1 INFO-VAX	Fri, 22 Mar 2002	Volume 2002 : Issue 160       Contents: Re: ADA95 for ia64/OpenVMS4 ANNOUNCE: Security Update kits for CSWS and CSWS_PHP( Another AntiSPAM RBL dies - ORBZ is gone Re: buffer I/O& Re: Compaq brand to survive the merger& Re: Compaq brand to survive the merger
 DCPI for VMS!  Re: DCPI for VMS!  Re: DCPI for VMS!  Re: DCPI for VMS! / Re: DEC C: why does exit(0) really exit with 1? / Re: DEC C: why does exit(0) really exit with 1? / Re: DEC C: why does exit(0) really exit with 1? / Re: DEC C: why does exit(0) really exit with 1? E Re: DECterm X client, ACCPORNAM (was: Re: 7.3a and 7.4 Release dates)  Re: Getting file date/time in C . Re: HELP: Changing redirected SYS$OUTPUT Width Re: Hidden files? A Re: How does one check to verify all nodes are fully operational? A RE: How does one check to verify all nodes are fully operational? P Re: How does one check to verify all nodes are fully operational? -> use a ClustP Re: How does one check to verify all nodes are fully operational? -> use a ClustG Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux G Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux G Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux G Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux F Re: HP's viewpoint on Linux, was: Re: Sun eating major helping ofLinuxF Re: HP's viewpoint on Linux, was: Re: Sun eating major helping ofLinuxF Re: HP's viewpoint on Linux, was: Re: Sun eating major helping ofLinux
 Re: Hypersort 
 Re: Hypersort 
 Re: Hypersort 
 Re: Hypersort  Re: LD062 Install Question Re: LD062 Install Question Re: LD062 Install Question Re: lpd printing + telnetqueue Microsoft bullies PC makers 8 Re: MONI SYS shows 80% CPU usage and 8 jobs in COM state* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...* Re: Once a fool, always a fool I guess ...$ Re: OpenVMS 7.2-1 and UCX 4.2 ECO 05# OpenVMS Technical Update for London  Patches for CSWS; Re: Problem with SMTP mail on TCPIP v5.1. Help is sought... ; Re: Problem with SMTP mail on TCPIP v5.1. Help is sought... 4 process priorities under v7.1-2, 7.2-1 Alpha servers8 Re: process priorities under v7.1-2, 7.2-1 Alpha servers8 Re: process priorities under v7.1-2, 7.2-1 Alpha servers8 Re: process priorities under v7.1-2, 7.2-1 Alpha servers8 Re: process priorities under v7.1-2, 7.2-1 Alpha servers8 Re: process priorities under v7.1-2, 7.2-1 Alpha servers" Question reguarding process crash. Re: Right down the middle P Re: Session abstracts and bio's for OpenVMS Technicial Updates in Europe (AustriP Session abstracts and bio's for OpenVMS Technicial Updates in Europe (Austria an$ Re: Shannon knows DEC -> CPQ -> ????$ RE: Shannon knows DEC -> CPQ -> ????$ Re: Shannon knows DEC -> CPQ -> ????$ Re: Shannon knows DEC -> CPQ -> ????$ Re: Shannon knows DEC -> CPQ -> ????$ Re: Shannon knows DEC -> CPQ -> ????" Re: System software status utility" Re: System software status utility" Re: System software status utility Upgrading from v7.2 % Re: VAX/VMS emulator or free version? % Re: VAX/VMS emulator or free version?  Why Delaware ? Re: Why Delaware ? Re: Why Delaware ? Re: Why Delaware ? Re: Why Delaware ? Re: Why Delaware ?  F ----------------------------------------------------------------------  % Date: Fri, 22 Mar 2002 16:41:25 +0100 = From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> # Subject: Re: ADA95 for ia64/OpenVMS ) Message-ID: <3C9B50A5.F49DD68D@gtech.com>    Martin Heller wrote:=  Perhaps of some interest (copied from the gcc mailing list):  > J > "I wanted to let people on this list know that Ada Core Technologies hasK >   signed a contract with Compaq to implement GNAT on OpenVMS for ia64. We J >   already have three ia64 machines in house, and are busy working on theF >   initial step of bootstrapping the current version of GNAT on ia64.  
 Good news.   For those that use Ada.   < And for the rest of us (it shows commitment to VMS Itanium).   Arne   ------------------------------  % Date: Fri, 22 Mar 2002 10:55:50 -0500 + From: "Rick Barry" <barry@star.zko.dec.com> = Subject: ANNOUNCE: Security Update kits for CSWS and CSWS_PHP 3 Message-ID: <RuIm8.1388$fL6.27212@news.cpqcorp.net>   9     Compaq Secure Web Server and PHP Security Update Kits   +     Update kits are can be downloaded from:     K http://www.openvms.compaq.com/openvms/products/ips/apache/csws_patches.html   ,     The kits address the following problems:  D         - Compaq Secure Web Server: mod_ssl buffer-overflow problem,6                 privileged port range control problem.  G         - PHP for Compaq Secure Web Server: file upload buffer-overflow                  problem.  D    Refer to the README.TXT files on the download site for additional information.  
 Rick Barry Compaq Secure Web Server OpenVMS System Software Group  Compaq Computer Corporation 
 Nashua, NH   ------------------------------  % Date: Fri, 22 Mar 2002 11:03:54 -0700 + From: "Barry Treahy, Jr." <Treahy@mmaz.com> 1 Subject: Another AntiSPAM RBL dies - ORBZ is gone ' Message-ID: <3C9B720A.7020607@mmaz.com>   E This is not completely off-topic in that if you use, or rather used,  D ORBZ for DNS anti-spam support, they where forced to shutdown under > duress of criminal charges and if your SMTP servers are still E referencing them, your inbound e-mail with most likely be blocked as  H ours have been for the base 36 hours; They shutdown on March 20, 2002.    H Other DNS RBL's still exist, one of the larger ones that are still free & is ORDB or you make subscribe to MAPS.   Regards,   Barry    --    @ Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028    ------------------------------  % Date: Fri, 22 Mar 2002 10:07:17 +0100 $ From: "labadie" <labadie_g@decus.fr> Subject: Re: buffer I/O 3 Message-ID: <WwCm8.1367$fL6.27115@news.cpqcorp.net>   2 "Joerg Spilling" <spi@equicon.de> wrote in message) news:a7ddkm$crh$01$1@news.t-online.com... 	 > Gerard,  > K > I'm also interesset on this theme. But I see your  specified link broken? 1 > Any ideas or can you provide the source itself?  >  Hello   J The link works, but the end of the line is .html and not .h, as my browser" shows, having split the line in 2.  L http://www.compaq.com/support/asktima/communications/CHAMP_SRC010730002768.h tml   , If you can't access, I will mail you a copy.   Regards    Grard   ------------------------------    Date: 22 Mar 2002 07:48:34 -0600B From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley)/ Subject: Re: Compaq brand to survive the merger 3 Message-ID: <0iW4dp+fWh3o@eisner.encompasserve.org>   t In article <BBum8.85704$ZR2.37219@rwcrnsc52.ops.asp.att.net>, "Terry C. Shannon" <terryshannon@mediaone.net> writes: > K > Some of 'em very well may. I suspect that the ProLiant name will live on, G > but not much is certain beyond that. MC was pretty emphatic about the ( > persistence of the brand name, though. >   N I wonder if there's any chance of getting the Digital name back as well... :-)   Simon.   --  B Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP       + Microsoft: The Lada of the computing world.    ------------------------------  # Date: Fri, 22 Mar 2002 15:13:36 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>/ Subject: Re: Compaq brand to survive the merger - Message-ID: <ASHm8.91152$af7.53895@rwcrnsc53>   G "Simon Clubley" <clubley@remove_me.eisner.decus.org-Earth.UFP> wrote in 5 message news:0iW4dp+fWh3o@eisner.encompasserve.org... I > In article <BBum8.85704$ZR2.37219@rwcrnsc52.ops.asp.att.net>, "Terry C. , Shannon" <terryshannon@mediaone.net> writes: > > I > > Some of 'em very well may. I suspect that the ProLiant name will live  on, I > > but not much is certain beyond that. MC was pretty emphatic about the * > > persistence of the brand name, though. > >  > L > I wonder if there's any chance of getting the Digital name back as well... :-)   E Well, that belongs to Digital India and Digital Networks now. Digital K Networks won't be affected by the acquisition, and I haven't a clue what'll # happen with the folks in Bangalore.    ------------------------------  # Date: Fri, 22 Mar 2002 11:44:14 GMT . From: ">>> ^P" <plj@NOSPAM.byron.ext.telia.se> Subject: DCPI for VMS!9 Message-ID: <3C9B1915.2E7372A2@NOSPAM.byron.ext.telia.se>    Not to be missed:   > http://www.openvms.compaq.com/openvms/products/dcpi/index.html    	 >>> ^P.Lj    ------------------------------  % Date: Fri, 22 Mar 2002 09:48:48 -0500 - From: "Peter Weaver" <peter.weaver@stelco.ca>  Subject: Re: DCPI for VMS!/ Message-ID: <wvHm8.647$je5.6271@nnrp1.uunet.ca>   9 ">>> ^P" <plj@NOSPAM.byron.ext.telia.se> wrote in message 3 news:3C9B1915.2E7372A2@NOSPAM.byron.ext.telia.se...  >  > Not to be missed:  > @ > http://www.openvms.compaq.com/openvms/products/dcpi/index.html    L Can anyone exlpain to me what this has to do with VMS? Besides the fact thatK this page is in the VMS section and the e-mail address given has VMS in it, C I can not find VMS on any page that this "announcement" references.    -- Peter WeaverL Opinions are my own, and do not reflect the opinions of my employer, nor theK company that it sub-contracts to, nor the company that it sub-contracts to.    ------------------------------  % Date: Fri, 22 Mar 2002 09:58:12 -0500 - From: "Peter Weaver" <peter.weaver@stelco.ca>  Subject: Re: DCPI for VMS!/ Message-ID: <lEHm8.650$je5.7012@nnrp1.uunet.ca>   8 "Peter Weaver" <peter.weaver@stelco.ca> wrote in message) news:wvHm8.647$je5.6271@nnrp1.uunet.ca...  >.. I > Can anyone exlpain to me what this has to do with VMS? Besides the fact  thatI > this page is in the VMS section and the e-mail address given has VMS in  it, E > I can not find VMS on any page that this "announcement" references.    >...  K Cancel that, I found the links on the left side of the page where the black  background is.   -- Peter WeaverL Opinions are my own, and do not reflect the opinions of my employer, nor theK company that it sub-contracts to, nor the company that it sub-contracts to.    ------------------------------  # Date: Fri, 22 Mar 2002 14:59:38 GMT . From: ">>> ^P" <plj@NOSPAM.byron.ext.telia.se> Subject: Re: DCPI for VMS!9 Message-ID: <3C9B46E0.F7A995DA@NOSPAM.byron.ext.telia.se>   L If you look to the left on the given page, a little further down, you'll see something like:      DCPI:  > Home > What's New > Install Software > Frequently Asked Questions > Documentation  > Publications
 > Feedback     Need I Say More!  	 >>> ^P.Lj      Peter Weaver skrev:   ; > ">>> ^P" <plj@NOSPAM.byron.ext.telia.se> wrote in message 5 > news:3C9B1915.2E7372A2@NOSPAM.byron.ext.telia.se...  > >  > > Not to be missed:  > > B > > http://www.openvms.compaq.com/openvms/products/dcpi/index.html > N > Can anyone exlpain to me what this has to do with VMS? Besides the fact thatM > this page is in the VMS section and the e-mail address given has VMS in it, E > I can not find VMS on any page that this "announcement" references.  >  > -- > Peter WeaverN > Opinions are my own, and do not reflect the opinions of my employer, nor theM > company that it sub-contracts to, nor the company that it sub-contracts to.    ------------------------------  % Date: Fri, 22 Mar 2002 15:52:55 -0000 / From: Michael Zarlenga <zarlenga@conan.ids.net> 8 Subject: Re: DEC C: why does exit(0) really exit with 1?/ Message-ID: <u9mkqn3t907lda@corp.supernews.com>   3 Jan C. Vorbrggen <jvorbrueggen@mediasec.de> wrote: C :> But if you want to write a user-friendly program, you should not A :> exit with zero but with a status identifying the nature of the C :> error, so the user can type HELP/MESSAGE as the next DCL command " :> and get a detailed explanation.  F : As I said - to a standard-conforming C program, exiting with 0 is an : indication of success.  = This may be seen as beating the proverbail dead horse, but I  < just don't believe that 0 is the standard-conforming success
 code in C.   Is that written anywhere?   = Many C RTL functions return 0 on failure, so how can 0 be any C kind of standard "success" value in C (though it is the traditional  success code for _Unix_).    --   -- Mike Zarlenga  >    "Now, throughout history, whenever people get wood, they'll'     think of Trojans!"  - Ned Flanders.    ------------------------------  + Date: Fri, 22 Mar 2002 17:12:13 +0100 (MET) B From: "Dr. Martin P.J. Zinser" <ZINSER@sysdev.deutsche-boerse.com>8 Subject: Re: DEC C: why does exit(0) really exit with 1?; Message-ID: <01KFO32C95EQ8ZRA5Q@sysdev.deutsche-boerse.com>    Hello,  E well how about the DEC C stdlib.h? The relevant part looks like this:    /*< ** Exit function failure definition changes corresponding to+ ** V7.0 "UNIX"-like exit function behavior.  */4 #if __CRTL_VER >=3D 70000000 && defined(_POSIX_EXIT) #   define  EXIT_FAILURE  (1)  #else $ #   define  EXIT_FAILURE  0x10000002 #endif #define  EXIT_SUCCESS  0  G Having said that, I do prefer to use the EXIT_* macros over values like G 1 or 0 since they express the intention much clearer in the code and=20 4 also allow an abstraction layer helping portability.   Greetings, Martin   J > From:	IN%"zarlenga@conan.ids.net"  "Michael Zarlenga" 22-MAR-2002 17:04= :41.337 > Subj:	RE: DEC C: why does exit(0) really exit with 1?   7 > Jan C. Vorbr=FCggen <jvorbrueggen@mediasec.de> wrote: E > :> But if you want to write a user-friendly program, you should not C > :> exit with zero but with a status identifying the nature of the E > :> error, so the user can type HELP/MESSAGE as the next DCL command $ > :> and get a detailed explanation. >=20H > : As I said - to a standard-conforming C program, exiting with 0 is an > : indication of success. >=20> > This may be seen as beating the proverbail dead horse, but I> > just don't believe that 0 is the standard-conforming success > code in C. >=20 > Is that written anywhere?  >=20? > Many C RTL functions return 0 on failure, so how can 0 be any E > kind of standard "success" value in C (though it is the traditional  > success code for _Unix_).  >=20 > -- > -- Mike Zarlenga >=20@ >    "Now, throughout history, whenever people get wood, they'll) >     think of Trojans!"  - Ned Flanders. J Dr. Martin P.J. Zinser                       zinser@sysdev.deutsche-boers= e.com 6 Deutsche Boerse Systems Inc.                       =20D Suite 1580                                   Tel: +1-312-408-3085=20A 141 West Jackson Blvd.                       FAX: +1-312-408-3071 / Chicago, IL, 60604                          =20 F USA                                          Private:  zinser@decus.de   ------------------------------   Date: 22 Mar 2002 17:03:45 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) 8 Subject: Re: DEC C: why does exit(0) really exit with 1?, Message-ID: <a7fo5h$2ffs$1@info.cs.uofs.edu>  / In article <u9mkqn3t907lda@corp.supernews.com>, 2  Michael Zarlenga <zarlenga@conan.ids.net> writes: |>  - |> Many C RTL functions return 0 on failure,    ? I am not aware of any C RTL functions that return 0 on failure. 2 Normaly they would return -1.  Can you name one??   @ |>                                           so how can 0 be anyF |> kind of standard "success" value in C (though it is the traditional |> success code for _Unix_).  @ 0 is not some universal success code either as a large number of= C RTL functions actually return meaningful values on success.   B While the problem pointed to by the original question is anomalous@ and interesting, it probably isn't really any kind of bug.  JustB another example of bad programming practice and further proof that> Unix programmers don't have that market cornered. (as was alsoB demonstrated by many of the respopnses that showed a definite lackA of understanding of how how C and it's myriad RTL's really work.)S   bill   -- nJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   i   ------------------------------  % Date: Fri, 22 Mar 2002 12:42:37 -0500y5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> 8 Subject: Re: DEC C: why does exit(0) really exit with 1?3 Message-ID: <K8Km8.1395$fL6.27224@news.cpqcorp.net>n  I With no standards, defacto standards emerge, and eventually are codified.iL While returning 0 for success is rather perverse (since the test for successL is something like (!status) - it probably stems from a lot of UNIX compilers& always initializing variables to zero.  G But again, with no standards, you can't expect consistancy, so it isn'td *universal*.  % Michael Zarlenga wrote in message ...p4 >Jan C. Vorbrggen <jvorbrueggen@mediasec.de> wrote:D >:> But if you want to write a user-friendly program, you should notB >:> exit with zero but with a status identifying the nature of theD >:> error, so the user can type HELP/MESSAGE as the next DCL command# >:> and get a detailed explanation.t > G >: As I said - to a standard-conforming C program, exiting with 0 is ani >: indication of success.j >h= >This may be seen as beating the proverbail dead horse, but Ig= >just don't believe that 0 is the standard-conforming successe >code in C.y >s >Is that written anywhere? >n> >Many C RTL functions return 0 on failure, so how can 0 be anyD >kind of standard "success" value in C (though it is the traditional >success code for _Unix_). >n >--s >-- Mike Zarlengao >a? >   "Now, throughout history, whenever people get wood, they'll.( >    think of Trojans!"  - Ned Flanders.   ------------------------------  % Date: Fri, 22 Mar 2002 18:23:04 -0000 ? From: "Malcolm MacArthur" <malcolmix@neverness.freeserve.co.uk> N Subject: Re: DECterm X client, ACCPORNAM (was: Re: 7.3a and 7.4 Release dates). Message-ID: <a7fso4$2uv$1@news8.svr.pol.co.uk>  ? "Hoff Hoffman" <hoffman@xdelta.zko.dec.nospam> wrote in message - news:6Ywm8.1357$fL6.26516@news.cpqcorp.net...oE > In article <a7d80t$va9$1@newsg2.svr.pol.co.uk>, "Malcolm MacArthur"f- <malcolmix@neverness.freeserve.co.uk> writes:- >nM > :And will we be able to tell who is on a DECterm using SHOW USERS in 7.3-1,r< > :or will we have to keep using ANALYZE/SYSTEM to find out? >eJ >   I will assume this is a continuation of the earlier "rant" (your word) >   on ACCPORNAM.a >rI >   Incoming IP connections were fixed in TCP/IP Services V5.1 and later, H >   and these now correctly identify themselves via the SYS$REM* logical* >   names and via the ACCPORNAM interface. >iI >   DECterm, however, is an X Windows client and thus targets an outgoing,L >   X Windows display, and a remote X Windows Server -- and this is not whatH >   is normnally found in the ACCPORNAM field of the (incoming) terminalH >   device -- that's the source of the connection, not the target of the >   connection.i  M OK, as you rightly point out, a DECterm is connected 'back-to-front' compared  to aJ normal terminal device. But, just because the connection was initiated the	 other way N around doesn't make the location it's displayed AT any less useful... in fact, it is much more L useful to know where a DECterm has appeared than to know where you initiated it from H (which may or may not be the same place)! Or to know nothing about it at all...  5 > The details of the connection target (stored in theeH >   workstation device) are and have been available via the SHOW DISPLAYK >   command, with SHOW DISPLAY/SYMBOL (undocumented) and the $qio interface I >   (also undocumented, examples available on the Freeware and elsewhere)eF >   provided details on the X Windows server that is the target of the >   DECterm display.  J Umm, yeah, right. So how does an ordinary unprivileged user know that, forH instance, FTA23: is using DECwindows pseudodevice WSA3: for its display?  N True, you can list the WSAn: display settings for each display device - but itM won't tell you which _terminals_ are associated with it. The only (supported?nC If it is even supported) method of finding out that information is:n  N 1) Use SHOW SYSTEM to find each DECW$TE_nnnn pseudoterminal controller processK 2) ANALYZE /SYSTEM SHOW PROCESS /ID=<pid of DECW$TE_nnnn process> /CHANNELSaM repeatedly for each controller in turn until you find the one controlling them terminal in question.S= 3) Note down the display device used by that process (WSAnn:)1C 4) Do SHOW DISPLAY WSAnn: to find out where that display points to.P  I Additionally, a WSAnn: device might not have a DECterm associated with it  anyway. It might< well be a non-DECterm application that is using the display.   >hI >   I regularly use the incoming connecting information that is stored inrL >   ACCPORNAM -- and the details on the DECterm escape and control sequencesM >   that is listed in the FAQ -- to set the DECterm icon and window displays.oN >   (I use a network connection to start a DECterm -- similar to the RDECW.COMK >   procedure that's been around for a while -- and the incoming connectionvK >   ACCPORNAM details are then used to target the X Windows display setting B >   back at the remote X Windows server; at the originating host.)  M There is another point in all this as well. Say someone tries to hack into myB system,OC and he is using X Windows. Now, he logs on, does a CREATE /TERMINAL H /DETACHED /NOLOGGED_IN, and logs in as (for instance) SYSTEM. Now, a fewI days later, I notice something unusual happened. So I go through my auditc records and L discover that SYSTEM logged on to terminal FTA23:. Now, from this historicalN perspective, how do I find out where pseudoterminal FTA23: was initiated from? In fact,% how can I even tell it was a DECterm?o  M Can use of PTD$CREATE (or device creation from a template) be audited? I just.K tried putting an alarm ACE on FTA0: (ALARM=SECURITY, ACCESS=WRITE+SUCCESS).VJ However, when your new DECterm is created, it inherits the security ACE on FTA0: - with theM result that every write to your new FTAnn: device fires off an audit alarm...  This occurs even if thesF ACL is set OPTIONS=NOPROPAGATE. I assume the same would be true of the DECwindows WSAnn:eH pseudodevice.... Looks like a bug to me (or an undocumented shortcoming, depending which way you look at it...) VMS 7.2-1.  D However, if I could find a way to audit creation of new devices from templates, I would atsJ least have a concrete audit trail. Sure, a DECterm will have an associated DECW$TE_nnnn process,oL but I just have to guess from there. I can't tell for sure if the session on FTA23: was initiatedI from DECW$TE_1234 or DECW$TE_5678 - not in a way that could stand up in a  (Scottish) court of law (if it came to that).  H So, as a compromise, I would settle for a method of auditing creation of pseudoterminals andi( DECwindows WSAnn: display pseudodevices.   >iH >   I do not expect that ACCPORNAM would ever be modified to contain theG >   target of a DECterm (X Windows) display.  (But if you were to writenH >   this product requirement onto the back of a cashier's check of largeH >   enough value or otherwise convincingly state the use of this change,3 >   local management might well change my mind. :-)e  	 Bah... :(t   >a >n( >  ---------------------------- #include' <rtfaq.h> ----------------------------- L >       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com, >  --------------------------- pure personal# opinion ----------------------------N >    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com >"   ------------------------------  % Date: Fri, 22 Mar 2002 16:44:35 +0100e= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>u( Subject: Re: Getting file date/time in C) Message-ID: <3C9B5163.D0127677@gtech.com>W   Dave Rich wrote:G > I have looked forwards and backwards through help|system_services ande
 > help|rtl...  > M > Anyone have examples of getting the date/time stamp from a file and storingIC > it in a char varaible or something of that nature for comparison?  > B > I ask because, The application i have written needs to re-read a7 > configuration file.. ONLY if a change has been made..'   C RTL : stat functionh: VMS   : RMS calls ($OPEN or $DISPLAY with appropriate XAB)   Arne   ------------------------------  % Date: Fri, 22 Mar 2002 07:45:34 -0700h0 From: "Mike Ober" <obermd.@.alum.mit.edu.nospam>7 Subject: Re: HELP: Changing redirected SYS$OUTPUT Width-/ Message-ID: <jsHm8.6$a47.20793@news.uswest.net>    Thanks,^  L I found the MARGIN and NOMARGIN statements late yesterday.  I ended up usingJ the NOMARGIN statement as it's the better fit for what I need.  It took meD over two and a half days to find the MARGIN and NOMARGIN statements.   Mike.-  - "John Santos" <JOHN@egh.com> wrote in message:0 news:1020321164912.36972A-100000@Ives.egh.com...' > On Thu, 21 Mar 2002, Mike Ober wrote:e >gL > > When using LIB$Spawn to redirect SYS$OUTPUT to a file, the file wraps at 72K > > characters.  I need to change this to 132 characters.  The program is a  VMS 8 > > Basic program that is running detached via loginoff. > >/ > > Thanks,/ > > Mike Ober. > G > You might need to use the MARGIN statement to change BASIC's built-in  > line wrapping. >ME > I think it normally uses the terminal's width, but if sys$output isVC > redirected to a file, it doesn't have a width, and so it uses thevD > default of 72.  I  think it inherited this from Dartmouth BASIC on< > a Teletype ASR33.  Isn't backwards compatibilty cool?  ;-) >t > --
 > John Santos  > Evans Griffiths & Hart, Inc. > 781-861-0670 ext 539 >2   ------------------------------  % Date: Fri, 22 Mar 2002 16:51:16 +0100R= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>a Subject: Re: Hidden files?) Message-ID: <3C9B52F4.C1886FD8@gtech.com>H   Dave Rich wrote:E > Can't seem to locate the standard include files on 6.2-1H3 OpenVMS.t0 > But, no problem finding them on 7.2 OpenVMS... > 
 > What gives?  > I > from the top of the system disk... dir [...]stdio.h reveals no matches.a > * > Where did dec hide the include files at? > / > I can find some, but not the 'standard' ones.p > < > Sorry if this is a stupid question, but I have to ask. ;-)   ????  ' The header-files are not files on VMS !e  : They are modules in libraries (SYS$LIBRARY:DECC$RTLDEF.TLB for stdio.h etc.).  @ If your system-manager answered yes to the rigth question during: installation of DEC C, then there will be a reference copy* of the files in SYS$COMMON:[DECC$LIB...] !   Arne   ------------------------------  % Date: Fri, 22 Mar 2002 09:05:35 +0100e) From: Bart Zorn <B.Zorn@xs4all.nospam.nl>sJ Subject: Re: How does one check to verify all nodes are fully operational?/ Message-ID: <3C9AE5CF.6040208@xs4all.nospam.nl>   
 JMK wrote:I > Given 50 workstations on clusters scattered over an are larger than you E > can walk comfortably in a day.  You reboot one or more clusters.  A9J > glance at $moni clus and you can see if the workstations are online, and8 > $sho dev d to see if the disks are online and mounted. > H > But what can you do to verify that logicals are defined, that paths toF > executables have been created, commands defined, etc. -- on each andH > every workstation.  This is a problem if you have a larger system withG > many comfiles which are run as part of the boot process, and existingwA > design is such that it is possible to not be ready when logicalnG > assignments are made, thus not being totally ready when the user sits  > down.- > G > Are there any existing software applications that address this topic?  >  > TIA, >  > Jeff Klopotic   L Like I responded at your other request, Cockpit Manager can do all you want.  	 Bart Zorns   ------------------------------  % Date: Fri, 22 Mar 2002 16:16:13 +0100B7 From: "Dijk, Jeroen van" <Jeroen.vanDijk@Getronics.com>lJ Subject: RE: How does one check to verify all nodes are fully operational?O Message-ID: <2795B75EF003D311801A00A0C906B511011C6DC1@cucexec.gbc.getronics.nl>a  M What was done at a site I have worked is putting together a simple .com file,nH that checks if the critical processes runs on every node in the network.O This was runned by an operator every morning on 4:00 before the users come in. r^ Simple by login on a captive account. Just three or four paged with x's that needs be checked.  @ Also where was some sort of watchdog for the systemadmistrators.` That displayed every critical and warning message from every thing you wanted to keep an eye on.P For example diskspace, critical processes, hardware failures, networklinks, etc.  * Both are custommade and worked very well.        > -----Original Message-----0 > From: JMK [mailto:jeffrey.m.klopotic@lmco.com]$ > Sent: donderdag 21 maart 2002 3:15 > To: Info-VAX@Mvb.Saic.ComaH > Subject: How does one check to verify all nodes are fully operational? >  > : > Given 50 workstations on clusters scattered over an are  > larger than you E > can walk comfortably in a day.  You reboot one or more clusters.  A*? > glance at $moni clus and you can see if the workstations are 3
 > online, ande8 > $sho dev d to see if the disks are online and mounted. > H > But what can you do to verify that logicals are defined, that paths toF > executables have been created, commands defined, etc. -- on each andH > every workstation.  This is a problem if you have a larger system withG > many comfiles which are run as part of the boot process, and existing A > design is such that it is possible to not be ready when logical2G > assignments are made, thus not being totally ready when the user sits1 > down.a > G > Are there any existing software applications that address this topic?  >  > TIA, >  > Jeff Klopoticu >    ------------------------------  % Date: Fri, 22 Mar 2002 09:49:59 +0100h$ From: "labadie" <labadie_g@decus.fr>Y Subject: Re: How does one check to verify all nodes are fully operational? -> use a Clusts3 Message-ID: <HgCm8.1365$fL6.26905@news.cpqcorp.net>h  4 "JMK" <jeffrey.m.klopotic@lmco.com> wrote in message! news:3C994225.9DA690B@lmco.com...aI > Given 50 workstations on clusters scattered over an are larger than youeE > can walk comfortably in a day.  You reboot one or more clusters.  ALJ > glance at $moni clus and you can see if the workstations are online, and8 > $sho dev d to see if the disks are online and mounted. >oH > But what can you do to verify that logicals are defined, that paths toF > executables have been created, commands defined, etc. -- on each andH > every workstation.  This is a problem if you have a larger system withG > many comfiles which are run as part of the boot process, and existing A > design is such that it is possible to not be ready when logical2G > assignments are made, thus not being totally ready when the user sits8 > down.h >gG > Are there any existing software applications that address this topic?  >  > TIA, >P > Jeff Klopotice  I If you have Vms 7.2 or later, and if you have a good handling of set noonr and so in your systartup_vms0 you can put at the end of your systartup_vms.com  ? def/table=lnm$cluster_table 'f$getsyi("nodename")_start_ok true.I This can also be put after a test (if appli1_ok and appli2_ok... then...)    After a reboot, ae' sh lo/table=lnm$cluster_table *start_ok  on any nodea% should show all your Cluster members. A If one is missing, you know something went wrong during the boot.    Regardsm   Grard   ------------------------------    Date: 22 Mar 2002 09:08:46 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)iY Subject: Re: How does one check to verify all nodes are fully operational? -> use a Cluste3 Message-ID: <ycFyq8anb8YX@eisner.encompasserve.org>r  Z In article <HgCm8.1365$fL6.26905@news.cpqcorp.net>, "labadie" <labadie_g@decus.fr> writes: > 6 > "JMK" <jeffrey.m.klopotic@lmco.com> wrote in message# > news:3C994225.9DA690B@lmco.com...fJ >> Given 50 workstations on clusters scattered over an are larger than youF >> can walk comfortably in a day.  You reboot one or more clusters.  A  A > def/table=lnm$cluster_table 'f$getsyi("nodename")_start_ok truetK > This can also be put after a test (if appli1_ok and appli2_ok... then...)e >  > After a reboot, ae) > sh lo/table=lnm$cluster_table *start_ok 
 > on any node ' > should show all your Cluster members.o  > The original problem statement (above) was for doing it across; multiple clusters without explicitly visiting each cluster.r   ------------------------------  % Date: Fri, 22 Mar 2002 07:15:58 -0500o2 From: rdeininger@mindspring.com (Robert Deininger)P Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of LinuxJ Message-ID: <rdeininger-2203020715580001@1cust56.tnt1.nashua.nh.da.uu.net>  9 In article <u9kairbebmln92@corp.supernews.com>, GreyClouds <mist@cumulus.com> wrote:!  , >Another question: why did they have to use H >proprietary ram simm packs instead of the off-the-shelf type of simms?? >r  E I don't know anything about that particular Sun system, but the usualaF reason for using "nonstandard" memory is to improve performance and/orE reliability.  Or occasionally to pack more memory into limited space.h  L >So far, the sales desk down in San Jose can't seem to answer any technical K              ^^^^^                                                ^^^^^^^^^e   >questions about anything.     And you are surprised?  * >They say they can't help me and then its 	 ><plonk>.t  4 Are you sure you didn't call Compaq by mistake?  :-)  % >Or am I talking to the wrong people?   H Likely.  These days, if you want competent technical sales folks in mostI any company, you have to jump through hoops, and show them your checkbook H first.  The checkbook needs to be the wide kind, to allow filling in big numbers.   >I wanted to buy a sun but  . >your people have made this almost impossible.  ; Aw, c'mon.  Buy an alpha system instead.  It sounds easier.r   ------------------------------  % Date: Fri, 22 Mar 2002 16:06:42 +0000 T From: Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com>P Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux& Message-ID: <3C9B5692.1070603@sun.com>   Rob Young wrote:   > In article <3C986D1C.2040805@sun.com>, Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com> writes:9 > C >>>"Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in messages0 >>>news:0vol8.1113$fL6.23977@news.cpqcorp.net... >>>l >>>iD >>>>Remember, the new Sun keyword is "commercial" UNIX as opposed toM >>>>"proprietary" UNIX ;-).  They'll try to find ways to belittle it and tellnN >>>>you why it isn't ready-for-prime-time.  I think I hear Rome burning in the! >>>>background as Andrew fiddles.e >>>> >>>> >>>> >>F >>Fred it would be better if you posted about something you understoodE >>rather than something you clearly don't. Perhaps you should confines8 >>yourself to alt.folklore or alt.alternative.universes. >>B >>However in the real world Linux isn't ready for primetime as an  >>enterprise class OS.   >> >  > F > 	Oh?  Is that from "talking points" on company literature or is that > 	real world? >       : An unfamiliar territory I know for you Rob but real world.? As I said in my post I am working with a team who are trying tot= harden Linux to meet their internal requirements. None of thet: distributions do so off the shelf. Considerable effort has9 been expended to do this. Neither of their two commerciall UNIX's require this.  9 This is before you understand just how modest a role theyw4 intend for linux, no DBMS's no mission critical apps6 no clustering, no medium sized or large systems. It is5 almost identical to the role that Win2000/NT delivers  in their current environment.      > Exec Summary:m > K > big sweep of its computer operations, replacing 60 different servers fromnM > different vendors with a single computer having 500 virtual servers runningt > Linux. > ; > http://www.open-mag.com/features/Vol_15/sonera/sonera.htm. > P > The deal is not the first for IBM in the telecom sector. Before Sonera Entrum,P > there was Sweden's telecom giant, Telia, that declared it was ready to go withO > virtual server approach. In December 2000, IBM and Telia announced that TeliauK > would create1,500 virtual Linux machines on an IBM mainframe with "Shark"eQ > storage to run Internet and IP-VPN service operations across Scandinavia. StoryiQ > has it that Telia was briefed by IBM at a Linux integration center and made theC0 > decision in a matter of days after the visit.  > O > The Sonera/Telia deal numbers are very similar. Telia replaced 70 web-hostingnO > Unix servers with 1,500 virtual Linux servers. Both are consolidating using a P > single IBM S/390/G6 (zSeries) server in conjunction with a Shark storage unit.Q > What's more, the motivation for both telecoms to consolidate was not to achieve I > cost savings alone, but high availability and reliability assurance fore
