1 INFO-VAX	Fri, 15 Nov 2002	Volume 2002 : Issue 632       Contents: %rms-f-dme on vms 7.3  Re: %rms-f-dme on vms 7.3  Re: %rms-f-dme on vms 7.3  Re: %rms-f-dme on vms 7.3  Re: %rms-f-dme on vms 7.3 # Re: 2002 Worldwide HP OpenVMS Study  Another press release  Re: Another Press release  Re: Atlant's rants...  Re: Atlant's rants...  Re: Atlant's rants...  Re: Atlant's rants...  Re: Atlant's rants...  Re: Atlant's rants...  Re: Atlant's rants...  Re: Atlant's rants...  Re: Atlant's rants...  Re: Atlant's rants... # backup: /since input file qualifier ' Re: backup: /since input file qualifier ' Re: backup: /since input file qualifier ' Re: backup: /since input file qualifier ' Re: backup: /since input file qualifier  Capellas is new WorldCom CEO  Re: Capellas to head up WorldCom Re: CSWS hangs Re: Detecting non-Files-11 disk + RE: DS10 600Mhz Special this month USD 3995 + RE: DS10 600Mhz Special this month USD 3995 ' Re: EVA and SPC and storage competition ' Re: EVA and SPC and storage competition ' Re: EVA and SPC and storage competition : Re: getting file metadata quickly (was: parsing file name)! Re: Getting the user owner string ! Re: Getting the user owner string @ Re: Going OT - cheese and bacon (was:Re: OT: Hamburger Flipping)@ Re: Going OT - cheese and bacon (was:Re: OT: Hamburger Flipping) Re: HP Advocacy Site Re: HP Advocacy Site+ Re: I've come back to build you a better OS + Re: I've come back to build you a better OS + Re: I've come back to build you a better OS + Re: I've come back to build you a better OS + Re: I've come back to build you a better OS + Re: I've come back to build you a better OS 2 Re: Identifying a MOP load client with DECnet/Plus' Re: Identifying an email user (VMSmail) ' Re: Identifying an email user (VMSmail) ' Re: Identifying an email user (VMSmail) ' Re: Identifying an email user (VMSmail)  Re: Jaw dropping EV7 systems Re: Jaw dropping EV7 systems RE: Jaw dropping EV7 systems Re: Jaw dropping EV7 systems Re: Jaw dropping EV7 systems Re: Linking problem - need help  Re: Linking problem - need help  Re: Linking problem - need help  Re: Linking problem - need help  Re: MajorDomo for VMS - Re: New AlphaServer models (ES47, GS1280) ??? - Re: New AlphaServer models (ES47, GS1280) ??? - Re: New AlphaServer models (ES47, GS1280) ??? " Re: OpenVMS books : seeking advice OpenVMS books : seeking advice Re: OT: Hamburger Flipping RE: OT: Hamburger Flipping, Re: parsing file name (refer to time thread) Re: Patch installation1 Re: Postscript printers and Mozilla/CSWB printing - Primary home block anywhere other than LBN 1? 1 Re: Primary home block anywhere other than LBN 1? 1 Re: Primary home block anywhere other than LBN 1? ; Problem with Hunter Goatley's ODS2 win version and MO drive A problem with InfoVAX---please reply to me (not to the group/list)  Re: Remote Console access # Re: Synchronization VMS/Unix clocks # Re: Synchronization VMS/Unix clocks  Re: system calls, RE time  Re: system calls, RE time  Re: system calls, RE time  Re: system calls, RE time  Re: system calls, RE time  Turnaround artist???
 Re: VMS Sales 
 Re: VMS Sales 5 RE: What EV am I running? (family name for processor) ) Re: Why Is ALTPRI Needed to Set Affinity? ) Re: Why Is ALTPRI Needed to Set Affinity? ) Re: Why Is ALTPRI Needed to Set Affinity?  Re: [OT] HP has big cheeses  Re: [OT] HP has big cheeses   F ----------------------------------------------------------------------  % Date: Fri, 15 Nov 2002 10:04:49 -0000 2 From: "Chris Sharman" <chris.sharman@sorry.nospam> Subject: %rms-f-dme on vms 7.34 Message-ID: <ar2go0$nh1$1$8300dec7@news.demon.co.uk>  ' I've seen %rms-f-dme a couple of times. E Now I'm seeing it all the time, for one ordinary captive user, during  sylogin.4 Initially it occurred doing a dir of her mail files.L I added some diagnostics (sh sym, sh log, sh tran, sh proc) and it occurs on them. % It seems there's a leak or something.   D piopages is 675 (default 575). The increase was "Entered by RMS (ASB/ increase)" via modparams - I don't recall when. @ VMS 7.3 Alpha, most required patches up until a month or so ago.   Help!    Thanks,  Chris    Here's the user:B Username: CROSTON                          Owner:  Kathryn CrostonL Account:  ARTDES                           UIC:    [205,13] ([MKTG,CROSTON])< CLI:      DCL                              Tables: DCLTABLES# Default:  DISK$S2020:[MKTG.CROSTON]  LGICMD:   SYS$MANAGER:MENU Flags:  Captive # Primary days:   Mon Tue Wed Thu Fri + Secondary days:                     Sat Sun F Primary   000000000011111111112222  Secondary 000000000011111111112222F Day Hours 012345678901234567890123  Day Hours 012345678901234567890123F Network:  ##### Full access ######            ##### Full access ######F Batch:    ##### Full access ######            ##### Full access ######F Local:    ##### Full access ######            ##### Full access ######F Dialup:   -----  No access  ------            -----  No access  ------F Remote:   ##### Full access ######            ##### Full access ######D Expiration:            (none)    Pwdminimum:  6   Login Fails:     2> Pwdlifetime:         90 00:00    Pwdchange:  15-NOV-2002 09:15> Last Login: 15-NOV-2002 09:16 (interactive), 15-NOV-2002 09:44 (non-interactive) 9 Maxjobs:         2  Fillm:       100  Bytlm:        64000 9 Maxacctjobs:     0  Shrfillm:      0  Pbytlm:           0 9 Maxdetach:    none  BIOlm:       150  JTquota:       4096 9 Prclm:           5  DIOlm:       150  WSdef:         2048 9 Prio:            4  ASTlm:       250  WSquo:         4096 9 Queprio:         0  TQElm:        10  WSextent:     16384 9 CPU:        (none)  Enqlm:      2000  Pgflquo:      50000  Authorized Privileges:   NETMBX       TMPMBX  Default Privileges:    NETMBX       TMPMBX = Identifier                         Value           Attributes -   COE_USER                         %X80010006 -   BUCENTAUR                        %X80010021 -   STOCK                            %X80010017 -   INVPRINTER                       %X8001001D -   ADPLANCHANGE                     %X8001003F -   ADSALESCODE                      %X80010043    Here's the log: $ $ captive = f$environment("captive")( $ remote = (5.eq.f$getjpi("","jobtype")) $ arch := Alpha  $ _reply :== reply $ if (arch.eqs."VAX")  $ else  vax := nL $       if f$search("sys$system:qreply.exe").nes."" then $ _reply :== qreply $ endif F $ _hello_ == 1 - ((f$getjpi("","uaf_flags").and.32)/32) ! NoDisWelcome! $ prog_only = (node.eqs."CYGNUS")  $ grp = 133  $ say :== write sys$output $ phdtfrac2 :== .2 ! .0448F $ phdtfrac7 :== .7 ! .1648 ! last print check (bu_com:bu-picklist.com)% $ phdtfrac9 :== .99999999 ! .23595999 A $ gosub mode_INTERACTIVE        ! BATCH,OTHER,NETWORK,INTERACTIVE  $mode_interactive: $ devname := _FTA104: - $ if (f$loc("WSA",devname).ne.f$len(devname)) 3 $ else  if (f$loc("MBA",devname).ne.f$len(devname)) 
 $       endif  $ endif  $ xterm :== @sys$manager:xterm) $ if f$type(session).nes."" then $ exit 1 = $ if f$trnlnm("DECW$DISPLAY").eqs."" then $ del/sym/glo xterm / $ rights = ","+f$getjpi(0,"PROCESS_RIGHTS")+"," < $ dialup = (f$locate(",DIALUP,",rights).lt.f$length(rights))
 $ bell[0,8]=7 4 $ location == f$getjpi("","tt_accpornam") - "Host: "9 $ location == f$ext(0,f$loc(" Port: ",location),location) D $ if .not. vax .and. ((f$loc("TNA",devname).ne.f$len(devname)) .or.-7                       ("LAT_".eqs.f$ext(0,4,location)))  $ endif  $ gosub node_EQUUS $node_equus: $node_draco:return $! $ if dialup .and..not.captive  $ endif " $ message=f$environment("message")$ $ set message/nofac/nosev/noid/notex
 $ set noon$ $ if _hello_ then $ r sys$system:inq $ if (location.nes."") $ thenD $       x = f$fao("!16AS !16AS !12AS !AS",location,f$cvtime(),user,-D                 f$getdvi("sys$command","device_type_name")-"Series")C $       open/read/write/share=write portuse sys$manager:portuse.dat G $       read/key="compman.ccagroup.co.uk"/error=norecord portuse oldrec  $norecord:  write portuse x  $       close portuse  $ endif 3 $ set message/FACILITY/SEVERITY/IDENTIFICATION/TEXT  $ set control=t C $! Any non-captive user with OPER privilege will be a tape operator  $ if captive.and._hello_ $ then $       tt = 113@ $       if (tt.ne.110).and.((tt.lt.112).or.(tt.gt.114)) then $ r sys$system:inq $       tt = 1137 $       if (tt.ne.110).and.((tt.lt.112).or.(tt.gt.114)) 
 $       endif  $ endif  $ if .not.captive  $ endif  $ set on+ $! Set up check for WordPerfect directories 5 $ scratch=f$parse("[.scratch]",,,,"SYNTAX_ONLY")-".;" = $ define/job wpcorp$scratch DISK$S2020:[MKTG.CROSTON.SCRATCH] + $ if f$parse(scratch) .eqs. "" then cre/dir ! DISK$S2020:[MKTG.CROSTON.SCRATCH] ( $! If user is GOD make mortal by defaultK $ IF .not.captive .and. (f$locate(",GOD,",rights).lt.f$length(rights)) THEN  $ set rights god/disable $! Personal message 1 $ sysmgr_message_file := sys$login:sysmgr.message + $ if f$search(sysmgr_message_file) .nes. ""  $ endif  $! Message of the day  $ local_user = (.not. remote)  $ if .not. remote  $ then $    local_user = "T"  $ else $ endif  $ if local_user  $ then5 $    announce_file := sys$manager:announcements.EQUUS - $    if f$search(announce_file).eqs."" then - <         $ announce_file := sys$manager:announcements.cluster& $    if f$search(announce_file).nes.""	 $    then J $       if _hello_ .or. dialup then $ type SYS$MANAGER:ANNOUNCEMENTS.EQUUS  L **************************************************************************** ****  $  Welcome to the EQUUS Alphaserver  !$  Welcome to the EQUUS Alphaserver  !  G         Any problems, contact the Computer Department, ext 2880-4, 2892   L **************************************************************************** ****  4 $       if captive .and. (user.nes."ARTWORK") .and.-K                         ("SYS$SYSROOT:[SYSMGR]".nes.f$env("default")) then- !                 $ gosub checkmail  $checkmail: ' $ wait_until := 15-NOV-2002 09:41:39.99  $ mailspec := [...]mail*.mai	 $ line := 1 $ if f$search(mailspec).eqs."" then $ goto nomail  $ if user.eqs."CROSTON"  $ then $       sh sym/all/loc3   ANNOUNCE_FILE = "SYS$MANAGER:ANNOUNCEMENTS.EQUUS"    ARCH = "ALPHA"   BELL = "."   CAPTIVE = "TRUE"   DEVNAME = "_FTA104:"2   DIALUP = 0   Hex = 00000000  Octal = 000000000001   GRP = 133   Hex = 00000085  Octal = 00000000205    LINE = ""    LOCAL_USER = "T"   MAILSPEC = "[...]MAIL*.MAI" 4   MESSAGE = "/FACILITY/SEVERITY/IDENTIFICATION/TEXT"   NODE = "EQUUS"	   P1 = "" 	   P2 = "" 	   P3 = "" 	   P4 = "" 	   P5 = "" 	   P6 = "" 	   P7 = "" 	   P8 = "" 5   PROG_ONLY = 0   Hex = 00000000  Octal = 00000000000 2   REMOTE = 0   Hex = 00000000  Octal = 00000000000
   RIGHTS =L ",CROSTON,INTERACTIVE,LOCAL,COE_USER,BUCENTAUR,STOCK,INVPRINTER,ADPLANCHANGE ,ADSALESCODE,"/   SCRATCH = "DISK$S2020:[MKTG.CROSTON.SCRATCH]" 2   SYSMGR_MESSAGE_FILE = "SYS$LOGIN:SYSMGR.MESSAGE"0   TT = 113   Hex = 00000071  Octal = 00000000161   USER = "CROSTON"   VAX = "N" /   VER = 0   Hex = 00000000  Octal = 00000000000 (   WAIT_UNTIL = "15-NOV-2002 09:41:39.99":   X = "pman.ccagroup 2002-11-15 09:41 CROSTON      VT400 " $       sh sym/all/glo   $RESTART == "FALSE"    $SEVERITY == "1"   $STATUS == "%X00030001" #   LOCATION == "pman.ccagroup.co.uk"    PHDTFRAC2 == ".2"    PHDTFRAC7 == ".7"    PHDTFRAC9 == ".99999999"   SAY == "WRITE SYS$OUTPUT" 4   _HELLO_ == 1   Hex = 00000001  Octal = 00000000001   _REPLY == "QREPLY" $       sh tran *    * = ""  (UNDEFINED)  $       sh log/job!/group    $STATUS == "%X100184D4"  $       sh log/group; %DCL-E-CAPTINT, captive account - interactive access denied 4   CROSTON      logged out at 15-NOV-2002 09:41:30.05   ------------------------------  % Date: Fri, 15 Nov 2002 12:58:30 +0100  From: "labadie" <g.g@127.0.0.1> " Subject: Re: %rms-f-dme on vms 7.3* Message-ID: <ar2nhi$26m$1@web1.cup.hp.com>  = "Chris Sharman" <chris.sharman@sorry.nospam> wrote in message . news:ar2go0$nh1$1$8300dec7@news.demon.co.uk...) > I've seen %rms-f-dme a couple of times. G > Now I'm seeing it all the time, for one ordinary captive user, during 
 > sylogin.6 > Initially it occurred doing a dir of her mail files.K > I added some diagnostics (sh sym, sh log, sh tran, sh proc) and it occurs  on > them. ' > It seems there's a leak or something.    Hello   ) check if this is related to Piopages with L http://www.compaq.com/support/asktima/operating_systems/CHAMP_SRC99030800167 1.html  % and if it is related to Ctlpages with L http://www.compaq.com/support/asktima/operating_systems/CY-1021490401-1.html   Regards    Grard   ------------------------------  % Date: Fri, 15 Nov 2002 10:31:40 -0000 2 From: "Chris Sharman" <chris.sharman@sorry.nospam>" Subject: Re: %rms-f-dme on vms 7.34 Message-ID: <ar2iab$pjb$1$830fa79d@news.demon.co.uk>  = "Chris Sharman" <chris.sharman@sorry.nospam> wrote in message . news:ar2go0$nh1$1$8300dec7@news.demon.co.uk...) > I've seen %rms-f-dme a couple of times. G > Now I'm seeing it all the time, for one ordinary captive user, during 
 > sylogin.6 > Initially it occurred doing a dir of her mail files.K > I added some diagnostics (sh sym, sh log, sh tran, sh proc) and it occurs  on > them. ' > It seems there's a leak or something.   < Situation fixed (for her, for now) by tidying her directory.G She had some 5000 or so files (mostly Powerhouse subfiles) in her login I directory, of which just two were mail files. I've no idea why sho log or  sho proc/quo gave rms-f-dme.  I I've seen %rms-f-dme once before recently (submitting a batch job, to the K system account I think), but it wasn't repeated, so I didn't follow up. I'm J now concerned that there's a latent problem here which could cause a bunch* of people to suddenly be unable to log in.   Chris    ------------------------------    Date: 15 Nov 2002 07:39:28 -0600- From: koehler@encompasserve.org (Bob Koehler) " Subject: Re: %rms-f-dme on vms 7.33 Message-ID: <pXbDhwHPfxQX@eisner.encompasserve.org>   i In article <ar2go0$nh1$1$8300dec7@news.demon.co.uk>, "Chris Sharman" <chris.sharman@sorry.nospam> writes: ) > I've seen %rms-f-dme a couple of times. G > Now I'm seeing it all the time, for one ordinary captive user, during 
 > sylogin.6 > Initially it occurred doing a dir of her mail files.N > I added some diagnostics (sh sym, sh log, sh tran, sh proc) and it occurs on > them. ' > It seems there's a leak or something.   B    Somebody's got a SET RMS command that's too big for the way the    system is tuned.    ------------------------------  % Date: Fri, 15 Nov 2002 15:15:51 -0000 ! From: Z  <zarlenga@conan.ids.net> " Subject: Re: %rms-f-dme on vms 7.3/ Message-ID: <uta3t7if4aj1cd@corp.supernews.com>   1 Chris Sharman <chris.sharman@sorry.nospam> wrote: K : I've seen %rms-f-dme once before recently (submitting a batch job, to the    $ help/message dme   DME,  dynamic memory exhausted  5 Facility:     RMS, OpenVMS Record Management Services   I Explanation:  OpenVMS RMS is unable to allocate additional process memory L               for the given request (usually an RMS $OPEN or $CONNECT systemH               service call). Such requests can require a large number of3               buffers, large buffer sizes, or both.   J User Action:  Increase the amount of available process memory, or decrease6               the size or number of requested buffers.  H               For process permanent files (such as DCL OPEN, DCL commandB               procedures using "@filename," SYS$INPUT, SYS$OUTPUT,K               SYS$ERROR, and batch log files), the size of available memory I               is governed by the SYSGEN parameter PIOPAGES. The number of H               buffers and their sizes is governed by the DCL command SETK               RMS. Only 63 process permanent files can be open at once; any D               attempt to open more such files produces this message.  D               For image files, process memory is governed by the UAFI               parameter PGFLQUOTA and the SYSGEN parameter VIRTUALPAGECNT G               (on VAX systems only). The size and number of buffers are H               controlled by the application, the DCL command SET RMS, or               both.   G               Image buffer space can also be controlled with the linker J               option of IOSEGMENT, which can determine the amount of fixedH               RMS memory allocated and the process region in which it isJ               allocated. See the OpenVMS Linker Utility Manual for further               information.   ------------------------------    Date: 15 Nov 2002 23:44:53 +0800, From: Paul Repacholi <prep@prep.synonet.com>, Subject: Re: 2002 Worldwide HP OpenVMS Study- Message-ID: <8765uybzwq.fsf@prep.synonet.com>   & Shane Smith <ssmith@icius.com> writes:  E > I suspect we have to accept the Alpha's gone. Yes, it's better, but A > the team's split up, the momentum's gone, and a lot of time has @ > passed with no progress. To bring it back would cost a fortune? > setting things up again and they'd lose more time while doing 7 > it. Then they'd have to remember where they left off.    D > I would like a "you're bunch of mindless idiots for killing Alpha" > button though.  H They did. The box where they asked about when you would move from Aplha.   "When Hell freezes over."   A I was going to add something like intel seeing the light, but why  waste the space?   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Fri, 15 Nov 2002 11:50:19 -0500 5 From: "Sue Skonetski" <susan.skonetski@hp.nospam.com>  Subject: Another press release* Message-ID: <ar38kj$ans$1@web1.cup.hp.com>  1 LJK/CDROM V1.0 supports RMS attributes on ISO9660   0 November 11, 2002, Cambridge, Massachusetts, USA  H LJK Software announces the release of LJK/CDROM, a new product to format  D ISO9660 volumes on VMS systems (VAX and Alpha, VMS V6.1 and higher).  D Unlike previously available software for formatting ISO9660 volumes,  A LJK/CDROM supports Section 6.10, Record structure, of the ISO9660   B specification, retaining information regarding RMS sequential file  B record structure (fixed or variable length records) on the ISO9660  E volume. Furthermore, LJK/CDROM provides the Digital System Identifier   B honored by VMS V6.1 and higher to support RMS relative and indexed   files from ISO9660 volumes.   B LJK/CDROM will create ISO9660 multivolume sets, either all at once  D (including splitting large files across multiple volumes) or through  4 adding on volumes to an existing ISO9660 volume set.  D Included in the product is a VERIFY function to validate an existing  ? CDROM against the ISO9660 standard, allowing a "second opinion"   > regarding volumes created with other tools. A COMPARE function  @ allows logical comparison between two ISO9660 volumes, primarily  4 to validate that minor differences are really minor.  E In accordance with the ISO9660 standard, LJK/CDROM can create volumes   F with multiple directory hierarchies to adhere to separate restrictions  H on file naming, including Rock Ridge specification of Posix information,  D Microsoft Joliet information, and VMS standard and extended filename  D parsing. Separate hierarchies created in that fashion share a single  $ copy of file data where appropriate.  : Besides honoring the OPENVMS-HOBBYIST PAK for general use,  ? LJK/CDROM also provides no-charge access to the VERIFY function   ? even for commercial users who do not wish to buy a full license    for the product.  C The standard license for LJK/CDROM costs $100 per single concurrent   E user for a perpetual license with a 1 year right-to-use-new-versions.   F Ordering information and full documentation is at http://www.ljk.com .   ------------------------------  % Date: Fri, 15 Nov 2002 12:49:00 -0500  From: norm.raphael@metso.com" Subject: Re: Another Press release? Message-ID: <OFCCB1386F.5AB92F5F-ON85256C72.0061E01F@metso.com>   ( Surely not just "Another Press release!"        E "Skonetski, Susan" <Susan.Skonetski@hp.com> on 11/15/2002 11:51:03 AM   = Please respond to "Skonetski, Susan" <Susan.Skonetski@hp.com>   ' To:    VMS-SIG@LISTSERV.ENCOMPASSUS.ORG  cc: ! Subject:    Another Press release     2  LJK/CDROM V1.0 supports RMS attributes on ISO9660  0 November 11, 2002, Cambridge, Massachusetts, USA  H LJK Software announces the release of LJK/CDROM, a new product to formatD ISO9660 volumes on VMS systems (VAX and Alpha, VMS V6.1 and higher). [Snip]   ------------------------------    Date: 14 Nov 2002 11:57:28 -0800 From: ar14ya@yahoo.com (Ar Ya) Subject: Re: Atlant's rants...