0 INFO-VAX	Sat, 10 Feb 2001	Volume 2001 : Issue 82      Contents:B Re: 64 bit systems with small memory, was: Higher sized memory for >Re: About your Credit Report  Re: Another missed opportunity DCE 1.5 & TCPWare ! Do you want to make a QUICK $100?  Re: Error messages on VMS 7.2  Re: Error messages on VMS 7.2 3 Re: Firmware upgrade for PWS 600au to use a ZLXp-L1 3 Re: Firmware upgrade for PWS 600au to use a ZLXp-L1 3 Re: Firmware upgrade for PWS 600au to use a ZLXp-L1  Re: Free space ? Re: GS build, no config.c 2 Re: How do you make a pointer to another directory; how to determine the initial queue a batch was submitted on ? Re: how to determine the initial queue a batch was submitted on / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours  Re: MMS/MMK Ghostscript problem  Re: MMS/MMK Ghostscript problem P Re: Not Being Published: OpenVMS and the Internet (was: Re: Notbeingpublished: O Re: Open VMS Cloning Re: set character set? Re: set character set? Re: Status of EV7  Re: Status of EV7  Re: Status of EV7  Re: Status of EV7  Re: Status of EV7 % Re: Suggested upgrade to VMS (DECnet)  Re: TK70 manual  Re: USAGE.DAT and missing files 9 Re: VAX/VMS 7.2 BACKUP/IMAGE behavior when a file is open 9 Re: VAX/VMS 7.2 BACKUP/IMAGE behavior when a file is open 9 Re: VAX/VMS 7.2 BACKUP/IMAGE behavior when a file is open 5 Was MMS/MMK Ghostscript problem - Now Humphrey Bogart 7 Re: What monitor and keyboad will work with dec3000/400 7 Re: What monitor and keyboad will work with dec3000/400   F ----------------------------------------------------------------------    Date: 10 Feb 2001 06:18:36 +0800, From: Paul Repacholi <prep@prep.synonet.com>K Subject: Re: 64 bit systems with small memory, was: Higher sized memory for - Message-ID: <87itmj1ow3.fsf@prep.synonet.com>     Dirk Munk <munk@home.nl> writes:  D > "You will also notice that memory size has an upper bound of 4 GB;@ > it is not clearly stated, but my interpretation is that at theC > moment, console firmware supports no more than 4 GB and more will  > come in the future."  = Go read the Typhoon/Tsunammi chipset manual. 4Gb is the limit  in the chipset.    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  * Date: Fri, 09 Feb 2001 11:16:23 +0120 (PM) From: y38yptDu3@netscape.net& Subject: >Re: About your Credit Report5 Message-ID: <B979920l4GtaQzu@SMTP-OUT003.ONEMAIN.COM>    YOU CAN HAVE PERFECT CREDIT   + What credit bureaus don't want YOU to know!    ERASE BAD DEBT - Legally!    The law is on YOUR side!  4 Say goodbye to those nasty credit rejection letters.  4 http://3506561038/debtrelief2002/perfect_credit_.htm       Transmissions to you by the sender of 'this' email will be stopped promptly by sending an e-mail with "REMOVE" in the subject line. Simply hit reply and send and we will remove you from our database. Please Note-This is a one time mailing.Thank you.    ------------------------------  % Date: Sat, 10 Feb 2001 09:11:25 +0100   From: Paul Sture <paul@sture.ch>' Subject: Re: Another missed opportunity + Message-ID: <VA.00000296.31aa58ae@sture.ch>   M In article <rdeininger-0902012323010001@user-2ivebcf.dialup.mindspring.com>,   Robert Deininger wrote: 4 > From: rdeininger@mindspring.com (Robert Deininger) > Newsgroups: comp.os.vms ) > Subject: Re: Another missed opportunity ' > Date: Fri, 09 Feb 2001 23:23:01 -0500  > R > In article <962d0i$u9$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> wrote: >  > P > > My point, which may be simply due to a faulty memory, was that I thought VMSN > > supported 'deny' ACEs as well as 'grant' ACEs, and that a 'deny' ACE couldN > > be used to curtail the scope of a 'grant' ACE (whether by virtue of coming0 > > earlier in the scan or not, I couldn't say). > I > Yes, ACEs can block access as well as grant it.  And the order matters.  > O I once heard a tale of Pathworks files which couldn't be deleted by the SYSTEM  0 account - solution - reinitialize the disk. Yuk. ___ 
 Paul Sture Switzerland    ------------------------------  % Date: Sat, 10 Feb 2001 17:26:11 +0300 4 From: "Ruslan R. Laishev" <Laishev@SMTP.DeltaTel.RU> Subject: DCE 1.5 & TCPWare0 Message-ID: <3A854F83.1C9AB5E7@SMTP.DeltaTel.RU>   Hi All! D 	Just for you information: the way is described in the documentationS "Compaq DCE for OpenVMS VAX and OpenVMS Alpha Release Notes"/"10 Using TCPware with  DCE" is not work at all.   	Be advised.   --   Cheers. = +---------------pure personal opinion-----------------------+ 9     RADIUS Server for OpenVMS project - www.radiusvms.com 6       vms-isps@dls.net - Forum for ISP running OpenVMS 		Mobile: +7 (901) 971-3222
   ------------------------------  + Date: Sat, 10 Feb 2001 02:20:44 -0500 (EST)  From: opportunity4u@usa.com * Subject: Do you want to make a QUICK $100?> Message-ID: <200102101025.TAA0000027534@as4000.munkyung.ac.kr>   Dear Friend,  8 May I have your permission to send you FREE information < regarding an exciting and very profitable SELF-RUN business?  D You can easily acheive $5,000 +  from the comfort of your own home,  and the BEST part is:    * - It makes no difference where you reside!# - You can begin making money TODAY! M - It does not take a lot of capital to begin. In fact, I bet you have enough  . money in your pocket RIGHT NOW to get started!  K Forget all the chain letters you read about, which by the way ARE illegal,  K no matter what postal regulations are quoted (do the research yourself and  J you will see). Forget about the pyramid scams,  matrixes,  MLM's and other  mathematically impossible stuff!  ; Find out how NORMAL people make $5,000 a month or MORE from B their home computers with NO COMPANY in between! Make $100 TODAY!  See the PROOF that it WORKS!       It is truly that SIMPLE!  	 AND.....     V You can try it for a 30 day period before you commit!  What could be easier than that!   V I spend less than 1 hour per day running this business and I make enough to support my entire family. You can too!    D To receive FREE VITAL INFORMATION please reply to the address below 1 with the word "information" in the subject field.    Mailto: ljfeeney1@flashmail.com    ) You WILL NOT be disappointed! Believe me!     Thank you and have a great day!     L.J.F. OL Ventures   N You will only receive additional information if you request it, therefore, no $ removal instructions are applicable.   ------------------------------  % Date: Fri, 09 Feb 2001 22:49:46 -0800 ! From: Koloth <koloth@tmisnet.com> & Subject: Re: Error messages on VMS 7.2+ Message-ID: <3A84E48A.6592085E@tmisnet.com>   & --------------1F782765EE2CB3BC1962428E* Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit   E This is DECnet V DTss (time service) complaining that it can't find 3 E DTSS time servers on the network.  You will most likely find that you * can't do a SET TIME on your system either.  ( To get around this you can disable DTSS.   $MC NCL  Disable DTSS  $MC NCL  Delete DTSS  G To prevent this from happening in the future you can either put an EXIT H at the front of SYS$MAMAGER:NET$DTSS_CLERK_STARTUP.NCL.  