1 INFO-VAX	Mon, 25 Sep 2000	Volume 2000 : Issue 537       Contents:' Re: Advanced OpenVMS training in Europe  Re: AlphaPC164 & decwindows ( Alphaserver 4100 + KGPSA (Fiber Channel) Re: Basic freeware CD problems* Re: BUT... (Re: Why are these files open?)* Re: BUT... (Re: Why are these files open?), CDD database no longer has its original name1 Re: Clustering VAXstation with dead serial driver  Re: Compaq VMS promo, Re: DCPS (was Re: OpenVMS features (future)) Re: Duplex Printing  elsa configuration problem Re: elsa configuration problem Re: elsa configuration problem/ Re: FILESERV@WKU: Updated VNCviewer for OpenVMS - Finding size of files inside a backup saveset 1 Re: Finding size of files inside a backup saveset - Finding size of files inside a backup saveset 1 Re: Finding size of files inside a backup saveset 1 Re: Finding size of files inside a backup saveset 1 Re: Finding size of files inside a backup saveset  FW: Suggested VMS new feature!" Re: FW: Suggested VMS new feature!' Re: Getting Compaq to advertise OpenVMS B How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectF Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirect Migrating from 7.1 to 7.2  Re: Migrating from 7.1 to 7.2  Re: Migrating from 7.1 to 7.2  Re: NTP with UCX 4.2 Re: Open File, or What? 6 Re: OpenVMS Systems as client to MS SQLserver  Server.6 Re: OpenVMS Systems as client to MS SQLserver  Server. OpenVMS Undelete Re: OpenVMS Undelete Re: OpenVMS Undelete Re: OSU scripts under Apache. Re: Pascal question: Can OUTPUT be [VOLATILE]?. Re: Pascal question: Can OUTPUT be [VOLATILE]?* Re: Porting Support (was Compaq VMS promo) Re: Press Release  Re: Press Release  Re: Press Release  Re: Press Release   Re: remote login using .com file  Re: remote login using .com file# Re: remote tailing of file with UCX 4 re: Replacing system disk, will licenses still work?4 Re: Replacing system disk, will licenses still work?! restart an Alpha ETN42-CA machine % Re: restart an Alpha ETN42-CA machine % Re: restart an Alpha ETN42-CA machine  RMS help required  Re: RMS help required  Re: RMS help required  Re: RMS help required $ Seeking OpenVMS Freeware Submissions( re: Seeking OpenVMS Freeware Submissions1 Re: Son of BUT... (Re: Why are these files open?)  Re: SRAM Corruption Thread Re: SRAM Corruption Thread Re: SRAM Corruption Thread Re: SRAM Corruption Thread Re: SRAM Corruption Thread Re: SRAM Corruption Thread Re: SRAM Corruption Thread Re: SRAM Corruption Thread Re: Suggested VMS new feature! Re: Suggested VMS new feature! Re: Suggested VMS new feature! Re: Suggested VMS new feature! Re: Suggested VMS new feature! Re: Suggested VMS new feature! Re: Suggested VMS new feature! Re: Suggested VMS new feature!! Re: Sun Hardware problems persist ! Re: Sun Hardware problems persist M Re: Top "F" Reasons OpenVMS isn't going to Die  (WAS:Re: [VMS] VAX is  dying) P Re: Top "F" Reasons OpenVMS isn't going to Die  (WAS:Re: [VMS] VAX isdying)dying, UNDELET (Was Re: Suggested VMS new feature!)0 Re: UNDELET (Was Re: Suggested VMS new feature!)0 Re: UNDELET (Was Re: Suggested VMS new feature!)0 Re: UNDELET (Was Re: Suggested VMS new feature!) Re: VMS and UNIX Re: VMS and UNIX
 Volume set Re: Volume set Re: Volume set Re: Volume set Re: Volume set Re: Volume set( Why is it so hard to port things to VMS?, Re: Why is it so hard to port things to VMS?, Re: Why is it so hard to port things to VMS?, Re: Why is it so hard to port things to VMS?, Re: Why is it so hard to port things to VMS?, Re: Why is it so hard to port things to VMS?, Re: Why is it so hard to port things to VMS?, Re: Why is it so hard to port things to VMS? Re: [JAVA] Adobe Acrobat Viewer   F ----------------------------------------------------------------------  % Date: Mon, 25 Sep 2000 17:59:19 +0200 0 From: Didier Morandi <Didier.Morandi@Easynet.fr>0 Subject: Re: Advanced OpenVMS training in Europe* Message-ID: <39CF7657.D233AE57@Easynet.fr>   Didier Morandi wrote:   A > Where can these course be attended, please? (in UK probably :-)   2 The answer is: Global Knowledge network in Europe.   D.   ------------------------------  + Date: Mon, 25 Sep 2000 13:14:22 +0000 (UTC) * From: Osmo Kujala <kujala@kanto.cc.jyu.fi>$ Subject: Re: AlphaPC164 & decwindows, Message-ID: <8qnj3e$e2u$1@mordred.cc.jyu.fi>  * yyyc186.illegaltospam_@flashcom.net wrote:  J > What autoconfig command did you use to get Decwindows to see the device?   $ mcr sysman io autoconfig   /OK    ------------------------------  % Date: Mon, 25 Sep 2000 09:34:15 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.br 1 Subject: Alphaserver 4100 + KGPSA (Fiber Channel) L Message-ID: <OF1F07811F.78E37B58-ON83256965.0044A56E@ep-bc.petrobras.com.br>  H Do you know if the Alphaserver 4100 (OpenVMS 7.2 and 7.1-1H2) works fine withC the KGPSA adapter. We will migrate our storage from a Storage Works 	 (nowadays  SCSI) to an EMC array (FC).    Regards,  