= Message-ID: <fbd58baf.0211141157.714629d6@posting.google.com>   d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<2MqdnVc6DMy9U0-gXTWc2Q@metrocast.net>.... > <VAXman- @SendSpamHere.ORG> wrote in message, > news:00A16EA5.2CD954EA@SendSpamHere.ORG... >  > ...  > M > > ... care to comment on who brought upon us that wonderous task every 15th M > > or thereabouts of the month of April?  Take that away and I wouldn't miss = > > any of the aforementioned since none of them apply to me.  > N > Exactly the kind of Libertarian attitude that makes many of us consider many  > of you self-centered assholes. >  > - bill  ; Because I don't want to pay for your pet government program @ (undoubtedly administered inefficiently at best and in a corrupt? fashion at worst), I'm an asshole?  I would like to borrow your B dictionary sometime and see what other words I've misunderstood my whole life.    - self-centered asshole    ------------------------------    Date: 15 Nov 2002 08:03:16 -0600; From: kaplow_r@eisner.encompasserve.org.TRABoD (Bob Kaplow)  Subject: Re: Atlant's rants...3 Message-ID: <9M8q$x0nmmch@eisner.encompasserve.org>   c In article <3DD3E8D9.F572CD7A@mindspring.com>, Atlant Schmidt <atlantnospam@mindspring.com> writes: : > And you still don't see why we need "Single-Payer Health; > Care" in this country, much like the civilised world has?   K That would work well. Then the government gets to decide what care you get, L who gets it, and who doesn't get it. With no way to appeal or go out and buyE what you want if they refuse to provide it. It's your worst HMO fears @ crammed down every one's throat with no other options available.  - I once saw the following on a bumper sticker:   K "National Health Care: The efficiency of the Post Office, the compassion of  the IRS, at Pentagon prices"   That pretty well sums it up.  1 	26-October, 2001: A day that will live in infamy 4 	Support Freedom: http://www.indefenseoffreedom.org/  I         "To enslave men, successfully and safely, it is necessary to have H         their minds  occupied with thoughts and aspirations short of theL         liberty of which they are  deprived.  A certain degree of attainableJ         good must be kept before them." Frederick Douglas, "My Bondage and         My Freedom," 1855    ------------------------------  + Date: Fri, 15 Nov 2002 17:25:23 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>  Subject: Re: Atlant's rants...; Message-ID: <01KOWKXZGHXU9ZLF2M@sysdev.deutsche-boerse.com>   < > > And you still don't see why we need "Single-Payer Health= > > Care" in this country, much like the civilised world has?  > H > That would work well. Then the government gets to decide what care you, > get, who gets it, and who doesn't get it.   F All get that which is medically necessary.  What other real choice is  there?   > With no way to appeal or go I > out and buy what you want if they refuse to provide it. It's your worst A > HMO fears crammed down every one's throat with no other options 
 > available.    @ The "appeal" is called an "election".  No problem if you have a H many-party as opposed to a "two-party" system.  Also, even in countries I with national health care, you can always buy more if you think you need   it.      ------------------------------  # Date: Fri, 15 Nov 2002 16:54:13 GMT ? From: Jim.Johnson@software-exploration.nospam.com (Jim Johnson)  Subject: Re: Atlant's rants...0 Message-ID: <3dd525a6.32782458@news.demon.co.uk>  8 On Fri, 15 Nov 2002 17:25:23 +0100 (MET), Phillip Helbig+ <HELBPHI@sysdev.deutsche-boerse.com> wrote:   = >> > And you still don't see why we need "Single-Payer Health > >> > Care" in this country, much like the civilised world has? >>  I >> That would work well. Then the government gets to decide what care you - >> get, who gets it, and who doesn't get it.   > G >All get that which is medically necessary.  What other real choice is   >there?   D Uhm.  Have you looked at the UK NHS lately?  They've certainly found another way.   >  >> With no way to appeal or goJ >> out and buy what you want if they refuse to provide it. It's your worstB >> HMO fears crammed down every one's throat with no other options >> available.  > A >The "appeal" is called an "election".  No problem if you have a  I >many-party as opposed to a "two-party" system.  Also, even in countries  J >with national health care, you can always buy more if you think you need  >it.    D That isn't going to help very much if your problem is a fast growing% cancer.  Or something of similar ilk.     F After having spent a number of years in the UK and the US, my personalE belief is that the national health care seems to work best up to, but E not beyond, 'gp level' care -- i.e. as a basic healt provision.  Past F that you run into the problem that no government can afford to provide? unlimited health care and no politician wants to be seen as not D providing the same.  Private provision is at least more honest about the actual inequalities.   IMHO.    Jim.   Jim Johnson  Software Exploration, Ltd.) (remove '.nospam' from the reply address)    ------------------------------  + Date: Fri, 15 Nov 2002 17:58:23 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>  Subject: Re: Atlant's rants...; Message-ID: <01KOWM14XQZOA23PDU@sysdev.deutsche-boerse.com>   A > > > > Care" in this country, much like the civilised world has? 6 > > >                                        ^^^^^^^^^L > > > That would work well. Then the government gets to decide what care you0 > > > get, who gets it, and who doesn't get it.  > > J > > All get that which is medically necessary.  What other real choice is 
 > > there? > F > Uhm.  Have you looked at the UK NHS lately?  They've certainly found > another way.  H The discussion was limited to the "civilised world".  :-)  The UK is in F a class by itself in this and many other aspects.  As most Brits will " tell you, it's not part of Europe.  G At the moment in Germany, there is a big discussion about cost-cutting  H in public health care (interestingly, shifting costs to the patients is 7 NOT an item being considered).  There is also a lot of  H party-politics--motivated mud-slinging, so that the opposition plans to G paint a worse picture than actually exists.  As was pointed out by the  B health minister herself, criticism claiming that it is worse than I country X is rather unbelievable from someone who, as most people do, is  A insured for being transferred to Germany in case of illness when  I travelling in country X.  I haven't seen anyone dying because they can't  G get treatment from cancer here.  It's like proportional representation  F with the anti-argument "look at the Weimar republic".  Apart from the I question of whether PR (and the lack of a 5% threshold) was the cause of  D its downfall, one can't ignore that PR functions much better than a 7 two-party (or a 1-party) system almost everywhere else.    ------------------------------  # Date: Fri, 15 Nov 2002 17:14:56 GMT ? From: Jim.Johnson@software-exploration.nospam.com (Jim Johnson)  Subject: Re: Atlant's rants...0 Message-ID: <3dd52af7.34143415@news.demon.co.uk>  8 On Fri, 15 Nov 2002 17:58:23 +0100 (MET), Phillip Helbig+ <HELBPHI@sysdev.deutsche-boerse.com> wrote:   B >> > > > Care" in this country, much like the civilised world has?7 >> > >                                        ^^^^^^^^^ M >> > > That would work well. Then the government gets to decide what care you 1 >> > > get, who gets it, and who doesn't get it.   >> >K >> > All get that which is medically necessary.  What other real choice is   >> > there?  >>  G >> Uhm.  Have you looked at the UK NHS lately?  They've certainly found  >> another way.  > I >The discussion was limited to the "civilised world".  :-)  The UK is in  G >a class by itself in this and many other aspects.  As most Brits will  # >tell you, it's not part of Europe.  > H >At the moment in Germany, there is a big discussion about cost-cutting I >in public health care (interestingly, shifting costs to the patients is  8 >NOT an item being considered).  There is also a lot of I >party-politics--motivated mud-slinging, so that the opposition plans to  H >paint a worse picture than actually exists.  As was pointed out by the C >health minister herself, criticism claiming that it is worse than  J >country X is rather unbelievable from someone who, as most people do, is B >insured for being transferred to Germany in case of illness when J >travelling in country X.  I haven't seen anyone dying because they can't H >get treatment from cancer here.  It's like proportional representation G >with the anti-argument "look at the Weimar republic".  Apart from the  J >question of whether PR (and the lack of a 5% threshold) was the cause of E >its downfall, one can't ignore that PR functions much better than a  8 >two-party (or a 1-party) system almost everywhere else.  B Can a public health service be run better than the UK NHS?  AlmostC certainly, and without personal knowledge of Germany's system, I'll  take your word for it.  A *Must* a public health service be efficient, equal, and effective E (which is the implication I read -- possibly incorrectly -- into your ; prior comment)?  The UK, at least, proves the answer is NO.    Jim.   Jim Johnson  Software Exploration, Ltd.) (remove '.nospam' from the reply address)    ------------------------------  # Date: Fri, 15 Nov 2002 17:00:32 GMT 1 From: LESLIE@JRLVAX.HOUSTON.RR.COM (Jerry Leslie)  Subject: Re: Atlant's rants...7 Message-ID: <QK9B9.1778$Yb.56164@twister.austin.rr.com>   # GreyCloud (cumulus@mist.com) wrote:  : Jerry Leslie wrote:  : > F : > Extending life is nice, but there needs to be QUALITY of life, not : > just quantity. : > F : > Nursing homes are EXPENSIVE and many are places one wouldn't leave : > their worst enemy. : > G : > Medicare has just announced a new site for comparing nursing homes:  : > 1 : >    http://www.medicare.gov/NHCompare/home.asp 1 : >    Medicare.gov - Nursing Home Compare - Home  : > D : > I've already seen some data there that doesn't match my mother's? : > experiences of one of the four nursing homes she's been in.  : > H : > When a large number of the baby-boomers reach nursing home age, theyG : > may find that the suicide chambers in "Soylent Green" an attractive ! : > option to many nursing homes.  : >  : ; : Been thru this with my mother-in-law.  She finally passed > : away in one of those Nursing Homes.  Near the end we managed< : to get Hospice in there to get better care, but we knew it; : was the end for her.  I'm sure I won't get too much older = : before croaking, and doubt I'll ever see a nursing home.  I  : hope.   > The costs of nursing home care is ENORMOUS. Most nursing homes? nickel and dime people to death, instead of including costs for A diapers, rubber gloves, etc. in the room rate, which runs between ( $ 150-$ 250 per day in the Houston area.  ? That's for the nursing homes where residents don't have to race  the rats for their meals.   2 --Jerry Leslie   (my opinions are strictly my own)9   Note: leslie@jrlvax.houston.rr.com is invalid for email    ------------------------------  # Date: Fri, 15 Nov 2002 17:18:26 GMT 1 From: LESLIE@JRLVAX.HOUSTON.RR.COM (Jerry Leslie)  Subject: Re: Atlant's rants...7 Message-ID: <C%9B9.2042$Yb.58768@twister.austin.rr.com>   < Bob Kaplow (kaplow_r@eisner.encompasserve.org.TRABoD) wrote:@ : In article <3DD3E8D9.F572CD7A@mindspring.com>, Atlant Schmidt ' : <atlantnospam@mindspring.com> writes: < : > And you still don't see why we need "Single-Payer Health= : > Care" in this country, much like the civilised world has?  : I : That would work well. Then the government gets to decide what care you oH : get, who gets it, and who doesn't get it. With no way to appeal or go J : out and buy what you want if they refuse to provide it. It's your worst L : HMO fears crammed down every one's throat with no other options available. : / : I once saw the following on a bumper sticker:. : < : "National Health Care: The efficiency of the Post Office, 0 : the compassion of the IRS, at Pentagon prices" :  : That pretty well sums it up. :a  M Perhaps the ethics of corporations and their attitude to workers is better...l  3    o http://money.cnn.com/news/specials/corruption/ )      Special Report: Corporate Corruptione  8    o http://www.chron.com/cs/CDA/story.hts/front/1380381J      HoustonChronicle.com - Enron utility's 'dead peasant' policies rankle  I     "When workers at Portland General die, there's a little more money toh9      spend on the top executives of the Enron subsidiary.   G      The utility has bought life insurance policies on the lives of its I      rank-and-file employees where the company is the beneficiary when annK      employee dies. That money goes for special compensation and retirement 3      benefits for its top executives and directors.r  E      That's a galling realization for workers, many of whom bet theirnG      retirement on Enron stock, which cratered last year as the Houstonl*      energy giant slid into bankruptcy..."  5    o http://www.aflcio.org/paywatch/case_polaroid.htme      AFL-CIO: PayWatch    t
     "Polaroide        [snip]'    mD        * Polaroid cut severance pay for downsized workers and health          benefits for retirees.oI        * Workers, who had owned 15 percent of Polaroid, lost their entired9          Employee Stock Ownership Plan (ESOP) savings..."l    E    o http://abcnews.go.com/sections/wnt/DailyNews/overtime020902.htmlg7      ABCNEWS.com : Workers Sue for Lack of Overtime Payi  J     "Sept. 2 - Even though the United States Census Bureau found one-thirdF      of American laborers worked more than 40 hours a week in the year@      2000, not everyone is being compensated for their toils..."  0    o http://www.fortune.com/articles/209973.htmlD      FORTUNE.COM - Fast Forward - Will the U.S. Fall Behind in Tech?  F     "At last week's Agenda technology conference, many in the audienceG      pricked up their ears when Pat Gelsinger, Intel's chief technologyaJ      officer and head of research, began passionately complaining on stageK      about U.S. technology competitiveness. "Perhaps the current downsizingbG      of the U.S. IT industry is not a temporary thing," Gelsinger said.hJ      "Maybe we are headed for becoming a second-class citizen in the worldG      of IT." He mentioned China and India among countries he said were pA      more committed to IT as a national priority than is the U.S.s  G      Perhaps Intel is more committed to India and China, and intends onk4      making the U.S. a second-class citizen of IT...  M      http://portland.bizjournals.com/portland/stories/2002/08/26/daily56.htmlEL      Intel holds job fairs for 'redeployed' employees, while hiring overseas  nJ     "Intel Corp., Oregon's largest technology employer, is proceeding withK      previously announced job cuts, while hiring software engineers for itse$      development center in India..."      e2 --Jerry Leslie   (my opinions are strictly my own)9   Note: leslie@jrlvax.houston.rr.com is invalid for emailr   ------------------------------  % Date: Fri, 15 Nov 2002 19:12:20 +0100n6 From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne@vajhoej.dk> Subject: Re: Atlant's rants...) Message-ID: <3DD53904.1000100@vajhoej.dk>m   G Everhart wrote:i  > > The term "liberal" originally meant one who favored liberty.% > Jefferson and others were so known.rB > Today however, many of us who believe in those values are calledF > Libertarian, because the word 'liberal' has come to mean an advocateB > of government and social control of the individual, and lost its> > original meaning.  This gets confusing to some because there> > are also "traditionalist" people who believe in governmental/ > control. They are not the same breed of cats.e  % And just to confuse things even more:   6 There are a world outside the US and in most countries/ the liberal versus conservative was replaced byC. social democrats versus liberal & conservative- after World War I ! And the terms liberal and - conservative mean the same to most people anda3 only the very political interested knows the subtle  differences.   Arne   ------------------------------  % Date: Thu, 14 Nov 2002 18:18:32 -0800o" From: GreyCloud <cumulus@mist.com> Subject: Re: Atlant's rants...( Message-ID: <3DD45978.FD243DD3@mist.com>   Jerry Leslie wrote:# > 5 > Atlant Schmidt (atlantnospam@mindspring.com) wrote:l% > : VAXman-, @SendSpamHere.ORG wrote:c > :#H > : > At the rate that the So-so security and medicare systems are beingN > : > pilfered, I'll not see a cent when my turn comes 'round so why shouldn'tO > : > I feel this way?  I'd rather take *my* money and provide for *my* latter- G > : > years welfare myself.  I don't need a top heavy gov't beaurocracy0K > : > skimming off the major-ity of my money providing and me me nothing ofn* > : > any substantial vaule in its return. > :1. > : You obviously have no idea how costly your7 > : "latter years" can be. Good luck finding a guardian . > : who meets your standards *AND* your price,/ > : and hope that when you become decripit, you - > : die quickly instead of lingering long and  > : expensively. > :  > D > Extending life is nice, but there needs to be QUALITY of life, not > just quantity. > D > Nursing homes are EXPENSIVE and many are places one wouldn't leave > their worst enemy. > E > Medicare has just announced a new site for comparing nursing homes:e > / >    http://www.medicare.gov/NHCompare/home.asp / >    Medicare.gov - Nursing Home Compare - HomeT > B > I've already seen some data there that doesn't match my mother's= > experiences of one of the four nursing homes she's been in.3 > F > When a large number of the baby-boomers reach nursing home age, theyE > may find that the suicide chambers in "Soylent Green" an attractive" > option to many nursing homes.= >   9 Been thru this with my mother-in-law.  She finally passeda< away in one of those Nursing Homes.  Near the end we managed: to get Hospice in there to get better care, but we knew it9 was the end for her.  I'm sure I won't get too much older ; before croaking, and doubt I'll ever see a nursing home.  I. hope.    ------------------------------  % Date: Thu, 14 Nov 2002 14:33:48 -0500t) From: Virginia Rogers <vrogers@umich.edu>o, Subject: backup: /since input file qualifier( Message-ID: <3DD3FA9C.4050602@umich.edu>  F I am using backup on OpenVMS 7.2-1, and I am trying to do incremental G backups.  The problem I'm having is that the /since qualifier does not r seem to be working.e  3 Here is an example of the backup command I'm using:tF BACKUP/RECORD DISK$USER:[VROGERS]*.*/MOD/SIN=1-NOV-2002 []TMP.BCK/SAVE  I But, when I do this it is backing up all the files in the directory, not fI just the ones that have been modified since Nov 1.  I'm going crazy with  F this.  It seems so simple and straightforward, but I can't figure out F what I'm doing wrong.  I've looked at release notes, and haven't seen I mention of a problem.  It did seem to work if I used /SIN=YESTERDAY, and oH   there were some other dates I tried that worked, but most of the time  it didn't work.   3 Am I doing something really dumb?  Any suggestions?o Thanks   ------------------------------  % Date: Thu, 14 Nov 2002 16:48:39 -0500s- From: "Peter Weaver" <peter.weaver@stelco.ca>20 Subject: Re: backup: /since input file qualifier5 Message-ID: <ar15nu$ecs3u$1@ID-141708.news.dfncis.de>    Virginia Rogers wrote: >...H > BACKUP/RECORD DISK$USER:[VROGERS]*.*/MOD/SIN=1-NOV-2002 []TMP.BCK/SAVE >kF > But, when I do this it is backing up all the files in the directory,C > not just the ones that have been modified since Nov 1.  I'm goingI >...  F Please take a look at section 5.41.1 of the VMS FAQ, that has the best explanation of this.   -- Peter WeaverD Opinions are my own, and do not reflect the opinions of my employer,A nor the company that it sub-contracts to, nor the company that itf sub-contracts to.t   ------------------------------    Date: 15 Nov 2002 07:20:48 -0600- From: koehler@encompasserve.org (Bob Koehler)d0 Subject: Re: backup: /since input file qualifier3 Message-ID: <HUYboYUUi7Np@eisner.encompasserve.org>c  T In article <3DD3FA9C.4050602@umich.edu>, Virginia Rogers <vrogers@umich.edu> writes:H > I am using backup on OpenVMS 7.2-1, and I am trying to do incremental I > backups.  The problem I'm having is that the /since qualifier does not   > seem to be working.b > 5 > Here is an example of the backup command I'm using:nH > BACKUP/RECORD DISK$USER:[VROGERS]*.*/MOD/SIN=1-NOV-2002 []TMP.BCK/SAVE > K > But, when I do this it is backing up all the files in the directory, not nK > just the ones that have been modified since Nov 1.  I'm going crazy with nH > this.  It seems so simple and straightforward, but I can't figure out H > what I'm doing wrong.  I've looked at release notes, and haven't seen K > mention of a problem.  It did seem to work if I used /SIN=YESTERDAY, and lJ >   there were some other dates I tried that worked, but most of the time  > it didn't work.* > 5 > Am I doing something really dumb?  Any suggestions?e  E    IIRC if the directory date updates, all files in it will be backed D    up.  I still don't know the rules for when the directory modifiedH    date gets updated (it's not when a direvtory entry is added/removed).   ------------------------------  % Date: Fri, 15 Nov 2002 10:17:05 -0500 ) From: Virginia Rogers <vrogers@umich.edu>.0 Subject: Re: backup: /since input file qualifier( Message-ID: <3DD50FF1.3090404@umich.edu>  & Thanks.  Where can I find the VMS FAQ?   Peter Weaver wrote:  > Virginia Rogers wrote: >  >>... H >>BACKUP/RECORD DISK$USER:[VROGERS]*.*/MOD/SIN=1-NOV-2002 []TMP.BCK/SAVE >>F >>But, when I do this it is backing up all the files in the directory,C >>not just the ones that have been modified since Nov 1.  I'm goingH >>...0 >  > H > Please take a look at section 5.41.1 of the VMS FAQ, that has the best > explanation of this. >  > -- > Peter WeaverF > Opinions are my own, and do not reflect the opinions of my employer,C > nor the company that it sub-contracts to, nor the company that it" > sub-contracts to.= >  >    ------------------------------    Date: 15 Nov 2002 08:58:01 -0600+ From: young_r@encompasserve.org (Rob Young)00 Subject: Re: backup: /since input file qualifier3 Message-ID: <+OIc0bUQURSR@eisner.encompasserve.org>E  T In article <3DD50FF1.3090404@umich.edu>, Virginia Rogers <vrogers@umich.edu> writes:( > Thanks.  Where can I find the VMS FAQ?  : 	The same place you can find most anything.  I start here:   			http://www.google.com/a  & 	If interested in more official links:   	http://www.openvms.compaq.com/o  @ 	Look for "OpenVMS faq" (lower right under Service and Support).   				Rob    ------------------------------  % Date: Fri, 15 Nov 2002 10:13:41 -0500a From: norm.raphael@metso.com% Subject: Capellas is new WorldCom CEOT? Message-ID: <OFF590E4CF.5C788481-ON85256C72.00539771@metso.com>1   Capellas is new WorldCom CEOJ Former head of Compaq takes reins of bankrupt telecom promising to restore
 integrity. November 15, 2002: 6:57 AM EST    H NEW YORK (CNN/Money) - Former Compaq Computer Corp. CEO Michael CapellasC was named the chairman and CEO of bankrupt telecom WorldCom Friday.     B The announcement came four days after Capellasleft his position asK president and chief operating officer of Hewlett Packard Co. (HPQ: ResearchsK , Estimates), the computer maker that bought Compaq earlier this year, amidy( reports that he was heading to WorldCom.    K WorldCom has been looking for a new CEO since the ouster of founder BernardoE Ebbers earlier this year, just ahead of the revelations of accounting 8 irregularities that plunged the company into bankruptcy.    I "I took this job because I am convinced that WorldCom has the assets, thecG customers and the people to regain a leadership role in this industry," J said a statement from Capellas. "In order to do this, we must first regainJ trust and win respect. Accordingly, together we will rebuild WorldCom into? a model of good corporate governance and management integrity."n    I Capellas replaces John Sidgmore, who became the interim CEO in April when-H Ebbers left amid questions about $366 million in his personal loans fromI the company and concerns about its finances and stock price. In June, the E company revealed it had improperly booked $3.8 billion in expenses; am# bankruptcy filing followed in July.e    E Capellas' contract with WorldCom, terms of which were not immediatelylI revealed, needs the approval of the bankruptcy court. The company said iteK could seek that approval as soon as Friday. The creditors' committee at the.3 company issued a statement endorsing his selection.,    H "The creditors' committee fully supports and endorses the appointment ofK Michael Capellas to lead WorldCom," said the statement from Irwin Gold, theoK committee's financial adviser. "We are optimistic that the skills he brings = will add significant value to the company and its customers."h    F Capellas' left HP Monday, accompanied by a statement from HP CEO CarlyK Fiorina that "We've reached a natural transition point." But investors sent,J shares of the computer maker 11 percent lower amid concerns over Fiorina's. direction for the company without him present.       Find this article at:hA http://money.cnn.com/2002/11/15/news/companies/capellas/index.htm8   ------------------------------    Date: 15 Nov 2002 07:34:01 -0600; From: kaplow_r@eisner.encompasserve.org.TRABoD (Bob Kaplow)a) Subject: Re: Capellas to head up WorldCom03 Message-ID: <i9LJvfH8BrmT@eisner.encompasserve.org>a  V In article <H5Fqr9.K5o@world.std.com>, bdc@world.std.com (Brian 'Jarai' Chase) writes:J > If I were Intel, I'd /really/ want to have Hammer-like processor waitingI > in the background in case IA-64 doesn't pull through, or in case HammerhK > starts eating into Intel's market.  But Intel would have to keep any suchsK > project completely secret to avoid undermining IA-64.  That's a hell of afK > position in which to find yourself.  You're potentially damned if you do,e& > and potentially damned if you don't.  6 In a way, Intel does have a backup 64 bit plan: Alpha!  1 	26-October, 2001: A day that will live in infamyc4 	Support Freedom: http://www.indefenseoffreedom.org/  I         "To enslave men, successfully and safely, it is necessary to have H         their minds  occupied with thoughts and aspirations short of theL         liberty of which they are  deprived.  A certain degree of attainableJ         good must be kept before them." Frederick Douglas, "My Bondage and         My Freedom," 1855    ------------------------------  # Date: Fri, 15 Nov 2002 16:44:34 GMTe+ From: "Rick Barry" <barry@star.zko.dec.com>r Subject: Re: CSWS hangs 2 Message-ID: <Sv9B9.20$TU7.728082@news.cpqcorp.net>  2 I'm not aware of any problem such as you describe.  L Please contact customer support for the quickest resolution to this problem.L There needs to be an exchange of more information to identify what's causingI the behavior you're seeing. It will be more efficient and effective goinga, through support than trying to do that here.  
 Rick Barry Compaq Secure Web Server OpenVMS System Software Groupn Hewlett-Packard Company 
 Nashua, NH  5 "Robbie Benton" <rrb35146@yahoo.com> wrote in messagef7 news:dba64bc2.0211141316.3b2713a6@posting.google.com...o8 > "Rick Barry" <barry@star.zko.dec.com> wrote in message- news:<ITNA9.8$Hl6.321622@news.cpqcorp.net>...y >eI > > When your PC file download stops workng, is the server able to handleE other 
 > > requests?  > >y? > The server at this point appears not to be able to handle anyoF > requests.  The default "Apache has been installed" page still existsF > as the default page for this server.  When the server gets into this= > situation, a browser is not even able to call up this page.i >tF > I have been able to find out a little more about the PC application.C > (It was not written here, but was developed elsewhere.) Each five A > minutes, it issues (almost simultaneously) two requests for two-= > different files.  This may be of interest and significance.  >l > Robbie >r > > Rick Barry > > Compaq Secure Web Server! > > OpenVMS System Software Group8 > > Hewlett-Packard Companym > > Nashua, NH   ------------------------------    Date: 14 Nov 2002 11:27:21 -0800! From: jsewell@iu.net (Joe Sewell) ( Subject: Re: Detecting non-Files-11 disk= Message-ID: <9e311bd1.0211141127.4d60a6d8@posting.google.com>b  U hoff@hp.nospam (Hoff Hoffman) wrote in message news:<aquoho$in3$1@web1.cup.hp.com>... c > In article <9e311bd1.0211070740.5012ca69@posting.google.com>, jsewell@iu.net (Joe Sewell) writes:dE > :I've been tasked to create a utility (written in DECwindows/Motif,h% > :though that shouldn't matter here)e > C >   DECwindows is less of an implementation language than a GUI andg >   associated libraries.t  - My apologies for the ill-phrased terminology.o  A > :                                  that can copy hard drives.    > C >   What is the particular problem you seek to solve?  Copying from B >   one disk to an identical disk, making a BACKUP of a disk, etc?  E Copying from one disk to an identical one, where the source disk may,u or may not, be a Files-11 disk.,  G > :                                                              I have2I > :the logic in place -- and working, even :) -- to do VMS copies out theuG > :wazoo.  I also have the logic in place to do disks created on a UN*X ( > :system, thanks to help provided here. > @ >   You know about BACKUP/PHYSICAL for block-oriented copies and@ >   probably also about the BACKUP callable interface.  I expect? >   there are also ports of dd around.  tar and zip and similar $ >   tools also have ports available.  A Correct.  That part is in place.  My question related to knowing,iF without asking the user, when to do a BACKUP/PHYSICAL and when to do a
 BACKUP/IMAGE.r  C > :My current roadblock is in detecting when a disk isn't Files-11.r > E >   MOUNT contains code that tries to auto-detect Files-11 ODS-2 and aD >   ODS-5 volumes, ISO-9660, and probably a couple of other formats.  D I had intended on using that, but something wasn't working properly, as I stated later.  G > :I thought I could try mounting the disk (using SYS$MOUNT), and if ituC > :failed, mount it /FOREIGN and force a physical backup.  For some0E > :reason, though, the mount hangs within SYS$MOUNT itself; I presumeeH > :it's hunting down a valid Files-11 homeblock, although there's no I/O > :on the drive. > C >   I would not expect a hang to arise, that would initially appearcD >   to be either a coding bug in the application, or an OpenVMS bug.  D After doing a home block search, I discovered I had neglected to add@ MNT$M_NOASSIST to the flags going into $MOUNT.  The hang was VMS7 trying to get the attention of a non-existent operator.a  C > :Short of reading the first physical block on the drive (which, IsI > :presume, would be done by opening a channel directly to the device andSI > :doing a SYS$QIO[W]), is there a better way of detecting a non-Files-11d > :disk? > H >   The first block (LBN 0) won't tell you if the disk is Files-11, thatH >   block is the operating system boot block for VAX, Alpha, and ItaniumJ >   consoles.  Some disks have boot blocks, some don't.  I'm in the midst G >   of working on LBN 0 as part of the OpenVMS port to Itanium, as well K >   -- that block will now contain an MBR record.  LBN 1 (the second block) I >   currently contains an ODS-format home block on ODS-2 and ODS-5 disks,-J >   but that might well change if we add GPT support for OpenVMS Itanium.   F Thanks for the info.  Again, I worded the thing poorly; I meant LBN 1,D not LBN 0.  (Sorry, Hoff & Larry.)  For our current situation, whereC we supply the full system (hardware and software), Itanium isn't ank issue.   ------------------------------  # Date: Fri, 15 Nov 2002 00:37:46 GMTa. From: peter@langstoeger.at (Peter LANGSTOEGER)4 Subject: RE: DS10 600Mhz Special this month USD 39952 Message-ID: <ulXA9.32393$6a.639959@news.chello.at>   In article <3D35AD137AAAD411A6BA0008C7B1B12D032C3DE9@MBCALBEXC03.BENDER.COM>, "Koska, John C. (LNG-ALB)" <John.C.Koska@lexisnexis.com> writes:K >This message is in MIME format. Since your mail reader does not understand-  2 Please, don't do this. No MIME and no TOFU here !!  L >It will cost you more to run in the long term without the VMS Base license. ><grin>8   Care to explain ?d  1 As a hobbyist, I'm interested in the answer also.cD And I don't see why you think a hobbyist license is not a license...   -- V Peter "EPLAN" LANGSTOEGERr% Network and OpenVMS system specialistA E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Fri, 15 Nov 2002 13:38:10 -0500n> From: "Koska, John C. (LNG-ALB)" <John.C.Koska@lexisnexis.com>4 Subject: RE: DS10 600Mhz Special this month USD 3995M Message-ID: <3D35AD137AAAD411A6BA0008C7B1B12D032C3E18@MBCALBEXC03.BENDER.COM>e  J This message is in MIME format. Since your mail reader does not understand< this format, some or all of this message may not be legible.  ' ------_=_NextPart_001_01C28CD6.25C172C0  Content-Type: text/plain;  	charset="iso-8859-1"n  K Please accept my apology on the MIME/rich text.  I had my billy box rebuilts to  H my corporation's current standard, and I forgot to flip the switch/check markL to plain text.  (Note to self: Try harder to get corporate standard changed)  I As to the content of my post, I was perhaps unclear and obtuse in my own DE amusing moment.  I was thinking the original poster might want to rung
 something D other (Linux, Tru64 or even perhaps an old NT Alpha image if that is possible) than ,L OpenVMS on the trimmed down system from Island. Yes, hobbyist is quite valid and  a great way to go.  L I guess I just plumb forgot that there was a hobbyist program out there for  OpenVMS.  My bad!u   :) jck
 John Koska   > -----Original Message-----: > From: peter@langstoeger.at [mailto:peter@langstoeger.at]+ > Sent: Thursday, November 14, 2002 7:38 PM  > To: Info-VAX@Mvb.Saic.Coma6 > Subject: RE: DS10 600Mhz Special this month USD 3995 >  > 
 > In article a@ > <3D35AD137AAAD411A6BA0008C7B1B12D032C3DE9@MBCALBEXC03.BENDER.CG > OM>, "Koska, John C. (LNG-ALB)" <John.C.Koska@lexisnexis.com> writes:f? > >This message is in MIME format. Since your mail reader does a > not understand > 4 > Please, don't do this. No MIME and no TOFU here !! > = > >It will cost you more to run in the long term without the p > VMS Base license.r	 > ><grin>n >  > Care to explain ?e > 3 > As a hobbyist, I'm interested in the answer also. F > And I don't see why you think a hobbyist license is not a license... >  > -- 2 > Peter "EPLAN" LANGSTOEGER ' > Network and OpenVMS system specialisto > E-mail  peter@langstoeger.atH > A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist >   ' ------_=_NextPart_001_01C28CD6.25C172C0l Content-Type: text/html; 	charset="iso-8859-1"i+ Content-Transfer-Encoding: quoted-printabled  1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">a <HTML> <HEAD>9 <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =i charset=3Diso-8859-1">@ <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
 5.5.2654.19">u: <TITLE>RE: DS10 600Mhz Special this month USD 3995</TITLE> </HEAD>  <BODY>  I <P><FONT SIZE=3D2>Please accept my apology on the MIME/rich text.&nbsp; =a% I had my billy box rebuilt to </FONT>bG <BR><FONT SIZE=3D2>my corporation's current standard, and I forgot to = ! flip the switch/check mark</FONT>eF <BR><FONT SIZE=3D2>to plain text.&nbsp; (Note to self: Try harder to =& get corporate standard changed)</FONT> </P>  G <P><FONT SIZE=3D2>As to the content of my post, I was perhaps unclear =J and obtuse in my own </FONT>F <BR><FONT SIZE=3D2>amusing moment.&nbsp; I was thinking the original =* poster might want to run something </FONT>H <BR><FONT SIZE=3D2>other (Linux, Tru64 or even perhaps an old NT Alpha =' image if that is possible) than </FONT>aI <BR><FONT SIZE=3D2>OpenVMS on the trimmed down system from Island. Yes, = # hobbyist is quite valid and </FONT>e, <BR><FONT SIZE=3D2>a great way to go.</FONT> </P>  I <P><FONT SIZE=3D2>I guess I just plumb forgot that there was a hobbyist =a program out there for </FONT> 0 <BR><FONT SIZE=3D2>OpenVMS.&nbsp; My bad!</FONT> </P>   <P><FONT SIZE=3D2>:) jck</FONT> $ <BR><FONT SIZE=3D2>John Koska</FONT> </P>  8 <P><FONT SIZE=3D2>&gt; -----Original Message-----</FONT>8 <BR><FONT SIZE=3D2>&gt; From: peter@langstoeger.at [<A =I HREF=3D"mailto:peter@langstoeger.at">mailto:peter@langstoeger.at</A>]</F=l ONT>@ <BR><FONT SIZE=3D2>&gt; Sent: Thursday, November 14, 2002 7:38 =	 PM</FONT>m8 <BR><FONT SIZE=3D2>&gt; To: Info-VAX@Mvb.Saic.Com</FONT>I <BR><FONT SIZE=3D2>&gt; Subject: RE: DS10 600Mhz Special this month USD =t 3995</FONT>u <BR><FONT SIZE=3D2>&gt; </FONT>i <BR><FONT SIZE=3D2>&gt; </FONT>l* <BR><FONT SIZE=3D2>&gt; In article </FONT> <BR><FONT SIZE=3D2>&gt; =,I &lt;3D35AD137AAAD411A6BA0008C7B1B12D032C3DE9@MBCALBEXC03.BENDER.C</FONT>=   F <BR><FONT SIZE=3D2>&gt; OM&gt;, &quot;Koska, John C. (LNG-ALB)&quot; =2 &lt;John.C.Koska@lexisnexis.com&gt; writes:</FONT>H <BR><FONT SIZE=3D2>&gt; &gt;This message is in MIME format. Since your = mail reader does </FONT>- <BR><FONT SIZE=3D2>&gt; not understand</FONT>r <BR><FONT SIZE=3D2>&gt; </FONT>lI <BR><FONT SIZE=3D2>&gt; Please, don't do this. No MIME and no TOFU here =v	 !!</FONT>  <BR><FONT SIZE=3D2>&gt; </FONT>-F <BR><FONT SIZE=3D2>&gt; &gt;It will cost you more to run in the long = term without the </FONT>0 <BR><FONT SIZE=3D2>&gt; VMS Base license.</FONT>/ <BR><FONT SIZE=3D2>&gt; &gt;&lt;grin&gt;</FONT>@ <BR><FONT SIZE=3D2>&gt; </FONT> 0 <BR><FONT SIZE=3D2>&gt; Care to explain ?</FONT> <BR><FONT SIZE=3D2>&gt; </FONT> E <BR><FONT SIZE=3D2>&gt; As a hobbyist, I'm interested in the answer =t also.</FONT>B <BR><FONT SIZE=3D2>&gt; And I don't see why you think a hobbyist =" license is not a license...</FONT> <BR><FONT SIZE=3D2>&gt; </FONT>t" <BR><FONT SIZE=3D2>&gt; -- </FONT>B <BR><FONT SIZE=3D2>&gt; Peter &quot;EPLAN&quot; LANGSTOEGER</FONT>D <BR><FONT SIZE=3D2>&gt; Network and OpenVMS system specialist</FONT>@ <BR><FONT SIZE=3D2>&gt; E-mail&nbsp; peter@langstoeger.at</FONT>- <BR><FONT SIZE=3D2>&gt; A-1030 VIENNA&nbsp; =tI AUSTRIA&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=o7 ;&nbsp;&nbsp; I'm not a pessimist, I'm a realist</FONT>g <BR><FONT SIZE=3D2>&gt; </FONT>s </P>   </BODY>  </HTML>s) ------_=_NextPart_001_01C28CD6.25C172C0--o   ------------------------------    Date: 15 Nov 2002 09:51:42 -0600+ From: young_r@encompasserve.org (Rob Young)a0 Subject: Re: EVA and SPC and storage competition3 Message-ID: <2pyh1yLlo67F@eisner.encompasserve.org>s  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:$MIQ0bTlS4QE@eisner.encompasserve.org... @ > In article <_H2dnXW5-aIbmkmgXTWcqA@metrocast.net>, "Bill Todd"  <billtodd@metrocast.net> writes:   ...a  H >> > Why not just use Win2K/XP's bundled volume manager (which is indeed	 >based ontI >> > Veritas code, but costs nothing extra)?  It should spread databases,n >filea9 >> > systems, or whatever across multiple LUNs just fine.  >> > >>B >> The answer to that is you can't failover disks if you are using4 >> the Win2000 Volume Manager to create "RAID" sets. >-L >Well, I guess I could believe that you couldn't do it dynamically (possiblyK >because Veritas wanted you to have to buy the full version), but I suspectDM >that you could do so as long as you were willing to boot the server that wasE8 >taking over (but that wouldn't be very transparent...). >e  H         Forgive me for working this angle but I claim there is substanceG         here... suggesting a reboot is very easy to refute from a sales I         angle.  Many of us have more than 1 datacenter even if the secondhE         data center is quite small.  To suggest a trip to the remote mG         data center (even if nearby) is a good opening for highlightingUI         "reboot for failover" weakness.  Secondly, (and more practically)oL         the server may be in the same room and you aren't there.  Maybe veryJ         good documentation makes up for this so that Operations can rebootF         and check failover server.  But many systems people don't let =         Operations come close to consoles for sundry reasons.l  H         Its bad enough fallover clusters have actually hijacked the termE         cluster.  But fallover reboot clusters is good for hystericalv         laughter.s     >... >p@ >> As all this functionality goes into storage hardware/softwareE >> (snaps, virtualization, remote snaps/bcvs and whatever else is hotgB >> technology) a Veritas is even less attractive as you are paying= >> for something that exists at the storage level and becomesd( >> highly redundant at the system level. >lM >Depends on how things are getting glued together.  When the storage box runseM >out of scalability and you start gluing them together at a higher level, theg, >Veritas-level features become useful again. >t  H         The higher level will *still* be storage, not bolt-on Veritas orG         similar "fixes" for lack of storage scalability - my opinion ofa         course.r  &         I shot this out here before.    P http://groups.google.com/groups?selm=X0YkNMvpPm1R%40eisner.encompasserve.org&oe= UTF-8&output=gplainn    K Although it isn't being shipped yet, Palo Alto, CA-based Compaq also has anlH innovative implementation of storage virtualization called the VersaStorL Executor. VersaStor uses intelligent agent technology to virtualize storage.H The agent, called a vector, resides in the host. The mapping informationN permanently resides in the VersaStor Executor. When the virtualization processK starts, mapping information is uploaded into the vector, where it's cached.CM Virtualization commands are sent to the vector in the host from the VersaStor5N Executor software that resides in the appliance. Then, the vector executes the