 > customers.   >     D Sorry it may be a reference but this does not mean that it is a goodA idea or that they couldn't have done it cheaper, quicker and witht2 less risk on commercial UNIX or even Linux on x86.  > Linux on a mainframe does not make commercial sense unless youA allready own the VM licenses and are paying the annual maitenancei@ on them. Even then given the option of turning the mainframe offE to remove the VM fees will pay for the equivalent UNIX infrastructureXC in a very short period of time. This is before you add in the otheri( feeding an housing costs of an Z series.    A Each virtual Linux server is a single point of management, do yous; seriously think that replacing 70 points of management with(9 1500 is a going to deliver cost reductions or reliabilityt
 improvements.p  < We have ISP's who host thousands of virtual WEB servers on a< single system, IBM seem to have managed to get a customer toB beleive that hosting thousands of virtual WEB servers on thousands= of separate systems. A wonderfull sales job but not something % that I would be personnally proud of.o  < Incedentally you seem to have caught the same "boosting IBM"0 illness thats effecting Freddy is is contageous.       > [snip] > P > Small wonder, then that Sonera Entrum made its move and announced in October aK > big sweep of its computer operations, replacing 60 different servers fromhM > different vendors with a single computer having 500 virtual servers running P > Linux. The vendors who gained from this decision: IBM impressed Sonera Entrum,N > with the promise of rock-solid and high performing hardware, along with SuSE; > Linux Enterprise Server and other Open Source solutions. i >  > * >>Apart from anything it does not scale to >>much more than 4 CPU's.  >> > % > 	In most cases, it doesn't need to.e >     - Except that Sun, HP and IBM make a  huge % of 2 their revenues and profits from machines with more2 than 4 CPU's. Compaq the vendor in whose choir you2 warble has the smallest stake in this market hence your interest in Linux.s    I > 	Won't kill it off?  Of course, you are leaving out the new terminologynB > 	to help in thinking.  "Linux is for edge computing."  Guess theD > 	folks at Sonera/Telia are rare exceptions with few to follow.  Or; > 	more likely, more than a few are knocking on IBM's door.r    < There you go again IBM boosting have you deserted OpenVMS so quickly ????   Regardsv   Andrew Harrisono   ------------------------------    Date: 22 Mar 2002 10:49:32 -0600+ From: young_r@encompasserve.org (Rob Young) P Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux3 Message-ID: <j1OmDn1cBSld@eisner.encompasserve.org>F  } In article <3C9B5692.1070603@sun.com>, Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com> writes:h >  >  > Rob Young wrote: >  >> In article <3C986D1C.2040805@sun.com>, Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com> writes: >> ?D >>>>"Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message1 >>>>news:0vol8.1113$fL6.23977@news.cpqcorp.net...m >>>> >>>>E >>>>>Remember, the new Sun keyword is "commercial" UNIX as opposed tosN >>>>>"proprietary" UNIX ;-).  They'll try to find ways to belittle it and tellO >>>>>you why it isn't ready-for-prime-time.  I think I hear Rome burning in thet" >>>>>background as Andrew fiddles. >>>>>T >>>>>i >>>>>i >>> G >>>Fred it would be better if you posted about something you understoodeF >>>rather than something you clearly don't. Perhaps you should confine9 >>>yourself to alt.folklore or alt.alternative.universes.- >>>-C >>>However in the real world Linux isn't ready for primetime as an . >>>enterprise class OS.  >  >  >> Exec Summary: >> eL >> big sweep of its computer operations, replacing 60 different servers fromN >> different vendors with a single computer having 500 virtual servers running	 >> Linux.  >> G< >> http://www.open-mag.com/features/Vol_15/sonera/sonera.htm >> vQ >> The deal is not the first for IBM in the telecom sector. Before Sonera Entrum,hQ >> there was Sweden's telecom giant, Telia, that declared it was ready to go withgP >> virtual server approach. In December 2000, IBM and Telia announced that TeliaL >> would create1,500 virtual Linux machines on an IBM mainframe with "Shark"R >> storage to run Internet and IP-VPN service operations across Scandinavia. StoryR >> has it that Telia was briefed by IBM at a Linux integration center and made the1 >> decision in a matter of days after the visit. C >> iP >> The Sonera/Telia deal numbers are very similar. Telia replaced 70 web-hostingP >> Unix servers with 1,500 virtual Linux servers. Both are consolidating using aQ >> single IBM S/390/G6 (zSeries) server in conjunction with a Shark storage unit.eR >> What's more, the motivation for both telecoms to consolidate was not to achieveJ >> cost savings alone, but high availability and reliability assurance for >> customers.  >> e >  > F > Sorry it may be a reference but this does not mean that it is a goodC > idea or that they couldn't have done it cheaper, quicker and with.4 > less risk on commercial UNIX or even Linux on x86. > @ > Linux on a mainframe does not make commercial sense unless youC > allready own the VM licenses and are paying the annual maitenancemB > on them. Even then given the option of turning the mainframe offG > to remove the VM fees will pay for the equivalent UNIX infrastructurebE > in a very short period of time. This is before you add in the other * > feeding an housing costs of an Z series. >  > C > Each virtual Linux server is a single point of management, do youD= > seriously think that replacing 70 points of management withs; > 1500 is a going to deliver cost reductions or reliability, > improvements.b >   @ 	Yes it is an improvement.... because you are dealing with less G 	hardware, cabling, networking infrastructure.  The management side of eA 	things these days is mostly a cinch to automate.  Cookie cutter  ? 	instances of Linux.  By the way, the concept is often referred B 	to as "Server consolidation."  In this case 1 to 1 consolidation,? 	or 1 to 2.  Can you consolidate servers if the server count is F 	acutally increased?  Sure.  If the servers live inside less hardware.  9 	So maybe more apropos would be "Hardware Consolidation."n  > > We have ISP's who host thousands of virtual WEB servers on a> > single system, IBM seem to have managed to get a customer toD > beleive that hosting thousands of virtual WEB servers on thousands? > of separate systems. A wonderfull sales job but not somethingv' > that I would be personnally proud of.h >   ? 	Sure... because servers crash on a daily basis when the serverh? 	count is that high.  You have a single kernel underneath those D 	thousands of virtual webservers.  When (not if) that kernel panics,H 	you have lost them all.  In the Linux model on the mainframe, you couldF 	have 10 crashes a day and effect less than 1% of your signed on users! 	if hosting 1500 Linux instances.y  > > Incedentally you seem to have caught the same "boosting IBM"2 > illness thats effecting Freddy is is contageous. >   B 	Ah... no.  I have been talking up Linux off line for over 2 years* 	now and have the email trail to prove it.  6 	A very good overview of the strategy (and part of theA 	reason for my buy-in) is found in this excellent NY Times piece t? 	(Librul that it is , you still have to enjoy it occasionally):r  F http://www.nytimes.com/library/tech/00/03/biztech/articles/20soft.html  A 	Contagious?  I'll tell you what is contagious... daring thinkingnC 	and putting your money where your mouth is.  IBM is not stale, forh 	sure!   >  > 	 >> [snip]  >> oQ >> Small wonder, then that Sonera Entrum made its move and announced in October aeL >> big sweep of its computer operations, replacing 60 different servers fromN >> different vendors with a single computer having 500 virtual servers runningQ >> Linux. The vendors who gained from this decision: IBM impressed Sonera Entrum,fO >> with the promise of rock-solid and high performing hardware, along with SuSEb< >> Linux Enterprise Server and other Open Source solutions.  >> u >> p+ >>>Apart from anything it does not scale ton >>>much more than 4 CPU's. r >>>  >> r& >> 	In most cases, it doesn't need to. >>   >  > / > Except that Sun, HP and IBM make a  huge % ofl4 > their revenues and profits from machines with more4 > than 4 CPU's. Compaq the vendor in whose choir you4 > warble has the smallest stake in this market hence > your interest in Linux.a >  >    	okay.. whatever that means.    J >> 	Won't kill it off?  Of course, you are leaving out the new terminologyC >> 	to help in thinking.  "Linux is for edge computing."  Guess therE >> 	folks at Sonera/Telia are rare exceptions with few to follow.  Orb< >> 	more likely, more than a few are knocking on IBM's door. >  > > > There you go again IBM boosting have you deserted OpenVMS so > quickly ???? >   < 	Not at all.  VMS is by far superior in most respects to allC 	other pretenders.  IBM management, has a lot of the same qualities < 	that VMS engineering engenders/endears ... high quality and
 	superiority.d   				Robc   ------------------------------  % Date: Fri, 22 Mar 2002 16:20:04 +0000HT From: Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com>P Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux& Message-ID: <3C9B59B4.8050904@sun.com>   Bill Todd wrote:  < > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3C99307F.2C879034@videotron.ca... >  > ...l >  > @ >>Will Linux really be able to eventually match the reliability, >> > N > Possibly.  That's the contention of the open-source movement (many eyes makeM > for more robust code), and there seems to be at least some evidence to thathL > effect if one waits for releases to settle significantly before committingM > to them (e.g., does anyone know whether any 2.4.x releases have yet reachedb8 > the stability of the later ones in the 2.2.x series?). >     F Depends where you think all the eyes are looking. Much is made of the H 1000's of programmers working on Linux. Most are not working directly onG the kernel, they are working on GNU, GNOME, Mozilla, KDE, Apache, SambahF etc only a comparatively small minority are working on the kernel evenE smaller if you exclude the people writing device drivers. At the samenI time intervals between Linux Kernel releases have got longer as the code n4 base has increased and the test matrix has expanded.    
 >  support > K > Possibly.  Many hands, this time - far, far more than are involved in theiM > development of VMS, but then you have the problem that from the outside (atdN > least while you're getting your feet wet and haven't learned the ropes) it'sI > a lot harder to know who's worth talking with and who isn't (save for avM > relatively small number of well-known people who are likely far too busy toY > help). >     G Again it depends on what you think Linux is, lots of people are workingeE on GNU, Gnome etc. This code happens to run on the Linux kernel as itm@ does on the commercial UNIX kernels. A much smaller minority are doing kernel engineering.s    H > So corporations may feel more comfortable dealing with commercial UnixG > vendors with a paid support staff:  though there's nothing to preventyG > commercial support groups for Linux as well (and in fact I think theylL > already exist for at least some of the major distributions), the idea thatN > the people providing you with a fix are the same people who will incorporateM > it into the OS distribution itself may give them a warmer, fuzzier feeling.c > I > Of course, one dimension of support that's almost completely lacking in-K > commercial offerings is the ability to tweak the OS itself to do what youoL > need.  Likely suitable only in rare instances, but once in a while perhaps > worth a lot. >  >  and > H >>scalability of the commercial Unix systems ? (or VMS for that matter). >> >     F I recently attended a meeting with the customer I am working with, oneC of the main Linux protagonists there was asked this question by onem5 of his collegues, the answer he gave was interesting.r    F He said Linux is basically somewhere between SunOS (One Lock) and the C first release of Solaris 2.0 (Somewhat threaded) which supported 8 r: CPU's. A lot of work will need to be done to improve this.   Solaris is 9 years old.e    J > Probably not.  IIRC Linus is on record as not wanting to burden the vastG > majority of Linux users with the mechanism (e.g., fine-grained locks)oN > required to make Linux scale significantly to large SMP configurations - andL > that kind of change is sufficiently pervasive that expecting it to turn upI > in an alternative distribution patch may well be silly.  OTOH, Linux is K > *heavily* into clustering, and versions that cluster many 4 - 8 processorhJ > 'nodes' hardware-partitioned out of a much larger SMP box will likely beI > available, lacking mostly only the software-partitionability of Galaxy.n >     E Hence my customers intention to use it for WEB, small apps, file and p# print and some small compute farms.r     Regards-   Andrew Harrison-   ------------------------------  % Date: Fri, 22 Mar 2002 15:44:16 +0000sT From: Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com>O Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping ofLinuxu& Message-ID: <3C9B5150.7040403@sun.com>   Fred Kleinsorge wrote:  4 > Andrew Harrison SUNUK Consultancy wrote in message > <3C98B3EC.8040101@sun.com>...s > * >>Andrew Harrison SUNUK Consultancy wrote: >> > ? >>Funny you accuse me of being non technical and a market-droideG >>but you do so in a series of postings full of supposition, stupidity,aA >>factual errors and name calling. If thats what it takes to gain > >>the "acolade" of being branded technical by you then call me >>a sales guy. >> >> > + > You mean you aren't a sales/market droid?s >  >     5 You can call me what you like Freddy it doesn't alterb: the fact that your attempts at FUD have been full of basic: errors, suppostion and what seems very like bits of Compaq; marketing bulletins. The antithesis of a reasoned technicalr? argument. So far all you have demonstrated is a very poor graspa@ of a wide range of subjects from Linux, relative CPU performance: and what it implies to applications through your own Alpha- business model to Sun's SPARC business model.o    ? One attribute of technical people is generally an understandingt< of what they indevidually know and don't know. This facility: seems to have deserted you, it seems to have happened soon7 after the merger. I can only hope for your sake that itc; returns. In the mean time the send button is your enemy andu not your friend.  < As I said if what you have spouted in this thread passes for, technical argument then call me a sales guy.   Regardsl Andrew Harrison            >    ------------------------------  % Date: Fri, 22 Mar 2002 13:22:54 -0500 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>0O Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping ofLinuxi3 Message-ID: <wKKm8.1397$fL6.27216@news.cpqcorp.net>e  2 Andrew Harrison SUNUK Consultancy wrote in message <3C9B5150.7040403@sun.com>...  >t= >As I said if what you have spouted in this thread passes fort- >technical argument then call me a sales guy.  >g  I Andy, I've always said you were a putz posing as someone technical.  YourtJ knowledge is gleaned from the latest marketing flyers and surfing the web.I You've passed up every opportunity to trot out your qualifications.  Haves  you ever written a line of code?  K Now let's get back to brass tacks.  I don't believe for a minute that LinuxiL is ready-for-prime-time in the enterprise today.  On the other hand, SolarisI isn't something to write home about either.  You want something ready for I the enterprise today - use VMS.  My note was in fact a gentle rattling ofGE your chain, and foundation - that somehow Linux will never be seriousm$ competition in the enterprise space.  F History has shown that marketing and the lemming effect, not technicalH quality and capabilities, very often wins the day.  Otherwise, Sun wouldJ have ended up in the dustbin of history.  Right now, Linux has the big mo'D going for it.  It doesn't matter what the *truth* is, marketing will2 convince people they need to get on the bandwagon.  L 9-10 years ago, I probably would have belittled Solaris (or SunOS) as a pileD of steaming crap, not worthy of considering as serious competition -J something only capable of being used as a techno-weenie workstation.  JustJ as you are now poo-pooing Linux.  The problem is that instead of needing 9F years to "evolve" like Solaris did, Linux will probably evolve to be a7 serious Enterprise threat in perhaps another 1-2 years.o  I Sparc is an industry joke as a perennial underperformer.  You can try andoF sell cheap, slow, old generation chips at the low end.  You think thatL competes with an IA32 system?  I'd take a sub $1000 Linux IA32 system over aI sub $1000 Slowaris system any day - and be able to run dual boot Windows.iJ At the high-end, you wont be able to keep up with Power and IA64 - you canE try to keep throwing more-and-more of your mediocre processors onto al1 system, but there is only so far you can take it.   H The only real thing holding Linux back right now is some serious backingF from a few vendors.  Once that happens...  buh-bye.  Cause guess what.L Solaris is just-another-UNIX.  Just like Linux is really just-another-UNIX -H but one that has a chance at fulfulling the Sun snake oil "open" claims.  I So take the hint.  Be one of those vendors.  Transform Sun into the LinuxoE company.  Canablize Solaris and Sparc before they become dead weight.w   ------------------------------  % Date: Fri, 22 Mar 2002 13:42:57 -0500i- From: JF Mezei <jfmezei.spamnot@videotron.ca>oO Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping ofLinux , Message-ID: <3C9B7B30.AC2E7903@videotron.ca>   Fred Kleinsorge wrote:K > isn't something to write home about either.  You want something ready forw! > the enterprise today - use VMS.l  K But Mr Fred, your superiors keep telling the media that wintel servers ruleoJ the enterprise. If windows crap is ready for the enterprise, Shirley Linux
