1 INFO-VAX	Tue, 27 Aug 2002	Volume 2002 : Issue 471       Contents: "inview" Article Re: "inview" Article Re: "inview" Article Re: "inview" Article Re: "inview" Article Re: "inview" Article Re: "inview" Article> Re: ??== Disk mount brackets 74-48510-01 for Alphastation 400." Re: Automating EDIT/FDL ->OPTIMIZE$ Re: DE101 - can it go to Full Duplex$ Re: DE101 - can it go to Full DuplexI Re: DECC (specifically v6.2-008, but older maybe acceptable) docs ONLINE?  K12 (Was Roadshow update)  Re: Low-level format SCSI disk Re: Low-level format SCSI disk Re: OpenVMS + Joint STARS  Re: OpenVMS + Joint STARS . Re: OT Mac OS X woes, but advice needed please# Re: PGP assets (mostly) sold by NAI  Re: Postscript printing on vms RE: Postscript printing on vms Re: Postscript printing on vms Re: Postscript printing on vms Re: Postscript printing on vms& Re: Problem with JDK1.3 on OpenVMS 7.3. Re: Procedure to enlarge system name required.4 Process Software Multinet NFS SRI file name encodingP Re: putenv()/setenv() was - Re: DECC (specifically v6.2-008, but older maybe acc  Quantum DLT 4000 and OpenVMS VAX$ Re: Quantum DLT 4000 and OpenVMS VAX Re: rename c rtl bug Re: Reverse LAT  Re: Reverse LAT  Re: Reverse LAT  Re: Reverse LAT  Re: Reverse LAT + Server Market Is Down in the Second Quarter   Re: set file/data_check (REPOST)  Re: set file/data_check (REPOST)" RE: silliest thing heard this week" Re: silliest thing heard this week" Re: silliest thing heard this week8 Re: Size of I/O count fields in SHOW PROC and SHOW SYS ?8 Re: Size of I/O count fields in SHOW PROC and SHOW SYS ?8 Re: Size of I/O count fields in SHOW PROC and SHOW SYS ? Re: SYS$EXAMPLES:ADDUSER.COM Using C - Api in PASCAL code  Re: Using C - Api in PASCAL code  RE: Using C - Api in PASCAL code  RE: Using C - Api in PASCAL code* Re: Want an alpha or 20?  Found an auction* Re: Want an alpha or 20?  Found an auction What's going on with Encompass? " Re: XFC patch for VMS 7.3 is up..." Re: XFC patch for VMS 7.3 is up..." Re: XFC patch for VMS 7.3 is up... RE: XFC v2 ECO, negative report  Re: XFC v2 ECO, negative report   F ----------------------------------------------------------------------  # Date: Mon, 26 Aug 2002 17:47:23 GMT ( From: Don Sykes <annonymous@pacbell.net> Subject: "inview" Article + Message-ID: <3D6A6A63.E34753D8@pacbell.net>   J Did anyone else see HP's "inview" bi-monthly for Aug. In it is an article,; cloaked as an interview with Jim Carlson. The title reads :  	"riding the wave N 	HP's Jim Carlson talks about explictly parallel instruction computing (EPIC),D 	the Intel Itanium processor family and the next wave of computing."  > In it he is asked, "Where is Itanium-based technology headed?"
 He respondes:  	...N 	HP is inventing where it makes sense, using our capital wisely to ensure thatO our multi-OS strategy enables a choice among three operating systems to address P varying business needs - HP-UX, Linux and 64-bit Windows - and be able to switch, operating systems as business needs change." 	...  K No where in the article, or as far as I could see, in the magazine, did the  letters VMS appear.   % Have to go now. I'm feeling nauseous.  --     Have VMS. Will Travel. Wire Paladin (@alphase.com) 
 San Francisco    ------------------------------    Date: 26 Aug 2002 15:22:40 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)  Subject: Re: "inview" Article 3 Message-ID: <Ghoo+BAClowz@eisner.encompasserve.org>   V In article <3D6A6A63.E34753D8@pacbell.net>, Don Sykes <annonymous@pacbell.net> writes: > L > Did anyone else see HP's "inview" bi-monthly for Aug. In it is an article,= > cloaked as an interview with Jim Carlson. The title reads :  > 	"riding the wave P > 	HP's Jim Carlson talks about explictly parallel instruction computing (EPIC),F > 	the Intel Itanium processor family and the next wave of computing." > @ > In it he is asked, "Where is Itanium-based technology headed?" > He respondes:  > 	...P > 	HP is inventing where it makes sense, using our capital wisely to ensure thatQ > our multi-OS strategy enables a choice among three operating systems to address R > varying business needs - HP-UX, Linux and 64-bit Windows - and be able to switch. > operating systems as business needs change." > 	... > M > No where in the article, or as far as I could see, in the magazine, did the  > letters VMS appear.    Well, he also left out NSK...    ------------------------------  % Date: Mon, 26 Aug 2002 14:58:52 -0600 $ From: Lee Y T Mah <lytmah@cha.ab.ca> Subject: Re: "inview" Article ) Message-ID: <3D6A968C.93ACEC45@cha.ab.ca>    Don,  X How reliable is this article?  How do I get at it?  We had our quarterly conference callY with HP this afternoon and I referred to the interview.  I offered to forward the article  to the HP TAM.     Don Sykes wrote:  L > Did anyone else see HP's "inview" bi-monthly for Aug. In it is an article,= > cloaked as an interview with Jim Carlson. The title reads :  >         "riding the waveW >         HP's Jim Carlson talks about explictly parallel instruction computing (EPIC), M >         the Intel Itanium processor family and the next wave of computing."  > @ > In it he is asked, "Where is Itanium-based technology headed?" > He respondes: 
 >         ... W >         HP is inventing where it makes sense, using our capital wisely to ensure that Q > our multi-OS strategy enables a choice among three operating systems to address R > varying business needs - HP-UX, Linux and 64-bit Windows - and be able to switch. > operating systems as business needs change."
 >         ...  > M > No where in the article, or as far as I could see, in the magazine, did the  > letters VMS appear.  > ' > Have to go now. I'm feeling nauseous.  > -- >  > Have VMS. Will Travel. > Wire Paladin (@alphase.com)  > San Francisco    -- Lee   ; Lee Y T Mah                        Capital Health Authority ? Email: lytmah@cha.ab.ca            Information Systems, RAH CSC 4 Phone:  (780) 477-4725, 477-4233   10240 Kingsway NW? Fax:      (780) 491-5119, 491-5619    Edmonton, AB, CAN  T5H3V9    ------------------------------  # Date: Mon, 26 Aug 2002 20:24:44 GMT # From: "John Smith" <a@nonymous.com>  Subject: Re: "inview" Article G Message-ID: <g8wa9.48138$bu81.143@news02.bloor.is.net.cable.rogers.com>   5 "Don Sykes" <annonymous@pacbell.net> wrote in message % news:3D6A6A63.E34753D8@pacbell.net...  > L > Did anyone else see HP's "inview" bi-monthly for Aug. In it is an article,= > cloaked as an interview with Jim Carlson. The title reads :  > "riding the waveG > HP's Jim Carlson talks about explictly parallel instruction computing  (EPIC), E > the Intel Itanium processor family and the next wave of computing."  > @ > In it he is asked, "Where is Itanium-based technology headed?" > He respondes:  > ... J > HP is inventing where it makes sense, using our capital wisely to ensure thatI > our multi-OS strategy enables a choice among three operating systems to  address K > varying business needs - HP-UX, Linux and 64-bit Windows - and be able to  switch. > operating systems as business needs change." > ...  > I > No where in the article, or as far as I could see, in the magazine, did  the  > letters VMS appear.  > ' > Have to go now. I'm feeling nauseous.  > --    ; http://www.hp.com/execcomm/inview/august02/aug02_epii4.html    The Preamble --------------D "In the near future, HP systems will bring down the cost of scale-upJ computing dramatically. HP is delivering the technology to get there, withG systems that support HP-UX, Windows and Linux. Jim Carlson, Director of K Marketing for Itanium-based systems, was involved from the start with Intel I in planning the marketing rollout of the Intel Itanium processor family."      In the body of the article ------------------------* "Where is Itanium-based technology headed?K Given that all vendors in the game will use the same Itanium processors and J in many cases the same systems boards or servers, hardware will not be theG primary differentiator in this new market. The emphasis will shift from K performance, price and features to total value. HP is the only major vendor J dedicated to moving the entire server line toward a single market-unifying architecture."  F Duh. Maybe today. Guess he never heard of an architecture called Alpha before.     I "HP is inventing where it makes sense, using our capital wisely to ensure L that our multi-OS strategy enables a choice among three operating systems toI address varying business needs - HP-UX, Linux and 64-bit Windows - and be L able to switch operating systems as business needs change. We are engaged inG a number of programs with our software partners to provide the greatest  options anywhere."  > To paraphrase that ad from the late '70's - 'Where's the VMS?"    K "One of our most successful programs is "Designing the Future," where we're B helping our partners prepare their HP-UX, 64-bit Windows and LinuxJ applications for Itanium. Through this program, our partners get access toI development tools, consulting, training and our Partner Technology Access L Centers. Why things connect becomes more important than how they connect, inK cases where information technology capabilities are provisioned, delivered, , metered, managed or purchased as a service."  J Guess HP isn't too interested in helping their VMS partners. With articlesH like these, it seems to show that HP doesn't WANT any VMS partners...the+ sooner they can be driven away, the better.     < "How would you recommend that customers decide which to use?L We know that there are three operating environments available, so we offer aL choice of platforms. And we work with the customers to build and maintain an@ agile infrastructure based on their heterogeneous or homogeneous
 environment."   E 3 environments ...the BSOD one, the not-ready for prime-time bet your B business penguin, and arguably the weakest overall commercial unixI implementation. What an outstanding menu to choose from. Since they don't B mention it, I'd guess VMS is owned by another company as of today.      H "We believe that having a choice of e-business platforms for applicationJ development, deployment and management is key to our customers' success in5 achieving or maintaining their competitive advantage. D HP with Itanium 2 provides flexibility for a choice of commercial orH technical platforms. HP with HP-UX 11i, Linux or Windows offers enhancedJ computing performance today and with a solid roadmap for the near future."    By inference, VMS has no future.      E "What's in store for HP customers in the near future, say, next year? B HP is moving forward with a roadmap featuring three microprocessorL architectures: PA-RISC, IA-32 and Itanium. HP will maintain parallel PA-RISC: and Itanium paths at least until the PA-8800 is announced.D It is possible that we will continue PA-RISC beyond the PA-8800. TheC decision will be based on customer feedback. Our goal is to provide G customers with the flexibility to move to Itanium at their own pace. HP J recognizes that our customers and partners will not respond favorably to aJ fixed deadline to transition to Itanium. With the recent HP/Compaq merger,E we will now phase the former Compaq product lines, the Alpha, VMS and F Non-Stop systems, into the overall plan to transition to Itanium-basedK products. We will continue to support the commitments made to Alpha and VMS I customers as we converge these systems into a single, unified product and  processor architecture."  L In other words...Itanium may be a piece of sh*t and they have to hedge theirL bets, and even if it isn't total cr*p, customers may not want it, and HP mayJ not be able to afford to ignore the money they wave at HP for PA8800-based machines instead of Itanic.   L And as to those 'commitments', HP have made no commitments for any other o/sE products to only support them until 2011 so you're safe buying those.       2 "Eventually, all HP systems will move to Itanium."  