 Fabio Cardoso  Analista de Suporte ( Compaq / Petrobras - Macae - RJ - Brasil   ------------------------------   Date: 25 Sep 2000 14:17:58 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)' Subject: Re: Basic freeware CD problems 6 Message-ID: <8qnmqm$422$2@mailint03.im.hou.compaq.com>  \ In article <8qh0mq$9so$1@news1.Radix.Net>, Thomas Dickey <dickey@saltmine.radix.net> writes:4 :Hoff Hoffman <hoffman@xdelta.zko.dec.nospam> wrote: : K :>   Freeware submissions for the next CD-ROM set are welcome.  Please see:  : J :I don't see the deadline (as usual I would like to submit some stuff, but/ :don't have a VAX handy to test-build/package).   8   The OpenVMS Freeware Submission deadline (for V5) is:        27-Nov-2000   H   I have just posted a note seeking new and updated submissions for the ;   next OpenVMS Freeware.  Submission details available via:   3     http://www.openvms.compaq.com/openvms/freeware/   G   If you know of new versions or new tools -- but are NOT the person or E   group responsible -- please let know know and I will work to secure I   permission to include the update or the new tool.   But I need to know  2   that the update or the tool even exists, first!)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 25 Sep 2000 16:24:28 +0100 - From: Tim Llewellyn <tim.llewellyn@bbc.co.uk> 3 Subject: Re: BUT... (Re: Why are these files open?) ) Message-ID: <39CF6E2C.3F5BA395@bbc.co.uk>    bill robertson wrote:   0 > So, how do I get rid of those INSTALL entries? >  > I tried INSTALL DELETE:  > 6 > $ install del dka200:[RLS_MTS_45.EXE]MENU_SIGNON.EXEW > %INSTALL-W-FAIL, failed to REMOVE entry for DISK$DEV1:<RLS_MTS_45.EXE>MENU_SIGNON.EXE 1 > -INSTALL-E-NOKFEFND, Known File Entry not found  > , > Well, let's replace it and then delete it: >  > $ copy menu_signon.exe *.*;1_ > %COPY-W-HIGHVER, higher version of           DKA200:[RLS_MTS_45.EXE]MENU_SIGNON.EXE;1 already  > exists9 > $ install repl dka200:[RLS_MTS_45.EXE]MENU_SIGNON.EXE;1 E > %INSTALL-W-NOPREV, no previous entry exists - new entry created for * > DKA200:[RLS_MTS_45.EXE]MENU_SIGNON.EXE;16 > $ install del dka200:[RLS_MTS_45.EXE]MENU_SIGNON.EXE >  > But: >  > $ sho dev/fil dka200B > Files accessed on device SFSC$DKA200: on 22-SEP-2000 11:29:48.54 >	% > Process name      PID     File name:0 >                 00000000  [000000]INDEXF.SYS;1= >                 00000000  [RLS_MTS_45.EXE]MENU_SIGNON.EXE;1)= >                 00000000  [RLS_MTS_45.EXE]MENU_SELECT.EXE;1 	 > ...etc.) >eD > Apparently, INSTALL has a list that pegs to the original File IDs. > Can I clean this out via DCL?V  E Are you SURE the disk is not mounted on another cluster member, whicheN also has the image installed? I mean, when you reaplace the image with installK it says no previous entry exists (on the node the command was executed on),W' and once it is installed it deletes OK.n       --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uk   A I speak for myself only and my views in no way represent those ofd MedAS or the BBC.n   ------------------------------  % Date: Mon, 25 Sep 2000 10:08:25 -0700M& From: bill robertson <wcr@pacbell.net>3 Subject: Re: BUT... (Re: Why are these files open?) + Message-ID: <39CF8689.62865CA2@pacbell.net>n  c I neglected to mention that this machine is not part of a cluster.  Thanks for the thought, though.    -billt   Tim Llewellyn wrote:   > bill robertson wrote:n > 2 > > So, how do I get rid of those INSTALL entries? > (yadayadayada)...LF > > Apparently, INSTALL has a list that pegs to the original File IDs.! > > Can I clean this out via DCL?o >oG > Are you SURE the disk is not mounted on another cluster member, whichcP > also has the image installed? I mean, when you reaplace the image with installM > it says no previous entry exists (on the node the command was executed on),n) > and once it is installed it deletes OK.e >o > --8 > Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project2 > MedAS at the BBC, Whiteladies Road, Bristol, UK.C > Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukr >3C > I speak for myself only and my views in no way represent those of1 > MedAS or the BBC.n   ------------------------------  # Date: Mon, 25 Sep 2000 17:43:58 GMTp' From: Paul <paul_clark0308@my-deja.com>S5 Subject: CDD database no longer has its original nameS( Message-ID: <8qo2sn$kc$1@nnrp1.deja.com>  ( Can anyone help me with this CDD errror.  
