1 INFO-VAX	Mon, 03 Nov 2003	Volume 2003 : Issue 610       Contents:9 2010: a VAX end of Space Odissey, but what about VAX/VMS? = Re: 2010: a VAX end of Space Odissey, but what about VAX/VMS? = Re: 2010: a VAX end of Space Odissey, but what about VAX/VMS? = Re: 2010: a VAX end of Space Odissey, but what about VAX/VMS?  Adaptec 29160 SCSI card and VMS , Re: Authorised reseller application - UPDATE cluster load balancing Re: cluster load balancing- Condos in Canada @ 75 cents on the US Dollar!  Re: Copy of SuperLat) Re: Database selection (was: Advertising) 2 Re: forwarding routine in C - Alpha assembly help?& Intergrity server first boot (windoze)* Re: Intergrity server first boot (windoze) Multiple telnet sessions Re: Multiple telnet sessions! Re: Open VMS Vax 7.X patch digest  Re: OpenVMS, NAS, Shadowing.... * Re: Passing a socket to a detached process& Passing a socket to a detached process Re: Show Memory/phys Re: Show Memory/phys Re: SIMH used with VMS?  Re: SIMH used with VMS?  Re: SIMH used with VMS?  Re: SIMH used with VMS?  Re: SIMH: used with VMS? Re: SIMH: used with VMS? Re: SIMH: used with VMS? Re: SIMH: used with VMS? Re: SIMH: used with VMS? Re: SIMH: used with VMS? Re: SIMH: used with VMS? Re: SIMH: used with VMS? Re: SIMH: used with VMS?' Re: simple advice needed on VAX upgrade ' Re: simple advice needed on VAX upgrade 1 Re: Speaking of Cincom (Was Re: World Wide Wake]) 1 Re: Speaking of Cincom (Was Re: World Wide Wake])  Re: Text library extensions  Re: Text library extensions - Re: To foreign or verb ? That is the question - Re: To foreign or verb ? That is the question - Re: To foreign or verb ? That is the question - Re: To foreign or verb ? That is the question - Re: To foreign or verb ? That is the question  Re: TS10, still around?  Re: TS10, still around? P Re: Veritas VMS client requested restores - solution found to dramatically  impr Re: [~OT] Last _production_ VAX   F ----------------------------------------------------------------------  % Date: Mon, 03 Nov 2003 16:20:12 +0100 " From: Didier Morandi <no@spam.com>B Subject: 2010: a VAX end of Space Odissey, but what about VAX/VMS?4 Message-ID: <3fa67236$0$13269$626a54ce@news.free.fr>  O I suddently realize that 2010 will ring the end of VAX support by HP, but what  N about VAX/VMS? The OpenVMS roadmap does have a serie of slides entitled "VAX, : Alpha and Itanium" but I did not find anything on VAX/VMS.  O The question is: if users replace their VAX by VAX-emulated systems, such like  1 the FutureVAX that I announced two days ago (see  N http://www.didiermorandi.com/vms/FutureVAX_3196.pdf), they solve that way the , VAX obsolescence issue, not the VAX/VMS one.  B Where is the interest for these Customers to swap from VAX/VMS to - mumblefratz/VMS if VAX/VMS is phased out too?    D.   ------------------------------  # Date: Mon, 03 Nov 2003 15:43:38 GMT " From:   VAXman-  @SendSpamHere.ORGF Subject: Re: 2010: a VAX end of Space Odissey, but what about VAX/VMS?0 Message-ID: <00A2857F.CFF50C5A@SendSpamHere.ORG>  Y In article <3fa67236$0$13269$626a54ce@news.free.fr>, Didier Morandi <no@spam.com> writes: P >I suddently realize that 2010 will ring the end of VAX support by HP, but what O >about VAX/VMS? The OpenVMS roadmap does have a serie of slides entitled "VAX,  ; >Alpha and Itanium" but I did not find anything on VAX/VMS.  > P >The question is: if users replace their VAX by VAX-emulated systems, such like 2 >the FutureVAX that I announced two days ago (see O >http://www.didiermorandi.com/vms/FutureVAX_3196.pdf), they solve that way the  - >VAX obsolescence issue, not the VAX/VMS one.   N Can you make it run on an Alpha iBook?  I would consider purchase but I *will* NEVER purchase Billywarez.   --  L VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)COM             5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  # Date: Mon, 03 Nov 2003 16:12:40 GMT 4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>F Subject: Re: 2010: a VAX end of Space Odissey, but what about VAX/VMS?/ Message-ID: <3FA67D68.A6A2F2D@blueyonder.co.uk>    Didier Morandi wrote:  > P > I suddently realize that 2010 will ring the end of VAX support by HP, but whatO > about VAX/VMS? The OpenVMS roadmap does have a serie of slides entitled "VAX, < > Alpha and Itanium" but I did not find anything on VAX/VMS. > P > The question is: if users replace their VAX by VAX-emulated systems, such like2 > the FutureVAX that I announced two days ago (seeO > http://www.didiermorandi.com/vms/FutureVAX_3196.pdf), they solve that way the . > VAX obsolescence issue, not the VAX/VMS one. >   9 If enough continue to buy VMX?VMS support, then why stop.   C If they don't buy support, they are running unsupported anyway :-).   C Maybe some 3rd partys might start offering support if/when HP phase  it out.    regards   C > Where is the interest for these Customers to swap from VAX/VMS to / > mumblefratz/VMS if VAX/VMS is phased out too?   > Presumably application availability is the major driving force for staying with VAX.   8 So, if the application has no migration path to Alpha or= Itanium, they will be left with 2 options (1) keep VAX system 3 unsupported (2) Start from scratch on new platform.    >  > D.   --   tim.llewellyn@blueyonder.co.uk   ------------------------------  * Date: Mon, 3 Nov 2003 12:03:48 -0500 (EST)+ From: Lord Isildur <isildur@andrew.cmu.edu> F Subject: Re: 2010: a VAX end of Space Odissey, but what about VAX/VMS?H Message-ID: <Pine.LNX.4.58-035.0311031202020.8565@unix44.andrew.cmu.edu>  N alpha ibook? i thought the only alpha portables ever made were the very short-9 lived alphabook from tadpole. i'd _love_ an alpha laptop!    isildur    > P > Can you make it run on an Alpha iBook?  I would consider purchase but I *will* > NEVER purchase Billywarez.   ------------------------------   Date: 3 Nov 2003 15:37:04 +0100 ' From: huber@mppmu.mpg.de (Joseph Huber) ( Subject: Adaptec 29160 SCSI card and VMS+ Message-ID: <IaKEsycW4fu$@vms.mppmu.mpg.de>   C  Some time ago I posted a question about an Adaptec 29160 Ultra160  " PCI SCSI adapter working with VMS.  @ Since then I went on and installed one in an XP1000, and operate1 a DLT80e on the external connector successfully.  G The 29160 is not recognized by the console SRM (V5.9-1), i.e. one can't $ boot from devices on this SCSI port.B VMS 7.3 (with update 2 in my case) works, if the interface card is/ configured via an entry in SYS$USER_CONFIG.DAT:      ! Adaptec SCSI card 29160 ! device          = "Adaptec 29160"    name          = PK#   driver        = sys$pkadriver.exe    adapter       = PCI    id            = 0x00809005   boot_class    = DK%   boot_flags    = HW_CTRL_LTR, UNIT_0 "   flags         = SCSI, PORT, BOOT
