1 INFO-VAX	Wed, 20 Mar 2002	Volume 2002 : Issue 156       Contents:* Re: Advertising CORBA services clusterwide* Re: Advertising CORBA services clusterwide Alphaserver 1000a Problem  Re: Andrew's back !  Re: Andrew's back ! ; Astat200-series memory (was: Re: WWW.DIGITAL-EQUIPMENT.COM)  BIND on VMS and new TLDs Re: BIND on VMS and new TLDs Re: BIND on VMS and new TLDs Re: BIND on VMS and new TLDs
 buffer I/O Re: buffer I/O Re: buffer I/O Re: buffer I/O% RE: Connecting to VMS Samba via win2k % RE: Connecting to VMS Samba via win2k / Re: DEC C: why does exit(0) really exit with 1? / Re: DEC C: why does exit(0) really exit with 1?  Re: FORTRAN ( Re: FORTRAN (was: Re: R.I.P. OpenVMS...)( Re: FORTRAN (was: Re: R.I.P. OpenVMS...)( Re: FORTRAN (was: Re: R.I.P. OpenVMS...)( Re: FORTRAN (was: Re: R.I.P. OpenVMS...) Free Airline Tickets< Re: How to install DSP3107L disk in a Storageworks canister? HP OfficeJet K80 and VMS Re: HP OfficeJet K80 and VMSG 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 P Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux LinuxLinF Re: HP's viewpoint on Linux, was: Re: Sun eating major helping ofLinux Re: LD062 Install Question4 MONI SYS shows 80% CPU usage and 8 jobs in COM state8 Re: MONI SYS shows 80% CPU usage and 8 jobs in COM state8 Re: MONI SYS shows 80% CPU usage and 8 jobs in COM state8 Re: MONI SYS shows 80% CPU usage and 8 jobs in COM state8 Re: MONI SYS shows 80% CPU usage and 8 jobs in COM state8 Re: MONI SYS shows 80% CPU usage and 8 jobs in COM state8 Re: MONI SYS shows 80% CPU usage and 8 jobs in COM state  OpenVMS 7.2-1 and UCX 4.2 ECO 05$ Re: OpenVMS 7.2-1 and UCX 4.2 ECO 05$ Re: OpenVMS 7.2-1 and UCX 4.2 ECO 05 Re: Plan B for Compaq  Right down the middle  Re: Right down the middle  RE: Right down the middle  Re: Right down the middle # Suggested enhancement. for DECterm. ' Re: Suggested enhancement. for DECterm.  Re: Talk about downtime....  telnetqueue  Re: Too close to call says BBC Re: Too close to call says BBC Too late to cry  Re: VMS DEBUG V7.2 bugs me! * Re: Witch streamer supported vax-4000-90 ? Re: Your mother is so fat...- Re: [Q]: Setting SCSI Cluster host SCIS ID's? - Re: [Q]: Setting SCSI Cluster host SCIS ID's?   F ----------------------------------------------------------------------  % Date: Wed, 20 Mar 2002 10:27:49 +0100 ' From: "Jakob Erber" <nospam@nospam.com> 3 Subject: Re: Advertising CORBA services clusterwide % Message-ID: <3c985616$1@news.post.ch>    Hello,  B Corba itself is not to strong in supporting failover without extraL programming. But what should be possible is, that you build the Corba ObjectH reference, using the TCPIP cluster alias. So if one nodes goes down, theH Corba Object on one of the remaining should be called. But this does not? help you, if not the node, but the corba Application goes down. I What I would probably do is, to let each Corba Application register there F own Object Ref in the Corba Naming Service. In case, one Object is notI reachable, the client hast do perform a reconnect to another Object. Then K remains the availablitiy of the Corba Naming Service :-). Best would be, to " contact it via cluster alias, too.   best regards   Jakob    --I What I publish in this group, is to be understood as my personal opninion A and must not be interpreted as a official statement of my company     3 "Anamika" <anamika@home.com> schrieb im Newsbeitrag 6 news:Xns91D6A06E0D501hemaniryahoocom@207.181.101.12... > Hi, G >    Iam interested in knowing whether CORBA services can be advertised  > clusterwide in a cluster. K > So, if a corba client requests for the service, any available node in the & > cluster will respond to the request.+ > This is for a failover situation in mind.  > 	 > Thanks,  > -A   ------------------------------  % Date: Wed, 20 Mar 2002 12:34:28 +0100 # From: "Manfred Reinart" <mr@ooc.de> 3 Subject: Re: Advertising CORBA services clusterwide 4 Message-ID: <a79s40$j1eop$1@ID-31703.news.dfncis.de>   Hi,   8 "Jakob Erber" <nospam@nospam.com> schrieb im Newsbeitrag news:3c985616$1@news.post.ch...  > Hello, > D > Corba itself is not to strong in supporting failover without extraG > programming. But what should be possible is, that you build the Corba  ObjectJ > reference, using the TCPIP cluster alias. So if one nodes goes down, theJ > Corba Object on one of the remaining should be called. But this does notA > help you, if not the node, but the corba Application goes down. K > What I would probably do is, to let each Corba Application register there H > own Object Ref in the Corba Naming Service. In case, one Object is notK > reachable, the client hast do perform a reconnect to another Object. Then J > remains the availablitiy of the Corba Naming Service :-). Best would be, to$ > contact it via cluster alias, too. >  > best regards >  > Jakob  >  > --K > What I publish in this group, is to be understood as my personal opninion C > and must not be interpreted as a official statement of my company  >  > 5 > "Anamika" <anamika@home.com> schrieb im Newsbeitrag 8 > news:Xns91D6A06E0D501hemaniryahoocom@207.181.101.12... > > Hi, I > >    Iam interested in knowing whether CORBA services can be advertised  > > clusterwide in a cluster. I > > So, if a corba client requests for the service, any available node in  the ( > > cluster will respond to the request.- > > This is for a failover situation in mind.  > >  > > Thanks,  > > -A >  >   I Another option would be usage of multiprofile IORs which contain the data @ for all running service instances on the differnt cluster nodes.I This permits client-side failover without referring to a central service. D An IMR (Implementation repository or a Load Balancer are other, more centralized approaches.   E Availability of transparent client side failover is ORB dependent, of  course.    Regards, Manfred    ------------------------------  # Date: Wed, 20 Mar 2002 16:00:41 GMT , From: "Patrick Coulier" <patrick@delight.be>" Subject: Alphaserver 1000a Problem6 Message-ID: <Jm2m8.291$GX2.1709@nreader1.kpnqwest.net>   Hello there,  L When trying to UPGRADE an old inherited Alphaserver 1000A 5/400 from NT to aH recent version of VMS, I get following message on the small LCD-display:   srom V1.0 cc   And the systems seems to hang.  = Any explanations or ideas about what is going on/wrong here ? , Any roadmaps available to do this operation.   Thanks in advance,   Patrick Coulier # System Engineer Delight I.S Belgium    ------------------------------  % Date: Wed, 20 Mar 2002 12:07:36 +0000 T From: Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com> Subject: Re: Andrew's back !& Message-ID: <3C987B88.3090707@sun.com>   Steve Reece wrote:  I > But would the memory be cache or main?  And is Rob likely to be static?  >  > :-)  >  > Steve     F Strangely enough I noticed a post to this group recently asking what aH particular failure that the poster had experienced on an Alpha CPU was. E One of the Compaq Choir unencumbered by any sense of irony helpfully  . told the poster that it was an ecache failure.   Regards  Andrew Harrison    .  >  > * > Andrew Harrison SUNUK Consultancy wrote: > I >>If you remember and I know this is tricky for you Rob, memory short and G >>long term not being one of your strong points you will also recollect E >>that I had a 3 month secondment into Product Marketing as part of a C >>job rotation. Most interesting it was to but I was glad to leave.  >>E >>You are a never ending souce of ammusement to me and despair to the ! >>OpenVMS community keep it up :)  >>	 >>Regards  >> >>Andrew Harrison  >> >    ------------------------------  # Date: Wed, 20 Mar 2002 13:09:43 GMT * From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: Andrew's back !@ Message-ID: <rS%l8.76435$1g.6183329@bin3.nnrp.aus1.giganews.com>  # "Andrew Harrison SUNUK Consultancy" = <andrew_nospam.remove_this.harrison@sun.com> wrote in message   news:3C987B88.3090707@sun.com... >  >  > Steve Reece wrote: > K > > But would the memory be cache or main?  And is Rob likely to be static?  > >  > > :-)  > > 	 > > Steve  >  > H > Strangely enough I noticed a post to this group recently asking what aI > particular failure that the poster had experienced on an Alpha CPU was. F > One of the Compaq Choir unencumbered by any sense of irony helpfully0 > told the poster that it was an ecache failure.  G Come back when you can identify that as a trend rather than an isolated J occurrence.  Any hardware can fail, but hardware that fails far more oftenL than its competition is what gets singled out for special notice, especiallyK when its vendor strives to hide the fact - witness the recent problems with ' IBM's 75GXP drives as a second example.    - bill   ------------------------------  + Date: Wed, 20 Mar 2002 12:24:42 +0000 (UTC) * From: Osmo Kujala <kujala@tukki.cc.jyu.fi>D Subject: Astat200-series memory (was: Re: WWW.DIGITAL-EQUIPMENT.COM), Message-ID: <a79v2a$fi4$1@mordred.cc.jyu.fi>  ! John Santos <JOHN@egh.com> wrote: * > On Fri, 15 Mar 2002, WILLIAM WEBB wrote:  7 >> They take 72 pin ECC SIMMs just like the 200 4/166s.  >>     ... ( > BTW, these are parity, not ECC, SIMMS.  ? Right, but in common language they are called ECC (Even more so " with SDRAM DIMMs ECC means 72-bit)   >  But I thought ECC: > memory worked by storing the ECC info in the parity bits8 > of multiple bytes (i.e. 8 ECC bits for each 64 bits of9 > memory), and the ECC calculation/comparison was done in : > the CPU memory adapter. In other words, there isn't such= > a thing as ECC memory per se, just a more sophisticated use  > of the memory parity bits.  J Of course there are/were real ECC SIMMs, but they are so rare, that let's  forget.   E BTW. Alphastation 200 doesn't even use all the ECC bits, only simple  E parity (33-bit memory). Maybe that doesn't apply to Alphastation 255, G because Astat200 mem is MSP01-AA,MSP01-AB... and for Astat255 MSP01-BA, G MSP01-BB... Commodity 36-bit (true parity) aka ECC SIMMs work for both.   I I seem to remember that AS200 takes SIMMS in pairs and AS255 in banks of   four. C Ahhh! Maybe the different order numbers only reflect the different   banking?   Osmo   ------------------------------  % Date: Wed, 20 Mar 2002 06:11:56 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> ! Subject: BIND on VMS and new TLDs , Message-ID: <3C986E62.9818A031@videotron.ca>  N With new TLDs such as .info, .biz, .aero coming to the internet, are there anyK changes that should be made to the BIND configuration files on a VMS system E that has its own BIND server, or does it get that type of information 7 automatically from the DNS server operated by the ISP ?    ------------------------------  % Date: Wed, 20 Mar 2002 08:17:06 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> % Subject: Re: BIND on VMS and new TLDs , Message-ID: <3C988BAB.52C24683@videotron.ca>   Roland Haider wrote:A > You have to update the list of root-servers. How depends on the / > tcpip stack, but i think you know what to do. D > Unless you add the root servers for a new TLD, all it's subdomains7 > are unreachable from computers using your nameserver.   K I just tried a site that is in the .info TLD ( http://www.stm.info ) and it N worked.  Is this because I have set forwarding server (my ISP's DNS servers) ?  L Looking at the file root.hint I don't really see anything specific to .TLDs,