 CDO> show all , %CDO-E-ERRDIRE, error displaying a directory4 -CDD-F-NOATTACH, unable to attach to dictionary data/ -RDB-F-DB_CORRUPT, database filename is corrupte -RDMS-F-BADROOTMATCH, root filesF "$1$DIA26:[DIR.DIR]CDD$DATABASE.RDB;1" no longer has its original name% "$1$DIA9:[DIR.DIR]CDD$DATABASE.RDB;1"t CDO>  C I'm in this situation because I performed a /image restore to a new?4 machine and the physical device names are different.  / Is there a utility to change the original name?n   Thanks in advance.    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  % Date: Mon, 25 Sep 2000 11:43:04 -0400e2 From: rdeininger@mindspring.com (Robert Deininger): Subject: Re: Clustering VAXstation with dead serial driverL Message-ID: <rdeininger-2509001143050001@user-2ive73h.dialup.mindspring.com>  W In article <G1EK34.AnK@world.std.com>, atlas@world.std.com (Alexander R Svirsky) wrote:r  G > I would like to cluster a VAXstation 3100 M30 with an apparently dead G > serial port driver.  Problem symptoms in serial console configuration  > include the following:     > Questions - K > How do I find the ethernet ID if I can't read test output on the console?sJ > Might there be a sticker on the board somewhere?  Also, before I heat upJ > my iron, do these symptoms describe a problem that may not be present in > the serial driver?    F I _think_ (no time to try it now) that you can set up a MOP bootserverO on your other node and it will tell you the ethernet address.  When you attempt)B an ethernet boot on the sick node, it sends some sort of multicastF request, and every MOP server within range will start to respond.  TheE the ethernet address of the requestor is known to the server, it willsE send the correct file over the net, otherwise it will end the service J request.  But the server logs an OPCOM message in either case, and I think; the message includes the ethernet address of the requestor.    -- o Robert Deininger rdeininger@mindspring.como   ------------------------------  % Date: Mon, 25 Sep 2000 14:14:37 +0000 - From: SysAdmin <djesys.nospam@fsi.net.mapson>e Subject: Re: Compaq VMS promon. Message-ID: <39CF5DCD.EF76F2B7@fsi.net.mapson>   Bill Gunshannon wrote: > / > In article <39C980B4.F6DDDC25@earthlink.net>, < >  "David J. Dachtera" <djesys.nospam@earthlink.net> writes: > |> Chris Scheers wrote:c > |> [snip]e/ > |> > Don't schools teach programming anymore?5 > |>; > |> My step daughter corrected me about this last weekend:s > |>J > |> "Computer 'Science'" students learn how to use "canned" applications. > |>E > |> "Computer Engineering" students learn hardware, architecture and2- > |> programming, including o.s. development.j >   > Not in any acredited program.   1 Contact University of Illinois, Champaign, IL USA$  $ > The only place we teach how to useF > canned applications is in Computer Literacy class.  A freshman level > course that everyone takes.   D ...and what about CAD students? ...other Computer Graphics students? ...office automation?0  H Apparently, these also fall under the "Computer Science" heading in some way.  E ...which probably explains why most of the network and e-mail folks I6: must deal with are lacking fundamental knowledge of serial communications and such.  % > I sure hope your not paying for herS
 > education. i  D Fortunately, no. Her birth father shoulders a chunk of it. Also, her> grand-parents have been giving her Cert.'s of Deposit for {her? birthday,Christmas,etc.} for many years. She's got her own bux.-  : > I would expect better than that from our local Community
 > College.   U of I, really.f   David J. Dachteras   ------------------------------   Date: 25 Sep 2000 14:55:26 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)5 Subject: Re: DCPS (was Re: OpenVMS features (future)):6 Message-ID: <8qnp0u$4av$2@mailint03.im.hou.compaq.com>  Z In article <8qkb8o$flb$1@aquila.news.mdx.ac.uk>, david20@alpha2.mdx.ac.uk (D.Webb) writes:h :In article <39cda2b4$0$30002$2c3edae7@news.voyager.net>, Jack Patteeuw <jjpatteeuw@voyager.net> writes: ..> :>What we really need is support for all of the "user" featureJ :>(single/double side, number up, ANSI, Postscript, forms, etc.) supportedG :>on the latest version of "popular" printer such as HP, Lexmark, etc. l$ :>Notice I said "latest versions".    L   HP apparently has multiple engineering teams replacing or upgrading their L   printers quite regularly.  Some of the upgraded and new printers are quiteJ   clearly relatively compatible with earlier ones, some incompatible, and I   the addition of new features in new printers also obviously means that -K   new knobs and new tweaks are needed if the features are to be used.  (HP  I   is not alone in the frequency of their device churn, of course -- this mH   product churn is prevalent throughout the entire computer and computerJ   peripheral industries, and the churn is clearly continously increasing.)  H :What I want is either for DCPS to support UCXs LPD or a facility to be F :added to VMS to easily provide printer-filters with standard filters D :text-postscript, pcl-postscript, etc  and number-up would be good.   +   DCPS Engineering is now (back) at Compaq.i  F   ps: I assume you mean a translator or a converter, and not a filter.   	--N  G   From my own experience, third-party printers are a black hole -- you  E   can throw as much time and effort and money at this engineering andeE   support as you want, and you still won't have all of what you need.   N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 25 Sep 2000 07:46:45 +0100N< From: John Macallister <j.macallister1@physics.oxford.ac.uk> Subject: Re: Duplex Printing7 Message-ID: <000925074645.ae4c@nplvms.physics.ox.ac.uk>    > > N > > Of course, even more paper can be saved by not printing many things in the > > R > I have a better idea!!  Install your favorite OpenVMS Web Server,  Pay me a week > orI > so worth of consulting (connectivity via interenet is least expensive)    T Almost all documents we use nowadays are already WWW-accessible. The Web does indeedS  provide a way of saving paper and other resources by not having multiple copies of U  documents. There's nothing VMS-specific about this: we make documents available from   VMS and many other systems.     John   ------------------------------  % Date: Mon, 25 Sep 2000 15:49:42 +0200 > From: "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr># Subject: elsa configuration problem 2 Message-ID: <8qnl72$d73$1@s1.read.news.oleane.net>  ) Here is a problem I got this week-end ... 5 New DS10 with factory installed 7.2-1 and Motif 1.2.5 - Elsa graphic adapter, with 21" Compaq Monitor    I added "     $ DECW$XSIZE_IN_PIXELS == 1280"     $ DECW$YSIZE_IN_PIXELS == 1024  in decw$private_server_setup.com  ? Since this modification, display is starting at one inch of the : left border of the screen. sounds like the frquency wasn't correctly set up.   1 I just tryed at my office with a DEC 21" monitor. ? The display is now like a barrel, with a larger top, bottom and  left margin (not right).  > What can I do to correct this and usr full 1280x1024 display ?   Cheers Jean-Franois Marchal  X9000 - LYON (FR)    ------------------------------   Date: 25 Sep 2000 15:48:16 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)' Subject: Re: elsa configuration problem , Message-ID: <8qns40$luq@gap.cco.caltech.edu>  s In article <8qnl72$d73$1@s1.read.news.oleane.net>, "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr> writes: * >Here is a problem I got this week-end ...6 >New DS10 with factory installed 7.2-1 and Motif 1.2.5. >Elsa graphic adapter, with 21" Compaq Monitor >  >I added# >    $ DECW$XSIZE_IN_PIXELS == 1280 # >    $ DECW$YSIZE_IN_PIXELS == 1024 ! >in decw$private_server_setup.com  > @ >Since this modification, display is starting at one inch of the; >left border of the screen. sounds like the frquency wasn't  >correctly set up. >   " Did you try adjusting the monitor?  2 >I just tryed at my office with a DEC 21" monitor.@ >The display is now like a barrel, with a larger top, bottom and >left margin (not right).  >   * That is also a monitor adjustment problem.  L Graphic card problems usually show up as things like "snow", missing lines, + flickering or blank displays, and the like.   ? >What can I do to correct this and usr full 1280x1024 display ?   H Adjust your monitor.  In the first case the image has not been centered.< In the second case you need to adjust the geometry settings.   Regards,   David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech     ------------------------------  % Date: Mon, 25 Sep 2000 18:29:10 +0200 > From: "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr>' Subject: Re: elsa configuration problem 2 Message-ID: <8qnugi$iob$1@s1.read.news.oleane.net>  B Thanks, but in both cases monitors are correctly setup for displayB with default DECwindows parameters ... If I change the monitor for: 1280x1024, the settings will be bad for default values ...? I must admit that correcting a monitor's gemetry is a real pain , when a user's changes have been done anyhow.  