 would be too.s   ------------------------------   Date: 22 Mar 2002 04:18 CSTr' From: carl@gerg.tamu.edu (Carl Perkins)m Subject: Re: Hypersort- Message-ID: <22MAR200204180820@gerg.tamu.edu>   1 Didier Morandi <Didier.Morandi@free.fr> writes...u }What is HyperSort?v }D.n   See HELP SORT:  )      High-Performance SORT/MERGE Utility:r  B         On Alpha systems, you can also choose the high-performanceD         Sort/Merge utility. It uses the same command line interface.?         Any differences are noted with the appropriate SORT and E         MERGE qualifiers. Use the SORTSHR logical to select the high-yF         performance Sort/Merge utility. Define SORTSHR to point to theC         high-performance sort executable in SYS$LIBRARY as follows:i  2         $ DEFINE SORTSHR SYS$LIBRARY:HYPERSORT.EXE  B         To return to SORT/MERGE, deassign SORTSHR. (The SORT/MERGE:         utility is the default if SORTSHR is not defined.)    D There is, as far as I know, no single place in the on-line help thatI talles you the differences beteen it and the regular sort. Each qualifierfF that is affected has a blurb about how it is different with hypersort.  F As for relative preformance, you really need to try it on some of your, actual data to find out if it will help you.   --- Carl   }"John N." wrote:o }> mL }> After years of blood sweat and tears (mostly tears), our move from VAX toD }> Alpha is gaining momentum.  I was asked about defining SORTSHR to: }> SYS$LIBRARY:HYPERSORT.EXE but I cannot find much on ito   ------------------------------  % Date: Fri, 22 Mar 2002 11:46:38 +0100 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>  Subject: Re: Hypersort' Message-ID: <3C9B0B8E.4BAA0C2D@aaa.com>e  5 From the Release Notes for Rdb (all later versions) :>    @ Do not use the OpenVMS Alpha High Performance Sort/Merge utility	 (selectednA by defining the logical name SORTSHR to SYS$SHARE:HYPERSORT) when G using LogMiner for Rdb. HYPERSORT supports only a subset of the library  sortH routines that LogMiner requires. Make sure that the SORTSHR logical name is not defined to HYPERSORT.0    B 2.3.3 Do Not Use HYPERSORT with RMU Optimize After_Journal Command  G The OpenVMS Alpha V7.1 operating system introduced the high-performances@ Sort/Merge utility (also known as HYPERSORT). This utility takes advantage of theF OpenVMS Alpha architecture to provide better performance for most sort andn merge operations.iD The high-performance Sort/Merge utility supports a subset of the SOR	 routines.eG Unfortunately, the high-performance Sort/Merge utility does not supporte
 several ofA the interfaces used by the RMU Optimize After_Journal command. Inr
 addition, thetD high-performance Sort/Merge utility reports no error or warning when being calledC with the unsupported options used by the RMU Optimize After_Journaly command.F Because of this, the use of the high-performance Sort/Merge utility is nothG supported for the RMU Optimize After_Journal command. Do not define the 0 logical name SORTSHR to reference HYPERSORT.EXE.    H It's not 100% clear if Hypersort *is* supported if you just do *not* use# these two features in Oracle Rdb...t     Best Regards Jan-Erik Sderholm.i   ------------------------------  % Date: Fri, 22 Mar 2002 15:22:13 +0100N= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>e Subject: Re: Hypersort) Message-ID: <3C9B3E15.37EBBF5D@gtech.com>e   Didier Morandi wrote:lR > What is HyperSort? I aways thought that the VMS SORT processing was built on theS > QuickSort algorithm which was mathematically demonstrated to be the fastest ever.e   Hmmm.,  = Both Quicksort and Heapsort has the n*log(n) characteristics.o  ; In fact Heapsort is theoretical better, because it does nott. have the same worst case problem as Quicksort.  A But yes - all practical implementations of Quicksort and Heapsort - shows that Quicksort is faster than Heapsort.e  8 Note that theese analysis is based on an assumption that7 the data are being sorted in memory. If temporary filesh3 becomes necesarry, then it is a different scenario.h   Arne   ------------------------------  % Date: Fri, 22 Mar 2002 11:37:03 -0500o- From: "Peter Weaver" <peter.weaver@stelco.ca>s Subject: Re: Hypersort- Message-ID: <%4Jm8.729$je5.28@nnrp1.uunet.ca>a  . "John N." <JNixon@cfl.rr.com> wrote in message9 news:Newm8.255436$TV4.34620903@typhoon.tampabay.rr.com... K > After years of blood sweat and tears (mostly tears), our move from VAX toaC > Alpha is gaining momentum.  I was asked about defining SORTSHR to 9 > SYS$LIBRARY:HYPERSORT.EXE but I cannot find much on it.e >lK > What benefits will I get?  What are the penalties?  Common sense seems tonI > indicate it will use more memory and run faster.  Are there any specialyK > sysgen considerations?  Does the improvement only effect large files?  IseL > there any effect when sorting small files?  Lots of small files?   Is more! > disk work space requred?  Less?s >h# > Thanks for knowledgable comments.. >  >   K In a previous life I recall being asked the same thing. One of the analystsvI insisted that we had to define the SORTSHR logical in SYLOGICALS since itdJ was so much better than the default, he couldn't figure out why DEC didn'tF make it the default and thought that we should. I refused to add it to9 SYLOGICALS until he could prove that nothing would break.s  J That analyst and one programmer did some reading and some testing over theK next 4 days and found that in 99% (or some number very close to 99%) of oursJ sorts the was no noticeable difference between doing a simple SORT commandI vs. using a DEFINE and a SORT command. Most of the other 1% broke becauseaJ they were using a feature not supported by Hypersort. In the end there wasH only one sort in one application (which was not owned by the analyst whoL started the whole project) that was large enough that using Hypersort made aL difference. The analyst who owned the application joked that since HypersortL cut 5 minutes off of a monthly job that we would save 1 hour of processing aH year. So the ROI would take 64 years. :( He never bothered to change his/ code to add the DEFINE command before the sort.   . After all that, the only advise I can give is;) 1. Do not define the logical system wide.tL 2. Let the programmers/analysts or whoever you have test each sort that they
 want to test. ; 3. Don't expect big returns unless you do a lot of sorting.o   -- Peter WeaverL Opinions are my own, and do not reflect the opinions of my employer, nor theK company that it sub-contracts to, nor the company that it sub-contracts to.r   ------------------------------  % Date: Fri, 22 Mar 2002 09:17:33 +0100>) From: Bart Zorn <B.Zorn@xs4all.nospam.nl> # Subject: Re: LD062 Install Questiono/ Message-ID: <3C9AE89D.6080808@xs4all.nospam.nl>    David J. Dachtera wrote:   > > > Yes - second the motion, taking a couple of queues from ZIP: > 7 > $ BACKUP/FIX_SAVESET[=FULL] saveset_filespec/SAVE_SET  > E > Repair the RMS attributes of a saveset (maybe). ["=FULL" means "try	
 > harder"] > 1 > $ BACKUP/TEST_SAVESET saveset_filespec/SAVE_SET1 > # > Check the integrity of a saveset.U > E > As usual, back-port the upgrades to V5.5-2 so no one gets left out.' >   I The second function is already present in Save Set Manager with the SAVE 9 VALIDATE command.o  	 Bart Zornd   ------------------------------  % Date: Fri, 22 Mar 2002 15:41:06 +0100 = From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>i# Subject: Re: LD062 Install Question ) Message-ID: <3C9B4282.2894066D@gtech.com>    Tom Linden wrote:nF > Just attempted to install LD062 on AXP running 7.3.  When the kit is
 > unzipped ittM > produces LD062.[A-K] but running VMSINSTALL only restores savesets A and E.t	 > Is thisa
 > correct?   Yes.  F I think A = install-script and the rest contains the binaries for each VMSe3 version supported - one saveset = one VMS version !r   Arne   ------------------------------    Date: 22 Mar 2002 09:59:04 -0800. From: SPAMSINK2001@YAHOO.COM (Alan E. Feldman)# Subject: Re: LD062 Install Questionh= Message-ID: <343f30ae.0203220959.269dfc8c@posting.google.com>   ` "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3C9AA4B1.A7F64B02@fsi.net>... [...]p> > Yes - second the motion, taking a couple of queues from ZIP: > 7 > $ BACKUP/FIX_SAVESET[=FULL] saveset_filespec/SAVE_SETh > E > Repair the RMS attributes of a saveset (maybe). ["=FULL" means "tryb
 > harder"] > 1 > $ BACKUP/TEST_SAVESET saveset_filespec/SAVE_SETi > # > Check the integrity of a saveset.     C You can use BACKUP/LIST to check the integrity of a save set. Check  out this article:t  
 [begin quote]s Search Result 1 ' From: Jerry Leichter (leichter@lrw.com) 9 Subject: RE: How To Verify Integrity of A BACKUP SaveSet n Newsgroups: comp.os.vms 5 View: Complete Thread (9 articles) | Original Format   Date: 1993-03-26 23:05:35 PST   o [original poster's problem:]E One of the groups using the VAXcluster that I manage is creating tape C  backups of disks to send out to our new sites.  Once the tapes aregF  created, they want to be able to verify that the tape is: 1) readableC  and 2) contains valid data.  The best solution I have been able tosF  come up with is to restore the entire tape to another disk, using the=  /Image qualifier (which was used when the tape was created).e   [Jerry's response:]tC A simple BACKUP/LIST verifies the contents of the save set.  Here'sl proof:C I created an on-disk saveset named FOO.A, then deliberately damagedw it.  Now watch:    $ back/list foo.a/sav  Listing of save set(s)h  s  Save set:          FOO.Ai   Written by:        LEICHTER    #  UIC:               [000110,000051]w+  Date:              26-MAR-1993 11:36:33.82r7  Command:           BACKUP/NOASSIST *.*; FOO.A/SAVE/LOG (  Operating system:  VAX/VMS version V5.2  BACKUP version:    V5.2  ...A  [LEICHTER.RUTGERS.CLASSES.315]MID93.DVI;7        23  24-MAR-1993c 20:52 A  [LEICHTER.RUTGERS.CLASSES.315]MID93.LIS;7         3  24-MAR-1993k 20:52 A  [LEICHTER.RUTGERS.CLASSES.315]MID93.LTX;14       24  26-MAR-1993i 07:57 - >>>> 1 error recovered by redundancy group inS3 >>>>   $DISK1:[LEICHTER.RUTGERS.CLASSES.315]FOO.A;10  o  Total of 21 files, 103 blocks  End of save set   F Before anyone comments that BACKUP only did this because I damaged the inter-E nal record structure, or something of that sort:  No, I was careful.   IaC changed one byte in the middle of the data from a file.  It was not 	 even at atB record boundary, not that that would matter.  In fact, here's what DIFFERB reports (after I've used FILE to set the the recordsize to 16 - to keep theC output to a reasonable length - in both FOO.A and FOO.B, which is a  savedy copy of FOO.A):,  2  File $DISK1:[LEICHTER.RUTGERS.CLASSES.315]FOO.A;1  s2  RECORD NUMBER 71 (00000047) LENGTH 16 (00000010)   l=   72612079 62207564 652E7372 3F677475 utg?rs.edu by ar 000000t  a2  RECORD NUMBER 72 (00000048) LENGTH 16 (00000010)   n=   7564652E 73726567 7475722E 73696D61 amis.rutgers.edu 000000l  ******f2  File $DISK1:[LEICHTER.RUTGERS.CLASSES.315]FOO.B;1  -2  RECORD NUMBER 71 (00000047) LENGTH 16 (00000010)   0=   72612079 62207564 652E7372 65677475 utgers.edu by ar 000000m  _2  RECORD NUMBER 72 (00000048) LENGTH 16 (00000010)   L=   7564652E 73726567 7475722E 73696D61 amis.rutgers.edu 0000003
  ************@  '  Number of difference sections found: 1r&  Number of difference records found: 1  a4  DIFFERENCES /IGNORE=()/MERGED=1/MODE=(HEXADECIMAL)-2      $DISK1:[LEICHTER.RUTGERS.CLASSES.315]FOO.A;1-1      $DISK1:[LEICHTER.RUTGERS.CLASSES.315]FOO.B;1e   [original poster:]F  I'm wondering if anyone has suggestion(s) as to how we can verify theF  contents of these tapes without having to allocate another whole diskC  to that group?  Is there a way to send the info to the NL: device?    [Jerry:]D If you want to be really, really, sure, restore the backup to a disk withA /VERIFY.  That will tell you all you can reasonably hope to learne	 about theb integrity of the saveset.d  * Me, I've found I can trust in BACKUP/LIST.        -- Jerrya   [end quote]r   --   Dislcaimer: JMHO Alan E. Feldmanu" afeldman atski gfigroup dotski com3 "Help me help you help me help you" --Bob Pattersone   ------------------------------  # Date: Fri, 22 Mar 2002 18:27:54 GMTp' From: Rick Dyson <Rick-Dyson@UIowa.EDU>s' Subject: Re: lpd printing + telnetqueuep) Message-ID: <3C9B77AA.6C24577C@UIowa.EDU>a   > > printer queue created byI > > INITIALIZE/QUEUE/PROCESSOR=TCPIP$TELNETSYM/on="ipadd:9100" PBEPQPTESTe  C 	Is it possible to use the Telnet Symbiot with TCPIP (or UCX?) withn1 the device name having a DNS defined name?  I.e.,n  ? 		Init/queu/proc=tcpip$telnetsym/on=VMS::"node.domain:9100" xxxt  I the node.domain is resolved from a remote DNS server, not on the VMS noden1 or from the static host table TCPIP can maintain.-  $ 	The error I always get for this is:  N %QUEMAN-F-INVQUAVAL, value 'VMS::"node.domain:9100"' invalid for /ON qualifier  E 	If I use the specific IP address instead of the Internet address, its works:  ; 		Init/queu/proc=tcpip$telnetsym/on=VMS::"1.1.1.2:9100" xxxU   	Thanks for any help!s   Rick   ------------------------------  % Date: Fri, 22 Mar 2002 07:46:36 -0500i- From: JF Mezei <jfmezei.spamnot@videotron.ca>c$ Subject: Microsoft bullies PC makers, Message-ID: <3C9B27AA.B1A62A63@videotron.ca>  8 http://www.nytimes.com/2002/03/22/technology/22SOFT.HTML  5 Ex-Gateway Executive Says Microsoft Bullies PC Makersy   By BLOOMBERG NEWSo  ' WASHINGTON, March 21 (Bloomberg News) iM The Microsoft Corporation (news/quote) bullies computer makers by withholding H discounts if they promote products that threaten its Windows monopoly, aN former Gateway Inc. (news/quote) executive testified today on behalf of states3 seeking tough antitrust remedies against Microsoft.   I  The executive, Peter Ashkin, said that the Justice Department's proposed L settlement of the four-year-old case would not stop Microsoft from punishingK computer makers like Gateway that develop rival products. The states calledtM Mr. Ashkin to testify on the fourth day of what is expected to be eight weeksu# of court hearings on their demands.o  N In 1999 and 2000, Microsoft reduced discounts Gateway received for the WindowsK operating system when Gateway developed an Internet-connection device and asE network computer powered by the rival Linux system, Mr. Ashkin  said.s  I   "Our decision to choose Linux over Windows was often cited by MicrosoftnM executives when explaining why they considered Gateway to be a `bad partner,'pO " Mr. Ashkin, who  was dismissed by Gateway in 2001, said in written testimony.e  I  The proposed settlement was negotiated last year after an  appeals courtnH ruled that Microsoft illegally protected its monopoly for Windows, whichI powers 95 percent of the world's personal computers. California, Iowa and L seven other states, along with the District of Columbia, refused to sign theL plan that would require Microsoft to give computer makers freedom to promote. certain types of products without retaliation.  E  The settlement would not prevent "Microsoft from retaliating in thisiP situation," said Mr. Ashkin, now an executive with AOL Time Warner (news/quote).  L  Judge Colleen Kollar-Kotelly is reviewing the settlement and the oppositionM by the states. She got the case after the appeals court overturned an earlierg% judge's order to break up Microsoft. X  E  Microsoft says the standard Windows contract for the top 20 computercL manufacturers under the settlement would free Gateway and other companies toN promote programs that compete with the Microsoft Internet Explorer Web browser or the Windows Media Player.  N  A Microsoft lawyer, Richard Pepperman, questioned Mr. Ashkin's assertion thatM Microsoft once  threatened to end its license to load Windows onto PC's if itnJ did not agree to continue selling the popular Microsoft Office set of word$ processing and spreadsheet programs.  N  Was there "any paper trail" documenting the threats? Mr. Pepperman asked. Mr.F Ashkin replied that he  did not write memos or e-mail messages to warn colleagues of the threat.t  N  The states want to force Microsoft to make a version of the Windows operatingN system that would let computer makers easily extract the Internet Explorer WebB browser and other applications and substitute  competing programs.   ------------------------------  % Date: Fri, 22 Mar 2002 20:13:30 +1100r. From: Burnie M <burniem.NOSPAM@ozemail.com.au>A Subject: Re: MONI SYS shows 80% CPU usage and 8 jobs in COM state 8 Message-ID: <ratl9ugednucrhkt1gukefl8f33n6pnjms@4ax.com>  - On Wed, 20 Mar 2002 11:42:04 -0500, "Syltrem"r' <syltremspammenot@videotron.com> wrote:   > >>Your load is about 80%.  So your average queue length should3 >>be around 5.  And that fits with what you report.  >iJ >So you find it normal that I have a queue? My understanding was that if II >have a queue, then I have people waiting to be served. If I don't have aRL >queue, I only have jobs waiting for IO etc and one CURrently executing. AndJ >that's good as it means my CPU is fast enough to keep up with the demand. >sK >Reducing QUANTUM would most probably make the queue shorter but would alsos@ >reduce actual throuput from the machine (although it may not beM >perceptible). Since interactive jobs have good response time, and only batchoK >jobs take longer to execute I don't think it's worth changing QUANTUM now.a9 >Also that problem only occurs at periods during the day.     E Rule of thumb is that up to 4 processes computable per CPU is fine ass long as they are not COMO.   ------------------------------    Date: 22 Mar 2002 08:59:42 -00004 From: Doc.Cypher <Use-Author-Address-Header@[127.1]>3 Subject: Re: Once a fool, always a fool I guess ...