 end_device  A The PCI device ID (0x00809005) may change with different firmware A versions of the card. The pci_bus_list program from Patrick Young 2 was very helpfull in determining the PCI card ID !   --  >    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/   ------------------------------  # Date: Mon, 03 Nov 2003 13:44:55 GMT 4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>5 Subject: Re: Authorised reseller application - UPDATE 0 Message-ID: <3FA65B2B.222C077A@blueyonder.co.uk>   H Vlems wrote:   H > This is the most disturbing message I've read for years in this group.    C And its not as if the group has been plagued (!) with good news in  
 recent years.    --   tim.llewellyn@blueyonder.co.uk   ------------------------------  $ Date: Mon, 3 Nov 2003 12:25:36 +01004 From: "Axel Haringa" <axel.haringa@nl.compuware.com> Subject: cluster load balancing < Message-ID: <newscache$o2xrnh$my2$1@news.emea.compuware.com>   Hello,  J I've got a question on cluster configurations and I hardly have experience% with that, so someone may be of help:   G I set up a 2 machine cluster of equal machines using tcpip (alpha 7.31, F tcpip5.3 on 10mb) and configured a cluster alias. As far as I can findJ information, it should be possible to have the cluster act as one host, atL this way being able to do 'load balancing' by spreading connections over the cluster.  J Sofar the theory. In practice this does work, but only when one of the twoL machines becomes unresponsive. In all other cases I repeatedly end up on theI same machine as I first connected to. (when I telnet to the cluster alias F Sachem, I get connected to system Elfin, when I shut down Elfin, I get connected to Satyr)   K Can anyone help me with this problem or maybe point me out a location where L I can find more info on this subject? The info I have now is as available inH the tcpip services manual and (scarcely) on the open vms cluster systems manual.          thanks in advance,   Axel Haringa   ------------------------------  # Date: Mon, 03 Nov 2003 11:30:23 GMT ( From: Tapani Rundgren <NoSpam@telia.com># Subject: Re: cluster load balancing 4 Message-ID: <j%qpb.36227$dP1.120604@newsc.telia.net>   Axel Haringa wrote:  > Hello, > L > I've got a question on cluster configurations and I hardly have experience' > with that, so someone may be of help:  > I > I set up a 2 machine cluster of equal machines using tcpip (alpha 7.31, H > tcpip5.3 on 10mb) and configured a cluster alias. As far as I can findL > information, it should be possible to have the cluster act as one host, atN > this way being able to do 'load balancing' by spreading connections over the
 > cluster. > L > Sofar the theory. In practice this does work, but only when one of the twoN > machines becomes unresponsive. In all other cases I repeatedly end up on theK > same machine as I first connected to. (when I telnet to the cluster alias H > Sachem, I get connected to system Elfin, when I shut down Elfin, I get > connected to Satyr)  > M > Can anyone help me with this problem or maybe point me out a location where N > I can find more info on this subject? The info I have now is as available inJ > the tcpip services manual and (scarcely) on the open vms cluster systems	 > manual.  >  >  >  >  > thanks in advance, >  > Axel Haringa >  > ? Use dynamic DNS, i.e update info in DNS about the current load. 8 LoadBroker and Metric-server information is in the docs.   --  $ Rundgren dot Tapani at telia dot com   ------------------------------  # Date: Mon, 03 Nov 2003 10:20:22 GMT  From: redgryphon@shaw.ca6 Subject: Condos in Canada @ 75 cents on the US Dollar!. Message-ID: <GZppb.276569$pl3.137907@pd7tw3no>  N Summer home, investment or retirement home, one or two bedroom condo suites inV a townhouse style in one of Alberta's most popular tourist areas, Sylvan Lake.<BR><BR>U Two blocks from the Marina and lake, boating, fishing, swimming, go-karts, waterslide @ park, good shopping and excellent clubs and restaurants.<BR><BR>O Only ten minutes from a medium sized city, Red Deer, with all its city features ) including some excellent casinos.<BR><BR> . <i>(Gambling is legal in Alberta!)</I><BR><BR>C Investors, ask for a Prospectus. Free info to all for an e-mail to  K <a href="mailto:redgryphon@shaw.ca">Red Gryphon</a> or visit our website at P <a href="http://members.shaw.ca/redgryphon/index.htm">Rainbow Park </a> and take	 the tour.    -- Posted by News Bulk Poster Unregistered version   ------------------------------   Date: 3 Nov 2003 09:49:47 -0800 ' From: memnon@mail.com (Memnon Emmanuel)  Subject: Re: Copy of SuperLat = Message-ID: <d2313732.0311030949.6f74f7f7@posting.google.com>   H > Well, something I would try is using reverse-TELNET, assigning logical" > names to mimic LTA device names. >  > For example: > B > LTA100 is currently a target on a DECserver. So, I set up TELNETI > listeners for the ports on the DEC server, install a third-party TCP/IP G > stack on the VMS machine (since UCX is very limited in this area) and J > set up, for example, NTY100 to target the appropriate TELNET listener onI > the DECserver, and do this just after I start up my network stacks, but & > before I start up the queue manager: > % > $ DEFINE/SYSTEM/EXEC LTA100 NTY100:  > I > Now, when the third-party spooler comes looking for LTA100, it finds it 6 > via the logical name, grabs it and sends data to it. > E > The experiment will tell you whether the third-party spooler breaks ? > enough rules that it ignores the logical names or tries to do I > LAT-specific things to what may or may not be an LAT device. If it just H > uses modified print symbionts, you may be able to play some games with@ > copying/renaming print symbionts to get the right combination. > . > Again, that's something I would try. YMMV...    C Very interesting solution David. I will get a hold of a third party  TCP/IP stack and have a go.   