 Jean-Franois     ? "David Mathog" <mathog@seqaxp.bio.caltech.edu> wrote in message & news:8qns40$luq@gap.cco.caltech.edu...J > Adjust your monitor.  In the first case the image has not been centered.> > In the second case you need to adjust the geometry settings. > 
 > Regards, >  > David Mathog > mathog@seqaxp.bio.caltech.edu @ > Manager, sequence analysis facility, biology division, Caltech   ------------------------------  # Date: Mon, 25 Sep 2000 09:39:33 GMT % From: A.Greig@virgin.net (Alan Greig) 8 Subject: Re: FILESERV@WKU: Updated VNCviewer for OpenVMS2 Message-ID: <39cf1cd4.1625076065@news.newsguy.com>  6 On Fri, 22 Sep 2000 23:19:22 -0500, "John E. Malmberg" <wb8tyw@qsl.net> wrote:   7 ><fabio_compaq#ep-bc.petrobras.com.br> wrote in message G >news:OFA8BCEB31.B77DCDEC-ON83256962.00579606@ep-bc.petrobras.com.br... B >> Would be much more interesting a Citrix ICA client for OpenVMS. > H >I have been told that there is a JAVA client that may run under OpenVMS >using Netscape.  @ Yes, the standard Citrix JAVA client runs on VMS. You can run ir> directly as well and don't need to run it as a browser applet.  D You can also direct a Terminal Server session to an X display on VMS   --
 Alan Greig   ------------------------------  # Date: Mon, 25 Sep 2000 06:26:02 GMT  From: keymaster@stlnet.com6 Subject: Finding size of files inside a backup saveset) Message-ID: <8qmr5d$ii0$1@nnrp1.deja.com>    Gang,   F This is my first post to this newsgroup so forgive me if I violate any rules...  G Does anyone know how to analyze a saveset and get the total size of the 1 saveset's contents and assign it to a VMS symbol?   F I want to write a routine that restores the contents of a saveset to aE device.  However, before I begin the restore, I'd like to compare the C size of the contents of the saveset to the size of the directory to D which the files will be restored.  If the freeblocks on the disk areF larger than the size of the saveset, I'll do the restore.  If they areB smaller, then I'll choose another destination for the restoration.  F I know how to determine the available free blocks on a disk.  However,D I don't know how to determine the size of the files in a saveset and assign it to a VMS symbol.  A I can display the size of the saveset files on my screen with the E backup/list command.  However, I think I need a lexical or some other D function to be able to work with the value in a DCL.  I can send theE screen output to a file and then play around with the contents to get 2 the file size which I'll only do as a last resort.  ' Thanks for any help anyone can give me!   
 Derek Cohn' Directory Information Technologies P.L.  Melbourne, Australia    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  # Date: Mon, 25 Sep 2000 11:06:56 GMT  From: JMan <jman@urlgray.com> : Subject: Re: Finding size of files inside a backup saveset8 Message-ID: <rccuss4ih3lcu8vd2qk4aqf04d6mp2q9sg@4ax.com>  = On Mon, 25 Sep 2000 06:26:02 GMT, keymaster@stlnet.com wrote:    >Gang, > G >This is my first post to this newsgroup so forgive me if I violate any 	 >rules...  > H >Does anyone know how to analyze a saveset and get the total size of the2 >saveset's contents and assign it to a VMS symbol? > G >I want to write a routine that restores the contents of a saveset to a F >device.  However, before I begin the restore, I'd like to compare theD >size of the contents of the saveset to the size of the directory toE >which the files will be restored.  If the freeblocks on the disk are G >larger than the size of the saveset, I'll do the restore.  If they are C >smaller, then I'll choose another destination for the restoration.  > G >I know how to determine the available free blocks on a disk.  However, E >I don't know how to determine the size of the files in a saveset and  >assign it to a VMS symbol.  > B >I can display the size of the saveset files on my screen with theF >backup/list command.  However, I think I need a lexical or some otherE >function to be able to work with the value in a DCL.  I can send the F >screen output to a file and then play around with the contents to get3 >the file size which I'll only do as a last resort.  > ( >Thanks for any help anyone can give me! >  >Derek Cohn ( >Directory Information Technologies P.L. >Melbourne, Australia  >  > ' >Sent via Deja.com http://www.deja.com/  >Before you buy.     Have a look at $backup\list   E You could direct this output to a temp file then use DCL to parse the % file to get what you are looking for.      HTH,     Jman...    jman@urlgray.com  3 No electrons were harmed in producing this message.c   ------------------------------  % Date: Mon, 25 Sep 2000 07:40:22 -0400F2 From: "Richard B. Gilbert" <DRAGON@compuserve.com>6 Subject: Finding size of files inside a backup saveset7 Message-ID: <200009250740_MC2-B47E-D4F9@compuserve.com>   H         How about SAVESET_SIZE=3DF$FILE_ATTRIBUTES("<filespec>", "EOF")?J That will get you the size of the saveset.  Unfortunately, this has only = a F tenuous relationship to the size of the files that will be restored! =  J Consider, for example, a large file marked NOBACUKP; it will occupy about=  J one block in the saveset because none of the data is saved but when it is=  F restored, it will be a large empty file with the same size, ownership," protections, etc, as the original.  