 commands.   E         There is more in the link.  The way it works is more than one,D         box contains the storage.  It appears the back-end won't run>         out of scalability (or we are heading that direction).   >>G >> Maybe Win2000 gets its act together and can failover host based RAID ' >> making even Dell storage attractive.h >>= >> Killer hot storage technology to drive a stake in Veritas?       [Differential LBN backups]h   >hK >Until file-level technology drives a stake into block-level storage (whichsK >it certainly should but shows little sign of doing unless I get off my asssE >and do it), volume-manager-level products will likely remain useful.G >n  .         Tell us where or how this will happen.  F         If applications work with blocks and those blocks keep gettingL         faster (or *at worst* have 8 ms random access), what advantage will #         file-level technology add?    M         In the last few years we've seen OS cachers get more intelligent sucheA         that sequential access is a no-brainer (pre-fetch at OS).   H         Where is the value-add on the file-level?  Very curious, thanks.     >> Differential LBN backups. >>H >> Backup only the blocks that changed on the storage.  Full weeklies ofD >> course if using legacy backup methods.  A state-of-the-art backupD >> solution tuned to differntial LBNs would do differentials foreverH >> (after initial full LBN scan, and I can see IBM's TSM supporting such' >> a method when it starts showing up).m > G >Backups will eventually go the way of the dodo, with the functionality 7 >folded into the (adequately redundant) storage itself.j >:           I'll buy that.   >>A >> Differential LBNs are a no-brainer if LBNs are vectored via anlI >> Executor ;-)  (my opinion, no special knowledge about backup futures).7 >iL >Semi-permanent snapshots, isolated by storage-level protection (i.e., hosts4 >can't corrupt them) and geographically distributed. >   H         That sounds very doable and a high goal for many.  So folks must0         be designing (or thinking about) it now.  #                                 Robn   ------------------------------    Date: 15 Nov 2002 09:59:21 -0600+ From: young_r@encompasserve.org (Rob Young)y0 Subject: Re: EVA and SPC and storage competition3 Message-ID: <I4VlrpOWitaV@eisner.encompasserve.org>s  7 	[The only difference between this one and the last onel< 	is attribution.  Editing offline, I made the mistake of not= 	having Bill above me making it appear as if I was respondingt 	to my own post]  , "Bill Todd" <billtodd@metrocast.net> writes:9 >"Rob Young" <young_r@encompasserve.org> wrote in messages- news:$MIQ0bTlS4QE@eisner.encompasserve.org...oA >> In article <_H2dnXW5-aIbmkmgXTWcqA@metrocast.net>, "Bill Todd"t! ><billtodd@metrocast.net> writes:    ...   H >> > Why not just use Win2K/XP's bundled volume manager (which is indeed	 >based ontI >> > Veritas code, but costs nothing extra)?  It should spread databases,o >filed9 >> > systems, or whatever across multiple LUNs just fine.e >> > >>B >> The answer to that is you can't failover disks if you are using4 >> the Win2000 Volume Manager to create "RAID" sets. > L >Well, I guess I could believe that you couldn't do it dynamically (possiblyK >because Veritas wanted you to have to buy the full version), but I suspectoM >that you could do so as long as you were willing to boot the server that waso8 >taking over (but that wouldn't be very transparent...). >t  H         Forgive me for working this angle but I claim there is substanceG         here... suggesting a reboot is very easy to refute from a salesaI         angle.  Many of us have more than 1 datacenter even if the second E         data center is quite small.  To suggest a trip to the remote  G         data center (even if nearby) is a good opening for highlighting/I         "reboot for failover" weakness.  Secondly, (and more practically)mL         the server may be in the same room and you aren't there.  Maybe veryJ         good documentation makes up for this so that Operations can rebootF         and check failover server.  But many systems people don't let =         Operations come close to consoles for sundry reasons.r  H         Its bad enough fallover clusters have actually hijacked the termE         cluster.  But fallover reboot clusters is good for hystericali         laughter.      >... >h@ >> As all this functionality goes into storage hardware/softwareE >> (snaps, virtualization, remote snaps/bcvs and whatever else is hotaB >> technology) a Veritas is even less attractive as you are paying= >> for something that exists at the storage level and becomess( >> highly redundant at the system level. >eM >Depends on how things are getting glued together.  When the storage box runs M >out of scalability and you start gluing them together at a higher level, ther, >Veritas-level features become useful again. >   H         The higher level will *still* be storage, not bolt-on Veritas orG         similar "fixes" for lack of storage scalability - my opinion ofl         course.0  &         I shot this out here before.    P http://groups.google.com/groups?selm=X0YkNMvpPm1R%40eisner.encompasserve.org&oe= UTF-8&output=gplain     K Although it isn't being shipped yet, Palo Alto, CA-based Compaq also has anfH innovative implementation of storage virtualization called the VersaStorL Executor. VersaStor uses intelligent agent technology to virtualize storage.H The agent, called a vector, resides in the host. The mapping informationN permanently resides in the VersaStor Executor. When the virtualization processK starts, mapping information is uploaded into the vector, where it's cached.eM Virtualization commands are sent to the vector in the host from the VersaStoraN Executor software that resides in the appliance. Then, the vector executes the
 commands.   E         There is more in the link.  The way it works is more than one D         box contains the storage.  It appears the back-end won't run>         out of scalability (or we are heading that direction).   >>G >> Maybe Win2000 gets its act together and can failover host based RAIDo' >> making even Dell storage attractive.a >>= >> Killer hot storage technology to drive a stake in Veritas?r      [Differential LBN backups]i   >tK >Until file-level technology drives a stake into block-level storage (whichsK >it certainly should but shows little sign of doing unless I get off my ass E >and do it), volume-manager-level products will likely remain useful.a >n  .         Tell us where or how this will happen.  F         If applications work with blocks and those blocks keep gettingL         faster (or *at worst* have 8 ms random access), what advantage will #         file-level technology add? e  M         In the last few years we've seen OS cachers get more intelligent such A         that sequential access is a no-brainer (pre-fetch at OS).t  H         Where is the value-add on the file-level?  Very curious, thanks.     >> Differential LBN backups. >>H >> Backup only the blocks that changed on the storage.  Full weeklies ofD >> course if using legacy backup methods.  A state-of-the-art backupD >> solution tuned to differntial LBNs would do differentials foreverH >> (after initial full LBN scan, and I can see IBM's TSM supporting such' >> a method when it starts showing up).a >wG >Backups will eventually go the way of the dodo, with the functionality 7 >folded into the (adequately redundant) storage itself.e >a           I'll buy that.   >>A >> Differential LBNs are a no-brainer if LBNs are vectored via an I >> Executor ;-)  (my opinion, no special knowledge about backup futures).n > L >Semi-permanent snapshots, isolated by storage-level protection (i.e., hosts4 >can't corrupt them) and geographically distributed. >s  H         That sounds very doable and a high goal for many.  So folks must0         be designing (or thinking about) it now.  #                                 Rob'   ------------------------------  % Date: Thu, 14 Nov 2002 20:57:09 -0500e* From: "Bill Todd" <billtodd@metrocast.net>0 Subject: Re: EVA and SPC and storage competition2 Message-ID: <jPycncfDxZvyyUmgXTWc2Q@metrocast.net>  ; "Robert Young" <young_r@encompasserve.org> wrote in messaget- news:$MIQ0bTlS4QE@eisner.encompasserve.org...t@ > In article <_H2dnXW5-aIbmkmgXTWcqA@metrocast.net>, "Bill Todd"  <billtodd@metrocast.net> writes:   ...g  G > > Why not just use Win2K/XP's bundled volume manager (which is indeed. based onH > > Veritas code, but costs nothing extra)?  It should spread databases, file8 > > systems, or whatever across multiple LUNs just fine. > >h >1A > The answer to that is you can't failover disks if you are usingu3 > the Win2000 Volume Manager to create "RAID" sets.s  K Well, I guess I could believe that you couldn't do it dynamically (possibly J because Veritas wanted you to have to buy the full version), but I suspectL that you could do so as long as you were willing to boot the server that was7 taking over (but that wouldn't be very transparent...).O   ...*  ? > As all this functionality goes into storage hardware/softwarekD > (snaps, virtualization, remote snaps/bcvs and whatever else is hotA > technology) a Veritas is even less attractive as you are paying < > for something that exists at the storage level and becomes' > highly redundant at the system level.o  L Depends on how things are getting glued together.  When the storage box runsL out of scalability and you start gluing them together at a higher level, the+ Veritas-level features become useful again.g   >tF > Maybe Win2000 gets its act together and can failover host based RAID& > making even Dell storage attractive. >0< > Killer hot storage technology to drive a stake in Veritas?  J Until file-level technology drives a stake into block-level storage (whichJ it certainly should but shows little sign of doing unless I get off my assD and do it), volume-manager-level products will likely remain useful.   > Differential LBN backups.f >sG > Backup only the blocks that changed on the storage.  Full weeklies of C > course if using legacy backup methods.  A state-of-the-art backupMC > solution tuned to differntial LBNs would do differentials foreverpG > (after initial full LBN scan, and I can see IBM's TSM supporting suchh& > a method when it starts showing up).  F Backups will eventually go the way of the dodo, with the functionality6 folded into the (adequately redundant) storage itself.   >e@ > Differential LBNs are a no-brainer if LBNs are vectored via anH > Executor ;-)  (my opinion, no special knowledge about backup futures).  K Semi-permanent snapshots, isolated by storage-level protection (i.e., hosts 3 can't corrupt them) and geographically distributed.d   - bill   ------------------------------    Date: 15 Nov 2002 08:51:49 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)iC Subject: Re: getting file metadata quickly (was: parsing file name)U3 Message-ID: <2Lcv04ky7Esu@eisner.encompasserve.org>   { In article <7c82b6954b.Alan.Adams@orchard-way.freeserve.co.uk>, Alan Adams <alan.adams@orchard-way.freeserve.co.uk> writes:o/ > In message <3DD4065E.451C1B98@pressenter.com>o; >           Lyndon Bartels <lbartels@pressenter.com> wrote:I >  >> Lyndon Bartels wrote: >> > s >> > Hello,h >> > s >> > Alpha VMS v7.3-1o >> > DEC C v6.2. (I think) >> > d >> =I >> OK.... I've been making progress. I've used a couple examples from them; >> dsn area. My goal is to get the same information as a "$i >> DIR/SIZE=ALL/DATE"g >> oI >> I've used $parse to get the full file spec. Then put $search in a loopeJ >> to get each of the file names. Inside the loop, I was using $open(&fab)% >> to get the xabfhc and xabdat info.e >>   >>  K >> But I don't like it... far too pokey. Too many I/Os... Is there a better2G >> way? A way of accessing the size and date of a file without actually: >> opening it? > K > Not to my knowledge.Why do you think DIR/SIZ takes a lot longer than DIR?c  A You can get better performance if you are going to do this over atB significant portion of the disk by accessing INDEXF.SYS separatelyB (perhaps even mapping it) and using RMS only to read the directory? entries themselves (getting the FID for a key into INDEXF.SYS).m  E LJK/Security uses that technique (looking mainly for file protections H rather than sizes), but that is in a context where the READALL privilegeG is a given and it is certain the whole disk will be examined.  I am notpG sure I would bother to do that for a personal utility, although it doesh& provide a useful programming exercise.  E Such examine-only approaches can be written to tolerate mismatches if-B an individual file header has changes by the time the program getsB around to reading it out of cache.  If writing were being done the@ full synchronization provided by the file system should be used.   ------------------------------  % Date: Thu, 14 Nov 2002 23:43:49 +0100M6 From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne@vajhoej.dk>* Subject: Re: Getting the user owner string) Message-ID: <3DD42725.8070201@vajhoej.dk>    Z wrote:  3 > Stephen Eickhoff <operagost@bogus.address> wrote:rN > : Is there a RTL of some sort to get the owner field from the current user's > : account? >  > $ write sys$output f$user()o    - What has that to do with the owner field ????d   Arne   ------------------------------    Date: 14 Nov 2002 23:39:14 -0800% From: Bart.Zorn@xs4all.nl (Bart Zorn)a* Subject: Re: Getting the user owner string= Message-ID: <a98cd882.0211142339.205a9c0a@posting.google.com>T  ! Have a look at the GETUAI tool ato; http://vms.process.com/scripts/fileserv/fileserv.com?GETUAIm   HTH,  	 Bart Zornt     "Stephen Eickhoff" <operagost@bogus.address> wrote in message news:<Q7TA9.6123$h37.2453@tornadotest1.news.pas.earthlink.net>...iL > Is there a RTL of some sort to get the owner field from the current user's
 > account? > K > I was hoping for something more elegant than using jpi_username to obtainl< > the username, then FINGERing it and parsing out the owner.   ------------------------------    Date: 15 Nov 2002 08:30:59 -0600; From: kaplow_r@eisner.encompasserve.org.TRABoD (Bob Kaplow)nI Subject: Re: Going OT - cheese and bacon (was:Re: OT: Hamburger Flipping) 3 Message-ID: <KkV+0ovXckwT@eisner.encompasserve.org>a  [ In article <Ni3TYK6WTOeC@rabbit>, sy18889@rabmbit.famrp.cosm (Bradford J. Hamilton) writes: K > Joking aside, there *are* domestic cheeses which are quite good, and bearrJ > little resemblance to the "cheese" being discussed here.  Goat cheese isN > somewhat popular in our neck of the woods (New England), and Vermont cheddarP > is quite tasty.  We may be "yahoos" out here in the "colonies", but not all of( > us can be painted with the same brush.  J For good made in the heart of the USA cheese, Check out Maytag Dairy FarmsK in Newton IA 800-247-2458. No web site. World famous blue cheese, plenty oft4 other varieties. Yes, it's the same Maytag family...  H And living in NE Illinois, it's only a short trip to plenty of WisconsinE cheese. My favorte cheese store is Bobby Nelson's Cheese shop, at theeI intersection of I94 and Route 142 (ignore the over commercial mars cheeseeL castle next door: why go to Wisconsin to buy cheese imported from around the world?)e  1 	26-October, 2001: A day that will live in infamyo4 	Support Freedom: http://www.indefenseoffreedom.org/  I         "To enslave men, successfully and safely, it is necessary to haveuH         their minds  occupied with thoughts and aspirations short of theL         liberty of which they are  deprived.  A certain degree of attainableJ         good must be kept before them." Frederick Douglas, "My Bondage and         My Freedom," 1855    ------------------------------  # Date: Fri, 15 Nov 2002 17:40:56 GMTn$ From: winston@SSRL.SLAC.STANFORD.EDUI Subject: Re: Going OT - cheese and bacon (was:Re: OT: Hamburger Flipping)78 Message-ID: <00A17013.8F757597@SSRL04.SLAC.STANFORD.EDU>  q In article <KkV+0ovXckwT@eisner.encompasserve.org>, kaplow_r@eisner.encompasserve.org.TRABoD (Bob Kaplow) writes: \ >In article <Ni3TYK6WTOeC@rabbit>, sy18889@rabmbit.famrp.cosm (Bradford J. Hamilton) writes:L >> Joking aside, there *are* domestic cheeses which are quite good, and bearK >> little resemblance to the "cheese" being discussed here.  Goat cheese iseO >> somewhat popular in our neck of the woods (New England), and Vermont cheddarpQ >> is quite tasty.  We may be "yahoos" out here in the "colonies", but not all of ) >> us can be painted with the same brush.d > K >For good made in the heart of the USA cheese, Check out Maytag Dairy Farms L >in Newton IA 800-247-2458. No web site. World famous blue cheese, plenty of5 >other varieties. Yes, it's the same Maytag family...r  M And the same Maytag family that's now responsible for the quite-decent Anchor  Steam Beer, incidentally.e   -- Alang   ------------------------------    Date: 14 Nov 2002 15:51:19 -0800: From: craig.berry@SignalTreeSolutions.com (Craig A. Berry) Subject: Re: HP Advocacy Site = Message-ID: <7f15589f.0211141551.305276ef@posting.google.com>-  n John McLean <mcleanj@swissonline.delete.ch> wrote in message news:<3DD2C825.B8A806DE@swissonline.delete.ch>...' > winston@SSRL.SLAC.STANFORD.EDU wrote:2 > > Q > > (And I think the whole JDBC deal is a diversion from John's original problem,lO > > which was that RMS access from Java was clumsy, slow, and awkward.  If Java J > > programs have to go through a JDBC layer and a SQL interpreter to do aO > > key-based lookup of an RMS file, that'll be clumsy, slow, and awkward too.)u  D Well, using JDBC is going to be a lot *less* clumsy and awkward than= calling the RMS services directly.  If that weren't true, theaC relational database would not have been invented some 30+ years agoeC and we'd still always use data access methods that require detailed"C knowledge of the file structure.  Whether a SQL layer on top of RMSn> would slow you down or speed you up would depend on how it wasC implemented and what exactly you're doing.  If you are joining data F from multiple files and/or searching on criteria that involve multipleD keys and if you had a query parser that does any cost calculation ofE the type all RDBMS's do for alternate query paths, then the SQL layeruF would more than make up for the overhead it introduces.  If you aren'tE doing a join and are just reading down a single key in a single file, @ of course the RMS services in a native compiled language are the3 fastest (if frequently the most awkward) way to go.o   <snip>  ) > I would like to see Java with low level J > hooks so that the various vendors could provide the interfacing to theirJ > various operating systems at this point rather than a singular interface> > which provides a stack-based mechanism to access C programs.  B I'm not sure what you mean by a "stack-based mechanism to access CB programs."  There is a well-documented standard called Java NativeB Interface (JNI) for making native libraries accessible from Java.  More info is here:  6 <http://java.sun.com/products/jdk/1.2/docs/guide/jni/>   There is a VMS example here:  H <ftp://ftp.digital.com/pub/Digital/info/java/JNI_EXAMPLE.BCK_DCX_AXPEXE>  F The docs assume C or C++, but you could probably write the code in anyC language with a pointer type, or you could do your heavy lifting in E FORTRAN or BLISS or whatever and do only the glue parts in C.  If youoC want to write completely general, robust wrappers to $PARSE, $FIND,sC $GET, $PUT, $WRITE, etc., then yes that will be a bit of a project,i4 though the hard part will be the RMS stuff, not JNI.   <snip>  9 > we seem to have third-party packages which will do someeF > of what we want - at a price - and yet end up in a similar situationI > viz, with code that is not transportable, so what have we really gainedc > ?l  F Code written for JBDC (or ODBC, or the Perl DBI for that matter) is in= fact quite portable to a variety of databases on a variety ofsC platforms.  If you used JDBC to access RMS, you could move the sameSA data into an RDBMS next year and have little or no change to yourK code.a  H > Should HP provide the interfacing to RMS (and perhaps other functions)> > ?   In the current situation I would say "maybe" rather than > "definitely".  l  E It would certainly be nice to have some more substantive JNI examplesdD giving access to system services and such.  I can see why it's not a high priority though.n  ; > On the other hand if Java could provide low level hooks -tJ > perhaps that enable me to run DCL with Java commands spread through it -D > then the answer would definitely be "yes" because of the degree of# > integration with DCL and the CLI.,  C Java's object-oriented nature makes it particularly ill-suited as an model for a command language.i  ,D > As for my original tasks I can see that it will not be possible toJ > create Java programs to perform that processing that I need unless I (orB > my employer) is prepared to invest significant time and money.    B You're about to do a bunch of low-level RMS programming, twiddlingD FABs and RABs and such, but it's too much work to wrap 20 lines of CC around one of your functions so it will be accessible from Java?  IeF don't follow this logic, but you may be right that Java isn't the best
