1 INFO-VAX	Wed, 30 Apr 2003	Volume 2003 : Issue 237       Contents: Active Telnet sessions ADDENDUM: Keyboard ? Re: ADDENDUM: Keyboard ? RE: ADDENDUM: Keyboard ? Re: ADDENDUM: Keyboard ? RE: ADDENDUM: Keyboard ? Re: Bug in f$search  Re: Bug in f$search + Re: Configuring FTP service On Open VMS 7.1 + Re: Configuring FTP service On Open VMS 7.1 + Re: Configuring FTP service On Open VMS 7.1 - Re: DCL question - null byte in string symbol / Does anyone have Perfect Disk 6.0 - 6.2 manuals & For European hobbyists with a DS10 .../ Gartner: HP is #1 vendor of servers for Q1 2003 3 Re: Gartner: HP is #1 vendor of servers for Q1 2003  Help with Vest Re: Help with Vest Re: Help with Vest9 Re: HELP: Running Out Of Vitrual Memory When Compiling... 9 Re: HELP: Running Out Of Vitrual Memory When Compiling... = Re: How Alpha will save Itanium - must reading for Bill Todd! = Re: How Alpha will save Itanium - must reading for Bill Todd! = Re: How Alpha will save Itanium - must reading for Bill Todd! $ Re: I learned about VMS from that...H Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyP Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopoly monopol2 Improvements in TCP/IP Services anti-spam features6 Re: Improvements in TCP/IP Services anti-spam features6 Re: Improvements in TCP/IP Services anti-spam features6 Re: Improvements in TCP/IP Services anti-spam features8 Re: Itanium Madison blasts Sun, IBM in encryption specs!8 Re: Itanium Madison blasts Sun, IBM in encryption specs!8 Re: Itanium Madison blasts Sun, IBM in encryption specs!8 Re: Itanium Madison blasts Sun, IBM in encryption specs!  Re: Johnny English is a VMS user
 Keyboard ? Re: Keyboard ?G Moving on / Leatning to let go. . .(Was: Re: Why did Digital go broke?) & Re: Not entirely OT: RSHELL to Solaris& Re: Not entirely OT: RSHELL to Solaris& Re: Not entirely OT: RSHELL to Solaris& Re: Not entirely OT: RSHELL to Solaris& Re: Not entirely OT: RSHELL to Solaris6 Re: OpenVMS Itanium port progressing well says Gorham!6 Re: OpenVMS Itanium port progressing well says Gorham!P Re: OpenVMS on non-HP Itanium boxes (was: Re: OpenVMS Itanium port progressing w9 opinions on implementing SNMP on OpenVMS 6.2 with UCX 4.2 % Random access to the TZ89 tape system ) Re: Random access to the TZ89 tape system ) Re: Random access to the TZ89 tape system ) Re: Random access to the TZ89 tape system  Re: RZ26-B in the real world Re: RZ26-B in the real world Re: RZ26-B in the real world Re: RZ26-B in the real world stupid keyboard question Re: stupid keyboard question Re: stupid keyboard question4 Re: TCPIP: SMTP sender design problem (bounced mail)* Two day OpenVMS seminar in Ottawa, Ontario
 Upgrade Query  Re: Upgrade Query 
 Re: VMS 7.3-1 
 Re: VMS 7.3-1   F ----------------------------------------------------------------------  % Date: Tue, 29 Apr 2003 14:56:56 -0700 * From: "Jack Peacock" <peacock@simconv.com> Subject: Active Telnet sessions 2 Message-ID: <L6OcnQzL2be3aDOjXTWcqw@mpowercom.net>  L Environment is VMS 7.2-1H1, TCPIP V5.0A ECO 3.  There are seven users loggedI in with Telnet from PCs, and a SHOW DEV TN displays seven online devices. J However, when I go to TCPIP and do a SHOW SERVICES/FULL TELNET it displaysB 44 active connections (limit set to 50).  What are all these extra1 connections and how can I find out what they are?    Service: TELNET -                            State:     Enabled G Port:               23     Protocol:  TCP             Address:  0.0.0.0 K Inactivity:          1     User_name: not defined     Process:  not defined C Limit:              50     Active:     44             Peak:      50   
 $ show dev tn   . Device                  Device           Error.  Name                   Status           Count. TNA0:                   Offline              0. TNA3159:                Online               0. TNA3162:                Online               0. TNA3176:                Online               0. TNA3177:                Online               0. TNA3194:                Online               0. TNA3195:                Online               0. TNA3197:                Online               0  F There is a valid user logged in on each TN device.  There are no print queues on reverse telnet.   D Is it possible to have an active telnet connection and no TN device?    Jack Peacock    ------------------------------  % Date: Tue, 29 Apr 2003 15:19:36 -0700 # From: "Tom Linden" <tom@kednos.com>  Subject: ADDENDUM: Keyboard ? 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIIEHEHBAA.tom@kednos.com>   8 This is in a cluster so I can get at from another system   >-----Original Message----- ) >From: Tom Linden [mailto:tom@kednos.com] & >Sent: Tuesday, April 29, 2003 3:11 PM >To: Info-VAX@Mvb.Saic.Com >Subject: Keyboard ? >  > F >Have an LK401 KB attached to a 3000/300LX running 6.2.  I, er, spilltL >some liquid on the KB so I quickly unplug to clean it.  When I plug it backE >it back in no response.  I am sure the KB is OK.  The LEDs are dark. B >Do I have to power down or is there some way to get this up?  The >attachment J >to the box is actually through one of those funny cables with DB15 at one >end and6 >a little box where both the KB and the mouse plug in. >  >  >---' >Outgoing mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.476 / Virus Database: 273 - Release Date: 4/24/2003  >  >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.476 / Virus Database: 273 - Release Date: 4/24/2003  >  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.476 / Virus Database: 273 - Release Date: 4/24/2003   ------------------------------  % Date: Tue, 29 Apr 2003 21:18:24 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)! Subject: Re: ADDENDUM: Keyboard ? L Message-ID: <rdeininger-2904032118240001@user-105n8do.dialup.mindspring.com>  F In article <CIEJLCMNHNNDLLOOGNJIIEHEHBAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> wrote:   9 >This is in a cluster so I can get at from another system  >  >>-----Original Message-----* >>From: Tom Linden [mailto:tom@kednos.com]' >>Sent: Tuesday, April 29, 2003 3:11 PM  >>To: Info-VAX@Mvb.Saic.Com  >>Subject: Keyboard ?  >> >>G >>Have an LK401 KB attached to a 3000/300LX running 6.2.  I, er, spillt M >>some liquid on the KB so I quickly unplug to clean it.  When I plug it back F >>it back in no response.  I am sure the KB is OK.  The LEDs are dark.C >>Do I have to power down or is there some way to get this up?  The  >>attachmentK >>to the box is actually through one of those funny cables with DB15 at one 	 >>end and 7 >>a little box where both the KB and the mouse plug in.   E Ok, on second thought, restarting the DECW server on the system might / execute a code path that re-inits the keyboard.   1 When that doesn't work, cycle the system's power.    ------------------------------  % Date: Tue, 29 Apr 2003 19:32:57 -0700 # From: "Tom Linden" <tom@kednos.com> ! Subject: RE: ADDENDUM: Keyboard ? 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIAEHKHBAA.tom@kednos.com>    >-----Original Message----- : >From: Robert Deininger [mailto:rdeininger@mindspring.com]& >Sent: Tuesday, April 29, 2003 6:18 PM >To: Info-VAX@Mvb.Saic.Com" >Subject: Re: ADDENDUM: Keyboard ? >  > G >In article <CIEJLCMNHNNDLLOOGNJIIEHEHBAA.tom@kednos.com>, "Tom Linden"  ><tom@kednos.com> wrote: > : >>This is in a cluster so I can get at from another system >> >>>-----Original Message----- + >>>From: Tom Linden [mailto:tom@kednos.com] ( >>>Sent: Tuesday, April 29, 2003 3:11 PM >>>To: Info-VAX@Mvb.Saic.Com >>>Subject: Keyboard ? >>>  >>> H >>>Have an LK401 KB attached to a 3000/300LX running 6.2.  I, er, spilltA >>>some liquid on the KB so I quickly unplug to clean it.  When I 
 >plug it back G >>>it back in no response.  I am sure the KB is OK.  The LEDs are dark. D >>>Do I have to power down or is there some way to get this up?  The
 >>>attachment L >>>to the box is actually through one of those funny cables with DB15 at one
 >>>end and8 >>>a little box where both the KB and the mouse plug in. > F >Ok, on second thought, restarting the DECW server on the system might0 >execute a code path that re-inits the keyboard.  K Well I know hot to kill it, but how to restart it?  Looking at the system I  have the following runningF 204000A4 DECW$SERVER_0   HIB      8     2126   0 00:04:09.22       683 874 F 204000A9 DECW$SESSION    LEF      6     1864   0 00:00:09.14      3120 912 F 204000AB DECW$TE_00AB    LEF      6      464   0 00:02:42.04      2072 921 F 204000AE DECW$MWM        LEF      4      433   0 00:00:05.27      1837 537    > 2 >When that doesn't work, cycle the system's power. >  >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.476 / Virus Database: 273 - Release Date: 4/24/2003  >  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.476 / Virus Database: 273 - Release Date: 4/24/2003   ------------------------------  % Date: Wed, 30 Apr 2003 05:54:45 +0200 2 From: martin@radiogaga.harz.de (Martin Vorlaender)! Subject: Re: ADDENDUM: Keyboard ? ; Message-ID: <3eaf4905.524144494f47414741@radiogaga.harz.de>   " Tom Linden (tom@kednos.com) wrote:4 > Robert Deininger <rdeininger@mindspring.com> wroteI > > Ok, on second thought, restarting the DECW server on the system might 3 > > execute a code path that re-inits the keyboard.  > : > Well I know hot to kill it, but how to restart it? [...]  $  $ @SYS$MANAGER:DECW$STARTUP RESTART   cu,    Martin --  G                            | Martin Vorlaender  |  VMS & WNT programmer 4 Microsoft isn't the Borg:  | work: mv@pdv-systeme.deK the Borg have proper       |       http://www.pdv-systeme.de/users/martinv/ ; networking.                | home: martin@radiogaga.harz.de    ------------------------------   Date: 29 Apr 2003 23:19 CDT ' From: carl@gerg.tamu.edu (Carl Perkins) ! Subject: RE: ADDENDUM: Keyboard ? - Message-ID: <29APR200323192969@gerg.tamu.edu>   ' "Tom Linden" <tom@kednos.com> writes... L }Well I know hot to kill it, but how to restart it?  Looking at the system I }have the following running G }204000A4 DECW$SERVER_0   HIB      8     2126   0 00:04:09.22       683  }874G }204000A9 DECW$SESSION    LEF      6     1864   0 00:00:09.14      3120  }912G }204000AB DECW$TE_00AB    LEF      6      464   0 00:02:42.04      2072  }921G }204000AE DECW$MWM        LEF      4      433   0 00:00:05.27      1837  }537  # $ @sys$manager:decw$startup restart   2 This will both shut it down and start it up again.   --- Carl   ------------------------------  # Date: Wed, 30 Apr 2003 01:24:10 GMT   From: Rob Brown <brown@gmcl.com> Subject: Re: Bug in f$searchL Message-ID: <Pine.LNX.4.44.0304291923370.22401-100000@localhost.localdomain>  D I also note that you are not specifying ";*" in your F$SEARCH calls.     --    / Rob Brown                        brown@gmcl.com A G. Michaels Consulting Ltd.      (866)438-2101 (voice) toll free! 6 Edmonton                         (780)438-9343 (voice)4                                  (780)437-3367 (FAX)1                                  http://gmcl.com/    ------------------------------  % Date: Tue, 29 Apr 2003 20:41:14 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: Bug in f$search' Message-ID: <3EAF29BA.1483246E@fsi.net>    brandon@dalsemi.com wrote: >  > In German no less... > G > > $ device = f$parse (f$search ("klima$datei:*.html_tmp"),,,"device") M > > $ directory = f$parse (f$search ("klima$datei:*.html_tmp"),,,"directory") C > > $ file = f$parse (f$search ("klima$datei:*.html_tmp"),,,"name")  > O > For starters, I would suggest that you remove the multiple search streams and  > have only one search stream: > 8 > $ res_filespec = f$search ("klima$datei:*.html_tmp",0)0 > $ device = f$parse ( res_filespec ,,,"device")6 > $ directory = f$parse ( res_filespec ,,,"directory"), > $ file = f$parse ( res_filespec ,,,"name")  D Exactly, and Doc Cypher explained why: Each invocation of F$SEARCH()E will return the next matching filespec. So, by repeating the function = call, you introduce the "bug" (it's yours, not F$SEARCH()'s).   D Also, I'd question the use of a logical in this application. It doesG work, but may not perform as expected in every case; and changes to DCL D and/or other system services in another VMS version may "break" your code.    I'd use:  4 $ klima$datei := DISK$USER:[MANSER.ARBEIT.global...]8 $ res_filespec = f$search ("''klima$datei'*.html_tmp",0)  ) ...but that's just me. YMMV, as always...   