.< I am pretty sure, that SMTP.NERIM.NET behaves as i described in my previous posting..: Is your PWS capable of doing host queries to a working DNS server ?5 If so, you can remove the Alternate gateway from yourt7 configuration, because your system will do an MX lookupe4 and then deliver mail directly to the target system.7 For example, when you send an email to somebody@hp.com,t SMTP will do the equivalent of $TCPIP sho MX hp.com1 and as a result recive the following information:o+ Gate address     Preference       Gate nameh- 192.151.81.14    10               smtp.HP.COMs. 15.81.176.20     30               smtpx.HP.COM0 156.153.255.246  50               palsmtp.HP.COM0 156.153.255.205  50               atlsmtp.HP.COM  7 Then the SMTP software will try to open ab SMTP session 1 with the gateway with the smallest priority valuel5 ( in this case 192.151.81.14 ), if this fails it willo4 try the gateways in the order of ascending priority,4 if they all fail, the SMTP mail will be retained and retried at a later time.    8 If you do not have the possibility to query a DNS server: capable of returing valid MX records you will be forced to< use NERIM.NET as outgoing email domain ( in your substitute- domain setting of SMTP ).- Peter-   ------------------------------  % Date: Tue, 18 Mar 2003 10:42:49 +0100r4 From: Didier Morandi <Didier.Morandi.nospam@Free.fr>( Subject: Re: send mail from an intranet?& Message-ID: <3E76EA19.4020209@Free.fr>   Peter Flunger wrote:  7 >"Didier Morandi" <Didier.Morandi.nospam@Free.fr> wrotes >  , >o$ >>Alternate gateway:  SMTP.NERIM.NET! >>General gateway:    not defineds >>     >>
 >There we go.o= >I am pretty sure, that SMTP.NERIM.NET behaves as i describedd >in my previous posting.; >Is your PWS capable of doing host queries to a working DNSt	 >server ?w6 >If so, you can remove the Alternate gateway from your8 >configuration, because your system will do an MX lookup5 >and then deliver mail directly to the target system.m8 >For example, when you send an email to somebody@hp.com, >SMTP will do the equivalent of  >$TCPIP sho MX hp.comi2 >and as a result recive the following information:, >Gate address     Preference       Gate name. >192.151.81.14    10               smtp.HP.COM/ >15.81.176.20     30               smtpx.HP.COM11 >156.153.255.246  50               palsmtp.HP.COMn1 >156.153.255.205  50               atlsmtp.HP.COMa >i8 >Then the SMTP software will try to open ab SMTP session2 >with the gateway with the smallest priority value6 >( in this case 192.151.81.14 ), if this fails it will5 >try the gateways in the order of ascending priority, 5 >if they all fail, the SMTP mail will be retained andt >retried at a later time.r >i > 9 >If you do not have the possibility to query a DNS servert; >capable of returing valid MX records you will be forced ton= >use NERIM.NET as outgoing email domain ( in your substitute-  >domain setting of SMTP ). >Peter > @ I have changed the gateway and removed the MX entry. Now I have:   SMTP ConfigurationJ                                                                    OptionsL Initial interval:   0 00:30:00.00       Address_max:    16       NOEIGHT_BITH Retry interval:     0 01:00:00.00       Hop_count_max:  16       NORELAYL Maximum interval:   3 00:00:00.00                                TOP_HEADERS  G Timeout             Initial       Mail    Receipt       Data  TerminatenG   Send:                   5          5          5          3         10o   Receive:                5    Alternate gateway:  not defined> General gateway:    NERIM.NETa  % Substitute domain:  HIDDEN, DTL02.DTLD Zone:               NERIM.NET    Postmaster:         TCPIP$SMTPC Log file:           SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOGp  0 Generic queue       Queues   Participating nodes  " TCPIP$SMTP_DTL02_00    1     DTL02  6 I do not find the command to remove the Hidden domain.7 When I send mail to didier.morandi@free.fr, I get this:o (thanks to JF)   D %%%%%%%%%%%%                   18-MAR-2003 10:30:55.39  %%%%%%%%%%%%% %TCPIP-I-SMTP_LOGSUC, using log file c/ SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOGt   TCPIP SMTP configuration data:% Server-Nodes                  : DTL02 3 Queue-Name                    : TCPIP$SMTP_DTL02_00p Alternate-Gateway             :y) General-Gateway               : NERIM.NET ) Substitute-Domain             : DTL02.DTLr) Zone                          : NERIM.NET9* Postmaster-Alias              : TCPIP$SMTP& Postmaster-Forwards-To        : SYSTEM Foreign-Transport-Synonyms    :t- Initial-Interval              : 0 00:30:00.000- Retry-Interval                : 0 01:00:00.00o- Retry-Maximum                 : 3 00:00:00.00y! Receive-Timeout               : 5a" Retry-Address                 : 16" Hop-Count                     : 16! Symbiont-Snapshot-Blocks      : 0u! Receiver-Snapshot-Blocks      : 0l! Utilities-Snapshot-Blocks     : 0 ! Send-Timeout-Init             : 5 ! Send-Timeout-Mail             : 5 ! Send-Timeout-Rcpt             : 5l! Send-Timeout-Data             : 3 " Send-Timeout-Term             : 10! Log-Level                     : 2c! Receiver-Debug                : 0r! Receiver-Trace                : 0e! Symbiont-Debug                : 0.! Symbiont-Trace                : 1 ! Utilities-Debug               : 0g! Utilities-Trace               : 0u! EF-Debug-Level                : 0d! Channel-Debug-Level           : 0 # Header-Placement              : TOPt% Eightbit                      : FALSEh% Relay                         : FALSE % Altgate-Always                : FALSE % Mx-If-Noaltgate               : FALSEm% No-Mx                         : FALSEE% No-Subs-Domain-Inbound        : FALSEn$ Smtp-Jacket-Local             : TRUE$ Cent-Sign-Hack                : TRUE$ Nosey                         : TRUE% Log-Line-Numbers              : FALSEr% Memory-Debug                  : FALSEa% Mail$Protocol-Debug           : FALSEt% CF-Debug                      : FALSEs% Parse-Debug                   : FALSE % Deliver-VMS-Def-To            : FALSE % Deliver-NoXVMS                : FALSEs% MTS-From-Hack                 : FALSEM% Rewrite-MTS-From              : FALSE % Local-Alias-Only              : FALSEd% Relay-Based-On-Mx             : FALSEh% Reject-Unbacktranslatable-IP  : FALSE % Accept-Unqualified-Senders    : FALSEe% Accept-Unresolvable-Domains   : FALSEh% SFF-Requires-Priv             : FALSE % 8BitMIME-Hack                 : FALSEh  " Other TCPIP SMTP environment data:* SMTP Software Username        : TCPIP$SMTP9 SMTP Software Default Director: SYS$SPECIFIC:[TCPIP$SMTP]r  Symbiont Log File             : / SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOGi   List of (4) local aliases: DTL02h	 DTL02.DTLt	 localhosto
 localhost.dtln      D %%%%%%%%%%%%                   18-MAR-2003 10:30:57.87  %%%%%%%%%%%%H %TCPIP-I-SMTP_SYMBRUN, symbiont is running the queue TCPIP$SMTP_DTL02_01  ( 18-MAR-2003 10:31:10.97: next open file @ _DTL02$DKA0:[SYS0.SYSMGR]03031810311062_SYSTEM.TCPIP_DTL02;1(60)( recv buf=220 mrelay1-1.free.fr ESMTP\d\a send buf=HELO dtl02.dtl\d\a " recv buf=250 mrelay1-1.free.fr\d\a) send buf=MAIL FROM:<system@dtl02.dtl>\d\aeG recv buf=553 sorry, your envelope sender domain must exist (#5.7.1)\d\aa send buf=QUIT\d\a " recv buf=221 mrelay1-1.free.fr\d\a  / Printing debug_level 2, Domains and recipients:U Domain: free.frr,    Recipient address: DIDIER.MORANDI@FREE.FR    Domain part:       free.fr"$    Local part:        DIDIER.MORANDI9    Address Status:    Not done, not delivered. (Requeued)o  ) Printing debug_level 2, relevant headers:  Return-Path: system@dtl02.dtlS+ Date: Tue, 18 Mar 2003 10:31:10 +0100 (MET) & Message-Id: <03031810311062@dtl02.dtl> From: system@dtl02.dtl To: DIDIER.MORANDI@free.fr Subject: test alphar% X-VMS-To: SMTP%didier.morandi@free.fr @ A permanent error makes this mail undeliverable. Must bounce it.9 553  %TCPIP-E-SMTP_COMMANDERR, SMTP command error free.fr   / Printing debug_level 2, Domains and recipients:n Domain: free.fru,    Recipient address: DIDIER.MORANDI@FREE.FR    Domain part:       free.fro$    Local part:        DIDIER.MORANDI9    Address Status:    Not done, not delivered. (Requeued)y  ) Printing debug_level 2, relevant headers:c Return-Path: system@dtl02.dtl>+ Date: Tue, 18 Mar 2003 10:31:10 +0100 (MET)r& Message-Id: <03031810311062@dtl02.dtl> From: system@dtl02.dtl To: DIDIER.MORANDI@free.fr Subject: test alphaU% X-VMS-To: SMTP%didier.morandi@free.fre  ( 18-MAR-2003 10:31:14.70: next open file D _DTL02$DKA0:[SYS0.TCPIP$SMTP]03031810311361_SYSTEM.TCPIP_DTL02;1(64)F recv buf=220 brinstar.nerim.net ESMTP Sendmail 8.12.6/8.12.5; Tue, 18 
 Mar 2003 1 0:32:09 +0100 (CET)\d\ae send buf=HELO dtl02.dtl\d\a C recv buf=250 brinstar.nerim.net Hello didiermorandi.net1.nerim.net n
 [213.41.147.1u 37], pleased to meet you\d\a send buf=MAIL FROM:<>\d\ae& recv buf=250 2.1.0 <>... Sender ok\d\a. send buf=RCPT TO:<didiermorandi@nerim.net>\d\a@ recv buf=250 2.1.5 <didiermorandi@nerim.net>... Recipient ok\d\a send buf=DATA\d\a = recv buf=354 Enter mail, end with "." on a line by itself\d\am8 send buf=Date: Tue, 18 Mar 2003 10:31:13 +0100 (MET)\d\a3 send buf=Message-Id: <03031810311361@dtl02.dtl>\d\as' send buf=From: TCPIP$SMTP@DTL02.DTL\d\aS! send buf=To: system@dtl02.dtl\d\a # send buf=Subject: Returned mail\d\ao
 send buf=\d\ae8 send buf=\a---- Transcript of session follows ----\a\d\aF send buf=553  %TCPIP-E-SMTP_COMMANDERR, SMTP command error free.fr\d\a1 send buf=\a---- Unsent message follows ----\a\d\aU8 send buf=Date: Tue, 18 Mar 2003 10:31:10 +0100 (MET)\d\a3 send buf=Message-Id: <03031810311062@dtl02.dtl>\d\a B send buf=From: system@dtl02.dtl (Alpha System Manager account)\d\a' send buf=To: DIDIER.MORANDI@free.fr\d\au  send buf=Subject: test alpha\d\a2 send buf=X-VMS-To: SMTP%didier.morandi@free.fr\d\a
 send buf=\d\ae send buf=.\d\aC recv buf=250 2.0.0 h2I9W9eO069431 Message accepted for delivery\d\a  send buf=QUIT\d\an< recv buf=221 2.0.0 brinstar.nerim.net closing connection\d\a  / Printing debug_level 2, Domains and recipients:s Domain: dtl02.dtlx&    Recipient address: system@dtl02.dtl    Domain part:       dtl02.dtls    Local part:        system-    Address Status:    Done, delivered. (Sent)n Domain: nerim.netc-    Recipient address: didiermorandi@nerim.net     Domain part:       nerim.nete#    Local part:        didiermorandi -    Address Status:    Done, delivered. (Sent)t  ) Printing debug_level 2, relevant headers:c Return-Path: From: TCPIP$SMTP@DTL02.DTL To: system@dtl02.dtl Subject: Returned mail  G I did get the error message back, because I forwarded incoming mail to t@ the SYSTEM account to didiermorandi@nerim.net, which works fine:  ' ---- Transcript of session follows ----j  9 553  %TCPIP-E-SMTP_COMMANDERR, SMTP command error free.fro    ---- Unsent message follows ----  + Date: Tue, 18 Mar 2003 10:32:12 +0100 (MET)y& Message-Id: <03031810321202@dtl02.dtl>5 From: system@dtl02.dtl (Alpha System Manager account), To: DIDIER.MORANDI@free.fr Subject: test alphah% X-VMS-To: SMTP%didier.morandi@free.fro   I loose my Latin  :-)A   ------------------------------  % Date: Tue, 18 Mar 2003 12:14:37 +0100 $ From: "Peter Flunger" <p-i-b@gmx.at>( Subject: Re: send mail from an intranet?0 Message-ID: <b56v2t$u2r$1@newsreader1.netway.at>  6 "Didier Morandi" <Didier.Morandi.nospam@Free.fr> wrote > I loose my Latin  :-)     si tacuisses ... :)   OK, try the follwoing setup:  ! TCPIP> set config smtp/gate=nogenl# TCPIP> set config smtp/gate=noaltert1 TCPIP> set config smtp/subs=(hidd,name=nerim.net)b  