 just letters.   E  Does this mean that if I want to resolve a .COM, it will first go to N C.ROOT-SERVERS.NET to find out who is responsible for .COM ? If so, that wouldH mean that BIND would just find .INFO by querying the I.ROOT-SERVER.NET ?  3 But how would it go about resolving a .ZA address ?    ------------------------------   Date: 20 Mar 2002 12:41:48 GMT0 From: roland.haider@at.bosch.com (Roland Haider)% Subject: Re: BIND on VMS and new TLDs 5 Message-ID: <91D78E154rolandhaideratboschc@127.0.0.1>   0 jfmezei.spamnot@videotron.ca (JF Mezei) wrote in" <3C986E62.9818A031@videotron.ca>:   E >With new TLDs such as .info, .biz, .aero coming to the internet, are I >there any changes that should be made to the BIND configuration files on G >a VMS system that has its own BIND server, or does it get that type of D >information automatically from the DNS server operated by the ISP ? >    Yes!  ? You have to update the list of root-servers. How depends on the - tcpip stack, but i think you know what to do. B Unless you add the root servers for a new TLD, all it's subdomains5 are unreachable from computers using your nameserver.    Regards,  
 Roland Haider    ------------------------------  % Date: Wed, 20 Mar 2002 13:43:03 -0000 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>% Subject: Re: BIND on VMS and new TLDs , Message-ID: <a7a3ks$17os@newton.cc.rl.ac.uk>  a "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message news:3C988BAB.52C24683@videotron.ca...   G >  Does this mean that if I want to resolve a .COM, it will first go to Q >  C.ROOT-SERVERS.NET to find out who is responsible for .COM ? If so, that would K >  mean that BIND would just find .INFO by querying the I.ROOT-SERVER.NET ?   H Nope. The root directory contains the top level directories (as on VMS).L All the root servers should know all the TLDs; it's just multiply redundant.   ------------------------------  % Date: Wed, 20 Mar 2002 22:35:22 +0800 + From: "Kenneth" <yeung_kenneth@hotmail.com>  Subject: buffer I/O * Message-ID: <a7a6nm$nll3@rain.i-cable.com>  K Beside terminal and network, what kind of IO will go to buffer IO? Is there ; any indication that if the process has very high buffer IO?    ------------------------------  % Date: Wed, 20 Mar 2002 17:39:55 +0100 ' From: "Joerg Spilling" <spi@equicon.de>  Subject: Re: buffer I/O / Message-ID: <a7aegv$d28$05$1@news.t-online.com>   G The driver programmes defines the kind of I/O per operation code of the G driver. A rule is, that DMA I/O is usually direct I/O. If you mean only I "usual" devices like disks, tapes, terminals, ... on an OpenVMS system, I @ think also that only terminal and network is using buffered I/O.  D To analyze the I/O traffic of a process I think you must look at the9 channels opened by it. In this way, you find the devices.      Joerg     < "Kenneth" <yeung_kenneth@hotmail.com> schrieb im Newsbeitrag$ news:a7a6nm$nll3@rain.i-cable.com...G > Beside terminal and network, what kind of IO will go to buffer IO? Is  there = > any indication that if the process has very high buffer IO?  >  >    ------------------------------  % Date: Wed, 20 Mar 2002 17:49:24 +0100 $ From: "labadie" <labadie_g@decus.fr> Subject: Re: buffer I/O 3 Message-ID: <p63m8.1260$fL6.25775@news.cpqcorp.net>   2 "Joerg Spilling" <spi@equicon.de> wrote in message) news:a7aegv$d28$05$1@news.t-online.com... I > The driver programmes defines the kind of I/O per operation code of the I > driver. A rule is, that DMA I/O is usually direct I/O. If you mean only K > "usual" devices like disks, tapes, terminals, ... on an OpenVMS system, I B > think also that only terminal and network is using buffered I/O. > F > To analyze the I/O traffic of a process I think you must look at the; > channels opened by it. In this way, you find the devices.   , To analyze the I/O traffic of a process, tryL http://www.compaq.com/support/asktima/communications/CHAMP_SRC010730002768.h tml    Regards    Grard   ------------------------------  # Date: Wed, 20 Mar 2002 17:25:08 GMT + From: Ryan Moore <rmoore@rmoore.dyndns.org>  Subject: Re: buffer I/O < Message-ID: <Pine.LNX.4.31.0203200924030.2027-100000@jaipur>  G Mailbox I/O (ie., MBAxxx: devices) are considered buffered I/O as well. F And not all network traffic is buffered I/O.  At least under Multinet,* IP network traffic shows up as direct I/O.  # On Wed, 20 Mar 2002, Kenneth wrote: M > Beside terminal and network, what kind of IO will go to buffer IO? Is there = > any indication that if the process has very high buffer IO?    ------------------------------   Date: 20 Mar 2002 03:53 CST ' From: carl@gerg.tamu.edu (Carl Perkins) . Subject: RE: Connecting to VMS Samba via win2k- Message-ID: <20MAR200203534998@gerg.tamu.edu>i  , "Bruin, J.M. de" <Bruin@WT.TNO.NL> writes...' }From: Roy Omond [mailto:Roy@Omond.net]eK }>As far as I'm aware, you'll need at least Samba 2.0.6 (or was it 2.0.8 ?)  }>to work with W2K.L }> e }>Roy Omond  } N }Well, if that is true, I do have a configuration here that is not supposed to }work, but works!  } L }According to SMBSTATUS, the version running on the VMSbox, is 1.9.16p2 !!!! }  }Marky  D I'm using 2.0.3 and it works (if imperfectly) with both Windows2000 C (I think - we only have maybe two W2K systems) and WindowsXP (which , we have about 4 of) as well as plain old NT.  @ We were at the same, or nearly the same, version as you but with? that version we didn't show up in the network neighborhood treekA even on NT (it did show up, then stopped for reasons that I neverc? actually tracked down). You could still map drives, but you hadg@ to know the "magic incatation" and just enter the right node and@ share name into the appropriate place in the Map Drive dialog in@ the proper format (it seems like 2/3 of the time I use the wrong= direction slashes on the first try). Upgrading to 2.0.3 fixedp that.    --- Carl   ------------------------------  % Date: Wed, 20 Mar 2002 11:19:01 +0100,( From: "Bruin, J.M. de" <Bruin@WT.TNO.NL>. Subject: RE: Connecting to VMS Samba via win2kC Message-ID: <EC85E7391071D511AC140008C7F37BC24DEE01@wt15.wt.tno.nl>a  P With 1.9.16p2 the shares show up in networkneighborhood as well as in the browse" of map network drive on my NT box.  2 I haven't checked it for W2K (that is my home-pc).   Mark   -----Original Message-----B From: carl@gergl1.gerg.tamu.edu [mailto:carl@gergl1.gerg.tamu.edu]" Sent: woensdag 20 maart 2002 10:53 To: Info-VAX@Mvb.Saic.Com . Subject: RE: Connecting to VMS Samba via win2k    , "Bruin, J.M. de" <Bruin@WT.TNO.NL> writes...' }From: Roy Omond [mailto:Roy@Omond.net]oK }>As far as I'm aware, you'll need at least Samba 2.0.6 (or was it 2.0.8 ?)s }>to work with W2K.D }> m }>Roy Omonds } N }Well, if that is true, I do have a configuration here that is not supposed to }work, but works!B } L }According to SMBSTATUS, the version running on the VMSbox, is 1.9.16p2 !!!! }  }Mark   D I'm using 2.0.3 and it works (if imperfectly) with both Windows2000 C (I think - we only have maybe two W2K systems) and WindowsXP (which , we have about 4 of) as well as plain old NT.  @ We were at the same, or nearly the same, version as you but with? that version we didn't show up in the network neighborhood treejA even on NT (it did show up, then stopped for reasons that I never5? actually tracked down). You could still map drives, but you hadn@ to know the "magic incatation" and just enter the right node and@ share name into the appropriate place in the Map Drive dialog in@ the proper format (it seems like 2/3 of the time I use the wrong= direction slashes on the first try). Upgrading to 2.0.3 fixed  that.u   --- Carl   ------------------------------  % Date: Wed, 20 Mar 2002 03:43:56 -0500t- From: JF Mezei <jfmezei.spamnot@videotron.ca>w8 Subject: Re: DEC C: why does exit(0) really exit with 1?, Message-ID: <3C984BC0.60F7B757@videotron.ca>   remembered this thread.O  E HELP CC RUN EXIT reveals that the behaviour is documented  exit(0) oreQ EXIT_SUCCESS are translated to SS$_NORMAL, all other values are passed unchanged.-   ------------------------------  % Date: Wed, 20 Mar 2002 15:20:24 +0100 = From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>p8 Subject: Re: DEC C: why does exit(0) really exit with 1?) Message-ID: <3C989AA8.EB365ED2@gtech.com>2   Michael Zarlenga wrote: * > Carl Perkins <carl@gerg.tamu.edu> wrote:H > : Since your C program is doing what C programs tend to do to indicateD > : success, it makes a lot more sense to report a status value that > : translates to the message  > ? > From my point of view, the program is telling the RTL to exit C > with a $STATUS of 0, nothing else.  It's assuming something about0) > the value 0 and changing it on me to 1.u  D I do not think that the ANSI C standard specifies that exit(0) shold exit with $STATUS = 0 on VMS !  A > Let's cut to the chase ... is there any easy way to change this B > (a different header file, a different RLT func., etc) so that an@ > exit(0) or return (0) (in main()) will return a ZERO to DCL as > the $STATUS?  > Just call SYS$EXIT. It exits to VMS with the code you specify.   Arne   ------------------------------  + Date: Wed, 20 Mar 2002 17:44:17 +0100 (MET)r9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>f Subject: Re: FORTRAN; Message-ID: <01KFLBET16OI8ZQP5J@sysdev.deutsche-boerse.com>,    > |> >         include 'xxx.inc' > |> rJ > |> And that is not exactly a new feature in DEC/Digital/Compaq Fortran ! > K > Not sure it was ever part of the language standard, either.  According to B > the Prime Programmer's Companion for Fortran77 they use $INSERT.  H INCLUDE was a common extension offered by many compilers, including the G venerable VAX FORTRAN compiler.  Syntax differed from vendor to vendor. 4 It was introduced into the standard with Fortran 90.  H Apart from its "obvious" uses, INCLUDE was often used to make sure that G COMMON block definitions were the same in all routines which used them.1D Since Fortran 90 also introduced modules and thus global variables, $ COMMON blocks were no longer needed.  