E John is also quite correct to mention that when F$SEARCH() returns an E empty string, that's the end of the list. Using the stream identifier . allows the use of multiple F$SEARCH() streams.   Good suggestions all...    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Tue, 29 Apr 2003 18:44:17 GMT . From: eplan@langstoeger.at (Peter LANGSTOEGER)4 Subject: Re: Configuring FTP service On Open VMS 7.13 Message-ID: <5Kzra.80897$v62.856490@news.chello.at>   m In article <20030429090953.19211.qmail@gacracker.org>, Doc.Cypher <Use-Author-Address-Header@[127.1]> writes: F >On Tue, 29 Apr 2003, peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) >wrote: ? >>In article <cb950e29.0304282214.31d8ccf2@posting.google.com>, 2 >>shakeelj2k@yahoo.com (shakeel-ur-rehman) writes: >>>I am new one to Open VMS  >>0 >>Then welcome to god's own operating system ;-) >> >>Start to see0 >>the VMS homepage	http://www.openvms.compaq.com0 >>the VMS doc	http://www.openvms.compaq.com/doc/C >>the VMS FAQ	http://www.openvms.compaq.com/wizard/openvms_faq.html  > > >I'm afraid the semi-sensible Compaq URLs cannot be relied on.  J I think, you can rely on them, more than you can on the hXY000.wwwZ.hp.com  I >These are the URLs the above redirect to, at least for the time being...     as you write "the time being..."   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 29 Apr 2003 20:36:13 -00004 From: Doc.Cypher <Use-Author-Address-Header@[127.1]>4 Subject: Re: Configuring FTP service On Open VMS 7.15 Message-ID: <20030429203613.9928.qmail@gacracker.org>   D On Tue, 29 Apr 2003, eplan@langstoeger.at (Peter LANGSTOEGER) wrote:B >In article <20030429090953.19211.qmail@gacracker.org>, Doc.Cypher, ><Use-Author-Address-Header@[127.1]> writes:  ? >>I'm afraid the semi-sensible Compaq URLs cannot be relied on.  > K >I think, you can rely on them, more than you can on the hXY000.wwwZ.hp.com   I If you take http://www.openvms.compaq.com and ask Lynx to retrieve it and ! log errors you get the following:   +    URL=http://www.openvms.compaq.com/ (GET) & STATUS=HTTP/1.1 301 Moved Permanently '    URL=http://h71000.www7.hp.com/ (GET)  STATUS=HTTP/1.1 200 OK  J >>These are the URLs the above redirect to, at least for the time being... > ! >as you write "the time being..."   I Yes, I'm not particularly impressed with these ugly looking things. I can A just imagine the http://bonzo.da-da.hp.com URL getting changed to B http://bonzo.do-da.hp.com and the redirects being forgotten about.  C I also wonder if HP have embraced VMS the same way they have Linux.   9 After all, http://www.linux.hp.com works. Whereas none of 1 http://www.openvms.hp.com, http://www.vms.hp.com, 7 http://www.hp.com/openvms, or http://www.hp.com/vms do.      Doc. --  6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                             https://vmsbox.cjb.net K                                                    http://althacker.cjb.net    ------------------------------  # Date: Tue, 29 Apr 2003 22:01:48 GMT 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)4 Subject: Re: Configuring FTP service On Open VMS 7.13 Message-ID: <gDCra.83628$v62.898553@news.chello.at>   l In article <20030429203613.9928.qmail@gacracker.org>, Doc.Cypher <Use-Author-Address-Header@[127.1]> writes:D >I also wonder if HP have embraced VMS the same way they have Linux.   Surely not.   : >After all, http://www.linux.hp.com works. Whereas none of2 >http://www.openvms.hp.com, http://www.vms.hp.com,8 >http://www.hp.com/openvms, or http://www.hp.com/vms do.  ( But http://www.hp.com/go/openvms does...   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 29 Apr 2003 19:39:24 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)6 Subject: Re: DCL question - null byte in string symbol= Message-ID: <b096a4ee.0304291839.6e88c758@posting.google.com>   b lewis@PROBE.mitre.org (Keith A. Lewis) wrote in message news:<b7mf5b$am1$1@newslocal.mitre.org>... > norm.raphael@metso.com writes in article <OFF16CBE7A.94AB9BF2-ON85256D0A.00519FD8@metso.com> dated Wed, 16 Apr 2003 11:13:09 -0400:  > >  > >Is this expected behavior?  >  > I'd call it a bug in DCL.    >  > >$ type null_test.com  > >$ null1[0,8]=0  > >$ show symbol null1D > >$ if "''null1'" .eqs. "0"  then  write sys$output "null1 is zero" > M > ASCII 0 is the only character that fails in that way, but ASCII 34 (") also  > produces a weird result.  D This is expected behavior for ASCII 34 ("). During the first scan of> the command line, the symbol and its enclosing apostrophes are( replaced by a quotation mark. So you get   $ if """ .eqs. "0" ...  C Now, since DCL just completed a symbol substitution, it rescans the D line. But now we have three consecutive quotation marks, after whichE DCL winds up in "string mode" or "quote mode" in which characters are F read verbatim (except for quotation marks!) and appended. But in orderE for the rest of the command to be interpreted properly, DCL has to be < in "command mode" in which single apostrophes trigger symbolE substitution and exclamation marks are treated as comment characters.   B IOW, you now are starting a quoted string with the first quotationE mark, the next two quotation marks become a single quotation mark and D the first character of that string, then the next few characters are@ appended verbatim until the next quotation mark which closes theB string, which is immediately followed by a 0. It is no wonder that this command fails.   E One must be careful doing symbol substitution within a quoted string. F If the value of the symbol being substituted contains quotation marks,F that could ruin things if not done carefully. (I actually had to writeD a DCL program that doubles all quotation marks in a given string forC use as a quoted DCL parameter -- it is based on a similar routine I  stumbled upon on c.o.v.)  B Strangely, another oddity is the fact that lexical functions being= substituted within a quoted string should contain only single  quotation marks!   For example:   $ SH SYM EV !   EV*ALUATE == "WRITE SYS$OUTPUT"  $ EV "''F$CVTIME("2:35")'" 2003-04-30 02:35:00.00 $ EV "''F$CVTIME(""2:35"")'"@ %DCL-W-SYMDEL, invalid symbol or value delimiter - check command syntax  \2\ $   < I once tried this with various lexical functions and various@ arguments, and I found that in some cases you *could* double theB quotation marks and all works fine, but there was no case in whichB doubled quotation marks worked and single didn't. If single didn'tD work, neither did double. AFAIK, this behavior of evaluating lexicalF functions within quoted strings is not documented except as part of an2 example whose purpose was something else entirely.  B I never tried null characters though! Thanks to this thread, now I know about them too! Thanks!   [...]    Disclaimer: JMHO Alan E. Feldman    ------------------------------    Date: 29 Apr 2003 11:26:49 -0700+ From: c00per11242001@yahoo.ca (Vic Mendham) 8 Subject: Does anyone have Perfect Disk 6.0 - 6.2 manuals= Message-ID: <f7a73cb1.0304291026.25bbebaa@posting.google.com>   F I have tried to get old Perfect disks 6.0 manuals from Raxco, but they? only have the latest. Does anyone have these in HTML of PDF for  OpenVMS?   Regards    c00per11242001@yahoo.ca    ------------------------------  % Date: Tue, 29 Apr 2003 23:38:24 +0200  From: Dirk Munk <munk@home.nl>/ Subject: For European hobbyists with a DS10 ... 2 Message-ID: <b8mrko$3v8$1@news1.tilbu1.nb.home.nl>  P As you may have noticed there is a new graphics patch for VMS 7.3-1. This patch J allows the use of ATi Radeon 7500 graphics cards, in PCI and AGP versions.  P The PCI version is only supported (= tested ?) on the DS10. Our good friends at N HP will be glad to provide you with such a card, however their price is a bit  steep.  Q In the US you can get a original ATi card for about $70, so that is nice for the  
 US hobbyists.   P We in Europe are a bit less fortunate, since ATi no longer sells graphics cards M in Europe. They leave the European market for other (mostly Asian) companies  O using their chipsets. And guess what, those companies don't bother producing a   Radeon 7500 PCI version.  O Ordering an original ATI card in the US is very difficult, since most Internet  O stores don't ship outside the US, or say they do so, but make it impossible to  P enter an address outside the US (very clever those guys...) I found one company J that would ship outside the US, but the shipping costs were more then $50.  N So I searched and searched the web untill I found a European company that has  such a card in its program.   + This is their URL: http://www.club-3d.nl/ .   Q I ordered the card through a local PC shop, installed it in a DS10, and it works  N like a dream. I payed less then 80 Euro for it, but I also had to buy a DVI > P VGA adapter for 15 Euro. (the card has two ports, a primary DVI and a secondary 
 VGA port).  O The card seems to be a 100% ATi clone, even the ATi brand name is on the board.   P So if you are the happy owner of a DS10, now you know where to get a moderately 4 priced and quite good graphics card for your system.   ------------------------------    Date: 29 Apr 2003 18:59:50 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris) 8 Subject: Gartner: HP is #1 vendor of servers for Q1 2003= Message-ID: <cf15391e.0304291759.5162026b@posting.google.com>   @ "HP (NYSE:HPQ) is the No. 1 vendor of servers worldwide, with 29A percent unit market share for the first calendar quarter of 2003, F according to preliminary results released today by Gartner Dataquest. 9 HP leads its closest competition by 9 percentage points."   9 http://www.hp.com/hpinfo/newsroom/press/2003/030428c.html    ------------------------------  % Date: Tue, 29 Apr 2003 21:30:49 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> < Subject: Re: Gartner: HP is #1 vendor of servers for Q1 2003' Message-ID: <3EAF3559.578908A0@fsi.net>    Keith Parris wrote:  > B > "HP (NYSE:HPQ) is the No. 1 vendor of servers worldwide, with 29C > percent unit market share for the first calendar quarter of 2003, G > according to preliminary results released today by Gartner Dataquest. ; > HP leads its closest competition by 9 percentage points."  > ; > http://www.hp.com/hpinfo/newsroom/press/2003/030428c.html   H I'm sure it's safe to assume that this is due to the sales of Proliants,H none - or a negligible percent - of which run VMS + some VAX emulator on another o.s.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 29 Apr 2003 13:43:07 -0700' From: pat.szuch@swagelok.com (PatrickS)  Subject: Help with Vest = Message-ID: <761a8bd8.0304291243.7691bfa9@posting.google.com>    Hello,  @ I have a single executable that I need to port from VMS to Alpha v7.3-1. 9 When I run the Vest/dependency command against it, I get:     $ Spvm02> vest /dependency csgscntl_21% %VEST-I-READIMAGE, Reading image file % SYS$OPSUSERS:[PSZUCH]CSGSCNTL_21.EXE; F %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]VAXCRTL.EXE; as input -RMS-E-FNF, file not foundD %VEST-I-READIMAGE, Reading image file SYS$COMMON:[SYSLIB]LIBRTL.EXE;E %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]MTHRTL.EXE; as input  -RMS-E-FNF, file not foundD %VEST-I-READIMAGE, Reading image file SYS$COMMON:[SYSLIB]FDLSHR.EXE;E %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]FORRTL.EXE; as input  -RMS-E-FNF, file not foundF %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]LIBRTL2.EXE; as input -RMS-E-FNF, file not foundD %VEST-I-READIMAGE, Reading image file SYS$COMMON:[SYSLIB]SCRSHR.EXE;E %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]PASRTL.EXE; as input  -RMS-E-FNF, file not foundE %VEST-I-READIMAGE, Reading image file SYS$COMMON:[SYSLIB]SORTSHR.EXE; @ %VEST-F-DEPENDFATAL, Complete dependency analysis was impossible    E As you can see, I have 5 RTL's that vest isn't finding.  If I look in 9 sys$common:[syslib], I find vested versions of these (ie. E VAXCRTL_D56_TV.EXE), but obviously, no native versions.  How do I get D Vest to recognize the current (translated) versions of these runtimeA libraries.  Is it as simple as defining logicals such as "$define ; vaxcrtl vaxcrtl_d56_tv"?  It would seem that this should be  unnecessary.  D Any ideas?  I have no experience with the Vest product, but it seemsF my only hope for getting the exectable in question ported to Alpha, as! the source code is not available.    Thanks! 