This is CompaqsH fix.  Or you can append the DISABLE DTSS and DELETE DTSS commands to theD end of this file.  I did the later because there was some problem ifC DTSS didn't start up first and then was shutdown.  I can't remember 2 exactly what the problem was but that was the fix.   Regards,   Cass Witkowski Senior Systtems Engineer SAIC   Gloria Griffith wrote:   > Hello All, > 9 > I have been searching through manuals and FAQ's for the E > reason/solution for these error message that I am getting on my VMS 9 > system but I have not been able to track them down yet.  > < > On my Alpha Server DS20E running VMS 7.2 I get this error:) >  (This machine is not part of cluster )  > $ : > %%%%%%%%%%%  OPCOM   9-FEB-2001 16:58:27.63  %%%%%%%%%%%$ > Message from user SYSTEM on GC02BS@ > Event: Too Few Servers Detected from: Node LOCAL:.GC02BS DTSS,/ >         at: 2001-02-09-16:58:27.632-06:00Iinf  >         Number Detected=0, >         Number Required=1 9 >         eventUid   C3F97DA7-FEAC-11D4-B2B8-474330324253 9 >         entityUid  2458844C-FA8C-11D4-83C9-AA0004000504 9 >         streamUid  242031AA-FA8C-11D4-83AA-AA0004000504  > H > it pops up every 7 minutes or so, what is it and how do I fix whatever > it is? > TIA  >  > Gloria Griffith  > Ericsson VMS System Admin  > 972-583-7052  & --------------1F782765EE2CB3BC1962428E) Content-Type: text/html; charset=us-ascii  Content-Transfer-Encoding: 7bit   > <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html>J This is DECnet V DTss (time service) complaining that it can't find 3 DTSSK time servers on the network.&nbsp; You will most likely find that you can't $ do a SET TIME on your system either.+ <p>To get around this you can disable DTSS.  <p>$MC NCL&nbsp; Disable DTSS  <br>$MC NCL&nbsp; Delete DTSS J <p>To prevent this from happening in the future you can either put an EXITM at the front of SYS$MAMAGER:NET$DTSS_CLERK_STARTUP.NCL.&nbsp; This is Compaqs F fix.&nbsp; Or you can append the DISABLE DTSS and DELETE DTSS commandsM to the end of this file.&nbsp; I did the later because there was some problem K if DTSS didn't start up first and then was shutdown.&nbsp; I can't remember 2 exactly what the problem was but that was the fix. <p>Regards,  <p>Cass Witkowski  <br>Senior Systtems Engineer <br>SAIC <p>Gloria Griffith wrote:   <blockquote TYPE=CITE>Hello All,J <p>I have been searching through manuals and FAQ's for the reason/solutionI for these error message that I am getting on my VMS system but I have not ! been able to track them down yet. = <p>On my Alpha Server DS20E running VMS 7.2 I get this error: 0 <br>&nbsp;(This machine is not part of cluster )" <br><font color="#FF0000">$</font>H <br><font color="#FF0000">%%%%%%%%%%%&nbsp; OPCOM&nbsp;&nbsp; 9-FEB-2001$ 16:58:27.63&nbsp; %%%%%%%%%%%</font>C <br><font color="#FF0000">Message from user SYSTEM on GC02BS</font> R <br><font color="#FF0000">Event: Too Few Servers Detected from: Node LOCAL:.GC02BS DTSS,</font>H <br><font color="#FF0000">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; at:( 2001-02-09-16:58:27.632-06:00Iinf</font>K <br><font color="#FF0000">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Number  Detected=0,</font>K <br><font color="#FF0000">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Number  Required=1</font> Y <br><font color="#FF0000">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; eventUid&nbsp;&nbsp; + C3F97DA7-FEAC-11D4-B2B8-474330324253</font> T <br><font color="#FF0000">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; entityUid&nbsp;+ 2458844C-FA8C-11D4-83C9-AA0004000504</font> T <br><font color="#FF0000">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; streamUid&nbsp;+ 242031AA-FA8C-11D4-83AA-AA0004000504</font> I <p>it pops up every 7 minutes or so, what is it and how do I fix whatever  it is? <br>TIA  <p>Gloria Griffith <br>Ericsson VMS System Admin  <br>972-583-7052</blockquote>  </html>   ( --------------1F782765EE2CB3BC1962428E--   ------------------------------  % Date: Sat, 10 Feb 2001 09:30:17 +0100 2 From: martin@radiogaga.harz.de (Martin Vorlaender)& Subject: Re: Error messages on VMS 7.2; Message-ID: <3a84fc19.524144494f47414741@radiogaga.harz.de>   0 Gloria Griffith (qusgagh@am1.ericsson.se) wrote:( > --------------5A86E8E4B172CF09A221C8FE, > Content-Type: text/plain; charset=us-ascii! > Content-Transfer-Encoding: 7bit   ? Please don't post MIME, or at least switch off HTML. Thank you.   I > I have been searching through manuals and FAQ's for the reason/solution G > for these error message that I am getting on my VMS system but I have ' > not been able to track them down yet.  > < > On my Alpha Server DS20E running VMS 7.2 I get this error:) >  (This machine is not part of cluster )  > $ : > %%%%%%%%%%%  OPCOM   9-FEB-2001 16:58:27.63  %%%%%%%%%%%$ > Message from user SYSTEM on GC02BS@ > Event: Too Few Servers Detected from: Node LOCAL:.GC02BS DTSS,  J It's DTSS (part of DECnet Phase V), the Network Time Services, complainingG that - following the default configuration - it was to find three other 0 nodes which would supply time information to it.  % To disengage DTSS, put the statements      $ RUN SYS$SYSTEM:NCL   DISABLE DTSS
   DELETE DTSS    $    in your SYSTARTUP_VMS.   cu,    Martin --J One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer7 One OS to find them           | work: mv@pdv-systeme.de J One OS to bring them all      |   http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------  % Date: Sat, 10 Feb 2001 11:23:36 +0000 ) From: Christof Brass <brass@infopuls.com> < Subject: Re: Firmware upgrade for PWS 600au to use a ZLXp-L1, Message-ID: <3A8524B8.33CAC0DF@infopuls.com>   Larry Kilgallen wrote: > Z > In article <3A848A63.49892BFF@infopuls.com>, Christof Brass <brass@infopuls.com> writes: > C > > Okay, the Mac offers a real convenient way of using multiple (?  > > - at least two) screens. > 8 > The Macintosh software will handle at least 6 screens. > H > How many PCI slots are in any particular Macintosh model is a separate > issue. > P > ==============================================================================P > Great Inventors of our time: Al Gore -> Internet; Sun Microsystems -> ClustersP > ==============================================================================  @ This is a strange thing. I always regarded one aspect of writing> software as a decision between the one and the many case (1 vs; n). I can't imagine how a system can work for more than one  (logically) and stop at 6.< Like my question about the limitation in the number of VMS's< ACEs where the answers induced that there were only physical@ limits like time needed to process the list of ACEs (but you can> decide if you want to spend that much time) or space needed to store them etc..   ------------------------------  % Date: Sat, 10 Feb 2001 08:02:56 -0500 2 From: "Richard B. Gilbert" <DRAGON@compuserve.com>< Subject: Re: Firmware upgrade for PWS 600au to use a ZLXp-L17 Message-ID: <200102100803_MC2-C4F8-A2DF@compuserve.com>   & Message text written by Christof BrassA >This is a strange thing. I always regarded one aspect of writing > software as a decision between the one and the many case (1 vs; n). I can't imagine how a system can work for more than one  (logically) and stop at 6.< Like my question about the limitation in the number of VMS's< ACEs where the answers induced that there were only physical@ limits like time needed to process the list of ACEs (but you can> decide if you want to spend that much time) or space needed to store them etc.. <   J You are thinking linearly.   Think interactions instead.  One graphics ca= rdH interacts with the system.  Two graphics cards interact with each other;J i.e. widgets moving from one to another.  I suspect that the complexity m= ayJ be 2^N or N! (factorial).  Those of you who are mathmatically inclined wi= llD probably see immediately that I'm not!!  Clearly, if you just want NJ independent X-windows servers, you can buy N X-terminals or N single head= edJ workstations and drive them from a honking BIG machine.  It's not the sam= e $ thing as an N headed workstation.  =    J I also suspect that there are issues of interrupts, memory bandwidth, PCI=  J slots, power, etc, etc, etc.  I suppose that Compaq could build a machine=  F that would handle sixteen graphics heads but who would buy such?  MostJ users find one head sufficient.  A few, probably less then one percent of=  J all workstations, would like two.  What percentage would want, and pay fo= r,F a sixteen headed workstation?  Would it make business sense to design,H manufacture, and write sofware for a sixteen headed machine if you could only sell five of them?    ------------------------------  % Date: Sat, 10 Feb 2001 12:12:18 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)< Subject: Re: Firmware upgrade for PWS 600au to use a ZLXp-L1L Message-ID: <rdeininger-1002011212190001@user-2ivebpu.dialup.mindspring.com>  W In article <3A8524B8.33CAC0DF@infopuls.com>, Christof Brass <brass@infopuls.com> wrote:p   > Larry Kilgallen wrote:: > > The Macintosh software will handle at least 6 screens. > > J > > How many PCI slots are in any particular Macintosh model is a separate
 > > issue.    B > This is a strange thing. I always regarded one aspect of writing@ > software as a decision between the one and the many case (1 vs= > n). I can't imagine how a system can work for more than onec > (logically) and stop at 6.  U He didn't say it stops at 6.  I guess 6 might be the most he has seen or heard about.s   From what I have seen on macs, the OS already knows how to handle very strange screen layouts.  I think the screens have to be logically contiguous, but I'm not even positive about that.  & I have seen a screen layout like this:   ------ |    | |    |  ---------- |    |  |        | ---------        | |       |        | |       |        | |       |        | ------------------  l... with one of the screens B/W and the others color.  Rectangular windows of pretty much arbitrary size could be moved pretty much anywhere on this set of screens.  So for example, a window might be on 1, 2 or 3 of the screens at once, with an odd-shaped part of the window not drawn at all.  Each part of the window is rendered correctly for the screen it is on.  Offhand, I don't see anything about >6 screens that would make drawing windows any more complex than this example.  So I expect it would work with more.  There might be a software limit of 16 or 256 screens.  But on any real Mac, you will run into hardware/slot/memory/power limits much sooner.  (I don't know if Mac device software is particularly adept at dealing with PCI/PCI bridges, and I doubt you'd get even 8 slots in a PCI bus without bridging.)   -- h Robert Deininger rdeininger@mindspring.com    ------------------------------  # Date: Sat, 10 Feb 2001 12:01:41 GMT = From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)o Subject: Re: Free space ?a0 Message-ID: <009F76B7.C8A63123@SendSpamHere.ORG>  ^ In article <wv_g6.601$uY2.12066@news2-hme0>, "John Griffiths" <johngriffiths@ic24.net> writes:H >Free space is shown in blocks, how do I work out the % of the full disc >space of 9.1gb ?  >Thanks in anticipation. >a >C  %  ((MAXBLOCK-FREEBLOCKS)/MAXBLOCK)*100      --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMy            eO city, n., 1. a place where trees are cut down and streets are named after them.U   ------------------------------    Date: 11 Feb 2001 00:32:02 +0800, From: Paul Repacholi <prep@prep.synonet.com>" Subject: Re: GS build, no config.c- Message-ID: <87g0hmts71.fsf@prep.synonet.com>t  4 mathog@seqaxp.bio.caltech.edu (David Mathog) writes:  ^ > In article <87r9171y8x.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes: > >i@ > >I grabbed the recently referenced version of GS the other day, > >and went to build it. But, no config.c...  < Sorry, looking at the wrong window and tired when posting :(   Make the GhostVIEW, not GS.e   -- 1< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda..@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------   Date: 10 Feb 2001 13:44:37 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)A; Subject: Re: How do you make a pointer to another directory 0 Message-ID: <963gk5$628$1@n.ruf.uni-freiburg.de>  j In article <3A849A10.50CE@nospam.chemistry.ucsc.edu>, Todd Wipke <wipke@nospam.chemistry.ucsc.edu> writes:@ >Installing Multinet with the new file structure requires makingD >a directory pointer to the vax_common directory in each homogeneousA >cluster satellite node, BUT they don't describe how to do it.  IeD >tried copying the syscommon link in the first node's directory, but >just got an empty directory.  t >sD >Could someone email me a com file that does this or describe how to@ >create a link?  Remove the nospam from my email address for the+ >correct email address.  Thanks in advance.-  > Normally Multinet does this during installation. Otherwise see   HELP SET FILE/ENTERv   Regards,    Christoph Gartmannc  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, FRG                                               |H +--------- http://www.immunbio.mpg.de/home/english/menue.html ---------+   ------------------------------  % Date: Sat, 10 Feb 2001 12:02:39 +0100-> From: "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr>D Subject: how to determine the initial queue a batch was submitted on. Message-ID: <9636ts$7ue$1@reader1.imaginet.fr>   Bonjour  tous !  9 I need to know on which queue a batch has been submitted,u: even if it has been requeued somewhere else for execution.  : The purpose is to enable the batch login procedure to load/ specific logicals names according to a context.   ; We originaly thought that the queue name would enable us to 5 determine the context, but some jobs were requeued... 