 This will:) Send outgoing mail as <vmsuser>@nerim.net )  ( which might be a problem, as i supposer0    system@nerim.net is not 'your' email address,     but should be OK for a test )6 and which will contact the target SMTP server directly   (dont't give up, yet...) Peter    ------------------------------  % Date: Tue, 18 Mar 2003 12:44:40 +0100S$ From: "Peter Flunger" <p-i-b@gmx.at>( Subject: Re: send mail from an intranet?0 Message-ID: <b570r9$uq6$1@newsreader1.netway.at>  ( "Peter Flunger" <p-i-b@gmx.at> (I) wrote+ >  ( which might be a problem, as i suppose 2 >    system@nerim.net is not 'your' email address," >    but should be OK for a test ), To avaoid that, you can use the logical name tcpip$smtp_from . $define tcpip$smtp_from "Didier.Morandi@...fr", in the login.com of the VMS account you use,- will send outgoing SMTP mail with that sender  address. Peterd   ------------------------------  % Date: Tue, 18 Mar 2003 10:53:29 -040000 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>( Subject: Re: send mail from an intranet?/ Message-ID: <3E7732E8.897EC477@vl.videotron.ca>    Didier Morandi wrote:nB > _DTL02$DKA0:[SYS0.SYSMGR]03031810311062_SYSTEM.TCPIP_DTL02;1(60)* > recv buf=220 mrelay1-1.free.fr ESMTP\d\a > send buf=HELO dtl02.dtl\d\ay$ > recv buf=250 mrelay1-1.free.fr\d\a+ > send buf=MAIL FROM:<system@dtl02.dtl>\d\a>I > recv buf=553 sorry, your envelope sender domain must exist (#5.7.1)\d\a>  N You are attempting to send a mail message from system@dtl02.dtl which is not a valid host name.  G The 553 message is issued by the remote SMTP server (which looks like a K relay). To prevent the relay from being "opened", they would restrict it to>> MAIL FROM containing only their own domain name (eg: free.fr).  N The "From:" in the body of the message is diffferent. The "MAIL FROM:"  in theG SMTP negotiation is part of the delivery envelope.  The RFC822 block oftK headers is not used to deliver the message, but the SMTP server does add toeQ it, notably the Received: line(s), and possibly the return-path: line at the top.m   ------------------------------  % Date: Tue, 18 Mar 2003 10:59:50 -0400i0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>( Subject: Re: send mail from an intranet?/ Message-ID: <3E773464.BB737CAC@vl.videotron.ca>t   Peter Flunger wrote:0 > $define tcpip$smtp_from "Didier.Morandi@...fr". > in the login.com of the VMS account you use,/ > will send outgoing SMTP mail with that senderu
 > address.    E last I checked, this affected only the domain name, not the username.    ------------------------------  + Date: Tue, 18 Mar 2003 17:13:51 +0100 (MET)a9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>o( Subject: Re: send mail from an intranet?; Message-ID: <01KTOEIUE7TA9H1MPS@sysdev.deutsche-boerse.com>h  2 > > $define tcpip$smtp_from "Didier.Morandi@...fr"0 > > in the login.com of the VMS account you use,1 > > will send outgoing SMTP mail with that senderh > > address. > G > last I checked, this affected only the domain name, not the username.   = No, it's just a logical, you can put whatever you want there.f  B /SUBSTITUTE_DOMAIN affects only the domain name, not the username.   ------------------------------  % Date: Tue, 18 Mar 2003 18:35:33 +0100s4 From: Didier Morandi <Didier.Morandi.nospam@Free.fr>( Subject: Re: send mail from an intranet?& Message-ID: <3E7758E5.9080707@Free.fr>   JF Mezei wrote:e  O >You are attempting to send a mail message from system@dtl02.dtl which is not ak >valid host name.o >nH >The 553 message is issued by the remote SMTP server (which looks like aL >relay). To prevent the relay from being "opened", they would restrict it to? >MAIL FROM containing only their own domain name (eg: free.fr).s > O >The "From:" in the body of the message is diffferent. The "MAIL FROM:"  in theoH >SMTP negotiation is part of the delivery envelope.  The RFC822 block ofL >headers is not used to deliver the message, but the SMTP server does add toR >it, notably the Received: line(s), and possibly the return-path: line at the top. >  d >v   IT WOOOOORKS !!!   MTP ConfigurationnJ                                                                    OptionsL Initial interval:   0 00:30:00.00       Address_max:    16       NOEIGHT_BITH Retry interval:     0 01:00:00.00       Hop_count_max:  16       NORELAYL Maximum interval:   3 00:00:00.00                                TOP_HEADERS  G Timeout             Initial       Mail    Receipt       Data  TerminatemG   Send:                   5          5          5          3         10i   Receive:                5u   Alternate gateway:  not defined  General gateway:    not defined   % Substitute domain:  HIDDEN, NERIM.NET  Zone:               NERIM.NET-   Postmaster:         TCPIP$SMTPC Log file:           SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOGi  0 Generic queue       Queues   Participating nodes  " TCPIP$SMTP_DTL02_00    1     DTL02  ' 18-MAR-2003 18:21:34.38: next open file.@ _DTL02$DKA0:[SYS0.SYSMGR]03031818213398_SYSTEM.TCPIP_DTL02;1(60)$ recv buf=220 mx1-1.free.fr ESMTP\d\a send buf=HELO dtl02.dtl\d\at recv buf=250 mx1-1.free.fr\d\a/ send buf=MAIL FROM:<Didier.Morandi@free.fr>\d\ai recv buf=250 ok\d\ae- send buf=RCPT TO:<DIDIER.MORANDI@free.fr>\d\an recv buf=250 ok\d\an send buf=DATA\d\ah recv buf=354 go ahead\d\an8 send buf=Date: Tue, 18 Mar 2003 18:21:34 +0100 (MET)\d\a3 send buf=Message-Id: <03031818213398@nerim.net>\d\a H send buf=From: Didier.Morandi@free.fr (Alpha System Manager account)\d\a' send buf=To: DIDIER.MORANDI@free.fr\d\aA  send buf=Subject: test alpha\d\a2 send buf=X-VMS-To: SMTP%didier.morandi@free.fr\d\aA send buf=X-VMS-True-From: system@nerim.net (Alpha System Manager s account)\d\a