 Patrick Szuch  DBA, Swagelok Company  pat.szuch@swagelok.com   ------------------------------  # Date: Wed, 30 Apr 2003 00:59:53 GMT 4 From: Craig A. Berry <craigberry@mac.com.spamfooler> Subject: Re: Help with Vest 7 Message-ID: <7af67ac918fc2ad740d5e61c7b77f7fd@TeraNews>   D In <761a8bd8.0304291243.7691bfa9@posting.google.com> PatrickS wrote: > Hello, > B > I have a single executable that I need to port from VMS to Alpha	 > v7.3-1. ; > When I run the Vest/dependency command against it, I get:   F I'm not experienced with the particular problem you are encountering, C but since you don't mention any version numbers I suggest you make  F yourself acquainted with the latest VAX-to-Alpha translation software . and documentation (if you haven't already) at:  = <http://h71000.www7.hp.com/openvms/products/omsva/omsva.html>    ------------------------------  % Date: Tue, 29 Apr 2003 20:49:14 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: Help with Vest ' Message-ID: <3EAF2B9A.FA8A4BDD@fsi.net>    PatrickS wrote:  >  > Hello, > B > I have a single executable that I need to port from VMS to Alpha	 > v7.3-1. ; > When I run the Vest/dependency command against it, I get:  > & > Spvm02> vest /dependency csgscntl_21' > %VEST-I-READIMAGE, Reading image file ' > SYS$OPSUSERS:[PSZUCH]CSGSCNTL_21.EXE; H > %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]VAXCRTL.EXE; as input > -RMS-E-FNF, file not foundF > %VEST-I-READIMAGE, Reading image file SYS$COMMON:[SYSLIB]LIBRTL.EXE;G > %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]MTHRTL.EXE; as input  > -RMS-E-FNF, file not foundF > %VEST-I-READIMAGE, Reading image file SYS$COMMON:[SYSLIB]FDLSHR.EXE;G > %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]FORRTL.EXE; as input  > -RMS-E-FNF, file not foundH > %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]LIBRTL2.EXE; as input > -RMS-E-FNF, file not foundF > %VEST-I-READIMAGE, Reading image file SYS$COMMON:[SYSLIB]SCRSHR.EXE;G > %VEST-F-OPENIN, Error opening SYS$COMMON:[SYSLIB]PASRTL.EXE; as input  > -RMS-E-FNF, file not foundG > %VEST-I-READIMAGE, Reading image file SYS$COMMON:[SYSLIB]SORTSHR.EXE; B > %VEST-F-DEPENDFATAL, Complete dependency analysis was impossible > G > As you can see, I have 5 RTL's that vest isn't finding.  If I look in ; > sys$common:[syslib], I find vested versions of these (ie. G > VAXCRTL_D56_TV.EXE), but obviously, no native versions.  How do I get F > Vest to recognize the current (translated) versions of these runtimeC > libraries.  Is it as simple as defining logicals such as "$define = > vaxcrtl vaxcrtl_d56_tv"?  It would seem that this should be  > unnecessary. > F > Any ideas?  I have no experience with the Vest product, but it seemsH > my only hope for getting the exectable in question ported to Alpha, as# > the source code is not available.   = Have a look at SHOW LOGICALS on OpenVMS Alpha, especially the  LNM$SYSTEM_TABLE table:     $ pipe sh log | sear sys$pipe tv   "ACLEDTSHR_TV" = "ACLEDTSHR"!   "BASRTL2_TV" = "BASRTL2_D53_TV"    "BASRTL_TV" = "BASRTL_D56_TV" #   "BLAS1RTL_TV" = "BLAS1RTL_D53_TV"    "CDDSHR_TV" = "CDDSHR""   "CMA$LIB_SHR_TV" = "CMA$LIB_SHR",   "CMA$OPEN_LIB_SHR_TV" = "CMA$OPEN_LIB_SHR"$   "CMA$OPEN_RTL_TV" = "CMA$OPEN_RTL"   "CMA$RTL_TV" = "CMA$RTL"   "COBRTL_TV" = "COBRTL_D56_TV"    "CONVSHR_TV" = "CONVSHR"   "CRFSHR_TV" = "CRFSHR"   "DBLMSGMGR" = "DBLMSGMGR_TV"   "DBLRTL_TV" = "DBLRTL_D56_TV"    "DCXSHR_TV" = "DCXSHR"   "DISMNTSHR_TV" = "DISMNTSHR"   "DTKSHR_TV" = "DTKSHR"   "EDTSHR_TV" = "EDTSHR"   "ENCRYPSHR_TV" = "ENCRYPSHR"   "EPC$SHR_TV" = "EPC$SHR"   "FDLSHR_TV" = "FDLSHR"   "FORRTL_TV" = "FORRTL_D56_TV"    "INIT$SHR_TV" = "INIT$SHR"   "LBRSHR_TV" = "LBRSHR"!   "LIBRTL2_TV" = "LIBRTL2_D56_TV"    "LIBRTL_TV" = "LIBRTL_D56_TV"    "MAILSHR_TV" = "MAILSHR"   "MONTORSHR_TV" = "MONTORSHR"   "MOUNTSHR_TV" = "MOUNTSHR"   "MTHRTL_TV" = "MTHRTL_D53_TV"    "NCSSHR_TV" = "NCSSHR"   "PASRTL_TV" = "PASRTL_D56_TV"    "PLIRTL_TV" = "PLIRTL_D56_TV"    "PPLRTL_TV" = "PPLRTL",   "PTD$SERVICES_SHR_TV" = "PTD$SERVICES_SHR"   "SCRSHR_TV" = "SCRSHR"%   "SECURESHR_TV" = "SECURESHR_JACKET"    "SMBSRVSHR_TV" = "SMBSRVSHR"   "SMGSHR_TV" = "SMGSHR"   "SORTSHR_TV" = "SORTSHR"   "SPISHR_TV" = "SPISHR"   "TECO32" = "TECO32_TV"   "TECOSHR" = "TECOSHR_TV"   "TPUSHR_TV" = "TPUSHR"#   "VAXCRTLG_TV" = "VAXCRTLG_D56_TV" !   "VAXCRTL_TV" = "VAXCRTL_D56_TV"    "VMSRTL" = "VMSRTL_TV"  C (from V7.3-1 on my ISP's Alpha machine - I can TELNET to it with my  usual username/password)   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Tue, 29 Apr 2003 14:16:55 -0500 2 From: Anonymous #314 <Anon314@mail.ourservers.net>B Subject: Re: HELP: Running Out Of Vitrual Memory When Compiling...2 Message-ID: <3EAE8957.E33FAE9@mail.ourservers.net>  8 Thanks to everyone that's helped, the problem is solved.  G I just doubled the PGFLQUO for my account (which someone suggested) and  that did the trick.    Thanks everyone.   ------------------------------  % Date: Wed, 30 Apr 2003 00:54:32 -0400   From: John Santos <JOHN@egh.com>B Subject: Re: HELP: Running Out Of Vitrual Memory When Compiling...5 Message-ID: <1030430005217.2697A-100000@Ives.egh.com>   . On 29 Apr 2003 briggs@encompasserve.org wrote:  h > In article <3EAE56F1.CCB6A3@mail.ourservers.net>, Anonymous #314 <Anon314@mail.ourservers.net> writes:
 > > Hello. > > J > > I'm kinda new to OpenVMS (I like what I've seen so far) and I'm tryingB > > to compile a C program and I keep getting the following error. > > ? > > Fatal: Insufficient vitrual memory to continue compilation. 1 > > %LIB-F-INSVIRMEM, insufficient virtual memory  > > C > > Its an Alpha machine, OpenVMS v7.2 (Hobbyist), 256MEG RAM, DECC  > > C > > What do I have to change where to solve this problem.  Is it a  2 > > user limits thing or more of a sysgen setting? > 3 > There are three things that limit virtual memory.  > H > 1.  Page file space.  If you don't have a big enough page file, you'llD > run out of virtual memory.  The symptom is a system hang, possiblyK > with messages such as "Pagefile badly fragmented, attempting to continue" @ > on the console.  That's not your problem now.  Though it could; > become your problem if you raise one of the other limits.  > C > 2.  Process page file quota.  In order to reduce the risk of page B > file exhaustion and the resulting system hang, VMS establishes aC > page file quota for each process (actually, I think it's a pooleddG > quota, but that's irrelevant for our purposes -- you're not using LSEb7 > and spawning the C compile, are you?).  You typicallyo6 > control page file quota using the AUTHORIZE utility: >  > 	$ SET DEFAULT SYS$SYSTEM  > 	$ RUN AUTHORIZE > 	UAF> SHOW your-username7 > 	UAF> MODIFY your-username /PGFLQUOTA=new-quota-valuee > E > 3.  SYSGEN parameter VIRTUALPAGECNT.  On VAX at least, process pageaD > tables live in the process header.  And the process headers of allD > resident (not swapped out) processes are in a virtually contiguousF > array with fixed size entries (balance slots).  You control the sizeC > of those process page tables with the VIRTUALPAGECNT paramter.  A D > very real constraint (on VAX) was that the process page tables forB > all resident processes had to fit into the 1 gigabyte S0 virtual@ > address space.  So the product of VIRTUALPAGECNT and BALSETCNT> > was critical.  On Alpha, that constraint has been lifted and3 > VIRTUALPAGECNT is not a serious issue any longer.  > B > Physical memory (your 256 meg) is essentially irrelevant when it) > comes to running out of virtual memory.  >  > K > On the other side of the problem, you have a C compilation that's blowingeF > out virtual memory.  Is there something about that C program that is@ > nasty?  Recursive include files, humonguous array definitions,? > convoluted compile time expression evaluation?  If you doublesC > your process page file quota a couple of times and bump your pagetH > file size up to match and still have the problem, you may need to look > at your C code for issues. >  > 	John Briggs  @ I think the C compiler just uses more memory than it used to.  IE was recently compiling SAMBA 2.2.8 on a VAX and had the same problem. / (In my case, I had to raise VIRTUALPAGECNT ...)s     -- B John Santoso Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  # Date: Tue, 29 Apr 2003 21:16:59 GMTd3 From: wallacethinmintr@eircom.net (Russell Wallace)dF Subject: Re: How Alpha will save Itanium - must reading for Bill Todd!0 Message-ID: <3eaeeb8b.344824410@news.eircom.net>  E On Mon, 28 Apr 2003 00:10:50 -0400, "Main, Kerry" <Kerry.Main@hp.com>i wrote:  9 >As to it not being the same architecture, there are many I >changes between the Merced, McKinley, Madison, Montecito etc in the samepH >manner as there are many architecture differences between EV4, EV5, EV6E >and EV7 Alpha architectures. I only stated that the IPF-2 (Mckinley)c@ >would be different than the initial Merced and it certainly is.  F Has the architecture changed? I thought Itanium 2 implemented the sameD architecture as Itanium 1, just with a different design, but I could	 be wrong.    -- 6 "Sore wa himitsu desu."  To reply by email, removet the small snack from address.p! http://www.esatclear.ie/~rwallacey   ------------------------------  % Date: Wed, 30 Apr 2003 01:10:33 +0200a From: Dirk Munk <munk@home.nl>F Subject: Re: How Alpha will save Itanium - must reading for Bill Todd!2 Message-ID: <b8n11j$eq7$1@news1.tilbu1.nb.home.nl>   Bob Ceculski wrote: 2 > here is how Alpha will save Itanium Bill ... you3 > can click link and read, but I especially cut outn0 > the paragraph for Bill how Alpha EV8-9 designs" > will fit nicely with Itanium ... >  > * > http://www.theinquirer.net/?article=9127 >  > . I have read the article and this whole thread.  O No one can accuse Bill of being the most diplomatic contributor to this group,  9 but I can certainly understand his anger and frustration..  D Over the past decades we customers had to deal with non-existing or L contra-productive marketing, broken promises, roller coaster like direction : changes in policy, and so on, from Digital, Compaq and HP.  O ("Yes customer, your are very wise! VMS is indeed the best operating system in iP the world ! Oh No, we will not market it. Please keep those qualities a secret, O since using VMS is Politicaly Incorrect. And if you like to migrate to Windooz   or Unix, we will help you ! ")  O I still have that historic pdf file somewhere, in which all the virtues of the  O Alpha over the Itanic are listed. And of course the solumn promise of a Compaq nN Vice President that the development of the Alpha would go on, and on, and on, 6 for years to come. There even was a roadmap to EV12 !!  O Then suddenly someone at Compaq found a wonderfull new CPU: The Itanic !!! Now rQ that is a processor, only years behind on schedule but who cares. It will be The sQ Industry Standard CPU. The Alpha development wasn't doing so well anyway, and it  7 was far to expensive to go ahead with this development.e  K And we the customers were suppose to believe that, after all the marketing c/ peptalk on the Alpha just a few months earlier.N  # This is what I think that happened.hM The CEO's of HP and Compaq were talking about a merger, or better a takeover eP from Compaq by HP. However there was one big problem. Both companies had a line O of midrange computer systems that were totaly incompatible. Keeping both lines 1P would have been impossible. At that time HP was working on the successor of the Q PA Risc architecture, and they did that in cooperation with Intel. So they had a mK   choice, using the superior Alpha architecture, or going for the unproven eJ Itanic. HP UX was without any doubt the more succesfull (not better) Unix P variant, and it would have meant a great loss of face for HP to stop the Itanic N development, and to port HP UX (with Tru64 components) to Alpha. So the Alpha E had to be killed, even if it was a far better design then the Itanic.r  Q And so it went. Compaq announced that wonderfull sudden discovery of the Itanic, aM and guess what, months later HP and Compaq announced they were having merger sG talks !! And all the analists said: "Wow, what a nice idea. And what a 2O coincidence that both companies have choosen to use the Itanic in their future  Q server line. Yes, these companies have the same ideas on this matter, very good! D   Go on with the merger !"  O Maybe it is possible in the end to build a kind of Frankenstein CPU with Alpha nO and Itanic components that will have sufficient performance to build succesors r to the present Marvel systems.  P But the all important question is: Will It Be A Industry Standard CPU. That was P the intention, and there HP and Compaq wanted to save money on CPU's and design!  H If AMD proofs to be succesfull with their 64 bit X86 design (with Alpha M influence), I have no doubt Intel will be forced to produce a 64bit Pentium. nP Unless they are very stupid, they already will have developed one, and they are M just waiting to see if they need it. If so, then that will be the end of the kO Industry Standard label for the Itanic, and it will be a niche CPU just as any eN other high end CPU like the Alpha, PA-Risc and Sparc. It that case HP will be Q stuck with a rather mediocre CPU design, that is just as propriarty as any other sE high-end CPU. Not a very nice prospect for HP and its customers......    ------------------------------  % Date: Tue, 29 Apr 2003 22:32:54 -0400W* From: "Bill Todd" <billtodd@metrocast.net>F Subject: Re: How Alpha will save Itanium - must reading for Bill Todd!2 Message-ID: <CEednX-B9_NFqDKjXTWcoQ@metrocast.net>  + "Dirk Munk" <munk@home.nl> wrote in message , news:b8n11j$eq7$1@news1.tilbu1.nb.home.nl... > Bob Ceculski wrote:e4 > > here is how Alpha will save Itanium Bill ... you5 > > can click link and read, but I especially cut outn2 > > the paragraph for Bill how Alpha EV8-9 designs$ > > will fit nicely with Itanium ... > >v > >c, > > http://www.theinquirer.net/?article=9127 > >I > >i0 > I have read the article and this whole thread. >oI > No one can accuse Bill of being the most diplomatic contributor to thise groupo  / I am cut to the quick.  I am crushed.  Crushed!   J It could be possible that one or two individuals might harbor ill feelingsE over some real or imagined long-forgotten slight, but surely there is J *someone* here who can recognize the fruits of my many years at the CutlerF School of Tact and Diplomacy and would be willing to accuse me in thisL manner.  Or perhaps, if not willing to make a such blanket accusation, couldI bring themselves to accuse me of being the most diplomatic contributor of L those who are still utterly disgusted at cHumPaq's perdify and incompetence.J Or at the very least accuse me of being the most diplomatic contributor ofL those who employ pithy Anglo-Saxon four-letter and other constructs in theirJ attempts to characterize people and situations with some reasonable degree+ of accuracy (since I try, God knows I try).k  L For I can assure you that I am not one of those boorish individuals (many ofH whom can be found in and around 1600 Pennsylvania Avenue in our nation'sL capital these days) whose reaction to the word 'diplomacy' is "I've heard ofH it."  Aside from a long-standing fondness for the board game of the sameJ name, I hold diplomacy in the more generally-accepted sense in the highestG regard, and make earnest attempts to apply it to any situation where iteH appears likely to be anything but actively counter-productive.  