 We shall see.    ------------------------------    Date: 26 Aug 2002 15:56:13 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)  Subject: Re: "inview" Article = Message-ID: <cf15391e.0208261456.74aa7b7e@posting.google.com>   [ Don Sykes <annonymous@pacbell.net> wrote in message news:<3D6A6A63.E34753D8@pacbell.net>... M > No where in the article, or as far as I could see, in the magazine, did the  > letters VMS appear.   C This journal has been published for some time, and it came from the D HP-classic side of the organization, and so was of course originallyA targetted at pre-merger HP customers, who ran HP-UX, Windows, and  Linux.  D I exchanged e-mail with the editor, asking whether the publication'sC scope was to be expanded post-merger to cover the pre-merger Compaq F products, and found out that change was already well underway -- as ofF the next issue (October) the journal will begin to have Compaq-classicD content.  In a quick look at the draft of that edition, I spotted anC article on AlphaServers and the EV7 chip that mentions both OpenVMS B and Tru64, an article on some StorageWorks product or another, and* another article that talked about NonStop.  E A high-quality printed journal like this has lead-times of at least 2 ? 1/2 to 3 months between authors writing articles and the glossy B printed magazine hitting the streets, so such changes do take some= time.  It's only been about 3 1/2 months since the merger was D finalized, so it's obvious they have moved fairly quickly to embrace( the inclusion of Compaq-classic content.. ----------------------------------------------. Keith Parris | parris at encompasserve dot org   ------------------------------  # Date: Mon, 26 Aug 2002 22:39:28 GMT ( From: Don Sykes <annonymous@pacbell.net> Subject: Re: "inview" Article + Message-ID: <3D6AAED9.54018D42@pacbell.net>    Lee Y T Mah wrote: >  > Don, > Z > How reliable is this article?  How do I get at it?  We had our quarterly conference call[ > with HP this afternoon and I referred to the interview.  I offered to forward the article  > to the HP TAM. >  <snip>P I assume it's very reliable! The "inview" magazine has the HP logo on it and has? a 2002 Copyright on it. Here's the link to the on line version: ; http://www.hp.com/execcomm/inview/august02/aug02_epii4.html P As to how you get a paper copy, I don't know. I never asked for it. It just came in the mail last Friday.  Q And as was mentioned by John, there was one mention of VMS in the last question,  & Q "What's in store for HP customers... A. ... N "With the recent HP/Compaq merger, we will now phase the former Compaq productO lines, the Alpha, VMS and Non-Stop systems, into the overall plan to transition N to Itanium-based products. We will continue to support the commitments made toK Alpha and VMS customers as we converge these systems into a single, unified $ product and processor architecture."   Key words here are  6 "we will now phase the FORMER Compaq product lines"... and . "We will continue to support...AS WE CONVERGE"   --     Have VMS. Will Travel. Wire Paladin (@alphase.com) 
 San Francisco    ------------------------------  # Date: Mon, 26 Aug 2002 23:33:13 GMT * From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: "inview" Article B Message-ID: <ZUya9.226142$2p2.9268474@bin4.nnrp.aus1.giganews.com>  > "Keith Parris" <keithparris_NOSPAM@yahoo.com> wrote in message7 news:cf15391e.0208261456.74aa7b7e@posting.google.com...    ...   G > A high-quality printed journal like this has lead-times of at least 2 A > 1/2 to 3 months between authors writing articles and the glossy D > printed magazine hitting the streets, so such changes do take some? > time.  It's only been about 3 1/2 months since the merger was F > finalized, so it's obvious they have moved fairly quickly to embrace* > the inclusion of Compaq-classic content.  K Then again, it's been well almost 12 months since the merger was announced, L at which time Carly & Curly stated clearly that the merged corporation wouldI be ready to hit the ground running the day it was approved.  This looks a + lot more like a stumble than a 'run' to me.    - bill   ------------------------------  # Date: Tue, 27 Aug 2002 05:05:06 GMT ' From: Stuart Fuller <stufuller@usa.net> G Subject: Re: ??== Disk mount brackets 74-48510-01 for Alphastation 400. 0 Message-ID: <sfueka.5hr.ln@dadsys1.fuller.local>   <posted & mailed>t   Aus, Hans Magnus wrote:   J > Where can I obtain two (2) disk mount brackets (aka:  Bracket, Exchange)" > 74-48510-01 for an Alphastation? >   L If these are the mounting brackets to put a 3.5" disk in the 5.25" areas at L the top of the Alphastation, then these are standard off-the-shelf brackets @ like you'd find for a PC case.  Check your local computer store.           Stue   ------------------------------  # Date: Mon, 26 Aug 2002 19:58:38 GMTT9 From: Hein van den Heuvel <hein_netscape@eps.zko.dec.com> + Subject: Re: Automating EDIT/FDL ->OPTIMIZE / Message-ID: <3D6A8623.91FCEFEC@eps.zko.dec.com>   	 Joe wrotei  G > Supposing one had a very large number of indexed files (slightly mores2 > than 17,000) that one wanted to create a FDL for   :   ? > COPY-W-INCOMPAT, Incompatible Attributes Copying Indexed Filec  J Hi, I was just passing thru and woudl like to clarify several loose end in this discussioncA First of all, for copy to fail with that particular error needs aX$ relativly rare set of circumstances:7 - The indexed file has an area which is not being used. 9 - That area has a bucket size larger than any other area.v  I If you run into a bunch of file meeting that, then you may want to figuree
 out how those_, files came to be and fix that for ever more!   > H > then run through EDIT/FDL - OPTIMIZE in an "automated" sort of manner.  B The main tool there, is a pointed out, EDIT/FDL/NOINTER/ANAL=stats input/OUT=output  F I would strongly discourage the use of faked interactive scripts afterI setting EDF$$PLAYBACK_INPUT. It's not worth the hassle. Use your energiess better.p  E I would encourage you to consider that EDIT/FDL/NOINT takes the basicP filesoJ layout (number of keys and attributes) from the 'input' file and uses only
 3 inputs from < the stats file to come up with the target: Those inputs are:   FILE;e      CLUSTER_SIZE            12  ANALYSIS_OF_KEY 0;       MEAN_DATA_LENGTH        123!      DATA_RECORD_COUNT       1234a  I So for automation purposes you may want to just fake those params to have- 'reasonable'B values, from ANAL/RMS output, or past converts or whatever and go!  # > We're moving from 7.1-1H2 to 7.3.e  H A permanent fix to this problem was submitted in Jan-2002 and supposedlyE has hit 'the streets'. OpenVMS support can certainly provide details.i  G > On our production system the indexed files aren't really optimized toa  > drive cluser size (long story)  I IMHO Optimization to clsuter size are over-rated, and at best only shouldv only becomegH a deciding factor when otherwise seemingly equal choices present itself:H For example, if your goal is no more than 2 levels of index, and you can
 get that withrH a bucket size greater than 14, and the cluster size is 17, just pick 17.- If the cluster size had been 9, then pick 18.   H My personal preference is to pick a 'nice', small cluster size of say 12I with many factors (2,3,4,6) which will 'guide' EDIT/FDL to make the righte? choice. Otherwise it will pick too big a bucket size too often.n  ? > Anyone run into this and addressed it other than via BACKUP??l  C Backup can indeed be used as a workaround, but the 'wasted' area is  copied.6J Of course some may simpkly prefer backup as the tool to do the copy in the first place.  G If this was my system, and 'all' ( more than 100 out of 17,000) ) of mya	 files hadAI the problem then I'd certainly focus on how those file came to be and see  if that could be fixed.s  G If only a few files ( less that 100 on 17,000 ?) have a problem I mighte just want to deal with those on an exception basis.  7 Reproducer below, and program to detect 'unused' areas.i Cheers,n Hein  *  $ convert/fdl=sys$input sys$input tmp.idx     file; organization indexed;i     area 0; bucket_size 3;     area 1; bucket_size 4;     key 0; seg0_l 4;	      Exits     aap.     noot     mies	      Exita     $ copy/log tmp.idx tmp.tmpJ     %COPY-W-INCOMPAT, U$1:[HEIN]TMP.IDX;1 (input) and U$1:[HEIN]TMP.TMP;18       (output)      have incompatible attributes8     %COPY-E-WRITEERR, error writing U$1:[HEIN]TMP.TMP;18A     -RMS-F-IOP, operation invalid for file organization or deviceo?     %COPY-W-NOTCMPLT, U$1:[HEIN]TMP.IDX;1 not completely copied      $ back tmp.idx tmp.tmp     $ type tmp.tmp     aapt     mies     noot     $-   /*8 ** Show_roots.c         Hein van den Heuvel, August 2002 **I **      Template program to verify whether all areas in a file have usage6G **      It relies on the fact that AREAs are required to be declared ind order. */   #include <rms> #include <stdio> #define MAX 255s   main (int argc, char *argv[])i {r struct FAB      fab; struct XABSUM   sum; struct XABKEY   xab[MAX]; 5 int             i, stat, lvl, keys, areas, area[MAX];y4 int             strlen(), sys$display(), sys$open();   fab = cc$rms_fab;  sum = cc$rms_xabsum; fab.fab$b_shr = FAB$M_SHRPUT;  fab.fab$b_fac = FAB$M_GET; fab.fab$l_fna = argv[1];" fab.fab$b_fns = strlen( argv[1] ); fab.fab$l_xab = (void *) &sum; stat = sys$open ( &fab );  if (!(stat&1)) return stat;i. if (fab.fab$b_org!=FAB$C_IDX) return RMS$_ORG;   areas = sum.xab$b_noa;" for (i=0; i<areas; i++) area[i]=1;   keys = sum.xab$b_nok; ! fab.fab$l_xab = (void *) &xab[0];t for (i=0; i<keys; i++)     {n     /*(     ** Init Xab Key for each defined key.     ** Point previous to current except first.     */     xab[i] = cc$rms_xabkey;w     xab[i].xab$b_ref = i;s1     if (i) xab[i-1].xab$l_nxt = (void *) &xab[i];w     }  /*2 ** Ask RMS to fill in the XABs hooked off the FAB. */ stat = sys$display ( &fab ); if (!(stat&1)) return stat;l   for (i=0; i<keys; i++) {#         area[xab[i].xab$b_dan] = 0;n#         area[xab[i].xab$b_ian] = 0; #         area[xab[i].xab$b_lan] = 0; 	         }p   for (i=0; i<areas; i++) {:    if (area[i]) { D         printf ("-- Area %d, not in use for file %s\n", i, argv[1]);         stat = RMS$_AID;	         }0    } return stat; }    ------------------------------   Date: 26 Aug 2002 19:05:34 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)- Subject: Re: DE101 - can it go to Full Duplexl* Message-ID: <akdu5u$pfq$5@web1.cup.hp.com>  q In article <6e2f14f4.0208221415.11467356@posting.google.com>, glenmark@utxvms.cc.utexas.edu (Glen Martin) writes:cF :Assuming that I can get my hands on a couple of DE500-XA cards (whichC :is iffy), does anyone know if there is a patch that allows them tohA :function under VMS 6.2? (Last time I checked, they required 7.x)t    5   Um, doesn't *anyone* read the OpenVMS FAQ any more?     N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Tue, 27 Aug 2002 02:34:17 GMTb1 From: "David J. Dachtera" <djesys.nospam@fsi.net> - Subject: Re: DE101 - can it go to Full Duplexr' Message-ID: <3D6AEA54.6A75D85F@fsi.net>a   Hoff Hoffman wrote:e > s > In article <6e2f14f4.0208221415.11467356@posting.google.com>, glenmark@utxvms.cc.utexas.edu (Glen Martin) writes:gH > :Assuming that I can get my hands on a couple of DE500-XA cards (whichE > :is iffy), does anyone know if there is a patch that allows them toiC > :function under VMS 6.2? (Last time I checked, they required 7.x)H > 7 >   Um, doesn't *anyone* read the OpenVMS FAQ any more?    For what it may be worth...i  C While it's a gold mine of priceless bits, the FAQ is getting ratherfE unwieldy. It's more than I can finish on my just-over-an-hour morning  train ride to work.-   -- e David J. Dachtera  dba DJE Systems? http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/e   ------------------------------   Date: 26 Aug 2002 17:41:38 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)R Subject: Re: DECC (specifically v6.2-008, but older maybe acceptable) docs ONLINE?* Message-ID: <akdp8i$pfq$1@web1.cup.hp.com>  l In article <ak2ppg$19e0@r02n01.cac.psu.edu>, jlw@psulias.psu.edu (j.lance wilkinson, (814) 865-1818) writes:D :Discovered that in their infinite wisdom, nobody bothered to secureD :DOCUMENTATION for DECC here.  While we've got the v6.2-008 compilerK :on OpenVMS v7.2-1h1, and v5.9-011 on Tru64 v4f(?), the only documentation XG :I can find still has the "d|i|g|i|t|a|l" logo prominently appearing on ! :the cover and is dated May 1994.e  K   Please read the OpenVMS FAQ; the FAQ has pointers to the C documentation.   N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Mon, 26 Aug 2002 22:27:34 GMTr( From: "C.W.Holeman II" <cwhii5@ACM5.org>" Subject: K12 (Was Roadshow update)C Message-ID: <qXxa9.1040$%D6.98644@newsread1.prod.itd.earthlink.net>    Sue Skonetski wrote:  > > I was not correct in my defination of what Public Sector is. > ; > Just a clarification Public Sector includes the followingt= > Federal, State & Local, Higher Education, and K12 Educationd  " How is VMS doing in the K12 world?   -- C.W.Holeman II< cwhii5@Julian5Locals.com                http://also.as/cwhii remove the fives# Send spam to junkmail@earthlink.neto   ------------------------------    Date: 26 Aug 2002 11:49:18 -0700! From: soterro@yahoo.com (Soterro) ' Subject: Re: Low-level format SCSI disk.< Message-ID: <d5440555.0208261049.ad138c9@posting.google.com>   Hello,  A I was in an exteeeended vacation so it is only today that I triednD again the thing. Thank you all for your answers, but I'm still stuckE after 8 hrs of reading man pages and web answers and trying things...o  B 1. Linux. I won't install Linux just for the formatting of a disk,E please understand me... I just ran the floppies to get hopefully somes hardware infos.n  + 2. Tru64. That scu is at least, hum, weird.vC     - the /dev/rdisk is not existing, so I had to figure out myselfe@ what rzxxx drive is my new disk drive (I read by chance also the= SCSI(7) man page), and how to use it. But scu is asking for ar character-mode device:   # scu -f rz18a4 scu: Device 'rz18a' is NOT a character device entry.  B so I have to use rrz18a (BTW, I have NO idea why I have to use the< partition 'a' for a device which is having NONE as it is notF formatted). Anyway, then the format command wants a block-mode device:   # scu -f rrz18at scu> format0C scu: 'rrz18a' is not a block device, EINVAL (22) - Invalid argumenth  = If I use the set device command in scu, the file system mountgC verification fails when trying to format (yeah of course, how could 6 one succeed in testing a non-existent file system :P).C Do you have any idea how is this program actually working? Have youi& ever actually used it and if yes, how?  > 3. OpenVMS. Heh, I'm waiting for any ideas on Tru64 but in the/ meanwhile I'll try the OpenVMS ideas. Tomorrow.l  ? I might be trying stupid things, but I'm not in the position of C studying the SCSI architecture, I just want to format a HDD and ther forget it...  
