1 INFO-VAX	Tue, 21 Oct 2003	Volume 2003 : Issue 583       Contents:, Re: Any VMS at ITU Telecom World in Geneva??! Re: Authentication via RSA Keyfob  Re: Clustering over GB Switches B Re: correspondence between access as defined in the UAF and F$MODEB Re: correspondence between access as defined in the UAF and F$MODE8 Re: Display Postscript (was: What is the performance...)P Re: HP announces new AlphaServer systems andenhancements to Tru64	=?iso-8859-1?QP Re: HP announces new AlphaServer systems andenhancements to Tru64	=?iso-8859-1?Q Re: HR/Payroll Package on VMS  Re: HR/Payroll Package on VMS  Kerry Main's e-mail  Re: LAN overload? $ Re: openVMS alpha 7.3.1 on PWS600au?$ Re: openVMS alpha 7.3.1 on PWS600au?- OpenVMS BootCamp BOF - Console Management ROI ! Re: OpenVMS in an HP advertisment , Re: OT: Why Scott McNeally is not worried..., RE: OT: Why Scott McNeally is not worried..., Re: OT: Why Scott McNeally is not worried..., Re: OT: Why Scott McNeally is not worried..., Re: OT: Why Scott McNeally is not worried..., RE: OT: Why Scott McNeally is not worried... Product (PCSI) Info Question  Re: Product (PCSI) Info Question  Re: Product (PCSI) Info Question  Re: Product (PCSI) Info Question" Re: question on patch nomenclature	 Re: rwast & Re: SMTP server behind an ADSL Switch?( Supporting RMS calls for "other" devices$ Re: SYSMAN> DEFINE/KEY/IF_STATE=GOLD7 VAX/VMS to Itanium Migration Letter available in French  Re: VMS v8.2 has a name  Re: VMS v8.2 has a name  Re: What is ACME_SERVER? Re: What is ACME_SERVER? Re: What is ACME_SERVER?% Re: What is the performance of iVMS ? % Re: What is the performance of iVMS ?  Re: [~OT] Last _production_ VAX  Re: [~OT] Last _production_ VAX   F ----------------------------------------------------------------------  % Date: Mon, 20 Oct 2003 20:56:47 +0200 B From: Michiel Erens <I.dont.want.spam@this.mailaddress.is.invalid>5 Subject: Re: Any VMS at ITU Telecom World in Geneva?? 7 Message-ID: <3F942FEF.573F@this.mailaddress.is.invalid>    Main, Kerry wrote: > G > I am not going to say that marketing could not be improved, but there F > are some (ok, not of the Wall Street Journal size)recent good thingsD > happening as well wrt OpenVMS promotions. Just look at the OpenVMSH > Technical Update sessions - which continue to be well received in many# > different areas around the globe.   B This reminds me (and I don't believe that it is already mentioned C here), that after the Dutch Technical Update days last month, there C was a half page article on OpenVMS in Computable magazine. Posting  A it here is not useful (it is in Dutch), but maybe Sue can give a  4 summary because she's mentioned five times in it :-)   --   ME Posted by news://news.nb.nu    ------------------------------  % Date: Mon, 20 Oct 2003 19:55:29 -0400 " From: "Hal Kuff" <kuff@tessco.com>* Subject: Re: Authentication via RSA Keyfob0 Message-ID: <k_OdnTra2cRp6AmiRVn-vw@comcast.com>  ' As always, a great answer... thanxs Rob   8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:ShXrr8EGtLos@eisner.encompasserve.org... = > In article <vp6dnbZJYvSUEg-iRVn-hQ@comcast.com>, "Hal Kuff"  <kuff@tessco.com> writes:  > > A > >     Is anyone doing Telnet authentication via RSA keyfob....?  >  > I know multinet, so... > . > I believe there are 3 card mehods supported. > ' > Regarding the most popular, SecureID. 6 > Multinet supports SecureID but not as a server,  you' > "point"  to an authentication server:  > 3 > http://www.opus1.com/mnetdoc/admin_guide/Ch12.htm  >  > H > If the MultiNet Secure/IP Client cannot contact the MultiNet Secure/IP Server,   > the following message appears: > = > Error opening connection to Secure/IP authentication server  >  > -- > > > Crypto Cards are pretty cheap, Stephen Arnold is a reseller: >  > CRYPTOCard > Arnold Consulting, Inc. ) > 2530 Targhee Street, Madison, Wisconsin  > 53711-5491 U.S.A. ( > Phone : 608-278-7700 Fax: 608-278-7701$ > Email: Stephen.L.Arnold@Arnold.Com@ > CRYPTOCard is a modern, SecureID-sized, SNK-compatible device. > > > Advantage there is Multinet is a CRYPTO server.  By the way,B > SecureID infrastructure isn't cheap, but you may have Enterprise > needs and budget.  > 9 > Likewise , there are SKEY cards out there but I haven't F > tracked them down.  Personally, I'm using SKEY generated keys storedD > in a spreadsheet on a PDA.  In fact, while I wasn't looking I justA > got booted on idle timeout so I have to login again, here is an 
 > example: >  >  > Username: myusernameB > Challenge: s/key 273 f8873evbi0u87zo        ! The SEED is munged& > Response: OR SWAY ASKS AVER MILE EAR > 4 >          Welcome to OpenVMS Alpha - Version V7.3-1 > ? >     Last interactive login on Monday, 20-OCT-2003 08:56:58.24 C >     Last non-interactive login on Monday, 13-OCT-2003 13:49:18.40  > , >             You have 15 new Mail messages. > C > Advantage here is Multinet is a SKEY server, and it is very cheap 5 > (my time), and one time pads are a beautiful thing.  >  > Rob  >    ------------------------------  + Date: Mon, 20 Oct 2003 20:26:27 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) ( Subject: Re: Clustering over GB Switches( Message-ID: <bn1gdj$ujs$1@pcls4.std.com>  2 dave.baxter@bannerhealth.com (Dave Baxter) writes:  w >keithparris_NOSPAM@yahoo.com (Keith Parris) wrote in message news:<cf15391e.0310171446.4b53aa6d@posting.google.com>...    >> > When I F >> > brought up the second node it attempted to create its own clusterG >> > (which caused me to do an immediate "HALT" on the system to try to G >> > forstall any partitioning issues).     This clearly indicated that E >> > NODE_2 could not communicate with the "up" node, NODE_1, via the  >> > switch. >>  F >> The other thing this seems to clearly indicate is that your settingE >> for EXPECTED_VOTES must be too low.  It should be the total of all F >> VOTES parameter values in the cluster, plus QDSKVOTES if you have a >> quorum disk.   F >I am not sure that I understand this reference to EXPECTED_VOTES.   IG >am not trying to configure a new cluster member, I am merely trying to A >reform a pre-existing cluster via a reconfigured interconnect.     J If EXPECTED_VOTES is set correctly, it is impossible to have tow instancesE of a cluster, but you (almost) did.  Thus it must be set incorrectly.   A >Surely, if EXPECTED_VOTES worked before, it should work after.   G >(NOTE: the "Point-to-Point" cluster has been rebooted many times, both G >one node at a time, and Cluster reboot, without any problems regarding  >EXPECTED_VOTES).   F VMS partially idiot-proofs clusters by automatically ratcheting upwardH the run-time value of EXPECTED_VOTES.  Assume the simplest case of a two> node cluster, each with one vote and no quorum disk, VOTES=1, J EXPECTED_VOTES=1. When they see each other, they realize there are a totalF of two nodes in the cluster and EXPECTED_VOTES "should" be 2, and theyB change the in-memory value to 2.  Therefore, if you yank the cableC between them, each node will hang because each doesn't have quorum, F rather than become 2 independent clusters because each 'thinks' it hasC quorum (VOTES=1, EXPECTED_VOTES=1).  However, you can still corrupt I your data this way, yank the cable between them and boot each node.  Each @ node will form its own one-node cluster rather than hang, since I EXPECTED_VOTES=1.  Each node doesn't know of the other.  Good bye data!    This is what you almost did.   --   -Mike    ------------------------------   Date: 20 Oct 2003 18:19 CDT . From: carl@nospam.gerg.tamu.edu (Carl Perkins)K Subject: Re: correspondence between access as defined in the UAF and F$MODE 4 Message-ID: <20OCT200318194598@nospam.gerg.tamu.edu>  T helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes...J }The UAF has different types of access: interactive, batch, local, dialup I }and remote.  F$MODE() can return INTERACTIVE, BATCH, NETWORK, AND OTHER.   . You missed the network access type in the UAF.  I }What is the correspondence between these two classes of types of access?  } J }In particular, is it possible to have an account which cannot log in via C }the network, but which can use SYSMAN on all nodes in a cluster?   F }(Ideally, it would have not TCPIP access, but LAT would be OK, maybe 7 }DECnet, but I doubt that it can be that fine-grained.)   / This is actually a mixing two different things.   D Look at the documentation for sys$getjpi and compare the values thatE can be returned for the "jpi$_jobtype" and "jpi$_mode" items with the  access data in the UAF.   