8 Jobs are queued from within cobol programs, so we need a9 solution with minimum coding .. (like an external define)c9 We cannot use P1..P8 because they are already beeing used  by the batch procedures.   Cordialement Jean-Franois Marchalo X9000 - LYON (FR)i   ------------------------------   Date: 10 Feb 2001 13:50:48 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)RH Subject: Re: how to determine the initial queue a batch was submitted on0 Message-ID: <963gvo$628$2@n.ruf.uni-freiburg.de>  o In article <9636ts$7ue$1@reader1.imaginet.fr>, "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr> writes:  >Bonjour  tous !  >e: >I need to know on which queue a batch has been submitted,; >even if it has been requeued somewhere else for execution.l >Y; >The purpose is to enable the batch login procedure to loadp0 >specific logicals names according to a context. > < >We originaly thought that the queue name would enable us to6 >determine the context, but some jobs were requeued...9 >Jobs are queued from within cobol programs, so we need a : >solution with minimum coding .. (like an external define): >We cannot use P1..P8 because they are already beeing used >by the batch procedures.   K Why not use the name of the job? Or just add a number or letter to the nameuI or even to one of the parameters. Of course you'll need a table somewheretN in order to have the relationship between these single characters and the real queue names.   Regards,    Christoph Gartman  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, FRG                                               |H +--------- http://www.immunbio.mpg.de/home/english/menue.html ---------+   ------------------------------  % Date: Sat, 10 Feb 2001 11:19:31 +0000e) From: Christof Brass <brass@infopuls.com>r8 Subject: Re: It's the end for VMS and other Wild Rumours, Message-ID: <3A8523C3.6747F94B@infopuls.com>  " Shane.F.Smith@Healthnet.com wrote: >  > Hmmmm, "no soup for YOU!". > J > Ok, English spellings are real strange, but since they are English, theyK > are at least /definitively/ strange. Being English we can decide to spell K > our words any way we wont tooh. :) And that Unix crack is even worse thana > the original American crack. > I > Ok, anyone who wants to continue this, let's take it off line. I'm sureaE > c.o.v. don't want to hear it any more than they want another Andrew% > Harrison tirade. >  > ShaneY >  > Hi,2 > ( > this is Christof's KEYBOARD typing ... > A > English coffee is even worse because the meals don't make it up 	 > for it. A > You may speak real English, but unfortunately you don't *write*dA > it because it can't - the writing of English is severely broken87 > like the UNIX commands. Please fix that first ... ;-)2  . Sorry, do you know the "ignore thread button"?@ If you look at the huge number of message related to an election= end of last year in a country which is politically completelyd@ rotten I wonder if you think that topic was more interesting for or related to this NG.? Basically you missed my point: you can define whatever spellingt? system you want but it is as smart as the UNIX way for definingi= a whole set of completely unrelated commands and options. The 8 English spelling has lost one basic attribute: the clear? relationship between spelling and pronounciation. If you have ap= collection of exeptions or ambiguities you waste your time ins learning them.> Thanks anyway for teaching me what other people are interested in ...   ------------------------------  % Date: Sat, 10 Feb 2001 12:52:25 +0000S  From: steven.reece@quintiles.com8 Subject: Re: It's the end for VMS and other Wild RumoursH Message-ID: <OF67003A68.C999ACA3-ON802569EF.00464EB9@qedi.quintiles.com>  & You mean he even beat Status Quo?  ;-)  H Mind you, if you write down the chord structure to something like "FortyK Five Hundred Times" or "Roadhouse Blues" you probably get the Unix commandsr/ of the day for about the next three weeks! :-))w   Steve (who is a Quo fan)  K Christof Brass (brass at infopuls dot com) wrote and quoted Tim Llewellyn :/ >>>i= > Yeah, Dylan invented rap 20 years before Grandmaster Flash.b6 > But he could play guitar and do it at the same time.  @ :-) play guitar? I defined a new standard of what guitar playing9 is otherwise he wouldn't regarded as "playing guitar". He#> established an unbeaten record: the most songs with the fewest different chords.<<<   ------------------------------  % Date: Sat, 10 Feb 2001 11:45:48 +0000g) From: Christof Brass <brass@infopuls.com>f( Subject: Re: MMS/MMK Ghostscript problem, Message-ID: <3A8529EC.3A8D6407@infopuls.com>  " Shane.F.Smith@Healthnet.com wrote:   [snip]  M > something's gone pear-shaped. Sometimes one trick fits, sometimes the other 7 > does. You pays your money, and you takes your choice.;*                ^                         ^*                |                         |  5 Is this correct English? (sorry, couldn't resist ...)#   >  > Shanen   ------------------------------  % Date: Sat, 10 Feb 2001 11:52:02 -05004  From: John Santos <JOHN@egh.com>( Subject: Re: MMS/MMK Ghostscript problem/ Message-ID: <1010210114604.46493A@Ives.egh.com>8  * On Sat, 10 Feb 2001, Christof Brass wrote:  $ > Shane.F.Smith@Healthnet.com wrote: >  > [snip] > O > > something's gone pear-shaped. Sometimes one trick fits, sometimes the other 9 > > does. You pays your money, and you takes your choice.i, >                ^                         ^, >                |                         | > 7 > Is this correct English? (sorry, couldn't resist ...)  >  > > 	 > > Shane-  B I just saw "The African Queen" the other day, and Humphrey Bogart,J pretending to be English, says this to Katherine Hepburn (also pretending,D a little more successfully, to be English.)  It's an old expression,A but I always assumed it was American.  And, no, it is not correct1# English (or even correct American.)p   -- h John Santosi Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  # Date: Sat, 10 Feb 2001 09:48:36 GMTr From: Dirk Munk <munk@home.nl>Y Subject: Re: Not Being Published: OpenVMS and the Internet (was: Re: Notbeingpublished: OI' Message-ID: <3A850E72.8B9A11A4@home.nl>n   Dean Woodward wrote:   > Dirk Munk wrote: > >7 > > But now a small question: Apache has its own user databases (if you want to use secure access). Is it possible to use the VMS authorize user database instead ?2 > H > The latest CSWS release claims to support it- I haven't tested it yet.  T You are absolutely right. We only only looked at the Apache documentation. Thanks !!     >u >c > <disclaimer>F > About 500 people are about to tell you how dangerous that is becauseA > it'll pass a user's password in cleartext.  