 Thanks again,  Memnon.    ------------------------------   Date: 3 Nov 2003 07:37:05 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 2 Subject: Re: Database selection (was: Advertising)3 Message-ID: <f74trpo9vXa1@eisner.encompasserve.org>   c In article <qiv6Q26EMyUg@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:   4 > That is the same mistake made by those who attemptE > to design genealogy software and provide a single field for date of 0 > birth, for instance, a VMS date-time quadword.  7    Wouldn't get you very many generations back, anyhow.    ------------------------------  % Date: Mon, 03 Nov 2003 10:58:55 +0100 D From: Thierry Lelegard <thierry.lelegard.NOSPAM@cptechno.NOSPAM.com>; Subject: Re: forwarding routine in C - Alpha assembly help? 3 Message-ID: <3FA626DF.1CCF74ED@cptechno.NOSPAM.com>   G > Thank you.  This is *almost* exactly what I am looking for.  The only E > problem we have with this is that it does not support dynamic image H > activation.  The image list is grabbed when the system is initialized.G >  Our system makes very heavy use of dynamic image activation - we can E > not simply relink our mainprog to activate all images at link time.  > C > What I need is for CallMon to hook into the lib$find_image_symbol H > routine and, as a post processing step, re-read the active image list.B >  I then need to get the current jackets reapplied to all the new" > images that were just activated. > H > Is this possible?  I guess as a first step, all I really need is a wayH > to tell CallMon to re-read the active image list.  If I had that, thenE > I could monitor all calls to lib$find_image_symbol, tell CallMon to C > re-read the active image list, and reapply all of my instrumented / > functions (I'd need to keep the list myself).  > E > Does this sound like a difficult feature for you to add?  Or, could  > you tell me how to add it?  B As I mentioned in my post, I wrote that a long time ago (1996) andD never modified it since. Since then, I left DEC, DEC is dead, CompaqE is dead, etc. My current job keeps me busy enough to not have time to $ come back into CallMon to modify it.  C However, after spending 2mn in the code, I think you could add some B "update image list" routine without difficulty, especially withoutB getting into the "dirty business" of CallMon instructon and memoryE hacker. There is a CallMon internal routine to load a shareable image 9 and some related data structures. This is done as part of C callmon$initialize. Look at this routine, look at the internal data B structures which are fully commented in a header file. This should be fine.  - Sorry not to have more time to spend on this.  -Thiery    ------------------------------  % Date: Mon, 03 Nov 2003 14:51:36 +0100 " From: Didier Morandi <no@spam.com>/ Subject: Intergrity server first boot (windoze) 4 Message-ID: <3fa65d72$0$13290$626a54ce@news.free.fr>  Q My company just booted Windows Server 2003 on "our" RS2600 Itanium2 server (just  P to have a look, you see, nothing else...) then stopped the system, until we can  run VMS.  P The shutdown window read: "Enter below the reason why you want to shutdown this  system (standalone backup)"   > This rings me a bell, but I do not remember right now what :-)   D.   ------------------------------  % Date: Mon, 03 Nov 2003 15:40:49 +0100 ) From: Michael Unger <unger@despammed.com> 3 Subject: Re: Intergrity server first boot (windoze) : Message-ID: <bo5pi3$18fk0b$1@ID-152801.news.uni-berlin.de>  , On 2003-11-03 14:51, "Didier Morandi" wrote:  S > My company just booted Windows Server 2003 on "our" RS2600 Itanium2 server (just  R > to have a look, you see, nothing else...) then stopped the system, until we can 
 > run VMS. > R > The shutdown window read: "Enter below the reason why you want to shutdown this  > system (standalone backup)"  > @ > This rings me a bell, but I do not remember right now what :-)  $ On VMS (VAX, 7.3) it's much shorter:  # > Reason for shutdown [Standalone]:    Michael    --  ; Real names enhance the probability of getting real answers. @ Please do *not* send "Security Patch Notifications" or "SecurityA Updates"; this system isn't running a Micro$oft operating system. = My e-mail account at DECUS Munich will expire on 31-Dec-2003.    ------------------------------  % Date: Mon, 03 Nov 2003 10:28:44 -0500 % From: Yakov Pekar <ypekar@purdue.edu> ! Subject: Multiple telnet sessions - Message-ID: <bo5s7c$rdd$1@mozo.cc.purdue.edu>    The problem:  I Want to determine the username of the user as soon as the user logs into  C the system.  Currently use the following set up by my predecessors:   ! user = F$EXTRACT(0,7,F$PROCESS())   I When user USERA a opens a telnet session to the system, "user" is set to  E "USERA".  In all other concurrent telnet sessions opened by the same  B user "user" is set to variations on "_TNA###" where ### are three D digits.  This makes sense as each telnet session runs as a separate F process.  I need a better way to obtain the username in case the user 4 opens more than one telnet session at the same time.   TIA, Yakov    ------------------------------  $ Date: Mon, 3 Nov 2003 10:53:17 -0500< From: "Peter Weaver" <WeaverConsultingServices@sympatico.ca>% Subject: Re: Multiple telnet sessions : Message-ID: <bo5tlf$11824k$1@ID-141708.news.uni-berlin.de>   Yakov Pekar wrote: >...> > process.  I need a better way to obtain the username in case the user6 > opens more than one telnet session at the same time. >...   Try;% $ user = "''f$getjpi("","USERNAME")'"    --   Peter Weaver Weaver Consulting Services Inc.  Canadian VAR for CHARON-VAX  www.weaverconsulting.ca    ------------------------------  $ Date: Mon, 3 Nov 2003 12:26:57 -0500 From: norm.raphael@metso.com* Subject: Re: Open VMS Vax 7.X patch digestQ Message-ID: <OFB1EB26A0.490D97CC-ON85256DD3.005F6F13-85256DD3.005F87D7@metso.com>   4 > Digest Name:  weekly Open VMS Vax 7.X patch digest+ >    Created:  Sun Nov  2  6:06:06 EST 2003   , In this document, the kit filename should be VAXDRIV03_072.A-DCX_VAXEXE not  VAXDRIV03_072.A_DCX-VAXEXE   >Document ID:  VAXDRIV03_072 Date Loaded:  200310306       Title:  DRIVER Facility kit for OpenVMS VAX V7.2   ****************************        ECO SUMMARY INFORMATION      ****************************       Kit Name: VAXDRIV03_072.A (               VAXDRIV03_072.A_DCX-VAXEXE$     Kit Applies To: OpenVMS VAX V7.2$     Approximate Kit Size: 452 blocks"     Installation Rating: INSTALL_1)     Reboot Required: Yes - rolling reboot $     Superseded Kits: VAXDRIV02_072.A/     Mandatory Kit Dependencies: VAXUPDATE01_072 #     Optional Kit Dependencies: None 3     VAXDRIV03_072.A_DCX-VAXEXE Checksum: 1407430498  <    ------------------------------  # Date: Mon, 03 Nov 2003 16:11:41 GMT # From: hoff@hp.nospam (Hoff Hoffman) ( Subject: Re: OpenVMS, NAS, Shadowing....2 Message-ID: <17vpb.8262$0b3.7098@news.cpqcorp.net>  < In article <6MCdnVqhu5E0gDmiRVn-vQ@comcast.com>, "Hal Kuff"  <kuff@tessco.com> writes:  : M :    What we would like to do is mirror two disks across a T-3 ..... as a T-3 J :will pull about 20gb/hour of uncompressed data it seems reasonable that aL :pair of HP NAS units could keep up with 10gb per hour..... We would setup aI :pair of the NAS boxes on each end of the T-3 and then NFS mount a volume H :from the NAS unit on an OpenVMS system at each end.  One OpenVMS systemJ :would write ot the NFS disk constantly and the blocks would be synched up :with the other system.  :  :    Is anyone doing this?  G   OpenVMS host-based Volume Shadowing and OpenVMS Cluster software can  H   and does operate across a T3, assuming the distance involved is withinH   the SPD-specified limit.  This configuration would be fully supported.  D   That written, OpenVMS host-based Volume Shadowing does not operateD   across the NFS protocol; it does not support data storage on disks   that are NFS-served.  F   There are also a DRS-based (hardware-based) approaches that can alsoH   remote mirroring (hardware folks often use the term mirroring, OpenVMSG   tends to use the term shadowing) of data.  You may or may not be able H   to access the data "hot" on both ends -- that said, the hardware-basedG   solutions I've seen have a primary/active and an off-line/hot-standby H   design.  (OpenVMS Volume Shadowing allows distributed/peer read-write 