 send buf=\d\ao send buf=.\d\a' recv buf=250 ok 1048008150 qp 15408\d\as send buf=QUIT\d\at recv buf=221 mx1-1.free.fr\d\a  / Printing debug_level 2, Domains and recipients:i Domain: free.frp,    Recipient address: DIDIER.MORANDI@FREE.FR    Domain part:       free.frh$    Local part:        DIDIER.MORANDI-    Address Status:    Done, delivered. (Sent)c  ) Printing debug_level 2, relevant headers:h# Return-Path: Didier.Morandi@free.fre+ Date: Tue, 18 Mar 2003 18:21:34 +0100 (MET)o& Message-Id: <03031818213398@nerim.net> From: Didier.Morandi@free.fr To: DIDIER.MORANDI@free.fr Subject: test alpha % X-VMS-To: SMTP%didier.morandi@free.fri  H What makes me "die out laughing", is "X-VMS-True-From: system@nerim.net"  A particular notion of Truth :-)   Thanks again to all.2 Hoff, you may introduce a new entry in the FAQ :-)   D.   ------------------------------  + Date: Tue, 18 Mar 2003 18:40:31 +0100 (MET)t9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>a( Subject: Re: send mail from an intranet?; Message-ID: <01KTOHGQRL409H1MPS@sysdev.deutsche-boerse.com>t  J > What makes me "die out laughing", is "X-VMS-True-From: system@nerim.net"" > A particular notion of Truth :-)  E The normal assumption is that only trusted users can tamper with the oE domain name.  In fact, I think TCPIP$SMTP_FROM was so long in coming eG since in some sense it allowed spoofing.  The X-VMS-True-From: I think .F is to include the "real" sender.  However, it will pick up the domain  from "trusted" information.s   ------------------------------    Date: 18 Mar 2003 05:17:08 -0800! From: rshkwee@hr.nl (R.S.H. Kwee) - Subject: Shareable data image and VMS-clustern= Message-ID: <7c610475.0303180517.5735aaaa@posting.google.com>l  F We have an application which uses shareable data images to share data.A The shareable data images are executable which will be installed.   E In this way the data can be shared between the application processes.eC Now we want to run our application on a cluster (OpenVMS 7.1-1H2 on  Alpha).e  C Can we put the shareble imagefiles on a shared disk and install thetD images on each clustermember, in order to share the data between the? processen on the different nodes? Or must we implement an other 
 mechanism?   ------------------------------    Date: 18 Mar 2003 07:41:43 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)N1 Subject: Re: Shareable data image and VMS-cluster 3 Message-ID: <OFUUKD05GKgA@eisner.encompasserve.org>i  a In article <7c610475.0303180517.5735aaaa@posting.google.com>, rshkwee@hr.nl (R.S.H. Kwee) writes:rH > We have an application which uses shareable data images to share data.C > The shareable data images are executable which will be installed.  > G > In this way the data can be shared between the application processes.sE > Now we want to run our application on a cluster (OpenVMS 7.1-1H2 oni	 > Alpha).n > E > Can we put the shareble imagefiles on a shared disk and install thehF > images on each clustermember, in order to share the data between theA > processen on the different nodes? Or must we implement an othero > mechanism?  E    I believe cluster-wide shareable images came out at about the same-$    time, and may be part of, Galaxy.   ------------------------------    Date: 18 Mar 2003 07:31:49 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)'H Subject: RE: Somebody didn't get the message?  HP shows it doesn't care.3 Message-ID: <ShCzXeqOGIKk@eisner.encompasserve.org>   _ In article <CIEJLCMNHNNDLLOOGNJIOEDEGNAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:n > K > Bob, I gave up on that in 1999.  Run a pc as a POP client, that is what Id > do.HI > I can still read the mail (if I want to) on one of my VMS platforms.  Is > don'tc. > think anybody will regard you as a quisling.  G    I munpack and FTP things to my Mac or PC, but mostly because I can'tP3    tell my customer what format to write things in.g  D    When I get curomt news from HP I expect them to support all theirF    platforms.  If VMS engineering wasn't so busy and mail such an easyE    target for security holes, I'd like to see them providing a bettere-    DECwindows interface, with HTML rendering.t  K    But right now I'd like to see them running DECwindows mail as is on IPF.    ------------------------------  % Date: Tue, 18 Mar 2003 11:02:59 -0400n0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>H Subject: Re: Somebody didn't get the message?  HP shows it doesn't care./ Message-ID: <3E773522.2EBFACEC@vl.videotron.ca>    Bob Koehler wrote:M >    But right now I'd like to see them running DECwindows mail as is on IPF.   N No, fixing up and sprucing up MAIL to be more complete and more internet savyyJ should be a priority well above that unwanted port to IA64 which can wait.L Nobody is using VMS on IA64 and won't be for a few years. Lots of people are& hindered by VMSmail's limitations NOW.   ------------------------------  % Date: Tue, 18 Mar 2003 18:09:36 +0100r$ From: Michael Unger <unger@decus.de>H Subject: Re: Somebody didn't get the message?  HP shows it doesn't care.