The factH that this exception has happened to apply to many aspects of discussionsJ relating to the actions of VMS's owner for the past couple of years shouldI not be misconstrued as any more general aversion to its application:  thelL tenets of diplomacy are part and parcel of my very soul, and anyone who says< otherwise is (well, it would not be diplomatic to say what).   - bill   ------------------------------  # Date: Wed, 30 Apr 2003 01:41:38 GMT7. From: "Tom Simpson" <simpsont@attbi.com.fubar>- Subject: Re: I learned about VMS from that...n/ Message-ID: <mRFra.692380$S_4.736130@rwcrnsc53>s  C I was wondering if you had selected the REBOOT Shutdown option (noth auto-reboot) on your initial attempt to reboot?   Regards, Tomo  2 "Dean Woodward" <deanw@rdrop.com> wrote in message" news:3EA975C2.1060208@rdrop.com...H > There are some lessons here- maybe some of them are obvious, hopefully= > some of them will be helpful to someone else down the road.v >rF > I spent the week in Boston at a customer's site. The last task on myG > list was to do a clean boot (sidenote: last reboot was 220 days ago),-H > just as a sanity check to make sure all the upgraded software I'd beenH > working on comes up cleanly so there wouldn't be any hassles when theyI > brought the machines down to move them to a new server room next month.n >hF > Believe me when I say that when you're scheduled to leave to catch aI > plane in 30 minutes, the last thing you want to see on a booting system0 > console is a message like4 >A > kernel stack error- halting  >6J > repeated on every cluster node. As a sidenote,  in *nix-land, this errorI > would probably begin with 'panic:'- apparently, VMS system managers are4A > supposed to be above panic. I tried it for a bit, but it wasn'tt > helping... =8-Oh >wJ > My working hypothesis is that a disk block underneath some critical bootD > file went bad. Based on that, and with a clue provided by hardwareI > support, I booted off the install CD and "upgraded" the OS- that got me4J > off the ground; then it was a matter of downloading the critical patchesJ > and applying them. The systems immediately came up and have been purringI >   along since. I maybe could have restored the system disk from tape as-I > well, but the CD was handy, and I have no idea how long my hypotheticalo > file was corrupt...  >  > What I learned:- >7F > 0) The hitchhiker's guide was right- DON'T PANIC. OK, not much- justJ > enough to get a little adrenaline flowing and the blood pumping is about- > right- after that, settle down and *think*.: >:G > 1) Just because one's boss has been twiddling VMS for 15 years longersG > than I have doesn't mean he's right; when we built this cluster three$F > years ago, I should have stuck to my guns about mirroring the systemF > disk- he doesn't like to do that ("performance"), and I think that's > what bit me. >iE > 2) Make sure the onsite staff have access to their SOFTWARE supporthJ > info; my customer's didn't, unless the MIS manager does- and he's out on > vacation. Grrr...p >=I > 2a) Make sure they know where distribution media is. They didn't, and I H > spent a frantic 15 minutes digging through closets looking for the VMSE > kit. (Apparently, their ConDists for the past three years have beenlB > disappearing into the aether- they haven't gotten a single one.) > D > 3) Make sure that all your critical patches are on-site somewhere-H > preferrably, on the system(s) they're applied to. It's a right PITA toI > try to load a patch onto a system that needs that patch to run it's NICe > properly.d >aJ > 4) PATHWORKS may be able to transfer a file when FTP can't. I don't know0 > why this is true, but I'm thankful that it is. >SJ > 5) When possible, rolling reboots are a good way to do this- if I'd madeE > sure one system could come back up cleanly before taking the otherso. > down, I'd have saved myself a lot of stress. >r@ > 6) VMS rocks when it comes to "upgrade" installs- It was done,H > autogenned, and coming back to life while I was still on hold with the- > airline trying to reschedule my flight out.  > I > 7) The Big Dig is beginning to be useful- it's now possible to get from:I > South Boston to the airport in < 15 minutes, especially if you know howeJ > (and are willing) to drive like a Boston native. And if you're in BostonJ > and got cut off by a maroon rental car w/ PA plates yesterday afternoon, > sorry 'bout that. ;-)c >l   ------------------------------  % Date: Tue, 29 Apr 2003 17:51:34 -0000 % From: Pete Fenelon <pete@fenelon.com>tQ Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolye/ Message-ID: <vatet6i59490fc@corp.supernews.com>s  @ In alt.folklore.computers Lars Duening <lars@bearnip.com> wrote: > F > The different BSDs at least have unique central themes guiding theirJ > development; whereas the central theme of most Linux distributions seems > to be 'Me Too!'.  ) BSD's a cathedral. Linux is a bazaar.  ;)   ? Thinking of Cologne, Notre Dame, Reims, York Minster, either ofm> Liverpool's, St Paul's, and many others - I know what I prefer architecturally.... ;)   pete -- nE pete@fenelon.com "there's no room for enigmas in built-up areas" HMHBi   ------------------------------  % Date: Tue, 29 Apr 2003 14:05:36 -0400"* From: JF Mezei <jfmezei.spamnot@istop.com>Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyR) Message-ID: <3EAEBEED.914DA770@istop.com>r   Brian Inglis wrote:-B > Heavy weight processes are great for long lead time development,8 > especially when user death is a critical factor, as in? > military/aerospace/medical machinery control systems, but for9A > non-critical process support systems, usability is normally thea > critical success factor.  G One area where documentation should be required is billing systems. For-M example, the ISP which I am about to dump in 3 hours when the telco activates G the DSL line implemented downloads limits last september without giving N customers 30 day warning. When we called to get some precise information, theyJ could not give us any answeres because nobody know how the programmers had actually implemented it.  N (for instance, the fee is documented in gig of usage, but nobody could tell usN if it was billed by the byte, kilobyte, megabyte or gig. (if you use 1 gig + 1J byte, how much would they charge). Nobody could tell us the actual billingJ periods (it would take a few days for the usage to appear on the web pagesK showing your concumption, so if they produce a bill on the 15th of a month,/H what happens to the consumption of the 13 14 and 15 that hasn't yet been	 counted ?a  M So the feeling is that they let some programmers decide how to implement thisc4 billing and they told nobody how they really did it.  I I had similar issues with mobile company in early days where the customer M support people didn't have access to "real" billing policies, especially with 1 regards to roaming onto other company's networks.h   ------------------------------  % Date: Tue, 29 Apr 2003 18:12:46 -0000h2 From: "Helmut P. Einfalt" <hp.einfalt@t-online.de>Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyE/ Message-ID: <b8mfjj$35e$00$1@news.t-online.com>s  8 >> We have one moored right ... in the center of Bremen,  5 > yeah, they were US ships from the second world war.@  F Actually, the one I was talking about is the "Treue", German-built, ofE 1943 vintage, the last of a series of 70. Both the hull (44 m long, 9uE m wide) and the superstructures are made from concrete. She served asfA cargo ship until 1962, then she was used as a floating motorcyclep> repair shop until 1998. Since 2000, she has been moored at theA Schlachte quai in Bremen (the first harbor going back to the 15th 2 century) and is being used as a disco/event stage.  6 Some images befor the last outfitting are available at http://www.ms-treue.de/  (click "Umbau" and "Nutzung")      Helmut# Your reliable source of informationk that no one needs...   ------------------------------  + Date: Tue, 29 Apr 2003 19:33:41 +0000 (UTC)b8 From: hawk@slytherin.ds.psu.edu (Dr. Richard E. Hawkins)Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolye. Message-ID: <b8mk2l$1a24$1@f04n12.cac.psu.edu>  / In article <vatet6i59490fc@corp.supernews.com>,.' Pete Fenelon  <pete@fenelon.com> wrote:0A >In alt.folklore.computers Lars Duening <lars@bearnip.com> wrote:n  G >> The different BSDs at least have unique central themes guiding theirlK >> development; whereas the central theme of most Linux distributions seems  >> to be 'Me Too!'.o  * >BSD's a cathedral. Linux is a bazaar.  ;)  @ >Thinking of Cologne, Notre Dame, Reims, York Minster, either of? >Liverpool's, St Paul's, and many others - I know what I prefer  >architecturally.... ;)i   :)  8 Just imagine a large structure built from open source :)  H Now, if only vmware 4 would run on FreeBSD.  I have my license, but have3 to use version 2, as it's all that has been ported!t   hawk --  K Richard E. Hawkins, Asst. Prof. of Economics    /"\   ASCII ribbon campaignaG dochawk@psu.edu  Smeal 178  (814) 375-4700      \ /   against HTML mailuD These opinions will not be those of              X    and postings. 6 Penn State until it pays my retainer.           / \      ------------------------------  % Date: Tue, 29 Apr 2003 22:02:23 -0400n- From: "Alan T. Bowler" <atbowler@thinkage.ca>uQ Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyi+ Message-ID: <3EAF2EAF.2CEE81B5@thinkage.ca>    Bob Koehler wrote: > i > In article <b8l9ql$88r$1@freenet9.carleton.ca>, ab528@freenet.carleton.ca (Heinz W. Wiggeshoff) writes:i > >a9 > >   (Seems that late 60's into early 70's, cement boatsi? > >   were in vogue.  I've been out of that loop for 30 years.)o > E >    Concrete shipos they were in use during WWII, and concrete boatso7 >    are still built and raced by various universities.I  6 When I was a student was ordanized a canoe/kayak race.6 The civil engineers entered their concrete canoe.  It 5 would have been fine except that the river was reallyl: low that year and they ended up lugging it for much of the route.   ------------------------------  # Date: Wed, 30 Apr 2003 02:27:53 GMTn4 From: Brian Inglis <Brian.Inglis@SystematicSw.ab.ca>Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopoly08 Message-ID: <g4duavkbko93lkp1lc8u6q6qjkkn1qcn5r@4ax.com>  = On Tue, 29 Apr 2003 17:51:34 -0000 in alt.folklore.computers, & Pete Fenelon <pete@fenelon.com> wrote:  A >In alt.folklore.computers Lars Duening <lars@bearnip.com> wrote:o >> uG >> The different BSDs at least have unique central themes guiding their K >> development; whereas the central theme of most Linux distributions seems  >> to be 'Me Too!'.b > * >BSD's a cathedral. Linux is a bazaar.  ;) >c@ >Thinking of Cologne, Notre Dame, Reims, York Minster, either of? >Liverpool's, St Paul's, and many others - I know what I preferr >architecturally.... ;)h  7 How much are the dates, figs, and olives, there, then? !  9 Thanks. Take care, Brian Inglis 	Calgary, Alberta, Canada- -- -F Brian.Inglis@CSi.com 	(Brian dot Inglis at SystematicSw dot ab dot ca),     fake address		use address above to reply   ------------------------------  # Date: Tue, 29 Apr 2003 19:22:35 GMTb+ From: Anne & Lynn Wheeler <lynn@garlic.com> Y Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopoly monopola) Message-ID: <ud6j5gl2j.fsf@earthlink.net>e  6 Brian Inglis <Brian.Inglis@SystematicSw.ab.ca> writes:< > CMU SEI CMM level 5 organizations are also said to produce" > consistently unusable programs. ? > Most developers I've met are lazy: some creatively so, otherseA > just bone idle, so I'd expect the thought of having to fill outaC > (virtual) paper to implement a feature would ensure not much gets7$ > added to or changed in a program. B > Heavy weight processes are great for long lead time development,8 > especially when user death is a critical factor, as in? > military/aerospace/medical machinery control systems, but for A > non-critical process support systems, usability is normally thet > critical success factor.    A i've often commented about service level application for businessa@ critical operation are 4-10 times the effort of just writing theE straightline application code ... and possibly four or more times theeA number of lines of code. Part of the issue is the degree that thea< underlying infrastructure have support for business critical
 operation.  ? we did a one week JAD a couple years ago with one of the objectsD application development suites regarding the business critical issueD ... and identified a 30 percent hit to all their existing operationsC and an additional 30 percent for new feature/function in support oft business criticial operation.t  E in any event, thinking about the business critical issues can be morel: than ten times the effort of the straight line applicationF implementation ... and whether it is four times or ten times more codeC depends on to what degree the underlying infrastructure has supportr  for business critical operation.  C when we were doing the payment gateway stuff in support of what has , become to be called electronic commerce ... , http://www.garlic.com/~lynn/aadsm5.htm#asrn2, http://www.garlic.com/~lynn/aadsm5.htm#asrn3F we developed a failure-mode grid that was used after the straight-lineD stuff had all been done and extensively regression tested. The issueF is the nominal service operation had the facility of doing first level9 problem determination and diagnostic within five minutes.o  C Part of the issue was that some amount of the payment protocol froml? circuit based infrastructure was morphed into a internet packetqB orientation .... and all the implicit diagnostic stuff (not in theF higher level protocol) that was part of end-to-end circuit environment> was gleefully ignored. The other part involved people that hadC experience doing stand-alone, desktop, straightline application ...:E trying to translate that into a business critical service environmentcB ... and all of a sudden there are a whole set of issues which theyA didn't even know that they didn't know (aka who cares if customerSE calls in with a problem and first level support spends three hours ont9 it and closes the trouble ticket: NTF; no trouble found).c  B in some sense this is analogous to the enormous increase in bufferA overflow exploits because the way string handling was implemented @ (the buffer exploits weren't seen in platforms with other string handling paradigms).  C note that lot of SEI has to do with total life cycle costs. many ofpE the facilities that have been shown to practically address production A quality & business critial rapid development and life cycle costsf0 ... tend to have steep up-front learning curves.     random business critical refs:8 http://www.garlic.com/~lynn/96.html#27 Mainframes & Unix8 http://www.garlic.com/~lynn/96.html#31 Mainframes & Unix8 http://www.garlic.com/~lynn/96.html#32 Mainframes & Unix? http://www.garlic.com/~lynn/97.html#15 OSes commerical, historytZ http://www.garlic.com/~lynn/98.html#18 Reviving the OS/360 thread (Questions about OS/360)_ http://www.garlic.com/~lynn/98.html#51 Mainframes suck? (was Re: Possibly OT: Disney Computing)S http://www.garlic.com/~lynn/99.html#49 Language based exception handling. (Was: Did Intel pay UGS to kill Alpha port? Or Compaq simply doesn't care?)i http://www.garlic.com/~lynn/99.html#224 X9.59/AADS announcement at BAI this weekhttp://www.garlic.com/~lynn/2000e.html#46 Where are they now : Taligent and PinkJ http://www.garlic.com/~lynn/2001b.html#25 what is interrupt mask register?q http://www.garlic.com/~lynn/2001b.html#60 monterey's place in computing was: Kildall "flying" (was Re: First OS?)MI http://www.garlic.com/~lynn/2001c.html#16 database (or b-tree) page sizesMD http://www.garlic.com/~lynn/2001d.html#56 Pentium 4 Prefetch engine?