   access.)  J   If you want to roll your own transactional and shadowing protocol, well,D   have at.  (I'll assume you are aware of the volume of work you areB   clearly inviting here, as this is not a trivial effort -- normalG   operations are typically easy to code and operate, it's the detection H   and recovery from errors and failures that's a far more difficult part   of the work involved.)    G   I'd expect that an NFS-based scheme might well need to verify the I/O H   using a read-after-write, and the simplest implementation of this willE   significantly reduce bandwidth.  (MD5- or checksum-based scheme for E   verifying the results -- with the assistance of computes and I/O on F   the remote server -- would obviously and potentially reduce the load   on the wire itself.)  G   There are "lesser" approaches available as well, which snapshot files H   and/or transactions across spindles and across a network, and that canH   allow the database server(s) to roll forward from the last checkpoint.H   This isn't as nice as mirroring or shadowing, but it can be easier andH   cheaper -- and it trades off the ability to hot-start during failures;I   some resynchronization effort and/or some restart-focused site-specific A   procedures are involved as part of a failure-restart operation.      N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------   Date: 3 Nov 2003 18:28:30 GMT 1 From: JONESD@er6.eng.ohio-state.edu (David Jones) 3 Subject: Re: Passing a socket to a detached process : Message-ID: <bo66oe$kpl$1@charm.magnus.acs.ohio-state.edu>  2 In message <tqwpb.8266$bo3.1655@news.cpqcorp.net>,,   "Greg Zymbaluk" <greg@no.spam.org> writes:K >One thing I noticed is that after the detached process exits the BG device H >is still there ($ sho dev BG5108:) even though the socket is closed. ($ >tcpip sho dev BG5108:= >                 %TCPIP-W-NODEVSOCK, device_socket not found J >Is there something else I need to do in the parent process to make the BGM >device go away? I tried closing the socket or deassigning the channel in the L >parent, but that seems to completely turn off the socket so the child can't >do anything with it.   L You need to deassign the channel in the parent, but only after the child hasM assigned its own channel to the device.  You will have to refine your handoff N protocol so that the parent receives a confirmation when the child assigns theF channel.  I know this works because I do it this way in some of my own network servers.    < David L. Jones               |      Phone:    (614) 292-6929- Ohio State University        |      Internet: L 140 W. 19th St. Rm. 231a     |               jonesd@er6s1.eng.ohio-state.edu: Columbus, OH 43210           |               vman+@osu.edu  1 Disclaimer: I'm looking for marbles all day long.    ------------------------------  # Date: Mon, 03 Nov 2003 17:40:41 GMT ( From: "Greg Zymbaluk" <greg@no.spam.org>/ Subject: Passing a socket to a detached process 2 Message-ID: <tqwpb.8266$bo3.1655@news.cpqcorp.net>  E I'm implementing an application on VMS using sockets. The application C consists of a process which creates a server socket and listens for L connections. When a socket connection arrives the "parent" process does someD initial communication with the connecting client and then if furtherL processing is needed it launches a detached process and passes the socket to9 that "child" for continued communication with the client.   L I have implemented the application and would like some advice on if there isK a better way to do this or if I am leaving anything out. Here is how I have 
 done this:   "Parent" process: L 1. Create a permanent mailbox to be used for communicating with the detached child process.K 2. Create a server socket using system service calls (sys$qiow) UCX$C_SHARE < is set on the socket to enable sharing with other processes.4 3. Accept socket from client, do some initial stuff.B 4. If needed, launch a detached subprocess with sys$creprc and the PRC$M_DETACH option.L 5. Write the BG device name to the mailbox and wait for the detached process to read it. G 6. After mailbox is read, go back to listen for next client connection.    "Child" process.7 1. create/ connect to mailbox created by parent process , 2. read the BG device name from the mailbox.: 3. Use sys$assign to assign a channel to that device name.7 4. read/write from/to the socket to service the client. E 5. when done with the client, shutdown the socket (call sys$qiow with E IO$_DEACCESS|IO$M_SHUTDOWN) then close the socket (call sys$qiow with = IO$_DEACCESS). Then deassign the I/O channel with sys$dassgn.  6. Exit the program.  J One thing I noticed is that after the detached process exits the BG deviceG is still there ($ sho dev BG5108:) even though the socket is closed. ($  tcpip sho dev BG5108: <                  %TCPIP-W-NODEVSOCK, device_socket not foundI Is there something else I need to do in the parent process to make the BG L device go away? I tried closing the socket or deassigning the channel in theK parent, but that seems to completely turn off the socket so the child can't  do anything with it.  G I'm afraid that with the way things currently are those BG devices will < never go away as long as the parent process remains running.   Thanks in advance,   Greg   ------------------------------  * Date: Mon, 3 Nov 2003 15:12:10 +0000 (UTC), From: lewis@PROBE.mitre.org (Keith A. Lewis) Subject: Re: Show Memory/phys . Message-ID: <bo5r8a$r53$1@newslocal.mitre.org>   bhushann@hotmail.com (Bhushan Narkhede) writes in article <8a3b834.0311022257.340a36e3@posting.google.com> dated 2 Nov 2003 22:57:58 -0800: < >  I have one VMS system that shows "how Mem/phy" output as D >Physical Memory Usage (pages):     Total        Free      In Use   	 >Modified G >  Private Memory (352.00Mb)        45056       31130       12872        > 1054A >  Shared Memory (32.00Mb)           4096        4001          95   J That's interesting, I haven't seen "Private Memory" before.  What hardware are you getting this on?  A >Can any one guide me on how to obtain main memory for VMS system  >showing the- >show mem/phy output as in the first case????    Maybe this: & $ write sys$output f$getsyi("MEMSIZE")  8 Or if you're using a compiled language, see sys$getsyiw.  0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------  # Date: Mon, 03 Nov 2003 15:37:42 GMT " From:   VAXman-  @SendSpamHere.ORG Subject: Re: Show Memory/phys 0 Message-ID: <00A2857E.FC022F56@SendSpamHere.ORG>  ] In article <bo5r8a$r53$1@newslocal.mitre.org>, lewis@PROBE.mitre.org (Keith A. Lewis) writes:  >bhushann@hotmail.com (Bhushan Narkhede) writes in article <8a3b834.0311022257.340a36e3@posting.google.com> dated 2 Nov 2003 22:57:58 -0800:= >>  I have one VMS system that shows "how Mem/phy" output as  E >>Physical Memory Usage (pages):     Total        Free      In Use    