H Fortran 90 also introduced array-valued functions and generic functions.F When writing one's own, all array ranks have to be covered, as well asH all data types.  Since this means a lot of repeated code, INCLUDE comes C in handy here.  Fortran 95 introduced ELEMENTAL procedures, so the nH repeated code for covering the various array ranks is no longer needed, D but still is for covering the various data types.  Thus, INCLUDE is C still "needed", though many other uses can now be done better with h modules.   ------------------------------  % Date: Wed, 20 Mar 2002 08:12:48 +0100r- From: Jouk Jansen <joukj@hrem.stm.tudelft.nl>a1 Subject: Re: FORTRAN (was: Re: R.I.P. OpenVMS...)e3 Message-ID: <3C984480.34B913E7@hrem.stm.tudelft.nl>    Steve Lionel wrote:u > 1 > On Mon, 18 Mar 2002 08:38:15 +0100, Jouk Jansen1$ > <joukj@hrem.stm.tudelft.nl> wrote: >  [snip] > I > > However, some Fortran compilers (even CompaqFortran on Tru64) support H > >proprocessing and thus the "#include <xxx.h>" syntax. For some reasonG > >this was never implemented in CompaqFortran for OpenVMS. Currently Id/ > >(mis)use the C-prepocessor for this purpose.d > F > Right - the UNIX compiler provides a cpp-style preprocessor, but notH > on VMS.  We never saw enough demand for this from VMS users to make it8 > work there, it would have been a non-trivial exercise. > A Probably most of us have both Fortran and C-compilers and in need H of preporecssing the C-preprocessor can be used, but including it in theE Fortran-command it would clean up a lot in my make-files for OpenVMS.b                          Jouk-   ------------------------------  % Date: Wed, 20 Mar 2002 15:16:06 +0100t= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>e1 Subject: Re: FORTRAN (was: Re: R.I.P. OpenVMS...) ) Message-ID: <3C9899A5.6A6C312C@gtech.com>y   Steve Lionel wrote:t, > On Fri, 15 Mar 2002 15:11:11 GMT, ">>> ^P"( > <plj@NOSPAM.byron.ext.telia.se> wrote:. > >Will we be able to use #include <xxx.h>   ? > ' > Wrong language.  In Fortran, you say:r >  >         include 'xxx.inc'n  E And that is not exactly a new feature in DEC/Digital/Compaq Fortran !n   Arne   ------------------------------   Date: 20 Mar 2002 15:47:24 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)>1 Subject: Re: FORTRAN (was: Re: R.I.P. OpenVMS...)d, Message-ID: <a7aauc$1hsg$1@info.cs.uofs.edu>  ) In article <3C9899A5.6A6C312C@gtech.com>, @  Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> writes: |> Steve Lionel wrote:/ |> > On Fri, 15 Mar 2002 15:11:11 GMT, ">>> ^P":+ |> > <plj@NOSPAM.byron.ext.telia.se> wrote:.1 |> > >Will we be able to use #include <xxx.h>   ?r |> > N* |> > Wrong language.  In Fortran, you say: |> > o |> >         include 'xxx.inc' |> 9H |> And that is not exactly a new feature in DEC/Digital/Compaq Fortran !  I Not sure it was ever part of the language standard, either.  According to @ the Prime Programmer's Companion for Fortran77 they use $INSERT.   bill   -- eJ 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>   n   ------------------------------  % Date: Wed, 20 Mar 2002 12:04:04 -0500r+ From: Steve Lionel <Steve.Lionel@intel.com>o1 Subject: Re: FORTRAN (was: Re: R.I.P. OpenVMS...) 8 Message-ID: <l6gh9u817i96g961e2f4ou97toutt7b08j@4ax.com>  < On 20 Mar 2002 15:47:24 GMT, bill@triangle.cs.uofs.edu (Bill Gunshannon) wrote:  + >|> > Wrong language.  In Fortran, you say:- >|> >  >|> >         include 'xxx.inc'0 >|> I >|> And that is not exactly a new feature in DEC/Digital/Compaq Fortran !m >rJ >Not sure it was ever part of the language standard, either.  According toA >the Prime Programmer's Companion for Fortran77 they use $INSERT.0  B It became standard in Fortran 90, though it was almost universally# implemented in Fortran 77 products.e      D Please send Visual Fortran support requests to vf-support@compaq.com   Steve Lionel Compaq Fortran Engineering Intel Corporation 
 Nashua, NH  . Compaq Fortran - http://www.compaq.com/fortranK Intel Fortran - http://developer.intel.com/software/products/compilers/f50/    ------------------------------  % Date: Wed, 20 Mar 2002 10:14:19 -0800aC From: "Special Offers" <SpecialOffers@star82.allbestcheapstuff.com>  Subject: Free Airline Ticketss9 Message-ID: <iss.4512.3c98b6a0.ca3c8.1@mx2.west.saic.com>e  4 <!-- saved from url=(0022)http://internet.e-mail --> <html>   <head>  , <title>Best Cheap Stuff Virtual Mall</title>  H <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">\ <link rel="stylesheet" href="http://www.bestcheapstuff.com/email/email.css" type="text/css">   <!--more scripts here-->   </head>C   <body bgcolor="#FFFFFF" text="#000000" leftmargin="0" topmargin="0" marginwidth="0" marginheight="0" link="#FF0000" vlink="#FF0000" alink="#FF0000">c <table width="500" border="1" align="center" bordercolor="#000000" cellpadding="0" cellspacing="0">0   <tr>     <td>S       <table width="500" border="0" cellpadding="0" cellspacing="0" align="center">                   <tr>
                      <td bgcolor="#000000" height="6" colspan="2"><img src="http://www.bestcheapstuff.com/images/transparent.gif" width="1" height="6" border="0"></td>
         </tr>t                  <tr valign="middle">
           
             <td width="240" class="heading">       <img src="http://www.bestcheapstuff.com/images/template/logo_ani.gif" height="114" border="0" alt="Best Cheap Stuff " vspace="0" hspace="0" width="211" align="left"></td>
           < <td align="center" class="heading">Free Airline Tickets</td>
         </tr>                   <tr valign="top">r
           
           <td height="25" background="http://www.bestcheapstuff.com/images/template/topbar_bkgd.gif" valign="middle" class="menu"><img src="http://www.bestcheapstuff.com/images/template/topbar_left.gif" width="198" height="25" border="0" align="left" vspace="0" hspace="0"></td>
            <td height="25" background="http://www.bestcheapstuff.com/images/template/topbar_bkgd.gif" valign="middle" class="menu" align="right">&nbsp;</td>s
         </tr>s(         <tr valign="top" align="center">C           <td height="25" valign="middle" class="menu" colspan="2">              <p>&nbsp;</p>d             o <table width="90%" height="50" cellpadding="2" cellspacing="0" border="0" align="center" bordercolor="#FFFFFF">                 <tr valign="middle">                 - <td align="center" class="menu" valign="top">mS <p align="left"><font face="Arial, Helvetica, sans-serif" size="2" color="#000000">d    r   </font>7? <table width="100%" border="0" cellspacing="0" cellpadding="5">9 <tr>  <td align="center" valign="top">M <p><font face="Arial, Helvetica, sans-serif" size="2" color="#000000"><b><br>a </b></font></p>b <hr width="80%"> <br> <font face="Arial, Helvetica, sans-serif" size="2" color="#000000"><b><font size="3">2 FREE round trip Airline Tickets</font> </b></font><font face="Arial, Helvetica, sans-serif" size="3" color="#000000"><b>to your choice of:<br>n <br>/   </b><font color="#FF0000" size="5">Hawaii<br>o Orlando<br>u
 Las Vegas<br>e London</font><b><br>4 <font size="2">and 20 other destinations!</font><br> <br> </b></font> J <font face="Arial, Helvetica, sans-serif" size="3" color="#000000"><b><br> </b></font><font face="Arial, Helvetica, sans-serif" size="2" color="#000000"><b>This offer is only valid for 48 hours so Act Now!</b></font>w <br> <br> <hr width="80%"> </td>e <td align="center">a <p><img   > src="http://www.ahamembership.com/assoc_images/fly-away1.jpg"   e width="199" height="297" border=0><font face="Arial, Helvetica, sans-serif" size="2" color="#000000">o <br>G <input type="text" name="name" size="30" value="info-vax@mvb.saic.com">y                 <br>               <font color="#FF0000"><a href="http://www.bestcheapstuff.com/cgi-bin/ahasub1.cgi?name=info-vax@mvb.saic.com"><font size="3"><font size="4">CLICK HERE!<br>5  to Opt-In For More Information</font></font></a><br>2 <br> </font></font></p> </td>  </tr>k       </table> </td>u </tr>b                                                                                                                    <tr>                 % <td valign="top" class="subcategory">b                   <p class="contentblack"><font color="#000066"><font face="Arial, Helvetica, sans-serif" size="2">You are receiving this because you are on  the special offers list of Bestcheapstuff.com. If you would no longer like to receive special offers from Bestcheapstuff.com, <a href="http://unsubscribe.bestcheapstuff.com" target="_blank">please click here</a> and you will be promptly unsubscribed. For additional information or comments <a href="mailto:info@bestcheapstuff.com">contact us</a>.</font></font></p> </td>r </tr>                                        </table>             <p>&nbsp;</p>v           </td> 
         </tr>               </table>	     </td>    </tr>i </table> </body>l </html>-   ------------------------------   Date: 20 MAR 2002 16:18:15 GMT+ From: Dave Greenwood <greenwoodde@ornl.gov>tE Subject: Re: How to install DSP3107L disk in a Storageworks canister?-2 Message-ID: <20MAR02.16181583@feda01.fed.ornl.gov>  V In a previous article, "news-server.rochester.rr.com" <kevino@rochester.rr.com> wrote:K > I  believe one of those jumpers is for drive spin up, at least this couldnD > explain the medium offline error. Does the drive spin on power up?N > if not try one jumper at a time and just power up until the drive spins?  OrJ > the spin up jumper could be on the bottom of the drive, same suggestion.  C That jumper is already set.  At least, I assume the jumper labelledtD "Spin Up" means to spin up on power up.  It's pretty hard to tell if< the disk is actually spinning up or not since the fan in the" StorageWorks pizza box is so loud.  G Isn't there a SCSI command to tell a drive to spin up?  Is there a toolp3 that would allow me to issue that command from VMS?P   Thanks,0 Dave --------------9 Dave Greenwood                Email: Greenwoodde@ORNL.GOV,H Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------  % Date: Wed, 20 Mar 2002 08:17:16 -05000* From: "Stanley F. Quayle" <stan@stanq.com>! Subject: HP OfficeJet K80 and VMS - Message-ID: <3C98458C.4755.5323B0B@localhost>   D I have an HP OfficeJet K80 connected to the network via a Jetdirect B 170X box.  The Jetdirect box allows PC's on the network to print,  scan, and fax.  D I'd like to be able to print from my VMS machines.  For non-windows D machines, the box supports FTP.  ASCII FTP transfer has the printer 3 assume text; binary FTP has the printer assume PCL.(  / For example, the following prints my LOGIN.COM:a  -   $ copy/ftp/ascii login.com 192.168.60.251::a  C I have DCPS installed on my cluster; can I use it to speak to this oF box?  Or is there a "dumb" print symbiont that I can use for text, at  least?    