C The possible "jobtype" values are: detached, network, batch, local,  dialup, and remote.   G The possible "mode" values are: other, network, batch, and interactive.   + F$MODE() is the same as F$GETJPI("","MODE")   L It is not clear exactly how the UAF types are mapped, but my guess would be: Interactive matches the mode. & Batch and network probably match both.0 Local, remote, and dialup all match the jobtype.  D So my guess is that it depends on the mode value except when that isD "other", in which case it checks the jobtype. Additionally, when theH type is "interactive" it also needs to determine if it is local, remote,L or dialup (which could be done by checking the jobtype). If both interactiveM and whichever of the other types it is (like "local" if the user is accessing J via a local terminal) - but I'm not sure how it then interprets the accessJ controls if one says it has access and the other says it doesn't; example:J it is 6pm and the interactive category access control says you have accessJ at 6pm, but the local access control says you don't have access as 6pm; doM you get in? I'm thinking no - it probably checks all relevant access controls J and only lets you in if none of them say "no". This is probably covered in$ the manuals, but I've never checked.  G Note that "detached" is a special case - there is no UAF access control F for it like there are for the other types, just a limit on how many anK account can have (including "none" and "infinite") with no hourly controls.    --- Carl   ------------------------------   Date: 20 Oct 2003 18:29 CDT . From: carl@nospam.gerg.tamu.edu (Carl Perkins)K Subject: Re: correspondence between access as defined in the UAF and F$MODE 4 Message-ID: <20OCT200318291610@nospam.gerg.tamu.edu>  y In article <bmsemt$8sa$3@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes... 4 }In article <bSJagfQFVdKT@eisner.encompasserve.org>,1 }Kilgallen@SpamCop.net (Larry Kilgallen) writes:   } L }> > Let me rephrase the question:  I have an account which has no network, H }> > dialup or remote access, but has local and batch access in the UAF. }> >  L }> > Even though it has the appropriate privs, it can't do any remote tasks H }> > via SYSMAN.  SYSMAN> DO WRITE SYS$OUTPUT F$MODE() results in OTHER. }>  E }> WRITE SYS$OUTPUT F$MODE() reflects the method of process creation.  }>  F }> If you get OTHER on an interactive login, something is quite wrong. }  }What I meant was: } % }SYSMAN> do write sys$output f$mode() 3 }%SYSMAN-I-OUTPUT, command execution on node ELIJAH  }OTHER3 }%SYSMAN-I-OUTPUT, command execution on node GLADIA  }OTHER3 }%SYSMAN-I-OUTPUT, command execution on node DANEEL  }OTHER }    Try   2 SYSMAN> do write sys$output f$getjpi("","JOBTYPE")  % It should say "0". "0" is "detached".   % From sys$library:starlet.req you find   > literal JPI$K_DETACHED = 0;             !  JPI$_JOBTYPE values> literal JPI$K_LOCAL = 3;                !  JPI$_JOBTYPE values> literal JPI$K_DIALUP = 4;               !  JPI$_JOBTYPE values> literal JPI$K_REMOTE = 5;               !  JPI$_JOBTYPE values  > (IIRC, 1 and 2 match the mode values, i.e. network and batch.)   --- Carl   ------------------------------    Date: 20 Oct 2003 15:13:07 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) A Subject: Re: Display Postscript (was: What is the performance...) 3 Message-ID: <jnrwy$CZEkaG@eisner.encompasserve.org>   n In article <h8Xkb.7439$x02.3388@news.cpqcorp.net>, "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> writes:  N > There are other ways to display postscript, and there are other formats thanD > postscript.  The only real people hurt here are those who actually$ > programmed to the DPS extension...  M And those who depend upon layered products that did so, such as DEC Document.    ------------------------------  % Date: Mon, 20 Oct 2003 13:30:02 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>Y Subject: Re: HP announces new AlphaServer systems andenhancements to Tru64	=?iso-8859-1?Q ) Message-ID: <3F941B99.A162564D@istop.com>    Sue Skonetski wrote:G > This just came out (5 minutes ago), which means please give it enough ) > time to reach the different HP offices.   	 Dear Sue,   K A comment: in that release, it does talk about EV7 and DS15, leading one to ) believe that DS15 machines are EV7 based.   M You may wish to clarify the text to make it very clear on which chip the DS15 B is based on and perhaps compare performance numbers with the DS10.   ------------------------------  % Date: Mon, 20 Oct 2003 20:34:52 +0200 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> Y Subject: Re: HP announces new AlphaServer systems andenhancements to Tru64	=?iso-8859-1?Q ' Message-ID: <3F942ACC.F1FB006E@aaa.com>    On this page :  8 http://h18002.www1.hp.com/alphaserver/ds15/overview.html  	 it says :   6 "The HP AlphaServer DS15 systems are available with a   1-GHz Alpha chip, the 21264C"    That's EV6-something, isn't it ?  	 Jan-Erik.    JF Mezei wrote:  > M > A comment: in that release, it does talk about EV7 and DS15, leading one to + > believe that DS15 machines are EV7 based.  > O > You may wish to clarify the text to make it very clear on which chip the DS15 D > is based on and perhaps compare performance numbers with the DS10.   ------------------------------    Date: 20 Oct 2003 17:34:45 -0700+ From: tonymcg27@yahoo.com.au (Tony McGrath) & Subject: Re: HR/Payroll Package on VMS= Message-ID: <521ebd9f.0310201634.5de984cf@posting.google.com>    Hi John,  * Check out http://www.frontiersoftware.com/  E They develop the "CHRIS" payroll package, which we currently use here F on one of our VMS systems. I think we are one of the last VMS sites to= still use it. VMS doesn't even get a mention in their list of 7 supported servers, which you'll find on their web site. ? Please don't ask how good it is, I have no idea, I don't use or C maintain it. I guess it can't be too bad, it has been on our system  since at least July 1998. 2 The package runs on the ACUCOBOL Cobol interpreter9 (http://www.acucorp.com/solutions/datasheets/acucobolgt/)    Hope that helps,   Tony ---  Tony McGrath OpenVMS Support Group  Toll Transport, IT Dept.$ Laverton North, VIC, Australia, 3006) tony_mcgrath (at) toll (dot) com (dot) au       a brandon@dalsemi.com (John Brandon) wrote in message news:<03102011260307@dscis6-0.dalsemi.com>... H > Anyone have a "supported" 3rd-party HR/Payroll package running on VMS? > ? > Any information concerning that product would be appreciated!  >  > TIA  >  >  > J*o*h*n B*r*a*n*d*o*n  > VMS Systems Administrator , > firstname.lastname.spam.me.not@dalsemi.com   ------------------------------    Date: 20 Oct 2003 22:02:29 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) & Subject: Re: HR/Payroll Package on VMS3 Message-ID: <GAGoLruvA+H7@eisner.encompasserve.org>   k In article <521ebd9f.0310201634.5de984cf@posting.google.com>, tonymcg27@yahoo.com.au (Tony McGrath) writes: 
 > Hi John, > , > Check out http://www.frontiersoftware.com/ > G > They develop the "CHRIS" payroll package, which we currently use here H > on one of our VMS systems. I think we are one of the last VMS sites to? > still use it. VMS doesn't even get a mention in their list of 9 > supported servers, which you'll find on their web site. A > Please don't ask how good it is, I have no idea, I don't use or  > maintain it.  7 The real question would be, do you get paid on time :-)    ------------------------------  % Date: Mon, 20 Oct 2003 15:19:21 -0400 % From: "John Vottero" <John@mvpsi.com>  Subject: Kerry Main's e-mail/ Message-ID: <vp8d9ppqsm4ibe@news.supernews.com>    Kerry's e-mail address is:   Kerry.Main@hp.com    ------------------------------  % Date: Mon, 20 Oct 2003 23:48:10 -0400   From: John Santos <JOHN@egh.com> Subject: Re: LAN overload?4 Message-ID: <1031020233914.403B-100000@Ives.egh.com>  D On Sun, 19 Oct 2003, Phillip Helbig---remove CLOTHES to reply wrote:  @ > In article <3F9295D2.D16AF012@blueyonder.co.uk>, Tim Llewellyn+ > <tim.llewellyn@blueyonder.co.uk> writes:   > 3 > > Phillip Helbig---remove CLOTHES to reply wrote:  > > > N > > > Just when I was settling down to do some long-needed maintenance work onI > > > my hobbyist cluster, one of the machines (a VAX 4000 100A) left the M > > > cluster and immediately came back (after rebooting).  As far as I know,   > > > this doesn't happen often. > > D > > So, why did it crash?  