- http://www.garlic.com/~lynn/2001n.html#11 OCOa> http://www.garlic.com/~lynn/2001n.html#85 The demise of compaq8 http://www.garlic.com/~lynn/2002.html#28 Buffer overflow8 http://www.garlic.com/~lynn/2002.html#32 Buffer overflowH http://www.garlic.com/~lynn/2002c.html#30 OS Workloads : Interactive etc_ http://www.garlic.com/~lynn/2002d.html#14 Mainframers: Take back the light (spotlight, that is)nF http://www.garlic.com/~lynn/2002f.html#23 Computers in Science FictionX http://www.garlic.com/~lynn/2002l.html#15 Large Banking is the only chance for Mainframe9 http://www.garlic.com/~lynn/2002p.html#6 unix permissionsh] http://www.garlic.com/~lynn/2003.html#38 Calculating expected reliability for designed system L http://www.garlic.com/~lynn/2003c.html#23 diffence between itanium and alpha   -- d3 Anne & Lynn Wheeler | http://www.garlic.com/~lynn/ ,A Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm-   ------------------------------  % Date: Tue, 29 Apr 2003 19:11:58 -0400i+ From: John Johnstone <jj_usenet2@yahoo.com> ; Subject: Improvements in TCP/IP Services anti-spam featuresa) Message-ID: <3EAECE7E.2EEEA526@yahoo.com>p  F There really should be a way to configure the handling of non-existentH addresses with HP's TCP/IP Services.  Mail to non-existent addresses hasF reached an intolerable level here.  Of all the email addresses that weC have, the ones that get the most spam are old addresses that are notG longer valid.  And it's incredibly ironic that the volume of mail beingeA sent to them is still on the increase.  There has even been a bigoE increase to one email address that has never existed.  The spam to itmE started coming from a few consistent IP addresses.  Since that's been C filtered, all of the mail to this address come from random IPs witht5 random return-paths that are completely unfilterable..  G Since 85% of our spam that isn't blocked has an invalid return address, ? that means all of those messages will be generating a bounce tohH Postmaster.  As Don Sykes noted, the side-effect of clogging up the mail3 queues with hopeless retries is also quite a drain.i  F It was great that they added anti-spam features to TCP/IP Services butD going just a little bit further would have a huge payoff.  How about* another field name such as Reject-Rcpt-To?  - Reject-Rcpt-To: old-email-address@domain-nameV  E With spam, looking for a signature to match on the mail message (i.e.RB Mail-From, source IP address, etc.) is always a moving target.  IfF specific non-existent addresses could be configured for a reject, thatG would block 100% of the spam sent to those addresses.  A reject messagePF Reject-Rcpt-To-Text could be tailored as desired if feedback for valid> mail was a concern.  In my case, if email is sent any of theseB particular old addresses, there's a 99.999% chance that it's spam.  F Since some spammers are thoughtful enough to put the recipient addressC in as a component of the return-path address, I've found that I canS block some spam with:t  1 Reject-Mail-From: *old-email-address*domain-name*n1 Reject-Mail-From: *domain-name*old-email-address*   E This regularly catches a small percentage of the spam sent to the old G email addresses.  You'd think that putting the recipient address in the F return-path address would indicate that the sender would be interestedE in knowing about invalid addresses.  That's clearly not the case withaG many spammers.  The bounce or reject does nothing to stop the flow sent.C from many sources even when the bounces are accepted by the sender.t   ------------------------------  # Date: Tue, 29 Apr 2003 23:38:44 GMTp' From: Don Sykes <anonymous@pacbell.net>e? Subject: Re: Improvements in TCP/IP Services anti-spam featuresh+ Message-ID: <3EAF0CFE.E3B64CF2@pacbell.net>    John Johnstone wrote:L > H > There really should be a way to configure the handling of non-existentJ > addresses with HP's TCP/IP Services.  Mail to non-existent addresses hasH > reached an intolerable level here.  Of all the email addresses that weE > have, the ones that get the most spam are old addresses that are noMI > longer valid.  And it's incredibly ironic that the volume of mail beingFC > sent to them is still on the increase.  There has even been a bigD7 > increase to one email address that has never existed."  H I've noticed the same thing. I bet someone with a big pipe (so to speak)B got a hold of one of those old 100M email address lists and is now& offering "spam services" to customers.D The way things are going, I expect spam to account for 99% of all IP' communications by the end of this year!      > The spam to itG > started coming from a few consistent IP addresses.  Since that's been E > filtered, all of the mail to this address come from random IPs with 7 > random return-paths that are completely unfilterable._  D I really don't want to throw the baby out with the bath water, but I) took Michael Austin's suggestion to add :R< Bad-Clients: 210.0.0.0/8,211.0.0.0/8,202.0.0.0/8,203.0.0.0/8F as he says, "Add these as well, these are entire Asia/Pacific! (China) entire networks.""  E I think it helped a little, but I could be loosing business from someV? good folks in Asia, so I'm keeping as a temporary measure only.V   > I > Since 85% of our spam that isn't blocked has an invalid return address,LA > that means all of those messages will be generating a bounce totJ > Postmaster.  As Don Sykes noted, the side-effect of clogging up the mail5 > queues with hopeless retries is also quite a drain.- > H > It was great that they added anti-spam features to TCP/IP Services butF > going just a little bit further would have a huge payoff.  How about, > another field name such as Reject-Rcpt-To? > / > Reject-Rcpt-To: old-email-address@domain-names > G > With spam, looking for a signature to match on the mail message (i.e.QD > Mail-From, source IP address, etc.) is always a moving target.  IfH > specific non-existent addresses could be configured for a reject, thatI > would block 100% of the spam sent to those addresses.  A reject message:H > Reject-Rcpt-To-Text could be tailored as desired if feedback for valid@ > mail was a concern.  In my case, if email is sent any of theseD > particular old addresses, there's a 99.999% chance that it's spam. >  <snip>  E Excellent idea! I think not too difficult to include either. It would F solve 99% of my problem too. I wonder whay it wasn't included already? (HP response???)   -- e   Have VMS, Will Travelr Wire paladin, San Franciscoi   (paladinATalphaseDOTcom)   ------------------------------  % Date: Tue, 29 Apr 2003 19:50:27 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>? Subject: Re: Improvements in TCP/IP Services anti-spam featurese) Message-ID: <3EAF0FBF.2BF26A4E@istop.com>s   John Johnstone wrote:fI > longer valid.  And it's incredibly ironic that the volume of mail being ) > sent to them is still on the increase. h  I Have you considered using an RBL service ? TCPIP 5.3's SMTP is capable of L interfacing to an RBL. This may give you more intelligent filtering based on( the IP address of the sending SMTP host.   ------------------------------  % Date: Tue, 29 Apr 2003 21:09:57 -0400f+ From: John Johnstone <jj_usenet2@yahoo.com>.? Subject: Re: Improvements in TCP/IP Services anti-spam features ) Message-ID: <3EAEEA25.3A94D3BD@yahoo.com>h   JF Mezei wrote:s >  > John Johnstone wrote:hK > > longer valid.  And it's incredibly ironic that the volume of mail beingr* > > sent to them is still on the increase. > K > Have you considered using an RBL service ? TCPIP 5.3's SMTP is capable ofsN > interfacing to an RBL. This may give you more intelligent filtering based on* > the IP address of the sending SMTP host.  F We were using RBLs but stopped due to upper management's concern aboutF the possibility of blocking legitimate mail.  If it was just up to me, I'd take that risk.r  C That reminds me of another problem.  The reject error response fromiG TCP/IP Services for an RBL match doesn't have the capability to include F the name of the block list that matched in it.  I can understand maybeG not wanting to provide that in the reject response that goes out to the G sending system, but it also isn't in the OPCOM message.  That is reallyr= critical in order to understand how your blocking is working.L   ------------------------------  % Date: Tue, 29 Apr 2003 13:54:01 -0400I* From: "Bill Todd" <billtodd@metrocast.net>A Subject: Re: Itanium Madison blasts Sun, IBM in encryption specs!t2 Message-ID: <o2OdndbTiqmhITOjXTWcqg@metrocast.net>  D "gregc at gregcagle.com" <"gregc at gregcagle.com"> wrote in message) news:vas7355o713e2c@corp.supernews.com...h > Bill Todd wrote: > >CI > > I'm curious as well.  Not all that long ago people were talking aboutiK > > entry-level single-processor Itanic2 development systems from HP in theoI > > $4500 range, with IIRC dual-processor systems starting at about $10K.z ButtG > > as best I can determine prices are *far* higher than that (though Ie admitsI > > that my success in navigating HP's pricing information on the Web hasa been > > spotty at best). >c/ >  From http://www.hp.com/workstations/itanium:o >m) > zx2000 (single CPU) 512M Linux = $3298.C< > zx6000 (dual CPU) 512M Linux single CPU installed = $4896. >8 > Example configs at:- >-H > http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1427 >c > andn >eH > http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1431  I Hmmm.  The last two citations above suggest that to get an actual running J system you need to spend a bit over $4K for a single-CPU (low-end Itanic2)F box (no monitor) and over $7K for a dual-CPU (low-end Itanic2) box (noK monitor).  So I'm wondering whether the first citation refers to bare-bonese& pricing (no memory, no disk, no ...?).  A Nonetheless, these prices are much more like the ones I thought IeJ remembered.  Perhaps it's just *server* pricing that's much higher (rx vs.G zx series?).  Unfortunately, the Opteron systems being compared againstmL *are* server systems (we'll have to wait until September to see how Athlon64 systems are priced).   - bill   ------------------------------  % Date: Tue, 29 Apr 2003 11:01:53 -0700a9 From: "gregc at gregcagle.com" <"gregc at gregcagle.com">-A Subject: Re: Itanium Madison blasts Sun, IBM in encryption specs! / Message-ID: <vatfh0epnhauc9@corp.supernews.com>w   Bill Todd wrote:F > "gregc at gregcagle.com" <"gregc at gregcagle.com"> wrote in message+ > news:vas7355o713e2c@corp.supernews.com...    <snip>  / >> From http://www.hp.com/workstations/itanium:a >>) >>zx2000 (single CPU) 512M Linux = $3298.o< >>zx6000 (dual CPU) 512M Linux single CPU installed = $4896. >> >>Example configs at:t >>H >>http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1427 >> >>ando >>H >>http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1431 >  > K > Hmmm.  The last two citations above suggest that to get an actual running L > system you need to spend a bit over $4K for a single-CPU (low-end Itanic2)H > box (no monitor) and over $7K for a dual-CPU (low-end Itanic2) box (noM > monitor).  So I'm wondering whether the first citation refers to bare-bonesi( > pricing (no memory, no disk, no ...?).  H No, the first citation is 512M RAM, 36G disk, and low end graphics card.G If you scroll down within the pages referred to in the second citationsW" you can find those configurations.  C > Nonetheless, these prices are much more like the ones I thought I L > remembered.  Perhaps it's just *server* pricing that's much higher (rx vs.I > zx series?).  Unfortunately, the Opteron systems being compared againstdN > *are* server systems (we'll have to wait until September to see how Athlon64 > systems are priced).  C Yes - HP's "server" pricing is much higher. The zx6000 makes a more E cost-efficient computational engine in clusters if you can do withouts the "server" features.   -- p
 Greg Cagle gregc at gregcagle dot com   ------------------------------  % Date: Tue, 29 Apr 2003 14:12:20 -0400n* From: "Bill Todd" <billtodd@metrocast.net>A Subject: Re: Itanium Madison blasts Sun, IBM in encryption specs!t2 Message-ID: <p5ucnc_2fdIaXTOjXTWcqg@metrocast.net>  D "gregc at gregcagle.com" <"gregc at gregcagle.com"> wrote in message) news:vatfh0epnhauc9@corp.supernews.com...w > Bill Todd wrote:H > > "gregc at gregcagle.com" <"gregc at gregcagle.com"> wrote in message- > > news:vas7355o713e2c@corp.supernews.com...b >  > <snip> >t1 > >> From http://www.hp.com/workstations/itanium:  > >>+ > >>zx2000 (single CPU) 512M Linux = $3298.i> > >>zx6000 (dual CPU) 512M Linux single CPU installed = $4896. > >> > >>Example configs at:h > >>J > >>http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1427 > >> > >>and, > >>J > >>http://www.smb.compaq.com/ctoBases.asp?ProductLineId=433&FamilyId=1431 > >a > > E > > Hmmm.  The last two citations above suggest that to get an actuali runningnE > > system you need to spend a bit over $4K for a single-CPU (low-endp Itanic2)J > > box (no monitor) and over $7K for a dual-CPU (low-end Itanic2) box (noD > > monitor).  So I'm wondering whether the first citation refers to