 tool for you.g   > I willE > probably opt for a CGI solution (probably using WASD on the server) G > because this will be quicker to develop, and probably more efficient.i  D As I think I said once before in this thread, whether or not you use, CGI is independent of what language you use.  G > It really seems that some tasks are fine for Java on VMS and some areeI > not.  It also seems that consideration of the use of Java should reallydG > be done during the design of projects because it does not readily fit  > all situations.w    That paragraph I can agree with.   ------------------------------  % Date: Fri, 15 Nov 2002 14:17:46 +0100oE From: Jan C. =?iso-8859-1?Q?Vorbr=FCggen?= <jvorbrueggen@mediasec.de>m Subject: Re: HP Advocacy Sites+ Message-ID: <3DD4F3FA.99238614@mediasec.de>G  F > Well, using JDBC is going to be a lot *less* clumsy and awkward than& > calling the RMS services directly.    6 An SQL SELECT statement is less clumsy and akward that    	READ (10, KEYGE = "Joe") Record   ?    > If that weren't true, thekE > relational database would not have been invented some 30+ years ago:  F I don't think that played a large role in the development of SQL. ACIDI and support of joins are likely more prominent in the list of priorities.n   	Jan   ------------------------------   Date: 14 Nov 2002 21:18:33 GMT5 From: koehler@bessta.gsfc.nasa.aspm.gov (Bob Koehler)K4 Subject: Re: I've come back to build you a better OS/ Message-ID: <ar13v9$jg5$1@skates.gsfc.nasa.gov>n  _ In article <jROdnfXi3-OblUmgXTWcog@metrocast.net>, "Bill Todd" <billtodd@metrocast.net> writes:n >eE >I believe he's supposed to have returned from the beyond (presumablyaH >restored to his prime).  It's a cute ad, but doesn't really stand up to >logical examination.i  H No, but notice on the car behind him they've brought back "holes in the 3 sides" (but of a jokes in one of VW's commercials).0  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation0= NASA GSFC Flight Software       | Federal Sector, Civil GroupoE speaking for no one else        | please remove ".aspm" when replying3   ------------------------------   Date: 15 Nov 2002 06:07 CST ' From: carl@gerg.tamu.edu (Carl Perkins)o4 Subject: Re: I've come back to build you a better OS- Message-ID: <15NOV200206071586@gerg.tamu.edu>-  4 JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes... }Jerry Leslie wrote:> }>      The legendary General Motors design chief, Harley Earl } D }>      him, by the time of his retirement in 1958, to a position ofH }>      influence unequalled by any automobile designer before or since. } G }Ok, so if the guy retired in 1958, how come they are airing ads with a C }not-so-old adult pretending to be him and saying he has returned ?t  G Because they can't use the real one since digging him out of the groundrG and propping him up would make for a bad commercial? (He died in 1969.)p  D They also probably couldn't find a 109 year old actor to play him at1 the age he would be today if he were still alive.b   --- Carl   ------------------------------    Date: 15 Nov 2002 06:58:52 -0600- From: koehler@encompasserve.org (Bob Koehler)-4 Subject: Re: I've come back to build you a better OS3 Message-ID: <4hDBd4YUWHHx@eisner.encompasserve.org>   _ In article <OdmcnWMiFsmbh0mgXTWc3g@metrocast.net>, "Bill Todd" <billtodd@metrocast.net> writes:a > D > Those are the famous Buick portholes, pardner:  show some respect.  >    I'd rather have my old VW than my uncle's old Buick, thanks
    anyway.   ------------------------------  % Date: Fri, 15 Nov 2002 08:23:28 -0800s% From: Dean Woodward <deanw@rdrop.com>I4 Subject: Re: I've come back to build you a better OS( Message-ID: <3DD51F80.6010508@rdrop.com>   Carl Perkins wrote:r6 > JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes... > }Jerry Leslie wrote:@ > }>      The legendary General Motors design chief, Harley Earl > } F > }>      him, by the time of his retirement in 1958, to a position ofJ > }>      influence unequalled by any automobile designer before or since. > } I > }Ok, so if the guy retired in 1958, how come they are airing ads with a1E > }not-so-old adult pretending to be him and saying he has returned ?t > I > Because they can't use the real one since digging him out of the groundHI > and propping him up would make for a bad commercial? (He died in 1969.)D   I think there are laws, too.   ------------------------------  # Date: Fri, 15 Nov 2002 17:38:45 GMT 1 From: LESLIE@JRLVAX.HOUSTON.RR.COM (Jerry Leslie) 4 Subject: Re: I've come back to build you a better OS7 Message-ID: <FiaB9.2311$Yb.61559@twister.austin.rr.com>b  1 JF Mezei (jfmezei.spamnot@vl.videotron.ca) wrote:F : Jerry Leslie wrote:t? : >      The legendary General Motors design chief, Harley Earl  : E : >      him, by the time of his retirement in 1958, to a position ofrI : >      influence unequalled by any automobile designer before or since.> : H : Ok, so if the guy retired in 1958, how come they are airing ads with aD : not-so-old adult pretending to be him and saying he has returned ? :n  G Because of his reputation and accomplishments (URL wrapped to 2 lines):(  0    http://www.automotivehalloffame.org/inductee/<    ViewInductee.cgi?Action=View&InductId=51&browser=top.html    Automotive Hall of Fame    m/    Earl , Harley J. (1893 - 1969) Inducted 1986O  H      * Established General Motors Art & Color Section in 1928, the first:        major styling department in the automotive industry  F      * Served GM as Vice President from 1940 through 1958, controlling<        the design of millions of automobiles and GM products  G      * Was instrumental in introducing annual styling changes to updatea2        and modernize vehicles at regular intervals        eC    Harley Earl reshaped the look of automobiles from the ground up.e  E    The career path of Harley Earl, the son of a California custom car1F    builder, became clear during a family camping trip in 1910. After aF    heavy rainfall, the 16-year-old Earl wandered from the campsite andG    found a clay-filled hollow. Earl sculpted the clay into fantasy carsNG    more elegant than any that existed at the time. Later, in the employ>H    of his father, Earl designed opulent cars for Hollywood clients untilG    a General Motors executive approached Earl with the words, Come withJ!    me son, Ive got a job for you.C  F    At GM, Earl introduced the use of clay models and obtained artisticF    freedom for himself and his staff in developing automotive designs.E    Following the success of the 1927 Cadillac LaSalle, Earl was namedOA    manager of GMs revolutionary Art & Color Section. Under Earl's F    guidance, GM stylists created vehicles that beautified the AmericanB    roadway and permanently changed the shape of automotive product    design."N    2 --Jerry Leslie   (my opinions are strictly my own)9   Note: leslie@jrlvax.houston.rr.com is invalid for emailI   ------------------------------  % Date: Fri, 15 Nov 2002 19:16:31 +0100>6 From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne@vajhoej.dk>4 Subject: Re: I've come back to build you a better OS) Message-ID: <3DD539FF.3030508@vajhoej.dk>e   JF Mezei wrote:o  N > Some car company in the USA has TV ads where some guy named Harley Earl saysN > "I've come back to build you a new car", making one believe that Harley EarlO > was a car designed in the 1950s/1960s and has returned to that car company tos > make even better cars3 >  > This got me to think.n > O > Imagine if HP were to get Dave Cuttler to quite Microsoft and "return" to HP,pM > even if her were just given an office to toy around and play. This way, the K > VMS folks could use Cuttler in TV ads that would directly spoof those car B > maker TV ads and say "I've come back to build you a better OS".  > P > That would generate lots of publicity, and especially would paint Microsoft inJ > a bad way since it would have lost a major architect for its Windows OS.  . 1)  I do not think that it wil have that big a      PR effect !  4      How many people outside our little groups knows      who Dave Cutler is ?e  8 2)  I do not think Dave Cutler has contributed that much      to Windows for many years.s  7      It may be more effective to hire Anders Hejlsberg.b   Arne   ------------------------------   Date: 15 Nov 02 05:10:57 +0100) From: p_sture@elias.decus.ch (Paul Sture)b; Subject: Re: Identifying a MOP load client with DECnet/Pluss) Message-ID: <qDFaZsZWdPu6@elias.decus.ch>a  R In article <3DD18DC4.4080507@iee.org>, Antonio Carlini <arcarlini@iee.org> writes: > Paul Sture wrote:WG >> So, does anyone know of a way using DECnet/Plus to get any more infoy* >> on the source of the MOP load requests? > ; > SET HOST/MOP may well get you somewhere ... at least some ; > devices will give you a clue as to exactly what they are.r > 9 > I canot remember the Phase V equivalent of SURVEILLANCEg9 > mode (but then I didn't use it often enough to remembere' > how to do it on Phase IV either ...).v >    Thanks for the pointer.    8 > The chance are, however, that it is a terminal server.5 > For most recent ones (even if they have loaded over  > bootp ...) you can doo3 > 	SET H/MOP/CIRC=whatever/ADDRESS=ethernet-addresso9 > then log in and see what the name is set to. If nothingk8 > special is set and all your servers are DS700 or DS900# > or whatever, serves you right :-)e >   ' Oh, to be sure, they all have names :-)i    k --  
 Paul Sture Switzerlandn   ------------------------------  + Date: Fri, 15 Nov 2002 14:22:17 +0000 (UTC)e, From: lewis@mazda.mitre.org (Keith A. Lewis)0 Subject: Re: Identifying an email user (VMSmail)- Message-ID: <ar2vup$q7$1@newslocal.mitre.org>    JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes in article <3DD4A92C.405E7EA2@vl.videotron.ca> dated Fri, 15 Nov 2002 02:58:38 -0500: 
 >Scenario: >rK >foreign mail protocol wants to pull a record from a separate profile to bee >used when sending an email. >sN >Now, when a user is on a cluster, he may be using any node to send an email.  >rO >So, assuming a  cluster alias of CHOCOLAT , with nodes SWISS, FRENCH, BELGIUM,  >GERMAN. >eO >User "chef" logs in on FRENCH where is starts to send an email. Is there a way L >to ensure that the foreign mail protocol application will see the sender as- >"CHOCOLAT::CHEF" instead of "FRENCH::CHEF" ?v  K OK, this is a Decnet mail question.  Until I saw the :: I was getting ready % to give you the Compaq tcp/ip answer.n  I What you need to do is enable the outgoing Decnet cluster alias.  I don'toB rembember how to do this, but it should be somewhere in the DecnetE networking manual.  Once you enable it, all your nodes will appear as A the cluster alias.  It saves me a lot of proxy records in SYSUAF!o  K >Is there a way , from nodPASTRY:: to find out that FRENCH:: is part of thea >CHOCOLAT:: cluster alias ?   G Not AFAIK.  You could write a script which runs on CHOCOLAT to do it --r? check F$GETSYI("CLUSTER_MEMBER","FRENCH") on any CHOCOLAT node.i  + --Keith Lewis              klewis$mitre.orgo> The above may not (yet) represent the opinions of my employer.   ------------------------------  + Date: Fri, 15 Nov 2002 16:54:37 +0100 (MET)w9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>s0 Subject: Re: Identifying an email user (VMSmail); Message-ID: <01KOWJY9FI769ZLF2M@sysdev.deutsche-boerse.com>s  L > > User "chef" logs in on FRENCH where is starts to send an email. Is thereK > > a way to ensure that the foreign mail protocol application will see ther; > > sender as "CHOCOLAT::CHEF" instead of "FRENCH::CHEF" ? g > G > OK, this is a Decnet mail question.  Until I saw the :: I was gettingt. > ready to give you the Compaq tcp/ip answer.  >lD > What you need to do is enable the outgoing Decnet cluster alias.    E I don't know much about DECnet.  Is there a TECHNICAL reason why one  E cannot "enable the outgoing TCPIP cluster alias"?  I really like the fE TCPIP cluster alias, but it would be nice if it could be the address f  used for outgoing stuff as well.   ------------------------------  + Date: Fri, 15 Nov 2002 16:13:08 +0000 (UTC)), From: lewis@mazda.mitre.org (Keith A. Lewis)0 Subject: Re: Identifying an email user (VMSmail). Message-ID: <ar36ej$16c$1@newslocal.mitre.org>   Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes in article <01KOWJY9FI769ZLF2M@sysdev.deutsche-boerse.com> dated Fri, 15 Nov 2002 16:54:37 +0100 (MET):F >I don't know much about DECnet.  Is there a TECHNICAL reason why one F >cannot "enable the outgoing TCPIP cluster alias"?  I really like the F >TCPIP cluster alias, but it would be nice if it could be the address ! >used for outgoing stuff as well.l  J I think there probably is a technical reason (beyond the obvious fact thatI Compaq tcp/ip doesn't support that feature).  Suppose a client running onhK your cluster sends a request, and the server (outside the cluster) needs to:K respond.  Which machine does the server send it to?  The ARP of the clusterN alias?  Yecch!  I I use a hidden SMTP substitute domain instead.  Works great for mail.  Ite. doesn't do anything for tcp/ip proxies though.   $ UCX HELP SET CONF SMTP /SUBSTs for more info.  + --Keith Lewis              klewis$mitre.org > The above may not (yet) represent the opinions of my employer.   ------------------------------  % Date: Fri, 15 Nov 2002 11:28:52 -0500 - From: "Peter Weaver" <peter.weaver@stelco.ca>-0 Subject: Re: Identifying an email user (VMSmail)5 Message-ID: <ar37cb$eo2oh$1@ID-141708.news.dfncis.de>E   JF Mezei wrote:a >...D > User "chef" logs in on FRENCH where is starts to send an email. IsG > there a way to ensure that the foreign mail protocol application willd@ > see the sender as "CHOCOLAT::CHEF" instead of "FRENCH::CHEF" ? >...  L Have you tried setting bit 0 of MAIL$SYSTEM_FLAGS? Normally in a cluster you0 would "Define /System /Exec MAIL$SYSTEM_FLAGS 7"   -- Peter WeaverD Opinions are my own, and do not reflect the opinions of my employer,A nor the company that it sub-contracts to, nor the company that itm sub-contracts to.s   ------------------------------    Date: 14 Nov 2002 17:26:34 -0800( From: bob@instantwhip.com (Bob Ceculski)% Subject: Re: Jaw dropping EV7 systemso= Message-ID: <d7791aa1.0211141726.5ba4abb7@posting.google.com>s  g JF Mezei <jfmezei.spamnot@vl.videotron.ca> wrote in message news:<3DD3FD2D.B07716BF@vl.videotron.ca>...> > Lyndon Bartels wrote:eK > > As long as there's money to be made with VMS and/or Alpha... they'll dot > > it.... Pride be damned.  > N > You're assuming that the company has no strategy except for making money. HP2 > has built an alliance with Intel and Microsoft.  > O > It was not to Digital's fiscal advantage to dump ALL-In-1, but Palmer decided O > it was strategic to dump its own profitable products in favour of Microsoft's , > products which yield 0 profit for Digital. > J > I think that there are plenty of examples of companies not acting in theA > interest of profit but in some interest of a supposed strategy.r  - what strategy?  I say Palmer was paid off ...n   ------------------------------    Date: 15 Nov 2002 07:48:00 -0600+ From: young_r@encompasserve.org (Rob Young)y% Subject: Re: Jaw dropping EV7 systemsm3 Message-ID: <NkJWfUOEaG1n@eisner.encompasserve.org>d  P In article <ar2c0s04oq@drn.newsguy.com>, Alan Greig <a.greig@virgin.net> writes:O > In article <zl$NCyYwigDt@eisner.encompasserve.org>, young_r@encompasserve.org 	 > says... E >>>	to acknowledge by now that much of the VMS base is overseas.  ThecH >>	growth is in (this is my speculation... I have no personal knowledge)I >>	two key segments I would say.  SMS (not in U.S. yet, sheesh) and cell n > K > On my last two trips to the States, my UK tri-band phone happily sent and Q > received SMS text messages. Roaming service was mainly provided by VoiceStream.iP > Of course it would be using my home SMS switching centre in the UK by default.Q > Are you saying that the US networks still don't provide their own SMS switchingg