 --Stan Quayle8! President, Quayle Consulting Inc.>  
 ----------G Stanley F. Quayle, P.E.   N8SQ   +1 614-868-1363   Fax: +1 614 868-1671o1 8572 North Spring Ct. NW, Pickerington, OH  43147A= Preferred address:  stan@stanq.com       http://www.stanq.comM   ------------------------------  # Date: Wed, 20 Mar 2002 14:07:12 GMT0( From: "Mark E. Levy" <mlevy70@attbi.com>% Subject: Re: HP OfficeJet K80 and VMS = Message-ID: <kI0m8.84018$uA5.78482@rwcrnsc51.ops.asp.att.net>   H You didn't say who's TCP/IP stack you have installed on your VMS box, soH you'll have to do a bit of research, but here's how to set up a queue to print to that printer:  F $ init/queue/start/on="192.168.60.251,9100"/processor=<processor name> <queue name>  H The processor name is either UCX$TELNETSYM or TCPIP$TELNETSYM for CompaqL TCPIP Services depending on the version, or TCPWARE$TELNETSYM for TCPware. II don't remember what Multinet uses. Also, you may need to change the comma.C before the port number (9100) to a colon. Check your documentation.a   DCPS can be used for this also.,  	 Mark Levyn SMA.    5 "Stanley F. Quayle" <stan@stanq.com> wrote in messagea' news:3C98458C.4755.5323B0B@localhost...AE > I have an HP OfficeJet K80 connected to the network via a Jetdirect C > 170X box.  The Jetdirect box allows PC's on the network to print,i > scan, and fax. > E > I'd like to be able to print from my VMS machines.  For non-windowsaE > machines, the box supports FTP.  ASCII FTP transfer has the printerg5 > assume text; binary FTP has the printer assume PCL.o >X1 > For example, the following prints my LOGIN.COM:a >a/ >   $ copy/ftp/ascii login.com 192.168.60.251::i >.D > I have DCPS installed on my cluster; can I use it to speak to thisG > box?  Or is there a "dumb" print symbiont that I can use for text, atD > least? >  >n > --Stan Quayle # > President, Quayle Consulting Inc.o >e > ----------I > Stanley F. Quayle, P.E.   N8SQ   +1 614-868-1363   Fax: +1 614 868-1671r3 > 8572 North Spring Ct. NW, Pickerington, OH  43147 ? > Preferred address:  stan@stanq.com       http://www.stanq.com0 >    ------------------------------  % Date: Wed, 20 Mar 2002 10:34:28 +0000vT 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: <3C9865B4.5060408@sun.com>   Fred Kleinsorge wrote:    > Bill Todd wrote in message ...  N > None of that has any bearing on your thesis.  While many of us believed thatN > converting VMS customers into NT customers was as bad an idea as the earlierL > attempt to convert them into UNIX customers -- the belief by many was thatL > VMS was a dying business, and NT was the wave of the future - and *it* wasM > what would driver Alpha volume.  So the strategy was to not try and reverserM > the VMS trend, but to try and capture as many of those customers with NT as( > an alternative.  >     F There is a very big different between to trying to reverse a trend andE actively trying to accelerate it. Digital and Compaq should be judged E on the basis of their actions rather than their words, and anyone whouC has ever delt with either organisation would can only conclude that2 words are cheap.  E Digitals decision to in effect stop selling All-In-One on OpenVMS in RI favour of Office and Exchange on NT is just one example of an depressing E= list of actions which served to accelerate OpenVMS's decline.l     Regardsu   Andrew Harrison    ------------------------------  % Date: Wed, 20 Mar 2002 11:06:04 +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: <3C986D1C.2040805@sun.com>  B > "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message/ > news:0vol8.1113$fL6.23977@news.cpqcorp.net...h > B >>Remember, the new Sun keyword is "commercial" UNIX as opposed toK >>"proprietary" UNIX ;-).  They'll try to find ways to belittle it and tell L >>you why it isn't ready-for-prime-time.  I think I hear Rome burning in the >>background as Andrew fiddles.S >> >> >     D Fred it would be better if you posted about something you understoodC rather than something you clearly don't. Perhaps you should confine 6 yourself to alt.folklore or alt.alternative.universes.  @ However in the real world Linux isn't ready for primetime as an = enterprise class OS. Apart from anything it does not scale to ? much more than 4 CPU's. One customer I work with is trialing it/? for compute servers and small web/apps servers. They have had a B team of engineers working on integrating OpenSource and Commercial@ tools, filesystems etc into their distribution to get Linux to aB state where they are confident that it will meet even these rather> limitted requirements. Almost all the work they have had to do? themselves or pay other people to do is unecessary on either of"> the two commercial UNIX's that they also use. They have had to) start at the filesystem and work upwards.i  ? How many of your customers would replace the OpenVMS filesysteml9 or the Tru64 filesystem because it isn't robust enough ??o  > The wonderfull thing about the BS you seem to have decended toC spouting is that is so applicable to Compaq. You are the archetypalw# thrower of stones in a glass house.e  ? Linux represents a much bigger threat to Compaq than it does to B Sun. Solaris has the largest number of commercial apps appart fromE Win32 and it offers excelent capabilites for running both Linux (GNU)vH and Win32 apps. OpenVMS offers neither and has a declining SW portfolio,C Tru64 has no future in HP so thats going, Tandem has a declining SW- base.-  B Linux will not kill off all the Commercial UNIX's any more than NT@ did. What it is most likely to do is kill the marginal platformsE are ISV's rationalise the number of ports that have and add Linux/x86 G to their platform this is what happened with NT. Compaqs brands appart n' from WinTel are all marginal platforms.:  D Most analysts are looking at Win32/64, Solaris, AIX and possibly but> not necessarely HP-UX depending on if HP bungle the PA to IA64C transition or not. Everything else is history and that I am affraida
 means you.  B If you can explain what your strategy is to avoid this then pleaseC feel free to do so, past attemps by you or any other members of the,* Compaq Choir have been less than tunefull.  A As I said earlier you seem to be spending an inordinate amount ofeD effort on criticising and FUDDING Sun is it because you realise that> your customers respect and understand Sun's strategy and don't understand or respect yours.     Regardst   Andrew Harrisond   ------------------------------    Date: 20 Mar 2002 08:41:14 -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: <xuvQjGn6fIQw@eisner.encompasserve.org>R  } In article <3C986D1C.2040805@sun.com>, Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com> writes:^ > C >> "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in messagei0 >> news:0vol8.1113$fL6.23977@news.cpqcorp.net... >> yC >>>Remember, the new Sun keyword is "commercial" UNIX as opposed to L >>>"proprietary" UNIX ;-).  They'll try to find ways to belittle it and tellM >>>you why it isn't ready-for-prime-time.  I think I hear Rome burning in theB  >>>background as Andrew fiddles. >>>( >>>R >> I >  > F > Fred it would be better if you posted about something you understoodE > rather than something you clearly don't. Perhaps you should confineh8 > 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. w    D 	Oh?  Is that from "talking points" on company literature or is that 	real world?  
 Exec Summary:   I big sweep of its computer operations, replacing 60 different servers fromEK different vendors with a single computer having 500 virtual servers runninga Linux.  9 http://www.open-mag.com/features/Vol_15/sonera/sonera.htmr  N The deal is not the first for IBM in the telecom sector. Before Sonera Entrum,N there was Sweden's telecom giant, Telia, that declared it was ready to go withM virtual server approach. In December 2000, IBM and Telia announced that Telia I would create1,500 virtual Linux machines on an IBM mainframe with "Shark" O storage to run Internet and IP-VPN service operations across Scandinavia. StorymO has it that Telia was briefed by IBM at a Linux integration center and made theg. decision in a matter of days after the visit.   M The Sonera/Telia deal numbers are very similar. Telia replaced 70 web-hosting M Unix servers with 1,500 virtual Linux servers. Both are consolidating using a N single IBM S/390/G6 (zSeries) server in conjunction with a Shark storage unit.O What's more, the motivation for both telecoms to consolidate was not to achieveeG cost savings alone, but high availability and reliability assurance form customers.     [snip]  N Small wonder, then that Sonera Entrum made its move and announced in October aI big sweep of its computer operations, replacing 60 different servers from K different vendors with a single computer having 500 virtual servers runningFN Linux. The vendors who gained from this decision: IBM impressed Sonera Entrum,L with the promise of rock-solid and high performing hardware, along with SuSE9 Linux Enterprise Server and other Open Source solutions. <  * > Apart from anything it does not scale to > much more than 4 CPU's.   # 	In most cases, it doesn't need to.8  ' One customer I work with is trialing itrA > for compute servers and small web/apps servers. They have had a D > team of engineers working on integrating OpenSource and CommercialB > tools, filesystems etc into their distribution to get Linux to aD > state where they are confident that it will meet even these rather@ > limitted requirements. Almost all the work they have had to doA > themselves or pay other people to do is unecessary on either oft@ > the two commercial UNIX's that they also use. They have had to+ > start at the filesystem and work upwards." > A > How many of your customers would replace the OpenVMS filesystemb; > or the Tru64 filesystem because it isn't robust enough ??g > @ > The wonderfull thing about the BS you seem to have decended toE > spouting is that is so applicable to Compaq. You are the archetypalc% > thrower of stones in a glass house.i > A > Linux represents a much bigger threat to Compaq than it does to" > Sun.    ? 	Not really.  Sun has the most installed Unix base by far.  ThegH 	most to lose.  That next application that runs on Linux, why not stick 9 	it on racks of Intel kit?  Cheaper hardware and free OS.0  = Solaris has the largest number of commercial apps appart fromcG > Win32 and it offers excelent capabilites for running both Linux (GNU) J > and Win32 apps. OpenVMS offers neither and has a declining SW portfolio,E > Tru64 has no future in HP so thats going, Tandem has a declining SWd > base.s > D > Linux will not kill off all the Commercial UNIX's any more than NTB > did. What it is most likely to do is kill the marginal platformsG > are ISV's rationalise the number of ports that have and add Linux/x86eI > to their platform this is what happened with NT. Compaqs brands appart l) > from WinTel are all marginal platforms.  >   G 	Won't kill it off?  Of course, you are leaving out the new terminologye@ 	to help in thinking.  "Linux is for edge computing."  Guess theB 	folks at Sonera/Telia are rare exceptions with few to follow.  Or9 	more likely, more than a few are knocking on IBM's door.    				Rob    ------------------------------  % Date: Wed, 20 Mar 2002 09:38:20 -0500"5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>aP Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux3 Message-ID: <Gf1m8.1256$fL6.25647@news.cpqcorp.net>g  2 Andrew Harrison SUNUK Consultancy wrote in message <3C986D1C.2040805@sun.com>...= >0C >> "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message"0 >> news:0vol8.1113$fL6.23977@news.cpqcorp.net... >>C >>>Remember, the new Sun keyword is "commercial" UNIX as opposed totL >>>"proprietary" UNIX ;-).  They'll try to find ways to belittle it and tellI >>>you why it isn't ready-for-prime-time.  I think I hear Rome burning inr the<  >>>background as Andrew fiddles. >>>b >>>f >> >e >"E >Fred it would be better if you posted about something you understood"D >rather than something you clearly don't. Perhaps you should confine7 >yourself to alt.folklore or alt.alternative.universes.o >=  J The amount of VMS knowledge you posess could fit on the head of a pin withJ room to spare - yet you have no problem making uninformed prognostications here.w   ------------------------------  % Date: Wed, 20 Mar 2002 16:08:12 +0000"' From: Andrew Harrison SUNUK ConsultancyaY Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping of Linux LinuxLin"& Message-ID: <3C98B3EC.8040101@sun.com>  ( Andrew Harrison SUNUK Consultancy wrote:   Fred Kleinsorge wrote:F  >When did you swallow the Compaq marketing line about SPARC being theL  >"proprietary" and your implied assumption that POWER and IA64 are not ????  >3  >Was it after the Alpha announcements of before ??   >  I  >I claim that Solaris is a proprietary UNIX.  IA64, POWER4 and Sparc are F  >proprietary hardware - there *is no* "open" chip architecture.  But 