 Thank you, Sorinf   ------------------------------  % Date: Mon, 26 Aug 2002 20:05:31 +0100b/ From: "Adam Price" <adam+usenet@pappnase.co.uk>e' Subject: Re: Low-level format SCSI disko6 Message-ID: <akdu5s$1hp6ga$1@ID-138239.news.dfncis.de>   Soterro wrote: > Hello, >tC > I was in an exteeeended vacation so it is only today that I tried F > again the thing. Thank you all for your answers, but I'm still stuckG > after 8 hrs of reading man pages and web answers and trying things...t >lD > 1. Linux. I won't install Linux just for the formatting of a disk,G > please understand me... I just ran the floppies to get hopefully somer > hardware infos.m >e- > 2. Tru64. That scu is at least, hum, weird.eE >     - the /dev/rdisk is not existing, so I had to figure out myselfeB > what rzxxx drive is my new disk drive (I read by chance also the? > SCSI(7) man page), and how to use it. But scu is asking for a  > character-mode device:  H OK That just means you are using a version of Tru64 older than V5 and so$ you are using the old device naming.E You can either do as you have done and user /dev/rrz* or you can workp  in a slightly different fashion.E You know which scsi bus you device is connected to, you know the scsig target id and you know the lun. / Start scu without pointing it at a file at all.e Then inside scu, type  set nexus bus B target T lun L3 Where BT and L are bus target and Lun respectively.)G The in program help of scu is quite good, read the help text for format  and it should give details.  HTHt Adam   ------------------------------  % Date: Mon, 26 Aug 2002 20:07:20 +0200 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>a" Subject: Re: OpenVMS + Joint STARS' Message-ID: <3D6A6E58.7EEBAC6C@aaa.com>n  ; AGA ("Acetylen Gas Aktiebolaget"), a company founded by thew7 Swedish inventor Gustaf Daln, famous for the automateda7 lighthouse equipment used around the world. www.aga.se.i   :-)'   Jan-Erik Sderholm.    Bob Koehler wrote: >  > J > Device AGA0:, device type AG 12403, is online, error logging is enabled. >s   ------------------------------   Date: 26 Aug 02 22:51:40 +0200) From: p_sture@elias.decus.ch (Paul Sture)a" Subject: Re: OpenVMS + Joint STARS) Message-ID: <dIe+VLhPta3z@elias.decus.ch>9  c In article <3D6A6E58.7EEBAC6C@aaa.com>, Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> writes:D= > AGA ("Acetylen Gas Aktiebolaget"), a company founded by the 9 > Swedish inventor Gustaf Daln, famous for the automatedd9 > lighthouse equipment used around the world. www.aga.se.d >  > :-)  > N Aga is also a brand of oven. These things boil kettles in a matter of seconds.  % http://www.aga-ranges.com/welcome.asp    :-)e   > Jan-Erik Sderholm.d >  > Bob Koehler wrote: >> e >> tK >> Device AGA0:, device type AG 12403, is online, error logging is enabled.a >> -- 4 __
 Paul Sture Switzerlanda   ------------------------------   Date: 26 Aug 02 22:36:14 +0200) From: p_sture@elias.decus.ch (Paul Sture) 7 Subject: Re: OT Mac OS X woes, but advice needed please ) Message-ID: <FhAfMFE3NbWW@elias.decus.ch>.  c In article <osQa97U5fOV8@eisner.encompasserve.org>, koehler@encompasserve.org (Bob Koehler) writes:aW > In article <NTS8+u3dG8cL@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes:ti >> Totally off-topic I realise, but I know there are VMS pals out there who have experience of Mac stuff.s > J >    Yeah, but could you get your posting down to less than 80 characters? >e  F What newsreader are you using? It appeared perfectly on Netscape 3.03.  7 I am using ANUNEWS and haven't seen the problem before.   8 Do all my posts show like that, or was it just that one?  - > ? >    I've has no such problem with OS X.  Get into the terminal F >    application (UNIX shell window) and use ls -l to determine if theH >    file system thinks the file exists, and how it's protection is set. > C >> Should I simply abandon OS X and go back to OS 9? - at present, 3 >> highly favoured.7 >  >    We're not.. > K >> What are your best recommendations on OS X backup strategies? (presentlys+ >> ftp to my Alpha before I lose work, thenn >> off to tape)e > G >    Right now I'm using NFS or an automated FTP to copy critical fileseM >    weekly to a VMS server which has an automated tape backup running.  Less,J >    often I backup the while thing to Jazz by dragging the hard disk onto >    the Jazz drive icon.  >   N Thanks for that. I see that Jaguar is now out, and receiving positive comments@ about speed and stability. I guess that's my best route forward.  d __
 Paul Sture Switzerlandr   ------------------------------  # Date: Mon, 26 Aug 2002 20:29:29 GMTh# From: "John Smith" <a@nonymous.com>-, Subject: Re: PGP assets (mostly) sold by NAII Message-ID: <Jcwa9.17477$6m61.16622@news04.bloor.is.net.cable.rogers.com>?  F "Brian Tillman" <tillman_brian@notnoone.notnohow.com> wrote in message news:3d6a6aec$1@news.si.com... > >Of interest to some here: > >c2 > >http://www.pgp.com/display.php?pageID=51#anch22 > > K > >Now if they can only be convinced to produce a fully supported 'businesse. > >server' version for VMS since NAI wouldn't. >MI > Except that PGP did not reacquire McAfee E-Business Server (once calledf PGPtI > Command Line), which is where I suspect PGP for VMS would fall.  Just ai > guess, though.    J True. But maybe the deal only specified the Business Server o/s'es that itJ currently runs on as being 'owned' by NAI. There might be some wiggle room+ for a VMS version from PGP rather than NAI.O  L But that said/speculated, there still won't be any VMS version unless eitherJ company sees VMS demand coming their way, and funnily enough, demand comesD from marketing VMS to customers who might need a robust platform forI e-commerce. Isn't that where HP said they would expend what little efforti they intended for VMS?   ------------------------------  % Date: Mon, 26 Aug 2002 14:02:36 -0400i; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>r' Subject: Re: Postscript printing on vmsr$ Message-ID: <3d6a6d76$1@news.si.com>  L >I have an Alpha station 200 with HP printer XL300 connected to the parallel port	 >(lra0:). I >Because dcps can'nt work with this parallel port is there another way too printm >postscript files?  J Perhaps find a DECserver 250.  $450 from eli.com, $250 from tcdigital.com,E $516 from www-legacy.digital.com.  Others also found in a web search.  --A Brian Tillman                   Internet: tillman_brian at si.com A Smiths Aerospace                          tillman at swdev.si.coml= 3290 Patterson Ave. SE, MS      Addresses modified to prevent < Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Mon, 26 Aug 2002 14:49:43 -0400t0 From: "Hank Vander Waal" <hvanderw@novagate.com>' Subject: RE: Postscript printing on vmss; Message-ID: <000901c24d31$57fd0090$cd96a8c6@manufact5l8vs8>,  > have a couple of those available if you are interested !!! : ) 200/300/90L+s too !!+ Contact me if you or anyone else interestedc! hvanderw-removethis-@novagate.comh Hank Vander Waal  / VMS - is there anything that DARES to compare??c! Unix  - for hackers - by hackers!N You get what you pay for !     -----Original Message-----@ From: Brian Tillman [mailto:tillman_brian@notnoone.notnohow.com]% Sent: Monday, August 26, 2002 2:03 PMm To: Info-VAX@Mvb.Saic.Comr' Subject: Re: Postscript printing on vms     L >I have an Alpha station 200 with HP printer XL300 connected to the parallel port	 >(lra0:).hI >Because dcps can'nt work with this parallel port is there another way toa printi >postscript files?  J Perhaps find a DECserver 250.  $450 from eli.com, $250 from tcdigital.com,E $516 from www-legacy.digital.com.  Others also found in a web search.H --A Brian Tillman                   Internet: tillman_brian at si.commA Smiths Aerospace                          tillman at swdev.si.comO= 3290 Patterson Ave. SE, MS      Addresses modified to preventt< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------   Date: 26 Aug 02 22:27:53 +0200) From: p_sture@elias.decus.ch (Paul Sture)<' Subject: Re: Postscript printing on vms:) Message-ID: <KNrivWKAaO6U@elias.decus.ch>a  i In article <3bbfbaa2.0208260038.50c371d9@posting.google.com>, meidanze@hotmail.com (meidan zemer) writes:s > Hello,R > I have an Alpha station 200 with HP printer XL300 connected to the parallel port
 > (lra0:).P > Because dcps can'nt work with this parallel port is there another way to print > postscript files?l > Thanks  I Does that model support a network card? That is the only way I can see ofoK getting my home system to talk directly to the printer (at the moment doing  the printing from Linux).  __
 Paul Sture Switzerlands   ------------------------------  % Date: Mon, 26 Aug 2002 14:15:09 -0700v+ From: "xenman" <xenman@sprynet.nospaam.com>a' Subject: Re: Postscript printing on vms 2 Message-ID: <ake5p6$ov5$1@slb2.atl.mindspring.net>  4 meidan zemer <meidanze@hotmail.com> wrote in message7 news:3bbfbaa2.0208260038.50c371d9@posting.google.com...  > Hello,D > I have an Alpha station 200 with HP printer XL300 connected to the
 parallel portw
 > (lra0:).J > Because dcps can'nt work with this parallel port is there another way to print  > postscript files?o > Thanks  9 Have you looked at ScriptServer from GrayMatter Software?a7 I know that it prints to any postscript printer on VMS.c   (I do not work for GrayMatter).u   ------------------------------  % Date: Mon, 26 Aug 2002 22:42:19 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> ' Subject: Re: Postscript printing on vms , Message-ID: <3D6AE707.A6BE65D4@videotron.ca>    re: printing via parrallel port.  L Coudln't one simply tweak the .COM files to recognize a parralel device nameJ as a serial device name, ignore any SET TERMINAL commands, and then let itN start the queue and assign a channel to that device ? Wouldn't the program not? really see a difference as long as the port is bi-directional ?e  L Or would a IO$SETMODE fail (tryiong to set terminal porrt features) and DCPS then stop right there ?-   ------------------------------  # Date: Mon, 26 Aug 2002 17:20:27 GMTS( From: Don Sykes <annonymous@pacbell.net>/ Subject: Re: Problem with JDK1.3 on OpenVMS 7.3n+ Message-ID: <3D6A6413.8EC20898@pacbell.net>k   Yong Liu wrote:6 >  > Hi,a > K > I got a problem when I was trying to run Java on my Alpha server. This ish > what happened: > (1) @java$1.3_setupu- > (2) define class$path 'My java lib path',[]e > (3) javac Test.java. > out of memery. exit  > & > What's happening? How do you fix it?  O Java imposes its own limitations. I'm only running 1.1.8 on my VMS machine, but-F there are options that allow you to specify stack and heap size - e.g:F     -ss<number>       set the maximum native stack size for any threadD     -oss<number>      set the maximum Java stack size for any thread4     -ms<number>       set the initial Java heap size4     -mx<number>       set the maximum Java heap size> I'd try working with those before altering VMS page file size. -- u   Have VMS. Will Travel. Wire Paladin (@alphase.com).