CLUEXIT? Hardware? I ran a bunch of thoseG > > boxes a while back and the only crashes were due to disk failure or I > > network troubles (multicast and older versions of UCX comes to mind).  > ' > OpenVMS (TM) VAX System dump analyzer  > ' > Dump taken on 18-OCT-2003 21:22:13.50 . > CLUEXIT, Node voluntarily exiting VAXcluster > G > > > The load on the system is not higher than it has been many times. L > > > Still, with a 10 Mb/s ethernet carrying ALL inter-machine traffic (IP,D > > > LAT, SCS---there are MSCP-served disks and shadow sets mountedK > > > distributed across nodes), I'm worrying if I am close to hitting some  > > > limit. > > K > > Wouldn't worry about that. Depending on the cause of the crash it could  > > be network problems. >  > Must have been a loose cable.  > H > I later did BACKUP/IMAGE/VERIFY from CD to disk.  At the beginning of  > the verify pass, I got   > ( >    Excessive packet losses on LAN path > E > for a few seconds, accompanied by errors on PEA0, then the messages ' > went away.  Maybe related, maybe not.   E If you don't already have one, I strongly recommend using an ethernet C switch, rather than a hub or traditional backbone, DELNI, thin-wire  repeater, etc.  B I don't think you need a fancy manageable one for a small network.  A The switch should keep unnecessary traffic off the VAX's segment, E reducing the load on its ethernet adapater and decreasing the chances . that it'll miss a cluster message and CLUEXIT.   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  + Date: Mon, 20 Oct 2003 19:06:22 +0000 (UTC) ? From: Graham Burley <burley.not-this@encompasserve-or-this.org> - Subject: Re: openVMS alpha 7.3.1 on PWS600au? 9 Message-ID: <3F943F41.2E23CD67@encompasserve-or-this.org>    netnews.comcast.net wrote: > < > According to http://ftp.digital.com/pub/DEC/Alpha/firmware > E > The latest SRM Console firmware for digital personal workstation is G > V7.2-1. It refers to my IDE cdrom as dka0, and the scsi disk as dkd0. + > I never see any DQ devices in "show dev".  >  > -Duy    F SET OS_TYPE VMS and INIT and you will see DQ devices instead of DK forB the IDE devices. No idea if it will help with your problem though.     Graham   ------------------------------  # Date: Tue, 21 Oct 2003 02:51:15 GMT 2 From: "netnews.comcast.net" <duon0035@comcast.net>- Subject: Re: openVMS alpha 7.3.1 on PWS600au? > Message-ID: <za1lb.511098$2x.222259@rwcrnsc52.ops.asp.att.net>   Graham Burley wrote:  H > SET OS_TYPE VMS and INIT and you will see DQ devices instead of DK forD > the IDE devices. No idea if it will help with your problem though. >  >  > Graham >     D This is it.  After changing OS_TYPE from OPENVMS to VMS, everything H works.  Thanks everyone for helping, especially Graham.  You're my hero 
 for today.   -Duy   ------------------------------  # Date: Tue, 21 Oct 2003 01:08:42 GMT + From: "Bill Johnson" <res0xcil@verizon.net> 6 Subject: OpenVMS BootCamp BOF - Console Management ROI8 Message-ID: <uG%kb.40949$mp1.25903@nwrddc02.gnilink.net>  2 http://h71000.www7.hp.com/symposium/tditx_bof.html   ------------------------------  % Date: Mon, 20 Oct 2003 20:41:21 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> * Subject: Re: OpenVMS in an HP advertisment' Message-ID: <3F948EC1.B0DA8679@fsi.net>    Fred Kleinsorge wrote: > 0 > "John Smith" <a@nonymous.com> wrote in messageG > news:UREjb.355155$Lnr1.235559@news01.bloor.is.net.cable.rogers.com...  > > L > > Serious question Fred - do you and your engineering colleagues in Nashua< > > ever wonder why HP does not do VMS-specific advertising? > D > Many engineers in the trenches would love to see some VMS-specificN > advertising.  We don't get a vote in it, nor are we privvy to the budget and+ > strategy planning for this type of thing.  > N > But this isn't something HP specific.  This complaint is at *least* a decade > old.   I was just thinking...  G Even if I can't come up with tech. session that gets accepted, I may go E to the next "DECUS" (by the nomme du jour) and pass out T-shirts that B say, "Own stock in HP? Ask them why they don't ADVERTISE OPENVMS".   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Mon, 20 Oct 2003 13:26:42 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>5 Subject: Re: OT: Why Scott McNeally is not worried... ) Message-ID: <3F941AD1.9E27E7BA@istop.com>    Fabio Cardoso wrote:F > b) Motorola (a.k.a. PowerPC) is splitting its Semiconductor Division > in another company;   I The Motorola version of PowerPc is essentially dead as a meanstream  CPU. N Apple has gone to IBM for the "real" Power chips now. Yes, they are labeled asK "PowerPC" and have a few Apple specific gadgets compared to the Power chips ; use in IBM servers, but they are still mostly IBM versions.   B > c) May be it will be acquired by AMD in the Future...Who knows ?  H Nop. AMD wouldn't gain much from acquiring Sun as a whole. What might beM possible is that Sun would "donate" its chip engineers to AMD. (remind you of 
 something ?).   G > In resume: Does not matter the processors anymore ! It becoming just 2 > a piece of hardware. What matters is it speed...  M And what matters is the architecture's flexibility and easy of increasing its M speed. If you have a bloated architecture that needs really special compilers L because half the optimisation is done by compilers, then big chip changes toL increase speed won't benefit software that is already out on the field until, the later is recompiled with a new compiler.  N On the other hand, if you have a simple architecture that makes it much easierF to increase speed, and where the smarts are in the chip instead of theG compiler, then you will have instant results when you upgrade machines.   I Consider the comments made by some VMS engineers in the debate over those N "demo" systems not being ultra fast. One of the arguments was that the currentI IA64 VMS is being compiled with Merced era compilers because the new ones K aren't on VMS yet. It isn't the chip that is important it is the compuilers  for that IA64 thing.  K With IA64, to benefit from better speeds, you need not only change the CPU, N but install new versions of software that was compiled with the new compilers.O Will the new compilers always be available as soon as the new chip comes out ?    N It is one thing for intel to claim X performance gains. But if those gains areL available only on one platform (lets say Linux) and it takes 2 years for theO new compilers to appear on VMS, then VMS will be 2 years behind in performance.   M Upgrading CPus to get better speeds will cost a lot more on IA64 since you'll M also have to upgrade compilers and/or 3rd party software. On other platforms, M you just plug in the new CPU and voila ! Instant performance gains. (although C in fairness, some new featuires may not be used by older sofwtare).    ------------------------------  % Date: Mon, 20 Oct 2003 11:35:12 -0700 # From: "Tom Linden" <tom@kednos.com> 5 Subject: RE: OT: Why Scott McNeally is not worried... 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIKELJIEAA.tom@kednos.com>   @ A more likely scenario is that Sun would abandon Sparc and go to$ X86 architecture, i.e. commodity HW.   >-----Original Message----- 2 >From: JF Mezei [mailto:jfmezei.spamnot@istop.com]( >Sent: Monday, October 20, 2003 10:27 AM >To: Info-VAX@Mvb.Saic.Com6 >Subject: Re: OT: Why Scott McNeally is not worried... >  >  >Fabio Cardoso wrote: G >> b) Motorola (a.k.a. PowerPC) is splitting its Semiconductor Division  >> in another company; > J >The Motorola version of PowerPc is essentially dead as a meanstream  CPU.@ >Apple has gone to IBM for the "real" Power chips now. Yes, they >are labeled as L >"PowerPC" and have a few Apple specific gadgets compared to the Power chips< >use in IBM servers, but they are still mostly IBM versions. > C >> c) May be it will be acquired by AMD in the Future...Who knows ?  > I >Nop. AMD wouldn't gain much from acquiring Sun as a whole. What might be ? >possible is that Sun would "donate" its chip engineers to AMD.  >(remind you of  >something ?). > H >> In resume: Does not matter the processors anymore ! It becoming just3 >> a piece of hardware. What matters is it speed...  > ? >And what matters is the architecture's flexibility and easy of  >increasing its < >speed. If you have a bloated architecture that needs really >special compilersB >because half the optimisation is done by compilers, then big chip >changes to A >increase speed won't benefit software that is already out on the  >field until- >the later is recompiled with a new compiler.  > C >On the other hand, if you have a simple architecture that makes it  >much easierG >to increase speed, and where the smarts are in the chip instead of the H >compiler, then you will have instant results when you upgrade machines. > J >Consider the comments made by some VMS engineers in the debate over thoseC >"demo" systems not being ultra fast. One of the arguments was that  >the currentJ >IA64 VMS is being compiled with Merced era compilers because the new onesL >aren't on VMS yet. It isn't the chip that is important it is the compuilers >for that IA64 thing.  > L >With IA64, to benefit from better speeds, you need not only change the CPU,@ >but install new versions of software that was compiled with the >new compilers. C >Will the new compilers always be available as soon as the new chip  >comes out ? > ? >It is one thing for intel to claim X performance gains. But if  >those gains are? >available only on one platform (lets say Linux) and it takes 2  >years for theC >new compilers to appear on VMS, then VMS will be 2 years behind in 