-6 Message-ID: <20020322085942.27771.qmail@gacracker.org>  > On 21 Mar 2002, Kilgallen@SpamCop.net (Larry Kilgallen) wrote:B >In article <20020321204945.10274.qmail@gacracker.org>, Doc.Cypher, ><Use-Author-Address-Header@[127.1]> writes:= >> On 21 Mar 2002, leslie@clio.rice.edu (Jerry Leslie) wrote:o4 >>>Paul Winalski (prune@ZAnkh-Morpork.mv.com) wrote:I >>>: It's not really that VMS is unhackable--it's more that it representsRE >>>: such a miiniscule part of the overall computer marketplace theser0 >>>: days that nobody bothers to try to hack it. >>>: >>>i5 >>>The best hackers in the country tried to hack VMS:  >>>p/ >>>  http://www.pointsecure.com/Defconwhite.pdf $ >>>  "Virtually Unhackable" DEFCON9: >> eN >> Compaq seems quite keen on having the "secure" tag attached to VMS with webL >> servers and browsers, shouldn't they have a system up and running with anL >> open challenge to hack it and a prize for doing so? It certainly wouldn'tL >> be expensive to set up such a challenge. Such would also gather some free' >> publicity for the operating system. a >f >They had that at DEFcon.e  2 Yes, but it wasn't an official Compaq team was it?  F Apparently Sun felt it was important enough to run their own team, butE Compaq just continue to demonstrate an uncaring attitude towards VMS.e     Doc. -- m6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                              http://vmsbox.cjb.neta   ------------------------------    Date: 22 Mar 2002 09:17:16 -0000= From: Doc.Cypher <Use-Author-Supplied-Address-Header@[127.1]>j3 Subject: Re: Once a fool, always a fool I guess ...t6 Message-ID: <20020322091716.28367.qmail@gacracker.org>  @ On Fri, 22 Mar 2002, "Dale King" <dalek@forpresident.com> wrote:9 >"Rob Young" <young_r@encompasserve.org> wrote in messageu. >news:Who+Lm+FuT4Q@eisner.encompasserve.org...? >> No more than I would have been impressed if someone had heldIA >> a gun to his head and got in.  Punk moves don't count.  If youc( >> have to resort to scumbag techniques: >>2 >>                              whoop-dee-doo-wah!L >Noone held a gun to anyone's head, that is exactly why these sort of social$ >engineering attacks ARE impressive. >tO >1: We OpenVMS (and yes, any OS for that matter) system managers are only humanm >and can be caught of guard. >rL >2: OpenVMS is still at a stage where passwords are sent over the network in >clear text. >aO >Early hackers oftentimes only had social engineering on their side.  Some good  >books in this area are: >r3 >Bruce Sterling's "The Hacker Crackdown" (get it at-F >http://www.eff.org/Publications/Bruce_Sterling/Hacker_Crackdown/) and >eL >Suelette Dreyfus' "Underground (get it at http://www.underground-book.com/) >e >Both worth the read.i >@O >Would you be prepared to put all of your VMS boxes on the internet to test thec >theory?  I I run one... vmsbox.cjb.net. Wanna try hacking it? Got any hacker friends J that want to give it a go? Hell, I even make it a little bit more easy forD people, I grant user accounts and host a copy of the VMS Hacking FAQ+ (http://vmsbox.cjb.net/VMS/vmsHackFAQ.txt).n  $ For information on user accounts see) http://vmsbox.cjb.net/shell-accounts.html   K Of course, the hacking community has little or no interest in VMS nowadays.sG There are many easy pickings elsewhere and VMS knowledge is thin on the?F ground there. I know, I asked alt.2600 if anybody wanted an account to" learn VMS - I got one application.     Doc. -- c6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                              http://vmsbox.cjb.net    ------------------------------  # Date: Fri, 22 Mar 2002 11:51:10 GMTaL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")3 Subject: Re: Once a fool, always a fool I guess ... 8 Message-ID: <00A0B4DD.44930ECD@SSRL04.SLAC.STANFORD.EDU>  T In article <a7dup2$r32$1@lore.csc.com>, "Dale King" <dalek@forpresident.com> writes:7 >"Rob Buxton" <rob.buxton@wcc.govt.nz> wrote in messagei+ >news:3c9a7a1b.71938081@news.wcc.govt.nz...a >>G >> No, it's part of the NAS license. Most or maybe even all Alphas cometH >> with a NAS license now. For some older systems these were additional,G >> but all of our old VAXes and Alphas seem to have some flavour of NASr >> license.a >aM >As I said, a TCPIP license does not come with VMS.  You end up paying for itpQ >separately to the OS whatever spin you put on it.  Then if you want SSH you haveuG >to buy ANOTHER IP stack if even if you've ended up with a NAS license.H  N Or install Dave Jones' SSH server, which we ran happily for over a year beforeO Process Software came out with one.  (Doesn't do SSH 2, which is a problem, buts$ is a whole lot better than nothing.)   -- Alano    O ===============================================================================n0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056 M  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210 O ===============================================================================a   ------------------------------    Date: 22 Mar 2002 05:42:03 -0800( From: bob@instantwhip.com (Bob Ceculski)3 Subject: Re: Once a fool, always a fool I guess ... < Message-ID: <d7791aa1.0203220542.d1b16b2@posting.google.com>  Y "Dale King" <dalek@forpresident.com> wrote in message news:<a7dqvt$ovg$1@lore.csc.com>...s, > <system@SendSpamHere.ORG> wrote in message, > news:00A0B44A.0BCE22D0@SendSpamHere.ORG...J > > In article <21MAR200206483608@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl >  Perkins) writes: 3 > > >"Dale King" <dalek@forpresident.com> writes...n; > > >}"Bob Ceculski" <bob@instantwhip.com> wrote in message = > > >}news:d7791aa1.0203201611.4943c0aa@posting.google.com...-6 > > >}> Analysts: Security flaws won't undermine LinuxM > > >}Wouldn't it be nice to have something like netfilter in the VMS kernel?l > > >o > > >No, it would not. > > >6M > > >The proper place for such a thing, if you want it, is not in the kernel.r > >gK > > Would you care to explain what this netfilter thing in the Linux kernelS > > does for one using Linux?A > Q > Netfilter is an advanced packet filter.  It lets you filter just about anythingtQ > IP, and allows you to do neat things like use your machine as a firewall or NAT 	 > router.  > A > If you want to know more, check out http://netfilter.samba.org/r > R > It is built into the kernel for speed, not security.  It is an age old debate asG > to whether this sort of thing should go into the kernel or userspace.a > Q > I think it's things like this that linux users get for free that stops VMS from.P > ever getting a look in.  VMS is very good at what it does: providing a secure,N > reliable platform for enterprise applications.  That is why I don't think weR > should dump on other OS's for being insecure, because they are trying to achieveF > something I don't think VMS ever will: to be on everybody's desktop.  0 I can already do that with TCPware ... big deal!   ------------------------------    Date: 22 Mar 2002 05:47:13 -0800( From: bob@instantwhip.com (Bob Ceculski)3 Subject: Re: Once a fool, always a fool I guess ...K= Message-ID: <d7791aa1.0203220547.6c97878f@posting.google.com>9  Y "Dale King" <dalek@forpresident.com> wrote in message news:<a7ebht$5j3$1@lore.csc.com>...m: > "Rob Young" <young_r@encompasserve.org> wrote in message/ > news:Who+Lm+FuT4Q@eisner.encompasserve.org...o@ > > No more than I would have been impressed if someone had heldB > > a gun to his head and got in.  Punk moves don't count.  If you) > > have to resort to scumbag techniques:  > > 3 > >                              whoop-dee-doo-wah!,M > Noone held a gun to anyone's head, that is exactly why these sort of socialJ% > engineering attacks ARE impressive.T > P > 1: We OpenVMS (and yes, any OS for that matter) system managers are only human > and can be caught of guard.  > M > 2: OpenVMS is still at a stage where passwords are sent over the network in 
 > clear text.  > P > Early hackers oftentimes only had social engineering on their side.  Some good > books in this area are:9 > 4 > Bruce Sterling's "The Hacker Crackdown" (get it atG > http://www.eff.org/Publications/Bruce_Sterling/Hacker_Crackdown/) and: > M > Suelette Dreyfus' "Underground (get it at http://www.underground-book.com/)  >  > Both worth the read. > P > Would you be prepared to put all of your VMS boxes on the internet to test the	 > theory?1  M funny, with TCPware and ssh/ssh2 I don't have to send password clear text ...    ------------------------------    Date: 22 Mar 2002 05:48:30 -0800( From: bob@instantwhip.com (Bob Ceculski)3 Subject: Re: Once a fool, always a fool I guess ...u= Message-ID: <d7791aa1.0203220548.1aa12e3f@posting.google.com>   Y "Dale King" <dalek@forpresident.com> wrote in message news:<a7eduj$6r0$1@lore.csc.com>... : > "Rob Young" <young_r@encompasserve.org> wrote in message/ > news:2uKa5O73vZJ$@eisner.encompasserve.org... > > > I take it you aren't using Multinet... SSH has been in use@ > > for a while.  They just rolled SSH2.  I use Multinet TCP/IP.O > Correct. UCX (or Digital/Compaq TCPIP Services) or whatever it will be calledF > tomorrow.R > E > > But I wouldn't mind putting one on there, configed like so in the5J > > following... and I would access it via SSH - always -.  If I couldn't,H > > or needed to access directly, I would unplug the ethernet connection* > > before using something other than SSH. > Q > Sounds like a good checklist.  Something like this should go into the "Guide to P > System Security", right after securing DECnet.  Security was something the OSI. > model was going to address for us wasn't it? > R > Something that just occurred to me: does anyone have any info on what the recentN > problems with Motif are?  I guess that would be vulnerability for VMS by the > sounds of the ECO info.G  D you could only exploit that bug if you were already a valid user ...   ------------------------------    Date: 22 Mar 2002 05:50:56 -0800( From: bob@instantwhip.com (Bob Ceculski)3 Subject: Re: Once a fool, always a fool I guess ...t= Message-ID: <d7791aa1.0203220550.19f1cf3e@posting.google.com>h  Y "Dale King" <dalek@forpresident.com> wrote in message news:<a7ebht$5j3$1@lore.csc.com>... : > "Rob Young" <young_r@encompasserve.org> wrote in message/ > news:Who+Lm+FuT4Q@eisner.encompasserve.org... @ > > No more than I would have been impressed if someone had heldB > > a gun to his head and got in.  Punk moves don't count.  If you) > > have to resort to scumbag techniques:u > >l3 > >                              whoop-dee-doo-wah!pM > Noone held a gun to anyone's head, that is exactly why these sort of social2% > engineering attacks ARE impressive.s > P > 1: We OpenVMS (and yes, any OS for that matter) system managers are only human > and can be caught of guard.  > M > 2: OpenVMS is still at a stage where passwords are sent over the network ind
 > clear text.S > P > Early hackers oftentimes only had social engineering on their side.  Some good > books in this area are:n > 4 > Bruce Sterling's "The Hacker Crackdown" (get it atG > http://www.eff.org/Publications/Bruce_Sterling/Hacker_Crackdown/) ande > M > Suelette Dreyfus' "Underground (get it at http://www.underground-book.com/)T >  > Both worth the read. > P > Would you be prepared to put all of your VMS boxes on the internet to test the	 > theory?   ? we are already doing that!  for 2 years now and no problems ...o   ------------------------------    Date: 22 Mar 2002 05:56:28 -0800( From: bob@instantwhip.com (Bob Ceculski)3 Subject: Re: Once a fool, always a fool I guess ...:= Message-ID: <d7791aa1.0203220556.4c464d3a@posting.google.com>u  Y "Dale King" <dalek@forpresident.com> wrote in message news:<a7dqct$ol8$1@lore.csc.com>...wA > "Robert Deininger" <rdeininger@mindspring.com> wrote in messageiG > news:rdeininger-2103020708470001@1cust225.tnt1.nashua.nh.da.uu.net... 7 > > In article <a7bk76$hiq$1@lore.csc.com>, "Dale King"n# > > <dalek@forpresident.com> wrote:  > >oN > > >Sure we can see VMS isn't susceptible to these attacks, but really, there >  isn'tN > > >a whole lot of this software functionality on VMS either.  Geez, we still >  havei > > >to pay for a TCP/IP stack!  > >TM > > Huh?  We've been though this many times in recent weeks.  TCP/IP serviceseI > > is bundled with most any new VMS system sale, and has been for years.p; > > There may be exceptions, but nobody has pointed to one.t > K > By bundled you mean in the Enterprise Integration Package right?  This ish > separate to a VMS license. > J > > >I would much rather see VMS get this sort of functionality than users >  braggingnJ > > >about how "unhackable" their OS is.  Windows/linux would be too if we
 >  didn't run  > > >anything on it ;) > > N > > So contact the TCP/IP services product manager at Compaq, or the folks whoN > > decide what to put in the products at Process Software, and tell them yourM > > needs.  That is the way new features get added to VMS.  Not via newsgroup2
 > > posts. > N > Example:  How long have we waited for a port of SSH?  That has been directlyQ > requested many I time.  I understand there is a good deal of engineering effortwD > to get this sort of stuff into VMS, and we will just have to wait.  D ssh has been on tcpware now for over a year ... where have you been?   ------------------------------   Date: 22 Mar 2002 14:28:08 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)t3 Subject: Re: Once a fool, always a fool I guess ...r, Message-ID: <a7ff1o$2b1i$1@info.cs.uofs.edu>  = In article <d7791aa1.0203220547.6c97878f@posting.google.com>,(+  bob@instantwhip.com (Bob Ceculski) writes:  |> vP |> funny, with TCPware and ssh/ssh2 I don't have to send password clear text ...  J Funny, with FreeBSD and telnet I don't send password clear text either ...     bill   -- yJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   e   ------------------------------    Date: 22 Mar 2002 09:17:16 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)n3 Subject: Re: Once a fool, always a fool I guess ...o3 Message-ID: <uFWRTP+g5zXI@eisner.encompasserve.org>i  m In article <20020322085942.27771.qmail@gacracker.org>, Doc.Cypher <Use-Author-Address-Header@[127.1]> writes: @ > On 21 Mar 2002, Kilgallen@SpamCop.net (Larry Kilgallen) wrote:C >>In article <20020321204945.10274.qmail@gacracker.org>, Doc.Cypher - >><Use-Author-Address-Header@[127.1]> writes:e> >>> On 21 Mar 2002, leslie@clio.rice.edu (Jerry Leslie) wrote:5 >>>>Paul Winalski (prune@ZAnkh-Morpork.mv.com) wrote:pJ >>>>: It's not really that VMS is unhackable--it's more that it representsF >>>>: such a miiniscule part of the overall computer marketplace these1 >>>>: days that nobody bothers to try to hack it.  >>>>:  >>>>6 >>>>The best hackers in the country tried to hack VMS: >>>>0 >>>>  http://www.pointsecure.com/Defconwhite.pdf% >>>>  "Virtually Unhackable" DEFCON9:@ >>> O >>> Compaq seems quite keen on having the "secure" tag attached to VMS with webdM >>> servers and browsers, shouldn't they have a system up and running with anmM >>> open challenge to hack it and a prize for doing so? It certainly wouldn'taM >>> be expensive to set up such a challenge. Such would also gather some freei( >>> publicity for the operating system.  >> >>They had that at DEFcon. > 4 > Yes, but it wasn't an official Compaq team was it? > H > Apparently Sun felt it was important enough to run their own team, butG > Compaq just continue to demonstrate an uncaring attitude towards VMS.b  A I doubt that corporate backing impresses the DEFCON constituency.    ------------------------------    Date: 22 Mar 2002 08:18:08 -0800( From: bob@instantwhip.com (Bob Ceculski)3 Subject: Re: Once a fool, always a fool I guess ...p= Message-ID: <d7791aa1.0203220818.4970204f@posting.google.com>e  Y "Dale King" <dalek@forpresident.com> wrote in message news:<a7dup2$r32$1@lore.csc.com>...y8 > "Rob Buxton" <rob.buxton@wcc.govt.nz> wrote in message, > news:3c9a7a1b.71938081@news.wcc.govt.nz... > >2H > > No, it's part of the NAS license. Most or maybe even all Alphas comeI > > with a NAS license now. For some older systems these were additional,aH > > but all of our old VAXes and Alphas seem to have some flavour of NAS > > license. > N > As I said, a TCPIP license does not come with VMS.  You end up paying for itR > separately to the OS whatever spin you put on it.  Then if you want SSH you haveH > to buy ANOTHER IP stack if even if you've ended up with a NAS license.  E so you would rather use insecure freeware then pay a little extra forT0 a secure environment?  You get what you pay for!   ------------------------------  % Date: Fri, 22 Mar 2002 13:26:58 -0500n5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> 3 Subject: Re: Once a fool, always a fool I guess ...s3 Message-ID: <lOKm8.1398$fL6.27040@news.cpqcorp.net>a  $ Larry Kilgallen wrote in message ...B >In article <20020322085942.27771.qmail@gacracker.org>, Doc.Cypher+ <Use-Author-Address-Header@[127.1]> writes:aA >> On 21 Mar 2002, Kilgallen@SpamCop.net (Larry Kilgallen) wrote: D >>>In article <20020321204945.10274.qmail@gacracker.org>, Doc.Cypher. >>><Use-Author-Address-Header@[127.1]> writes:? >>>> On 21 Mar 2002, leslie@clio.rice.edu (Jerry Leslie) wrote:o6 >>>>>Paul Winalski (prune@ZAnkh-Morpork.mv.com) wrote:K >>>>>: It's not really that VMS is unhackable--it's more that it representsBG >>>>>: such a miiniscule part of the overall computer marketplace these-2 >>>>>: days that nobody bothers to try to hack it. >>>>>: >>>>>y7 >>>>>The best hackers in the country tried to hack VMS:c >>>>>O1 >>>>>  http://www.pointsecure.com/Defconwhite.pdf & >>>>>  "Virtually Unhackable" DEFCON9: >>>>L >>>> Compaq seems quite keen on having the "secure" tag attached to VMS with webnK >>>> servers and browsers, shouldn't they have a system up and running withe anE >>>> open challenge to hack it and a prize for doing so? It certainlyI wouldn'tI >>>> be expensive to set up such a challenge. Such would also gather somer free( >>>> publicity for the operating system. >>>  >>>They had that at DEFcon.N >>5 >> Yes, but it wasn't an official Compaq team was it?  >>I >> Apparently Sun felt it was important enough to run their own team, buteH >> Compaq just continue to demonstrate an uncaring attitude towards VMS. >tB >I doubt that corporate backing impresses the DEFCON constituency.  