You've been warned.M > </disclaimer>T   ------------------------------  % Date: Sat, 10 Feb 2001 12:38:40 +0000n  From: steven.reece@quintiles.com Subject: Re: Open VMS CloninggH Message-ID: <OFA7ED894F.F1954540-ON802569EF.00452D6F@qedi.quintiles.com>  H This is something I've been playing with in recent days for a completely different project.....G The "minimum" install that is generated using the PCSI procedure that's F optionally fired off on Alpha when one uses @sys$update:stabackit willH still attempt to write to files, even if they're only the error log dump) file, system dump file and the page file.a   Steve.: Robert Deininger (rdeininger at mindspring dot com) wrote:K >>>I dunno.  VMS certainly has a lot of files open, and I think quite a fewaH are written.  What if some e-mail arrives?  That will cause some writes.  I How about putting a "minimal" VMS installation on the disk, in a separatePG system root?  (This is only supported on alpha, IIRC.)  If you boot the G minimal VMS (which I haven't done for a while), you may get a read-only=J environment similar to a CD.  That would certainly be safer that a regular: disk boot.  And faster than a CD boot, on most systems.<<<   ------------------------------  # Date: Sat, 10 Feb 2001 13:59:52 GMT 8 From: Veli =?iso-8859-1?Q?K=F6rkk=F6?= <korkko@decus.fi> Subject: Re: set character set? ( Message-ID: <3A8504C8.4D5ECF23@decus.fi>  4 DEC MCS is very close to ISO-LATIN-1. ISO-LATIN-9 is: almost same but contains the new euro symbol. Check if you9 can setup your printers to use ISO-LATIN-1 character set.    _velik   Fred A G wrote:o > < > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3A81F74E.DFFC749C@videotron.ca... > > Fred A G wrote:-A > > > I'm having some problems printing from VMS 7.2-1 via telnetm > symbiont. I > > > The character symbols don't come out right (using only the standardaH > > > device control library). I guess the character set on the VMS side > isC > > > some DEC/VT variant, but how do I find out exactly which one?l > >r > >e+ > > Which symbols are you trying to print ?i > F > Character symbols with code in the 128-255 range (the reverse telnet* > printing is not the cause here, though). > ) > > And where are the documents generatedt > > using what software ?  > >w > J > On the VMS side, by hosted application software or with EDIT. All "plain > text" files. > ? > > Is the printer in postscript mode, or some HPCL variation ?  > ( > Could be PS, PCL or some line-printer. > 	 > Regardsl > /Fad   ------------------------------  % Date: Sat, 10 Feb 2001 15:54:45 +0100s+ From: "Fred A G" <nospam@allowed.localhost>t Subject: Re: set character set?e3 Message-ID: <rCch6.423$hi2.673@nntpserver.swip.net>e  0 "Veli Krkk" <korkko@decus.fi> wrote in message" news:3A8504C8.4D5ECF23@decus.fi...6 > DEC MCS is very close to ISO-LATIN-1. ISO-LATIN-9 is< > almost same but contains the new euro symbol. Check if you; > can setup your printers to use ISO-LATIN-1 character set.i  E Perhaps that's the "best bet". I guess most printers have support for 3 Latin-1 (or some close enough home-brewed variant).r   Regardsy /Fad   ------------------------------   Date: 10 Feb 2001 10:27:08 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: Status of EV70 Message-ID: <96351s$bjc$1@pegasus.csx.cam.ac.uk>  D In article <rW1h6.3876$ZJ5.324048@newsread2.prod.itd.earthlink.net>,. H.W. Stockman <stockman3@earthlink.net> wrote: >y> >Now you have piqued my curiosity.  Didn't the original IBM RTH >use a Nat Semi 3208x (I'm fuzzy on whether x=1 or 2) for the FPU?  That< >FPU was the choice for many 68000-based systems, before the: >68881 came out.  I think the RT actually switched over to# >a 68881 shortly before its demise.   B I don't know, but it was apparently not so much the floating-pointE chip that was so slow, as the wet string it was connected to the mainr CPU by.   6 >Do you really mean emulation, or do you mean basic FPM >calculation, using Motorola's software (integer instruction)  "fast floating  >point"w3 >library?  I recall benchmarks that showed an 8 MHzI' >68000, with FFP, beating a 5 MHz 8087.d  C I thought that it was doing it to full precision, but it was a longe5 time ago, and I can't remember whether I tested that.b  : >The problem, was that the 8087 was doing the calculations> >to at least 64 bits, and was achieving a Savage accuracy test? >score of about 1 part in a trillion.  The FFP library achieveds; >an accuracy score of 1 part in 7.  Fast, just not correct.c  D Yes, but the time has never been in the precision, but the overheadsB of getting into the code in the first place.  We can see this evenC today, with most modern chips degrading by factors of hundreds when-; they have to take an interrupt to handle exceptional cases.   D I really DO mean that I could write a fully-conforming software IEEEB library and a benchmark, and my library would beat the hardware onD most modern machines.  All right, it would be 'cheating' because theE benchmark would be carefully chosen to concentrate on the exceptionalm) cases, but this doesn't change the facts.,  = No, I am not going to.  It would be a damn-fool activity ....r  F The other point is to remember that single precision (i.e. 6-7 digits)B was not totally stupid on a machine the size of the 68000 or 8087.A There were quite a lot of calculations where you ran out of spaceiC before you ran out of precision.  Now, today, the situation is veryo5 different and there are many fewer such calculations.p     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QG, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679    ------------------------------  % Date: Sat, 10 Feb 2001 13:46:50 -00001& From: "Thomas Womack" <tom@womack.net> Subject: Re: Status of EV7/ Message-ID: <963gme$ldk$2@newsg3.svr.pol.co.uk>   * "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote  B > The Hitachi SR2201 has 4 MB/sec per MFlop, all the way from main@ > memory (actually it bypasses cache in pseudovectorising mode).B > Experience is that this was more than adequate (except for a few> > inner loops), but that dropping below 2 starts to be a major7 > bottleneck.  What is the Itanium supposed to deliver?-  I Well, the bottleneck for current microprocessors is always the front-sidetD bus, and the Itanium documentation I have says the front-side bus isE 2133MB/s, significantly slower than Intel's own P4, whilst the FPU isoI dual-issue and running at >666MHz. Well, the lowest speed grade is only a 3 little under your target in a uni-processor system.i  L Are there any little birds out there prepared to conjecture about McKinley'sD FSB? "Quick, steal Willamette's" is I hope a reasonable lower bound.  K By your measure, the only modern microprocessor not majorly bottlenecked isiI the EV7, and that's significantly more vaporous than the Itanium. Oh, and B possibly the CyrixIII, but that runs at 500MHz with a single-issue/ unpipelined FPU with several cycles of latency.    Tom5   ------------------------------  # Date: Sat, 10 Feb 2001 14:32:37 GMT # From: Chris Morgan <cm@mihalis.net>  Subject: Re: Status of EV71 Message-ID: <874ry2d34u.fsf@dumpster.mihalis.net>-  ( "Thomas Womack" <tom@womack.net> writes:  M > By your measure, the only modern microprocessor not majorly bottlenecked istC > the EV7, and that's significantly more vaporous than the Itanium.g  B It's true that EV7 is still vapour, but it does't strike me as theF Thing That Should Not Be as much as Itanium. If it really is the 21264E with cache memory, memory controller, I/O controller, comms links etc_; on-chip, I doubt it will be _quite_ so late as the original,? 21264. Plus we know the fab partners for it have the technologys