J         The only way you will get the size of the files being restored is=  J to do a BACKUP/LIST, parse the output for the file size and keep a runnin= g- total.  5 Message text written by INTERNET:keymaster@stlnet.com G >This is my first post to this newsgroup so forgive me if I violate anyf rules...  G Does anyone know how to analyze a saveset and get the total size of thet1 saveset's contents and assign it to a VMS symbol?f  F I want to write a routine that restores the contents of a saveset to aE device.  However, before I begin the restore, I'd like to compare the C size of the contents of the saveset to the size of the directory tohD which the files will be restored.  If the freeblocks on the disk areF larger than the size of the saveset, I'll do the restore.  If they areB smaller, then I'll choose another destination for the restoration.  F I know how to determine the available free blocks on a disk.  However,D I don't know how to determine the size of the files in a saveset and assign it to a VMS symbol.  A I can display the size of the saveset files on my screen with thesE backup/list command.  However, I think I need a lexical or some other.D function to be able to work with the value in a DCL.  I can send theE screen output to a file and then play around with the contents to geta2 the file size which I'll only do as a last resort.  ' Thanks for any help anyone can give me!b <    ------------------------------  % Date: Mon, 25 Sep 2000 14:36:12 +0000 - From: SysAdmin <djesys.nospam@fsi.net.mapson>D: Subject: Re: Finding size of files inside a backup saveset. Message-ID: <39CF62DC.8C1607BB@fsi.net.mapson>   "Richard B. Gilbert" wrote:s > H >         How about SAVESET_SIZE=F$FILE_ATTRIBUTES("<filespec>", "EOF")?L > That will get you the size of the saveset.  Unfortunately, this has only aF > tenuous relationship to the size of the files that will be restored!K > Consider, for example, a large file marked NOBACUKP; it will occupy aboutsK > one block in the saveset because none of the data is saved but when it istH > restored, it will be a large empty file with the same size, ownership,$ > protections, etc, as the original. > K >         The only way you will get the size of the files being restored is:L > to do a BACKUP/LIST, parse the output for the file size and keep a running > total.  B Remember also that the target disk clustersize will play a role inF determining whether the contents of a specific saveset can be restored to a specific disk volume.   David J. Dachteraa   ------------------------------  % Date: Mon, 25 Sep 2000 15:43:51 +0100f2 From: Chris Sharman <Chris.Sharman@CCAgroup.co.uk>: Subject: Re: Finding size of files inside a backup saveset. Message-ID: <39CF72B7.71A7DD12@CCAgroup.co.uk>  I > Does anyone know how to analyze a saveset and get the total size of theo3 > saveset's contents and assign it to a VMS symbol?h  > If you can settle for a rough ballpark figure, look at lexical f$file("saveset","eof")r  G It'll be bigger than its contents depending on the backup group & blockt sizes you used.e   ------------------------------   Date: 25 Sep 2000 15:27:12 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog): Subject: Re: Finding size of files inside a backup saveset, Message-ID: <8qnqsg$luq@gap.cco.caltech.edu>  F In article <8qmr5d$ii0$1@nnrp1.deja.com>, keymaster@stlnet.com writes: > H >Does anyone know how to analyze a saveset and get the total size of the2 >saveset's contents and assign it to a VMS symbol?  J To a first approximation, use the size of the saveset.  They are actually B somewhat larger than the original directory structure (there's no  compression.)  Example::   $ ds/grand [.utils...]  : Grand total of 2 directories, 152 files, 7677/8241 blocks.) $ backup [.utils...]*.*.* []test.bck/savea
 $ ds test.bck   ) Directory USRDISK:[USERS.MATHOG.PROGRAMS]n  ! TEST.BCK;1              8820/8820o  B You could of course always list the contents of the save set with:   $ backup/list TEST.BCK;1/savec   which will show you lines like:   P [USERS.MATHOG.PROGRAMS.UTILS]ABI.TXT;24                    52   4-AUG-2000 10:33  D You could conceivably direct the list to a file and then parse it toE pick up the exact file sizes.  But unless disk space is really tight,d? I'd just use the estimate from the size of the save set itself.S    G >I know how to determine the available free blocks on a disk.  However,tE >I don't know how to determine the size of the files in a saveset and  >assign it to a VMS symbol.s   $ HELP LEX F$FILE_ATTRIBUTES  3 $ savesetsize = f$file_attributes("TEST.BCK","ALQ")o $ write sys$output savesetsize 8820  / Sounds like you're new to VMS.  Have a look at:y  2   http://seqaxp.bio.caltech.edu/www/vms_sheet.html:   http://seqaxp.bio.caltech.edu/www/vms_beginners_faq.html  , and the User's and System Manager's manuals.   Regards,   David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech g   ------------------------------  % Date: Mon, 25 Sep 2000 09:36:32 -0400S; From: "Everhart, Glenn (FUSA)" <GlennEverhart@FirstUSA.com>r' Subject: FW: Suggested VMS new feature!fN Message-ID: <4B279CF3578CD211B945009027178017046CFF8C@swilnts809.wil.fusa.com>  @ There seem to be a number of articles suggesting a VMS Undelete.