K It also woke up a lot of people.  It would not suprise me to see it be dones officially next time.e   ------------------------------  % Date: Fri, 22 Mar 2002 13:35:18 -0500H5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>c3 Subject: Re: Once a fool, always a fool I guess ...e3 Message-ID: <aWKm8.1399$fL6.27260@news.cpqcorp.net>e  $ Bill Gunshannon wrote in message ...( >In article <a7dgih$sp3$1@joe.rice.edu>,- > leslie@clio.rice.edu (Jerry Leslie) writes:h5 >|> Paul Winalski (prune@ZAnkh-Morpork.mv.com) wrote:hJ >|> : It's not really that VMS is unhackable--it's more that it representsF >|> : such a miiniscule part of the overall computer marketplace these1 >|> : days that nobody bothers to try to hack it.b >|> :S >|>a6 >|> The best hackers in the country tried to hack VMS: >|>v0 >|>   http://www.pointsecure.com/Defconwhite.pdf% >|>   "Virtually Unhackable" DEFCON9:n >eA >It took hackers years of work to find the holes in Unix (most ofSA >which have fixes available within hours) and because this wasn'ti? >duplicated on VMS in a week you think VMS is un-hackable??  It.? >really is time people came to the realization that the primaryiB >thing keeping VMS un-hackable is it's not being readily available@ >to hackers due to a very limited presence in the real world.  AD >presence I fear is going to decrease over the next few years rather >than increase.  :-( >g  G That's only partially true.  While I agree that Windows and UNIX have a0J harder time because of familiarity and availability -- VMS was designed toI be secure, and is old enough, and mature enough to have had a significantbK number of it's vulnerabilities found and fixed.  In addition, it isn't likezL UNIX hasn't been around a *long* time, but still seems to be reinventing new! ways to revive the same old bugs.r  H The participants of DEFCON9 didn't "qualify" their opinion - they reallyJ said it was cool and unhackable.  VMS did something that nobody else thereJ could do - including providing an account on the system to log into, and aI running web server to try and hack.  The best way to try and attack a VMS E system is the Mitnik way - convince someone to give you the password.    ------------------------------  % Date: Fri, 22 Mar 2002 14:02:41 -0500o- From: JF Mezei <jfmezei.spamnot@videotron.ca>'3 Subject: Re: Once a fool, always a fool I guess ...b+ Message-ID: <3C9B7FCF.BDA24B4@videotron.ca>=   Fred Kleinsorge wrote:L > harder time because of familiarity and availability -- VMS was designed toK > be secure, and is old enough, and mature enough to have had a significantP2 > number of it's vulnerabilities found and fixed.       M Not to belittle the quality of VMS' security, but should one characterize the=J TCPIP Services for VMS (which came from Unix) to be "old enough and mature enough" ????  N In today's world, the TCPIP stack seems to be the one big conduit for incoming hacking attempts.a   ------------------------------  # Date: Fri, 22 Mar 2002 15:08:44 GMTl% From: jlsue <jlsuexxxz@insightbb.com>n- Subject: Re: OpenVMS 7.2-1 and UCX 4.2 ECO 05.8 Message-ID: <f1im9uofushiv25dq6k0m4trtg0i16pglf@4ax.com>   Fabio,  F I have done upgrades for customers where I take a copy of their systemD drive and perform all software upgrades on our own systems while the; customer's systems continue to run production applications.l  A After I complete the upgrades, we schedule a time to "reboot" the3B servers at the customer site and swap-in the upgraded system disksD (taking care to copy any *important* "modified" files on the current disk first).  F This allows a relatively short re-boot period of downtime, and then ofE course some testing on the production applications (taking care to do 9 full backups of the systems prior to the "upgrade" boot).   B I have done upgrades on standalone servers that have only cost 2-4D hours of application downtime, and upgrades on VMSclusters that haveE had *zero* application downtime (with proper planning and execution).e      7 On Wed, 20 Mar 2002 07:48:06 -0800 (PST), Fabio Cardoso ! <fabiopenvms@yahoo.com.br> wrote:    >David >l5 >This machine is not clustered, and the users give mee6 >just a few hours per weekend to stop it. It is a 24x76 >machine, and I dont have conditions now to upgrade to2 >VMS 7.3 because we didnt test our systems in this2 >version. It is better to upgrade "slowly", I will0 >go to V7.2-1 because all the other machines in 3 >the company (other sites) are in this max version.h5 >I dont want to cause high impact in this upgrade, itn >ise5 >because I prefer to just patch the UCX, but if there . >is no condition I will upgrade to TCPIP 5.0A. >k5 >I have 1500 printer and server  queues running under! >UCX.- >- >Regards >- >FC  >  >a > 4 >--- David Beatty <David.Beatty@qwertysasasdfgh.com> >wrote:l >>  4 >> Any reason why you would not upgrade to V7.2-2 or >> V7.3 of VMS?1 >> 27 >> Any reason why you would not upgrade to V5.0 or V5.1l4 >> of Compaq TCP/IP Services?  I don't think V4.2 is% >> supported higher than V7.1 of VMS.u >> e >> David R. Beatty >> n2 >> On Wed, 20 Mar 2002 04:27:45 -0800 (PST), Fabio
 >> Cardoso$ >> <fabiopenvms@yahoo.com.br> wrote: >>  ' >> >Anyone is using OpenVMS 7.2-1 with n >> >UCX 4.2 ECO 05 ? >> >6 >> >I am planning to upgrade one Alphaserver this next3 >> >weeked. The actual versions are OpenVMS 7.1-1h2d >> >and UCX 4.2 ECO 1. >> >) >> >Reading the README of the UCX ECO at s) >> >ftp.support.compaq.com, it is saying m+ >> >that UCX ECO 05 is compatible just with  >> >OpenVMS 7.1 (max version). >> >, >> >Is that right ? Is the V7.2-1 compatible" >> >only with TCPIP Services 5.x ? >> > >> >Regardse >> > >> > >> >	 >> >=====  >> >========================== >> >Fbio dos Santos Cardoso >> >OpenVMS System Manager >> >Rio de Janeiro - Brazila >> >fabiopenvms@yahoo.com.br >> >========================== >> >6 >> >__________________________________________________ >> >Do You Yahoo!?/ >> >Yahoo! Sports - live college hoops coverage/ >> >http://sports.yahoo.com/ >>   >h >r >===== >==========================a >Fbio dos Santos Cardosob >OpenVMS System Managere >Rio de Janeiro - Brazil >fabiopenvms@yahoo.com.br2 >==========================d >e3 >__________________________________________________e >Do You Yahoo!?4, >Yahoo! Sports - live college hoops coverage >http://sports.yahoo.com/   1 Not speaking for anyone, certainly not DEC/Compaqa- (get rid of the xxxz in my address to e-mail)D   ------------------------------  % Date: Fri, 22 Mar 2002 09:49:21 -0500.2 From: "Sue Skonetski" <susan.skonetski@compaq.com>, Subject: OpenVMS Technical Update for London3 Message-ID: <XtHm8.1382$fL6.27202@news.cpqcorp.net>i   Dear Newsgroup,h  E The following is the agenda for the London tech update as well as theT: registration site.  I look forward to meeting many of you.  
 Warm Regards,  Suel  ' AlphaServer and OpenVMS Technical Forumm+ Venue  - London Marriott Regents Park Hotel   ' http://www.compaq-conference.com/site3/t  $  Technical Update (day 1) 11th April  * 0900 - 0915                        Welcome  L 0915 - 1000                        OpenVMS Strategy and Directions (Keynote) Steve Stebulis  ( 1000 - 1015                        Break  L 1015 - 1115                        OpenVMS Technical Update 7.3.1 and beyond
 Steve HoffmanO  D 1115 - 1215                        OpenVMS and IPF Systems (base OS) Burns Fisher  ( 1315 - 1315                        Lunch  I 1315 - 1415                        Best Practices for VMS Performance for 0 2002                                Steve Lieman  ( 1415 - 1430                        Break  K 1430 - 1530                        Installing and configuring Fibre Channel6* storage                          Rick Lord  @ 1530 - 1630                        Integration of OpenVMS Data &8 Applications                               Sunil Kumaran  # Technical Update (Day 2) 12th Aprilo    * 0830 - 0845                        Welcome  G 0845 - 0945                        The next generation of Alpha systemsL8 (Keynote)                                  Richard Smith  ( 0945 - 1000                        Break  L 1000 - 1100                         Porting your OpenVMS Applications to IPF Burns Fisher  G 1100 - 1200                        Compaq Secure Web Server Apache withh6 Tomcat                                   Sunil Kumaran  ( 1200 - 1300                        Lunch  A 1300 - 1400                        OpenVMS COE & Unix Portability ; Initiatives                                   Brad McCuskerf  3 1400 - 1500                        Volume Shadowings	 Rick Lords  ( 1500 - 1515                        Break  : 1515 - 1545                        OpenVMS Hints and Kinks
 Steve Hoffman    1615 - 1630            Closure   ------------------------------  % Date: Fri, 22 Mar 2002 17:54:01 +0200o* From: Mike Rechtman <rechtman@tzora.co.il> Subject: Patches for CSWSD* Message-ID: <3C9B5399.9000102@tzora.co.il>   To anyone running CSWS:  checkwK http://www.openvms.compaq.com/openvms/products/ips/apache/csws_patches.htmlh2 Looks as if its not urgent, but nice to feel safe.   ~Mikea -- o  & New to c.o.vms? allow me to recommend:6 http://eisner.encompasserve.org/~rechtman/post_hlp.htmE ---------------------------------------------------------------------oE Usual disclaimer: All opinions are mine alone, perhaps not even that.3? Mike Rechtman                            *rechtman@tzora.co.il*oE Kibbutz Tzor'a.                          Voice (home):(972)-2-9908337"C    "20% of a job takes 80% of the time, the rest takes another 80%"3E ---------------------------------------------------------------------k   ------------------------------    Date: 22 Mar 2002 08:20:00 -0800- From: pdafniotis@yahoo.com (Petros Dafniotis)eD Subject: Re: Problem with SMTP mail on TCPIP v5.1. Help is sought...= Message-ID: <e54adf36.0203220820.5e6b982e@posting.google.com>    Matt-   + here it is what gets bounced back to YAHOO.?K ===========================================================================s& Date: Fri, 22 Mar 2002 10:21:08 -0500    ) From: Postmaster@gatekeeper.mycompany.come  p To: pdafniotis@yahoo.com e  o Subject: Undeliverable mail    	          r; Your message was not delivered to the following recipients:M  > dafniotis@gvcrd1.email.mycompany.com: 550 Closing transmission channel.    # Attachment: Message/delivery-statush    + Reporting-MTA: dns;gatekeeper.mycompany.com   ? Original-Recipient: rfc822;dafniotis@gvcrd1.email.mycompany.comc< Final-Recipient: rfc822;dafniotis@gvcrd1.email.mycompany.com Action: failed
 Status: 5.1.1t- Remote-MTA: dns;gvcrd1.somename.mycompany.comt7 Diagnostic-Code: smtp;550 Closing transmission channel.d  - --------------------------------------------- ; Clearly the line "Remote-MTA" indicates that the gatekeepers; successfully resolved the external name to an internal one.i  F Here is what was logged on my alpha machine. It seems that my Alpha is stopping& the listening process. This the log in) TCPIP$SMTP_COMMON:TCPIP$SMTP_RECV_RUN.LOGe  
 $ Set NoOn/ $ VERIFY = F$VERIFY(F$TRNLNM("SYLOGIN_VERIFY"))0B send buf=220 gvcrd1.somename.mycompany.com V5.1-15, OpenVMS V7.2-2 Alpha ready at Fri, 2l$ 2 Mar 2002 16:28:37 +0100 (MET) \d\a* recv buf=EHLO gatekeeper.mycompany.com\d\a0 send buf=250 gvcrd1.somename.mycompany.com Hello9 gatekeeper.mycompany.com, pleased to meet you, friend\d\av- recv buf=MAIL FROM:<pdafniotis@yahoo.com>\d\aa. send buf=550 Closing transmission channel.\d\a8   TCPIP$SMTP   job terminated at 22-MAR-2002 16:28:38.90     Accounting information:lF   Buffered I/O count:                185      Peak working set size:       6912F   Direct I/O count:                  118      Peak virtual size:         176496F   Page faults:                       666      Mounted volumes:                0C   Charged CPU time:        0 00:00:00.60      Elapsed time:       0i 00:00:03.19o $e        z "Matt Muggeridge" <Matt.Muggeridge@compaq.com> wrote in message news:<bVqm8.17015$Hz2.68392@news-server.bigpond.net.au>...N > So your outbound mail is sent successfully, your inbound mail bounces at the
 > gateway? > K > What reject message do your remote senders receive when the mail bounces?t > 3 > >the gatekeeper rejects email sent to the addressg > H > I cannot imagine how your changes could cause the gatekeeper to reject > inbound mail.. >  > Matt.r >  > --? > -------------------------------------------------------------> > OpenVMS TCP/IP Engineering > Enterprise Computing Group > Compaq Computer Corporationi > Gold Coast, AUSTRALIA ? > -------------------------------------------------------------  >  > < > "Petros Dafniotis" <pdafniotis@yahoo.com> wrote in message9 > news:e54adf36.0203210516.1c3b7640@posting.google.com...wG > > I have a big problem after I have upgraded to TCPIP v5.1 (I had 5.0q > > before). TheF > > problem is on both of my Alphas (one running OpenVMS 7.2-2 and the > > other 7.3).v > >yJ > > I am sending email behind a firewall. So on my computer (name: GVCRD1). > > my real domain name inside the company is: > > ! > > gvcrd1.somename.mycompany.com  > >gA > > I have to ask SMTP to send the email to the outside world as:h > >d > > gvcrd1.email.mycompany.com > >aB > > When response arrives at the gatekeeper of my company, this is > > automatically changed to:r > >t! > > gvcrd1.somename.mycompany.com  > >s# > > and gets delivered to my Alpha.S > > F > > Now here is the problem. This was all working great till I did the > > upgrade. NowI > > it does not function anymore in the sense that the gatekeeper rejectsB	 > > email)2 > > sent to the address:   gvcrd1.email.dupont.comH > > However I do not think that anythings has changed on the gatekeeper.G > > Some setting must be screwed up on my computer but I cannot find it. > > out. > > I > > Of course delivery from my machine to the outside world works but not3
 > > the othero > > way around.  > >  > > Here is the output:n > >o! > > $ tcpip show config smtp/full  > >h > > SMTP Configuration > >o > > Optionst> > > Initial interval:   0 00:30:00.00       Address_max:    16 > > NOEIGHT_BIT > > > Retry interval:     0 01:00:00.00       Hop_count_max:  16 > > NORELAYw% > > Maximum interval:   3 00:00:00.00e > > TOP_HEADERSS > >e@ > > Timeout             Initial       Mail    Receipt       Data
 > > Terminate @ > >   Send:                   5          5          5          3 > > 10 > >   Receive:                5l > >r0 > > Alternate gateway:  GATEKEEPER.MYCOMPANY.COM# > > General gateway:    not defined- > > : > > Substitute domain:  HIDDEN, GVCRD1.EMAIL.MYCOMPANY.COM# > > Zone:               not definedu > >:" > > Postmaster:         TCPIP$SMTPG > > Log file:           SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOGi > >e4 > > Generic queue       Queues   Participating nodes > >m' > > TCPIP$SMTP_GVCRD1_00   1     GVCRD1o > >r
 > > and also:. > >r > > $ show service smtp/full > >n > > Service: SMTP 1 > >                            State:     Enabled0B > > Port:               25     Protocol:  TCP             Address: > > 0.0.0.0eB > > Inactivity:          5     User_name: TCPIP$SMTP      Process: > > TCPIP$SMTPG > > Limit:              10     Active:      0             Peak:       1  > >h6 > > File:         TCPIP$SYSTEM:TCPIP$SMTP_RECV_RUN.COM > > Flags:        Listen > >m > > Socket Opts:  Rcheck Scheckd4 > >  Receive:            0     Send:               0 > > J > > Log Opts:     Acpt Actv Dactv Conn Error Exit Logi Logo Mdfy Rjct TimO > > AddrB > >  File:        SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG > >> > > Security > >  Reject msg:  not definede > >  Accept host: 0.0.0.0M > >  Accept netw: 0.0.0.0a > >o > > -------------e' > > Any help appreciated. Kind regards,  > >o
 > > Petros > > ---  > > Petros Dafniotis, PhD? > > pdafniotis@yahoo.com   ------------------------------  % Date: Fri, 22 Mar 2002 18:25:22 +0100 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>eD Subject: Re: Problem with SMTP mail on TCPIP v5.1. Help is sought...& Message-ID: <3C9B6902.81A1353@aaa.com>  ; I'd try to put on more logging in the SMTP receiver to find  out what happens between  1 "> recv buf=MAIL FROM:<pdafniotis@yahoo.com>\d\a"a   andr  2 "> send buf=550 Closing transmission channel.\d\a"  : For some reason, it looks as your box decides to close the
 transmission.i   Jan-Erik Sderholm.-   ------------------------------    Date: 22 Mar 2002 02:37:25 -0800% From: andrew_art@hotmail.com (Andrew)t= Subject: process priorities under v7.1-2, 7.2-1 Alpha serverss= Message-ID: <6ce640e8.0203220237.566d008f@posting.google.com>   C Since several of our supported sites ( who run a ticketing softwaree? application ) upgraded to VMS 7.1-2 and 7.2-1 we've noticed theW@ cur/base priority level of users are rising to never before seenA levels, e.g 56 or 40. On previous versions of VMS they never wenta above 4.  C I'm not sure if this is anything to worry about but we seem to havey@ had far more instances of Alpha machines 'hanging' and needing aA reboot ( Compaq were analysing one of the dump files and said theeA crash "might be caused by DECNET not de-allocating space from theI non-paged pool" )s  F Anyway, has anyone else noticed this priority strangeness, if so is it> a problems ? And, if you have the time to write, are any usersE experiencing more 'hanging' machines or crashes under 7.2-1 ( & 7.1-2o )e   Thanks for any advice, Andrew   ------------------------------  % Date: Fri, 22 Mar 2002 11:54:16 +0100r9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>eA Subject: Re: process priorities under v7.1-2, 7.2-1 Alpha serverst' Message-ID: <3C9B0D58.CB060D7E@aaa.com>t   Upgraded from what version ?  What are these "users" running ? Other layered products ? Some 3'rd part middleware ?d. Checked all patches for current VMS versions ?  & Isn't 32 the maximum prioroty in VMS ?1 And anything above 16 is concidered "real-time" ?d  7 What "tool" (e.g. "SHOW SYS") are you using when seeingm this priorities ?t  5 I'm running a 7.2-1 box, and it's stable as a rock...    Jan-Erik Sderholm.E   PS.i@ Why not use the same version on all boxes, when you upgraded all anyway ? DS.   