 necessary. -- eH Chris Morgan <cm at mihalis.net>                  http://www.mihalis.net        Temp sig. - Enquire within   ------------------------------   Date: 10 Feb 2001 15:14:10 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: Status of EV70 Message-ID: <963ls2$p2n$1@pegasus.csx.cam.ac.uk>  / In article <963gme$ldk$2@newsg3.svr.pol.co.uk>,d% Thomas Womack <tom@womack.net> wrote:-+ >"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote- >-C >> The Hitachi SR2201 has 4 MB/sec per MFlop, all the way from main4A >> memory (actually it bypasses cache in pseudovectorising mode). C >> Experience is that this was more than adequate (except for a fewt? >> inner loops), but that dropping below 2 starts to be a major 8 >> bottleneck.  What is the Itanium supposed to deliver? >uJ >Well, the bottleneck for current microprocessors is always the front-sideE >bus, and the Itanium documentation I have says the front-side bus isrF >2133MB/s, significantly slower than Intel's own P4, whilst the FPU isJ >dual-issue and running at >666MHz. Well, the lowest speed grade is only a4 >little under your target in a uni-processor system.  E Not quite, as I did mean per MFlop, and I am counting fused multiply/eD add instructions as two operations.  This gives the Itanium only 0.7 MB/sec per MFlop.c  M >Are there any little birds out there prepared to conjecture about McKinley'suE >FSB? "Quick, steal Willamette's" is I hope a reasonable lower bound.eA I can conjecture :-)  As it would be based on a position of totalaC ignorance (and not merely near-total ignorance), I shall spare you.o9 I should be extremely interested to know the answer, too.b  L >By your measure, the only modern microprocessor not majorly bottlenecked isJ >the EV7, and that's significantly more vaporous than the Itanium. Oh, andC >possibly the CyrixIII, but that runs at 500MHz with a single-issuem0 >unpipelined FPU with several cycles of latency.  @ Yes, a good many microprocessor designers have told me that theyC know their designs are seriously bandwidth limited for vectorisablea? (i.e. large array) applications, but that the cost of providing C enough bandwidth is prohibitive.  Sometimes blocking algorithms cane help, sometimes they can't.i     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QG, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679    ------------------------------   Date: 10 Feb 2001 17:31:57 GMT From: jfc@mit.edu (John F Carr)  Subject: Re: Status of EV7A Message-ID: <3a857b0d$0$17484$b45e6eb0@senator-bedfellow.mit.edu>y  D In article <rW1h6.3876$ZJ5.324048@newsread2.prod.itd.earthlink.net>,. H.W. Stockman <stockman3@earthlink.net> wrote: >r6 >"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message+ >news:961lbq$3nn$1@pegasus.csx.cam.ac.uk...o6 >> In article <GQ75cTCbCNfr@eisner.encompasserve.org>, >[...]E >> Not necessarily.  When the original IBM RT came out, I benchmarked-C >> it at 25% faster at floating-point than an 8 MHz 68000 that used F >> emulation.  There were already 10 MHz 68010s around, and they wouldG >> have beaten the RT handsomely, even without floating-point hardware.  > > >Now you have piqued my curiosity.  Didn't the original IBM RTH >use a Nat Semi 3208x (I'm fuzzy on whether x=1 or 2) for the FPU?  That< >FPU was the choice for many 68000-based systems, before the: >68881 came out.  I think the RT actually switched over to# >a 68881 shortly before its demise.o  D The original RT had a 170 ns cycle time, load and store took about 5A cycles minimum, and an optional NS-<mumble> math coprocessor on aeG separate board on the processor bus.  Early AOS (BSD) offered a librarygD that you could link to use the math hardware.  Later AOS had a weirdE software FP model that pretended there were 8 high precision floatingdC point registers, but with the early FPA one of them didn't actually C exist in hardware.  (I think it had 24 single precision registers.)t  C The "APC" RT (model 115/125) ran at 10 MHz, allowed two outstandingaE loads or stores for a significant decrease in CPI, and had an onboardYF 68881.  I consider this the major model of the RT.  I used to know theG as-installed 68881 instruction timings accounting for the odd interface F hardware.  There was a glue chip that handled the Motorola coprocessorC protocol for the common cases but it was still slower than the sametC chip used with a 68020.  On the other hand, it was competitive withw+ contemporary 68K Suns for integer programs.t  ? In the model 135 the processor ran at about 12 MHz, onboard RAMiA displaced the 68881, and the FPU moved back onto a separate card.u   -- t     John Carr (jfc@mit.edu)g   ------------------------------  % Date: Sat, 10 Feb 2001 09:11:27 +0100   From: Paul Sture <paul@sture.ch>. Subject: Re: Suggested upgrade to VMS (DECnet)+ Message-ID: <VA.00000297.31aa5f04@sture.ch>n  < In article <3A8318E2.E35F5C57@virgin.net>, Alan Greig wrote:' > From: Alan Greig <a.greig@virgin.net>l > Newsgroups: comp.os.vmsy0 > Subject: Re: Suggested upgrade to VMS (DECnet)' > Date: Thu, 08 Feb 2001 22:08:34 +0000  >  >  >  > "antonio.carlini" wrote: >  > > "David J. Dachtera" wrote:N > > > Actually, what I'd like to see is a "management layer" on top of all theJ > > > NCL and other nonsense. We need a way to do things (so it's actually= > > > POSSIBLE!) that doesn't get in the way of productivity.d > >n2 > > I was told (a few years ago) that the original5 > > plan was to have a GUI front end to NCL (possiblys" > > a cut down version of DECmcc). > > 9 > > Obviously this never happened ... whether it ever wasf% > > a real plan or not, I don't know.! > >w > N > I believe it said this somewhere in the original DECNET VAX Extensions earlyI > Phase V release documentation. I remember when that extremely large boxGM > arrived and my horror when I discovered that it really was full of manuals.- > What? Just for DECNET!...w > E You were lucky! On my first two encounters with Phase V, much of the yN documentation had gone missing (presumably folks had taken it home to try and O get to grips with it). IIRC the documentation CDs I had at the time refused to 12 let me in because I didn't have an Ultrix license!   ___f
 Paul Sture Switzerlando   ------------------------------  % Date: Sat, 10 Feb 2001 13:17:07 +0100m' From: benchabane <benchabane@wissal.dz>o Subject: Re: TK70 manual6 Message-ID: <200102101217.f1ACH7f14706@mail.wissal.dz>   Not scanned , sorry !!  # At 10:05 08/02/01 +0800, you wrote: 2 >Does anyone out there have a scanned TK70 manual? > 