 San Franciscon   ------------------------------  % Date: Mon, 26 Aug 2002 22:32:08 -0400c- From: JF Mezei <jfmezei.spamnot@videotron.ca>t7 Subject: Re: Procedure to enlarge system name required.e, Message-ID: <3D6AE4A5.877AE643@videotron.ca>   Johno wrote: > b > JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3D605225.19D94A1@videotron.ca>... > > Johno wrote:L > > > What I had hoped was that the output on the screen would be duplicatedH > > > on the la70 console printer but this solution does not produce the > > > desired result.s > >rH > If you can find the escape sequence for setting and resetting the la70D > to the suggested character spacings (5 and 10) I will be grateful.  ? This is for the LA75, but i *think* it should work on the LA70.a   To set to 5 cpi:  CSI 5 w  	6 cpi:	CSI 6 w3 	8 cpi:	CSI 8 wr 	8.55cpi 	CSI 1 2 wn  : All of the above generate double width horizontal pitches.  2 Default pitch: CSI w	or CSI 0 w		( usually 10 cpi)   10 cpi:	CSI 1 w> 12 cpi:	CSI 2 w  16.5 cpi:  CSI 4 w 17.1 cpi:	CSI 11 w  I CSI is  either the 8 bit CSI character or the 2 character sequence ESC  [a   ------------------------------  % Date: Tue, 27 Aug 2002 08:59:05 +0930o/ From: Mark Daniel <Mark.Daniel@wasd.vsm.com.au>z= Subject: Process Software Multinet NFS SRI file name encoding 3 Message-ID: <akedue$cbf$1@fang.dsto.defence.gov.au>   J Can anyone clarify whether Multinet's NFS SRI file name encoding algorithm  L    http://www.multinet.process.com/ftp/docs/html/admin_guide/Ch27.htm#E13E57  L changes case (using the leading '$' symbol before an alphabetic) on a whole / of path basis, e.g. the *nix file specificationo      /mnt/vms/this/is/A/TEst  ( would be represented as (something like)      DISK_NFS:[THIS.IS.$A]TE$ST.  F -or- (which seems more reasonable to me) on a per-path-component basis      DISK_NFS:[THIS.IS.$A]$TE$ST.)   or perhaps evens       DISK_NFS:[THIS.IS.$A$]$TE$ST.  9 I do not have MultiNet to experiment with. As always TIA.r   ------------------------------   Date: 26 Aug 2002 18:13:12 GMT= From: jlw@psulias.psu.edu (j.lance wilkinson, (814) 865-1818)eY Subject: Re: putenv()/setenv() was - Re: DECC (specifically v6.2-008, but older maybe acc-, Message-ID: <akdr3o$115k@r02n01.cac.psu.edu>  k In article <3D6542D6.50007@dskwld.zko.dec.compaq>, "John.Malmberg" <Malmberg@dskwld.zko.dec.compaq> writes:T) ~j.lance wilkinson, (814) 865-1818 wrote:z7 ~> In article <3D651A20.6010409@dskwld.zko.dec.compaq>,l; ~> "John.Malmberg" <Malmberg@dskwld.zko.dec.compaq> writes:t, ~> ~j.lance wilkinson, (814) 865-1818 wrote: ~> ~ >A ~> ~ > While we've got the v6.2-008 compiler on OpenVMS v7.2-1h1, # ~> ~ > and v5.9-011 on Tru64 v4f(?)l ~> ~C ~> ~Any C RTL ECOs installed could also be very significant in thisl
 ~>  ~area. ~> uB ~> No idea what may have been installed.  Finding that out, unlessB ~> there's some procedure that I can invoke to list them, is about ~t ~Try:a ~rE ~$DIRECTORY SYS$HELP:*.RELEASE* !for a list of all release notes that A ~have been installed on your system.  Note that it is possible tofC ~install the release notes with out actually installing the rest ofi	 ~the kit.a ~" ~> ~D ~> ~On OpenVMS, after a spawn, vfork(), system() or other event thatH ~> ~creates a child process, the parent process will not see any changes9 ~> ~in the environment variables made by a child process.I ~> oC ~> I know that -- unless one TELLS the LIB$SPAWN routine to include  ~> the symbol tableE ~/B ~No, what I mean is after the child is born, the enviroment tablesB ~are forever separate.  On UNIX there may be some sharing.  If you8 ~have not been able to tell, I am not a UNIX programmer. ~eC ~This type of sharing needs to use the LNM$JOB table, which is onlyA! ~readable to getenv() on OpenVMS.D ~U4 ~> -- but since putenv() on Tru64 is BEING USED HEREA ~> to affect the parent environment of the program that calls it,A@ ~> I'm assuming the intended affect of putenv() is to affect theD ~> outside/ parent environment, which would be create a LOCAL symbolD ~> for VMS by default, and that would work, or a GLOBAL (which would0 ~> also be known to a subsequent child process). ~tB ~Both local and global DCL symbols existing at birth are copied to ~the child process.) ~ I ~> ~>  While getenv() for OpenVMS seems to clearly be able to access both M ~> ~>  DCL-presented SYMBOLS (LOCAL and GLOBAL) and LOGICAL NAMES in the sameiL ~> ~>  way that getenv() for Unix accesses Shell-based ENVIRONMENT VARIABLESJ ~> ~>  and make their values accessible to a running program, and putenv()I ~> ~>  for Unix [Tru64] clearly is able to allow a running program to putoM ~> ~>  a variable out for later use in the same shell script that invoked thetJ ~> ~>  program, I can't seem to find any of the environment variables thatL ~> ~>  my OpenVMS-based putenv() calls are creating after the program exits. ~> ~D ~> ~It appears that the environment array that setenv() and putenv()4 ~> ~modifies is local to the application in OpenVMS. ~> ~J ~> ~>  -Even these docs sited by Craig here fail to address this question.M ~> ~>   IS IT EVEN SUPPOSED TO WORK?  Or should I rewrite that chunk of code,i ~> ~>   to have: ~> ~G ~> ~You could put #ifdef __VMS conditionals in your code as you stated.l ~> ~K ~> ~I would recommend creating a separate module named dcl_putenv that doesiI ~> ~the LIB$SET_SYMBOL instead, and takes the same arguments as putenv().l ~> ~G ~> ~Then for the OpenVMS build of the modules that call putenv() add a sI ~> ~/DEFINE=("putenv=dcl_putenv") to the CC line.  Then the OpenVMS buildk ~> ~will have a extra module.s ~> ~J ~> ~I can not find putenv() in "The Open Group" Single UNIX specification,K ~> ~but it is documented in the current Compaq C Run-time Library manual ini" ~> ~the URL you previously quoted. ~oE ~Correction, I made a transposition error.  putenv() is documented inr> ~"The Open Group" Single UNIX specification.  setenv() is not. ~i- ~setenv() appears to be easier to use though.T ~$A ~> Yeah, but you've probably noted that it says nothing about howpE ~> one could then retrieve such an environment variable after puttingoF ~> it.  And presumably one would use putenv() to affect an environment@ ~> shared by other programs, not as an alternative to a variableC ~> shared by other routines in the same program.  So where the heckr ~> does putenv() stash it? ~c ~It is documented on OpenVMS:  ~aF ~http://www.openvms.compaq.com/commercial/c/5763p041.html#index_x_1218 ~;6 ~     "The putenv function uses the malloc function to ~      enlarge the environment.  ~ 2 ~      A potential error is to call putenv with an4 ~      automatic variable as the argument, then exit5 ~     the calling function while string is still partn ~     of the environment." ~tA ~So putenv just stashes a pointer to the string passed in program  ~locally allocated memory. ~ @ ~It probably might be desirable to have putenv()/setenv() update: ~the local symbol table, I can pass that suggestion along. ~m> ~A search of the comp.lang.c frequently asked questions turned	 ~up this:  ~n ~16.6:  @ ~How can a process change an environment variable in its caller? ~r ~A: > ~In general, it cannot.  Different operating systems implement= ~	name/value functionality similar to the UNIX environment in]< ~	different ways.  Whether the "environment" can be usefully: ~	altered by a running program, and if so, how, is system- ~	dependent. ~i< ~	Under UNIX, a process can modify its own environment (someA ~	systems provide setenv() and/or putenv() functions to do this),0@ ~	and the modified environment is usually passed on to any child: ~	processes, but it is _not_ propagated back to the parent
 ~	process. ~ > ~As I understand UNIX, the shell executes a program as a child= ~process and waits for it to end, unless the "&" character ism' ~used to make it run in the background.f ~)> ~This FAQ entry indicates that a value set with putenv() under< ~UNIX also may not be visible to the shell after the program ~exits.6 ~sB ~But I do not have direct experience with doing this type of thing ~under UNIX. ~tD ~> Before giving up and going the OpenVMS/Tru64 specific routes withC ~> #ifdef/#else/#endif, I'm going to try to see if setenv() has anynE ~> better impact. It certainly doesn't have any better DOCUMENTATION.' ~a@ ~I would expect that setenv() and putenv() to use the same local ~memory and behave the same. ~-@ ~The documentation for putenv() in the Compaq C Run time library7 ~contains the same information as in the "Official UNIX-> ~Specification" maintained by the Open Group, and a little bit! ~more for site specific behavior.i ~r6 ~You can read the official specification your self at: ~-= ~http://www.opengroup.org/  Allows online browsing after free  ~registration. ~o ~m: ~So it appears that the OpenVMS implementation putenv() is: ~compliant with the official standard, even if it does not ~do what you want it to do.4 ~0C ~I can certainly understand that if putenv() works the way that you ? ~would like to under UNIX, that under OpenVMS the same behaviorc; ~would be desired.  Again, I can pass the suggestion along.  ~t ~-John" ~malmberg@dskwld.zko.dec.compaq.hp ~Personal Opinion Only    ? 	OK.  Further discussion with the Tru64/*Oracle* programmer whor? 	showed me her code containing calls to putenv() have clarified ? 	that the sole purpose of her use of putenv() was NOT to affecte@ 	the parent program's environment, but to extend the environment? 	with oracle parameters before making system() calls to ORACLE J? 	scripts.  So her demonstrated examples of putenv() working theMC 	way I thought it should clearly were misunderstood on my part, and-B 	the Tru64 version of putenv() (and, likely, setenv()) are workingA 	per the FAQ cited here and the observations of the same routines0C 	in the OpenVMS CRTL.  My suspicion is that were I inclined to testnB 	the behavior and impact these routines have in OpenVMS on spawnedB 	subprocesses (for which I have no need at present), I'd find they) 	have the same functional effect as well.e  @ 	So....  in the famous words of Emily Latella, "Never Mind...".	  P +----"Never Underestimate the bandwidth of a station wagon full of mag tapes"--+B | J.Lance Wilkinson ("Lance")		InterNet:  Lance.Wilkinson@psu.edu < | Systems Design Specialist - Lead	AT&T:      (814) 865-18189 | Digital Library Technologies		FAX:       (814) 863-3560e2 | 3 Paterno Library				"I'd rather be dancing..." B | Penn State University		    A host is a host from coast to coast,K | University Park, PA 16802	    And no one will talk to a host that's closevC | <postmaster@psulias.psu.edu>	    Unless the host that isn't closeu: | EMail Professional since 1978	    Is busy, hung or dead.P +---------"He's dead, Jim. I'll get his tricorder. You take his wallet."-------+9                 [apologies to DeForest Kelley, 1920-1999]-3 <A Href="http://perdita.lcs.psu.edu">home page</a> 2J <a Href="http://perdita.lcs.psu.edu/junkdec.htm">junk mail declaration</a> --	/"\ 	\ /	ASCII RIBBON CAMPAIGN 	 X	AGAINST HTML MAIL3 	/ \   ------------------------------  % Date: Mon, 26 Aug 2002 13:26:54 -0400e; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>w) Subject: Quantum DLT 4000 and OpenVMS VAXr$ Message-ID: <3d6a6518$1@news.si.com>  K Over the weekend I installed an HSD30 on a cluster to that I could add some5J drives.  I also found a Compaq-branded Quantum DLT4000 that I hooked up toI the HSD.  It works, but is identified as a TZ01 within VMS.  