 Andrew wrote:t > E > Since several of our supported sites ( who run a ticketing software A > application ) upgraded to VMS 7.1-2 and 7.2-1 we've noticed theaB > cur/base priority level of users are rising to never before seenC > levels, e.g 56 or 40. On previous versions of VMS they never wenta
 > above 4. >    ------------------------------  # Date: Fri, 22 Mar 2002 11:13:27 GMTj1 From: "Kari Keronen" <kari.keronen@radiolinja.fi>eA Subject: Re: process priorities under v7.1-2, 7.2-1 Alpha servers:5 Message-ID: <rlEm8.12416$zZ5.259681@news.kpnqwest.fi>   3 "Jan-Erik Sderholm" <aaa@aaa.com> wrote in message ! news:3C9B0D58.CB060D7E@aaa.com...n  ( > Isn't 32 the maximum prioroty in VMS ?   No, on Alpha it's 63.o   -Kari-   ------------------------------  % Date: Fri, 22 Mar 2002 12:17:06 +0100e9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> A Subject: Re: process priorities under v7.1-2, 7.2-1 Alpha serversb' Message-ID: <3C9B12B2.67E9606E@aaa.com>n  ' Ah, well, I'v never been "up there" :-)-  	 Jan-Erik.r     Kari Keronen wrote:M > 5 > "Jan-Erik Sderholm" <aaa@aaa.com> wrote in messagee# > news:3C9B0D58.CB060D7E@aaa.com...g > * > > Isn't 32 the maximum prioroty in VMS ? >  > No, on Alpha it's 63.p >  > -Kari-   ------------------------------  % Date: Fri, 22 Mar 2002 11:21:36 -0000u* From: "Richard Brodie" <R.Brodie@rl.ac.uk>A Subject: Re: process priorities under v7.1-2, 7.2-1 Alpha serverso, Message-ID: <a7f43l$11rm@newton.cc.rl.ac.uk>  < "Kari Keronen" <kari.keronen@radiolinja.fi> wrote in message/ news:rlEm8.12416$zZ5.259681@news.kpnqwest.fi...n  * > > Isn't 32 the maximum prioroty in VMS ? >  > No, on Alpha it's 63.y  + Are you sure? That's not what the docs say.    ------------------------------  # Date: Fri, 22 Mar 2002 11:51:12 GMTu1 From: "Kari Keronen" <kari.keronen@radiolinja.fi>oA Subject: Re: process priorities under v7.1-2, 7.2-1 Alpha serverso5 Message-ID: <QUEm8.12420$zZ5.259612@news.kpnqwest.fi>V  5 "Richard Brodie" <R.Brodie@rl.ac.uk> wrote in messaget& news:a7f43l$11rm@newton.cc.rl.ac.uk... >n> > "Kari Keronen" <kari.keronen@radiolinja.fi> wrote in message1 > news:rlEm8.12416$zZ5.259681@news.kpnqwest.fi...h > , > > > Isn't 32 the maximum prioroty in VMS ? > >e > > No, on Alpha it's 63.  > - > Are you sure? That's not what the docs say.o >n Here is example from our GS160:e   KEROKA_HURJA>sho sys/proc=*lck* J OpenVMS V7.2-2  on node HURJA  22-MAR-2002 13:41:10.33  Uptime  1 15:42:58J   Pid          Process Name              State      Pri      I/O       CPU Page flts  PagesF 20600806 LCKMGR_SERVER   CUR  1  63       48   1 13:08:50.17        88 210c  L This is the process that in Wildfire ( GS80/GS160/GS320) environments uses 1G CPU alone. That CPU handles all spinlocks and nothing else thus freeingsL other CPUs from doing MPsync. Uses allways maximum priority that being 63 as you can see.   -Kari-   ------------------------------  % Date: Fri, 22 Mar 2002 11:50:02 -0600y% From: "Dave Rich" <drich@nucorar.com>.+ Subject: Question reguarding process crash.d5 Message-ID: <a7fqpt$k5rg1$1@ID-136223.news.dfncis.de>S  B The processes I am running does this, after a little over 2 hours.  5 Process CAS_PROLST started at 21-MAR-2002 15:18:04.77r4 Open Error: non-translatable vms error code: 0x1C002" %rms-e-acc, acp file access failed8   RICH         job terminated at 21-MAR-2002 17:37:38.99   Accounting information: K   Buffered I/O count:             640         Peak working set size:  19408 K   Direct I/O count:             16511         Peak page file size:    81904 K   Page faults:                   1268         Mounted volumes:            0iA   Charged CPU time:           0 00:01:10.28   Elapsed time:     0  02:19:35.29a
 [End of file]>  J Is there a way to get a tad more info? Like 'where' in the application the problem is occuring, etc etc?   C I can / will post the src code, if anyone wants to look it over for=	 problems.=   ------------------------------  + Date: Fri, 22 Mar 2002 09:09:29 +0000 (UTC)Y* From: Osmo Kujala <kujala@tukki.cc.jyu.fi>" Subject: Re: Right down the middle, Message-ID: <a7esc9$abt$1@mordred.cc.jyu.fi>  4 Fred Kleinsorge <kleinsorge@star.zko.dec.com> wrote:  J > I am perplexed why any VMS customer should be concerned about the merger > other than generic FUD.    Generic FUD like: < http://www.hp.com/hpinfo/execteam/speeches/fiorina/idc01.htm  5 > I believe that it has a lot of upside potential for J > VMS.  You don't like how Compaq has handled VMS, why not let HP try it'sN > hand.  From your arguments, aside from outright cancellation - can it be any > worse than the current state?o  H Maybe not. VMS is already dead as OS, only kind of special niche system J left. But there are still invisible VMS users, like we small universities,2 who insist to continue running VMS. Stupid we are.E (I don't mean there is something wrong with VMS, only in the way it'sn	 treated.)oE Do you think we will want to convert to IA64, when even the VAX-Alpha  transition isn't ready yet.nF Porting to IA64 is interesting and good thing, but without future with Alpha doesn't feel good.   Osmo   ------------------------------  # Date: Fri, 22 Mar 2002 15:40:45 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>Y Subject: Re: Session abstracts and bio's for OpenVMS Technicial Updates in Europe (Austrie> Message-ID: <1gIm8.102440$uA5.86835@rwcrnsc51.ops.asp.att.net>  = "Sue Skonetski" <susan.skonetski@compaq.com> wrote in messageu- news:rdIm8.1386$fL6.27158@news.cpqcorp.net... J > There are a couple of session abstracts missing but this will give you aE > good idea on what to expect.  And I could be partial but this is an3 > excellent group of engineers.  <snip>  K Yes indeed. I attended the Technical Update Days in Europe last spring. All C were well-attended, and all were very useful. Audience feedback was I overwhelmingly positive. If you live in EMEA and have an interest in VMS,S& these events are well worth attending.   ------------------------------  % Date: Fri, 22 Mar 2002 10:40:00 -0500 2 From: "Sue Skonetski" <susan.skonetski@compaq.com>Y Subject: Session abstracts and bio's for OpenVMS Technicial Updates in Europe (Austria an 3 Message-ID: <rdIm8.1386$fL6.27158@news.cpqcorp.net>   H There are a couple of session abstracts missing but this will give you aC good idea on what to expect.  And I could be partial but this is an  excellent group of engineers.S  
 Warm Regards,2 Sue1     Session Abstracts K OpenVMS Strategy and Directions (Keynote) - This session will cover OpenVMS L strategies and directions.  Additionally this session will cover a number of0 different programs in place in the OpenVMS GroupF  OpenVMS Technical Update 7.3.1 and beyond - This session will provideL information and roadmaps for OpenVMS, OpenVMS Galaxy, Clustering, eBusiness,L TCP/IP Services, and other areas.  Technical details of current and upcomingK OpenVMS operating system releases including V7.2-2, V7.3, V7.3-1, V7.4, theDD Itanium Port, and on on-going advanced development projects, will be featured in the presentation.c    K OpenVMS and IPF Systems (base OS) -  This session describes how some of theeK internals of VMS will be ported to the Itanium Processor Family. MechanismsoJ such as ASTs, IPLs, page protections, synchronization methods, and contextL switching are presented in terms of the Intel fundamentals used to implement them.n    J Best Practices for VMS Performance for 2002 - This session highlights someI new and evolving OpenVMS capabilities, tools, and best practices that youiL can put to use to help with performance management, tuning, troubleshooting,I and capacity planning on your largest and most important OpenVMS systems.-I The session relates these new practices to the standard set of industrial:L strength performance practices that have proven themselves to be universallyI applicable and indispensable to attaining consistent performance success.3G The session will also briefly illustrate how these combined old and newuL practices can set the stage for success with the next generation Alphaserver and IPF systems.  3  Installing and configuring Fibre Channel storage -o    - Integration of OpenVMS Data & Applications  -t    G The next generation of Alpha systems (Keynote) At the end of this year,eJ Compaq will introduce a new family of modular AlphaServer systems based onH the new, highly integrated Alpha EV7 CPU.  This session will explain theL unified architecture for hardware, software, and system management that makeJ these systems easy to manage, reliable, and scalable from low price, entryD level dual processor systems to high capacity 128 processor systems.L Systems, assembled from a small number of simple building blocks modules andI drawers, will offer the flexibility to expand processing, memory, and I/0-1 independently, and over a wide range of capacity.t      E Porting your OpenVMS Applications to IPF - The port of OpenVMS to theiH Itanium(tm) Processor Family was announced in June, 2001. This technicalC session will cover the current OpenVMS Itanium(tm) Processor Family:I schedule, plans for porting layered products, plans for compilers, binaryIA translators, and other preliminary information about porting yourp. applications to Itanium-based OpenVMS systems.  + Compaq Secure Web Server Apache with Tomcate    G OpenVMS COE & Unix Portability Initiatives - This session describes theeG current efforts under way to simplify porting of applications from UNIXMH systems to OpenVMS.  Specific areas addressed include the extensive workI being done to update and enhance the OpenVMS C Runtime Library (CRTL) andgJ the UNIX tools and utilities on OpenVMS, and to improve the performance ofJ ported applications.  A portion of this session will allow for interactive: discussion of customer needs and directions for the future  H  Volume Shadowing -  This session will provide information on a new HBVSI feature in V7.3, Mini Copy.  This new feature will enable a former shadow G set member to be returned to the shadow set in a much shorter time than1G normal, based on the level of write activity while it was removed.  ThenJ session will also include an update on new DCL functionality, that enablesD new per shadow set member read cost and member time out values to be) specified, and new DCL recovery semanticsh  L OpenVMS Hints and Kinks - This technical session provides an insiders' guideH on taking advantage of OpenVMS strengths, covering the following topics:F Optimizing programs for effective failover; Failover locking; HandlingL security features correctly; Designing shareable images; User-written SystemI Services; Message utility and command definition  utility; OpenVMS Galaxy D and OpenVMS Cluster; Insights into C, C++, and the OpenVMS Debugger.L Details on certain useful (and undocumented) features will also be included.    Speaker Bio's    K Steve Stebulis, is a Director in the High Performance Server Division, withoJ over 20 years of experience in software engineering and product managementK with a focus on operating system and technical development.  Steve had beenyJ one of the driving forces behind the OpenVMS strategy and has expanded hisD role to cover Tru64, Lunix, High Performance Technical Computing andJ Computer Special systems working for Rich Marcello, VP and General ManagerA of the HPSD organization.  He is also responsible for key accounth2 relationships and customer architectural planning.  D  Previously, Steve has held a number of positions in OpenVMS productJ management, from managing the OpenVMS product management group to focusingK on specific business lines including the base operating system and businessiL critical technologies group.  Steve's other experience includes positions asL product manager for OpenVMS POSIX and Open Standards, and product managementH and engineering positions in workstation engineering, networks, database$ development and computer operations.  L Steve received a B.S. in Forest Sciences from Penn State University, an M.A.F in Computer Science and an M.E. in Recreation Administration both from& Temple University in Philadelphia, PA.  G  Stephen "Hoff" Hoffman, Is a Consulting Engineer in the Compaq OpenVMS K Engineering group, with a variety of responsibilities.  Areas of experience C include voice applications and telephony, factory floor networking,iL databases, device drivers and ACPs, web and internet technologies, hardware,I bad puns, clustering, with other areas too numerous to mention.  Writings L include the second edition of the "Writing Real Programs in DCL" book and anI update(presently underway) to another Digital Press OpenVMS-related book.   F  Burns Fisher, Since joining VMS Engineering in 1986, Burns Fisher hasD worked in many engineering and project leading roles.  These includeB contributing to the MIT X11 protocol, porting DECwindows to Alpha,I contributing to the System Code Debugger and SCSI clusters, and technical K leadership of the Extended File Specifications project.  More recently as a I member of the VMS Exec group, he helped develop Galaxy concepts, designedtJ and wrote RAD-based scheduling and RAD-based pool allocation for GS-seriesJ Alphaserver systems.  He is currently designing and writing code to do lowJ level interruption handling and software interrupts for VMS on the Itanium Processor Family.     L Before joining VMS, Burns worked as an engineer in Digital's ESG EngineeringJ group, and as a system manager, coder, and general computer engineer for aK research project at the University of Rochester (New York, USA).  He has anyK Master's degree in Electrical Engineering from the University of Rochester.E  F  Steve Lieman, (OpenVMS Performance Group, Compaq).  Steve is an awardB winning author who has more than 30 years large system performanceG experience with the last 7 years focusing on OpenVMS.  He has delivered F seminars on practical performance management at more than 30 locationsD worldwide.  He is currently working on various OpenVMS benchmarking,L performance characterization, and tool enhancement project and is frequentlyH called in to help troubleshoot customer performance problems.  Steve has; been a long-time student of the best performance practices.o  L  Rick Lord - Joined Digital in 1983 in the Mid-Range Systems Business Group,K working on CAD support, the VAXBI DRB32 VMS device driver and VAX RTA (RealoJ Time Accelerator). part of Digital's After that Rick was part VAXTCC (TimeL Critical Computing) startup in Munich. Then continuing as the technical leadJ on the VAX 8400 console port before joining Digital's Telecom group, whereG he worked on an NMF (Network Management Forum) interoperability project6H between DECmcc and British Telecom's Concert network management package.L Rick joined the OpenVMS group in 1991. Rick now works with FibreChannel, andK is  largely responsible for the design & implementation of our FibreChanneln drivers.  L  Richard Smith has served as the Product Manager on the development team forI a new family of AlphaServers for the past 3 years. In 20 years at DigitaliF and Compaq, he led business and engineering for PC network software inK Europe and managed Business Development for MicroVAX minicomputers.  BeforeNB joining Digital, Richard worked at AT&T and in various engineeringL assignments at Systems Control, Inc.  He earned MS and BS degrees in controlL systems engineering from MIT and an MBA from Harvard Business School.  After> six years living and working in France, he is fluent in French  L  John Apps - Is a senior technical architect in the High Performance ServersK Division of Compaq.  He has over 30 years in the IT industry, mainly in thej6 area of transaction processing and middleware systems.  > He works with the OpenVMS team which architects and implementsF Internet/eBusiness solutions for OpenVMS and Alpha customers. Areas ofL expertise include: designing for scalability and integration of both WindowsE and Tru64 UNIX based eBusiness and integration solutions into OpenVMScI environments using industry application server technologies from Compaq'sn& partners such as Oracle, BEA and IONA.  L  He consults worldwide and has implemented a number of Internet projects for$ existing OpenVMS and Alpha customers  H  Sunil Kumaran has 15 years experience in the IT industry, the last 6 ofJ which has been with Compaq. He has worked in a number of fields, from realE time systems, to Scada systems to Stock Exchange systems. He also hastL experience developing Oracle applications as well as RTR based applications.I Most of his work has been with Compaq platforms and Operating systems and-J applications that run on them. He is currently a Technology Consultant forE the High Performance Systems division.    Areas of expertise include:7H designing for scalability and integration of both Windows and Tru64 UNIXI based eBusiness and integration solutions into OpenVMS environments using0G industry application server technologies from Compaq's partners such aso Oracle and BEA.3    L He consults worldwide and has participated in the implementation a number of: Internet projects for existing OpenVMS and Alpha customers  I  Brad McCusker -  has been with the OpenVMS Engineering group since 1995.eI Currently, Brad is the Project Leader for the C Run-Time Library (C RTL),eE responsible for leading the upgrade of the C RTL code to support UNIXlE portability.  In addition, Brad has a long history with PATHWORKS andgL Advanced Server engineering where he continues to be the engineering ProjectE Manager, responsible for directing all engineering efforts.  Prior to0D Digital and Compaq, Brad spent eight years in an OpenVMS development@ environment, primarily for the US Navy in the field of real-timeK anti-submarine warfare simulation.  Brad frequently speaks in North Americao4 and Europe about OpenVMS and Windows NT integration.   ------------------------------  + Date: Fri, 22 Mar 2002 13:22:38 +0000 (UTC)n* From: Osmo Kujala <kujala@tukki.cc.jyu.fi>- Subject: Re: Shannon knows DEC -> CPQ -> ????-, Message-ID: <a7fb6u$hj3$1@mordred.cc.jyu.fi>  " John Smith <a@nonymous.com> wrote:  B > "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message/ > news:eepm8.1325$fL6.26745@news.cpqcorp.net...@ >> >>7 >> You are reading tea leaves that simply aren't there.i   > Fred, I > So why has there been a distinct silence thus far (since the merger wasr  G Not only silence, but Carly talks about Market-unifying standards which.J means of course industry standards. Clearly VMS doesn't belong to that but: "Itanium, UNIX/Linux, NT, open source, open APIs and open  connectivity...". C VMS would be in "proprietary, inflexible, old-world approaches ..."   N > announced) in statements by Carly/Curly et al. about the fate of VMS, unlikeC > definitive announcements pertaining to the death of Tru64 and thet > continuation of NSK?  I > Please tell us that you know what Carly is thinking and committed to in M > terms of growing VMS. Or failing the merger, what Curly plans to do in thisa	 > regard.e   Osmo   ------------------------------  % Date: Fri, 22 Mar 2002 06:08:15 -0800a# From: "Tom Linden" <tom@kednos.com> - Subject: RE: Shannon knows DEC -> CPQ -> ????-9 Message-ID: <CIEJLCMNHNNDLLOOGNJICEGMEGAA.tom@kednos.com>m  * Is this the Shannon knows Algol story? :-)   > -----Original Message-----; > From: Terry C. Shannon [mailto:terryshannon@mediaone.net] ( > Sent: Thursday, March 21, 2002 7:16 PM > To: Info-VAX@Mvb.Saic.Comr/ > Subject: Re: Shannon knows DEC -> CPQ -> ????7 >  >n >r1 > "Mark Hittinger" <bugs@pu.net> wrote in messagee? > news:Mewm8.190902$uv5.16444189@bin6.nnrp.aus1.giganews.com...- > > Shannon knows Unisys(4) :-)i > >f >aC > Well, I did know one half of the so-called "Power of Two." I usedt > to programL > Burroughs L-Series minicomputers in SL/3 and SL/5 assembler, and had a bit3 > of experience with big Burroughs systems as well.e >o >e >    ------------------------------  % Date: Fri, 22 Mar 2002 09:12:29 -0500 # From: "Tom Cole" <tom.cole@sas.com> - Subject: Re: Shannon knows DEC -> CPQ -> ????e/ Message-ID: <a7fe4e$85e$1@license1.unx.sas.com>e  G "Simon Clubley" <clubley@remove_me.eisner.decus.org-Earth.UFP> wrote in 5 message news:2nZVaifpLEPr@eisner.encompasserve.org...aA > So, anyone want to guess what Terry calls his publication now ?l >uJ > My suggestion: Shannon knows DEC, where DEC = Digital Enhanced ComputingI > (a variant on the DEC = Digital Enhanced Compaq suggested when CPQ took: > over DEC)r  * CHP?  No wait, that's Estrada Knows CHP...   -- Tom Cole! HPUX+TRU64=TRUX?  Silly rabbit...e   ------------------------------  % Date: Fri, 22 Mar 2002 09:19:28 -0500h# From: "Tom Cole" <tom.cole@sas.com>f- Subject: Re: Shannon knows DEC -> CPQ -> ????h/ Message-ID: <a7fehg$8a6$1@license1.unx.sas.com>s  H Oops!  My mistake... I took the posting to be somewhat whimsical, beforeK reading the rest of the followups and seeing that (like most threads in covS? today) it is YET ANOTHER TWO DOZEN RANTS ABOUT THE POLITICS ANDn PERSONALITIES OF THE MERGER.  1     "I have a question about why SYS$GETJPIW sets !        the IOSB the way it does."   2     "Because Carly is a spawn of satan, and worked2      out a devil's deal with Curly to sell all our      futures down the river."r  /     "No, I think a more reasonable and moderater2      approach is to imagine that Carly cares about4      IOSBs despite her complete silence on the topic
      so far."e  9     "You are a moron who fails to see the facts clearly."e  6     "No you are a moron, and a tool of the evil ones."  