 >>ModifiedH >>  Private Memory (352.00Mb)        45056       31130       12872       >> 1054 B >>  Shared Memory (32.00Mb)           4096        4001          95 > K >That's interesting, I haven't seen "Private Memory" before.  What hardware  >are you getting this on?    Can you say, Galaxy?     --  L VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)COM             5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Mon, 03 Nov 2003 10:21:58 +0100 ( From: Andreas Davour <ante@update.uu.se>  Subject: Re: SIMH used with VMS?0 Message-ID: <cs9ad7dolhl.fsf@tempo.update.uu.se>  4 martin@radiogaga.harz.de (Martin Vorlaender) writes:  C > P.S.: Has anyone yet looked into porting the pcap library to VMS? B > (needed for SIMH to have Ethernet access, e.g. when running SIMH > under OpenVMS Alpha).   H Note that there are limitations with the pcap library. A better solutionG would be to hack simh to use tun/tap and get it to work on VMS also, if  it doesn't already.    /andreas   ------------------------------   Date: 3 Nov 2003 07:31:45 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)   Subject: Re: SIMH used with VMS?3 Message-ID: <VV7+beGu7Zry@eisner.encompasserve.org>   m In article <f30679fb.0311011603.8ee88e5@posting.google.com>, fabiopenvms@yahoo.com.br (Fabio Cardoso) writes:  > By the way ... > 6 > Why HP dont liberate MicroVMS ! I remember it was a 5 > version of  VMS which used to consume low memory !   >   +   Me think's your memory is shy a few bits.    ------------------------------   Date: 3 Nov 2003 07:13:17 -0800 . From: fabiopenvms@yahoo.com.br (Fabio Cardoso)  Subject: Re: SIMH used with VMS?= Message-ID: <f30679fb.0311030713.3fe85840@posting.google.com>   ` Roger Ivie <rivie@ridgenet.net> wrote in message news:<slrnbq96og.j1o.rivie@Stench.no.domain>...M > In article <f30679fb.0311011603.8ee88e5@posting.google.com>, Fabio Cardoso   > wrote:8 > > Why HP dont liberate MicroVMS ! I remember it was a 7 > > version of  VMS which used to consume low memory !   > I > I'm no expert, but IIRC the whole point of MicroVMS was to make an RX50 G > distribution possible. The pieces provided with MicroVMS are selected D > bits of normal VMS. For example, MicroVMS included support for theD > MicroVAX II, but omitted support for the VAX-11/780. It was just a7 > tailored distribution for a specific set of machines.  > J > Even tailored to support just a small number of machines, MicroVMS still( > made an impressive pile of floppies...    F What about a bootable CDROM with MicroVMS ?  Or an USB Pen Drive ? :-)1 Or maybe some ready demos of SIHM or Charon-VAX?     Regards    FC   ------------------------------  # Date: Mon, 03 Nov 2003 16:09:07 GMT 4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>  Subject: Re: SIMH used with VMS?0 Message-ID: <3FA67C96.435EEF81@blueyonder.co.uk>   Fabio Cardoso wrote: > b > Roger Ivie <rivie@ridgenet.net> wrote in message news:<slrnbq96og.j1o.rivie@Stench.no.domain>...N > > In article <f30679fb.0311011603.8ee88e5@posting.google.com>, Fabio Cardoso
 > > wrote:9 > > > Why HP dont liberate MicroVMS ! I remember it was a 8 > > > version of  VMS which used to consume low memory ! > > K > > I'm no expert, but IIRC the whole point of MicroVMS was to make an RX50 I > > distribution possible. The pieces provided with MicroVMS are selected F > > bits of normal VMS. For example, MicroVMS included support for theF > > MicroVAX II, but omitted support for the VAX-11/780. It was just a9 > > tailored distribution for a specific set of machines.  > > L > > Even tailored to support just a small number of machines, MicroVMS still* > > made an impressive pile of floppies... > H > What about a bootable CDROM with MicroVMS ?  Or an USB Pen Drive ? :-)2 > Or maybe some ready demos of SIHM or Charon-VAX? >     MicroVMS went away with VMS 5.0.  6 Do REALLY want to run VMS 4.x? No tcp/ip for starters.  # The MicroVMS docs were good though.    --   tim.llewellyn@blueyonder.co.uk   ------------------------------  # Date: Mon, 03 Nov 2003 13:35:59 GMT 4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>! Subject: Re: SIMH: used with VMS?2/ Message-ID: <3FA65914.1518224@blueyonder.co.uk>    Didier Morandi wrote:t >  > Robert Alan Byer wrote:A > F > > As the maintainter of SIMH for OpenVMS I can say that the SIMH VAX8 > > emulator runs OpenVMS pretty well on a fast machine. > Q > Super. So you can tell me where to start from. I have a Sim> prompt and a wholerP > bunch of text displayed in response to the HELP command, but found nowhere how0 > to load my Hobbyst VAX/VMS CD and install VMS. > 	 > Thanks,  >   
 Sim> Boot cpue  $ is I believe the voodoo you require.  2 Then you should get something that looks familiar.  3 This assumes you have got the microcode loaded OK. s  [ You need to attach at least one hard drive and your CD drive/image to start installing VMS.p  G Agreed the docs are wriiten by experts for experts, no simple "how to".e    d   -- s tim.llewellyn@blueyonder.co.uk   ------------------------------  % Date: Mon, 03 Nov 2003 14:47:57 +0100S" From: Didier Morandi <no@spam.com>! Subject: Re: SIMH: used with VMS?i4 Message-ID: <3fa65c97$0$13290$626a54ce@news.free.fr>  N KA655.bin is missing, although the HOWTO sez "is included in the package". In Q the Ouindoze "package" I found pdp11.exe and vax.exe and nothing else (and I had e4 to figure out how to find the packet.dll library...)   D.   Roger Ivie wrote:c  D > Apparently, the magic is to attach the CDROM -r. There's already aJ > HOWTO over at http://www.wherry.com/gadgets/retrocomputing/vax-simh.html   ------------------------------  $ Date: Mon, 3 Nov 2003 15:16:13 +0100- From: "Martin Vorlaender" <mv@pdv-systeme.de>s! Subject: Re: SIMH: used with VMS?e9 Message-ID: <bo5nvj$18789i$1@ID-56200.news.uni-berlin.de>s   Didier Morandi wrote:r > Roger Ivie wrote:s > E >> Apparently, the magic is to attach the CDROM -r. There's already ai >> HOWTO over at= >> http://www.wherry.com/gadgets/retrocomputing/vax-simh.htmlT >aB > KA655.bin is missing, although the HOWTO sez "is included in theG > package". In the Ouindoze "package" I found pdp11.exe and vax.exe andmB > nothing else (and I had to figure out how to find the packet.dll > library...)   A You realized, of course, that the URL Roger mentioned refers to a B Linux system; there "the package" - of course also - refers to theB source distribution. There is no binary Linux distribution, AFAIK.   cu,d   Martin -- XF   OpenVMS:                | Martin Vorlaender  |  VMS & WNT programmer3    The operating system   | work: mv@pdv-systeme.de F    God runs the           |   http://www.pdv-systeme.de/users/martinv/:    earth simulation on.   | home: martin@radiogaga.harz.de   ------------------------------  # Date: Mon, 03 Nov 2003 14:46:05 GMTr4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>! Subject: Re: SIMH: used with VMS?S0 Message-ID: <3FA6696A.90CD16AC@blueyonder.co.uk>   Didier Morandi wrote:0 > O > KA655.bin is missing, although the HOWTO sez "is included in the package". InXR > the Ouindoze "package" I found pdp11.exe and vax.exe and nothing else (and I had6 > to figure out how to find the packet.dll library...) >  > D. >   B I remember some grief with ka655.bin, I had to copy it from ts10 I% think (or maybe the other way round).b  3 I can email u a copy if u like. Its only 128 kbyte.    regardss        -- d tim.llewellyn@blueyonder.co.uk   ------------------------------  % Date: Mon, 03 Nov 2003 16:08:13 +0100a" From: Didier Morandi <no@spam.com>! Subject: Re: SIMH: used with VMS? 4 Message-ID: <3fa66f68$0$13303$626a54ce@news.free.fr>   please.u' Zip it first to protect its checksum...e   Thanks,s   D.G firstnamelastname English for "chapeau" without the "h" nerim point net? :-)o   Tim Llewellyn wrote:   >  > Didier Morandi wrote:y > O >>KA655.bin is missing, although the HOWTO sez "is included in the package". IndR >>the Ouindoze "package" I found pdp11.exe and vax.exe and nothing else (and I had6 >>to figure out how to find the packet.dll library...) >> >>D. >> >  > D > I remember some grief with ka655.bin, I had to copy it from ts10 I' > think (or maybe the other way round).M > 5 > I can email u a copy if u like. Its only 128 kbyte.i   ------------------------------  # Date: Mon, 03 Nov 2003 16:07:41 GMT 4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>! Subject: Re: SIMH: used with VMS? 0 Message-ID: <3FA67C41.A7EC9F33@blueyonder.co.uk>   Didier Morandi wrote:  > 	 > please. ) > Zip it first to protect its checksum...  > D OK, done, should be with u soon assuming I decoded you email address