 > centres? >   > 	That wasn't very clear.  Perhaps there is SMS going on in the@ 	U.S.  If there is, I'm not aware of anyone using it.  The folksC 	around me have Sprint, VoiceStream, Cellular One (and at that, thee= 	names may be different, they change names often) and none ofaA 	about 20 people I know with cellphones are using SMS.  Every use A 	of a cell phone (watching hundreds perhaps) as far as I can tellpF 	is with it pressed against a head.  If folks are using SMS, I suspectA 	I would see it in their lap fiddling with their thumbs.  I don'ta4 	think you do SMS with it pressed against your head.   				Robo  D >>	phone billing.  Both of those are very intensive growth oriented G >>	segments.  Huge IO needs and very uptime centric, perfect for large h >>	VMS clusters. >>	 >>				Robm >>L >>(1)  pre-shipping ES40 orders were large.  In hindsight , it may have beenA >>a lot of JSTARS purchases (speculation, no personal knowledge).d >> >    ------------------------------  % Date: Thu, 14 Nov 2002 22:49:47 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com>l% Subject: RE: Jaw dropping EV7 systemsoT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660B5F@kaoexc01.americas.cpqcorp.net>   Rob,  ? >>>  I personally know about very large customers that are very  impressed. <<<  F Check this external link out: [no - I don't know "exactly" what system  was used, but I have an idea ..]   :-)I  1 http://www.wallstreetsystems.com/news/hpbench.htmKB "WALL STREET SYSTEMS BREAKS ONE MILLION FX TRANSACTIONS PER DAY IN BENCHMARKING TESTS WITH HP=20   E New York and London, Tuesday, 1st October 2002 - Wall Street Systems,c? Inc., a leading provider of global treasury and capital marketsgH solutions and services, today announced the results of benchmark testingG of The Wall Street System(r) treasury engine performed under laboratory E conditions at the Hewlett-Packard Company's New Hampshire engineering E benchmark center. The Wall Street System treasury engine exceeded one E million FX transactions per day, representing a five-fold improvement G from the previously measured peak performance. It processed a peak load + of approximately 45,000 new deals per hour.   B Over a period of several months, the Wall Street Systems Technical@ Design Team, in collaboration with HP's OpenVMS operating systemH developers, redesigned the FX deal processing modules of The Wall StreetH System treasury engine, with a view to further increasing FX transactionF processing rates. The FX deal processing modules take advantage of newA operating system features created by HP expressly for Wall Streetr Systems.  G Mark Tirschwell, Chief Technology Officer, Wall Street Systems, states:iG "The tremendous performance gains can be attributed to a combination of A HP AlphaServer systems and software architecture that enables our G solution to operate resiliently in a high performance environment. ThisoH is particularly important as The Wall Street System treasury engine mustG be able to process a high-volume of transactions for clients 24 hours ai day, seven days a week.""   H The global requirements of the customers of Wall Street Systems, runningG an enterprise-wide solution, demand ongoing investment in benchmarking. C In the specific area of FX trading, the need to process significant2> numbers of transactions is essential in running a worldwide FXD operation. During the tests, new, advanced techniques were developed> that enable The Wall Street System treasury engine to run manyH concurrent FX position processing queues and maximize FX throughput. TheH new structure allows deals to be submitted and completely processed fromF a front-office perspective in a single step, ensuring users around the+ world are working with real-time positions.e  D This industry leading performance was achieved running on the latestH version of HP OpenVMS (Version 7.3-1) running on HP AlphaServer systems.C "I'm delighted to see this performance breakthrough, resulting fromsG close collaboration between one of our leading Financial Services ISVs,sF and the latest technology from OpenVMS and our AlphaServer engineeringF teams," said Mark Gorham, vice president, HP's OpenVMS Software Group.   :-)c  
 Kerry Main Senior Consultantl Hewlett-Packard (Canada) Co.! Consulting & Integration Servicesb Voice: 613-592-4660) Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)a     -----Original Message-----5 From: Rob Young [mailto:young_r@encompasserve.org]=20a  Sent: November 14, 2002 10:55 AM To: Info-VAX@Mvb.Saic.Comt% Subject: Re: Jaw dropping EV7 systemsr    8 In article <3DD3522B.A1C06AC2@vl.videotron.ca>, JF Mezei) <jfmezei.spamnot@vl.videotron.ca> writes:  > Rob Young wrote:2 >> Paul DeMone (pdemone@igs.net) on 11/9/02 wrote: >> ---------------------------H >> I think the EV7 systems are going to cause a lot of jaws to drop, the  H >> POWER4 design to be re-evaluated in a different, less positive light,  I >> and maybe even stir up controversy again about Compaq killing Alpha=20lI >> in favour of IA64 (I mean by people that aren't already obsessed by=20o >> this topic).t >>=20 @ >>         Might I note he has his reputation on the line there. >=20 >=20H > It has often been said that HP won't want to flaunt EV7 systems and=20H > their performance, and even that they will delay their introduction as   > much as possible.h >=20 > Is that really realistic ? >=20E > Let say that know they erred in killing Alpha, and know that the=20iH > public knows.  Wouldn't it be better for them to be up front about it,  H > release EV7 as soon as possible and at reap as much profits from it as  1 > they could and get some limelight/PR from it  ?l >=20D > It would seem to me that hiding/delaying EV7 will only increase=20I > speculation on HP's real motives and certaintly not help in building=20 ; > the trust between HP and the unhappy VMS/Tru64 customers.b    : 	There *may* be a myriad of reasons.  Paul speculates in a	 follow-upcA 	that a lowered body count may be delaying things.  Either way, Ia  > 	will wager that EV7 pre-shipping orders are quite good(1).  I
 personally= 	know about very large customers that are very impressed.  We  have= 	to acknowledge by now that much of the VMS base is overseas.t The<; 	growth is in (this is my speculation... I have no personals
 knowledge)A 	two key segments I would say.  SMS (not in U.S. yet, sheesh) and1 cell=20eA 	phone billing.  Both of those are very intensive growth orientedo  > 	segments.  Huge IO needs and very uptime centric, perfect for large=20 	VMS clusters.   				Rob.  E (1)  pre-shipping ES40 orders were large.  In hindsight , it may havedD been a lot of JSTARS purchases (speculation, no personal knowledge).   ------------------------------    Date: 15 Nov 2002 08:11:06 -0800% From: Alan Greig <a.greig@virgin.net>s% Subject: Re: Jaw dropping EV7 systems ( Message-ID: <ar36aq0gsr@drn.newsguy.com>  M In article <NkJWfUOEaG1n@eisner.encompasserve.org>, young_r@encompasserve.orgk says...a >o >>? >	That wasn't very clear.  Perhaps there is SMS going on in theiA >	U.S.  If there is, I'm not aware of anyone using it.  The folks   I If you have a GSM cellphone feel free to email me the number and I'll trycN sending you an SMS. In the UK all networks have been able to receive SMS sinceK about day one but not all could send. Also you can change the SMS switchingpM centre number in the phone setup. You would probably find that if you changedhN the number on a US phone to (say) the T-Mobile SMS switch number in the UK, itN would work. In other words the basic SMS protocols are operational (being partK of the GSM spec) but no (or few) (VMS!!) switches are available to actuallyx handle the text.  P In Dallas I had no problems sending whatsoever. In Chicago I had some reports ofN folks receiving two or three copies of every text I sent back to the UK. I wasN even able to send an SMS message from my UK cellphone to a Norwegian cellphone# when both of us were in the States.t  D >	around me have Sprint, VoiceStream, Cellular One (and at that, the> >	names may be different, they change names often) and none ofB >	about 20 people I know with cellphones are using SMS.  Every useB >	of a cell phone (watching hundreds perhaps) as far as I can tellG >	is with it pressed against a head.  If folks are using SMS, I suspect B >	I would see it in their lap fiddling with their thumbs.  I don't5 >	think you do SMS with it pressed against your head.   J So if you're not all texting you won't have discovered the delights of SMSI predictive text input. Guaranteed to produce some hilarious unintentional ' sentences if you don't check carefully.a   >r >				Rob >BE >>>	phone billing.  Both of those are very intensive growth oriented 8H >>>	segments.  Huge IO needs and very uptime centric, perfect for large  >>>	VMS clusters.v >>>o
 >>>				Rob >>>fM >>>(1)  pre-shipping ES40 orders were large.  In hindsight , it may have beenrB >>>a lot of JSTARS purchases (speculation, no personal knowledge). >>>s >> w   ------------------------------  % Date: Thu, 14 Nov 2002 20:12:45 -0500e. From: Lyndon Bartels <lbartels@pressenter.com>% Subject: Re: Jaw dropping EV7 systemsh. Message-ID: <3DD403BD.4134750A@pressenter.com>   Bill Todd wrote: > = > "Lyndon Bartels" <lbartels@pressenter.com> wrote in messagee* > news:3DD35157.55288B34@pressenter.com...0 > > Bottom line is just that... the bottom line. > >: > >GD > > Whatever give a company the most profits.... that's what they'll > > probably do. > L > Sure - that's why Compaq killed their most profitable hardware platform 17
 > months ago.o >  > - bill    D But..... There may be more factors determining "profit." And there'sF also shortterm versus longterm... And from what perspective.  Like theE stock-holders.  A sale of the company, could reap the stock-holders agH huge short-term profit... Where-as, a slow, long term profitable product line, wouldn't."    H And that's another reason why Curly could be seen as a "successful CEO".F Because he earned the stockholders a good chunk of short-term profits.      : It may not be "right" it may not be "fair." It's business.     Lyndon   -- sG My opinions are mine and mine alone. They seldom align with those of mya	 employer.     H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.   ------------------------------  % Date: Fri, 15 Nov 2002 14:32:58 -0000 * From: "Craig Cooke" <ccooke@beta.dabs.com>( Subject: Re: Linking problem - need help% Message-ID: <3dd4ffb3$1@194.70.94.92>S  D "Robert TRAWISKI" <robert.trawinski@softax.com.pl> wrote in message( news:ar2uoi$1u2$1@bozon.softax.com.pl... > Hi,o >vC > I've upgraded VMS 7.2 to version 7.2-2. After that when I link my  > application I get message: >t > %LINK-F-OPENIN, error openingr; > DKA100:[USERS.ADAM.MRL.ASM2]LIBTHREADUTILS.EXE;2 as inputd >c: > -RMS-F-CHN, assign channel system service request failed >-. > -SYSTEM-F-NOIOCHAN, no I/O channel available >1D > On OpenVMS 7.2 linking was succesfull. VMS help says that I should increase2 > CHANNELCNT system parameter. How can I do that ? >- >  >A > Robert >7 >R >.   Hi,A   $ MC SYSMANn USE CURRENT+ PARAM SET CHANNELCNT 'value' PARAM WRITE CURRENT. EXIT  C It's a non dynamic parameter so you'll need to re-boot the machine.b   Regards9   Craig Cookec   ------------------------------  % Date: Fri, 15 Nov 2002 17:00:36 +0200h* From: Mike Rechtman <rechtman@tzora.co.il>( Subject: Re: Linking problem - need help* Message-ID: <3DD50C14.2090602@tzora.co.il>  : The advise given will work, but it is wrong, nevertheless.    1. Edit SYS$SYSTEM:MODPARAMS.DAT= 2. Look for previous occurences (if any) of CHANNELCOUNT, and  erase or merge them.
 3. Add a line  MIN_CHANNELCOUNT = <new_value>$ (prefereble to hard-coding a value.)C and add a comment (starts "!" .... ) stating by whom, when and why.e6 4. @SYS$UPDATE:AUTOGEN SAVPARAMS REBOOT CHECK_FEEDBACK    
 Good luck. Mike   Craig Cooke wrote:F > "Robert TRAWISKI" <robert.trawinski@softax.com.pl> wrote in message* > news:ar2uoi$1u2$1@bozon.softax.com.pl... >  >>Hi,  >>C >>I've upgraded VMS 7.2 to version 7.2-2. After that when I link my. >>application I get message: >> >>%LINK-F-OPENIN, error opening7; >>DKA100:[USERS.ADAM.MRL.ASM2]LIBTHREADUTILS.EXE;2 as inputa >>: >>-RMS-F-CHN, assign channel system service request failed >>. >>-SYSTEM-F-NOIOCHAN, no I/O channel available >>D >>On OpenVMS 7.2 linking was succesfull. VMS help says that I should > 
 > increase > 2 >>CHANNELCNT system parameter. How can I do that ? >> >> >> >>Robert >> >> >> >  >  > Hi,v > 
 > $ MC SYSMANv
 > USE CURRENT  > PARAM SET CHANNELCNT 'value' > PARAM WRITE CURRENTe > EXIT > E > It's a non dynamic parameter so you'll need to re-boot the machine.  > 	 > Regardso > 
 > Craig Cookeu >  >  >      --    & New to c.o.vms? allow me to recommend:6 http://eisner.encompasserve.org/~rechtman/post_hlp.htm= Other useful links at http://eisner.decusserve.org/~rechtman/aE ---------------------------------------------------------------------rE Usual disclaimer: All opinions are mine alone, perhaps not even that.u? Mike Rechtman                            *rechtman@tzora.co.il*rE Kibbutz Tzor'a.                          Voice (home):(972)-2-9908337gC    "20% of a job takes 80% of the time, the rest takes another 80%"rE ---------------------------------------------------------------------t   ------------------------------    Date: 15 Nov 2002 07:54:13 -0600- From: koehler@encompasserve.org (Bob Koehler)o( Subject: Re: Linking problem - need help3 Message-ID: <0sUcgJwamYZ6@eisner.encompasserve.org>a  j In article <ar2uoi$1u2$1@bozon.softax.com.pl>, "Robert TRAWISKI" <robert.trawinski@softax.com.pl> writes: > Hi,e > C > I've upgraded VMS 7.2 to version 7.2-2. After that when I link my  > application I get message: >  > %LINK-F-OPENIN, error opening ; > DKA100:[USERS.ADAM.MRL.ASM2]LIBTHREADUTILS.EXE;2 as inputs > : > -RMS-F-CHN, assign channel system service request failed > . > -SYSTEM-F-NOIOCHAN, no I/O channel available > M > On OpenVMS 7.2 linking was succesfull. VMS help says that I should increaseg2 > CHANNELCNT system parameter. How can I do that ?  *    1) edit modparams.dat and enter one of:       CHANNELTCNT = number       MIN_CHANNELCNT = numberp       ADD_CHANNELCNT = numberi  6    2) use sys$update:autogen.com to retune your system  D    Avoid simply using SYSGEN to change parameters.  That can lead to2    poorly tuned systems and hard to track changes.   ------------------------------  % Date: Fri, 15 Nov 2002 16:16:17 +0100t9 From: "Robert TRAWISKI" <robert.trawinski@softax.com.pl>I( Subject: Re: Linking problem - need help. Message-ID: <ar332u$gqf$1@bozon.softax.com.pl>  @ It works ! But there was small error in your hint. It should be:   $ MC SYSMANa PARAM USE CURRENTo PARAM SET CHANNELCNT 'value' PARAM WRITE CURRENTd EXIT   Thank you !a   Robert  D Uytkownik "Craig Cooke" <ccooke@beta.dabs.com> napisa w wiadomoci news:3dd4ffb3$1@194.70.94.92...  >C >r >eF > "Robert TRAWISKI" <robert.trawinski@softax.com.pl> wrote in message* > news:ar2uoi$1u2$1@bozon.softax.com.pl... > > Hi,C > >cE > > I've upgraded VMS 7.2 to version 7.2-2. After that when I link myv > > application I get message: > >a! > > %LINK-F-OPENIN, error openingc= > > DKA100:[USERS.ADAM.MRL.ASM2]LIBTHREADUTILS.EXE;2 as inputs > >i< > > -RMS-F-CHN, assign channel system service request failed > >c0 > > -SYSTEM-F-NOIOCHAN, no I/O channel available > >pF > > On OpenVMS 7.2 linking was succesfull. VMS help says that I should
 > increase4 > > CHANNELCNT system parameter. How can I do that ? > >  > >e > >o
 > > Robert > >  > >e > >  >t > Hi,C >i
 > $ MC SYSMANs
 > USE CURRENT  > PARAM SET CHANNELCNT 'value' > PARAM WRITE CURRENTs > EXIT >gE > It's a non dynamic parameter so you'll need to re-boot the machine.n >r	 > Regardsp >i
 > Craig Cooket >a >n >o   ------------------------------  % Date: Thu, 14 Nov 2002 18:43:58 -0800 % From: Dean Woodward <deanw@rdrop.com>v Subject: Re: MajorDomo for VMS( Message-ID: <3DD45F6E.6020104@rdrop.com>  	 I wibble:r  I > Has this been updated for TCPIP yet? Last time I had time to dink with oH > it (not much), it was still UCX-based, and I didn't have enough time, / > energy, or need to dink it into shape myself.r  D I grabbed the kits; at a quick glance, the answer is still 'no', if E anyone else were wondering. And I still don't have time to muck with  
 it right now.b   ------------------------------  % Date: Fri, 15 Nov 2002 12:28:55 +0100h$ From: Michael Unger <unger@decus.de>6 Subject: Re: New AlphaServer models (ES47, GS1280) ???+ Message-ID: <00A1702B.074BDE51.14@decus.de>s  5 "Robert Deininger" <rdeininger@mindspring.com> wrote:m > [...]m >aL > HP hasn't said anything "publicly", but there are still lots of people whoK > have "private" information about the upcoming EV7 systems.  I don't thinkeJ > any of this is unusual for a new family of computer systems from a major	 > vendor.w  ! A Web page is "public", isn't it?-   Michaelr   ------------------------------    Date: 15 Nov 2002 09:03:18 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)H6 Subject: Re: New AlphaServer models (ES47, GS1280) ???3 Message-ID: <ky$4uyQhIWO$@eisner.encompasserve.org>s  _ In article <rK6B9.5$PC7.245791@news.cpqcorp.net>, "Mike Kier" <michael.kier@compaq.com> writes:s  J > But keep in mind that there's a *whole* lot more to a product launch forK > systems designed for the VMS and Tru64 markets than just getting the base M > hardware running.  You've got a sales force to train around configuring theSN > systems (and this includes OEMs and distributors), a service organization toG > train, a parts logistics system to put in place, there may be changes:B > occurring at the firmware level (think about partitioning a meshN > architecture, for instance) and a sea of documentation to create, and on andL > on.  I'm not privy to any of it, just being a field guy who volunteered toF > work at ETS, but I can bet there is a hurricane of activity going on, > preparing for the official product launch.  C I am aware of at least one 3rd party vendor who was not involved inoC testing the new systems but got called in when there was a question'D regarding operation of their product on a test system.  This was notE a product one would expect to behave differently on the new hardware,-C and the last I heard the resolution was not a matter either for the D hardware or the third party software, but having some time for thoseB with test systems to throw everything they have at it is goodness.   ------------------------------  % Date: Thu, 14 Nov 2002 20:17:50 -0500 . From: Lyndon Bartels <lbartels@pressenter.com>6 Subject: Re: New AlphaServer models (ES47, GS1280) ???. Message-ID: <3DD404EE.4EBB117B@pressenter.com>   Bradford J. Hamilton wrote:r > ( > Aren't these new machines MARVELous??? >  > :-)d  H Yup.... Marvel machines... I was given a preview presentation of them..." They look enticing. Very enticing.  C I can't say anything except that if what the salesmen said is true,o' these are going to be some real honeys.i     Lyndon   -- eG My opinions are mine and mine alone. They seldom align with those of myn	 employer.t    H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.   ------------------------------    Date: 15 Nov 2002 08:55:10 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) + Subject: Re: OpenVMS books : seeking advicei3 Message-ID: <s2BX$yyQbeRg@eisner.encompasserve.org>   T In article <3DD4CD56.D2DDDFF0@127.0.0.1>, Nic Clews <sendspamhere@127.0.0.1> writes: > "Georges A. Tomazi" wrote: >> a >> Hi -S >> sI >> I'm looking for a good introductory book on OpenVMS, from newbie levelbB >> to intermediate/advanced covering usage and administration, for, >> someone with some strong Unix background. >  >> Any recommendation ?c > D > Amazingly, I believe the user documentation is pretty good. You'reF > asking a tall order to go from basic user to administration, becauseJ > administration functions are quite presumptuous over your level of basic > knowledge.  F While I agree, my degree of amazement is much less because I have comeG to expect that from the VMS documentation.  I still email my complaintsLL just as often, but my typical documentation these days is about a particularJ reference being missing from the Index.  (Complete indices are _extremely_ hard to achieve.)e   ------------------------------  % Date: Fri, 15 Nov 2002 01:52:15 +0100 ) From: Georges A. Tomazi <gt@diapason.com>r' Subject: OpenVMS books : seeking advice 8 Message-ID: <phg8tukfkjbpdhm9d8b492nu7id0ugsmpf@4ax.com>   Hi -  F I'm looking for a good introductory book on OpenVMS, from newbie level? to intermediate/advanced covering usage and administration, forD) someone with some strong Unix background.-  E I've found at least three that may fit (and available here in France)L :J  D * The Operating systems handbook: Unix, OpenVMS, OS/400, VM, MVS (D.! DuCharme - McGraw-Hill GB, 1994).a  D * Unix for OpenVMS users, second edition (P. Bourne - Digital Press, 1998).  @ * Introduction to OpenVMS, 5th edition (L. Rice - Digital Press, 1999).   Any recommendation ?   Thx,   Georgesh   --$ Georges A. Tomazi - gt@sunwizard.net   ------------------------------  % Date: Thu, 14 Nov 2002 16:43:07 -0500b- From: "Peter Weaver" <peter.weaver@stelco.ca>e# Subject: Re: OT: Hamburger Flippingd5 Message-ID: <ar15dj$do6q7$1@ID-141708.news.dfncis.de>a  ( Andrew Harrison SUNUK Consultancy wrote: >...8 > Perhaps the Canadians do better cheese, they certainly > do better bacon. >...  H Yes, we do. When I lived in New York City I mentioned a few times that IH really missed good cheddar cheese (and Poutine, but I never got close toG finding that in NYC). I was told to look for Wisconsin cheese, I almost . choked on it when I tasted it, terrible stuff.  H If you like a nice mild cheese try a good Canadian Cheddar, or a Havarti! Light or Lapi if you can find it.   7 I don't believe it, Andrew is right for once :):):):):)    -- Peter WeaverD Opinions are my own, and do not reflect the opinions of my employer,A nor the company that it sub-contracts to, nor the company that it  sub-contracts to.i   ------------------------------   Date: 15 Nov 02 06:53:16 +0100) From: p_sture@elias.decus.ch (Paul Sture) # Subject: RE: OT: Hamburger Flippingt) Message-ID: <24EX1VbYcUX9@elias.decus.ch>e  W In article <01C28BB6.F8AF4520@sulfer.icius.com>, Shane Smith <ssmith@icius.com> writes:tD > That's an omission on my part Roy, I agree. The French cheeses areE > excellent. I'm a particular fan of Brie, which once saved me from alG > customs official at Dover (Long story). However, the Bloody ColonialsaF > don't make crappy knockoffs of the French cheeses (at least not thatC > I've seen), which is probably why I didn't think to mention them.i >  Ah, French cheeses...f  6 From various pages found by Google, Charles de Gaulle:  J "How can one conceive of a one-party system in a country that has over 200 varieties of cheeses?"  H "How can anyone govern a nation that has 246 different kinds of cheese?"  = But from http://www.schweizerkaese.ch/ (German/French pages),s 367 types of Swiss Cheese!  J And no, not all Swiss Cheese has holes in it. The one famous for its holes@ is Emmentaler. The US appears to have difficulty even with that:  E http://abcnews.go.com/sections/world/DailyNews/switzerland010222.html=  K "The U.S. Department of Agriculture (USDA) is reducing the size of holes indO domestically-produced Swiss cheese, cutting the time-honored standard size eye,  or hole, by nearly half.    I Some Swiss are taking the new standard as an affront to one of their mostiC beloved exports - one which they export over 20,000 tons of a year.e   ...i  J American cheesemakers asked for the change because cheese with large holesL often crumble in high-speed slicing machines, used for large commercial food service operations.f  M What used to be the minimum size of a hole, 11/16ths of an inch in diameter - ? about the size of a cherry - is now the maximum size of a hole.t  O And the new minimum size of a hole is 3/8ths of an inch in diameter - less thano the size of a dime.   D Cheese that fails to meet that standard - and that will include mostE traditionally-produced Emmentalers - may not merit a USDA "A" grade."s  : Read the link for the full article, if you can stomach it.  < Oh, we get French and Italian cheese here in .ch as well :-)  s -- s