     "Am not."V       "Are too."  4 COV isn't as much fun these days as it used to be...      . "Tom Cole" <tom.cole@sas.com> wrote in message) news:a7fe4e$85e$1@license1.unx.sas.com...  >sI > "Simon Clubley" <clubley@remove_me.eisner.decus.org-Earth.UFP> wrote in 7 > message news:2nZVaifpLEPr@eisner.encompasserve.org...nC > > So, anyone want to guess what Terry calls his publication now ?  > > L > > My suggestion: Shannon knows DEC, where DEC = Digital Enhanced ComputingK > > (a variant on the DEC = Digital Enhanced Compaq suggested when CPQ tookn
 > > over DEC)  >v, > CHP?  No wait, that's Estrada Knows CHP... >v > --
 > Tom Cole# > HPUX+TRU64=TRUX?  Silly rabbit...e >n >  >- >-   ------------------------------  # Date: Fri, 22 Mar 2002 15:14:43 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>- Subject: Re: Shannon knows DEC -> CPQ -> ????c> Message-ID: <DTHm8.102243$uA5.86877@rwcrnsc51.ops.asp.att.net>  . "Tom Linden" <tom@kednos.com> wrote in message3 news:CIEJLCMNHNNDLLOOGNJICEGMEGAA.tom@kednos.com...o, > Is this the Shannon knows Algol story? :-)  % Nope. SKC, as in Shannon Knows CORAL.e  ( "Or COBOL," he said in a JOVIAL fashion.   ------------------------------  % Date: Fri, 22 Mar 2002 10:31:32 -0500l- From: JF Mezei <jfmezei.spamnot@videotron.ca>d- Subject: Re: Shannon knows DEC -> CPQ -> ???? , Message-ID: <3C9B4E53.5B8845DA@videotron.ca>   Tom Cole wrote:c6 > COV isn't as much fun these days as it used to be...  H If Compaq/ and/or HP (depending on oiutcome of merger) provide a *clear*K statement with regards to VMS, and if Compaq/HP follow through with actionsgI that support that statement, then there won't be much need for discussionh either way.o  L It is the uncertainty that is killing VMS. *If* they announce the retirementL of VMS  and provide an honest reason  then there won't be much of a point to discuss it.e  I The debates about Alpha were mostly due to Compaq providing a reason thato> lacked credibility and it was clear that Compaq wasn't honest.  J As long as the owner provides an honest reason and not some concucted lame. excuse, there won't be much need to speculate.  L However, if HP simply ignores VMS but Gorham and Hoffman say that VMS is OK,? we will continue to really wonder what HP has in store for VMS.t   ------------------------------  % Date: Fri, 22 Mar 2002 09:03:01 +0100p) From: Bart Zorn <B.Zorn@xs4all.nospam.nl>O+ Subject: Re: System software status utilitye/ Message-ID: <3C9AE535.1080003@xs4all.nospam.nl>t  
 JMK wrote:H > Does anyone know of an existing utility that can tell me if a computerI > on the network has all its software up and running?  Hardware utilities J > abound to determine if computers are up, disks, routers, etc are online,7 > but I'm not aware of any like product for software.  e >  > Are any out there? >  > TIA, >  > Jeff Klopotic   C I am not very familiar with Linux and NT, but for OpenVMS there is  E Cockpit Manager from Compaq in Belgium. It allows you to monitor and e? manage almost everything in and around OpenVMS clusters. It is y8 especially targeted at disaster tolerant configurations.  	 Bart Zorn    ------------------------------   Date: 22 Mar 2002 14:44:09 GMT/ From: "Mark H. Wood" <mwood@mhw.ULib.IUPUI.Edu> + Subject: Re: System software status utility,- Message-ID: <a7ffvp$901$1@hercules.iupui.edu>p  E You can get the process table from Windows 2000 via SNMP, if you haveiC the SNMP agent loaded on the target host.  IIRC there are some SNMP > manager tools in either the Win2k kit or the Resource Kit.  (IC wouldn't know their names; I use Net-SNMP on Linux for this sort ofo
 snoopery.)   -- O6 Mark H. Wood, Lead System Programmer   mwood@IUPUI.EduE MS Windows *is* user-friendly, but only for certain values of "user".S   ------------------------------  # Date: Fri, 22 Mar 2002 15:02:47 GMTc% From: jlsue <jlsuexxxz@insightbb.com>l+ Subject: Re: System software status utilityt8 Message-ID: <gmhm9uogpdke2oi5oictch3vmq0uq6n42n@4ax.com>  E There are many Enterprise Management tools available.  It all dependsnC on how much you want to get into it, and how much you're willing too spend.A The old System Watchdog tool was sold to CA and is now "Unicentere% System Watchdog for OpenVMS" from CA. D There's also the tools that provide an enterprise-wide console, likeB Unicenter TNG, BMC Patrol, and Tivoli Enterprise Manager.  Most ofD these also include options for alerting via paging, e-mail, etc., as3 well as automatic script execution based on events.    HTHr    ' On Tue, 19 Mar 2002 14:41:44 -0800, JMKm+ <jeffrey.m.klopotic@lmco.nospam.com> wrote:r  G >Does anyone know of an existing utility that can tell me if a computeraH >on the network has all its software up and running?  Hardware utilitiesI >abound to determine if computers are up, disks, routers, etc are online, 6 >but I'm not aware of any like product for software.   >n >Are any out there?  >t >TIA,s >e >Jeff Klopotic  1 Not speaking for anyone, certainly not DEC/Compaqp- (get rid of the xxxz in my address to e-mail)n   ------------------------------    Date: 22 Mar 2002 13:11:38 -00004 From: Doc.Cypher <Use-Author-Address-Header@[127.1]> Subject: Upgrading from v7.24 Message-ID: <20020322131138.657.qmail@gacracker.org>  H I've just had a user ask if it was possible for me to install the v1.3.1H Java kit on VMSbox. I spent this morning checking up on and applying ECOJ kits because when I scanned the Java pages in the Compaq website there wasI mention of needing to be fairly up to date on those. Of course, because IeJ didn't read the page thoroughly enough the first time I didn't notice thatF I can't install it because I'm on the Hobbyist version - OpenVMS v7.2.H That's not to say that applying the patches wasn't worthwhile - I shouldF have done it a while ago, but have been lazy about it and not made the time.e  H Now, if I recall correctly, there was a PCSI kit to upgrade from v6.2 toF v7.2, and this prompted me to look for a similar kit to get me up to aE version where I can install Java 1.3.1. I can't see one. Am I missingaJ anything, or correct that no such kit exists? Are there any plans to offerJ such a kit? After all, I'm sure I'm not the only Hobbyist who'd like to be a little more up to date.g  I Now, I know I can borrow media - but I've checked with the VMS sites I'venK done work at in the recent past and none are up to date enough to lend me arI recent install kit. :-( So if there's anyone in Belgium who can lend me apK up to date kit I'd appreciate it. Contact me by mail if you can oblige, andiI I'll obviously owe several fine Belgian beers to anyone that can help outr here.i     Doc. -- e6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                              http://vmsbox.cjb.nett   ------------------------------  % Date: Fri, 22 Mar 2002 11:02:11 -0000I? From: "Malcolm MacArthur" <malcolmix@neverness.freeserve.co.uk>,. Subject: Re: VAX/VMS emulator or free version?. Message-ID: <a7f2tj$tmf$1@news7.svr.pol.co.uk>  ' <sword7@speakeasy.org> wrote in message,) news:u9l02smcdof8d7@corp.supernews.com...t@ > Malcolm MacArthur <malcolmix@neverness.freeserve.co.uk> wrote:N > > A guy called Tim Stark has been working on a PDP-10 and VAX emulator. Last > > post I saw from him here, M > > he had just about got it to boot VMS. But it is still a work-in-progress.B SeeF) > > http://sourceforge.net/projects/ts10.6 >uC > Yeah.  That is me.  Latest updates are available to my ftp site - B > ftp://ftp.firesword7.net/pub/ts10/develop and get latest one.  ID > still am working on it with DELQA emulation.  Now I am able telnetJ > and ftp into my TS10 emulator through DELQA emulation but DELQA has someG > problems with interrupt issue or lag time.  Also, I was able downloadaL > a 1.4 MB file from TS10 emulator at 250 to 500 kbps depending on processor& > speed.  That is much like DSL speed. >.F > Also, there is another VAX emulator called Bob's simh emulator.  For7 > more information, check http://simh.trailing-edge.comm >r > -- Tim Stark >  > --. > Timothy Stark <>< Inet: sword7@speakeasy.orgL > --------------------------------------------------------------------------G > "For God so loved the world, that he gave his only begotten Son, thatNJ > whosoever believeth in him should not perish, but have everlasting life.0 > Amen." -- John 3:16 (King James Version Bible)   ------------------------------  # Date: Fri, 22 Mar 2002 11:46:01 GMTh. From: ">>> ^P" <plj@NOSPAM.byron.ext.telia.se>. Subject: Re: VAX/VMS emulator or free version?9 Message-ID: <3C9B1980.30B25A63@NOSPAM.byron.ext.telia.se>    Hey, don't forget: http://simh.trailing-edge.com/  	 >>> ^P.Ljo     mykrowyre skrev:  = > Is there a free opensource version or emulator for VAX/VMS?e >p > -tom   ------------------------------  % Date: Fri, 22 Mar 2002 09:22:43 +0000e From: Roy Omond <Roy@Omond.net>s Subject: Why Delaware ?m) Message-ID: <3C9AF7E3.CE5C444B@Omond.net>i  . Could someone please explain why HP and Compaq# are both incorporated in Delaware ?    Just out of interest ...  	 Roy Omondo Blue Bubble Ltd.   ------------------------------  % Date: Fri, 22 Mar 2002 06:55:28 -0500i2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: Why Delaware ?sJ Message-ID: <rdeininger-2203020655290001@1cust56.tnt1.nashua.nh.da.uu.net>  > In article <3C9AF7E3.CE5C444B@Omond.net>, Roy@Omond.net wrote:  / >Could someone please explain why HP and Compaq $ >are both incorporated in Delaware ? >n >Just out of interest ...F  D Generally speaking, Delaware is more corporation-friendly than otherC states.  Taxes, fees, and red tape are generally less than in otherP states.n  G Credit card companies and mutual funds seem to be incorporated in DE in  most cases as well.a   ------------------------------  # Date: Fri, 22 Mar 2002 12:13:01 GMTfL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") Subject: Re: Why Delaware ?c8 Message-ID: <00A0B4E0.521C01A0@SSRL04.SLAC.STANFORD.EDU>  K In article <3C9AF7E3.CE5C444B@Omond.net>, Roy Omond <Roy@Omond.net> writes:i/ >Could someone please explain why HP and Compaq $ >are both incorporated in Delaware ? >i >Just out of interest ...i >r
 >Roy Omond >Blue Bubble Ltd.r >cM I remembered that a lot of companies were incorporated in Delaware because itvH had a particularly corporate-friendly set of laws, but I didn't know theB details.  Not all of these apply, but I'd think some of them do.    K A quick google search (on "Incorporation in Delaware") turned these up at acH site that says these are unique advantages of incorporating in the stateJ of Delaware.  (It doesn't mention that incorporation fees are cheaper thanJ any other state, but this appears to the case on one of the charts I saw.)  > No minimum capital is required to form a Delaware corporation.  F No corporate income tax for companies formed in Delaware and not doingK business in the state.  [I presume that's no _state_ corporate income tax.]e  4 Corporate records can be kept anywhere in the world.  K No formal meetings are required and shareholders need not be U.S. citizens.s  0 Any legal business can be conducted in Delaware.  = Ownership of a Delaware corporation is strictly confidential.a  F It is possible to have a corporation where one person acts as the only3 officer, director and shareholder of a corporation.t     Hope this helps!   -- Alanh    O ===============================================================================c0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056nM  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210rO ===============================================================================i   ------------------------------  # Date: Fri, 22 Mar 2002 11:31:15 GMTs( From: "Ken Farmer" <kfarmer@openvms.org> Subject: Re: Why Delaware ? < Message-ID: <7CEm8.2488$a3.1926898@typhoon.southeast.rr.com>  0 > Could someone please explain why HP and Compaq% > are both incorporated in Delaware ?l   Just a few reasons...o  : From: http://www.corpamerica.com/delawarecorporations.html  E - Delaware now allows for conversion of corporate entities - LLCs canrA convert into corporations and corporations can convert into LLCs!aG - Delaware state income tax is not levied on corporations which are notn doing business in Delaware.iL - Annual franchise tax is low (minimum is $30 tax plus $20 filing fee, total $50).i@ - One person can be the only Officer, Director, and Shareholder.D - Officers and Directors can be indemnified, limiting their personal
 liability.@ - Corporate books and records may be kept anywhere in the world.+ - No minimum amount of capital is required.a: - Non-resident shareholders pay no Delaware tax on shares.E - Shareholders are protected by takeover statue, which limits abusivem hostile takeover tactics.L% - Directors need not be shareholders.>; - Service from the State of Delaware is fast and efficient.r - Incorporation costs are low.K - Most Delaware corporations can be formed within minutes and documents arel  available within 24 to 48 hours.A - Delaware corporation law has well-established legal precedent*.'K - Delaware courts are respected nationwide for their expertise in corporatee matters.L - Voting provisions requiring greater-than-majority approval may be enacted.C - Liberal choice of corporate name provisions and ease of reservingt corporate name. 9 - Corporation may pay dividends from profits and surplus. G - Shareholders, directors and/or committee members may act by unanimousa, written consent in place of formal meetings.= - Directors may be given the power to make and alter by-laws.nI - Corporation may hold stocks, bonds or securities of other corporations,2K real and personal property, within or outside the state, with no limitation:
 as to amount.D? - Different kinds of business may be carried on in combination.@C - Corporation may fix quorum of board of directors -- not less thanOK one-third of the whole board; two if only two shareholders; one if only onen shareholder.5 - Voting trusts and voting agreements may be created.iB - Generally, stockholder liability is limited to stock held in the corporation.5 - Delaware law includes Close Corporation provisions.0+ - Classes of stock may be issued in series..  % Aslo see: http://www.delawarecorp.comf     -- Kenneth Farmer http://www.Tru64.org http://www.OpenVMS.org      , "Roy Omond" <Roy@Omond.net> wrote in message# news:3C9AF7E3.CE5C444B@Omond.net...d0 > Could someone please explain why HP and Compaq% > are both incorporated in Delaware ?- >- > Just out of interest ... >a > Roy Omond2 > Blue Bubble Ltd. >l   ------------------------------  % Date: Fri, 22 Mar 2002 14:34:38 +0000s From: Roy Omond <Roy@Omond.net>x Subject: Re: Why Delaware ? ) Message-ID: <3C9B40FE.5FF5BC41@Omond.net>F  * Alan Winston - SSRL Admin Cmptg Mgr wrote:   > [...snip...]H > No corporate income tax for companies formed in Delaware and not doingM > business in the state.  [I presume that's no _state_ corporate income tax.]   F Gosh !  Is that an incitement for HP and Compaq to NOT do any business
 in Delaware ?   J I can imagine the telephone conversation at the State Supercomputer Center of Delaware:   "Hello, is that Compaq ?"-   "Yes, Compaq Sales Force."  = "We'd like to place an order for 2,048 brand new Marvel boxesc
 with VMS."  8 "Sorry, you're in Delaware.  We won't sell them to you."  	 Roy Omondc Blue Bubble Ltd.  2 P.s. thanks to you and others for the explanation.   ------------------------------  % Date: Fri, 22 Mar 2002 15:23:04 +0100a= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>P Subject: Re: Why Delaware ?F) Message-ID: <3C9B3E48.68F90DF7@gtech.com>t   Roy Omond wrote:0 > Could someone please explain why HP and Compaq% > are both incorporated in Delaware ?k   I think it is a tax issue !h   Arne   ------------------------------   End of INFO-VAX 2002.160 ************************