 >Thanks, Tim.| > E >-------------------------------------------------------------------- D >"You can download an atmosphere and dial up a groove, but there's aC >certain magic when three musicians and a dyslexic get together andi >play in a room."- >   -- Bono1E >--------------------------------------------------------------------: >    ------------------------------   Date: 9 Feb 01 06:48:05 PSTh From: mckinneyj@cpva.saic.come( Subject: Re: USAGE.DAT and missing files( Message-ID: <Le1YAURke1Au@cpva.saic.com>  ) In article <95v8qr$joo$1@nnrp1.deja.com>, -  Edward Heller <ejheller@my-deja.com> writes: 5 > Here is a puzzler, if anyone can take a stab at it:u > H > I recently had a disk that approached the max files and started givingB > me errors. I ran ANALYZE/DISK and solved the problem, however inE > reviewing the options for ANALYZE, I noticed the option to create aoI > USAGE.DAT file. The system documentation does not provide a descriptionn@ > of the file, however, being persistent, I determined the basicJ > structure. I deciphered enough to get it into an editable file. Here are2 > the questions I have, if anyone has the answers:G > 1) What is the structure of the binary header data on each record andB> > where might I find documentation on the meaning of the data?  = Extracting a comment that I wrote in an ancient piece of coder  C ; See supplemental section D of the  Analyze/Disk_Structure UtilitycC ; in System Management Volume 2 of the VMS V5.n  documentation  fori* ; complete description of DISK_USAGE file. ;sC ; Simply, the first record in a USAGE file is a disk identificationa= ; record and all succeeding records are file summary records.l6 ; conditionally followed by variable length packet(s).    $ The record layout is defined as such  7 $ lib/extr=$usgdef/mac/out=sys$output sys$share:starlets         .MACRO  $USGDEF,$GBL         $DEFINI USG,$GBL $EQU    USG$K_IDENT     1- $EQU    USG$K_FILE      2a $EQU    USG$K_IDENT_LEN 61 $EQU    USG$C_IDENT_LEN 61 $EQU    USG$S_USGDEF    61 $EQU    USG$B_TYPE      0- $EQU    USG$L_SERIALNUM 18 $EQU    USG$S_STRUCNAME 12 $EQU    USG$T_STRUCNAME 5n $EQU    USG$S_VOLNAME   12 $EQU    USG$T_VOLNAME   17 $EQU    USG$S_OWNERNAME 12 $EQU    USG$T_OWNERNAME 29 $EQU    USG$S_FORMAT    12 $EQU    USG$T_FORMAT    41 $EQU    USG$S_TIME      8u $EQU    USG$Q_TIME      53 $EQU    USG$K_FILE_LEN  423  $EQU    USG$C_FILE_LEN  423h $EQU    USG$S_USGDEF1   423m $EQU    USG$L_FILEOWNER 1e $EQU    USG$W_UICMEMBER 1g $EQU    USG$W_UICGROUP  3g $EQU    USG$L_ALLOCATED 5o $EQU    USG$L_USED      9o $EQU    USG$W_DIR_LEN   13 $EQU    USG$W_SPEC_LEN  15 $EQU    USG$S_FILESPEC  406hB $EQU    USG$T_FILESPEC  17              ;  File spec "[dir]nam.typ         $DEFEND USG,$GBL,DEF
         .ENDMr      I > 2) Why does the USAGE.DAT file have more records (i.e,, files) than areiI > shown with a DIR/GRAND [*...] on the disk? Not only that, but there area6 > more directory records in the file than on the disk.B > Note that this is not a system disk and does not have duplicated > directory trees. > 	 > Thanks, 	 > Edward.  >  > -- > Edward Hellere > TransCore, ITS > Atlanta, GA, USA >  >  > Sent via Deja.comc > http://www.deja.com/   ------------------------------  % Date: Sat, 10 Feb 2001 08:12:56 -0500S2 From: "Richard B. Gilbert" <DRAGON@compuserve.com>B Subject: Re: VAX/VMS 7.2 BACKUP/IMAGE behavior when a file is open7 Message-ID: <200102100813_MC2-C4F8-A30C@compuserve.com>e  ( Message text written by Robert DeiningerJ >I kind of doubt that backup is worrying about read order.  The files see= m'J to be accessed in regular directory (alphabetical) order.  Backup DOES op= enJ as many files at once as possible, but I assume it lets the disk and driv= erJ worry about optimizing the ordering.  In fact, I don't see how backup cou= ldH possibly have the info needed to do this.  I think it just makes sure itJ keeps a lot of I/Os queued to the disk, with the goal of keeping the disk=   utterly busy all the time. <s  J         Backup may access the files in *any* order and it makes little or=  H no difference.  My argument was that the blocks in the files need not beJ accessed in sequential order.  The blocks appear to be written to the sav= e J set in sequential order but that does not necessarily have anything to do=  & with the order in which they are read.   ------------------------------    Date: 11 Feb 2001 00:13:30 +0800, From: Paul Repacholi <prep@prep.synonet.com>B Subject: Re: VAX/VMS 7.2 BACKUP/IMAGE behavior when a file is open- Message-ID: <87ofwatt1x.fsf@prep.synonet.com>   4 "Richard B. Gilbert" <DRAGON@compuserve.com> writes:    > Message text written by Ken Ho  F > When doing an /IMAGE backup under VAX/VMS 7.2, if a file is open forE > write when BACKUP gets to it, does it get reliably backed up *as itI > existed at that time*?   M >         Well, yes and no.  The problem is that a large file, say 2Gb, might K > take 15 minutes to backup!  If that file is being written to, what do you-F > have when you are done?  If it's a sequential file and data is beingM > written at the end you might think it's ok but I wouldn't trust it.  BACKUPeK > retrieves data IN THE ORDER IN WHICH THE BLOCKS APPEAR ON DISK.  There iseL > no guarantee that the beginning of the file is retrieved first and the endH > last.  If the file is fragmented it might be retrieved, worst case, in* > something approximating reverse order!!!  A Well, perhaps.  Most SCSI and MSCP disks/controllers will do someeF degree of optimization, generally with some fairness tweek. The commonB optimization is the 'elevator' sequence; go from low to high, thenG go back, repeat forever... So backup could get to the extents in either E order. Add to this, that IO will be in /block chunks, and that if youoD cross a header extent, the IO is split,... Well, you get the idea ;)  ? So the time from backup reading block N+1 from block N could bec( several seconds. Positive or NEGATIVE...   -- e< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.e@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Sat, 10 Feb 2001 12:27:25 -0500i2 From: rdeininger@mindspring.com (Robert Deininger)B Subject: Re: VAX/VMS 7.2 BACKUP/IMAGE behavior when a file is openL Message-ID: <rdeininger-1002011227250001@user-2ivebpu.dialup.mindspring.com>  k In article <200102100813_MC2-C4F8-A30C@compuserve.com>, "Richard B. Gilbert" <DRAGON@compuserve.com> wrote:h    L >         Backup may access the files in *any* order and it makes little or= > J > no difference.  My argument was that the blocks in the files need not beL > accessed in sequential order.  The blocks appear to be written to the sav= > eeL > set in sequential order but that does not necessarily have anything to do= > ( > with the order in which they are read.   Fair enough.  I have watched Backup's progess with SHOW DEV/FILE, and while I have seen Backup with >100 files open at once, they were always near neighbors in the directory-order sense.   Why would Backup presume to know the optimum read order better than the driver and disk?  Backup knows that it's going to read the whole (non empty portion of) the disk; the lower levels don't know that.  But there's never room in RAM to buffer the WHOLE disk, so a global best-read-order optimization doesn't seem possible given the constraint that the save set is in order.  I guess Backup concentrates on a "few" files at once, and the headers for some more files.  In effect he tells the driver, "I'm gonna w  