  >SPARC isL  >old and slow, and not owned by a chip maker that can drive the >technologyI  >and process - and the performance gap is widening to the point that therI  >*despite* shaving your margins to non-existant - you are losing market l  >shares)  >(IBM now clains to have surpassed you).e  ? There you go again. Fred when will you learn, stick to what you/? know, I havn't discovered what it is yet from your posts but it<+ isn't the BS you are posting at the moment.r  B SPARC is not old and slow, who gives a rats arse about SPECint and@ SPECfp certainly not any of your commercial customers. They care? about applications performance, price, if the app actaully runs<? on the platform you are proposing to them etc. Guess what "old" @ SPARC based systems beat you hands donw in all these areas. They> are faster, look at SAP, PeopleSoft, Oracle Apps etc, they areC cheaper, the apps actually run on SPARC Solaris and we can even getL> sales and technical people in front of customers who know what? our boxes do. Sadly you can't in any of these areas. Where doeshD Compaq lead any benchmarking table excluding SPEcint and SPECfp, ohhE I forgot you can get the most 8 CPU machines into a benchmark center.e  D In fact the only thing that is old is your argument, you claim SPARCD is slow when pushed you will respond with SPECint and SPECfp numbersD which no one uses to buy systems. You will then thrash arround a bit@ and find no other evidence to support your conjecture and so theC argument will end. Whats so boring and sad is that this is the same A argument Digital was using to fail to sell Alphas all those yearst1 ago, they never learn't and you obviously havn't.C    ,  > Solaris is a lock-in to Sparc and to Sun.  A So explain to me why a customer should buy OpenVMS, Tandem, HP-UXs> or Tru64. The first two are a total lockin once on you have to= fight to escape.Moving from Linux to Solaris or the other wayL< arround is trivial, moving from either of your cash cow OS's, to anything else is much more problematical.  < Incedentally there is no one version of Linux there are lots: and they differ in where things are installed and how they@ are managed, RedHat is different to Suse etc. This is remarkably? like the Commercial UNIX market, similar if not identical API's = different to manage. We no see customers "standardising" on a 9 particular Linux distribution which then excludes others.r    I  > Where is the high end single stream Sparc to compete with even the > .rI  > x86? When IA64 prices start to drop, you will not have the high end or G  > the low end.  I'd love to see exactly what your margins are on thesemH  > sub $1k systems.  Just trying to hang on to market share at any cost?$  > With the lowest performing chips?  D The margins are OK we make money, thin but OK, what about Compaqs PCG business, thin to negative margins. Actually thats the crux of why your8D argument is bankrupt. Only three vendors make money at being Wintel B vendors, Dell, Intel and Microsoft. HP don't you don't, IBM don't.C How long do you expect people to continue to finance something thatM$ basically desn't deliver any return.  D As for performance I see you are leading towards SPECint and SPECfp,A when you find that the majority of your customers care about this A get back to me. Incedentally the low end appliance space that ouraB Sub 1000 dollar boxes play in and yours don't is not generally CPUF intensive unless you think serving up HTTP or files is a CPU intensiveF operation. But even if we do use SPECint and SPECfp SPARC has slightlyA worse SPECint performance than the best Wintel box and better FP.R  F Your chosen platform IA64 has much worse SPECint performance which mayF or may not be fixed by Mckinley. On the downside IA64 is expensive andA very hot, when its slowness, heat and price are fixed it might bea# competitive at the moment it isn't.e    B  > Ooooh.  Again.  What's your margins here?  You give away cheap E Solaris > on the low-end, and sell hardware that is slow, but cheap. d3 Hoping to > make it up on service and the high end?   D Solaris is basically the same across the whole range, we make enoughA money from the entire range to fund its development and providinguC cheap boxes saves our customers from the nightmare you would imposer@ on them of running Win32 while encouraging SW vendors to port to@ Solaris. The fact that this seems inexplicable or strange to you- says much about your powers of comprehension.   >  > IBM's gonna eat your lunch with it.  I think the IBM Linux  commercials > arehF  > crap.  But I bet that the people who actually spend the money are >  > being convinced.g  H Has IBM suddenly become a partner of yours ?? they are after your market6 sahre as well which is much more vunerable than Sun's.      >D  >You also betray your breathtaking ignorance about SPARC processorsG  >when you say that SUN fabs SPARC CPU's, we never have, SUN has always2J  >been a FABless CPU vendor. Our CPU's are FABBED by Texas and a number ofG  >other manufacturers. Sadly this one fact alone destroys your argument:0  >since we don't need to invest in the FAB Ramp.  >  I  > Who said you directly fabbed your chips?  In fact, this is one of the e  > bigI  > problems you have - you have to rely on others to fab your chips, and h  > havevG  > little control over being able to drive the chip process.  So while   IBM and K Intel will relentlessly drive faster and cheaper processes for their chips,   G Again more uninformed FUD,  Texas's process is pretty much in step withw@ IBM's and if Sun wanted to we could go to IBM they would be only too happy to FAB SPARC's.t  H In the meantime in the real world IA64 (Mkinley) (YOUR CHOSEN CPU) gets  slower and later.a  5  > Alpha cost a fortune to keep on the bleeding edge.a  A The Alpha FAB cost a fortune and Digital never managed to fill itdD with non Alpha work. Sun does not own a FAB and Texas builds SPARC'sC and other processors in their FABs. Your argument is getting almosteB to dumb to bother responding to, even you must be able to rememberB that it was the FAB costs that crippled Alpha and Digital as well.      G  > Ooooh.  Come on.  You pretend to be technical, while being a blatentaD  > market-droid.  Why can't I take a few pokes at Sun - the king of @ snake > oil sales - now that what goes around is comming around.  = Funny you accuse me of being non technical and a market-droidhE but you do so in a series of postings full of supposition, stupidity,t? 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.          >  E  > You're running on vapors.  Your high-end ain't so high, and your >"
  > low-endI  > ain't so low.  You've lost your .COM bubble, and are selling into yourbI  > installed base.  Your margins are thin, and you're losing money.  You e  > haveuI  > no long-term story in HW.  Your "open systems" rethoric has come back l  > totH  > bite you in the arse in the form of Linux.  You are poo-pooing it, >  like DEC  > poo-poo'd the PC, and UNIX.  E I guess that just proves my point. Sun's high end is larger by almostt- any measure you care to use than Compaq/HP's.s  ? At the low end most of the posters to this group would call you = a liar, they would gag for a sub 1000 dollar Tru64 or OpenVMSi box but they don't exist.   > Sun lost money in the last two quarters, nothing compared withA Compaqs loss making record which you managed to sustain while thei= rest of the industry was booming, as I suggested before don'tc& throw stones you are in a glass house.    F  >I see no indication that Compaq intends to address the "right price"C  >issues that have crippled OpenVMS and the transition to IA64 withoD  >all its attendant issues is only likely to reduce the portfolio ofI  >software available for OpenVMS. Given this you should be expending moreOI  >efforts trying to save OpenVMS and less time questioning Sun's strategy?C  >which is one that many of your customers admire and wish that you !  >had had the courage to emulate.o  >  I  > NONE of my note had ANYTHING to do with Compaq, or with VMS.  So I'll e1  > ignore your attempt to deflect the discussion.s  = Odd this is a discussion group about OpenVMS not Sun or SPARCr= you seem to have proven my point without me having to respondt3 thanks Freddy you have just decended another notch.r   regardsd Andrew Harrisono   ------------------------------  % Date: Wed, 20 Mar 2002 12:23:18 -0500 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>IO Subject: Re: HP's viewpoint on Linux, was: Re: Sun eating major helping ofLinuxf3 Message-ID: <kG3m8.1261$fL6.25801@news.cpqcorp.net>h  2 Andrew Harrison SUNUK Consultancy wrote in message <3C98B3EC.8040101@sun.com>...d) >Andrew Harrison SUNUK Consultancy wrote:a  > >Funny you accuse me of being non technical and a market-droidF >but you do so in a series of postings full of supposition, stupidity,@ >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.s >e  ) You mean you aren't a sales/market droid?    ------------------------------  % Date: Wed, 20 Mar 2002 11:14:41 +0100.2 From: martin@radiogaga.harz.de (Martin Vorlaender)# Subject: Re: LD062 Install Questions; Message-ID: <3c986111.524144494f47414741@radiogaga.harz.de>o  0 David J. Dachtera (djesys.nospam@fsi.net) wrote:F > Wonder if there are any new features in LD V6.3, or is it just a VMS > version support upgrade?   New in V6.3:   $ LD TRACE/FDT (Alpha only)_B $ LD TRACE/ACCURATE (Alpha only) - accurate timing of I/O requests- Container files my now reside on NFS volumes.  $ LD CREATE/CONTINUOUS; Restriction on the maximum size of the logical disk withouts   specifying geometry lifted.C< Deleted container files which are still connected can now be   disconnected without /ABORT.   cu,e   Martin   P.S.:w? > MAN! *THAT* is a *GREAT* sig! I gotta show that one around....   ;-)  How 'bout this one:S -- yH    Emacs would be a great   | Martin Vorlaender  |  VMS & WNT programmer5    operating system,        | work: mv@pdv-systeme.deyL    if only it came with     |       http://www.pdv-systeme.de/users/martinv/<    a decent editor...       | home: martin@radiogaga.harz.de   ------------------------------  % Date: Wed, 20 Mar 2002 10:28:03 -0500 0 From: "Syltrem" <syltremspammenot@videotron.com>= Subject: MONI SYS shows 80% CPU usage and 8 jobs in COM state.4 Message-ID: <qS1m8.4206$a04.22365@tor-nn1.netcom.ca>  L I'm sure there is a good reason, but how do you explain I have (at times) upF to 8 processes in COM state and the CPU usage shows 158% with 2 CPUs ?   Thanks for your input!   --   SyltremlI http://pages.infinit.net/syltrem (OpenVMS related web site - en franais)h> To reply to myself directly, remove .spammenot from my address   ------------------------------    Date: 20 Mar 2002 09:39:24 -0600+ From: young_r@encompasserve.org (Rob Young)EA Subject: Re: MONI SYS shows 80% CPU usage and 8 jobs in COM statea3 Message-ID: <D6DeAXabpgqd@eisner.encompasserve.org>b  g In article <qS1m8.4206$a04.22365@tor-nn1.netcom.ca>, "Syltrem" <syltremspammenot@videotron.com> writes:wN > I'm sure there is a good reason, but how do you explain I have (at times) upH > to 8 processes in COM state and the CPU usage shows 158% with 2 CPUs ? >   A 	Granularity.  Somewhere it says don't pay attention to COM queue ? 	as that is a snapshot in time... more indicative is CPU usage.   @ 	If you want to make your system more responsive at a user levelG 	lower QUANTUM.  When set to default, 20 ms - you can get CPU intensivetB 	processes hogging up the CPUs.  By shuttling them in and out moreD 	often, (by lowering QUANTUM) you get folks that need a tiny bit of C 	CPU (application startup on fast alphas) off the COM queue quickeri* 	hence the system feels faster.  