Compaq's webrK site has nothing when I search for "TZ01", "TK88", or "DLT 4000".  I'd like  info on what a "TZ01" is.i  K How do I control the density when INITing or MOUNTing tapes.  HELP INIT and J HELP MOUNT both show various /DENSITY=TK8x as being allowed, but (althoughL they don't say it) apparently only on Alphas.  I tried /DENSITY=6250 and theJ "20G" light came on, but I'd like to be able to choose the lower densitiesH as well, both with and without compaction enabled (via the /MEDIA_FORMAT qualifier).b --A Brian Tillman                   Internet: tillman_brian at si.comoA Smiths Aerospace                          tillman at swdev.si.coma= 3290 Patterson Ave. SE, MS      Addresses modified to prevent?< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------   Date: 26 Aug 2002 20:12:14 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)- Subject: Re: Quantum DLT 4000 and OpenVMS VAX * Message-ID: <ake22u$sk2$1@web1.cup.hp.com>  b In article <3d6a6518$1@news.si.com>, "Brian Tillman" <tillman_brian@notnoone.notnohow.com> writes:L :Over the weekend I installed an HSD30 on a cluster to that I could add someK :drives.  I also found a Compaq-branded Quantum DLT4000 that I hooked up tomJ :the HSD.  It works, but is identified as a TZ01 within VMS.  Compaq's webL :site has nothing when I search for "TZ01", "TK88", or "DLT 4000".  I'd like :info on what a "TZ01" is.  H   That's what the tape identified itself as to OpenVMS.  The widget veryG   likely does not have the TZ88 firmware; it is (of course) not a TZ88.'  L :How do I control the density when INITing or MOUNTing tapes.  HELP INIT andK :HELP MOUNT both show various /DENSITY=TK8x as being allowed, but (althoughPM :they don't say it) apparently only on Alphas.  I tried /DENSITY=6250 and thetK :"20G" light came on, but I'd like to be able to choose the lower densities(I :as well, both with and without compaction enabled (via the /MEDIA_FORMATn :qualifier).  7   6250 is a 9-track magtape density keyword, of course.n  0   The Quantum DLT4000 is the basis for the TZ88.  D   Most density selection keywords apply to OpenVMS Alpha only -- theB   OpenVMS help files were updated to relect this for OpenVMS V7.3.  A   Off-hand, I do not know if there is a way to change the defaultg'   settings of the DLT4000 series drive.n  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 26 Aug 2002 18:49:10 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: rename c rtl bugt* Message-ID: <akdt76$pfq$3@web1.cup.hp.com>  l In article <81c810ad.0208220751.7b7d580@posting.google.com>, rdls22@jnatx.com (Stephen L. De Rudder) writes:D :I found the following bug (feature) in the rename c rtl.  Here is aB :small smipit of code that produces the bug.  Yes I realize that I: :could fix the code but my preference is to get VMS fixed.  )   Then please contact the support center.o   ..@ :Notes: OpenVMS Alpha V7.2, DEC C V6.0-001 on OpenVMS Alpha V7.2  K   Expect to be asked to move to a supported configuration with current ECOsrH   applied, as a start.  (Though I'd expect that will not be of immediate;   use here, this sequence will fix various known problems.)h  J   There are known inconsistencies with handling the leading-dot UNIX-styleH   file notation, and ANSI C largely leaves the details of file naming upH   to the host operating system.  In this case, the OpenVMS rename() callG   does not ancipate a UNIX-style second argument, particularly with the F   leading dot and leading double-dot notations -- the OpenVMS rename()=   call is trying to provide the expected filename stickiness.o  E   There has been work in this area in later releases, and there are amF   whole pile of logical names around that control the behaviour of theF   C run-time library.  (The current ECO kit for the C RTL will get youC   at least some of this work, but you'll definitely want to move toaB   V7.2-2 or (better) to V7.3-1 -- the latter releases is the firstF   "mainline" OpenVMS release that has various of the DII COE changes.)    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 26 Aug 2002 13:56:21 -0400c> From: "Michael Raspuzzi" <raspuzzi#spamblaster.emesen-dot-com> Subject: Re: Reverse LAT+ Message-ID: <ev8zHnSTCHA.2208@cpimsnntpa03>:  I If you create a service on the DS200 (password protected is recommended),6J you could then connect to the console through the DECserver port using SETK HOST /LAT.  When I worked at Digital, I used to do that for my debugging so:: that I would not have to run the console manager software.  J It has been quite a few years so I do not have exact syntax but I can give, you the general idea of what you need to do.  H - Decide which DECserver port will be connected to the VAX/Alpha console line.sJ - Create a service on the DECserver that will be advertised to the networkH (something like SYSTEM_NAME_CONSOLE where "SYSTEM_NAME" might be easy to
 remember)./ - Supply a password for the SYSTEM_NAME_CONSOLE ) - Set up the group codes for the service.h0 - Associate the DECserver port with the service.  J On OpenVMS systems that will be connecting to the console (I recommend the= latest and greatest OpenVMS software for this functionality):t  K - Set outgoing group codes to match the DECserver's service group codes fore SYSTEM_NAME_CONSOLE: - Enable outgoing connections   F Use SET HOST /LAT SYSTEM_NAME_CONSOLE and you should be prompted for aI password.  Supply the correct password and you should now be connected to  the console of the computer.   On other DECservers:  I - Enable group codes so you can connect locally to the remote DECserver'sm service.  J Use CONNECT SYSTEM_NAME_CONSOLE and you should be prompted for a password.  I Warning - older OpenVMS systems can use some form of SET HOST /DTE to get/H around the password on the DECserver port.  This is bad as it means thatK someone with an old system on your network now has access to the computer'saJ console port.  The Console Management software used to always be connectedG to the port so that helped keep undesirables out if they knew where thetK server and port were located.  Now, you would be advertising the service to - the LAN and you know how curiosity can get...e  D Check DECserver documentation and OpenVMS documentation for details.   Mike Former OpenVMS LAT developer        ? "Lucas, Edward A (SAIC)" <Edward.Lucas@bp.com> wrote in messageD9 news:EF1DC894691AD5118AF000508BB85FDE034CC7E7@AMCLVX11...e > Hello everyone >pI > I need to replace my current "CA Console manager SW"  The SW enables me6 (andJ > my team) to access all system like being on a console but remotely (i.e.I > >>>).  My plan is to replace the SW, plug (I think) a serial cable from> theW/ > TS (DS200) to the console port on the server.7 >1G > Can someone please point me in the correct direction. I need to learnaL > everything about this and then configure. I need to get this working ASAP. >a > Thank you in advance.i >n >h- > The over worked, under paid,  unappreciatedI   ------------------------------  % Date: Mon, 26 Aug 2002 13:51:57 -0400R6 From: "John.Malmberg" <Malmberg@dskwld.zko.dec.compaq> Subject: Re: Reverse LAT2 Message-ID: <3D6A6ABD.50205@dskwld.zko.dec.compaq>   Lucas, Edward A (SAIC) wrote:l > Hello everyone > N > I need to replace my current "CA Console manager SW"  The SW enables me (andJ > my team) to access all system like being on a console but remotely (i.e. > L >>>>).  My plan is to replace the SW, plug (I think) a serial cable from the >>>-1 > TS (DS200) to the console port on the server.    > G > Can someone please point me in the correct direction. I need to learneL > everything about this and then configure. I need to get this working ASAP.  G If you want to use CA Console Manager with out using the console ports  F connected through a terminal server, this is documented in one of the , manuals.  Or is this what you are doing now?  A When the product was named Polycenter Console manager, IIRC this o, information was in the Online Documentation.  I Also check the release notes, and there is example code in VCS$EXAMPLES: e% for using LAT instead of a DecServer.A  F You will no longer be able to issue console commands while OpenVMS is 5 down, but all other functionallity will be available.     F If you just want to stop using the Console Manager, but still use the C terminal server, then all you need to do is bind a LAT port to the rF DecServer port and SET HOST/LAT to it from any host that supports LAT.  H Some one will probably post the specific LAT commands if you need them, F or you can look up LAT in the OpenVMS Master Index to find where they  are documented.g  F If you use a newer terminal server that supports TCP/IP, then you can  telnet to the console ports.  H Please note the obvious security risks of having the console accessable I to the network, and take what ever steps are needed for your environment.   E It also may be advisable to disable the console from halting when it tE gets a break signal.  Such signal could be generated spuriously by a u reboot of the terminal server.       -Johni malmberg@dskwld.zko.dec.compaq Personal Opinion OnlyC   ------------------------------  % Date: Mon, 26 Aug 2002 21:13:39 +0200R" From: "Hans Vlems" <hvlems@iae.nl> Subject: Re: Reverse LAT6 Message-ID: <akdul7$1hefbr$1@ID-143435.news.dfncis.de>  A "Lucas, Edward A (SAIC)" <Edward.Lucas@bp.com> schreef in berichtr9 news:EF1DC894691AD5118AF000508BB85FDE034CC7E7@AMCLVX11...B > Hello everyone > I > I need to replace my current "CA Console manager SW"  The SW enables meh (andJ > my team) to access all system like being on a console but remotely (i.e.I > >>>).  My plan is to replace the SW, plug (I think) a serial cable from  theo/ > TS (DS200) to the console port on the server.p >fG > Can someone please point me in the correct direction. I need to learnrL > everything about this and then configure. I need to get this working ASAP. >pL The picture I get from your description is that you're (for some reason) are moveing away from)J a CA product and replace it with what is basically a hardware alternative.  F If that is the situation then this is probably what you're asking for:  H Connect the serial console of the VAX or Alpha to a terminal server that supports reverse LAT.(G Not all models do, most notably the DS100 and DS5x0 and (IIRC) the DS90a don't have that facility.eB The DS200, DS300, all DS700's and all DS900's do have reverse LAT.J If you have more than 8 devices, a DS700 or an old DS300 are good choices, especially because theJ remote client can also use IP, TELNET, to connect to the DECserver. Useful for those systems  that have no LAT.v  L The HELP command on a DS300 returns quite good information. You CREATE a LAT service on a port.H That service is advertised on the LAN and any connection to that service, will connect directly to that physical port.J A password may be added to keep less knowledgable users away from an OPA0: port; typing ^P will haltm a VAX.I Furthermore, have a look at the SET PORT BREAK command, you might want top use a different character than the DECserver default.  " This is what you were looking for?   ------------------------------   Date: 26 Aug 2002 19:49:27 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: Reverse LAT* Message-ID: <ake0o7$pfq$8@web1.cup.hp.com>  v In article <EF1DC894691AD5118AF000508BB85FDE034CC7E7@AMCLVX11>, "Lucas, Edward A (SAIC)" <Edward.Lucas@bp.com> writes:  H :...My plan is to replace the SW, plug (I think) a serial cable from the0 :TS (DS200) to the console port on the server.   :mF :Can someone please point me in the correct direction. I need to learnK :everything about this and then configure. I need to get this working ASAP._  H   Wire the DECserver port to the target console (usually 9600 8N1), nameJ   and enable the DECserver port for remote access via DECserver port-levelH   commands, and then use SET HOST/LAT[/DESTINATION_PORT] to connect from6   the OpenVMS host to the target LAT service and port.  E   Please see LAT$SYSTARTUP.TEMPLATE and the OpenVMS system managementiE   documentation for details of setting up for SET HOST/LAT operationstF   on the OpenVMS host.  LCP SET NODE /CONNECTIONS=BOTH, LAT$LINK, etc.  G   Please be aware that the DECserver hardware power-up can potentially tB   generate spurious characters on the serial lines until the powerF   stabilizes, and this can be misinterpreted as a halt by the console.H   Please see the FAQ for some discussions of HALT and BREAK and related.  G   Some folks prefer to use opto-isolators on the serial lines, to avoida!   the unexpected flow of current.a  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------    Date: 26 Aug 2002 16:05:03 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)w Subject: Re: Reverse LAT< Message-ID: <cf15391e.0208261505.dbac4f6@posting.google.com>  { "Lucas, Edward A (SAIC)" <Edward.Lucas@bp.com> wrote in message news:<EF1DC894691AD5118AF000508BB85FDE034CC7E7@AMCLVX11>...oN > I need to replace my current "CA Console manager SW"  The SW enables me (andJ > my team) to access all system like being on a console but remotely (i.e.M > >>>).  My plan is to replace the SW, plug (I think) a serial cable from thei1 > TS (DS200) to the console port on the server.  e  A I'm a bit confused as to what you mean by "replace the SW".  I'lle@ assume "SW" means "Software".  Did you mean upgrade your currentC version of 'Unicenter Console Management for OpenVMS' software withiE the latest version, replace it with another product, or get rid of ith	 entirely?   B If you're looking for alternative products, there are ConsoleWorksC from TecSys Development, Inc., RoboCentral from Heroix, and Commandu/ Line Interface Manager (CLIM) from Ki Networks.f. ----------------------------------------------. Keith Parris | parris at encompasserve dot org   ------------------------------  # Date: Tue, 27 Aug 2002 02:44:45 GMT-# From: "John Smith" <a@nonymous.com>f4 Subject: Server Market Is Down in the Second QuarterH Message-ID: <xIBa9.19223$6m61.1706@news04.bloor.is.net.cable.rogers.com>  L http://news.yahoo.com/news?tmpl=story2&cid=569&ncid=738&e=1&u=/nm/20020827/t c_nm/tech_servers_dc_1  + Server Market Is Down in the Second Quarter  Mon Aug 26,10:25 PM ET  K SAN FRANCISCO (Reuters) - International Business Machines Corp. widened itsnF lead in a shrinking market for powerful server computers in the secondE quarter, raking in nearly 30 percent of revenue worldwide, technology7, researcher Gartner Dataquest said on Monday.  H But rival Sun Microsystems Inc. showed the fastest rise in market share,K Hewlett-Packard Co. kept the lead in unit volume, thanks to its acquisition.L of Compaq Computer in May, and Dell Computer Corp. also increased its share.  L The global market for computers which manage corporate networks and serve upH data to other machines shrank to $10.1 billion from $11.6 billion in theL second quarter a year earlier, and revenue from all top four vendors shrank,
 Gartner said.T  F IBM led the pack with 29.6 percent of the market revenue, up from 27.9L percent, followed by HP with 24.7 percent, down from 27.4, and Sun with 18.4 percent, up from 16.6 percent.  H Dell increased its revenue share to 7.2 percent from 6.5 percent and wasL second by unit volume to HP. The two companies are leaders in the market for lower end servers.     -------------------------Z  = The above is an apples to oranges comparison for HP and Dell.  Dell only sells Wintel.EI It'd be interesting to know what happened to HP's Wintel server sales vs.h Dell.t  E Do the above figures account for just server boxes or do they include L attached, optional extra storage like external RAID sold at the same time as the server?d   ------------------------------    Date: 26 Aug 2002 11:36:23 -0700$ From: tshoppa@wmata.com (Tim Shoppa)) Subject: Re: set file/data_check (REPOST)a= Message-ID: <8327b166.0208261036.62fc4fa6@posting.google.com>e  s "Bill Todd" <billtodd@metrocast.net> wrote in message news:<qWqa9.217771$SS.8766688@bin3.nnrp.aus1.giganews.com>... 3 > "Tim Shoppa" <tshoppa@wmata.com> wrote in message 9 > news:8327b166.0208260357.4734d5b5@posting.google.com...:9 > > "Bill Todd" <billtodd@metrocast.net> wrote in message @ >  news:<nFga9.209265$SS.8445357@bin3.nnrp.aus1.giganews.com>...> > > > "Tim Shoppa" <shoppa@trailing-edge.com> wrote in message1 > > > news:3D694FA3.248F3039@trailing-edge.com...o > > > > Bill Todd wrote:M > > > > > That's something I've wondered about too, in the context of 'end too >  end'CH > > > > > validation.  I've always thought it would be nice for disks to
 >  support >  8 -E > > > > > 16 bytes of ancillary data associated with each disk sectorT > > > >8 > > > > Most SCSI drives do; > > >n$ > > > If that's read-long/write-long > >eI > > No, it's not; a low-level format can be used to give 520 to 528 bytestL > > per sector on most SCSI drives after issuing the appropriate mode select> > > commands.  (With a corresponding fewer sectors per drive.) > H > Oh.  I already mentioned that AS/400 and some storage systems do that.  + Actually, a number of other systems do too.N   >  ButK > it's not 'ancillary' data but in-line, which makes it somewhat awkward toe > deal with   L But, for end-to-end integrity checking, the more 'in-line' it is the better.M Making it ancillary means that applications will not have to check/update it.u  I Having the OS/driver level strip it out isn't the worst thing you can do,l' but isn't our goal end-to-end checking?   ; >(as well as not being an available feature on IDE drives).f  N Building reliable resilient systems out of "commodity" least-common-denomiatorK hardware is an interesting task, but I didn't think we were playing that at- this moment.  Were we?   Tim.   ------------------------------  # Date: Mon, 26 Aug 2002 23:26:34 GMT * From: "Bill Todd" <billtodd@metrocast.net>) Subject: Re: set file/data_check (REPOST)fB Message-ID: <KOya9.213920$m91.8876718@bin5.nnrp.aus1.giganews.com>  1 "Tim Shoppa" <tshoppa@wmata.com> wrote in message27 news:8327b166.0208261036.62fc4fa6@posting.google.com...f7 > "Bill Todd" <billtodd@metrocast.net> wrote in messagea= news:<qWqa9.217771$SS.8766688@bin3.nnrp.aus1.giganews.com>....5 > > "Tim Shoppa" <tshoppa@wmata.com> wrote in messageL; > > news:8327b166.0208260357.4734d5b5@posting.google.com... ; > > > "Bill Todd" <billtodd@metrocast.net> wrote in messageuB > >  news:<nFga9.209265$SS.8445357@bin3.nnrp.aus1.giganews.com>...@ > > > > "Tim Shoppa" <shoppa@trailing-edge.com> wrote in message3 > > > > news:3D694FA3.248F3039@trailing-edge.com...0 > > > > > Bill Todd wrote:L > > > > > > That's something I've wondered about too, in the context of 'end to	 > >  end' J > > > > > > validation.  I've always thought it would be nice for disks to > >  support > >  8 -G > > > > > > 16 bytes of ancillary data associated with each disk sector 	 > > > > >  > > > > > Most SCSI drives do; > > > >a& > > > > If that's read-long/write-long > > >oK > > > No, it's not; a low-level format can be used to give 520 to 528 bytesnG > > > per sector on most SCSI drives after issuing the appropriate moden select@ > > > commands.  (With a corresponding fewer sectors per drive.) > >gJ > > Oh.  I already mentioned that AS/400 and some storage systems do that. > - > Actually, a number of other systems do too.l >i > >  ButJ > > it's not 'ancillary' data but in-line, which makes it somewhat awkward to
 > > deal with  > F > But, for end-to-end integrity checking, the more 'in-line' it is the better.tK > Making it ancillary means that applications will not have to check/updatet it.m  J Applications shouldn't have to check/update it:  once it's in main memory,J the OS can do that just fine and present it to the application without theK check data.  Which is why having the check data embedded is awkward:  you'd J like to be able just to dump the requested data into memory and handle theK check data elsewhere (or, when writing, accept the data stream as presentedcB by the application and add the check data on the side).  Yes, SCSID scatter/gather does make it possible to embed the data in over-sizedI sectors, but not only is that awkward to manage but it means that *every*:H disk-accessing facility (right out to the application, if handled as you+ describe) must understand how to handle it.h  J 'End-to-end' usually refers to hardware transit of the data:  if you can'tJ trust it to remain valid in main memory when passed from the kernel to the= application, then you have far worse problems to worry about.    >eK > Having the OS/driver level strip it out isn't the worst thing you can do, ) > but isn't our goal end-to-end checking?e  
 See above.   >c= > >(as well as not being an available feature on IDE drives).e >v8 > Building reliable resilient systems out of "commodity" least-common-denomiatorhJ > hardware is an interesting task, but I didn't think we were playing that at > this moment.  Were we?  H EMC certainly has started to (see their 'Centera' storage box), and theyF define higher-end, reliable storage as much as anyone.  The quality ofI commodity IDE drives from reputable vendors matches that of the best SCSIgK drives available just a few years ago, and isn't all that far below that ofSL the best drives available today:  given that they cost less than 15% as muchA per MB and less than 40% as much for a given level of (aggregate)sD random-access performance (with the bonus of yielding over twice theK capacity at that performance level), they're damn attractive if one managesoG them with an eye to their limitations.  Long ago, they used significant L short-cuts (e.g., not checking the integrity of transfers over the bus) thatI made their use much more questionable, but that hasn't been true for manyr
 years now.   - bill   ------------------------------  % Date: Mon, 26 Aug 2002 14:56:19 -0400o; From: "Webb, William W Raleigh, NC" <wwebb1@email.usps.gov>a+ Subject: RE: silliest thing heard this weekoK Message-ID: <BA52530E3149734A9BAABDBBFA808E49028BEA33@rlghncst964.usps.gov>e  ) Four ES40s fully populated, 16 GB apiece.e5 I don't know how much more I'm permitted to disclose.t   WWWebb   -----Original Message-----2 From: "Paul Sture" [mailto:p_sture@elias.decus.ch]% Sent: Sunday, August 25, 2002 4:33 AM  To: "Info-VAX@Mvb.Saic.Com" + Subject: RE: silliest thing heard this weekD    K In article <BA52530E3149734A9BAABDBBFA808E49028BEA28@rlghncst964.usps.gov>,c= "Webb, William W Raleigh, NC" <wwebb1@email.usps.gov> writes:  >y > Not. >c= > Usernames and node identities redacted for obvious reasons.> >  > REDACTED$ SHOW USERS/PAGE, >u9 >  Username      Node     Interactive  Subprocess   Batch 9 >       OpenVMS User Processes at 23-AUG-2002 13:33:04.13t? >     Total number of users = 2278,  number of processes = 2646t >i   <snip>  H Exactly. Care to share how much memory, which processor(s), and how many nodes?  K Sharon has raised an excellent point though. On my web travels I constantly-K come across technically good guys and girls who quote VAX/VMS experience insC their resumes, and haven't a clue of the capabilities of VMS today..  J I even spotted someone recently saying that Distaster Tolerance was only aH buzzphrase used by consultants (implied slur there), while in real life, people2 were concentrating on Disaster Recovery solutions.   Sheesh!u used by consultants    __