+ Message-ID: <00A1D101.FFAFB2F0.52@decus.de>   > "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote:  M >    "Today I received an illegible email from dspp.hewlett-packarg@hp.com. IaL >    think it says something about being customized news. Since it's full ofM >    HTML markups it can't easily be read through my HP OpenVMS mail system, c+ >    so I know it's not customized for me."e > 7 >    I sent the above to the DSPP contact and got back:a > N >    ... "The newsletters are tested and supported for the most popular email F >    systems that the majority of the newsletter subscribers currently >    use." ... > H >    Followed by instructions on how to get the info via the web and how( >    to unsubscribe from the email list.  ? I had similar experiences with the other successor of "the old tF Hewlett-Packard" (i.e., Agilent), but finally they got the message ...  H A few months ago I got a printed mailing suggesting to subscribe to the @ new electronic newsletters in addition to receiving the printed E material. I visited the URL mentioned just to discover that I wasn't  D able to subscribe with a secured browser (lots of JavaScript). So I F complained to the feedback address given at the web pages, explaining H that at least some users of computers are aware of security and privacy  issues.3  F Some days later I got an apology by e-mail together with an extensive H list of topics to choose from for a "customized" newsletter. I selected I my preferences and sent it back to them. About two weeks later I got the rH first newsletter (they are distributed at regular intervals) -- full of H HTML and nearly unreadable. So I complained again; the first answer was G a rather short one directing me to a web page where I should change my sG preference from "HTML" to "plain text". As already expected I couldn't  C do that myself for the very same reason I couldn't subscribe. So I eF complained again. Now I'm getting the newsletter in plain text format.  1 Agilent obviously listens to customer complaints.i   Michael    ------------------------------    Date: 18 Mar 2003 02:36:03 -0800- From: denis.fayaud@netspace.mc (Denis Fayaud)l Subject: strange ENQ rates= Message-ID: <93820504.0303180236.4042d807@posting.google.com>l   Hello,$ I have a strange problem on a DS 10.1 Issuing the command monitor lock, I can see that:>F the two parameters "Converted ENQ Rate" and "ENQs Not Queued Rate" areD increasing continuously, I mean they are increasing of about 10 in a day... and they neveri	 decrease. ? The DS10 is in a mixed VAX/ALPHA cluster, 3 alphas and 4 vaxes.t
 any idea ? thanx in advance   ------------------------------  % Date: Tue, 18 Mar 2003 11:10:46 +0000h( From: Nic Clews <sendspamhere@127.0.0.1> Subject: Re: strange ENQ rates) Message-ID: <3E76FEB6.194CF3DA@127.0.0.1>    Denis Fayaud wrote:   & > I have a strange problem on a DS 10.  ? Not necessarily, it may just be the way you are doing things...i  3 > Issuing the command monitor lock, I can see that:lH > the two parameters "Converted ENQ Rate" and "ENQs Not Queued Rate" areF > increasing continuously, I mean they are increasing of about 10 in a > day... and they never  > decrease.hA > The DS10 is in a mixed VAX/ALPHA cluster, 3 alphas and 4 vaxes.  > any idea ?    What are your programmers doing?  B Converted locks mean that a lock is already held at a lower level,B typically NL, and converted to some level up (CW, PW, CR, EX etc.)  E As to ENQs not queued, are you [your application] trying to acquire a-G lock, finding someone else already has it, and then abandoning the locknF request from the application's point of view? (Is it a sum of new lockE requests and/or conversions not queued?) I'm not certain, but I think2B this would contribute to the figures, but I'm open to correction /
 confirmation.t  F DEQ rate is the rate of 'lowering' a lock, or releasing it altogether.  ) Have you looked at your DLOCK statistics?2   -- 9? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencess nclews at csc dot com6   ------------------------------  % Date: Tue, 18 Mar 2003 12:51:17 -0000 - From: "Steve Spires" <Steve.Spires@torex.com>y% Subject: SYS_CHECK equivelant on VMS. E Message-ID: <91947A84607D9D48B8E674A5FAB54DA63CAE6A@tahiti.tinuk.com>(  H Is there anything like SYS_CHECK which you have with Tru64 which runs on VMS?   Steve Spires Technical Consultant Torex Health [T] +44 (0) 1295 274200S [F] +44 (0) 1295 275131a
 www.torex.coms   ------------------------------    Date: 18 Mar 2003 07:48:09 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)o) Subject: Re: SYS_CHECK equivelant on VMS.r3 Message-ID: <FaHxLdvGY+kx@eisner.encompasserve.org>   u In article <91947A84607D9D48B8E674A5FAB54DA63CAE6A@tahiti.tinuk.com>, "Steve Spires" <Steve.Spires@torex.com> writes:oJ > Is there anything like SYS_CHECK which you have with Tru64 which runs on > VMS?  C The SHOW TIME command is like SYS_CHECK in that is begins with "S".h  @ If you were thinking of _functionally_ "like", perhaps you could explain what SYS_CHECK is :-)r   ------------------------------  + Date: Tue, 18 Mar 2003 15:25:32 +0100 (MET)e9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> + Subject: TCPIP: how to change the MTU valuer; Message-ID: <01KTOAQ7HWIA9H1MPS@sysdev.deutsche-boerse.com>d  G TCPIP SHOW [CONFIG] INTERFACE shows an MTU value.  How does one change   this?u  A (The default seems to be 1500, but since 8 are lost due to PPPoE  E overhead, it might make sense to change it to 1492 when using PPPoE).h   ------------------------------    Date: 18 Mar 2003 08:03:10 -08001 From: susan_skonetski@hotmail.com (Sue Skonetski)t' Subject: Tuesday March 18 OpenVMS Pearlo= Message-ID: <857e9e41.0303180803.26417147@posting.google.com>d   Dear Newsgroup,a+ Here was the pearl I sent out this morning.w   sue 3 ___________________________________________________s   Folks,    @ I thought you might be interested in this persons change of viewB because of the IPF port.  This is a great example of how technical5 information contributes to the companies bottom line.s  