Snappy!!!  @ 	There have been numerous QUANTUM disucssions in c.o.v. over the 	years:s 			www.deja.comf   				Robe   ------------------------------  + Date: Wed, 20 Mar 2002 16:39:43 +0100 (MET)n9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> A Subject: Re: MONI SYS shows 80% CPU usage and 8 jobs in COM statee; Message-ID: <01KFL9FYQRM48ZQP5J@sysdev.deutsche-boerse.com>n  D > I'm sure there is a good reason, but how do you explain I have (atI > times) up to 8 processes in COM state and the CPU usage shows 158% with  > 2 CPUs ? r  F I'm not sure what the problem is.  The state is a snapshot, while the H CPU usage (maximum 200% for your system) is an average.  From HELP SHOW  SYSTEM:o  I               Note that the SHOW SYSTEM command examines the processes on I               the system without stopping activity on the system. In thisiJ               example process information changed during the time that theH               SHOW SYSTEM command collected the data to be displayed. AsI               a result, this display includes two processes, named GAWAINoI               and S. Whiplash, with the state CUR on the same CPU, CPU IDl               6 in the example.e  D Similarly, MONITOR might indicate more than one process in the same ! state, when actually they aren't.    ------------------------------  + Date: Wed, 20 Mar 2002 07:43:50 -0800 (PST) . From: Fabio Cardoso <fabiopenvms@yahoo.com.br>A Subject: Re: MONI SYS shows 80% CPU usage and 8 jobs in COM stateN@ Message-ID: <20020320154350.20257.qmail@web20210.mail.yahoo.com>   What kind of process ?  5 Batch process ? Network process ? Interactive processs ?     5 About 158%, I believe is is because your process donth3 demaand more than this. And of course the other 42%i1 are needed - at least a small part - to manage tob  OpenVMS Operating System itself.   Regardsf   FC i3 --- Syltrem <syltremspammenot@videotron.com> wrote:o1 > I'm sure there is a good reason, but how do youg > explain I have (at times) up5 > to 8 processes in COM state and the CPU usage showsT > 158% with 2 CPUs ? >  > Thanks for your input! >  > -- > 	 > SyltremS3 > http://pages.infinit.net/syltrem (OpenVMS related  > web site - en franais)f5 > To reply to myself directly, remove .spammenot fromt > my address >  >  >      =====a ========================== Fbio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazilh fabiopenvms@yahoo.com.br ==========================  2 __________________________________________________ Do You Yahoo!?+ Yahoo! Sports - live college hoops coverage' http://sports.yahoo.com/   ------------------------------    Date: 20 Mar 2002 10:13:40 -0600 From: briggs@encompasserve.orgA Subject: Re: MONI SYS shows 80% CPU usage and 8 jobs in COM state:3 Message-ID: <sOshxe+iVMfh@eisner.encompasserve.org>G  g In article <qS1m8.4206$a04.22365@tor-nn1.netcom.ca>, "Syltrem" <syltremspammenot@videotron.com> writes:0N > I'm sure there is a good reason, but how do you explain I have (at times) upH > to 8 processes in COM state and the CPU usage shows 158% with 2 CPUs ?  F And presumably sometimes you will see no processes in the COM state at all.  G One gotcha with MONITOR is that your snapshot of the system will alwayssB have at least one process in the CUR state.  The process doing theB monitoring.  So your snapshots tend to have a +1 bias on the total2 reported number of computable processes (CUR+COM).  B If we were to take a simple queueing theory model (exponential jobB completion distribution and Poisson arrival rate) you would expectB the average queue length to be roughly approximated by 1/(1-load).E The fact that you have two CPU's changes this slightly, but it's beenrB way too long since I took queueing theory to remember exactly how.  < Your load is about 80%.  So your average queue length should1 be around 5.  And that fits with what you report.:  A So we need not invoke anything special to explain your situation.E   	John Briggs   ------------------------------  % Date: Wed, 20 Mar 2002 11:42:04 -0500>0 From: "Syltrem" <syltremspammenot@videotron.com>A Subject: Re: MONI SYS shows 80% CPU usage and 8 jobs in COM statet4 Message-ID: <QX2m8.4225$a04.22520@tor-nn1.netcom.ca>  = >Your load is about 80%.  So your average queue length should 2 >be around 5.  And that fits with what you report.  I So you find it normal that I have a queue? My understanding was that if I H have a queue, then I have people waiting to be served. If I don't have aK queue, I only have jobs waiting for IO etc and one CURrently executing. And I that's good as it means my CPU is fast enough to keep up with the demand.k  J Reducing QUANTUM would most probably make the queue shorter but would also? reduce actual throuput from the machine (although it may not bedL perceptible). Since interactive jobs have good response time, and only batchJ jobs take longer to execute I don't think it's worth changing QUANTUM now.8 Also that problem only occurs at periods during the day.   --   SyltremeI http://pages.infinit.net/syltrem (OpenVMS related web site - en franais)p> To reply to myself directly, remove .spammenot from my address  ; <briggs@encompasserve.org> a crit dans le message de news: ( sOshxe+iVMfh@eisner.encompasserve.org...@ > In article <qS1m8.4206$a04.22365@tor-nn1.netcom.ca>, "Syltrem"( <syltremspammenot@videotron.com> writes:F > > I'm sure there is a good reason, but how do you explain I have (at	 times) up J > > to 8 processes in COM state and the CPU usage shows 158% with 2 CPUs ? >aH > And presumably sometimes you will see no processes in the COM state at > all. >eI > One gotcha with MONITOR is that your snapshot of the system will alwayssD > have at least one process in the CUR state.  The process doing theD > monitoring.  So your snapshots tend to have a +1 bias on the total4 > reported number of computable processes (CUR+COM). >oD > If we were to take a simple queueing theory model (exponential jobD > completion distribution and Poisson arrival rate) you would expectD > the average queue length to be roughly approximated by 1/(1-load).G > The fact that you have two CPU's changes this slightly, but it's beenuD > way too long since I took queueing theory to remember exactly how. >n> > Your load is about 80%.  So your average queue length should3 > be around 5.  And that fits with what you report.  >hC > So we need not invoke anything special to explain your situation.. > 
 > John Briggse   ------------------------------    Date: 20 Mar 2002 11:45:50 -0600 From: briggs@encompasserve.orgA Subject: Re: MONI SYS shows 80% CPU usage and 8 jobs in COM statey3 Message-ID: <6yFXWOPfXGgS@eisner.encompasserve.org>n  g In article <QX2m8.4225$a04.22520@tor-nn1.netcom.ca>, "Syltrem" <syltremspammenot@videotron.com> writes: > >>Your load is about 80%.  So your average queue length should3 >>be around 5.  And that fits with what you report.  > K > So you find it normal that I have a queue? My understanding was that if IdJ > have a queue, then I have people waiting to be served. If I don't have aM > queue, I only have jobs waiting for IO etc and one CURrently executing. AndrK > that's good as it means my CPU is fast enough to keep up with the demand.o   Yes, I find it normal.  F Consider the difference between average load and peak load.  You don'tA have enough CPU to handle peak load (measured at a 20 millisecond,F granularity).  You do have enough CPU to handle average load (measured at 10 second granularity).  D Note that you have two CPUs.  You should have two CURrent processes.  L > Reducing QUANTUM would most probably make the queue shorter but would alsoA > reduce actual throuput from the machine (although it may not benN > perceptible). Since interactive jobs have good response time, and only batchL > jobs take longer to execute I don't think it's worth changing QUANTUM now.: > Also that problem only occurs at periods during the day.  E For a job mix with exponential service times, reducing QUANTUM has non effect on queue length.b  F Since, like 99% of today's generation, you have no clue about queueing1 theory, let me give you a tiny bit of background.   B Take the simplest model.  You've got jobs coming in.  The jobs areC characterized by the rate at which they arrive and by the length of F time they will take to complete.  Both are random with some underlyingC distribution.  You've got one or more servers that can do the work. C And you've got a queue to hold the jobs that haven't been done yet.s  E Maybe we're talking about dirty dishes coming in via busboy and beingt cleaned at the sink.  H Maybe we're talking about customers coming out of the aisles and queuing up at the cash register.  G Maybe we're talking about VMS processes becoming computable and needingfD to do a little processing before waiting for the next I/O operation.     Exponential service time:   G    When a job is worked, it will take some time before it is done.  How B much is variable.  We model it as being random with an exponentialB distribution.  Which means that there is some probability that theD job will take one second.  And some identical probability that if it6 doesn't take 1 second that it will take 2.  And so on.   p(t)	.1	.081	.729	etc. t	1	2	3R  G That's an exponential distribution.  And it's nice to work with becauseh= it has the property that it doesn't matter how long a job hasyE been worked.  The distribution of "time left to work" stays the same.e  B That simplifies matters because the state of the queuing system is4 purely a function of how many jobs are in the queue.  B You can work problems in queuing theory using different models for= service time distribution.  But the computations are messier.t   Poisson arrival rate:l  D     If you have jobs coming in with a fixed incremental rate, you'llC find that the number of jobs coming in over some interval will haveeC a Poisson distribution.  There is actually an intimite relationship > between an exponential distribution and a Poisson one.  If you? look at the rate at which jobs with an exponential service time(? are completed, you'll find that the number of jobs completed incG a fixed time interval has a Poisson distribution.  So this distributionu5 is a very natural one to assume for job arrival rate.a  E If you look at job arrival rate over a very short sample time, you'llo= find that you have a distribution with a lot of variance (thebC standard deviation will be a large fraction of the mean).  You willh> have peaks that exceed your service capacity.  There _will_ be jobs in the queue.  D If you look at job arrival rate over a very long sample time, you'llE find that you have a distribution with comparitively little variance. A The standard deviation will be only a small fraction of the mean.b you'll be in serious trouble).   	John Briggs   ------------------------------  + Date: Wed, 20 Mar 2002 04:27:45 -0800 (PST)d. From: Fabio Cardoso <fabiopenvms@yahoo.com.br>) Subject: OpenVMS 7.2-1 and UCX 4.2 ECO 05 @ Message-ID: <20020320122745.38241.qmail@web20204.mail.yahoo.com>  # Anyone is using OpenVMS 7.2-1 with a UCX 4.2 ECO 05 ?  2 I am planning to upgrade one Alphaserver this next/ weeked. The actual versions are OpenVMS 7.1-1h2o and UCX 4.2 ECO 1.  % Reading the README of the UCX ECO at u% ftp.support.compaq.com, it is saying p' that UCX ECO 05 is compatible just witht OpenVMS 7.1 (max version).  ( Is that right ? Is the V7.2-1 compatible only with TCPIP Services 5.x ?   RegardsI       =====  ========================== Fbio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazil/ fabiopenvms@yahoo.com.br ==========================  2 __________________________________________________ Do You Yahoo!?+ Yahoo! Sports - live college hoops coverage  http://sports.yahoo.com/   ------------------------------  % Date: Wed, 20 Mar 2002 08:07:54 -0500m5 From: David Beatty <David.Beatty@qwertysasasdfgh.com>b- Subject: Re: OpenVMS 7.2-1 and UCX 4.2 ECO 05n2 Message-ID: <3oiYPH6vs+TDDvWANQeTMi6GlPoK@4ax.com>  1 Any reason why you would not upgrade to V7.2-2 or  V7.3 of VMS?  4 Any reason why you would not upgrade to V5.0 or V5.11 of Compaq TCP/IP Services?  I don't think V4.2 is " supported higher than V7.1 of VMS.   