 Paul Sture Switzerland    ------------------------------  % Date: Thu, 14 Nov 2002 20:23:58 -0500d. From: Lyndon Bartels <lbartels@pressenter.com>5 Subject: Re: parsing file name (refer to time thread)-. Message-ID: <3DD4065E.451C1B98@pressenter.com>   Lyndon Bartels wrote:. >  > Hello, >  > Alpha VMS v7.3-1 > DEC C v6.2. (I think)e >   F OK.... I've been making progress. I've used a couple examples from the8 dsn area. My goal is to get the same information as a "$ DIR/SIZE=ALL/DATE"  F I've used $parse to get the full file spec. Then put $search in a loopG to get each of the file names. Inside the loop, I was using $open(&fab)O" to get the xabfhc and xabdat info.    H But I don't like it... far too pokey. Too many I/Os... Is there a betterD way? A way of accessing the size and date of a file without actually opening it?I     Thanks,t   Lyndon   -- rG My opinions are mine and mine alone. They seldom align with those of myp	 employer.     H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.   ------------------------------  % Date: Thu, 14 Nov 2002 13:47:08 -0500r' From: "Dale A. Marcy" <dqm@y12.doe.gov>  Subject: Re: Patch installation , Message-ID: <ar0r3d$4gp$1@sws1.ctd.ornl.gov>  I I used to frequently install patches during the day and then reboot afteraI the last user logged off that evening.  Several years ago, I did that and-G the system started behaving erratically and I was forced to do a rebootiJ early in the afternoon.  Unfortunately, I cannot remember the specifics ofG what VMS version and/or patches (probably somewhere around the VMS V6.1oL time).  Since then, I do not install the patches until I am ready to perform the reboot.2  
 Dale A. Marcy0    = "Chris Sharman" <chris.sharman@sorry.nospam> wrote in messager. news:aqtv84$bu8$1$8300dec7@news.demon.co.uk...J > It's that time of year again, when no way nohow can I schedule downtime. >sB > Patches arrive with reboot required, so I keep them for January.L > Would it be safe to install them & not reboot, or am I better keeping them > uninstalled ?tL > Installing without rebooting seems to go against the stock warning carried
 > by patches.sH > I'm never sure whether the warning is because the patch is ineffectiveH > before reboot, or because the system is unstable in the window between patcho/ > & reboot, or whether it's just belt & braces.d >sG > I've never had a problem, either installing several patches & doing a  singleI > reboot after, or installing a patch & scheduling a reboot for some less G > civilised hour, but I'm a bit chary of installing a patch requiring a 1 > reboot, & then not doing the reboot for months.r >n, > For example, the sys 5 patch just arrived.K > I could not install it, and the system could crash, and it still wouldn'tw be > installed.L > If I installed it, & didn't reboot, at least it would be installed after a% > crash or power failure or whatever.i >iJ > I'm just after maximising uptime - potentially I could see a patch for aH > crash, and then several crashes for that reason or another, or a powerJ > failure. At least if the patch is installed, the problems it fixes won't- > cause more than one crash (the reboot one).- > 	 > Thanks,7 > Chrisy >v >u >h   ------------------------------  # Date: Fri, 15 Nov 2002 17:51:13 GMTn' From: Colin Blake <colin@theblakes.com> : Subject: Re: Postscript printers and Mozilla/CSWB printing, Message-ID: <3DD533DF.9070805@theblakes.com>   Rich Jordan wrote:  
 >Assuming:N >You can catch the print file - a permanently stopped queue may be useful here >pH Not a problem. Mozilla/CSWB have a "print to file" option for capturing  the PostScript output.   ------------------------------    Date: 14 Nov 2002 11:15:00 -0800! From: jsewell@iu.net (Joe Sewell)l6 Subject: Primary home block anywhere other than LBN 1?= Message-ID: <9e311bd1.0211141115.700d0e59@posting.google.com>t  A According to Kirby McCoy's VMS File System Internals book, a home1E block can be on any of the 1+n*delta (where delta is calculated based 
 on geometry).:  D Will VMS detect a valid home block if the one on LBN 1 of the volumeD is not a valid home block?  (My preliminary tests indicate it won't,( but that contradicts McCoy's statement.)  ? I ask this because I'm working on a tool that needs to detect aTE Files-11 disk vs. a "UNIX" disk.  I've had to go to reading LBN 1 andnB doing a check, again based on McCoy's guidelines, for a valid homeB block.  I'm wondering, though, if I need to scan 1+n*delta blocks,
 just in case?e   ------------------------------  % Date: Fri, 15 Nov 2002 17:33:48 -0000s* From: "Richard Brodie" <R.Brodie@rl.ac.uk>: Subject: Re: Primary home block anywhere other than LBN 1?+ Message-ID: <ar3b5t$slg@newton.cc.rl.ac.uk>e  . "Joe Sewell" <jsewell@iu.net> wrote in message7 news:9e311bd1.0211141115.700d0e59@posting.google.com...l  F > Will VMS detect a valid home block if the one on LBN 1 of the volumeF > is not a valid home block?  (My preliminary tests indicate it won't,* > but that contradicts McCoy's statement.)  K There is the option of using /OVERRIDE=LIMITED. I'm guessing the filesystemyC internals book predates the (new default) limited search algorithm.-   ------------------------------    Date: 15 Nov 2002 11:20:20 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)f: Subject: Re: Primary home block anywhere other than LBN 1?3 Message-ID: <A95e9mN6IDFb@eisner.encompasserve.org>   \ In article <87of8qai72.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:% > jsewell@iu.net (Joe Sewell) writes:w > D >> According to Kirby McCoy's VMS File System Internals book, a homeB >> block can be on any of the 1+n*delta (where delta is calculated >> based on geometry). >   G >> Will VMS detect a valid home block if the one on LBN 1 of the volumewG >> is not a valid home block?  (My preliminary tests indicate it won't,i+ >> but that contradicts McCoy's statement.)  >  oB >> I ask this because I'm working on a tool that needs to detect aD >> Files-11 disk vs. a "UNIX" disk.  I've had to go to reading LBN 1D >> and doing a check, again based on McCoy's guidelines, for a validB >> home block.  I'm wondering, though, if I need to scan 1+n*delta >> blocks, just in case? > G > Is that an ODS-1, ODS-2 or ODS-5 value of Files-11? Ther is a utilityt3 > that does just what you want, it is called MOUNT.'  : No, that is not an adequate test of a valid Files-11 disk. You need ANALYZE/DISK.   ------------------------------    Date: 15 Nov 2002 09:07:31 -0800  From: seremil@email.it (Seremil)D Subject: Problem with Hunter Goatley's ODS2 win version and MO drive= Message-ID: <1c729d32.0211150907.23cc5d9a@posting.google.com>r   Hi allF I'm tryng to use old MO 512byte/sector cartridges writed on a Microvax; VMS 6.2 machine using the Hunter's ODS2 reader for windows. ? I'm tried with windows2000, windows98 but allways with the samer< result: when i mount the drive letter at the $ prompt i have  $ "Could not find suitale ASPI device" " Mount failed with 8"     error.B The MO drive works fine: i can read and write in windows formatted
 cartridges The ASPI check give me all OKs   Suggestions?  
 Thanks to all4 Byen Emilio   ------------------------------  + Date: Fri, 15 Nov 2002 12:18:38 +0100 (MET).9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> J Subject: problem with InfoVAX---please reply to me (not to the group/list); Message-ID: <01KOWABX1NZ6A23PDU@sysdev.deutsche-boerse.com>w  I Due to the fact that I received very few messages today, I would like to eC know if there are problems with the InfoVAX mailing list (which is hE gatewayed to comp.os.vms).  Thus, if you see this before 16 November nD 2002, please reply to the email address in the From: header (not to  comp.os.vms or to InfoVAX).b   ------------------------------   Date: 15 Nov 02 05:44:21 +0100) From: p_sture@elias.decus.ch (Paul Sture) " Subject: Re: Remote Console access) Message-ID: <R4YJJPcmqAHm@elias.decus.ch>r  [ In article <3DD31F77.816ADEAE@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:a > Bill Johnson wrote:e >> mL >> Its interesting, I haven't received a single offer to upgrade anyone fromM >> the NT SMC for wildfire. After all the complaining about it being bad formnJ >> and something should be done, and after I make the offer to upgrade forM >> free, and yet we have had no single taker. What happened to the folks thateO >> wanted something done? I am a vendor that is eager to help and eager to make	O >> sure folks are happy with the product. How do I get some participation or iss2 >> there something else we should do to reach out? > % > Corporate wheels turn slowly, Bill.	 >o   Indeed.1  pF > That whole mind-set thing takes a *VERY* long to establish, and once8 > established, takes an *EXTREMELY* long time to change. >   A Since Bill's offer, we haven't even got through a single cycle of2G weekly meetings, far less the bi-weekly or monthly ones. It does indeedJ
 take time.  I > Remember - we're not the ones controlling the purse-strings, we just doa, > as we're told in most cases. <gentle snip> >   = Not only that, but it takes a good deal of persuasion to make C any change to a working production system, and if given a go-ahead,mB quite some planning to execute the change. 6 months is not unusual$ for either the former or the latter.   -- S
 Paul Sture SwitzerlandX   ------------------------------    Date: 15 Nov 2002 08:53:19 -0800( From: carlo.pettirossi@libero.it (Carlo), Subject: Re: Synchronization VMS/Unix clocks< Message-ID: <553e9e4b.0211150853.a353625@posting.google.com>  	 Hi again,n  A I actually want to synch both computers at the UTC time. The unixeF shows the local time now just cause I used ntpdate to set its clock to8 the VMS "sh time" date and time (which is the UTC time).E Anyway I launched the advanced configuration although the VMS is busysC (it's unfortunately always in use). It sends the local time anyway.iD I admit I'm not an expert at all in it. I'd be thankful if you could5 tell me a good site where to find some documentation.s   thanks again   Carlot   ------------------------------    Date: 15 Nov 2002 10:55:50 -0800( From: carlo.pettirossi@libero.it (Carlo), Subject: Re: Synchronization VMS/Unix clocks< Message-ID: <553e9e4b.0211151055.689eabd@posting.google.com>   > G > I have a suspicion. In Italy local time is UTC plus one hour. If yourgM > machine has its timezone set for Italy, then sho time is showing local timeyN > and the underlying clock is one hour earlier. Which makes my theory rubbish,# > as it needs to be one hour later.  > 0 > However I still suspect the time differential. > O > Are you running Decnet Plus (AKA Decnet OSI, aka Decnet Phase V). If so, try   > K > NCL> sho dtss time. The last part of the display is the time differentialt) > factor, which for you should be +001.00o > ' > If it isn't you need to find out why.i > D > I think NTP will be exporting the part of the time before the TDF. > I > Your unix system will then add back the TDF before displaying the time.m >    Hi Alan,  F you're right. The TDF wasn't properly configured. Thanks for the hint.F Now the only thing missing is to discover why a stratum 1 server has a -11 precision. Thanks a lot to Rob too. n byed   Carloc   ------------------------------   Date: 14 Nov 2002 21:21:50 GMT5 From: koehler@bessta.gsfc.nasa.aspm.gov (Bob Koehler)u" Subject: Re: system calls, RE time/ Message-ID: <ar145e$jg5$2@skates.gsfc.nasa.gov>c  b In article <3DD3FFBB.AC14521C@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:  >briggs@encompasserve.org wrote:C >> LIB$ADD_TIMES and LIB$SUB_TIMES are (in my opinion only) uselessh< >> complication on top of a fundamentally trivial operation. >hI >From a philosophical perspective, aren't date-time values supposed to bel; >"opaque" objects that you shouldn't manipulate directly ? n >i  F    No, these have long been documented.  (Where do you think I learned$    the stuff I said I'd been using?)  B    The UTC services return values which are supposed to be opaque.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation:= NASA GSFC Flight Software       | Federal Sector, Civil Group"E speaking for no one else        | please remove ".aspm" when replying@   ------------------------------    Date: 15 Nov 2002 07:08:19 -0600- From: koehler@encompasserve.org (Bob Koehler)y" Subject: Re: system calls, RE time3 Message-ID: <9tdh3WeXY1AW@eisner.encompasserve.org>.  { In article <784258954b.Alan.Adams@orchard-way.freeserve.co.uk>, Alan Adams <alan.adams@orchard-way.freeserve.co.uk> writes:  > J > As far as I can recall, with LIB$ADDX adding a delta time to an absolute< > REDUCES the absolute (because Delta is a negative number). >hN > I THINK LIB$ADD_TIMES subtracts the Delta, making the result larger, and the: > converse for SUB_TIMES. Check it - I may remember wrong.  D    No.  The LIB$__TIMES routines know the sign convention and do theH    logically right thing.  Abs time ADD delta = later abs time, abs time     SUB delta = earlier abs time.  t  G >> You can use LIB$SUBX and check the value of the quadword difference.u >>  , >> High order longword negative 	: less than. >> Else high order longword > 0	: greater than- >> else low order longword > 0	: greater than  >> else 				: equal. >   A    This complication is not needed.  Since all VMS systems use 2soC    compliment integers, the high longword will look negative if theaE    result is negative.  There is no need to look at the low longword.h   ------------------------------    Date: 15 Nov 2002 07:36:54 -0600 From: briggs@encompasserve.org" Subject: Re: system calls, RE time3 Message-ID: <FiFRD3F$WZlS@eisner.encompasserve.org>*  { In article <784258954b.Alan.Adams@orchard-way.freeserve.co.uk>, Alan Adams <alan.adams@orchard-way.freeserve.co.uk> writes: C >> LIB$ADD_TIMES and LIB$SUB_TIMES are (in my opinion only) useless > >> complication on top of a fundamentally trivial operation.   >  > Be careful however.w > J > As far as I can recall, with LIB$ADDX adding a delta time to an absolute< > REDUCES the absolute (because Delta is a negative number).  C Yes.  LIB$ADD_TIMES and LIB$SUB_TIMES layer a bunch of special casep? logic on top of an underlying quadword addition or subtraction.   = If you LIB$ADD_TIMES two positive quadwords, you get an error L If you LIB$ADD_TIMES a positive and a negative, you get positive + -negativeL If you LIB$ADD_TIMES a negative and a positive, you get positive + -negativeK If you LIB$ADD_TIMES a negative and a negative, you get negative + negative0C If a negative is > 9999 days and you have an unpatched pre-1999 O/S: then you get an error.  C If you LIB$SUB_TIMES two positive quadwords you get the negation ofnE the difference if that difference is positive and an error otherwise.>K If you LIB$SUB_TIMES a positive and a negative, you get positive + negativefJ If you LIB$SUB_TIMES a negative and a positive, you get an error (I think); If you lib$SUB_TIMES a negative and a negative, you get theeE difference if that difference is negative and an error otherwise (ands I may have the cases reversed).oH If a negative is > 9999 days and you have an unpatched pre-1999 O/S then you get an error.   N > I THINK LIB$ADD_TIMES subtracts the Delta, making the result larger, and the: > converse for SUB_TIMES. Check it - I may remember wrong.  F You're right.  And it is precisely the need to check the documentationG that makes me reject LIB$SUB_TIMES and LIB$ADD_TIMES for any real world. use.  = I _KNOW_ what LIB$ADDX and LIB$SUBX will do.  The behavior ofmB LIB$ADD_TIMES and LIB$SUB_TIMES is much less obvious and much lessD well documented.  The 9999 day thing still sticks in my craw -- whatA brainless engineer thought that particular blade guard was a goodtB idea?  It was bloody frigging obvious that that restriction should" have only been enforced on output.  > I don't know what happens when you LIB$SUB_TIMES two identicalB quadwords.  I do know what happens when you LIB$SUBX two identical' quadwords.  And that's important to me.	  @ I rather like the notion of being able to compute the difference; between a third Thursday and the third Wednesday and having 7 the result come out as either +1 day or -6 days without 1 having to fight with my date arithmetic routines.v   	John Briggs   ------------------------------    Date: 15 Nov 2002 07:52:47 -0600 From: briggs@encompasserve.org" Subject: Re: system calls, RE time3 Message-ID: <HWlZOXCDfTS+@eisner.encompasserve.org>r  b In article <3DD3FFBB.AC14521C@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:! > briggs@encompasserve.org wrote:>C >> LIB$ADD_TIMES and LIB$SUB_TIMES are (in my opinion only) uselesss< >> complication on top of a fundamentally trivial operation. > J > From a philosophical perspective, aren't date-time values supposed to be< > "opaque" objects that you shouldn't manipulate directly ?   D Yes, but then you get into the 9999 day debacle.  Some idiot decidedE that the documented valid delta times were the only valid delta timesn4 and that all other values warranted an error status.  ; If you followed the opaque philosophy, you then had no good 8 way to deal with intervals of longer than 27 years.  And= if you decided to break opacity and generate such an intervalr? anyway but still use LIB$ADD_TIMES and LIB$SUB_TIMES, the bladei! guards in those routines bit you.s  E One rational response to that problem was to stop using LIB$ADD_TIMES" and LIB$SUB_TIMES.  9 Another rational response was to remove the blade guards..  $ Both approaches were actually taken.  