 bare-bones* > > pricing (no memory, no disk, no ...?). >2J > No, the first citation is 512M RAM, 36G disk, and low end graphics card.I > If you scroll down within the pages referred to in the second citationsr$ > you can find those configurations.  E Mea culpa - I *did* scroll down, but in the first instance missed thePI lower-priced systems and in the second missed your original qualification J that the price was for a dual-CPU-capable box populated with only a singleG CPU (so technically my assertion above about the lowest dual-CPU systemo= price was correct, but it did not serve to contradict yours).s  I I should know better than to try to respond in detail when I'm rushed forh time...e   - bill   ------------------------------  # Date: Tue, 29 Apr 2003 23:09:04 GMTe& From: Rick Jones <foo@bar.baz.invalid>A Subject: Re: Itanium Madison blasts Sun, IBM in encryption specs!., Message-ID: <kCDra.20$3F.9@news.cpqcorp.net>  P Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote:I >> Further, HP has not (to my knowledge) announced prices for the 1.5 GHz  >> systems.r# > 1.5 MB cache not 1.5 GHZ systems.s  B That comment was refering to the systems with 1.5 GHz Itanium 2 6M, CPUs (aka Madison), not the older McKinleys.  8 > And of course its a very horizontally scalable service  B Indeed, web services can be quite horizontally scalable.  However,E there have been no clustered SPECweb99_SSL results published to date.9D That would seem to suggest that either the workload characterized byB the benchmark is either not horizontally scalable, or isn't easilyA horizontally scalable. (SPECweb99_SSL does not prohibit clustered  results)  F > box that costs 34K with 2 CPU's hardly makes much sense when you can9 > buy a 5 x 2 way V210's with cryptos for the same price.   B Do you mean to imply that it would take 5 of the two-way V210's toA match the SPECweb99_SSL performance of a single two-CPU non-SPARCc systems?  E > The V210/V240 are new low cost low end 1/2U 1 GHz UltraIIIi servershD > BTW, similar capacity to a Sun V280 but well under half the price.  D Sparring with you in Usenet is always fun because you are so fast onA your keyboard :) You were saying how much more expensive than therC Opterons the HP IPF systems were.  I then said SPARC systems aren'tdA that much better compared to the Opterons (in otherwords, the pote@ calling the kettle).  You've come-back instead with a comparisonB between the Sun systems and the IPF boxes instead of the Opterons.  @ So, lets then look at those V240s you brought-up (there being noD SPECweb99_SSL figures published for the V210, and I doubt you reallyB meant to imply that it would take 5 hardware-accelerated V210's toF match a two-CPU non-SPARC system at SPECweb99_SSL) and compare them to those Opterons.  a  I The SPECweb99_SSL result for a two-CPU V240 was 833 [1]. So, to match theuF two-CPU Opteron results one would need two of those (actually, two andD a fraction - 1783/833 = 2.14) and to match the four-CPU result would  require 4 (actually 4.2 - ie 5).  D Base price per Sun's online pricing pages for the V240 appears to beF $6,495.  The SCA 500 adds another $695, so box cost - without the fullB 8GB of RAM - is $7190.  Two of those to keep pace with the two-CPUD Opteron would be $14380, IIRC, you have asserted previously (openingD this pricing can) that the two-CPU Opteron systems (unspecified RAM)F were going for 5K?  Four, dual-CPU V240s (should that really be five?)D to keep-up with the single, four-CPU Opteron system would be $28760.  F Bill has taken the TPC-C pricing from the RackSaver and concluded thatB the Opteron SPECweb99_SSL config would be around $49000.  I do notF know if that price is truly "real" or not (seems everything is bundledF into one price, no per-line item pricing, so it is hard to check), butD for now lets go with that and then add the 8GB of RAM to those V240sF since that $49000 figure includes the RAM.  That is it seems $1795 forE 2, 1GB DIMMs, so 8GB of RAM would be $7180.  I would trust that therepF is a return credit for the four 512MB boards included in the base V240C config, so credit back that price (795x2) and so the add-on to eachAB system is 7180 - 1590 or $5590.  So, each box is then 7190+5590 orC 12,780, four of those would then be $51,120, five would be $63,900.o  B Heck, if we really want to have fun, then we compare the number ofF Gigabit ports used by the V240-based solution - 8 to 10 - to that usedB for the single-system four-CPU systems - 4.  The V240's are 2U, soE four of them is 8U of rack space to the four-CPU system's 4U. Etc etc  etc...  @ >> http://store.sun.com/catalog/doc/BrowsePage.jhtml?catid=55844 >> hD >> Of course, pricing is always a fun game isn't it - what is in theF >> base config, how much to get that to the tested config, how much ofE >> the tested config was required to get the result. For example, waslC >> the entire 12x36GB StorEdge 3310 SCSI array and dual Ultra3 SCSIsE >> HBA required required to hit that 1008 number, what is the pricinggF >> of ZWS compared to Sun ONE Web Server (although it was probably theB >> SNCA doing all the real work as an in-kernel accelerator rather> >> than the web server...), how much does that SCA 1000 or the( >> GigaSwifts cost.  All that fun stuff.  D > What has this got to do with the pricing of the server itself. YouE > seem to have dissapeared off on an alarming Kerry Main type tangenth1 > which is very unlike your normal posting style.h  7 I'm afraid I don't understand the Kerry Main reference.d  F If we really want to get into the ever so fun and twisty discussion ofD pricing which (iric) you brought-up, and if the _solution_ put-forthC by a vendor to achieve a given SPECweb99_SSL result really did neednA all those things from the SPECweb99_SSL disclosure (and there areoC indeed times when a SPECweb99_SSL result had stuff it didn't reallyoD need but just happened to be there) then comparing pricing should beF comparing _solution_ pricing not just box pricing.  And after that, weE could go futher into the pricing hole and start arguing about whetherhC it costs more to admin four, two-CPU V240s or one, four-CPU system, C support prices etc etc etc...  Probably one of the reasons why SPECn< has never been especially keen on that maze of twisty little passages...o  
 rick jones  B [1] SPECweb99_SSL figures per http://www.spec.org/ as of April 26,3 2003, SPECweb being a trademark of SPEC etc etc etce   -- hH Wisdom Teeth are impacted, people are affected by the effects of events.F these opinions are mine, all mine; HP might not want them anyway... :)A feel free to post, OR email to raj in cup.hp.com  but NOT BOTH...a   ------------------------------  % Date: Tue, 29 Apr 2003 13:56:30 -0400e* From: JF Mezei <jfmezei.spamnot@istop.com>) Subject: Re: Johnny English is a VMS user ) Message-ID: <3EAEBCCC.37622524@istop.com>    Phillip Helbig wrote:aG > I think there is an episode of Battlestar Galactica which has a VT100T > terminal.   J The NCC1701-A technical reference manual CD shows a VT100 display inside aN shuttle's aft area. Probably meant to look like some high tech sensor display,N but to me, it is just a character cell display that has survived centuries andL coexists with transporters, warp drives etc. An ode to the legacy of Digital :-) :-) :-)t   ------------------------------  % Date: Tue, 29 Apr 2003 15:10:54 -0700o# From: "Tom Linden" <tom@kednos.com>t Subject: Keyboard ?t9 Message-ID: <CIEJLCMNHNNDLLOOGNJIOEHDHBAA.tom@kednos.com>n  E Have an LK401 KB attached to a 3000/300LX running 6.2.  I, er, spilltiK some liquid on the KB so I quickly unplug to clean it.  When I plug it backiD it back in no response.  I am sure the KB is OK.  The LEDs are dark.L Do I have to power down or is there some way to get this up?  The attachmentI to the box is actually through one of those funny cables with DB15 at one  end andr5 a little box where both the KB and the mouse plug in.s     ---e& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.476 / Virus Database: 273 - Release Date: 4/24/2003   ------------------------------  % Date: Tue, 29 Apr 2003 21:17:02 -0400g2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: Keyboard ?sL Message-ID: <rdeininger-2904032117030001@user-105n8do.dialup.mindspring.com>  F In article <CIEJLCMNHNNDLLOOGNJIOEHDHBAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> wrote:r  F >Have an LK401 KB attached to a 3000/300LX running 6.2.  I, er, spilltL >some liquid on the KB so I quickly unplug to clean it.  When I plug it backE >it back in no response.  I am sure the KB is OK.  The LEDs are dark.zM >Do I have to power down or is there some way to get this up?  The attachmentiJ >to the box is actually through one of those funny cables with DB15 at one >end and6 >a little box where both the KB and the mouse plug in.   Been there, done that.  G I'm not sure the KB is hot-plugable on that system.  If it doesn't justr; work, I think you'll need to cycle the power on the system.t   ------------------------------  + Date: Tue, 29 Apr 2003 19:33:16 +0000 (UTC)o3 From: "Richard Maher" <maher_rj@hotspamnotmail.com> P Subject: Moving on / Leatning to let go. . .(Was: Re: Why did Digital go broke?)/ Message-ID: <b8mk1q$p29$1@titan.btinternet.com>d  	 Hi Kevin,t  ; > There's currently a 32 byte limit in the private messagesn= > that are passed by DTI in order to retrieve the participante5 > information. So increasing the 32 byte limit to 256s4 > limit would mean a bumping of the message version. >i6 > As a result, you then get into the whole question of< > testing in mixed version clusters. Yes, these are messages3 > that can go between instances of TPSERV processesN > around the cluster.  >u> > In most cases the messaging is restricted to the local node,
 > but.....  H I was thinking more along the lines of a whole new message rather than aH new version and layout for the existing message. And, to take your mixedF version cluster scenario, if a node was on an earlier version that hadI never heard of V2 messages then it would (hopefully) just reply somethinge like ss$_unsupported?i  H As I suggested to Jim here a while back, a new search item code like dtiJ $_search_verbose could tell sys$getdti to permit an output item list other" than dti$_transaction_information.  B Anyway, I am the first to acknowledge that this new and additionalH functionality falls within the realm of an enhancement request. So let's* leave that to one side for the moment. . .  = > I would suggest that I try and help you by lobbying for then< > GETDTI functionality to be fixed. Getting back 32 bytes of= > data liked you asked for would go a long way to helping youw > with your problem.  ; That would be outstanding! Tell me though, you seem to be a J moderate/temperate sort of a bloke not given to exaggeration or the makingK of outlandish claims (and there's not many of us left :-) but how would youp( rate the chances of success on this one?  I I want to go with my existing strategy, and desperately need this bug fixeI for that, but I can also get by without it if I have to. There are also arD couple of benefits to maintaining my own journal. 1) I don't have toK translate any DNS names till QUERY time, which maybe useful. 2) I'm waitingdH to have the format of the Tandem NSK Transaction Identifier confirmed toG me, but it looks to have a few additional bytes and may not compress as:' well into 32 bytes as the OleTx Id did.d  K I'll keep both versions of my code in synch with any new functionality (I'm-E coding t3$pull and looking at cluster recovery at the moment) but I'd1I obviously like to drop one as soon as possible. All things going well. donH you have any idea of a time frame for a new version of dti$share? Weeks? Months? A thousand cuts?   > It looks like a bug to mea  K Vindication at last! This, by itself, almost makes the exercise worthwhile.a  I (Please forgive my welling-up. I'll try to compose myself and soldier on.c! It's just all been so difficult.)   5 > If I can help get you a fix for DTI will that help?t   Most definitely!  H If it looks like your gonna get a budget then there are also a couple ofG bugs with $trans_event's status reporting that you may want to look at.o8 (And are probably already aware of after The XA Gateway)  @ It looks like you're UK based? Do you get anywhere near RichmondI or the City? Regardless of what happens with this I'd really like to talki6 to you about the XA Veneer stuff. Perhaps over a beer?  H Anyway, no one else seems interested so I'll get in touch directly a bit@ later on. (I'm busy trying to secure my cash flow at the moment)  0 Anything you can do to help is much appreciated!   Cheers Richard  H Kevin Playford <kevin.playford@nospam.software-exploration.com> wrote in6 message news:b88mc7$b9f$1$830fa7b3@news.demon.co.uk...F > > But please tell me can I, like LMCP, just access DECdtm's log file
 > directlyK > > in Read-Only mode? I'd also use $getdti and $setdti for wildcarding andaK > > updates, but use your Log Read routine to retrieve the full participantn	 > > name.' >cK > I would recommend against that. If nothing else you would need to get theo > full descriptionK > of the log format, and given it is private to DECdtm I would suggest thatt > was not the route  > to go. >IC > > Yes that would be ideal and, as you're aware, 256 was the limit 	 specifiedn > onL > > $declare_rm and $join_rm in the old Functional Spec before it was cut to > 32I > > bytes by "RB" from Software-Exploration. (Although I've only ever got- LMCP > > to return under 132 bytes?)3 >pJ > The problem that I see with wanting to extend the amount of data from 32
 > bytes toJ > 256 bytes is that you wanting either a functional change or a functional
 > enhancementk5 > to the way in which the data is currently returned.2 > K > There's currently a 32 byte limit in the private messages that are passedr by > DTI inJ > order to retrieve the participant information. So increasing the 32 byte > limit to 256J > limit would mean a bumping of the message version. As a result, you then gett > into the wholeH > question of testing in mixed version clusters. Yes, these are messages that > can go between3 > instances of TPSERV processes around the cluster.s >,G > In most cases the messaging is restricted to the local node, but.....e > E > > Given that $getdti was the only game in town for getting what wash written? > toD > > the log file back out, I tailored my storage requirements to the 32-bytesD > > that dti$_transaction_information was able to return. Only to be
 > white-antedsC > > by the *bug* in $getdti that results in the dti$t_part_name andpL > > dti$b_part_name_len fields in the output ITMLST argument being populatedJ > > *not* with the data that has just been retrieved from the log file but= > > instead with a straight copy of their counterparts in thesF > > dti$_search_resolved_state structure of the input SEARCH argument. >iD > I would suggest that I try and help you by lobbying for the GETDTI > functionality toH > be fixed. Getting back 32 bytes of data liked you asked for would go a long > way to  > helping you with your problem. > I > > I then asked for a favour from senior VMS people, and even senior Rdbn > peopleD > > asked for this bug to be fixed, but all we were greeted with wasJ > > intransigence, recalcitrance and outright snootiness from the mothballL > > moguls. (You must have read their now infamous line about it not being a > buge% > > because you did it deliberately?)c >s$ > I think that's going a bit far.... >aK > > I won't re-state all of my reasons why the current behaviour of $getdtie isL > > crap (unless you missed them the first time round) but seeing as how youI > > were probably there when the designs were being engraved in stone ande thetI > > code was being cut, others may prefer to hear your version. Is this ar bug. > or8 > > just something spawned from a hallucinogenic frenzy? >s8 > It looks like a bug to me and not a hallugenic frenzy. >cI > > I have trouble reading BLISS but can you please look at lines 2354 too 2381E > > in the DTIC_SERVICES.LIS file. Does it make sense to you? "! Copyi resourceJ > > manager name from transaction record." That doesn't make sense the .TR isK > > the copy of the input item list isn't it? "! Copy resource manager namem > fromI > > DTIB" So which one is it? The Transaction Record or the DTRIB? Or didt theyC > > mean Commit Domain? I see evidence of confusion. How 'bout you?p >tD > First rule when reading anybody elses code, read the code, NOT the	 comments.oL > If code is ever fixed the majority of people go and fix the code but leave > the comments@ > untouched. How many times have you done that yourself Richard? >rC > If you look carefully at the code lines you mention there are two 	 different  > pieces ofsJ > data being copied. The first is the resource manager name, the second is > commit domaine: > name even though the comment says resource manager name. > K > > Anyway I don't know who's running the asylum in VMS engineering anymoren > andcI > > I don't care! I can work around this problem with my own journal filei andi > anG > > extra four I/O per transaction. No it's not ideal but I don't thinkc we'lle > > ever see ideal again :-( >.5 > If I can help get you a fix for DTI will that help?i >s) > > PS. Has Jim really gone to Microsoft?o >.' > I'll let Jim answer that one himself.e >c	 > regardsv > Kevin Playford > Software Exploration Ltd.  >  >h   ------------------------------  % Date: Tue, 29 Apr 2003 20:22:43 -0500h1 From: "David J. Dachtera" <djesys.nospam@fsi.net>i/ Subject: Re: Not entirely OT: RSHELL to Solarisc' Message-ID: <3EAF2563.9F34427E@fsi.net>   ( Andrew Harrison SUNUK Consultancy wrote:- > [Edited for grammar and sentence structure]p. > What kind of data is it[? Is it] in a file ?  G It's list of tape volumes by owner. The STK VAR provided a shell scriptaD which in turn invokes ACSLS's report writer to produce the list. The data is written to stdout.   -- s David J. Dachteraw dba DJE Systemsa http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/a   ------------------------------  % Date: Tue, 29 Apr 2003 20:24:47 -0500p1 From: "David J. Dachtera" <djesys.nospam@fsi.net>x/ Subject: Re: Not entirely OT: RSHELL to Solarise' Message-ID: <3EAF25DF.C6FD5CAB@fsi.net>h   David Beatty wrote:x > 9 > On Mon, 28 Apr 2003 21:09:41 -0500, "David J. Dachtera"r  > <djesys.nospam@fsi.net> wrote: > K > >Okay, now that we have our STK L700E libraries running, our ACSLS serveroJ > >talking to DCSC and thence to SLS, I need to get data from ACSLS, which > >runs on Solaris.o > >n > >My question is: > >eJ > >Is there some way to eliminate the need to have passwords in clear textH > >on RSHELL commands in our DCL procedures? Is there some kind of proxyJ > >and/or trusted host relationship I can set up on Solaris so that I need- > >only pass a username on my RSHELL command?g > >. > >Info:
 > >VMS V7.3-1h > >Multinet V4.4 Rev A-X > >Solaris/8 (Sparc) > >ACSLS V6.0.1. > E >     How about an entry in the file ~UNAME/.rhosts on the Unix side,O* > where UNAME is the username, for example >  > vmssys.vms.comp.com vmsuserM > < > This seems to work fine going from VMS V7.2-2 using either, > TCP/IP V5.1 or Multinet V4.3 to Solaris 8. > E >     If the username is the same on both systems, you don't need thep > /username qualifier.  E I'll have to look for "man rhosts" in the morning, but do you know iffF the username parameter in there can be a wildcard? ...or do I just not enter it to allow any user?r   -- / David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/F   ------------------------------  % Date: Tue, 29 Apr 2003 20:28:32 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>0/ Subject: Re: Not entirely OT: RSHELL to Solariso' Message-ID: <3EAF26C0.CFF5697D@fsi.net>    mckinneyj@cpva.saic.com wrote: > [snip]F >         http://vms.process.com/scripts/fileserv/fileserv.com?PRIVDCL > A >         This it does not address plain text over the net but iti7 >         doesn't appear as though that's your concern.-  H Well, I'm trying not to over-complicate this too much. Yes, I would likeA to avoid sending clear-text passwords (healthcare / HIPPA issues,TE y'know), but I'd also like to keep it clean and relatively simple. MyoE partner needs to be able to understand what I've done so we can cover-H for each other. He's good, really good, but I'm more of a DCL coder than he is.   -- e David J. Dachterav dba DJE Systemst http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/g   ------------------------------  % Date: Tue, 29 Apr 2003 21:37:14 -0500t1 From: "David J. Dachtera" <djesys.nospam@fsi.net>e/ Subject: Re: Not entirely OT: RSHELL to Solarisn' Message-ID: <3EAF36DA.89AEEA26@fsi.net>o  
 Milton wrote:o > 9 > On Mon, 28 Apr 2003 21:09:41 -0500, "David J. Dachtera"e  > <djesys.nospam@fsi.net> wrote: > K > >Okay, now that we have our STK L700E libraries running, our ACSLS server5J > >talking to DCSC and thence to SLS, I need to get data from ACSLS, which > >runs on Solaris.s > >  > >My question is: > >hJ > >Is there some way to eliminate the need to have passwords in clear textH > >on RSHELL commands in our DCL procedures? Is there some kind of proxyJ > >and/or trusted host relationship I can set up on Solaris so that I need- > >only pass a username on my RSHELL command?  > >e > >Info:
 > >VMS V7.3-1c > >Multinet V4.4 Rev A-X > >Solaris/8 (Sparc) > >ACSLS V6.0.1i > ; > I believe you need to create a mcf file on the sun servere
 > with the > access=user_id > hostname= host_nameoB > with that being the user_id and server that is requesting access > E > Read  the Sun SAM-FS and Sun SAM-QFS Storage and Archive Management  > Guide 5 > http://docs-pdf.sun.com/816-2544-10/816-2544-10.pdfr > or4 > ftp://docs-pdf.sun.com/816-2544-10/816-2544-10.pdf  H I'll take a look at those at work tomorrow. I read this group via google during the day.a   > According to this document:u4 > http://h18000.www1.hp.com/info/SP3359/SP3359PF.PDF > ? > The DCSC kit includes a Configuration File Editor to create aeH > configuration file used to describe the transport and ACS environ-mentH > to DCSC. DCSC reads this file to determine the operation, network, and! > transport resource allocations.   5 I have become quite familiar with that facility, yes.r  D > If I understand this correctlly that is what is needed to create a > connection tween VMS and Sun.i   Yes, that is quite correct.   F The issue now is getting data from ACSLS to make up for short-falls inH SLS and DCSC, and issues that neither DCSC nor SLS were ever intended to address.   --   David J. Dachtera- dba DJE Systems- http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/I   ------------------------------  # Date: Wed, 30 Apr 2003 02:11:28 GMT.% From: Milton <mbhewitt@optonline.net>e/ Subject: Re: Not entirely OT: RSHELL to Solariso8 Message-ID: <ulauavssv4dut054vf75nro3vlvj352vc4@4ax.com>  7 On Mon, 28 Apr 2003 21:09:41 -0500, "David J. Dachtera"e <djesys.nospam@fsi.net> wrote:  I >Okay, now that we have our STK L700E libraries running, our ACSLS server H >talking to DCSC and thence to SLS, I need to get data from ACSLS, which >runs on Solaris.f >i >My question is: >hH >Is there some way to eliminate the need to have passwords in clear textF >on RSHELL commands in our DCL procedures? Is there some kind of proxyH >and/or trusted host relationship I can set up on Solaris so that I need+ >only pass a username on my RSHELL command?i >" >Info: >VMS V7.3-1t >Multinet V4.4 Rev A-X >Solaris/8 (Sparc)
 >ACSLS V6.0.1f  9 I believe you need to create a mcf file on the sun server 	 with the m access=user_id hostname= host_nameoA with that being the user_id and server that is requesting access m  C Read  the Sun SAM-FS and Sun SAM-QFS Storage and Archive Managementh Guidec3 http://docs-pdf.sun.com/816-2544-10/816-2544-10.pdfr or e2 ftp://docs-pdf.sun.com/816-2544-10/816-2544-10.pdf   According to this document:a2 http://h18000.www1.hp.com/info/SP3359/SP3359PF.PDF  = The DCSC kit includes a Configuration File Editor to create aiF configuration file used to describe the transport and ACS environ-mentF to DCSC. DCSC reads this file to determine the operation, network, and transport resource allocations.   B If I understand this correctlly that is what is needed to create a connection tween VMS and Sun.s   Cheers,w Milton   ------------------------------  % Date: Tue, 29 Apr 2003 14:00:33 -0400o* From: "Bill Todd" <billtodd@metrocast.net>? Subject: Re: OpenVMS Itanium port progressing well says Gorham!a2 Message-ID: <eHydne7ePelYIDOjXTWcqA@metrocast.net>  7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messagew# news:3EAE2947.79EEEC63@istop.com...R   ...c  J > However, in fairness, in the early 1990s, the 8086 wasn't yet capable ofJ > taking on VMS. It wasn't until 80386 that it had enough modes etc etc toI > become serious enough for a VMS port. But by then, Alpha was a reality.t  F Not in *our* reality:  the 80386 appeared in 1986 (unless my memory is  faulty and it appeared in 1985).   - bill   ------------------------------  % Date: Tue, 29 Apr 2003 20:17:36 -0500f1 From: "David J. Dachtera" <djesys.nospam@fsi.net>i? Subject: Re: OpenVMS Itanium port progressing well says Gorham!9' Message-ID: <3EAF2430.84EB2FEC@fsi.net>    David Froble wrote:  >  > Dave Weatherall wrote: > J > > On Sun, 27 Apr 2003 06:52:44 UTC, "Bill Todd" <billtodd@metrocast.net>
 > > wrote: > K > >>If Compaq had kept the Alpha team and actually backed Alpha, that never N > >>would have happened:  Intel would have been stuck with a power-inefficientK > >>architecture that it had no idea how to push forward and that therefore:N > >>would never have looked attractive compared with Alpha (though AMD64 stillJ > >>might, for the low end), and EV8 would be coming next year to solidifyL > >>Alpha's high-end leadership.  The current situation with the engineering/ > >>forced-march port is no favor to customers.i > >> > >5G > > Again no disagreement. However, one of Fred's arguments for IA64 isnG > > that it provides CHumPaq with a route to provide Reasonable, if not F > > Low, cost workstations. (Yeah, I too feel that he has contradicted& > > himself on occasion in this area). > N > Keep in mind that there have been low cost VMS workstations in the past, andQ > that they could have been even lower priced if DEC hadn't worried so much about   > eroding the 'high end'. [snip]  F Y'know, I've always chafed at that argument: put out small, low-priced, machines and we'll kill our high-end sales.   ? Seems to me a case of the tail wagging the dog. If you know howaH consumers will respond, don't bemoan it and price your product right out of the market, EXPLOIT IT!  + ...but that's just me. YMMV considerably...e   -- e David J. Dachterat dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/o   ------------------------------  # Date: Tue, 29 Apr 2003 19:17:24 GMTL' From: Don Sykes <anonymous@pacbell.net>lY Subject: Re: OpenVMS on non-HP Itanium boxes (was: Re: OpenVMS Itanium port progressing wo+ Message-ID: <3EAECFBC.2448C9BA@pacbell.net>s   Hoff Hoffman wrote:y >  <snip>H >   The nice thing about industry standards is that there are so many to >   choose from.     Hoff,s Great line! = This statement should be etched in bronze and attached to the0D cornerstone of every hardware & software development company! :):):)   -- d   Have VMS, Will Traveln Wire paladin, San Francisco    (paladinATalphaseDOTcom)   ------------------------------    Date: 29 Apr 2003 11:24:58 -0700+ From: c00per11242001@yahoo.ca (Vic Mendham)eB Subject: opinions on implementing SNMP on OpenVMS 6.2 with UCX 4.2; Message-ID: <f7a73cb1.0304291024.ca6fbc@posting.google.com>   D I'm looking for opinions and thoughts on implementing SNMP using UCX 4.2 and OpenVMS 6.2.A When I do a Show version, I don't see any eco kits. Are there any  specific ones I should use?gF Also I think UCX 4.2 uses SNMPv1 and TCP/IP 5.3 only uses SNMPv2. If I; can't get to SNMPv3, does it really matter which one I use?tD If I can't upgrade to OpenVMS 7.2 or 7.3 can I still use TCP/IP 5.3, by upgrading UCX 4.2.x2 Is there a path to follow to for UCX 4.2 upgrades?F Can anyone give me their versions (o/s and UCx + patches ) so i can do$ a comparision to what I should have?  E HP just sent me the pdf's for UCX 4.2, so If anyone needs them let mep know.t   RegardsF   c00per11242001@yahoo.cal   ------------------------------  % Date: Tue, 29 Apr 2003 15:07:32 -0600n$ From: Lee Y T Mah <lytmah@cha.ab.ca>. Subject: Random access to the TZ89 tape system) Message-ID: <3EAEE993.D5361F78@cha.ab.ca>x  F I have a TZ89 DLT 35/70GB 10-tape system.  I'm using BACKUP to archiveE our files in a sequential manner.  Up to now sequential access to thepF tapes has been adequate.  However, I have encountered situations whereE if I'm working on tape 5, it would be nice to be able to dismount thetA current tape and get at the first tape again.  Are there DCL tapesF commands to do this?  Is such software available on the Freeware CD's?F We do have a VAX SLS license which could be transferred to the current4 Alpha cluster if SLS has the random tape capability.   -- Leec  5 L Y T Mah                    Email:  lytmah@cha.ab.ca  Capital Health Authority Information Systems, RAH CSC Edmonton, Alberta, CANADA    ------------------------------  # Date: Wed, 30 Apr 2003 03:42:47 GMTr1 From: Michael Austin <maustin@firstdbasource.com>s2 Subject: Re: Random access to the TZ89 tape system2 Message-ID: <3EAF387B.EEAC34B9@firstdbasource.com>   Lee Y T Mah wrote: > H > I have a TZ89 DLT 35/70GB 10-tape system.  I'm using BACKUP to archiveG > our files in a sequential manner.  Up to now sequential access to thelH > tapes has been adequate.  However, I have encountered situations whereG > if I'm working on tape 5, it would be nice to be able to dismount theuC > current tape and get at the first tape again.  Are there DCL tapetH > commands to do this?  Is such software available on the Freeware CD's?H > We do have a VAX SLS license which could be transferred to the current6 > Alpha cluster if SLS has the random tape capability. >  > -- > Lee  > 7 > L Y T Mah                    Email:  lytmah@cha.ab.ca  > Capital Health Authority > Information Systems, RAH CSC > Edmonton, Alberta, CANADAs   Lee,  E you can put the tape drive in Random Mode and use MRU???  to manuallyeG load/unload tapes using the ROBOT command.  Unfortunately this drive isuG either random or sequential.  The downside is that with Random mode YOUtF have to explictly load/unload tapes which means that when your current; tape finishes and BACKUP requests the next tape it will notyF automatically load the next slot.  I like the little TZ7 4mm DAT tapedD drive with 4 slots because it would automatically load the next tape5 (A2100 home system with 12GB raid5 using 2GB drives).r     --   Regards,  6 Michael Austin            OpenVMS User since June 19847 First DBA Source, Inc.    Registered Linux User #261163l   ------------------------------  % Date: Tue, 29 Apr 2003 22:25:55 -0400p' From: Stuart Fuller <stufuller@usa.net>O2 Subject: Re: Random access to the TZ89 tape system0 Message-ID: <j7cn8b.pdv.ln@dadsys2.fuller.local>   Lee Y T Mah wrote:  H > I have a TZ89 DLT 35/70GB 10-tape system.  I'm using BACKUP to archiveG > our files in a sequential manner.  Up to now sequential access to thecH > tapes has been adequate.  However, I have encountered situations whereG > if I'm working on tape 5, it would be nice to be able to dismount theCC > current tape and get at the first tape again.  Are there DCL tapegH > commands to do this?  Is such software available on the Freeware CD's?H > We do have a VAX SLS license which could be transferred to the current6 > Alpha cluster if SLS has the random tape capability. >   L When you bought your TL891 tape library system, did it come with a software J package called MRU?  This package provides the ROBOT command set to allow * you to manipulate your tape library robot.   -- o           Stub   ------------------------------  % Date: Tue, 29 Apr 2003 21:51:41 -0600o$ From: Lee Y T Mah <lytmah@cha.ab.ca>2 Subject: Re: Random access to the TZ89 tape system) Message-ID: <3EAF484C.61757290@cha.ab.ca>r  J End of my search.  I require the sequential mode with the autoload feature; more than the convenience of random access.  Thanks to all.      Michael Austin wrote:s   > Lee Y T Mah wrote: > >-J > > I have a TZ89 DLT 35/70GB 10-tape system.  I'm using BACKUP to archiveI > > our files in a sequential manner.  Up to now sequential access to thesJ > > tapes has been adequate.  However, I have encountered situations whereI > > if I'm working on tape 5, it would be nice to be able to dismount thetE > > current tape and get at the first tape again.  Are there DCL tape J > > commands to do this?  Is such software available on the Freeware CD's?J > > We do have a VAX SLS license which could be transferred to the current8 > > Alpha cluster if SLS has the random tape capability. > >n > > -- > > Lee  > >r9 > > L Y T Mah                    Email:  lytmah@cha.ab.can > > Capital Health Authority  > > Information Systems, RAH CSC > > Edmonton, Alberta, CANADA> >s > Lee, >.G > you can put the tape drive in Random Mode and use MRU???  to manuallyeI > load/unload tapes using the ROBOT command.  Unfortunately this drive is I > either random or sequential.  The downside is that with Random mode YOUeH > have to explictly load/unload tapes which means that when your current= > tape finishes and BACKUP requests the next tape it will notsH > automatically load the next slot.  I like the little TZ7 4mm DAT tapedF > drive with 4 slots because it would automatically load the next tape7 > (A2100 home system with 12GB raid5 using 2GB drives).  >  > --