 correctly.     p   -- d tim.llewellyn@blueyonder.co.uk   ------------------------------  % Date: Mon, 03 Nov 2003 17:13:13 +0100M" From: Didier Morandi <no@spam.com>! Subject: Re: SIMH: used with VMS? 4 Message-ID: <3fa67ea5$0$13287$626a54ce@news.free.fr>  Q you did... now I need to find that page where I will understand how to configure h my VMS RL02 from :-)   D.   Tim Llewellyn wrote:   >  > Didier Morandi wrote:r > 	 >>please. ) >>Zip it first to protect its checksum...t >> > F > OK, done, should be with u soon assuming I decoded you email address > correctly.   ------------------------------  # Date: Mon, 03 Nov 2003 16:19:39 GMTe4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>! Subject: Re: SIMH: used with VMS? 0 Message-ID: <3FA67F0A.73469E16@blueyonder.co.uk>   Didier Morandi wrote:- > R > you did... now I need to find that page where I will understand how to configure > my VMS RL02 from :-) >   C A couple of sample .INI files should be in your mailbox soon. Stick? with RA disks I recommend:-).    regards,   > D. >  > Tim Llewellyn wrote: >  > >? > > Didier Morandi wrote:o > >e > >>please.e+ > >>Zip it first to protect its checksum...o > >> > >vH > > OK, done, should be with u soon assuming I decoded you email address > > correctly.   -- r tim.llewellyn@blueyonder.co.uk   ------------------------------  % Date: Mon, 03 Nov 2003 17:46:21 +01001" From: Didier Morandi <no@spam.com>! Subject: Re: SIMH: used with VMS?t4 Message-ID: <3fa68668$0$13298$626a54ce@news.free.fr>   Thanks for the files.mN SUCCESS ! http://www.didiermorandi.com/vms/SimHboot.jpg (671x340, 45369 bytes)- I even succeeded to set host to my VMS box...dF Let's go eVAX now, when I terminate the translation of the newsletter.   D.   Tim Llewellyn wrote:  M > A couple of sample .INI files should be in your mailbox soon. Stick with RAe > disks I recommend:-).    ------------------------------  * Date: Mon, 3 Nov 2003 07:12:40 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)0 Subject: Re: simple advice needed on VAX upgrade$ Message-ID: <bo4v58$gt4$1@online.de>  ; In article <3FA5B27E.C6EEFAE2@fsi.net>, "David J. Dachtera"   <djesys.nospam@fsi.net> writes:   G > > > > It talks about booting from [SYSF] during the upgrade.  Is thisIG > > > > something which is invoked automatically somehow by the upgrade ? > > > > procedure, or do I need to specify this at the console.n > > >iQ > > > Either way, the upgrade procedure should be telling you something like "the-Q > > > upgrade procedure will now shutdown your system and reboot XXXX (where XXXX S > > > may be "automatically from a different root" or "manually, you'll have to askl( > > > the system to boot from root "F"). > > H > > So I just have to @SYS$UPDATE:VMSINSTAL and follow the instructions. > > Sounds easy enough.o > >  >  > >>> B/R5:F0000000   I Right, I know how to boot from a given root.  The question was whether I sG should expect to have to do this, or expect the VAX to boot from where -G it needs to automatically.  If the former, should I change the default rF boot flags for the duration of the upgrade (certainly needed if there F are automatic reboots from the default root which should be [SYSF] at D some phase of the upgrade) or not (which assumes I'll always have a / chance to boot manually at the console prompt).m   ------------------------------  % Date: Mon, 03 Nov 2003 11:53:17 +0000b- From: John Laird <nospam@laird-towers.org.uk>a0 Subject: Re: simple advice needed on VAX upgrade8 Message-ID: <f6gcqvglosagvlmqknrhqkefkpt3eohman@4ax.com>  J On Sun, 2 Nov 2003 08:33:37 -0500, "Chris Moore" <just@my.twocents> wrote:  J >On the original query, almost all VAX h'ware will boot automatically fromG >the correct roots at the various stages, though the warning message isi >displayed regardless.  H Having OPCCRASH set some console setting to allow an automatic reboot onD another root would be really neat, but I've never seen it done ?  MyL experience is that the b/[r5:]f0000000 command is needed.  I always rememberD it as "eff-ty million", to avoid getting the number of zeroes wrong.   -- i John Mail john rather than nospam...    ------------------------------  $ Date: Mon, 3 Nov 2003 09:53:13 -0500 From: norm.raphael@metso.com: Subject: Re: Speaking of Cincom (Was Re: World Wide Wake])Q Message-ID: <OFD5189237.B919AC79-ON85256DD3.00519265-85256DD3.0051748A@metso.com>j   Hi Paul,H Sorry about that attack.  We are victims of  corporate NotesMail, which=   MIMEs without notice.cH Cincom is still privately held by Tom Nies, and should be listed as an = HP partner. -Norme  B From:  Paul Sture <nospam@sture.homeip.net> on 11/02/2003 03:38 AM  6 Please respond to Paul Sture <nospam@sture.homeip.net>       To:    Info-VAX@Mvb.Saic.Com cc:b  = Subject:    Re: Speaking of Cincom (Was Re: World Wide Wake])i     norm.raphael@metso.com wrote:p  ( Aaargh, you had an attack of HTML there!  , >This is a multipart message in MIME format.' >--=3D_alternative 0000BB1185256DD2_=3Ds1 >Content-Type: text/plain; charset=3D"ISO-8859-1"d, >Content-Transfer-Encoding: quoted-printable     > ) > Funny you should mention them, Paul....  >hE > From:        "Cincom Systems" <owner-pr_english@lm01.cincom.com> ono > 10/30/2003 01:47 PM  >aF > Subject:        Cincom's 2003 Fiscal ROIC Performance Breaks 35-Year > Record >i= > Cincom's 2003 Fiscal ROIC Performance Breaks 35-Year Recordl >nD >     Cincom's  ROIC of 106.8 percent was five times better than itsH >  previous record-breaking 2002 performance. No other software company=  - >                            even came close.N >b@ > CINCINNATI, Ohio - October 29, 2003 - Cincom announced today aE > record-breaking 106.8% Return On Invested Capital (ROIC) for fiscal- > year 2003. >-H > ROIC cuts past the glamour of both profit and ROI figures to paint an=  C > unvarnished picture of how well the company is run. It is the rawbB > measure of profits generated relative to the capital employed to > generate that profit.m >nA >    * Cincom's 106.8% ROIC is 2.6 times greater than that of theuH >      industry-standard bearer Oracle=AE (ORCL), whose numbers for the=  # >      same period were just 40.8%.r >dB > Cincom achieved this remarkable feat despite the challenging andE > turbulent economic climate. The tumultuous technology sector of theyF > economy in the last couple of years had most technology vendors justH > trying to "survive."  Cincom not only survived, but thrived: calendar=  < > 2001 and 2002 were its best two years in a row, ever (see: > www.cincom.com/pr/financial).i >aB > Cincom's CEO Tom Nies attributes the ongoing success to Cincom'sH > "high-value, low-cost, and rapid ROI" offerings. "In the 'irrationall= yv@ > exuberant' economy of the last several years, few seemed to beH > concerned about long-delayed and minimized ROI, immense implementatio= npH > costs, and long development cycles where lots could go wrong - and to= oe? > often did. Customers are returning to more rational and soundcD > decision-making processes that focus much more on cost, value, and > rapid ROI," says Nies. > H > He continues, "We will continue to follow Cincom's golden rule: 'Help=  F > our customers grow their businesses faster and more profitably, withC > far less upfront investment, much less risk, and much greater andr< > quicker ROI.' Simple rule. It's worked well for 35 years." >hH > Cincom's operating success is also a testimony to the effectiveness o= fuH > CEO Tom Nies' approach to fiscal management that has, over the last 3= 5hC > years, steered Cincom to its present status as one of the leadingoE > software companies, now ranked in the top 5 percent of all softwareeH > companies in the world.  