A look at the source code for Backup would answer this, but the only version I have access to is around VMS 5.2.  That old version is probably irrelevant for today's Backup.  If you've looked at the guts of backup, then clearly you're on the right track and I'm not.  MI have the (dis)advantage of having just read Ken Bates' book "VAX I/O Subsystems: Optimizing Performance", which is now 10 years old.  My mind is slightly dizzy with all the tricks they put into the high-end controllers.  This was before the "cheap and dumb" era of SCSI, so I don't know how much applies to today's storage systems.a   -- f Robert Deininger rdeininger@mindspring.comw   ------------------------------  % Date: Sat, 10 Feb 2001 18:12:40 +0100H  From: Paul Sture <paul@sture.ch>> Subject: Was MMS/MMK Ghostscript problem - Now Humphrey Bogart+ Message-ID: <VA.00000298.3399dffa@sture.ch>e  B In article <1010210114604.46493A@Ives.egh.com>, John Santos wrote: > Newsgroups: comp.os.vmso" > From: John Santos <JOHN@egh.com>+ > Subject: Re: MMS/MMK Ghostscript problem i' > Date: Sat, 10 Feb 2001 11:52:02 -0500e > , > On Sat, 10 Feb 2001, Christof Brass wrote: > & > > Shane.F.Smith@Healthnet.com wrote: > > 
 > > [snip] > > Q > > > something's gone pear-shaped. Sometimes one trick fits, sometimes the othera; > > > does. You pays your money, and you takes your choice.h. > >                ^                         ^. > >                |                         | > > 9 > > Is this correct English? (sorry, couldn't resist ...)o > >  > > >  > > > Shaneo > D > I just saw "The African Queen" the other day, and Humphrey Bogart,L > pretending to be English, says this to Katherine Hepburn (also pretending,F > a little more successfully, to be English.)  It's an old expression,C > but I always assumed it was American.  And, no, it is not correct % > English (or even correct American.)i >bJ Years since I saw that film (or is it already Easter in the US? :-) :-) ).  M Before you mentioned the film, I was thinking along the lines of the London   < Cockney accent (which Hollywood used to get horribly wrong). ___n
 Paul Sture Switzerland-   ------------------------------  % Date: Sat, 10 Feb 2001 12:34:13 +0000M  From: steven.reece@quintiles.com@ Subject: Re: What monitor and keyboad will work with dec3000/400H Message-ID: <OFBA0311D3.743D9347-ON802569EF.004470F1@qedi.quintiles.com>  7 The PMAGD-AA graphics adapter also works under OpenVMS.TK Suitable monitors should be the VRC16 and VRT19.  The VRC21 might also workaC as this is a valid monitor option for the DEC 3000-600 workstation.  Steve.          C rdeininger@mindspring.com (Robert Deininger) on 10-02-2001 04:11:32i   To:   Info-VAX@Mvb.Saic.Com ) cc:    (bcc: Steven Reece/QRED/Quintiles)s  A Subject:  Re: What monitor and keyboad will work with dec3000/400     B In article <017101c09300$8e1c8250$c1b64d0c@nstar.net>, gary prarat" <g.prarat@worldnet.att.net> wrote:  * > I am about to become practicing hobbyist# > (due to layoff)and am acquiring an$ > Dec Alpha 3000/400 running unknown) > version of VMS. Can  anyone please telln% > me what monitor and keyboard I needa) > to use (or point me to the right faq oru > web site to find out).  J The tricky part with keyboards and mice on a DEC 3000 is the adapter cableF you need.  They are often lost before the system gets to the hobbyist.# Here is an example, with a picture:m  C   http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1212015277i   (No, I'm not the seller.)n  - Anything over $15 or so is probably too much.i  J The same keyboards that work on DEC terminals should work on the DEC 3000,G as long as the connector is compatible.  (The keyboard connector on therJ above cable is the telephone-like modular plug.  The round plug is for theI mouse.)  I think LK401 and LK411 work, but that's from memory.  I know ans LK201 will work.  G I don't remember the number of the mouse.  There are a few that work; Is: think any of the DEC mice with the round connector are ok.  > I think the various graphics options all require monitors withK sync-on-green; the cable has 3 BNC connectors (red, green, blue).  The mostvB common (by far) graphics device in a DEC 3000 is the PMAGB-B, alsoG sometimes known as "HX", which works fine with VMS.  If the machine was,H originally a "server", it might not have a graphics card at all.  AFAIK,G the only difference between server and workstation is the name, and the- graphics card.  I Most or all of the older turbochannel graphics won't work with VMS. We've H just been having discussions about the newer, high-end graphics options,I and we don't really know if they work with current VMS, but they probablyd do.h  D As someone else pointed out, you can also use a PC/Mac/whatever withK X-windows, and not put a monitor/keyboard on your DEC 3000.  But that won'tx work for initial configuration.7   Possibly useful links:E    http://www.phys.ufl.edu/~prescott/linux/alpha/dec3000-sysinfo.htmloB    http://www.phys.ufl.edu/~prescott/linux/alpha/dec3000-docs.html=    http://www6.compaq.com/alphaserver/archive/axp/dec3000.pdf6  G The last is the owner's manual for the DEC 3000/600 and /700, which aretH almost identical to your machine (only faster).  I've seen a hardcopy of) the /400 manual, and they are very close.   H Turbochannel options are pretty interchangeable between all the DEC 3000I models.  The 400/600/700/500/800/900 all use the same memory, but the 300a models use different memory.  D If your system has been neglected for a while, it likely will need aJ firmware upgrade and a VMS upgrade.  Having a CD drive makes both of these (much) easier.    H (I should probably gather up all my tidbits, pool with a couple of otherF folks, and make up a FAQ for turbochannel alphas and VMS.  But there's never enough time...)p   -- Robert Deininger rdeininger@mindspring.comn   ------------------------------    Date: 11 Feb 2001 00:18:53 +0800, From: Paul Repacholi <prep@prep.synonet.com>@ Subject: Re: What monitor and keyboad will work with dec3000/400- Message-ID: <87k86ytssy.fsf@prep.synonet.com>   / gary prarat <g.prarat@worldnet.att.net> writes:8  * > I am about to become practicing hobbyist$ > (due to layoff)and am acquiring a $ > Dec Alpha 3000/400 running unknown) > version of VMS. Can  anyone please tella% > me what monitor and keyboard I needp) > to use (or point me to the right faq or  > web site to find out).  ' LK401s work fine, and a VSXXX-GA mouse.t   -- d< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------   End of INFO-VAX 2001.082 ************************