< In the Safety package I implemented this, for vax and alpha,D several years ago. The sigtapes are the best source (the Spring 2000A tapes are out btw.) since there was some additional coding neededWD to support 7.2 systems with privs stored in PSBs. The undelete stuffB needed no mods. Safety gets control at FDT time, so as long as theE step 2 driver interface remains the same, it needs little alteration.,  F You can get it now and use it now. Of course Compaq can reimplement itA now too if they choose. Safety however leaves the VMS delete codeiC and all VMS commands alone...just adds some processing ahead of theaB filesystem. It can do a number of other neat tricks (file hiding, @ moving things around behind your back and restoring when you ask for them etc.)  D Safety instead implements a delete and expunge command. Delete means@ that files get the delete treatment (which can be to put them in wastebasket,D store them (possibly compressed) somewhere or (based on various userG selections) possibly just delete them.  When free space runs out on thetC disk the normal action is that files deleted more than nnnn seconds,F ago (normally 5-10 minutes) get actually deleted. Expunge is a sort ofF delete/i_mean_it command which deletes files irreversibly. The idea is@ that you let normal deletes move files to save areas and let theB wastebasket automatically empty when the disk needs space, and runI a cleanup every so often to remove older files. That can be done whenevert
 one likes.   Thus you can * delete a file RIGHT NOWmI * Have a file move to wastebasket or some other backing store in the formo 	the system manager wantsnJ * Clean out wastebasket stuff of things deleted more than nnnn seconds ago5 	(normally something that would run at least nightly)-K * Allow cleanout to run when the disk fills, so creation or extend requestsf) 	do NOT fail because of old deleted stuffzE * specify not to bother saving things like .lis, .tmp, .map, or other0
 classes ofA 	files that don't need to be protected (so that deletions of suchu filesR 	just take place normally-G * undelete a file that was protected (puts it back exactly where it hade been;eA 	wildcards in the file path are ok and will get whatever matches)a  H This is of course all available right now. The macro32 code that enables this< is "on there" so you can see what is going on if interested.  E I thought this would be quite popular when I wrote it, but nobody hasc botheredJ to tell me they use it. I have found it works smoothly (as have some otherE organizations), runs on vax or alpha, and suggest that the "rename toa wastebasket"J mode of use is the least intrusive. If the system has to use other methodsE to handle storage (esp. firing up a command file, even with a process  sittingaI and waiting for it as is there) deletion can be noticeably slowed. It canrG take a noticeable time to compress a file or copy it over a network foreH storage. I supply command files to compress (and decompress on undelete)> but you need to be patient if you let it do so on huge files.)  K You should note the native VMS delete code is there, untouched, and is usedSK as is. My code gets in there first, then passes to delete. (This also meansyH that when Safety protects you from unauthorized deletion, it REALLY doesJ it, since it can block the IO$_delete from ever being seen by the acp/xqp.E Packages that check io$_access (used by the library delete call) and bI especially those that trigger off access calls seen in the audit log, mays' not be able to block deletes reliably.)n  L Thus Safety does NOT rewrite the VMS delete functions. Rather, it adds a bitJ of added processing just ahead of whatever filesystems you want it to workI on. You can specify some disks get the added delete processing and othersc nottC to, and the ones you don't add the Safety processing to just remainF	 untouchedcI by the code. Thus for instance a news disk really is not a good candidatei8 for saving deleted stuff; don't tell safety to touch it.  K Safety renames files in the wastebasket using file id and some of the name, L so these names remain unique, yet can be moved back by hand if anyone reallyG needs to. A complete cross reference is kept also by Safety so it has aaE record of the full original file path. This is too long to keep in a rH renamed filename. Safety also records when the file was deleted. That isH necessary in order to be able to delete files deleted more than xxx time ago.  J Yes, you can replace the delete command with a DCL script to rename files,K emptying the wastebasket periodically, but that won't handle deletions fromhE programs or for that matter other commands. Safety can catch anythinga deletedhJ with a usermode io$_delete getting rid of it. Thus if you use RSX FOCAL inL compat mode and say "Lib Erase somefile.sometype", Safety will still happily5 catch it for you. Ditto for less unusual :-) methods.i   Glenn Everhart everhart@gce.com   ------------------------------  % Date: Mon, 25 Sep 2000 15:07:54 +0000d- From: SysAdmin <djesys.nospam@fsi.net.mapson>-+ Subject: Re: FW: Suggested VMS new feature! . Message-ID: <39CF6A4A.301548F3@fsi.net.mapson>   "Everhart, Glenn (FUSA)" wrote:  > B > There seem to be a number of articles suggesting a VMS Undelete.> > In the Safety package I implemented this, for vax and alpha,F > several years ago. The sigtapes are the best source (the Spring 2000C > tapes are out btw.) since there was some additional coding neededuF > to support 7.2 systems with privs stored in PSBs. The undelete stuffD > needed no mods. Safety gets control at FDT time, so as long as theG > step 2 driver interface remains the same, it needs little alteration.. [snip]  