 >performance.  > A >Upgrading CPus to get better speeds will cost a lot more on IA64 
 >since you'll C >also have to upgrade compilers and/or 3rd party software. On other  >platforms, = >you just plug in the new CPU and voila ! Instant performance  >gains. (although D >in fairness, some new featuires may not be used by older sofwtare). >  >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.524 / Virus Database: 321 - Release Date: 10/6/2003  >  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.524 / Virus Database: 321 - Release Date: 10/6/2003   ------------------------------  % Date: Mon, 20 Oct 2003 15:29:30 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>5 Subject: Re: OT: Why Scott McNeally is not worried... ) Message-ID: <3F943791.B2E78C39@istop.com>    Tom Linden wrote:  > B > A more likely scenario is that Sun would abandon Sparc and go to& > X86 architecture, i.e. commodity HW.  L Since McNealy doesn't like Intel or Microsoft, wouldn't it be more strategic+ for Sun to jump into the Power bandwagon ?    K In the short term, going with AMD's 64 bit 8086 will hurt HP a lot since it L will be one more bit of pressure to push Intel to abandon IA64 and produce a 64 bit 8086.  N But once this is done and intel has standardized on the 8086, then SUN will beK stuck in the same basket as HP, Dell and Microsoft with no differetiator to  make Sun better than HP.    J Carly and Curly and before him Palmer are very generous people, willing toL hurt themselves for the good of Microsoft and Intel. But Sun is out there toK compete. Its decisions aren't designed to help others, they are designed to N make Sun better than others. And I admire Sun for being one of the few ones toK stand up and fight the wintel monopoly, while drones such as HP are burning 1 their own houses to keep Microsoft nice and warm.   I How IBM plays its Power architecture in the next year or two will be very K interesting. Will IBM want to keep its Power to itself and Apple so that it K retains a technological edge over competitors, or will IBM want to start to N make Power more popular in order to gain market share which would further helpM Power grow and make the 8086 look like the toy controlled it started off as ?    ------------------------------    Date: 20 Oct 2003 15:50:19 -0600+ From: young_r@encompasserve.org (Rob Young) 5 Subject: Re: OT: Why Scott McNeally is not worried... 3 Message-ID: <nn9vihX8ckDC@eisner.encompasserve.org>   V In article <3F943791.B2E78C39@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes: > Tom Linden wrote:  >>  C >> A more likely scenario is that Sun would abandon Sparc and go to ' >> X86 architecture, i.e. commodity HW.  > N > Since McNealy doesn't like Intel or Microsoft, wouldn't it be more strategic- > for Sun to jump into the Power bandwagon ?   >   : 	You've got a point.  Sure would make those Solaris to AIX 	migrations go a lot smoother.  M > In the short term, going with AMD's 64 bit 8086 will hurt HP a lot since it N > will be one more bit of pressure to push Intel to abandon IA64 and produce a > 64 bit 8086.  : 	Word leaks out Solaris-x86 is faster and cheaper it couldA 	freeze certain SPARC segments, i.e. 4 processors and under.  But ; 	probably not.  Much of the erosion in low-end SPARC is due > 	to Dell Windows/Linux.  Faster and cheaper.  Sun's hope is to= 	stem the bleeding by offering a faster/cheaper x86 offering.   K > How IBM plays its Power architecture in the next year or two will be very M > interesting. Will IBM want to keep its Power to itself and Apple so that it M > retains a technological edge over competitors, or will IBM want to start to P > make Power more popular in order to gain market share which would further helpO > Power grow and make the 8086 look like the toy controlled it started off as ?   E 	You certainly don't see anyone lined up to migrate to Power, do you? 9 	Power systems are a high-performing high-dollar niche.      				Rob    ------------------------------  % Date: Mon, 20 Oct 2003 17:13:31 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>5 Subject: Re: OT: Why Scott McNeally is not worried... ) Message-ID: <3F944FEB.D6846509@istop.com>    Rob Young wrote:N >         You certainly don't see anyone lined up to migrate to Power, do you?@ >         Power systems are a high-performing high-dollar niche.  L Since apple is using IBM Power chips now (instead of Motorola PowerPC as wasR the case before), I don't think that they could be considered "high-dollar niche".  N And apple is going for the desktop. IA64 doesn't even dare touch the desktop. = So if Sun had to choose between those two, it would be Power.   ; Now, factor in a 64 bit 8086 and things get a bit muddled.    K At this point in time, Power probably has a big edge over the 8086 since it G allows more complex systems to be built for enterprise. Big question is K whether AMD (or Intel when it switches from IA64 to a 64 bit 8086) will add L "enterprise system" features to its 8086s to allow those large systems to be/ built with 8086s. (think Tandem with lockstep).   H Intel could go both ways. The "Digital" mentality would have Intel buildF crippled 64 bit 8086s for desktop only, preserving the IA64 market for. expensive chips that need enterprise features.  J Or Intel could bite the bullet and produce a killer 64 bit 8086 that couldN really go from desktop to datacentre with both desktop and enterprise features? (perhaps in different versions of the chip but with same core).   L If Intel chooses to preserve the expensive IA64, then AMD only needs to jumpN into the fray and produce a 64 bit enterprise ready 8086 with all the features needed for large systems.y   ------------------------------  % Date: Mon, 20 Oct 2003 17:23:42 -0700s# From: "Tom Linden" <tom@kednos.com>o5 Subject: RE: OT: Why Scott McNeally is not worried...$9 Message-ID: <CIEJLCMNHNNDLLOOGNJIOEMDIEAA.tom@kednos.com>.   >-----Original Message-----e3 >From: Rob Young [mailto:young_r@encompasserve.org] ' >Sent: Monday, October 20, 2003 2:50 PMa >To: Info-VAX@Mvb.Saic.Com6 >Subject: Re: OT: Why Scott McNeally is not worried... >P >a3 >In article <3F943791.B2E78C39@istop.com>, JF Mezei $ ><jfmezei.spamnot@istop.com> writes: >> Tom Linden wrote: >>>PD >>> A more likely scenario is that Sun would abandon Sparc and go to( >>> X86 architecture, i.e. commodity HW. >>@ >> Since McNealy doesn't like Intel or Microsoft, wouldn't it be >more strategicc- >> for Sun to jump into the Power bandwagon ?t  = Like?  What has like got to do with?  It is a business issue,oD and Mcnealy is a businessman.  I am sure Microsoft is pleased to seeD HP going down the Itanium lane (1) it was a costly adventure and (2)C it ensures that they will never be able to compete on price.  It is D odd that HP has never taken inventory of what their real assets are,D namely SW. Now they will be squeezed on one side by Microsoft and on> the other by Dell and on a third front by IBM.  What a pickle.  A Power PC is a great architecture, but That is not the determiningSF factor.  Sun can not afford to continue developing Sparc, so they needF to find another horse to throw their saddle on and X86 wins hands down' as an example how about a 3.666 GHz P4.a  L http://www.extremeoverclocking.com/reviews/motherboards/ABIT_IC7_MAX3_3.html   >> >-; >	You've got a point.  Sure would make those Solaris to AIX  >	migrations go a lot smoother.a >eA >> In the short term, going with AMD's 64 bit 8086 will hurt HP a?