 Warm Regards,-   Sue-      4 http://www.shado.info/blog/archives/cat_hacking.html  
 VMS Lives ? When I started at NASA in 1986, I started working "for real" onPD Digital Equipment Corporation VAX/VMS Systems. Prior to that, I usedC Mainframes. VMS is what I did for next 7 years, and it was the bestpE operating system I ever used. But for a variety of reasons, I steppedaF away from VMS in 1993 expecting for the Operating System to die a slow) death with the march of UNIX and Windows.l  @ Well, on the 31st, the brains on Spitbrook Road (Home to the VMSE Engineering Group) in Nashua, NH let us know they were still there...r  3 From: Sue Skonetski (susan.skonetski@hp.nospam.com)d3 Subject: OpenVMS Boots on Itanium on Friday Jan 31 r Newsgroups: comp.os.vms  Date: 2003-01-31 14:50:17 PST    -----Original Message-----   From: Grant, Clair& Sent: Friday, January 31, 2003 3:40 PM" To: OpenVMS Systems Software Group Subject: VMS boots on Itaniumu  @ We are please to announce that OpenVMS has booted on an HP i2000 ItaniumhB system. By successfully executing a DIRECTORY command the official date and: time for the Boot Contest is January 31, 2003 3:31 PM EST.  F Many, many people have contributed to this effort, including managers,E supervisors, project leaders, consultants, development infrastructuree group,D calling standard committee, engineers working on post-boot projects, etc.> All will receive their well-deserved credit as we proceed. The following is: a very specific set of people; these are the engineers whoE designed/wrote/debugged the code that comprises the running operatingM system.   5 (list of wonderful engineers removed by Sue Skonetskie   ------------------------------  % Date: Tue, 18 Mar 2003 09:46:04 -0000o5 From: "Peter Ibbotson" <spambox@ibbotson.demon.co.uk>n Subject: Re: unixf4 Message-ID: <b56pss$o4f$1$8300dec7@news.demon.co.uk>  4 "Carl Perkins" <carl@gerg.tamu.edu> wrote in message' news:17MAR200316402046@gerg.tamu.edu... ; > "Peter Ibbotson" <spambox@ibbotson.demon.co.uk> writes... L > }b) The resolution is "fixed", they'll support lower resolutions, but they doI > }it by interpolating between the pixels, so if the screen resolution is L > }1024x768 you'ld better set your computer to the same or it'll be "fuzzy".K > }(Relevant for those who are getting older and like to work in 800x600 or ! > }640x480 to get "bigger" fonts)l > K > If you are lowering the resolution to get bigger fonts then you are doing)G > it wrong. To get bigger fonts, you should set it to use bigger fonts.s >     H Lots of apps don't care. (Excel for one!) the toolbar never gets bigger.G Many form filling apps don't work with larger fonts. Many of the worldssK flash enabled web sites use fonts that are too small. Basically life is lotnH easier by changing the resolution, it acts as a 20% zoom for those whoseH eyesight is going off. My mum for example complains that 1024x768 is tooG small, maybe she needs reading glasses but in the meanwhile it's a good  solution for her.    --F Work peteri@lakeview.co.uk.plugh.org | remove magic word .org to replyH Home peter@ibbotson.co.uk.plugh.org  | I own the domain but theres no MX   ------------------------------  ! Date: Tue, 18 Mar 03 11:14:56 GMT0 From: jmfbahciv@aol.com  Subject: Re: unixA, Message-ID: <b571r9$8e6$11@bob.news.rcn.net>  - In article <17MAR200319530725@gerg.tamu.edu>, +    carl@gerg.tamu.edu (Carl Perkins) wrote:3! >bfranchuk@jetnet.ab.ca writes...  >}Carl Perkins wrote:e+ >}> Then set the menus to use bigger fonts.  >}Why?   >u" >So that you can read them. (Duh.)  4 <GRIN>  Careful, you may have a bit god by the tail.   >M- >}I thought software was to my my life easierP
 >}not harder?F >.> >(First I'd like to point out that this statement may indicate< >that you have not been paying much attention to reality...)   ROTFL.   >,8 >Software is not now, nor has it ever been, telepathic.   ; That's good because the computer biz would have disappeared ( in a puff of blue smoke a long time ago.   > ..It can't< >know what font size you want. You have to tell it. This has: >always been the case, although once upon a time there was? >nothing you could do about it (without modifying the hardware, 	 >anyway).o  = HOWEVER, in the olden days, a mere computer user did not havei; to be a typeset expert in order to print an ASCII file that > contained a list of characters.  Once again, the PC generation@ has forced users to become experts at everything before the turn on the system.   >a@ >You know, you have to set the screen resolution and color depth= >that you want to use too (unless you happen to get lucky andi@ >the default is what you actually want). It does not set itself. >1B >Thus you have a choice: you can set the resolution low to make itE >have bigger (and uglier) text, or you can set the fonts to be bigger.B >(and relatively nice). In either case, you are setting something. > E >This ugly-nice resolution dependant comparison is true on a CRT typeaF >display, but it is even more so on an LCD. Running an LCD at anythingF >other than it's native resolution is a bad plan - that blurryness mayC >not be your bad eyesight, it may be the interpolation that you are  >making it use.i  < I guess the best buying plan would be to make sure there was< a (it won't work on my machine) return policy before I shell	 out $800.b   /BAH  ' Subtract a hundred and four for e-mail.