 Paul Sture Switzerlandh   ------------------------------   Date: 26 Aug 02 22:41:00 +0200) From: p_sture@elias.decus.ch (Paul Sture)a+ Subject: Re: silliest thing heard this weekc) Message-ID: <mYAh1SaZQPqX@elias.decus.ch>t  Z In article <3D690C73.1060704@qsl.network>, "John E. Malmberg" <wb8tyw@qsl.network> writes:  > system@SendSpamHere.ORG wrote:X >> In article <Fl$U3cyEau7v@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes: >> h >>>aH >>>I hate to say this, but you can change the IP address (on NT) without
 >>>rebooting.o >>  ! >> NT4SP6a.  I need to reboot it.t >>  1 > NT pops up a message that a reboot is required.e > J > It is documented that the reboot is not required, and if you click [NO] I > on the reboot now box, you will find that the new I.P address has been t > activated. > K > I do not know if this bug is documented in the on-line database.  It has  0 > been know for quite a bit of time before SP6a. > F I must admit to being suprised when I last did it that a reboot wasn't/ required. I will try again to see what happens.l   __
 Paul Sture SwitzerlandT   ------------------------------  % Date: Mon, 26 Aug 2002 21:55:55 -0400I- From: JF Mezei <jfmezei.spamnot@videotron.ca>f+ Subject: Re: silliest thing heard this weeks, Message-ID: <3D6ADC2B.59602E5E@videotron.ca>   John Smith wrote: M > back in 1982 I was running up to 30 simultaneous users on a 750 with 4mb ofn  > memory, running a Fortran app.  N In 1987, I was running 12 users on ALL-IN-1 (WPSPLUS) on a Microvax II with 16K meg of memory and a 154 meg disk drive (RD54).  Back then, they though thatlN the 12 or so seconds to start A1 was long. Now they complain about the minutes- it takes to login to NT and start everything.-   ------------------------------  + Date: Mon, 26 Aug 2002 21:49:21 +0000 (UTC)v* From: bleau@umtof.umd.edu (Lawrence Bleau)A Subject: Re: Size of I/O count fields in SHOW PROC and SHOW SYS ?d0 Message-ID: <ake7p1$2l2$1@grapevine.wam.umd.edu>  _ In article <3D687BC8.36E49932@trailing-edge.com>, Tim Shoppa <shoppa@trailing-edge.com> writes:e [snip]= >For systems with moderate uptimes and processes with lots ofyC >I/O (especially network I/O) the displayed counters would overflow.D >and print as "*********" after a few weeks using 1990-era machines. >oE >On a modern busy machine with just a few hours of uptime and networko@ >I/O intensive processes, it must be the case that the displayedD >counts overflow very quickly.  Have recent versions of VMS expandedC >the displayed fields?  If not, are there any plans in the works too >expand these fields?e  I While you're at it, how about changing the field widths on the ACCOUNTING 9 suymmary report?  This overflows regularly on my reports.b   Lawrence Bleau University of Maryland" Physics Dept., Space Physics Group 301-405-6223 bleau@umtof.umd.edur   ------------------------------  % Date: Mon, 26 Aug 2002 16:20:31 -0600)$ From: Lee Y T Mah <lytmah@cha.ab.ca>A Subject: Re: Size of I/O count fields in SHOW PROC and SHOW SYS ?.) Message-ID: <3D6AA9AE.1AB3086D@cha.ab.ca>o  / You could set the display width to 132 columns?e     Guy Peleg wrote:   > Tim, >,H > There is not much extra space in $SHOW SYSTEM, however $SHOW PROC/CONT > has a lot of extra space? > in the Direct I/O field. If you can escalate this formally toe0 > engineering through your local support center, > I will fix this. >n
 > Regards, >e > Guy Pelegt > OpenVMS Utilities Group. >. > Tim Shoppa wrote:d > A > > Since the dawn of time (well, I started using VMS in the late D > > 3.x and early 4.x days) the SHOW SYS output has had 9 digits forA > > the I/O count and SHOW PROC/CONT has had 8 digits for the I/OrH > > count.  For systems with moderate uptimes and processes with lots ofF > > I/O (especially network I/O) the displayed counters would overflowG > > and print as "*********" after a few weeks using 1990-era machines.f > >eH > > On a modern busy machine with just a few hours of uptime and networkC > > I/O intensive processes, it must be the case that the displayed.G > > counts overflow very quickly.  Have recent versions of VMS expanded F > > the displayed fields?  If not, are there any plans in the works to > > expand these fields? > >. > > Tim.   -- Leee  ; Lee Y T Mah                        Capital Health Authorityn? Email: lytmah@cha.ab.ca            Information Systems, RAH CSCl4 Phone:  (780) 477-4725, 477-4233   10240 Kingsway NW? Fax:      (780) 491-5119, 491-5619    Edmonton, AB, CAN  T5H3V9e   ------------------------------  # Date: Tue, 27 Aug 2002 03:49:21 GMTh" From: Guy Peleg <guy.peleg@hp.com>A Subject: Re: Size of I/O count fields in SHOW PROC and SHOW SYS ?i& Message-ID: <3D6AF51C.F0ECB817@hp.com>  A Again, easy fix, just escalate this to your local support center.E   GuyC   Lawrence Bleau wrote:.  a > In article <3D687BC8.36E49932@trailing-edge.com>, Tim Shoppa <shoppa@trailing-edge.com> writes:r > [snip]? > >For systems with moderate uptimes and processes with lots ofiE > >I/O (especially network I/O) the displayed counters would overflowiF > >and print as "*********" after a few weeks using 1990-era machines. > >sG > >On a modern busy machine with just a few hours of uptime and networkaB > >I/O intensive processes, it must be the case that the displayedF > >counts overflow very quickly.  Have recent versions of VMS expandedE > >the displayed fields?  If not, are there any plans in the works toi > >expand these fields?- >-K > While you're at it, how about changing the field widths on the ACCOUNTINGm; > suymmary report?  This overflows regularly on my reports.: >- > Lawrence Bleau > University of Maryland$ > Physics Dept., Space Physics Group > 301-405-6223 > bleau@umtof.umd.edu4   ------------------------------   Date: 26 Aug 2002 19:25:37 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)% Subject: Re: SYS$EXAMPLES:ADDUSER.COMd* Message-ID: <akdvbh$pfq$6@web1.cup.hp.com>  i In article <f019af2c.0208241458.746e3962@posting.google.com>, helbig@multivax.de (Phillip Helbig) writes:)   ..@ :I know it's just a (sys$)example.  It works fine with PRIV=ALL.  G   DCL that uses f$setprv (or SET PROCESS/PRIVILEGE) is arguably busted.aF   It should either try the operation and recover from the error, or itF   should check for the privileges required and return an error if not.H   Silently enabling privileges is something that has always troubled me.  F :Sure, the change required is trivial, but since SYS$EXAMPLES is oftenF :a place to start looking to avoid re-inventing the wheel, it would be+ :nice if everything there were very robust.i  5   This is hardly a controversial position, of course.-  A   Many of the existing SYS$EXAMPLES: example modules target oldera9   compilers and older syntax, and can be quite out-dated.u  A   I've spent quite a chunk of my "spare" time cleaning up variousuB   SYS$EXAMPLES: modules and SYS$STARTUP: modules over the past fewD   years.  It's slow going; I've cleaned up just three or four of the@   existing modules over the past couple of months.  (The one I'mB   presently working on has consumed 16 hours so far, and I'm stillC   debugging it.  And -- as I mentioned recently to some other folksa@   that were seeking better examples -- I have the IA-64 port and5   other code development projects officially queued.)r  F   I was tempted to purge SYS$EXAMPLES: entirely and start over, thoughE   that would likely generate as many complaints as would the existingw1   and out-dated examples and associated problems.r  C   The http://askq.compaq.com/ website often has far newer examples.a  4   Please see the FAQ for other pointers to examples.    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 26 Aug 2002 19:46:50 +0200d, From: "Peter Grevel" <p.grevel@inter.nl.net>% Subject: Using C - Api in PASCAL code 7 Message-ID: <3d6a695d$0$231$19deed1b@news.inter.NL.net>m   Hello Everyone    L Is there a way of using an API or example code written in C to use in PASCAL- (both OpenVMS)?. And if this is possible HOW?o  K I want to use an CA-Console Manager example written in C with some specificoJ functions from an include file (Console.h). Because PASCAL is familiar and I'm new in using C.      Thank you in advance.l     Peterh   ------------------------------  # Date: Mon, 26 Aug 2002 18:57:34 GMTh& From: John Reagan <john.reagan@hp.com>) Subject: Re: Using C - Api in PASCAL coder# Message-ID: <3D6A7839.70409@hp.com>u   Peter Grevel wrote:a > Hello Everyone >  > N > Is there a way of using an API or example code written in C to use in PASCAL/ > (both OpenVMS)?. And if this is possible HOW?1 > M > I want to use an CA-Console Manager example written in C with some specificoL > functions from an include file (Console.h). Because PASCAL is familiar and > I'm new in using C.@ >  >  > Thank you in advance.  >  >  > Peter9 >  >   < You can certainly call C from Pascal and call Pascal from C.  ? For structs/records, you'll have to make sure that your struct 9G declarations in C match your record declarations in Pascal.  You might iG want to look at Pascal's /LIST/SHOW=STRUCTURE_LAYOUT for insight as to  9 how the Pascal compiler allocated fields in your records.k  H If you passing nul-terminated strings back and forth, you'll have to be I careful of that on the Pascal side (Pascal has several builtin functions  H and a builtin type, C_STR_T to help manipulate nul-terminate strings in  Pascal).  I Other areas to watch:  C passes scalar parameters by by-immediate-value; g8 Pascal always uses by-reference (you can override that).  L If you have more detailed questions, ask them here and I'm sure we can help.   --   John Reaganm' Compaq Pascal/{A|I}MACRO Project Leader  Hewlett-Packard Companyt   ------------------------------  % Date: Mon, 26 Aug 2002 12:49:50 -0700t# From: "Tom Linden" <tom@kednos.com>n) Subject: RE: Using C - Api in PASCAL code 9 Message-ID: <CIEJLCMNHNNDLLOOGNJICEOJFJAA.tom@kednos.com>    >-----Original Message-----t. >From: John Reagan [mailto:john.reagan@hp.com]' >Sent: Monday, August 26, 2002 11:58 AM' >To: Info-VAX@Mvb.Saic.Com* >Subject: Re: Using C - Api in PASCAL code >t >  >Peter Grevel wrote: >> Hello Everyonei >>   >> mB >> Is there a way of using an API or example code written in C to  >use in PASCAL0 >> (both OpenVMS)?. And if this is possible HOW? >> eA >> I want to use an CA-Console Manager example written in C with u >some specificA >> functions from an include file (Console.h). Because PASCAL is f
 >familiar and( >> I'm new in using C. >> 2 >> $ >> Thank you in advance. >> n >> o >> Peter >> p >> s >p= >You can certainly call C from Pascal and call Pascal from C.a >o@ >For structs/records, you'll have to make sure that your struct H >declarations in C match your record declarations in Pascal.  You might H >want to look at Pascal's /LIST/SHOW=STRUCTURE_LAYOUT for insight as to : >how the Pascal compiler allocated fields in your records.  1 Why not use SDL?  Isn't that why it was invented?    >:I >If you passing nul-terminated strings back and forth, you'll have to be pJ >careful of that on the Pascal side (Pascal has several builtin functions I >and a builtin type, C_STR_T to help manipulate nul-terminate strings in m	 >Pascal).  >dJ >Other areas to watch:  C passes scalar parameters by by-immediate-value; 9 >Pascal always uses by-reference (you can override that).  >pD >If you have more detailed questions, ask them here and I'm sure we 