Among these larger software firms, during th= esD > fiscal years 2001-2002, Cincom led the industry in both ROI and in! > earnings-per-share improvement.j >  > About CincomH > Cincom, the world's most experienced software company, builds, sells,=  F > and supports software for 1) data access and integration, 2) processB > automation, 3) manufacturing business solutions, and 4) businessE > communications. Thousands of major clients around the world rely onnD > Cincom's 35 years of experience to provide innovative solutions toB > some of their toughest and most complex business problems.  HighA > quality, rapid implementation and excellent service give Cincom E > clients greater productivity, faster speed-to-market, and a quicker6 > Return on Investment (ROI).r > H > For more information about Cincom products and services, contact us a= teB > 1-800-2CINCOM (US), send e-mail to info@cincom.com, or visit the& > company's website at www.cincom.com. >r > Media Contacts:t > Steve Kayser > Cincom Systems, Inc. > 513-612-2348 > skayser@cincom.com >-
 > Donna Hedge- > Cincom Systems, Inc. > 513-612-2305 > dhedge@cincom.com  >e  H Very interesting. Thanks for that. I had assumed that they had probably=  H gone the way of many others and got bought out somewhere along the way.=        H > From:        Paul Sture <nospam@sture.homeip.net> on 11/01/2003 02:28=   > AM >o8 > Please respond to Paul Sture <nospam@sture.homeip.net> > " > To:        Info-VAX@Mvb.Saic.Com > cc:s >s& > Subject:        Re: World Wide Wake] >  > Bill Todd wrote:A >  A variant of Cincom's TOTAL (no shame there - DEC's first DBMSlE > > was a reworked version of Cullinane's IDMS), it was only somewhateC > > heavier-weight than RMS but provided real database capabilitiesn
 > without theoF > > overhead of other hierarchical or network DBMSs of the time (or of > theira > > relational successors).w > >mC > > If Cincom still sells TOTAL it would presumably be a reasonable  > migrationlA > > alternative from IMAGE.  There are certainly cases in which am > relational > > database would not be. >iE > Holy mackerel Batman! I was just talking about TOTAL to a colleague  > whenE > we were in "reminicences mode" a couple of days ago. Way back when,o > thehH > company I worked for bought TOTAL, coupled with SHADOW (a lightwieght=  < > form of CICS), when IBM were trying to push DL/1 and CICS. >iH > One of the best IT courses I ever attended was the TOTAL one, given b= yS > at; > Cincom guy. Sod PowerPoint, those were the days when truet > professionalsyD > did it with carefully prepared flipcharts, notes and handouts, and > coulda3 > answer any technical question you cared to throw.s > E > And were prepared to admit that they didn't know the answer to someeE > questions, but rather than ignoring you, would be on the phone thatr" > evening to HQ to find an answer. >oB > Mind you, the DEC lady who ran an RSX course I attended about 18 > monthsH > later came an extremely close second. Same deal really, both lecturer= s ) > beat IBM's best of the time hands down.e >r > -- > Paul Sture   --   --
 Paul Sture   =    ------------------------------  % Date: Mon, 03 Nov 2003 17:01:22 +0000 * From: Nic Clews <sendspamhere@[127.0.0.1]>: Subject: Re: Speaking of Cincom (Was Re: World Wide Wake])' Message-ID: <bo61jd$3rm$1@lore.csc.com>7   norm.raphael@metso.com wrote:  > K > Cincom is still privately held by Tom Nies, and should be listed as an HPt
 > partner.  E I've used the C:M (Control:Manufacturing) database using SUPRA on VMS  and it's fairly bombproof.  G I've studied and used it from a performance point of view, and the filetD layouts (multiple primary and related) can lend themselves to making quite a neat database system.-  E I understand you can also use the Oracle database engine, but I was ai DBA for a while using SUPRA.   -- u? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencesb nclews at csc dot com1   ------------------------------  % Date: Sun, 02 Nov 2003 17:40:54 -0600y5 From: "Martin P.J. Zinser" <zinser@zinser.no-ip.info>0$ Subject: Re: Text library extensions: Message-ID: <bo44m6$16mqr1$2@ID-209632.news.uni-berlin.de>   JF Mezei wrote:M5 > OK, I know that text libraries are by default, .TLBa? > I know that there is also .HLB which are also text libraries.  > B > Are other extensions that are commonly used for text libraries ?   MLB for Macro library?   ------------------------------   Date: 3 Nov 2003 06:05:51 -0800-- From: djesys@earthlink.net (David J Dachtera)n$ Subject: Re: Text library extensions= Message-ID: <66a00d01.0311030605.777a215b@posting.google.com>s  Z JF Mezei <jfmezei.spamnot@istop.com> wrote in message news:<3FA5CB5C.84501C6@istop.com>... > "David J. Dachtera" wrote:G > > .HLB is not necessarily purely a "text library". The library formataK > > differs little; however, the extension has meaning to the Librarian andD > > the LBR$ RTL routines. > G > Yes, but ou can still extract the text contents of a module. Right ? n  A /EXTRACT is a documented and supported qualifier of the LIBRARIANt1 command. I would expect it to work as documented.P  D Have you looked at a .HLP file? (.HLP is the default input extension/ when adding modules to a /HELP library (.HLB)).n   -- David J Dachtera dba DJE Systemss   ------------------------------  * Date: Mon, 3 Nov 2003 07:14:24 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)6 Subject: Re: To foreign or verb ? That is the question$ Message-ID: <bo4v8g$gt4$2@online.de>  D In article <bo4his$12i$1@newslocal.mitre.org>, lewis@PROBE.mitre.org (Keith A. Lewis) writes: m  N > If it's a VMS-forever program, my personal preference is verbs.  Verbs allowO > abbreviation where it's smart.  That's about the only advantage to the user. oL > I can think of lots of other reasons why it's better for the programmer to% > use verbs, but you didn't ask that.   H It has been pointed out in this thread that the DCL-verb-style look and   feel is possible without a verb.   ------------------------------  % Date: Mon, 03 Nov 2003 08:51:29 +0100o" From: Didier Morandi <no@spam.com>6 Subject: Re: To foreign or verb ? That is the question4 Message-ID: <3fa6090b$0$13285$626a54ce@news.free.fr>   JF Mezei wrote:s   > Didier Morandi wrote:C > ; >>may I suggest to see how PCSI could answer your question?m >  > O > Until you can specify multiple directories and devices as target for software L > PRODUCT remains useless to me.  I don't want to have freeware dropped intoN > sys$system or other system directories without making a very explicit prompt' > to the system manager during install.o > H > PRODUCT may be fine for Digital procucts that are coordinated with VMSK > engineers so the later will know what to expect to find in the vms$common 5 > structure, but not for other unknown aopplications.o   PRODUCTA  