 > Regards, >t8 > Michael Austin            OpenVMS User since June 19849 > First DBA Source, Inc.    Registered Linux User #261163m   -- Leey  5 L Y T Mah                    Email:  lytmah@cha.ab.ca  Capital Health Authority Information Systems, RAH CSC Edmonton, Alberta, CANADA4   ------------------------------   Date: 29 APR 2003 17:51:17 GMT+ From: Dave Greenwood <greenwoodde@ornl.gov>l% Subject: Re: RZ26-B in the real worlde2 Message-ID: <29APR03.17511794@feda01.fed.ornl.gov>  F In a previous article, "David D Miller" <ddmiller@raytheon.com> wrote: >  P > Folks. >   J > A clarification is in order.  The VAX 3100/48 is one of those boxes thatL > has a 1 GB system disk limitation as several of you pointed out.  The SCSIL > controller is Narrow-Slow.  However the /48 was designed to accomodate the< > "full height" drives so just about any solution will work. >  nJ > Many thanks to those that suggested the Hawk and provided model numbers.2 > BTW, I tried a Quantum Fireball without success. >  V > I'm off to eBay,  H What VMS version are you running?  IIRC, versions <=7.0 require that the% disk report geometry values such thatt  I  total blocks = Sectors per track * Total cylinders * Tracks per cylinder   I The DEC-branded disks did that but "generic" disks did not necessarily dotI so.  I've had problems with disks purchased on the open market because ofeJ this.  You get a "%INIT-F-INTDIV, arithmetic trap, integer divide by zero"F error when trying to initialize the disks if the geometry values don't
 "compute".   Dave --------------9 Dave Greenwood                Email: Greenwoodde@ORNL.GOVeH Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------  % Date: Tue, 29 Apr 2003 12:34:00 -0700 . From: "David D Miller" <ddmiller@raytheon.com>% Subject: Re: RZ26-B in the real world F Message-ID: <OFA100F257.CAC6189C-ON07256D17.006B5080@rsc.raytheon.com>  H Yikes!  Never heard of a problem like that and I've been doing 3100s for over 10 years.   I'll be installing 7.3 on it.h   dave.n   Dave Greenwood wrote ...F In a previous article, "David D Miller" <ddmiller@raytheon.com> wrote: >  > Folks. >tJ > A clarification is in order.  The VAX 3100/48 is one of those boxes thatG > has a 1 GB system disk limitation as several of you pointed out.  Thew SCSIH > controller is Narrow-Slow.  However the /48 was designed to accomodate theb< > "full height" drives so just about any solution will work. >dJ > Many thanks to those that suggested the Hawk and provided model numbers.2 > BTW, I tried a Quantum Fireball without success. >  > I'm off to eBay,  H What VMS version are you running?  IIRC, versions <=7.0 require that the% disk report geometry values such that   I  total blocks = Sectors per track * Total cylinders * Tracks per cylinder4  I The DEC-branded disks did that but "generic" disks did not necessarily do I so.  I've had problems with disks purchased on the open market because oftJ this.  You get a "%INIT-F-INTDIV, arithmetic trap, integer divide by zero"F error when trying to initialize the disks if the geometry values don't
 "compute".   Dave --------------9 Dave Greenwood                Email: Greenwoodde@ORNL.GOV H Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------  % Date: Tue, 29 Apr 2003 22:51:31 +0200 ( From: "H Vlems" <hvlems.nieuw@zonnet.nl>% Subject: Re: RZ26-B in the real world 5 Message-ID: <b8moko$bes46$1@ID-143435.news.dfncis.de>v  K An IBM 0662 A10 might be want you want. It is just under 1.05 GB and workedt8 well for me on a a VAXstation 3100-M48 as a system disk.  ; "David D Miller" <ddmiller@raytheon.com> schreef in berichtn@ news:OF3C182AB0.F412D3C5-ON07256D17.004C4B41@rsc.raytheon.com... > Folks: > K > My recently acquired VAXstation 3100/48 needs a system disk.  What is thenJ > Seagate/Quantum/Maxtor equivalent of a 1 GB drive (RZ26-B I believe)?  I3 > want to search eBay with the proper model number.a >n > dave.h >e >g   ------------------------------  % Date: Tue, 29 Apr 2003 19:10:15 -0500e/ From: Chris Scheers <chris@applied-synergy.com>w% Subject: Re: RZ26-B in the real world 3 Message-ID: <3EAF1467.731A4837@applied-synergy.com>h   David D Miller wrote:l >  > Folks: > K > My recently acquired VAXstation 3100/48 needs a system disk.  What is thenJ > Seagate/Quantum/Maxtor equivalent of a 1 GB drive (RZ26-B I believe)?  I3 > want to search eBay with the proper model number.r >  > dave.e  > If you can handle a 5 1/4" disk, the ST41200N works very well.  = If you need a BA42 box for a 5 1/4" disk, please let me know.k  G -----------------------------------------------------------------------,$ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com d   Fax: 817-237-3074e   ------------------------------  % Date: Tue, 29 Apr 2003 23:48:34 -0400)! From: Beyonder <beyonder@vrx.net>j! Subject: stupid keyboard question 8 Message-ID: <dqhuav4ubclr0hv60rcsaj6fli8dki3vpg@4ax.com>  4 I have an LK201 keyboard and it just dawned on me...  ( where the heck is the ESC (escape) key ?  C doesn't seem to be labelled. How do I do it? ctrl-27 ? compose-27 ?a  E I know its a real stupid question, and I hadn't noticed before today.T   any help would be appreciated.   B.   ------------------------------   Date: 29 Apr 2003 23:21 CDTf' From: carl@gerg.tamu.edu (Carl Perkins)h% Subject: Re: stupid keyboard question - Message-ID: <29APR200323213212@gerg.tamu.edu>v  % Beyonder <beyonder@vrx.net> writes...e5 }I have an LK201 keyboard and it just dawned on me...- } ) }where the heck is the ESC (escape) key ?  } D }doesn't seem to be labelled. How do I do it? ctrl-27 ? compose-27 ? } F }I know its a real stupid question, and I hadn't noticed before today. }  }any help would be appreciated.  }  }B./  + Try control-3. Or control-[, if you prefer.    --- Carl   ------------------------------  % Date: Wed, 30 Apr 2003 01:29:06 -0400e  From: John Santos <JOHN@egh.com>% Subject: Re: stupid keyboard questionr5 Message-ID: <1030430011453.2697B-100000@Ives.egh.com>t  $ On Tue, 29 Apr 2003, Beyonder wrote:  6 > I have an LK201 keyboard and it just dawned on me... > * > where the heck is the ESC (escape) key ? > E > doesn't seem to be labelled. How do I do it? ctrl-27 ? compose-27 ?( > G > I know its a real stupid question, and I hadn't noticed before today.t >   > any help would be appreciated. >  > B.  B IIRC, this is the original VT220 keyboard.  How to make it send anB escape depends on what it is plugged into.  I think all VT320s and? later generations, but not VT220s, can be configured (in SETUP)a@ to send an escape when the upper left key (tilda~/grave` key) isA pressed.  This is how I've always set mine.  VT220s and later can3B be set to send an escape when the F11 function key is pressed, but- this is far less convenient to us TECO users.   4 When all else fails, ctrl/[ should work on anything.  > If the keyboard is plugged into a workstation, then it depends@ on the application.  If DECTerm, it is an option in the "Options@ /Keyboard" menu.  If xterm, one of the many XMODMAP's may do the> trick.  If your LK201 is plugged into a PC and you are using a@ terminal emulator, key mapping is done by the emulator; RTFM for
 the emulator.   ; P.S.  Since the favored method of using the (~/`) key makes-< those characters go away, I usually set the comma and period9 keys to send < and > when shifted (just like VT100s did),m9 and set the (</>) key (lower left, next to the shift key)h< send tilda and grave.  The options to make these changes are: also in SETUP on VT320s and later, and in the Options menu in DECTerm.p   -- m John Santosh Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Tue, 29 Apr 2003 19:55:48 -0400l+ From: John Johnstone <jj_usenet2@yahoo.com> = Subject: Re: TCPIP: SMTP sender design problem (bounced mail)n) Message-ID: <3EAED8C4.13521567@yahoo.com>    JF Mezei wrote:n >  > VAX VMS 7.2,  TCPIP 5.3 ECO 2s > P > When a remote SMTP server rejects a username, the SMTP server simply returns aT > non-delivery notification without the actual text of the rejection, and just says: > ) > ---- Transcript of session follows ----iC > %TCPIP-E-SMTP_NOSUCHUSER, no such user, USER@HOST.THAT.BOUNCED.MEa% > USER@HOST.THAT.BOUNCED.ME (bounced)d >  > But here is the dialogue: + > send buf=RCPT TO:<jfmezei@host.kom>\0d\0aiR > recv buf=554 <modemcable084.123-201-24.mtl.mc.videotron.ca[24.201.123.84]>: ClieR > nt host rejected: listed in RBL DUL, see <URL:http://www.five-ten-sg.com/blackho > le.php>\0d\0ar > I > It would be very useful if the TCPIP folks would include the actual 5nnnM > message generated instead of generating a vms-style TCPIP-E-SMTP_NOSUCHUSERo) > which, in this case is very misleading.n  D Nearly all of the error messages that a system returns are mapped toB the "no such user" response.  Other systems thoughfully tell aboutD problems such as mailboxes being full but none of that makes its way back to the sender.u  G It's really interesting to see the results of mail sent between two VMSrG systems running TCP/IP Services when one system is configured to send asB reject to the other.  It's nice that you've now got the ability toC configure a reject message text.  It's too bad that the user on theeE system that sent the message will never see it.  His bounce has just:   G 550  %TCPIP-E-SMTP_COMMANDERR, SMTP command error recipient-domain-namen  : As you mentioned, the reject message isn't passed through.  G On the subject of bounce messages, the ones produced by TCP/IP Services , still have <LF> characters embedded in them.  / <LF>---- Transcript of session follows ----<LF>eG 550  %TCPIP-E-SMTP_COMMANDERR, SMTP command error recipient-domain-name'( <LF>---- Unsent message follows ----<LF>  F This causes the bounces to be rejected by some mail servers.  It's notG good when you produce a bounce message that's a candidate for rejections by its recipient.i   ------------------------------  % Date: Tue, 29 Apr 2003 19:42:20 -0400a) From: "Neil Rieck" <n.rieck@sympatico.ca>t3 Subject: Two day OpenVMS seminar in Ottawa, Ontario,9 Message-ID: <x5Era.4164$NX1.354890@news20.bellglobal.com>s  L The Association, in collaboration with HP Canada and HP Corporation, will beH hosting a 2-day technical seminar in Ottawa on June 25 and 26, 2003. AllL members, users and customers of HP technology from across Canada are invitedM to attend this information-packed seminar for IT Professionals involved in HPmN technology platforms. Registration is now open. Please visit our seminar pages) regularly to find out the latest updates.    http://www.canacu.org/) http://www.communitywerke.com/seminar.htmt  
 Neil Rieck Kitchener/Waterloo/Cambridge,e Ontario, Canada.! http://www3.sympatico.ca/n.rieck/o   ------------------------------  # Date: Tue, 29 Apr 2003 18:59:34 GMTI From: not@this.time (FullName) Subject: Upgrade Query2 Message-ID: <tZGcnd1Oj6_gVjOjXTWcoQ@metrocast.net> Keywords: upgrade, patches  	 greetingsl5      am about to upgrade OpenvVMS from 7.1-2 to 7.2. g;      have not installed several 'non-critical' patches thata%      are available for version 7.1.  s9      is it imperative that all patches for one version bea8      installed before an upgrade, or can the 'fixes'  be-      expected as features of the new version?t      thanks in advance.   ! b underscore clark at usa dot nets   ------------------------------  % Date: Tue, 29 Apr 2003 16:11:46 -0400f From: norm.raphael@metso.com Subject: Re: Upgrade Query? Message-ID: <OFF0C362F2.E11A36D5-ON85256D17.006EBC6D@metso.com>w  E In general, the patches for any version are dropped and you will need J to get and install the new patches after/during the upgrade for the higherE version.  Unless the patches are to PCSI, they most likely would be ayE waste of time/energy to do before the upgrade - providing you are noth having crashes, of course.  C You have read the release notes, and will make several good backupst before proceding, yes?    6 From:  not@this.time (FullName) on 04/29/2003 02:59 PM  * Please respond to not@this.time (FullName)   To:    Info-VAX@Mvb.Saic.Com cc:p   Subject:    Upgrade Querye    	 greetingsn4      am about to upgrade OpenvVMS from 7.1-2 to 7.2.;      have not installed several 'non-critical' patches thatb#      are available for version 7.1.e9      is it imperative that all patches for one version ben8      installed before an upgrade, or can the 'fixes'  be-      expected as features of the new version?s      thanks in advance.   ! b underscore clark at usa dot netr   ------------------------------  % Date: Tue, 29 Apr 2003 21:27:13 -0500o1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: VMS 7.3-1' Message-ID: <3EAF3481.F9A7C363@fsi.net>    "John N." wrote: > [snip]J > Changing stacks is not a short term option, even if we were so inclined.N > This reflects badly on VMS itself since the problem is only occurring on VMSH > 7.3-1 nodes and provides ammunition to those that want to supplant yet0 > another large, but tenuous VMS bastion. [snip]  : Well, speaking only for myself and my own preferences, ...  D My first TCP/IP stack was CMU/IP. Being rather green at it, I didn'tG understand enough of it to actually get it to work, but I did study they doc.'s quite a bit.   A Then, I was exposed to UCX, circa. V3.2 or so. I saw how they did H printers and how to BS spooled devices using LAT, and I figured "o.k., I can follow this".f  G Then, I was exposed to Multinet and found myself back in what I thoughteB was a better place. Still had UCX on VAXstations and such, but the production boxes were Multinet.t  F Later on, I got some exposure to TCPware. Just enough to be dangerous.  H Of the three, I'd have to say that I prefer Multinet, maybe because I've5 used it the most and am the most comfortable with it.a  A ...but yes, changing stacks is not likely an option in your case.   G I'd be interested in knowing the answer to your issue, so I hope you'lleF post again if comes by off-line mail. I can see network issues causingD sessions to be dropped - firewalls dropping VCs and such, congestionG resulting in severe packet loss, etc., but there's always something newo
 to learn, eh?n   -- z David J. Dachteral dba DJE Systemsn http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/4   ------------------------------  # Date: Wed, 30 Apr 2003 04:29:22 GMT 4 From: Craig A. Berry <craigberry@mac.com.spamfooler> Subject: Re: VMS 7.3-17 Message-ID: <8d3167028601fae77d825827788e31ee@TeraNews>   B In <669ra.32711$My6.441574@twister.tampabay.rr.com> John N. wrote:  G > We also upgraded one of our Alphas (a GS140)  from VMS 7.2-1 w TCPIP iH > version 5.0 to VMS 7.3-1 w TCPIP v 5.3 eco 2 and a slew of other ECOs.  F > Occasionally (about 20 times yesterday to three different users), a D > telnet session hangs, then disconnects, returning the user to the , > original system from which they telnetted. >e( > Any ideas?  Anything obvious to check?  G Check your sysgen parameter MAXBUF per the release notes to the latest   TCP/IP ECO:s  G <http://ftp1.support.compaq.com/public/vms/axp/v7.3-1/tcpip/5.3/axpvms-s tcpip_eco-v0503-181-4.README>   F That's just a wild guess but it is a known problem with telnet in the  environment you describe.i   ------------------------------   End of INFO-VAX 2003.237 ************************