 >can help. >o >--  >John Reagan( >Compaq Pascal/{A|I}MACRO Project Leader >Hewlett-Packard Company >- >---' >Incoming mail is certified Virus Free.o; >Checked by AVG anti-virus system (http://www.grisoft.com).o@ >Version: 6.0.381 / Virus Database: 214 - Release Date: 8/2/2002 >  ---n& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.381 / Virus Database: 214 - Release Date: 8/2/2002t   ------------------------------    Date: 26 Aug 2002 15:21:45 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) ) Subject: RE: Using C - Api in PASCAL codee3 Message-ID: <zAAa2sW3EjIB@eisner.encompasserve.org>t  _ In article <CIEJLCMNHNNDLLOOGNJICEOJFJAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:s   >>Peter Grevel wrote:u >>> Hello Everyone >>>  >>> C >>> Is there a way of using an API or example code written in C to w >>use in PASCAL 1 >>> (both OpenVMS)?. And if this is possible HOW?i >>> B >>> I want to use an CA-Console Manager example written in C with  >>some specifictB >>> functions from an include file (Console.h). Because PASCAL is  >>familiar and >>> I'm new in using C.t  3 > Why not use SDL?  Isn't that why it was invented?d  D But it seems possible CA did not use SDL to define their interfaces,D so their structure definitions would not be the way SDL would do it.   ------------------------------  # Date: Mon, 26 Aug 2002 17:45:02 GMTu# From: "John Smith" <a@nonymous.com>E3 Subject: Re: Want an alpha or 20?  Found an auctiondI Message-ID: <yOta9.16859$6m61.16330@news04.bloor.is.net.cable.rogers.com>m  7 "Jay E. Morris" <morrisj@epsilon3.com> wrote in messageo7 news:8600a73e.0208260930.1ff3991a@posting.google.com...l5 > While skimming about found this auction in England.  >'J > http://www.dovebid.com/search/searchresults.asp?Query=alpha&EventID=1450 >aD > 1200, 8400, 300, DS10, DS20, ES40.  Might be a chance to pick up aE > hobbiest system.  Never used dovebid.com before but they state theynD > handle "Business auctions" so might be up against some deep pocket
 > bidders.    C Dove Bid is a highly reputable auctioneer for industrial/commercial 
 equipment.   [dig on]L Some of the bidders are likely to have deep pockets but there is unlikely toJ be many of them. With HP's lack of marketing of OpenVMS, and the executionK of Tru64, the only people you are likely to run up against in an auction ofiK Alpha's are the existing, but dwindling, VMS customer base (411,000 systemsn0 or less), the Tru64 base, and a few Linux types.  I It won't be PC pricing, but you should be able to pick something up quitea inexpensively.
 [/dig off]   ------------------------------  % Date: Mon, 26 Aug 2002 17:05:12 -0400n( From: David Froble <davef@tsoft-inc.com>3 Subject: Re: Want an alpha or 20?  Found an auction , Message-ID: <3D6A9808.5090701@tsoft-inc.com>   John Smith wrote:   9 > "Jay E. Morris" <morrisj@epsilon3.com> wrote in message-9 > news:8600a73e.0208260930.1ff3991a@posting.google.com...0 > 5 >>While skimming about found this auction in England.g >>J >>http://www.dovebid.com/search/searchresults.asp?Query=alpha&EventID=1450 >>D >>1200, 8400, 300, DS10, DS20, ES40.  Might be a chance to pick up aE >>hobbiest system.  Never used dovebid.com before but they state theyoD >>handle "Business auctions" so might be up against some deep pocket
 >>bidders. >> >  > E > Dove Bid is a highly reputable auctioneer for industrial/commercial  > equipment. > 
 > [dig on]N > Some of the bidders are likely to have deep pockets but there is unlikely toL > be many of them. With HP's lack of marketing of OpenVMS, and the executionM > of Tru64, the only people you are likely to run up against in an auction ofoM > Alpha's are the existing, but dwindling, VMS customer base (411,000 systemsv2 > or less), the Tru64 base, and a few Linux types. > K > It won't be PC pricing, but you should be able to pick something up quiteo > inexpensively. > [/dig off]    P Usually these people are selling large lots of stuff.  So, instead of that nice Q little AlphaStation you want, you might have to bid on a lot of say, 50 systems, rJ and not all of them Alphas.  At least, this is what I've seen in the past.   Dave   ------------------------------    Date: 26 Aug 2002 22:57:36 -0700# From: kostik@beenet.ru (Konstantin)s( Subject: What's going on with Encompass?= Message-ID: <9c4388dd.0208262157.25b288d7@posting.google.com>e   Hi!c  N I'm trying to register as Encompass Associate for 3 weeks without any success.B The only thing I've got is confirmation that I've filled web-form.H Then I wrote a mail to membership@encompassus.org and still no answer...  % Ppl, what is going on with Encompass?e   ------------------------------   Date: 26 Aug 2002 19:32:28 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)+ Subject: Re: XFC patch for VMS 7.3 is up...R* Message-ID: <akdvoc$pfq$7@web1.cup.hp.com>  U In article <z8rHzGPwnQHX@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes:, :In article <rdeininger-2408020930110001@1cust82.tnt2.nashua.nh.da.uu.net>, rdeininger@mindspring.com (Robert Deininger) writes: :s; :While on the topic of XFC, has anyone experience of taking 8 :the V7.3-1 option of changing the system disk to ODS-5?  H   The related "five or ten or more minutes" listed for switching to hardJ   links -- if you choose that option -- is rather optimistic.  The systemsA   I've switched have taken roughly an hour to complete this step.   H   I'm running ODS-5 on most of my local OpenVMS Alpha system disks.  (It1   makes porting off-platform code rather easier.)   I :I think I need to do this to make sure my bit of software doesn't break.u  E   Please review the extended file naming documentation in the OpenVMSuD   manual set.  There are cases where existing code can have specificE   assumptions that are broken by ODS-5.  (I've one local package thatu>   is tied into ODS-2 syntax far more tightly than I'd prefer.)    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 26 Aug 02 22:24:39 +0200) From: p_sture@elias.decus.ch (Paul Sture) + Subject: Re: XFC patch for VMS 7.3 is up... ) Message-ID: <DgVQ+hjg4i9+@elias.decus.ch>l  l In article <akco14$hgp3@doiweb4.volkswagen.de>, Karl Rohwedder <extern.karl.rohwedder@volkswagen.de> writes: > Paul Sture wrote:] >> In article <rdeininger-2408020930110001@1cust82.tnt2.nashua.nh.da.uu.net>, rdeininger@mindspring.com (Robert Deininger) writes: >> a= >> While on the topic of XFC, has anyone experience of takinge: >> the V7.3-1 option of changing the system disk to ODS-5? >>  K >> I think I need to do this to make sure my bit of software doesn't break.d >> __t
 >> Paul Sture  >> Switzerland > X > I am running the VMS V7.3-1 EFT with the systemdisk changed to ODS5 with hardlinks and9 > access date enabled and encountered no problems so far.b   Thanks.g   mfg :-)w __
 Paul Sture Switzerlandm   ------------------------------   Date: 26 Aug 02 23:04:45 +0200) From: p_sture@elias.decus.ch (Paul Sture),+ Subject: Re: XFC patch for VMS 7.3 is up...o) Message-ID: <CWvOTeZU93bt@elias.decus.ch>-  _ In article <akdvoc$pfq$7@web1.cup.hp.com>, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:  > W > In article <z8rHzGPwnQHX@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes:  > :In article <rdeininger-2408020930110001@1cust82.tnt2.nashua.nh.da.uu.net>, rdeininger@mindspring.com (Robert Deininger) writes: > :h= > :While on the topic of XFC, has anyone experience of takingr: > :the V7.3-1 option of changing the system disk to ODS-5? > J >   The related "five or ten or more minutes" listed for switching to hardL >   links -- if you choose that option -- is rather optimistic.  The systemsC >   I've switched have taken roughly an hour to complete this step.  >    That's useful to know.  SJ >   I'm running ODS-5 on most of my local OpenVMS Alpha system disks.  (It3 >   makes porting off-platform code rather easier.)m > K > :I think I need to do this to make sure my bit of software doesn't break.d > G >   Please review the extended file naming documentation in the OpenVMSeF >   manual set.  There are cases where existing code can have specific) >   assumptions that are broken by ODS-5./   Will do.    >   (I've one local package that@ >   is tied into ODS-2 syntax far more tightly than I'd prefer.) >   E This software was originally written to target a variety of unices ass& well as OpenVMS, so that is a concern. __
 Paul Sture Switzerlande   ------------------------------   Date: 26 Aug 2002 18:34:59 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)( Subject: RE: XFC v2 ECO, negative report* Message-ID: <akdscj$pfq$2@web1.cup.hp.com>   In article <3D35AD137AAAD411A6BA0008C7B1B12D032C3B9E@MBCALBEXC03.BENDER.COM>, "Koska, John C. (LNG-MBC)" <John.C.Koska@lexisnexis.com> writes:  J   re: XFC interactions with unspecified privileged-mode software component  K :Just a guess, but probably lack of quadword alignment when doing async I/OoF :in the "other locally-relevent privileged-mode software component".    I   I'm not immediately familiar with the reference.  For user-mode code, IeI   have not needed to consider quadword alignment for asynch I/O -- thoughsI   there are certainly discussions around quadword granularity that are ofdJ   interest for asynchronous operations, please see the OpenVMS FAQ for one   such discussion.  J   As mentioned earlier, this appears to be an adverse reaction between theJ   XFC V2.0 baselevel and the use of inner-mode driver-level APIs within a J   site-locally-supported privileged-mode software component.  The initial K   investigation points to a potential "disagreement" within the kernel-mode-H   I/O completion path of this privileged-mode code within the context ofJ   XFC, but (as noted in my earlier posting) we've not yet duplicated this ,   particular problem within the OpenVMS lab.  D   Please see my previous post in this thread for additional details.  L   Once the actual problem is identified, I will post additional information.  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 26 Aug 2002 19:02:47 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)( Subject: Re: XFC v2 ECO, negative report* Message-ID: <akdu0n$pfq$4@web1.cup.hp.com>  f In article <3D653F0C.1411E30D@clarityconnect.com>, "Mark D. Jilson" <jilly@clarityconnect.com> writes:D :Please report this to your local CSC.  AUTOGEN has had a history of1 :problems with figuring out S0 space consumption.f :e :Jim Strehlow wrote:F :> (Note: AUTOGEN does not detect an error if you specify too large of+ :> a MIN_VCC_MAXSIZE value in MODPARAMS.DATm  I   There is a requirement to correctly set the reserved memory, please seedJ   the documentation for the details and the SYSMAN RESERVED_MEMORY syntax.J   (I should have more on this -- there are some expected but quite obscureJ   behaviours here both for VIOC and for XFC -- added into my presentationsH   in time for the upcoming fall events.  There are OpenVMS documentation   updates underway, as well.)d    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   End of INFO-VAX 2002.471 ************************