 >lot since it A >> will be one more bit of pressure to push Intel to abandon IA64  >and produce a >> 64 bit 8086.m >e; >	Word leaks out Solaris-x86 is faster and cheaper it couldfB >	freeze certain SPARC segments, i.e. 4 processors and under.  But< >	probably not.  Much of the erosion in low-end SPARC is due? >	to Dell Windows/Linux.  Faster and cheaper.  Sun's hope is to > >	stem the bleeding by offering a faster/cheaper x86 offering. >bL >> How IBM plays its Power architecture in the next year or two will be veryC >> interesting. Will IBM want to keep its Power to itself and Appleo >so that itzB >> retains a technological edge over competitors, or will IBM want >to start to> >> make Power more popular in order to gain market share which >would further helpn? >> Power grow and make the 8086 look like the toy controlled it  >started off as ?  >rF >	You certainly don't see anyone lined up to migrate to Power, do you?8 >	Power systems are a high-performing high-dollar niche. >a >				Rob >t >---' >Incoming mail is certified Virus Free.e; >Checked by AVG anti-virus system (http://www.grisoft.com).eA >Version: 6.0.524 / Virus Database: 321 - Release Date: 10/6/2003c >, --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.524 / Virus Database: 321 - Release Date: 10/6/2003   ------------------------------  # Date: Mon, 20 Oct 2003 21:56:27 GMT ' From: Rick Dyson <rick-dyson@uiowa.edu>e% Subject: Product (PCSI) Info Questiont) Message-ID: <3F945A0B.780BD8E1@uiowa.edu>e  I With the old VMSInstal, there was a field kept in the "HISTORY file" that M gave the status of the install and the IVP (if applicable).  I.e., whether it  was successful or failed, etc.  N Does such info exist in the PCSI database?  I can't find any reference to thisM in the SysMan docs I can access on-line.  "Product Show Product /Full" has an0@ "operation" (like Install or Remove), but I can't find a status.  N Can I assume that if it is logged in the database, that it was, by definition, successful?e  M (I am not asking to know if the patch or LP actually works!  Just whether thea+ install process was or was not successful.)a   Thanks!s Rick   ------------------------------  # Date: Mon, 20 Oct 2003 22:24:54 GMTt4 From: brad@.gateway.2wire.net (Bradford J. Hamilton)) Subject: Re: Product (PCSI) Info Questionn/ Message-ID: <WgZkb.831588$Ho3.234892@sccrnsc03>   S In article <3F945A0B.780BD8E1@uiowa.edu>, Rick Dyson <rick-dyson@uiowa.edu> writes: J !With the old VMSInstal, there was a field kept in the "HISTORY file" thatN !gave the status of the install and the IVP (if applicable).  I.e., whether it !was successful or failed, etc.l !eO !Does such info exist in the PCSI database?  I can't find any reference to this0N !in the SysMan docs I can access on-line.  "Product Show Product /Full" has anA !"operation" (like Install or Remove), but I can't find a status.  !d  " (Apologies if this posts twice...)  O Try "$product sho history/full".  There is a column labeled "err" which shows 0 I for all my installations; I'm assuming that means "0 errors", rather than2 "$status=0".	:-)   !snip! !Thanks! !Rick.  J __________________________________________________________________________A Bradford J. Hamilton                    "All opinions are my own"cK bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-' e0                                          with @"   ------------------------------  # Date: Mon, 20 Oct 2003 22:32:27 GMTa4 From: brad@.gateway.2wire.net (Bradford J. Hamilton)) Subject: Re: Product (PCSI) Info Question > Message-ID: <g7Zkb.321314$mp.260594@rwcrnsc51.ops.asp.att.net>  S In article <3F945A0B.780BD8E1@uiowa.edu>, Rick Dyson <rick-dyson@uiowa.edu> writes: J !With the old VMSInstal, there was a field kept in the "HISTORY file" thatN !gave the status of the install and the IVP (if applicable).  I.e., whether it !was successful or failed, etc.v !eO !Does such info exist in the PCSI database?  I can't find any reference to thiseN !in the SysMan docs I can access on-line.  "Product Show Product /Full" has anA !"operation" (like Install or Remove), but I can't find a status.x  O Try "$product sho history/full".  There is a column labeled "err" which shows 0eM for all my installations; since I'm not *that* incompetent, I take it to meand' that 0 = success.  How un-VMS-like!	:-)S   !snip! !Thanks! !Rickr  J __________________________________________________________________________A Bradford J. Hamilton                    "All opinions are my own"rK bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-' d0                                          with @"   ------------------------------    Date: 20 Oct 2003 22:00:45 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)n) Subject: Re: Product (PCSI) Info Question 3 Message-ID: <0wl2rTwT4kqn@eisner.encompasserve.org>t  S In article <3F945A0B.780BD8E1@uiowa.edu>, Rick Dyson <rick-dyson@uiowa.edu> writes:iK > With the old VMSInstal, there was a field kept in the "HISTORY file" thatdO > gave the status of the install and the IVP (if applicable).  I.e., whether itc  > was successful or failed, etc. > , > Does such info exist in the PCSI database?  C As for IVP, PCSI has properly recognized that IVPs must be separate B from installation since there is no guarantee at installation time% that the product has been configured.S   ------------------------------  # Date: Mon, 20 Oct 2003 18:47:16 GMTt2 From: "George Pagliarulo" <georgepag@adelphia.net>+ Subject: Re: question on patch nomenclature 2 Message-ID: <U4Wkb.7427$Ak2.4264@news.cpqcorp.net>  J A full product kit means that the kit installs the complete product, it isI not just a patch.   Some layered products ship this way rather than doingTK partial ECO kits.  Even though the kit installs the full product, you wouldd6 not be able to run it unless you also had the license.   George Pagliarulo$ ECO Release ProcessE OpenVMS EngineeringR Hewlett-Packard Companys    L "Phillip Helbig---remove CLOTHES to reply" <helbig@astro.multiCLOTHESvax.de>/ wrote in message news:bmsfn7$8sa$5@online.de...c >     ECO NUMBER:   DFGE01026  >     ----------H >     PRODUCT:      COMPAQ Disk File Optimizer for OpenVMS Version 2.6-1 >     ---------- >,E >     UPDATED PRODUCT: COMPAQ Disk File Optimizer for OpenVMS Versionb 2.6-E01t >     ---------------n >r# >     APPROX BLK SIZE: 30051 blocks  >     ---------------- >oE >     Cover Letter for COMPAQ Disk File Optimizer for OpenVMS Version  2.6-E01e >o >     Kit Description: >     ----------------$ >     This kit is a full kit update. >sF > What does "full kit update" mean?  Does it mean that it replaces allB > files in the 2.6 kit?  Or does it mean that I can install 2.6 byD > installing this patch, even if I am running a lower version of the > product or even none at all? >    ------------------------------  % Date: Mon, 20 Oct 2003 21:18:58 +0200a" From: Didier Morandi <no@spam.com> Subject: Re: rwast2 Message-ID: <3f943526$0$227$626a54ce@news.free.fr>   mckinneyj@cpva.saic.com wrote:M > You might find the following article useful (paste the two lines together).h > > > http://h18000.www1.hp.com/support/asktima/operating_systems/ > 0094A663-57DAB060-1C0069.htmlu   Or use tinyurl.com.tY http://h18000.www1.hp.com/support/asktima/operating_systems/0094A663-57DAB060-1C0069.htmlh = http://tinyurl.com/rmu2    D.   ------------------------------  % Date: Mon, 20 Oct 2003 21:08:29 -0500 5 From: "Martin P.J. Zinser" <zinser@zinser.no-ip.info>x/ Subject: Re: SMTP server behind an ADSL Switch?c9 Message-ID: <bn24eu$rpb14$1@ID-209632.news.uni-berlin.de>e   Didier Morandi wrote: D > What should I do to have an SMTP/POP server running on my VMS box % > instead of using the one of my ISP?n > F > The box is behind an ADSL router. I NATted the FTP, HTTP and TELNET 8 > ports to it, but is there such SMTP port to "NAT" too? > + > A good HOW2 would be greatly appreciated.  > 	 > Thanks,r >  > D.H This being VMS I declared my VMS system to be the DMZ host in my router,B i.e. all ports are forwarded to my system. Adding an MX record is H required as others have mentioned already. I use No-ip for both DNS and H MX and that works fine for me. Other dynamic DNS providers might support MX records too.t   Greetings, Martine   ------------------------------  % Date: Mon, 20 Oct 2003 13:40:32 -0400a* From: JF Mezei <jfmezei.spamnot@istop.com>1 Subject: Supporting RMS calls for "other" devicess) Message-ID: <3F941E0E.EA05290B@istop.com>m  J Ok, with RMS, one can open a file, open a mailbox device, a decnet task orH file, and even a TX/TT device. And for some devices, namely tape, an ACP process is created.c  G What exactly does RMS need to see to determine that it will or will noti. SYS$OPEN to such a file specification/device ?  K Is it possible to write some device driver that RMS will find compatible asTK soon as it sees it, or is the list of supported device types hardcoded intoi RMS  ?  L If one wanted to allow RMS to access ICC services as a client, what would be? the best way to implement this ? (same applies to TCPIP links).%   ------------------------------   Date: 20 Oct 2003 17:41 CDTM. From: carl@nospam.gerg.tamu.edu (Carl Perkins)- Subject: Re: SYSMAN> DEFINE/KEY/IF_STATE=GOLD 4 Message-ID: <20OCT200317415128@nospam.gerg.tamu.edu>  y In article <bms7hb$1ul$3@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes...aG }I'm hoping to save myself some time in the future by coming up with a OI }sensible SYSMANINI file.  It seems to work OK as far as it goes, but it  G }doesn't go far enough.  In particular, /IF_STATE=GOLD doesn't seem to   }work. } H }I am assuming that PF1 is GOLD and that I don't have to define this keyC }to be the GOLD key or whatever.  This approach seems to work OK in0 }MAIL$KEYDEF.INI.  } . }After some experimentation, I have found that } $ }   DEFINE/KEY/SET_STATE=GOLD PF1 "" }  }seems to work.e } J }So, let me rephrase my question: why is the above command apparently the , }implicit default in MAIL but not in SYSMAN?  F I would guess that it follows the DEFINE/KEY command in DCL. With thatD there is no "gold key". The state can be any alphanumeric string, itG doesn't have to be "gold". There can be many different states that makex  other keys function differently.  ) In fact, the help in SYSMAN says as much:E   DEFINE     /KEY       /SET_STATE             /SET_STATE           /NOSET_STATE  E        Associates a state with the key you are defining. A state name A        can be any alphanumeric string. If you omit or negate thiseH        qualifier, the current state remains unchanged. You cannot defineA        a key using both the /SET_STATE and /TERMINATE qualifiers.   F This is perhaps slightly less clear than the DCL help's equivalent, btF still note that there is no such thing as a specific "GOLD" state, and0 therefore there is no such thing as a "GOLD key"  I PF1 is PF1, not GOLD. It can be defined to be anything you want it to be. 0 Any key can be defined to set the state to GOLD.   --- Carl   ------------------------------  % Date: Mon, 20 Oct 2003 21:28:16 +0200 " From: Didier Morandi <no@spam.com>@ Subject: VAX/VMS to Itanium Migration Letter available in French2 Message-ID: <3f943754$0$253$626a54ce@news.free.fr>  H http://h21007.www2.hp.com/dspp/ne/ne_PartnerNewsList_IDX/1,1406,,00.html  5 FYI, French speaking Europe, North-Africa and Canada.H   D.   ------------------------------  # Date: Mon, 20 Oct 2003 18:04:19 GMT"3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond)   Subject: Re: VMS v8.2 has a name2 Message-ID: <DsVkb.7421$7c2.3389@news.cpqcorp.net>  4 In article <jnTOTE31caAP@eisner.encompasserve.org>, / Kilgallen@SpamCop.net (Larry Kilgallen) writes:a  G >I would presume the VAX VMS 8.2 and Alpha VMS 8.2 versions do not have  >that name.u  G Both OpenVMS Alpha 8.2 and OpenVMS I64 8.2 are built from the same code D base (with build/compile time differences recognized).  So they will both habe the Topaz code name.  I OpenVMS VAX 8.2 will be built from its own code stream; it may or may not  have the same code name.   -- wJ       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------    Date: 20 Oct 2003 14:18:33 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)   Subject: Re: VMS v8.2 has a name3 Message-ID: <C2JEEhXIvMrf@eisner.encompasserve.org>J  h In article <DsVkb.7421$7c2.3389@news.cpqcorp.net>, hammond@not@peek.ssr.hp.com (Charlie Hammond) writes:6 > In article <jnTOTE31caAP@eisner.encompasserve.org>, 1 > Kilgallen@SpamCop.net (Larry Kilgallen) writes:D > H >>I would presume the VAX VMS 8.2 and Alpha VMS 8.2 versions do not have >>that name. > I > Both OpenVMS Alpha 8.2 and OpenVMS I64 8.2 are built from the same codesF > base (with build/compile time differences recognized).  So they will  > both habe the Topaz code name. > K > OpenVMS VAX 8.2 will be built from its own code stream; it may or may not  > have the same code name.  < But my reference to "that name" was to the original poster's- "OpenVMS for Integrity Servers" or some such.e   ------------------------------    Date: 20 Oct 2003 14:16:10 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) ! Subject: Re: What is ACME_SERVER?c3 Message-ID: <R+0EtT5RlKWs@eisner.encompasserve.org>d  e In article <SiUkb.601186$cF.265400@rwcrnsc53>, brad@.gateway.2wire.net (Bradford J. Hamilton) writes:te > In article <WWFpsp1ODl1v@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:hi > !In article <nvRkb.188893$%h1.179230@sccrnsc02>, brad@.gateway.2wire.net (Bradford J. Hamilton) writes:  > !gR > !> this correctly.  I don't think I'll ever be using ACM Services on the machineL > !> at work (now or in the future), so I can't see the need for the server. > !aN > !So all your software vendors have guaranteed they will never call SYS$ACM ? > K > Our software "vendors" are limited to those providing software for faxes.rG > No guarantee, but I don't see faxing software as needing that kind ofs