David R. Beattye  7 On Wed, 20 Mar 2002 04:27:45 -0800 (PST), Fabio Cardosoa! <fabiopenvms@yahoo.com.br> wrote:g  $ >Anyone is using OpenVMS 7.2-1 with  >UCX 4.2 ECO 05 ?s >m3 >I am planning to upgrade one Alphaserver this nexta0 >weeked. The actual versions are OpenVMS 7.1-1h2 >and UCX 4.2 ECO 1.V > & >Reading the README of the UCX ECO at & >ftp.support.compaq.com, it is saying ( >that UCX ECO 05 is compatible just with >OpenVMS 7.1 (max version).e >y) >Is that right ? Is the V7.2-1 compatibles >only with TCPIP Services 5.x ?o >r >Regards >e >d >  >===== >==========================- >Fbio dos Santos Cardoso2 >OpenVMS System Manager  >Rio de Janeiro - Brazil >fabiopenvms@yahoo.com.brm >==========================n >n3 >__________________________________________________e >Do You Yahoo!?1, >Yahoo! Sports - live college hoops coverage >http://sports.yahoo.com/t   ------------------------------  + Date: Wed, 20 Mar 2002 07:48:06 -0800 (PST)w. From: Fabio Cardoso <fabiopenvms@yahoo.com.br>- Subject: Re: OpenVMS 7.2-1 and UCX 4.2 ECO 05e@ Message-ID: <20020320154806.39227.qmail@web20203.mail.yahoo.com>   Davidc  4 This machine is not clustered, and the users give me5 just a few hours per weekend to stop it. It is a 24x7s5 machine, and I dont have conditions now to upgrade to 1 VMS 7.3 because we didnt test our systems in this 1 version. It is better to upgrade "slowly", I willr/ go to V7.2-1 because all the other machines in a2 the company (other sites) are in this max version.4 I dont want to cause high impact in this upgrade, it is4 because I prefer to just patch the UCX, but if there- is no condition I will upgrade to TCPIP 5.0A.u  4 I have 1500 printer and server  queues running under UCX.   RegardsE   FC T      3 --- David Beatty <David.Beatty@qwertysasasdfgh.com>  wrote: > 3 > Any reason why you would not upgrade to V7.2-2 ort > V7.3 of VMS? > 6 > Any reason why you would not upgrade to V5.0 or V5.13 > of Compaq TCP/IP Services?  I don't think V4.2 is $ > supported higher than V7.1 of VMS. >  > David R. Beatty  > 1 > On Wed, 20 Mar 2002 04:27:45 -0800 (PST), Fabioo	 > Cardosoa# > <fabiopenvms@yahoo.com.br> wrote:h > & > >Anyone is using OpenVMS 7.2-1 with  > >UCX 4.2 ECO 05 ?| > > 5 > >I am planning to upgrade one Alphaserver this nextv2 > >weeked. The actual versions are OpenVMS 7.1-1h2 > >and UCX 4.2 ECO 1.e > >s( > >Reading the README of the UCX ECO at ( > >ftp.support.compaq.com, it is saying * > >that UCX ECO 05 is compatible just with > >OpenVMS 7.1 (max version).S > >m+ > >Is that right ? Is the V7.2-1 compatibleM! > >only with TCPIP Services 5.x ?o > > 
 > >Regards > >g > >  > >8 > >===== > >==========================  > >Fbio dos Santos Cardosob > >OpenVMS System Managere > >Rio de Janeiro - Brazil > >fabiopenvms@yahoo.com.brs > >==========================  > >s5 > >__________________________________________________t > >Do You Yahoo!?M. > >Yahoo! Sports - live college hoops coverage > >http://sports.yahoo.com/m >      =====d ========================== Fbio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazil. fabiopenvms@yahoo.com.br ==========================  2 __________________________________________________ Do You Yahoo!?+ Yahoo! Sports - live college hoops coverage2 http://sports.yahoo.com/   ------------------------------    Date: 20 Mar 2002 11:47:46 -0600+ From: kuhrt@encompasserve.org (Marty Kuhrt)e Subject: Re: Plan B for Compaq3 Message-ID: <kL4QOdh3dmk2@eisner.encompasserve.org>s  t In article <Blxl8.71764$uA5.70404@rwcrnsc51.ops.asp.att.net>, "Terry C. Shannon" <terryshannon@mediaone.net> writes: > @ > "Steve Reece" <SYSTEM@ipl.demon.co.nospam.uk> wrote in message2 > news:3C969522.7118AE3B@ipl.demon.co.nospam.uk...I >> Who says that this isn't how the takeover^H^H^H^Hmerger got thought up  >> in the first place???  :-oo >>I >> Anyway, to answer JF's initial question.  Curly has said that he has a(G >> Plan B, he just isn't telling anyone what it is (presumably since heeG >> might be able to get consultancy fees from Compaq to disclose them).u >> > K > Yeah, as far as plans go,  in the killer worm-vs-humankind cinematic epic-G > "Tremors," actor Kevin Bacon had a whole bunch of plans to handle thelK > earthworms on steroids. And I seem to recall that Dick Nixon had a secret N > plan to end the war in Viet Nam. That must have been a big secret, because IN > held a top secret/crypto/si clearance and didn't hear a thing about the plan0 > even though I was in Viet Nam at the time. ;-} > N > And didn't Carly say something about burning boats or bridges or whatnot way > back last September?  C I didn't think that Fiorina was a Norse name, but I'm guessing thataC the burning boats she referred to were Viking Funeral Pyre boats.  SA If this merger doesn't work she already has a burning boat to putt. her career into when they push her out to sea.   ------------------------------  % Date: Wed, 20 Mar 2002 09:07:07 -0800 ' From: David Mathog <mathog@caltech.edu>e Subject: Right down the middle+ Message-ID: <3C98C1BB.9B52108B@caltech.edu>   M It's always odd when a vote turns out very near 50/50.  The usual reaons thatn happens are:  I 1.  There were two great candidates and the voters can't decide which onet6 they want more.  (Exceedingly rare, in my experience.)  M 2.  The choices are equally bad so the voters can't figure out which one they3
 dislike more.   P 3.  It's impossible to make an accurate assessment of the relative merits of the two choices. (So everybody flips a coin.)   4.  Pure coincidence.m  P In the current HP/Compaq collision there seems to be elements of scenarios 2, 3, and 4 at work.J Interesting how in the final analysis people voting their own money (small
 shareholders,tN retirement funds, the two families) came out largely against and people voting othertK peoples money (the big investment funds) largely came out for.  And the twom balanced each othero out almost exactly.e  M Unintentionally Carly ran a referendum on herself.  The one clear result fromp	 this vote M was that the HP employees hated it (2:1 against it in the employee retirement  plan and theO polls by Hewlett). I've not seen the numbers for Compaq, but it's probably safei to sayK that with the axe swinging in their direction Compaq employees hate it evene more. All ofL this bodes very ill for the post merger work environment.  As if working for
 Carly, Curly,iE and Winkler wasn't already a big enough triple curse for any company!=   Regards,   David Mathog mathog@caltech.edu   ------------------------------  % Date: Wed, 20 Mar 2002 12:28:35 -0500t5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>-" Subject: Re: Right down the middle3 Message-ID: <hL3m8.1263$fL6.25826@news.cpqcorp.net>Y  @ David Mathog wrote in message <3C98C1BB.9B52108B@caltech.edu>...   >@I >Unintentionally Carly ran a referendum on herself.  The one clear resulto from
 >this voteC >was that the HP employees hated it (2:1 against it in the employeer
 retirement
 >plan and theiK >polls by Hewlett). I've not seen the numbers for Compaq, but it's probably  safe >to saysL >that with the axe swinging in their direction Compaq employees hate it even
 >more. All ofcI >this bodes very ill for the post merger work environment.  As if workinge forn >Carly, Curly,F >and Winkler wasn't already a big enough triple curse for any company! >h  J Well, I guess I'll buck the trend here.  I am all in favor of it, and look forward to it.  J You are right that it is polarizing however, people either love it or hate it, with little in-between.o   ------------------------------  % Date: Wed, 20 Mar 2002 09:46:12 -0800-# From: "Tom Linden" <tom@kednos.com>U" Subject: RE: Right down the middle9 Message-ID: <CIEJLCMNHNNDLLOOGNJIIEDAEGAA.tom@kednos.com>   C There has been a lot of analysis as to the merits of the merger andhE the like, but I believe it boils down to Carly want to maintain a fatnI paycheck.  I think that had she not undertaken this she would likely have-K lost her job.  So this is a smokescreen to cover up poor performance at HP. @ Plus as others have pointed out the CEOs would profit handsomely: from the merger, even if the combined companies did poorlyD So, if this were true, then she had nothing to lose.  Clever gamble.   > -----Original Message-----0 > From: David Mathog [mailto:mathog@caltech.edu]) > Sent: Wednesday, March 20, 2002 9:07 AM  > To: Info-VAX@Mvb.Saic.Comh  > Subject: Right down the middle >e >eC > It's always odd when a vote turns out very near 50/50.  The usualt
 > reaons thata > happens are: >eK > 1.  There were two great candidates and the voters can't decide which oneh8 > they want more.  (Exceedingly rare, in my experience.) >u@ > 2.  The choices are equally bad so the voters can't figure out > which one they > dislike more.i >); > 3.  It's impossible to make an accurate assessment of thet > relative merits of the > two choices. > (So everybody flips a coin.) >. > 4.  Pure coincidence.o >rB > In the current HP/Compaq collision there seems to be elements of > scenarios 2, 3,M > and 4 at work.L > Interesting how in the final analysis people voting their own money (small > shareholders,oB > retirement funds, the two families) came out largely against and > people votingr > othern@ > peoples money (the big investment funds) largely came out for.
 > And the twoi > balanced each other  > out almost exactly.o >rC > Unintentionally Carly ran a referendum on herself.  The one clears
 > result from  > this votes; > was that the HP employees hated it (2:1 against it in they > employee retirements > plan and theC > polls by Hewlett). I've not seen the numbers for Compaq, but it'st > probably safew > to say@ > that with the axe swinging in their direction Compaq employees > hate it even > more. All ofB > this bodes very ill for the post merger work environment.  As if
 > working forR > Carly, Curly,wG > and Winkler wasn't already a big enough triple curse for any company!i >d
 > Regards, >  > David Mathog > mathog@caltech.edu >    ------------------------------  # Date: Wed, 20 Mar 2002 18:19:24 GMT * From: "Bill Todd" <billtodd@metrocast.net>" Subject: Re: Right down the middle? Message-ID: <Mo4m8.3941$2q2.308615@bin4.nnrp.aus1.giganews.com>o  @ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message- news:hL3m8.1263$fL6.25826@news.cpqcorp.net...h   ...w   >dL > Well, I guess I'll buck the trend here.  I am all in favor of it, and look > forward to it.  ' Company man to the bitter end, eh Fred?0  L Sometimes I wonder whether your vigorous defense of the indefensible isn't aL reaction against a little voice inside you that is just as pissed off as theE rest of us are at a company that has left you in a dead-end job, on ayK dead-end product, in a corporate environment dedicated to mediocrity - wheneH things could have been so very different.  But it would help explain whyG your main ambition these days seems to be to retire to Florida and playa golf.i  K Or not:  you don't leave anywhere nearly the wealth of publicly-collectibleaK information in your wake that Compaq does, so devining your motives is puren* speculation rather than informed analysis.   - bill   ------------------------------  % Date: Wed, 20 Mar 2002 11:26:02 +0100e% From: "Fred Zwarts" <F.Zwarts@KVI.nl> , Subject: Suggested enhancement. for DECterm.. Message-ID: <a79o3r$r95$1@info.service.rug.nl>  9 For easier tracing the location of the users of a system,1: it would be advantageous to define the remote port info=207 of the FTA device of a DECterm to the properties of thew1 related DISPLAY, e.g.: TCP/host.domain:1.0, or=20  DECNET/host:0.0.   What do you think of it?                                F.Z.   ------------------------------  # Date: Wed, 20 Mar 2002 12:21:59 GMT  From: system@SendSpamHere.ORGo0 Subject: Re: Suggested enhancement. for DECterm.0 Message-ID: <00A0B368.62F7E255@SendSpamHere.ORG>  V In article <a79o3r$r95$1@info.service.rug.nl>, "Fred Zwarts" <F.Zwarts@KVI.nl> writes:: >For easier tracing the location of the users of a system,; >it would be advantageous to define the remote port info=20t8 >of the FTA device of a DECterm to the properties of the2 >related DISPLAY, e.g.: TCP/host.domain:1.0, or=20 >DECNET/host:0.0.r >e >What do you think of it?e >e! >                            F.Z.r >o  I This was discussed here several years ago.  If you want this now, here isn some code to do it for you:t  " ftp://ftp.tmesis.com/accpornam.zip    H YOU SHOULD NOT CHANGE THE MARKER (xtnd) IN ACCPORNAM_INIT NOR SHOULD YOUI CHANGE ANY OF THE ACCPORNAM_INIT.MAR CODE!  