    INSTALL        /DESTINATION5  4            /DESTINATION=device-name:[directory-name]  D         Specifies a location where the utility will install softwareD         product files. If you omit the device name, the utility usesH         your current default device. If you omit the directory name, theI         utility uses the [VMS$COMMON] directory as the default directory.   F         If you do not use this qualifier to specify a destination, theH         utility installs the software in the location defined by logicalG         name PCSI$DESTINATION. If this logical name is not defined, the H         utility installs the software in SYS$SYSDEVICE:[VMS$COMMON], the7         top-level directory for software product files.t   D.0 PS: "multiple directories destination"? Weird...   ------------------------------  % Date: Sun, 02 Nov 2003 17:37:47 -0600o5 From: "Martin P.J. Zinser" <zinser@zinser.no-ip.info>o6 Subject: Re: To foreign or verb ? That is the question: Message-ID: <bo44gc$16mqr1$1@ID-209632.news.uni-berlin.de>   VAXman- wrote:t > In article <bo38qu$17fk11$1@ID-209632.news.uni-berlin.de>, "Martin P.J. Zinser" <zinser@zinser.no-ip.info> writes: > {...snip...} > I >>- I prefer foreign commands, since a.) Most programs do not have a VMS iG >>CLI anyhow, b.) Using a CLD bloats the definition process c.) If you rH >>want to put it in DCLTABLES (which avoids problem b) you do need again >>privs. >  > G > DCL command syntax parsing does not require entries in DCLTABLES.  ItrH > is far easier to use the CLI$DCL_PARSE call and tables linked into theH > image.  Done this way, a foreign command can invoke the program and itH > can then treat the command line like a DCL command string with qualif-4 > iers, parameters, keywords, alternate syntax, etc. >   F So you still end up with a symbol defining the foreign command, not a F verb in this particular case. Oh, for the really adventurous there is B still DCL$PATH ;-) (And no, I never used this and do not like the  concept too much).   Greetings, Martinl   ------------------------------  # Date: Mon, 03 Nov 2003 10:45:58 GMT 5 From: rdeininger@mindspringdot.com (Robert Deininger)t6 Subject: Re: To foreign or verb ? That is the questionL Message-ID: <rdeininger-0311030546270001@user-uinj07o.dialup.mindspring.com>  2 In article <3FA479E9.6352DE4B@istop.com>, JF Mezei" <jfmezei.spamnot@istop.com> wrote:  M >What do system managers and VMS engineers prefer when it comes to installinge >freeware applications ? >lI >Are there guidelines for when an application should be invoked as a verb N >versus a foreign command that takes arguments in DCL style versus application% >that takes arguments in unix style ?i >yM >While I am at it, there really needs to be some standard established to makeiO >it easy to distribute software. To this end, if Digital/HP were to release thecK >tool that generates DCX_EXE files, it would be a great thing for VMS whichfI >would get its own distribution format, in the same vein as APPLE has itsuO >binhex or stuffit formats, Debian Linux has its own proprietary format etc. Oru) >is ZIP the official format now for VMS ?f  G I believe the DCX_EXE tool (DTSV, a.k.a. SPOOL) will be on Freeware V6.s   ------------------------------  # Date: Mon, 03 Nov 2003 13:12:15 GMT-" From:   VAXman-  @SendSpamHere.ORG6 Subject: Re: To foreign or verb ? That is the question0 Message-ID: <00A2856A.AA6FA218@SendSpamHere.ORG>  r In article <bo44gc$16mqr1$1@ID-209632.news.uni-berlin.de>, "Martin P.J. Zinser" <zinser@zinser.no-ip.info> writes: >VAXman- wrote:ou >> In article <bo38qu$17fk11$1@ID-209632.news.uni-berlin.de>, "Martin P.J. Zinser" <zinser@zinser.no-ip.info> writes:s >> {...snip...}P >> gJ >>>- I prefer foreign commands, since a.) Most programs do not have a VMS H >>>CLI anyhow, b.) Using a CLD bloats the definition process c.) If you I >>>want to put it in DCLTABLES (which avoids problem b) you do need againt	 >>>privs.c >> r >>  H >> DCL command syntax parsing does not require entries in DCLTABLES.  ItI >> is far easier to use the CLI$DCL_PARSE call and tables linked into theDI >> image.  Done this way, a foreign command can invoke the program and ittI >> can then treat the command line like a DCL command string with qualif-e5 >> iers, parameters, keywords, alternate syntax, etc.  >>  G >So you still end up with a symbol defining the foreign command, not a tG >verb in this particular case. Oh, for the really adventurous there is cC >still DCL$PATH ;-) (And no, I never used this and do not like the e >concept too much).r >c >Greetings, Martin >e   Define Verb THE_VERB!         Image           THE_IMAGE !         Cliflags        (Foreign)h     --  L VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)COM            95   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Mon, 03 Nov 2003 08:46:57 +0100u" From: Didier Morandi <no@spam.com>  Subject: Re: TS10, still around?4 Message-ID: <3fa607fb$0$13288$626a54ce@news.free.fr>  % Thank you Tim, nice to hear from you.  Keep on the good job :-)   D.   Timothy Stark wrote:  	 > Didier,j > K > Yes, I still am doing developments.  I am porting my TS10 emulator to C++w > language in progress.lE > Also, my latest snapshots are at ftp://ftp.firesword7.net/pub/ts10.s   ------------------------------  # Date: Mon, 03 Nov 2003 13:37:33 GMT.4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>  Subject: Re: TS10, still around?0 Message-ID: <3FA65973.B5CF2AF2@blueyonder.co.uk>   Didier Morandi wrote:n > N > Does anyone know if TS10 is still a project under development? Does not lookO > like, according to the SourceForge site http://sourceforge.net/projects/ts10/e > 0 > Does anyone know if VMS once booted with TS10?  T Yes, but (1) not easy/possible to build on Windows (I tried and failed, but did run F it on Linux as it had network support before Simh did) (2) development% seems to have fallen off as you note.e   regardst   >  > Tx,l >  > D.   -- $ tim.llewellyn@blueyonder.co.uk   ------------------------------  % Date: Mon, 03 Nov 2003 14:16:42 +0000m* From: Nic Clews <sendspamhere@[127.0.0.1]>Y Subject: Re: Veritas VMS client requested restores - solution found to dramatically  impr & Message-ID: <bo5nuk$f2$1@lore.csc.com>  
 Nobody wrote:  > L > Working on the premise that the issue is not the networks but inter-system> > buffer sizing, I ran an experiment on one of our test nodes. >  ... G > If you have any questions or concerns about this, please let me know.g >  > Thank you, >  > Andrew Roberta  C The documentation states that the best transfer rate is achieved byhF having the buffer sizes matching (128k is default) but the buffer sizeG on the VMS server's IP stack (default 32k) overrides that unless set ine the NET_BUFFER.SZ.  D Experimentation is explicitly allowed, but advice against values too1 large due to instability and diminishing returns.s  D I expect your server is probably at 128k, so I would think that yourG server is now at a disadvantage, and backing the size off to 128k wouldt0 be beneficial for saves as well as the restores.  D The speedup you've observed is relatively easy to understand, if theE server is pushing 128k blocks down to a "receive" buffer of just 32k,oF there's a lot of overhead in splitting down what the server sends, butB 128k into 131k goes very neatly. It's just you may have introduced< pushing 131k blocks into 128k blocks for the save operation.  
 Just my 0.02.i -- a? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencesv nclews at csc dot comT   ------------------------------  $ Date: Mon, 3 Nov 2003 22:21:45 +1100U From: "Antony Wardle" <remove_clothes_antony.wardle@_remove_clothers_optusnet.com.au> ( Subject: Re: [~OT] Last _production_ VAX; Message-ID: <3fa63a56$0$2393$afc38c87@news.optusnet.com.au>-  * wasn't the last new one in december 1999 ?   kiwi      3 "B Fisher" <none@antispamming.com> wrote in message 3 news:72dkb.54389$832.26759@twister.rdc-kc.rr.com...rH > Slightly off topic but:  Does anyone have an idea of when the last VAXL > was/is produced?  My recollection is that the last introduction was aroundH > 1992 or so, but I am more interested in the last _produced_.  We still haveF > a few but I suspect when they get repaired, they are refurbed boards >  > Just a curiosity, Thanks >p > scottt >f >t     --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.534 / Virus Database: 329 - Release Date: 31/10/2003g   ------------------------------   End of INFO-VAX 2003.610 ************************