 > capability.r  E Fair enough.  Just beware of the possibility that for some version of-E VMS the non-SYS$ACM version of LOGINOUT might no longer be supported.    ------------------------------  % Date: Mon, 20 Oct 2003 15:51:54 -04000* From: JF Mezei <jfmezei.spamnot@istop.com>! Subject: Re: What is ACME_SERVER?e) Message-ID: <3F943CCF.8911B5B7@istop.com>F   someone wrote:G > Fair enough.  Just beware of the possibility that for some version of:G > VMS the non-SYS$ACM version of LOGINOUT might no longer be supported.   I Since ACM is not supported on VAX, as long as VMS continues to be able to J cluster with Alphas, wouldn't it be logical to state that non-ACM LOGINOUTB will be supported at least until HP stops development on VAX VMS ?   ------------------------------   Date: 20 Oct 2003 19:40 CDT . From: carl@nospam.gerg.tamu.edu (Carl Perkins)! Subject: Re: What is ACME_SERVER?e4 Message-ID: <20OCT200319400193@nospam.gerg.tamu.edu>  ! brad@.gateway.2wire.net writes...od }In article <FRmojpvbR+nC@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:w }!In article <SpDkb.505706$2x.211633@rwcrnsc52.ops.asp.att.net>, brad@.gateway.2wire.net (Bradford J. Hamilton) writes:hg }!> In article <V8YK+mItTUvk@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:  }!( }!> !Are sure you are running VMS V7.3 ? }!> !s }!>  }!> Yup. }!7 }!Do you have system parameter STARTUP_P2 set to "VD" ?i } O }Yes, I do.  It looks as though the startup routine for ACME_SERVER just startsf4 }up, without any reference to checking the logicals: } 1 }RABBIT::BRAD$ search sys$system:startup.log acmePE }$ DEFINE ACME$TO_BE_STARTED FALSE                       ! ACM Server,K }$ DEFINE NTA$NT_ACME_TO_BE_STARTED FALSE                ! NT ACM ExtensiondI }$if f$search("SYS$STARTUP:VMS$CONFIG-050_ACME_SERVER.COM").nes."" then -eL }$if stdrv$execut then $ @SYS$STARTUP:VMS$CONFIG-050_ACME_SERVER.COM  "FULL" }"VD"h } "" "" "" "" "" ""r }$set server acme/start/logc, }$set server acme/config=(name=vms,cred=vms) }$set server acme/enable } G }Notice the if statements; they only seem to check for the existence oftK }ACME_SERVER.COM, and to some logical called "stdrv$execut".  I've followeddK }this back as far as I can, but I can't see that the logic takes the *ACME*- }logicals into account at all. } < }The same behavior exhibits itself in a V7.3 system at work. } K }__________________________________________________________________________.B }Bradford J. Hamilton                    "All opinions are my own"L }bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-' 1 }                                         with @"t  A Interesting: I appear to ahve exactly the opposite problem. On myoE system I can find absolutely nothing that would stop it from running.oB I can also see that the system startup proceedure shows that it is1 supposed to be starting it. But it isn't running.S  J This is on VMS7.2-1, which has two different startup files for this thing.  N The file SYS$STARTUP:VMS$CONFIG-050_ACME_SERVER.COM does not check the logicalG names to see if it should start or not. It will start the server unlesss1 you are runing a minimum boot or an upgrade boot.o  L The file SYS$STARTUP:ACME$STARTUP.COM does actually check the logical names.@ If it gets passed the checks, it then uses the other file above.  	 If you dow $ mcr sysman. SYSMAN> start set database startup$startup_vmsG %SYSMAN-I-NEWCOMPFIL, current component file is now STARTUP$STARTUP_VMSh SYSMAN> start show file > %SYSMAN-I-COMFIL, contents of component database on node GERGX Phase        Mode   File [...] 2 CONFIG       DIRECT VMS$CONFIG-050_ACME_SERVER.COM  J So the first question is, why is it directly running the first one and not* the second one like it seems it should be?  = Another question is, if the system startup is already runningm6 VMS$CONFIG-050_ACME_SERVER.COM, then why does the fileG SYS$STARTUP:VMS$END-050_COORDINATED.COM (which is the last thing listedoH in the SYSMAN startup list) then run ACME$STARTUP.COM? It should alreadyJ be running since the earlier file starts it unless you are booting minimum or upgrade.n  I Yet another question is, given all this, why is it that it is NOT runningmJ on my system? The SYSMAN startup list clearly shows that it is supposed toC be running VMS$CONFIG-050_ACME_SERVER.COM, which certainly does theo   $set server acme/start/log+ $set server acme/config=(name=vms,cred=vms)s $set server acme/enables  D sequence of commands. Then the VMS$END-050_COORDINATED.COM should beE running ACME$STARTUP.COM. So why is the ACME server not running? I doyG not have the logical name defined, so there are (at least) two separatecC places in the system startup where it should apparently be started.i  B I can, in fact, start it manually with the same command that is inC VMS$CONFIG-050_ACME_SERVER.COM ($ set server acme/start/log). ThereuD is no ACME$SERVER.LOG file around unless I do that command manually.  H But it isn't running and there is never an ACME$SERVER.LOG file anywhereI (I even used DFU to search the entire disk, in case it was someplace that/ is not obvious).  D It seems to me that the correct behavior should be somehwere between your situation and mine.   --- Carl   ------------------------------  % Date: Mon, 20 Oct 2003 14:36:30 -04009* From: JF Mezei <jfmezei.spamnot@istop.com>. Subject: Re: What is the performance of iVMS ?) Message-ID: <3F942B28.F8528B76@istop.com>r   Fred Kleinsorge wrote:F > In my opinion, it was a mistake to make the VAX and Alpha code basesI > seperate - but even without that split - there are some tough problems.w  K Out of curiosity, could you have used the current exercise of reviewing allnN the code to make sure it runs on that IA64 thing  to re-integrate VAX into theF same code base ? I realise that some of the stuff would be better leftI separate due to lack of 64 bits, but wouldn't a LOT of the code on VAX bed= easily integrated with that of Alpha  in a common code base ?     L > It's a tough call.  Why isn't PPP available?  Assuming that it isn't, it'sH > probably because the customer who needed it - didn't need it for VAX.   L Ironic because it is far more likely that a VAX would have serial ports thanJ an Alpha which would have terminal servers with native PPP support. So VAX# would need PPP far more than Alpha.h  M At least if you guys were to fix the SLIP drivers on vax because we are force 6 tyo use this antiquated technology due to lack of PPP.  N > The last VAX was sold years ago.  It's probably still doing what it was soldN > to do just fine.  But at what point when you push it beyond what it was sold9 > to do, do you have to bite the bullet and move forward.r  D Nop. Just because you arte forced to remain on VAX due to one or twoM applications doesn't mean that your machine is stangnanbt and not getting new K stuff on it when/if it becomes available. Remember that VMS was designed tocF support many things at the same time, unlike wintel boxes that run one application per box.  K > [BTW - we didn't "abandon" Display Postscript - Adobe forced it as a siden > effect of the merger].  N So essentially, Compaq , being a wintel slave, suffered because of Microsoft's dislike of Adobe.s   ------------------------------  # Date: Mon, 20 Oct 2003 19:59:09 GMTh9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>i. Subject: Re: What is the performance of iVMS ?2 Message-ID: <h8Xkb.7439$x02.3388@news.cpqcorp.net>  7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messagem# news:3F942B28.F8528B76@istop.com...r > Fred Kleinsorge wrote:H > > In my opinion, it was a mistake to make the VAX and Alpha code basesK > > seperate - but even without that split - there are some tough problems.a >pI > Out of curiosity, could you have used the current exercise of reviewinge alloL > the code to make sure it runs on that IA64 thing  to re-integrate VAX into the1H > same code base ? I realise that some of the stuff would be better leftK > separate due to lack of 64 bits, but wouldn't a LOT of the code on VAX bee? > easily integrated with that of Alpha  in a common code base ?a >a  L Nope.  Not only would the re-integration be a lot of work, the testing wouldH dwarf the engineering work.  Do you *really* want instability in the VAX