O > I realise that the odds of the internal representation changing are very low,lO > with Hoff and Fred so busy playing with intel toys :-), but out of principle, M > is direct arithmetic/comparison something which gets official sanction frome > the VMS gods ?  > The VMS quadword date/time format is documented.  Accordingly,A direct arithmetic comparison is guaranteed valid.  At least untilm the documentation changes.   	John Briggs   ------------------------------    Date: 15 Nov 2002 08:29:27 -0600 From: briggs@encompasserve.org" Subject: Re: system calls, RE time3 Message-ID: <OCZ2PmgCgqM5@eisner.encompasserve.org>:  c In article <9tdh3WeXY1AW@eisner.encompasserve.org>, koehler@encompasserve.org (Bob Koehler) writes:s? [attribution missing.  >>> text was written by me, John Briggs]tH >>> You can use LIB$SUBX and check the value of the quadword difference. >>> - >>> High order longword negative 	: less thani/ >>> Else high order longword > 0	: greater thano. >>> else low order longword > 0	: greater than >>> else 				: equal.. >> M > C >    This complication is not needed.  Since all VMS systems use 2ssE >    compliment integers, the high longword will look negative if theaG >    result is negative.  There is no need to look at the low longword.   > Yes indeed.  You only need to check both longwords if you want? to get a three way comparison (greater, less or equal).  If youS> only want to know "is A less than B", you can make do with one comparison.a  C If you want to know "is A greater than B", you can make the obviousdE move and invert the operand order, asking "is B less than A" instead.o  B Similarly, "greater than or equal" and "less than or equal" can beB obtained by negating the sense of the result and inverting operand order as required.  C Four operators (<, >, <=, >=) and four possibilities (invert, don't$> invert, negate, don't negate).  That seems to cover the cases.D One quadword subtraction and one longword compaison will do the job.  = If you want to test for equality (=, !=) then you'll need twod longword comparisons, though.s   	John Briggs   ------------------------------  % Date: Fri, 15 Nov 2002 09:17:27 -0800 ' From: David Mathog <mathog@caltech.edu>  Subject: Turnaround artist???i8 Message-ID: <20021115091727.46e6dff9.mathog@caltech.edu>  @ ZDNET today discusses Capellas starting as CEO of Worldcom here:  ,   http://zdnet.com.com/2100-1106-965953.html  5 In this piece they say a few things which must be then+ "analysts" view of the man.  In particular:I  H  "Capellas is credited with stabilizing Compaq before it was sold to HP"  5  "If Capellas is successful, he will secure his placee   as a turnaround artist."  L It never seemed to me that Compaq stabilized under Capellas and it certainlyJ never "turned around".  Compaq was more like a slow motion car crash, withN the vehicle flipping end over end, and big chunks flying off. Still, Curly didI manage to sell Carly the wreck.  Which makes him an artist of sorts - but' not a turnaround artist.  K CEO of Worldcom is the perfect job for him - it's a no lose situation.  ThevF company's such a basket case financially that if it never comes out ofK chapter 11 nobody will blame him, and if it manages to pull itself together-L (with or without his input) he'll be hailed as a miracle worker.  Either wayG he gets paid a bundle for at least a couple more years.  My bet is thattC all he plans to do is spin off MCI with himself as CEO, pay off theaI Worldcom creditors (to the extent that they will ever be paid) with stocktD in this "new" enterprise, and let the rest of Worldcom disintegrate.N In other words, to mix too many metaphors, he's going ride out of Dodge on the. cash cow while the town burns down behind him.   Regards,   David Mathog mathog@caltech.edu> Manager, Sequence Analysis Facility, Biology Division, Caltech   ------------------------------  + Date: Thu, 14 Nov 2002 19:28:08 +0000 (UTC)e6 From: John Wallace <johnwallace4@yahoo.spamfree.co.uk> Subject: Re: VMS Sales3 Message-ID: <3DD3F98E.5080608@yahoo.spamfree.co.uk>m   JF Mezei wrote:  <snip>   > M > My education on IA64 started with lots of presentations by Digital staff onIN > the Alpha architecture and why Alpha was much better than IA64, and why IA64N > was flawed in its architecture. Are you telling me that on June 25 2001, all$ > those flaws magically disapeared ? >    <snip>    C Folks wanting a refresh on why IA64 was inferior to Alpha, and who mH haven't kept any local copies of DIGITAL stuff, should try Googling for H "alpha ia64 parallelism executive summary". You should hopefully find a E copy of a nice DIGITAL paper from 1999 which compared and contrasted n! EPIC vs Alpha. Or go straight to  E http://www.cs.trinity.edu/~mlewis/CSCI3294-F01/Papers/alpha_ia64.pdf -F It's 80kb, 32 pages - it's a whitepaper not a presentation. IIRC this > paper first appeared very quietly on the Alphapowered website.  H In August 2001, Richard Tomkins posted a text-only version of the paper F on comp.org.decus (find using same keywords). The comp.org.decus post E also had a Compaq hpc address for the pdf file (404 now, of course). NG Richard also asked the same question JF asks: what has changed to make t@ IA64 better? I guess it's a question on a lot of people's minds.  D What the 1999 paper obviously couldn't describe is what snags later I Alphas might hit before coming to market. Opinions on that seem to vary,   and are hard to verify.o  F Anyway, now that MC is out of the way, can't Carly just blame him for C the decision and revisit it as and when appropriate ? She probably LF doesn't care greatly that HP's folks have invested years of effort in  EPIC and IA64.   regardsa john  + Imagine.... being the one without Alpha AXPa   ------------------------------    Date: 16 Nov 2002 00:11:10 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: VMS Sales- Message-ID: <871y5mbyox.fsf@prep.synonet.com>   8 John Wallace <johnwallace4@yahoo.spamfree.co.uk> writes:   > JF Mezei wrote:T  F > > My education on IA64 started with lots of presentations by DigitalF > > staff on the Alpha architecture and why Alpha was much better thanF > > IA64, and why IA64 was flawed in its architecture. Are you tellingC > > me that on June 25 2001, all those flaws magically disapeared ?c   > <snip>  D > Folks wanting a refresh on why IA64 was inferior to Alpha, and whoE > haven't kept any local copies of DIGITAL stuff, should try GooglinghF > for "alpha ia64 parallelism executive summary". You should hopefullyB > find a copy of a nice DIGITAL paper from 1999 which compared and- > contrasted EPIC vs Alpha. Or go straight tolF > http://www.cs.trinity.edu/~mlewis/CSCI3294-F01/Papers/alpha_ia64.pdfB > It's 80kb, 32 pages - it's a whitepaper not a presentation. IIRCE > this paper first appeared very quietly on the Alphapowered website.>  uC > In August 2001, Richard Tomkins posted a text-only version of theA9 > paper on comp.org.decus (find using same keywords). ThegD > comp.org.decus post also had a Compaq hpc address for the pdf fileE > (404 now, of course). Richard also asked the same question JF asks:rD > what has changed to make IA64 better? I guess it's a question on a > lot of people's minds.  D Nothing, and THAT is a big worry. Considering that the IA-64 projectB is about as old as Alpha, what have we seen? Very little, and veryF behind the 8-ball. Now, befor you go on about the FP performance beingD way up there, yes, I have noticed that. And guess what. That is JUSTF what you would expect from a re-implemntation of one of the (up to) 20D year old WLIW/MPP machines. That is, no progress other than what the7 FAB people bring to the mix! This is *not* a good sign.   C In the same time, Alpha has shipped 4 major variants, 064, 066, 164tE and 264. (The EV7 and 8 had not shipped in that timeframe, though EV7c had taped out I think.)r  A If anything, the comparison supports the *principles* Sites et aliF layed out in the AAM. Intel seems to havew been unable to get anythingB other than minor tweeks around the edges and clean-ups done in theB same time frame. I fear the the baggage around the designers necks) will sink them even more as time goes on.m  tE > What the 1999 paper obviously couldn't describe is what snags lateraD > Alphas might hit before coming to market. Opinions on that seem to > vary, and are hard to verify.u  .C > Anyway, now that MC is out of the way, can't Carly just blame himi? > for the decision and revisit it as and when appropriate ? SheeF > probably doesn't care greatly that HP's folks have invested years of > effort in EPIC and IA64.  B Alpha = cost is how she will see it I suspect. Rather than getting@ out front and being a true leader, hp will continue on till they( have become the dellmat of the industry.   -- o< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.1@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Fri, 15 Nov 2002 07:11:39 -0500g' From: "Main, Kerry" <Kerry.Main@hp.com>h> Subject: RE: What EV am I running? (family name for processor)T Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660B62@kaoexc01.americas.cpqcorp.net>   Mark,   E As a fyi - Here is what the latest Alpha ES45 1.25Mhz system produces C when a $show cpu/full is entered: (dual CPU system). Hopefully, thef0 formatting does not get munged with email, but -   $ sho sys/noproc  A OpenVMS V7.3-1  on node XXXXX  14-NOV-2002 17:25:39.74  Uptime  0  18:50:44   $ sho cpu /fullK  ( System: XXXXX, AlphaServer ES45 Model 2B  .   SMP execlet   =3D 3 : Enabled : Streamlined.   Config tree   =3D Version 6<5   Primary CPU   =3D 0                             =202=   HWRPB CPUs    =3D 4                                     =20 =   Page Size     =3D 8192                                  =20r5   Revision Code =3D                               =20u    Serial Number =3D 4240KSVZA001   Default CPU Capabilities:b         System: QUORUM RUN=20d   Default Process Capabilities:e         System: QUORUM RUN=20e  G CPU 0    State: RUN                CPUDB: 81030000     Handle: 00000B90tB          Owner: 00000640         Current: 00000640     Partition 0 (XXXXXX)>         ChgCnt:        0           State: Present, In-Console, Reassignable        Process: * None *   Capabilities:h*         System: PRIMARY QUORUM RUN RAD0=20   Slot Context: 99ECA180<      CPU     -  State..........: RC, PA, PP, CV, PV, PMV, PLI                 Type...........: EV68 (21264/EV68CB 21264C),  Minor =3D =a 6t)                 Speed..........: 1250 Mhz-B                 Variation......: VAX FP, IEEE FP, Primary Eligible+                 Serial Number..: JA23603284:#                 Revision.......:=20k"                 Halt Request...: 0$                 Software Comp..: 8.4(      PALCODE -  Revision Code..: 1.96-01#                 Compatibility..: 40 "                 Max Shared CPUs: 4I                 Memory  Space..: Physical =3D 00000000.00000000  Length =y =3D( 0 I                 Scratch Space..: Physical =3D 00000000.00000000  Length =p =3D, 0/   Bindings:(B      NETACP           PID =3D 2100041E  Reason: PRIMARY Capability   Fastpath:n         FGB0=20i   Features:9      Autostart - Enabled.p4      Fastpath  - Selection enabled as Preferred CPU.  G CPU 1    State: RUN                CPUDB: 81107680     Handle: 00000E90nB          Owner: 00000640         Current: 00000640     Partition 0 (XXXXX)0>         ChgCnt:        0           State: Present, In-Console, Reassignable2        Process: _NTA3:               PID: 21000453   Capabilities:e"         System: QUORUM RUN RAD0=20   Slot Context: 99ECA400<      CPU     -  State..........: RC, PA, PP, CV, PV, PMV, PLI                 Type...........: EV68 (21264/EV68CB 21264C),  Minor =3D =  6e)                 Speed..........: 1250 Mhz B                 Variation......: VAX FP, IEEE FP, Primary Eligible+                 Serial Number..: JA23602849 #                 Revision.......:=20l"                 Halt Request...: 0$                 Software Comp..: 8.4(      PALCODE -  Revision Code..: 1.96-01#                 Compatibility..: 40 "                 Max Shared CPUs: 4I                 Memory  Space..: Physical =3D 00000000.00000000  Length =o =3D  0 I                 Scratch Space..: Physical =3D 00000000.00000000  Length =t =3Di 0    Bindings:     * None *   Fastpath:          FGA0=206   Features:       Autostart - Enabled.l4      Fastpath  - Selection enabled as Preferred CPU.   $    Regards,  
 Kerry Main Senior Consultant< Hewlett-Packard (Canada) Co.! Consulting & Integration Servicesm Voice: 613-592-4660a Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)i     -----Original Message-----/ From: Mark Bowman [mailto:mb301@hotmail.com]=20, Sent: November 13, 2002 3:23 AMa To: Info-VAX@Mvb.Saic.ComF> Subject: Re: What EV am I running? (family name for processor)    = rdeininger@mindspring.com (Robert Deininger) wrote in message G news:<rdeininger-1211020740310001@1cust184.tnt2.nashua.nh.da.uu.net>...c? > In article <1d08b916.0211120431.3d10da03@posting.google.com>,i( > mb301@hotmail.com (Mark Bowman) wrote: >=20E > >Does anyknow how to find out what type of cpu you have on openVMS?c > >e  > >In Tru64 Unix its like this:- > >  > ># sizer -implver! > >EV5 >=20 > From a priv'd account: >=20 > $ ANALYZE/SYSTEM > SDA> CLUE CONFIG >=20H > This is the most complete information I know of, in partly-readable=20 > form.o  H The fields called 'CPU Type' under Per-CPU Slot Processor Information is! blank. Any other way to find out?f   OpenVMS V6.2 AlphaServer 2100 5/300=20e  # Per-CPU Slot Processor Information:i CPU ID         00    =20% CPU Type       EV5  Pass 4 (21164)=20  PAL Code       1.19-2       =20    The ES40 displayse $ sh sys/noprochC OpenVMS V7.2-2  on node FALCON 13-NOV-2002 08:19:18.06  Uptime  425s 12:52:38   CPU ID         00       =20s
     <<blank>>l PAL Code       1.93-101 =20e   Regardsa   Mark   ------------------------------    Date: 15 Nov 2002 01:08:55 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)c2 Subject: Re: Why Is ALTPRI Needed to Set Affinity?3 Message-ID: <eILD+ZZpyq7U@eisner.encompasserve.org>   j In article <3DD43D5C.50105@yahoo.spamfree.co.uk>, John Wallace <johnwallace4@yahoo.spamfree.co.uk> writes: > Stuart, Ed wrote: N >> Why should someone have to have elevated privileges to set their process toM >> have affinity with a particular CPU?  If the program is single threaded it K >> should perform better if the overhead of moving it between processors isS >> eliminated, right?  -  E The ability to interfere with the scheduling algorithms chosen by them5 system manager should require some sort of privilege.:   ------------------------------  % Date: Fri, 15 Nov 2002 14:20:14 +0100pE From: Jan C. =?iso-8859-1?Q?Vorbr=FCggen?= <jvorbrueggen@mediasec.de> 2 Subject: Re: Why Is ALTPRI Needed to Set Affinity?+ Message-ID: <3DD4F48E.B130DB0E@mediasec.de>p  H > If the program is single threaded then yes it should perform better if& > it doesn't get shunted between CPUs.  G The current scheduler already takes this into account. Indeed, with theaK introduction of SMP into VMS, the scheduler is no longer strictly priority-i based and round-robin.  " > On a bad day, you get deadlocks.   Not with a well-formed program.    	Jan   ------------------------------  + Date: Fri, 15 Nov 2002 00:17:33 +0000 (UTC)d6 From: John Wallace <johnwallace4@yahoo.spamfree.co.uk>2 Subject: Re: Why Is ALTPRI Needed to Set Affinity?1 Message-ID: <3DD43D5C.50105@yahoo.spamfree.co.uk>s   Stuart, Ed wrote: M > Why should someone have to have elevated privileges to set their process to L > have affinity with a particular CPU?  If the program is single threaded itJ > should perform better if the overhead of moving it between processors is > eliminated, right?   >  > EdG > **Please apply a generous amount of all the usual disclaimers here.**n >   E Hmmm. No takers in a fortnight. Let me try this then, in the hope of lI provoking corrections, clarifications, or comments. Apologies if this is   stating the obvious.  F My vague recollection is that it's something to do with prevention of H users shooting themselves in the feet without prior authorisation (like H lots of privileges are...). On an SMP machine, a starting assumption is F that identical CPUs are equally available to all (non-priv) users. If G some application says "only this CPU will do for me", the scheduler no  G longer has free choice of what to put where. 2 apps both being granted p? affinity for the same CPU on a 2CPU machine may result in user oE disappointment. The requirement for ALTPRI means that someone has to rG consciously be granted permission to do this; ie someone's supposed to  B have thought whether CPU affinity is A Good Thing in the relevant 3 circumstances. Sometimes it is, sometimes it isn't.e  G If the program is single threaded then yes it should perform better if cD it doesn't get shunted between CPUs. But multiprocessing life isn't B usually that simple. If some other application has also requested F affinity for the same CPU, the scheduler may have to leave a CPU idle F while the two apps compete for their preferred CPU. At best, this can C lead to disappointing performance. On a bad day, you get deadlocks.l  	 Any help?    ------------------------------  % Date: Fri, 15 Nov 2002 15:53:04 +0000 ( From: Martyn <mpattKNICKERS@bigfoot.com>$ Subject: Re: [OT] HP has big cheeses* Message-ID: <3DD51860.1000903@bigfoot.com>   Mike Kier wrote:   >>John Smith wrote:y >> >>     >> >>>[...snip...]s >>>d	 >>>      a >>>sF >>>>>HP's big cheeses have repeatedly said that only industry standard >>>>>stuff will survive. >>>>>          n >>>>>aE >>>>Aha, JF, I *knew* we'd come full circle and get back to the topicvE >>>>of cheeses :-)  Do you think that HP's big cheeses are all of the: >>>>American cheese variety ?? >>>>         >>>>H >>>Yes. They are mass-market cheeses - featureless and bland, lacking in >>>distinction.@	 >>>      o >>> A >>Yep, I was aware of that.  Mind you, some of them do sound like  >>European cheeses:u >>+ >>Fiorina (soft, older cheese with a bite),a: >>Marcello (mild cheese suitable for those with no teeth),? >>Stallard (Stahlhardt in the original German, a cheese so hard2G >>   that it's impossible to eat, and even more impossible to stomach),hE >>Winkler (another German cheese designed to "eviscerate, eviscerate,7 >>   eviscerate")g >>< >>and then there are the good cheeses, many of which seem to >>be Germanic in origin: >>A >>Hoffman (a very pleasant cheese, with much hope for the future))B >>Kleinsorge (a good cheese at heart, with very "little worry" :-)? >>Goldstein (the TECO cheese, very suitable as a Backup cheese)eE >>Skonetski (a bubbly Polish cheese, very popular among ambassadorialu >>   types)g >>@ >>And how about some of the other cheeses found in comp.os.vms ? >>D >>Harrison (a cheese with a chip-on-its-shoulder, which often leaves >>   a bitter taste)G >>Mezei (a cross between French and German cheeses, energetic, suitablek >>   for bicyclists) >> >> >>Any more suggestions ? >> >>Roy Omond ! >>Blue Bubble Cheese Factory Ltd.d >>> >>(Blue Bubble Cheese ?  Now there's something that would have >>potential in the USofA :-) >>     >> >@L >Well, there's Kilgallen... it is rumored to be a well-dressed, mildly aged,G >New England cheese of Celtic heritage, but it is kept under such tight ( >security that no one has ever found it. >o >-- 
 >Mike Kier1 >Consultant, HP Consulting & Integration Services  >Cincinnati, OH, USA >mike.kier@hp.coma >e( >"Practice random acts of VMS marketing" >E >M >    > : This group gets more like a Monty Python sketch every day.     -- i, Remove KNICKERS before replying by Email ;-)   ------------------------------  % Date: Fri, 15 Nov 2002 16:27:41 +0000e( From: Nic Clews <sendspamhere@127.0.0.1>$ Subject: Re: [OT] HP has big cheeses) Message-ID: <3DD5207D.ED393BAD@127.0.0.1>s  
 Martyn wrote:a >  > Mike Kier wrote: >  > >>John Smith wrote:c > >>    < > This group gets more like a Monty Python sketch every day. >   4 Except there wasn't any cheese in their Cheese Shop.  H I suppose I resemble a rather mature piece of Stilton, well past its USE< BY date which emits a periodic strong but distinctive odour.   -- h? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Scienceso nclews at csc dot coma   ------------------------------   End of INFO-VAX 2002.632 ************************