-   ------------------------------  % Date: Tue, 18 Mar 2003 07:15:36 -0500 * From: "J. Clarke" <jclarke@nospam.invalid> Subject: Re: unixn; Message-ID: <pan.2003.03.18.12.15.33.266040@nospam.invalid>f  9 On Tue, 18 Mar 2003 09:46:04 +0000, Peter Ibbotson wrote:d  6 > "Carl Perkins" <carl@gerg.tamu.edu> wrote in message) > news:17MAR200316402046@gerg.tamu.edu...e< >> "Peter Ibbotson" <spambox@ibbotson.demon.co.uk> writes...H >> }b) The resolution is "fixed", they'll support lower resolutions, but >> }they > doJ >> }it by interpolating between the pixels, so if the screen resolution isD >> }1024x768 you'ld better set your computer to the same or it'll beJ >> }"fuzzy". (Relevant for those who are getting older and like to work in- >> }800x600 or 640x480 to get "bigger" fonts). >>F >> If you are lowering the resolution to get bigger fonts then you areG >> doing it wrong. To get bigger fonts, you should set it to use bigger 	 >> fonts.  >> >> >>J > Lots of apps don't care. (Excel for one!) the toolbar never gets bigger.  J I just tried it and it does appear to get bigger on Excel 2K under Windows 2000.   6 > Many form filling apps don't work with larger fonts.  I This is an unfortunate truth--developers seem not to test with nondefaultv fonts.   > Many of the worlds7 > flash enabled web sites use fonts that are too small.d  G IMO Bin Laden picked the wrong buildings--he should have hit MacromediaeI headquarters.  However we are stuck with the damned thing, at least untillE someone does the world that particular service.  One more reason thati Flash is eeeeevil.   > Basically life is lotaJ > easier by changing the resolution, it acts as a 20% zoom for those whoseJ > eyesight is going off. My mum for example complains that 1024x768 is tooI > small, maybe she needs reading glasses but in the meanwhile it's a good  > solution for her.e  D FWIW, under Windows Matrox video boards implement zoom with a singleC keystroke, which gives you a magnified window into a larger virtualaG desktop.  ATI used to do this as well, but for some reason their recentSG drivers no longer implement the single keystroke zoom.  I have a friendt@ with poor vision for whom I set one of these up and he loves it.  I Another thing that may help--if you are using Windows and have a Logitech " mouse try the IBM mouse drivers atn <http://www-1.ibm.com/support/docview.wss?rs=0&q=31P7405&uid=psg1MIGR-4HVR6Q&loc=en_US&cs=utf-8&cc=us&lang=en>H (or go to support and search for 31P7405 or just buy the mouse with thatJ model number--it's a nice mouse).  Those drivers have a nice little pop-upJ magnifier that can be assigned to a mouse button--on the IBM mouse there'sE a true center button, on a 2-button Logitech I assign it to the wheel% click.   --   --John- Reply to jclarke at ae tee tee global dot netT# (was jclarke at eye bee em dot net)F   ------------------------------  % Date: Tue, 18 Mar 2003 13:12:47 -0000e5 From: "Peter Ibbotson" <spambox@ibbotson.demon.co.uk>  Subject: Re: unix$4 Message-ID: <b5760f$7vd$1$8302bc10@news.demon.co.uk>  5 "J. Clarke" <jclarke@nospam.invalid> wrote in message 5 news:pan.2003.03.18.12.15.33.266040@nospam.invalid... ; > On Tue, 18 Mar 2003 09:46:04 +0000, Peter Ibbotson wrote:t >t8 > > "Carl Perkins" <carl@gerg.tamu.edu> wrote in message+ > > news:17MAR200316402046@gerg.tamu.edu...3> > >> "Peter Ibbotson" <spambox@ibbotson.demon.co.uk> writes...J > >> }b) The resolution is "fixed", they'll support lower resolutions, but
 > >> }they > > doL > >> }it by interpolating between the pixels, so if the screen resolution isF > >> }1024x768 you'ld better set your computer to the same or it'll beL > >> }"fuzzy". (Relevant for those who are getting older and like to work in/ > >> }800x600 or 640x480 to get "bigger" fonts)m > >>H > >> If you are lowering the resolution to get bigger fonts then you areI > >> doing it wrong. To get bigger fonts, you should set it to use bigger  > >> fonts.e > >> > >> > >>L > > Lots of apps don't care. (Excel for one!) the toolbar never gets bigger. > L > I just tried it and it does appear to get bigger on Excel 2K under Windows > 2000.e    K Actually it's because the text in the combo boxes got bigger. If you pick a L toolbar with no text (The picture one doesn't normally) and put it on a lineJ on it's own then it doesn't resize. (To be fair to MS there is a big iconsH checkbox hidden away under View/Toolbars/Customise/Options/Large icons.)  F > FWIW, under Windows Matrox video boards implement zoom with a singleE > keystroke, which gives you a magnified window into a larger virtualAI > desktop.  ATI used to do this as well, but for some reason their recent   J One of our support guys uses a dual monitor matrox config where the secondJ monitor chases the mouse, but then he's registered blind so needs a betterD solution that just changing the resolution. Since I can see over hisK shoulder, it tends to be the toolbar stuff that doesn't scale well (or feel- nice in "high contrast" mode)e --F Work peteri@lakeview.co.uk.plugh.org | remove magic word .org to replyH Home peter@ibbotson.co.uk.plugh.org  | I own the domain but theres no MX   ------------------------------  % Date: Tue, 18 Mar 2003 08:03:18 -0500t' From: Howard S Shubs <howard@shubs.net>C Subject: Re: unixw< Message-ID: <howard-CB24D5.08031818032003@enews.newsguy.com>  4 In article <b56pss$o4f$1$8300dec7@news.demon.co.uk>,7  "Peter Ibbotson" <spambox@ibbotson.demon.co.uk> wrote:0  J > Lots of apps don't care. (Excel for one!) the toolbar never gets bigger.I > Many form filling apps don't work with larger fonts. Many of the worldsgM > flash enabled web sites use fonts that are too small. Basically life is lotLJ > easier by changing the resolution, it acts as a 20% zoom for those whoseJ > eyesight is going off. My mum for example complains that 1024x768 is tooI > small, maybe she needs reading glasses but in the meanwhile it's a goods > solution for her.e  J I've noticed that in older people.  