 code base?  K > > The last VAX was sold years ago.  It's probably still doing what it was  soldK > > to do just fine.  But at what point when you push it beyond what it wast sold; > > to do, do you have to bite the bullet and move forward.M >rF > Nop. Just because you arte forced to remain on VAX due to one or twoK > applications doesn't mean that your machine is stangnanbt and not getting  neweJ > stuff on it when/if it becomes available. Remember that VMS was designed toH > support many things at the same time, unlike wintel boxes that run one > application per box. >t  I But you have to understand that the hardware *is* stagnant.  When you put J new stuff on it, you inevitably take the risk that the underlying hardwareJ just won't cut it.  Supporting multiple things has nothing to do with thisK fact of life - you can't expect old HW to run SW designed for hardware that I is 5 or 10 times faster.  And from our (the vendor) point of view, makinglJ investments in that older hardware base becomes a low payback investment -I we won't sell new VAXes to run the SW.  And every new place we might pushiJ the VAX, will simply increase the risk of breaking something that's worked for decades.  H > > [BTW - we didn't "abandon" Display Postscript - Adobe forced it as a side > > effect of the merger]. >aD > So essentially, Compaq , being a wintel slave, suffered because of Microsoft'se > dislike of Adobe.r  L Eh?  Even if I somehow agree with the slander that we are a "wintel slave" IK don't see how you get from here to there.  Adobe doesn't want to do DisplayiL Postscript for X11 anymore, they took advantage of a poorly written contractJ (IMHO) that was specific to Compaq, and not to any company that might take@ it over.  Microsofts "dislike" for Adobe if there is any, or theI characterization of Compaq as a wintel slave - didn't have anything to do  with it.  L There are other ways to display postscript, and there are other formats thanB postscript.  The only real people hurt here are those who actuallyL programmed to the DPS extension...  and frankly, this was always something IG found weak about X11 -- extensions are not supposed to be counted on byBJ applications (extensions are optional), but far too many do count on them,J and far too many can't function without them.  So they become requirements instead of optional extensions.m   Oh well.   ------------------------------  % Date: Mon, 20 Oct 2003 22:36:14 +0200y( From: "H Vlems" <hvlems.nieuw@zonnet.nl>( Subject: Re: [~OT] Last _production_ VAX9 Message-ID: <bn1h44$s2s92$1@ID-143435.news.uni-berlin.de>e  = "William Webb" <al5vf03p02@sneakemail.com> schreef in berichtm7 news:d5ce4b06.0310200557.320a441b@posting.google.com...a5 > "B Fisher" <none@antispamming.com> wrote in messages5 news:<q_nkb.47525$%C5.11359@twister.rdc-kc.rr.com>... 	 > > Andy,a [snip] > > >s >aJ > We had some VAX 4000 Model 108s with serial numbers starting with NI8220 whichtD > indicates that the machines were manufacatured on Day 220 of 1998. >s > WWWebb  ! Wouldn't that be week 22 of 1998?-   ------------------------------   Date: 20 Oct 2003 18:36 CDTm. From: carl@nospam.gerg.tamu.edu (Carl Perkins)( Subject: Re: [~OT] Last _production_ VAX4 Message-ID: <20OCT200318363402@nospam.gerg.tamu.edu>  : "Andy Bustamante" <a_c_bustamante@earthlink.net> writes...4 }"B Fisher" <none@antispamming.com> wrote in message4 }news:72dkb.54389$832.26759@twister.rdc-kc.rr.com...I }> Slightly off topic but:  Does anyone have an idea of when the last VAXwM }> was/is produced?  My recollection is that the last introduction was aroundsI }> 1992 or so, but I am more interested in the last _produced_.  We stilld }haveeG }> a few but I suspect when they get repaired, they are refurbed boardsh }> }> Just a curiosity, Thanksn }> }> scottK }I remember writing  a reassuring letter to our customers pointing out thatHF }Alpha was the natural replacement; Compaq wasn't selling any more VAX> }Systems just as Intel didn't care to sell any more 486 chips. } H }" . . . MicroVAX 3100 platforms will be orderable through September 30,0 }2000. "  -Terry Shannon on the last VAX orders. } 7 }http://h18002.www1.hp.com/alphaserver/vax/vax_now.html  }  }--g }Andy Bustamante  E That was systems though. They stopped making the actual VAX CPU chipsoD years before that - I don't remember when but I'm pretty sure it wasH at least 2 years earlier, and possibly 4. They produced a largish numberC of them and just drew them out of the warehouse as needed for quiteo a while.   --- Carl   ------------------------------   End of INFO-VAX 2003.583 ************************her helpM Power grow and make the 8086 look like the toy controlled it started off as ?    ------------------------------    Date: 20 Oct 2003 15:50:19 -0600+ From: young_r@encompasserve.org (Rob Young) 5 Subject: Re: OT: Why Scott McNeally is not worried... 3 Message-ID: <nn9vihX8ckDC@eisner.encompasserve.org>   V In article <3Fo"E8G|'ɲ\MOWC$]1Q>cRc67ʒ8*<$OrEl	75I;. ,v8-*[hwhGZvܞsVC=榝%j!)l&_B;:e}/jDzǋ;is_&9QoWp'@1r#A&ߕJ~ז{ڶq`ƚŭ֥~Gtr-Ėc|U'Cv;G9<ݎ:9?}t,-S}xA)ŏyX?Mz{asCd<vxg+LO;eZ 	Rn+7<1<Wmxgx/dۛ'bHZ#")g~,Vл1qX٘SEv,F",I<$\Sl	>ԧtrKqCE#ɶh[ߐ۱B:NsK'tK\p)}.sQ{cۓӉcp+
lτ3=zW u(^JΗ[zWp8tTI4g;JĩF'A<eKD~JNsZVT}ގ2\FψE(k9py:!^wVg".A4/M9N"ɩjS/s"K{MOU@yl侫Fy"asPcx2^mx>3!2\GX.Ǳ,8| m<3brl)O7PN`d_53TphG
>:QQQ\x}XcߨElO;?C1gzQd7r;x԰U1?[MrvflU70zzxo_![q]^%(uh7}|Hvzai6 l0m?b:9e>5lnzp֪S?TfK.9<#P,Gp5Eг
n"C-L nhJޫTtJhz>U޲UԆx-sN? Y율ܨA7I'+d<ujy&KƈǏM_p<23{@Ab_lMxg0m\i쁩G_Ri#/4yR/@uL᯾?Ż4H~9rg@nKfUcߴǤB|gV%Ws,{6"yZ9m9>_XSrfe>6u?GP&?{]埥?_a?Ɇd?w4<M)-ɿToH^K%c">H<]"
6)-x(~x?.Ň
.Xuh`󡱂M<y>2\y5T5T'Jp>MuQTpEm)b)}$T8)3jpk[ukb@CچkEX#%FᨧPᎥt}+%{r	
r125b 
E:,Q3=5:	+#s!{q!(o44
6ukW46	dOᨧPh΄KV9uc]!
5Nٳ(y̞
'ґ9u4YkZ`sr>HNpRd~4\}
yhiink>sθNB3ᤇ1is7|N3`JG榇!
q޾Xx%0:.STMyͮoo.=C\ƦIpR:2G==,[6Q۳JسZu&\=k4o{lpQskp{
9pQǵÔ-ۙg[tdnz8-th<aM+7ٮMPpI4~'C_nj)Qs+Y3ssJ禬tdnz8-u<fS:.K rakFo9j&Th'R:z:hJcIg*ezNU, Mji8ۙ<CK&p!n
\fMV%[ӹ)8Ip<)yp[2nx{շ4qIpR:pspNokiNOjЊBtnvhkNADqS_!sTDO7ݙ+8ptWIۯbp8ۙ*S:\2UNz8ĭkUdlI먖D|\eMMPᔧtdnz8mcKm50uͼGScV'tdnz8ZMv7oƟh3%q[u:rIp:S:<ਫtn5Z*[j:+q!)jmh	nr"mᨧPk΄K[4o֭
@C+u

Vlu8&T3׸ήr;`\3qQG)2ډ3q>n8qjۖhk
:1qjlPQ:N[Aݞґw{5xJvt8^ξWu&\=9P<=G?=GS:2G=57tnf>sHpsp깎p>:ڞU9ٞ͟ݞє?!Qs;p;9
i:3V@nKy@6Rt<qt9ԢtN/I@Ur͟h,Y@C5m^XuSh$T8Cz8jK