Presently, there are two com-oK mercial products that have employed this code to define an access port nameaK field for the FTA.  If you change things, you could cause yourself grief ifo; you have either of these products installed on your system.e   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMn            hJ   "And of course, I'm a genius, so people are naturally drawn to my fiery I   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbese   ------------------------------  # Date: Wed, 20 Mar 2002 13:07:06 GMTh( From: "Mark E. Levy" <mlevy70@attbi.com>$ Subject: Re: Talk about downtime....- Message-ID: <_P%l8.75148$af7.49868@rwcrnsc53>k  J Weren't the movers insured for any damage they might have caused? It wouldI seem that if the machine was working before the move and there was damage 5 that required repair afterward, they would be liable.g  	 Mark Levyu SMAa  > "Keith Parris" <keithparris_NOSPAM@yahoo.com> wrote in message7 news:cf15391e.0203191741.6e801a21@posting.google.com... A > Tim Llewellyn <tim.llewellyn@blueyonder.co.uk> wrote in messager, news:<3C962F97.DD3979EC@blueyonder.co.uk>... > > Robert Deininger wrote:e > > >  DECmove help in all7 > > > these areas, but I never heard that it was cheap.s > >i> > > yup, in my limited experience it definitely was NOT cheap. >n@ > I know of one move where DECmove service was suggested but wasC > rejected as too costly.  As best we could determine afterward, ityF > appears the movers must have dropped one of the 8400 system boxes atE > some point.  A backplane ended up twisted and had to be replaced tonE > get the box working at the new site, at a cost of more for that onev, > single part than the entire DECmove quote.0 > ----------------------------------------------0 > Keith Parris | parris at encompasserve dot org   ------------------------------    Date: 20 Mar 2002 09:45:20 -0800, From: cmussche@volvocars.com (Carlo Mussche) Subject: telnetqueue< Message-ID: <ab891bd3.0203200945.b8394ac@posting.google.com>  6 Suppose I have defined on my vms machine a telnetqueue  A INIT /QUEUE/PROCESSOR=TCPIP$TELNETSYM/ON=xxx::"<ip-address>:9100"l /start xxx_QUEUE  @ How can I print correctly to that queue from for example a nt-pc   ------------------------------  # Date: Wed, 20 Mar 2002 09:07:33 GMTh* From: "Bill Todd" <billtodd@metrocast.net>' Subject: Re: Too close to call says BBC8C Message-ID: <pjYl8.167408$uv5.14128474@bin6.nnrp.aus1.giganews.com>i  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3C98314A.F8FA5E3C@videotron.ca...   ...   D > I don't really know what drove Carly to want to buy a sick Compaq,  	 How aboutu  G 1.  A sick HP (that to a significant degree she had made that way), and   K 2.  The promise that she'd have at least an additional year or so to make atI mess of things due to merger confusion (and an outside chance of actually=J pulling it off), rather than get kicked out sooner for her existing record at HP.  	  but I dooI > have to admire her guts to really put her job on the line for something_ she_ > believes in.  J What Carly believes in is anybody's guess.  What is pretty much certain isJ that this merger will make her a hell of a lot richer than she already is.  H Her job already was on the line:  it didn't take so much guts to convertJ that into an opportunity as just a great deal of cunning (which she has inD abundance).  Like many despicable figures in history, Carly has some7 remarkable strengths - but I don't admire her for them.    - bill   ------------------------------    Date: 20 Mar 2002 05:45:38 -0800( From: bob@instantwhip.com (Bob Ceculski)' Subject: Re: Too close to call says BBCs= Message-ID: <d7791aa1.0203200545.7611686f@posting.google.com>V  a JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3C98314A.F8FA5E3C@videotron.ca>.... > Bill Todd wrote:O > > Typical Carlyese.  The HP shareholders indeed made a choice today (whateverOJ > > it turns out to have been), but to all appearances by the narrowest ofJ > > margins rather than any kind of 'embrace' (let alone outright charge). > K > Not sure if it was Carly or someone else who said it, but I heard that HP 6 > really need only 50% + 1 vote to declare this a win. > 8 > If Carly gets 50%+1 , can she really declare victory ? > M > She may be able to legally go ahead with the takeover, but she will have to_L > take into consideration that half of the votes are against her, but if oneK > shareholder does switch against carly, then a vote of no-confidence woulde > succeed in throwing her out. > M > I don't really know what drove Carly to want to buy a sick Compaq, but I do1M > have to admire her guts to really put her job on the line for something sheoN > believes in. If this takeover doesn't work out as she predicted, she will be* > out faster than Pfeiffer was kicked out.  H what drove her to buy Q is that only their printer line makes money, theH server/OS end is losing money ... they hope Alpha/Itanium VMS unix linuxI and the 4 billion vms makes in revenues will change that so that they cann; spin off the printer end and have a viable services end ...e   ------------------------------  # Date: Wed, 20 Mar 2002 18:13:22 GMTU# From: "John Smith" <a@nonymous.com>i Subject: Too late to cryI Message-ID: <6j4m8.16806$oG11.15251@news01.bloor.is.net.cable.rogers.com>   L It's pretty much a foregone conclusion that Compaq shareholders will vote in favor of the merger.  G Why? Because they see the exchange of Compaq shares into HP shares as auK plus, mostly from a monetary perspective as to how it affects their wallets4 today.    L As to Compaq enterprise systems (read OpenVMS/Tru64) owners/users who wantedI to ensure that they didn't have to spend megabucks porting or moving from>E their chosen o/s platform, the smartest thing they could have done to J protect their investment was to have purchased some HP shares and voted NO merger.a  J Hell, they would have even made money on the deal - buying HP low throughtJ the uncertainty leading up to the vote, and dumping HP after it bounced up after a NO merger result.s  > Anyway, it'll be a week or two before we know the real result.    	 /dream onmH Who knows, maybe a majority of HP directors will want to back out of the1 deal even if the HP vote is 50.1 - 53%% in favor.   K It might not be a bad idea for HP to just to pay Compaq the $650 million toz kill the deal.  
 /dream off   ------------------------------    Date: 20 Mar 2002 08:33:49 -0800 From: ohm62@hotmail.com (OHM)D$ Subject: Re: VMS DEBUG V7.2 bugs me!= Message-ID: <9d337b47.0203200833.44baed58@posting.google.com>   A > Try the debugger from VMS 7.3. It's much better and you can get  > it from Compaq for VMS 7.2.- >  >  > Mark Williamso  - Thank you for the suggestion.  I'll try that!_  
   -- Olivier._   ------------------------------   Date: 20 Mar 2002 04:27 CSTe' From: carl@gerg.tamu.edu (Carl Perkins)o3 Subject: Re: Witch streamer supported vax-4000-90 ?2- Message-ID: <20MAR200204274703@gerg.tamu.edu>r  5 "David J. Dachtera" <djesys.nospam@fsi.net> writes...-E }....which raises the question: is/was there a QIC format program for$I }whatever tapes the TZK10 uses (I have exactly one, and that has a backupa3 }of SCO UNIX on it)? Does that program run on VMS? o }--  }David J. Dachtera  5 Have you completely forgotten the INITIALIZE command?.   --- Carl   ------------------------------  # Date: Wed, 20 Mar 2002 08:25:12 GMTh1 From: "Boris Dynin" <borisSPAMNO@PLEASEmovil.com>'% Subject: Re: Your mother is so fat...t3 Message-ID: <IHXl8.5782$44.45645@typhoon.sonic.net>h  - I disabled JScript in IE. It won't affect me.t   Boriss0 "Greg Lamb" <greg@linuxbin.com> wrote in message% news:3C96D2A8.4040508@linuxbin.com... E > Just in case anybody freaks out over this who hasn't heard about itnH > already, the following is all this idiot is doing, it's not a virus... > it's a Javascript loop >x >h > function tgos()n > {n >     while (true)E >         window.alert("** WARNING ** VirusScan has detected the TGOSpC > virus on your hard drive. If you have recently opened an email or A > newsgroup message and see this alert your system is infected.")t > }a >t   ------------------------------  % Date: Wed, 20 Mar 2002 10:44:15 -0800tM From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@spam-be-gone.intel.com.net> 6 Subject: Re: [Q]: Setting SCSI Cluster host SCIS ID's?: Message-ID: <3C98D87F.BEFEFC2B@spam-be-gone.intel.com.net>   sms@antinode.org wrote:e  O > From: "Kenneth H. Fairfield" <Kenneth.H.Fairfield@spam-be-gone.intel.com.net>0< > > So...does anyone know how to set/change the host SCSI ID: > > on an AlphaStation 200?  I didn't _notice_ any console > > variable.  [...] >y* >    I vote for "pka0_host_id".  Mine's 7.  ? Bingo!  That did it, Steven. :-)  The cluster is up and runningc3 (again) with the host controllers at SCSI ID 6 & 7.y  = Thanks to the others who posted as well.  Like I said before,i; I inherited this cluster and the User Manual isn't anywheret6 near thorough enough...and with no hard-copy device toB look at the console output, I (easily) missed the pkao* variables.  : FYI, these two Alpha Station 200s are using their internal5 SCSI adapters, NCR 810's they say at the console.  Atf9 least one system's CD is on the single, shared, SCSI bus.n5 I didn't open up the second system to see whether itsh. CD had been disconnected (it does have power).   -Ken --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfieldu! F20 Automation VMS System Support  kenneth.h.fairfield#intel.com    ------------------------------    Date: 20 Mar 2002 12:53:24 -0600+ From: kuhrt@encompasserve.org (Marty Kuhrt) 6 Subject: Re: [Q]: Setting SCSI Cluster host SCIS ID's?3 Message-ID: <ftPBBX2kw2Z$@eisner.encompasserve.org>I   In article <3C97E2C1.D4DD6B7E@spam-be-gone.intel.com.net>, "Kenneth H. Fairfield" <Kenneth.H.Fairfield@spam-be-gone.intel.com.net> writes:< > I have inherited a two-node SCSI cluster comprising a pair9 > of AlphaStation 200 4/166's (yep, they're a bit long ink > tooth :-). > : > The configuration seems a bit odd to me in that one host5 > is configured at SCSI ID 6 (OK) and the other is at-7 > SCSI ID 3 (very odd!).  Also, whoever configured thism: > cluster managed put/keep the CDrom at DKA400 (SCSI ID 4)7 > for both systems...I _believe_ that only one system's57 > CD is actually on the SCSI bus, but I could easily be3 > mistaken.0 > 7 > The two systems share an external StorageWorks shelf,n9 > but given the one host at SCSI ID 3 and the CD at ID 4,w8 > two of the external slots are not usable.  Furthermore9 > I worry that there may be some performance issues given'! > the unfair arbitration of SCSI.  > : > So...does anyone know how to set/change the host SCSI ID8 > on an AlphaStation 200?  I didn't _notice_ any console: > variable.  I also couldn't see anything _obvious_ on the3 > motherboard (yep, I did open 'er up and look :-).n  A The console variables that affect the scsi controllers start with < pk.  If you do a show pk* from the >>> prompt you should see something like...r   pka0_disconnect  1 pka0_fast        1 pka0_host_id     7  > If you type set pka0_host_id 6, followed by an init, it should+ reset the host id for that controller to 6.0   ------------------------------   End of INFO-VAX 2002.156 ************************le9 Message-ID: <CIEJLCMNHNNDLLOOGNJIIEDAEGAA.tom@kednos.com>   C There has been a lot of analysis as to the mer`    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `     `    `    `    `    `    `    `    `    `    	`    
`    `    `    
`    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `     `    !`    "`    #`    $`    %`    &`    '`    (`    )`    *`    +`    ,`    -`    .`    /`    0`    1`    2`    3`    4`    5`    6`    7`    8`    9`    :`    ;`    <`    =`    >`    ?`    @`    A`    B`    C`    D`    E`    F`    G`    H`    I`    J`    K`    L`    M`    N`    O`    P`    Q`    R`    S`    T`    U`    V`    W`    X`    Y`    Z`    [`    \`    ]`    ^`    _`    ``    a`    b`    c`    d`    e`    f`    g`    h`    i`    j`    k`    l`    m`    n`    o`    p`    q`    r`    s`    t`    u`    v`    w`    x`    y`    z`    {`    |`    }`    ~`    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    `    