640 x 480 on a 17" monitor and such.  6 A bigger monitor means bigger pixels for these people.   -- s4 Today, on Paper-view: The World Origami Championship   ------------------------------  % Date: Tue, 18 Mar 2003 21:52:37 +0800M4 From: Bernd Felsche <bernie@innovative.iinet.net.au> Subject: Re: unixn3 Message-ID: <5b875b.ftd.ln@innovative.iinet.net.au>s  7 "Peter Ibbotson" <spambox@ibbotson.demon.co.uk> writes:T  5 >"Carl Perkins" <carl@gerg.tamu.edu> wrote in messaged( >news:17MAR200316402046@gerg.tamu.edu...< >> "Peter Ibbotson" <spambox@ibbotson.demon.co.uk> writes...P >> }b) The resolution is "fixed", they'll support lower resolutions, but they doJ >> }it by interpolating between the pixels, so if the screen resolution isM >> }1024x768 you'ld better set your computer to the same or it'll be "fuzzy". L >> }(Relevant for those who are getting older and like to work in 800x600 or" >> }640x480 to get "bigger" fonts) >>B >> If you are lowering the resolution to get bigger fonts then youD >> are doing it wrong. To get bigger fonts, you should set it to use >> bigger fonts.  A >Lots of apps don't care. (Excel for one!) the toolbar never getsrC >bigger.  Many form filling apps don't work with larger fonts. ManyMD >of the worlds flash enabled web sites use fonts that are too small.D >Basically life is lot easier by changing the resolution, it acts as= >a 20% zoom for those whose eyesight is going off. My mum for0> >example complains that 1024x768 is too small, maybe she needsC >reading glasses but in the meanwhile it's a good solution for her.,  C I honestly think those "flash" developers use 640 x 480 resolution,0A or have screens with 1mm dot-pitch. A hi-res screen with < 0.25mmaD dot pitch and sized to suit, results in three-point print size; half0 the size of the classified ads in the newspaper.  C The best solution is to let the webmasters know that you can't read3( what the hell they do. And go elsewhere. -- 0B /"\ Bernd Felsche - Innovative Reckoning, Perth, Western Australia4 \ /  ASCII ribbon campaign | I'm a .signature virus!;  X   against HTML mail     | Copy me into your ~/.signatureo/ / \  and postings          | to help me spread!P   ------------------------------  % Date: Tue, 18 Mar 2003 15:04:50 -0000m% From: Pete Fenelon <pete@fenelon.com>t Subject: Re: unix / Message-ID: <v7edcigbbjp3bb@corp.supernews.com>3  O In alt.folklore.computers Bernd Felsche <bernie@innovative.iinet.net.au> wrote:  > E > I honestly think those "flash" developers use 640 x 480 resolution,e  G I don't care what they use. "Skip Intro" and/or "Add to Block List" are\6 your friends when encountering unwanted Flash content.   pete -- tE pete@fenelon.com "there's no room for enigmas in built-up areas" HMHBe   ------------------------------  + Date: Tue, 18 Mar 2003 15:25:23 +0000 (UTC)s8 From: hawk@slytherin.ds.psu.edu (Dr. Richard E. Hawkins) Subject: Re: unix . Message-ID: <b57dp3$1fl4$9@f04n12.cac.psu.edu>  - In article <873clpejjg.fsf@prep.synonet.com>,o. Paul Repacholi  <prep@prep.synonet.com> wrote:; >hawk@slytherin.ds.psu.edu (Dr. Richard E. Hawkins) writes:p  = >> I can put the plain postscript up if anyoen is interested..  A >So why are you in a snit with LaTeX? PS is a good second best...   F I usually write in lyx, not latex.  THat particular paper got exported@ to latex to use an article class that didn't work well with lyx.  F Two reasons for using lyx over latex (and, yes, vim/latex is much more/ my natural inclination, if it wasn't obvious :)5  E 1) Equation display. Mine tend to be too complicated to sensibly editeG without the display.  A couple of minutes with lyx several years ago isg@ what switched me from mac (pre-6 word) to unix as my regular os.  E 2)  I'm a keystroke whore.  There are a *lot* less keystrokes in lyx. H I suppose if I had spare time I could write a package for vim, but . . .   hawk -- -K Richard E. Hawkins, Asst. Prof. of Economics    /"\   ASCII ribbon campaignpG dochawk@psu.edu  Smeal 178  (814) 375-4700      \ /   against HTML mail2D These opinions will not be those of              X    and postings. 6 Penn State until it pays my retainer.           / \      ------------------------------  % Date: Tue, 18 Mar 2003 08:39:53 -0700= From: bfranchuk@jetnet.ab.ca Subject: Re: unixI+ Message-ID: <3E773DC9.5040902@jetnet.ab.ca>b   Carl Perkins wrote:t > . > }I thought software was to my my life easier > }not harder? > ? > (First I'd like to point out that this statement may indicates= > that you have not been paying much attention to reality...)P > A > Software is not now, nor has it ever been, telepathic. It can't-= > know what font size you want. You have to tell it. This has ; > always been the case, although once upon a time there wasn@ > nothing you could do about it (without modifying the hardware,
 > anyway).  = No but software could be smart and check window size on startu> and have resonable defaults but alas most stuff is hard coded. Ben.   ------------------------------   End of INFO-VAX 2003.152 ************************