 Hi, Glenn,  H I think the point was that the requester wanted something that would notH require the use of a different command (proc.'s, probably thousands, are# already coded with DELETE in them).x  F When we look at DELETE.CLD (or is it in DCLINT.CLD? I forget just now)G from VMS062.A (or just extract it from your command tables using VERB),:D we see that DELETE.EXE actually only handles RMS-file deletes. OtherC forms of the DELETE command invoke QUEMAN or some other program, or0 invoke a DCL cliroutine.  F So we're not talking about "rewriting" the ENTIRE DELETE command, only- that portion of it that deals with RMS files.w  H I don't normally advocate writing "trojans" for OpenVMS. However, takingD an example from MTI's TapeControl is where I came up with the idea I posted (the long post).    David J. Dachterai   ------------------------------  % Date: Mon, 25 Sep 2000 11:26:12 -0500a1 From: "Dave Gudewicz" <david.gudewicz@abbott.com>r0 Subject: Re: Getting Compaq to advertise OpenVMS7 Message-ID: <8qnu4m$5g$1@fizban.fizban.pprd.abbott.com>e  H It took Mr. Harrison about 50 years to convince the British government'sC science commission that he had a pretty good idea on how to measurecL longitude with the accurate measure of time.  Some of the other ideas on how> to do this were unbelievable with the aide of 20-20 hindsight.  L His persistence paid off and we now have GPS and other marvels as extensions of his original idea.      (another) Dave...h  B "David J. Dachtera" <djesys.nospam@earthlink.net> wrote in message' news:39CE2D66.C87E9536@earthlink.net...o > "William H." wrote:e > >w > > If you want to promote# > > advertising of OpenVMS, write ao% > > letter on your Company Letterheadz > > to the people who make the > > descisions at Compaq.  Keep , > > it professional and keep it to one page. >o0 > I think I'm starting to get William's message: >l: > Keep writing and writing and writing and writing and ... > I > (Ever seen "The Shawshank Redemption"? Remember how the buy got funding 8 > for the prison library? It took years, but he got it!) >t > -- > David J. Dachteram > dba DJE Systemsa > http://www.djesys.com/ >e< > Unofficial Affordable OpenVMS Home Page and Message Board:! > http://www.djesys.com/vms/soho/  >cH > This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings > is to be expected. >pB > Feel free to exercise your rights of free speech and expression. >aH > However, attacks against individual posters, or groups of posters, are > strongly discouraged.e   ------------------------------  # Date: Mon, 25 Sep 2000 06:54:19 GMTd, From: "Arto Tanninen" <arto.tanninen@icl.fi>K Subject: How to connect Digital line printer (LP29 or LP37) to HP JetDirectr, Message-ID: <vECz5.33$bb.4319@read2.inet.fi>   Hi,   J I would like to connect Digital LP29 and LP37 to network via HP JetDirect.F Printers are now connecting to DECserver 250 with cable BC27A. I'm nowL looking for parallel cable DB25-DB37 or DB25-DB50 or some kind of converter.   Could you help me.   Arto arto.tanninen@icl.fi   ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200a" From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirect 0 Message-ID: <8qn5ml$gg3$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, andlF not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in berichto& news:vECz5.33$bb.4319@read2.inet.fi... > Hi,a >eL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind ofa
 converter. >I > Could you help me. >u > Arto > arto.tanninen@icl.fi >  >h   ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200 " From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirect 0 Message-ID: <8qn7eu$kr0$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, andeF not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in bericht & news:vECz5.33$bb.4319@read2.inet.fi... > Hi,i >sL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind oft
 converter. >a > Could you help me. >a > Arto > arto.tanninen@icl.fi >h >w   ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200i" From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectl0 Message-ID: <8qn977$p98$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, andhF not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in berichth& news:vECz5.33$bb.4319@read2.inet.fi... > Hi,o >hL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind ofs
 converter. >  > Could you help me. >o > Arto > arto.tanninen@icl.fi >t >d   ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200I" From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirect 0 Message-ID: <8qnavg$tj3$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, andsF not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in bericht & news:vECz5.33$bb.4319@read2.inet.fi... > Hi,  >aL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind ofl
 converter. >k > Could you help me. >3 > Arto > arto.tanninen@icl.fi >  >c   ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200B" From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirecte0 Message-ID: <8qncnq$28i$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, and F not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in berichtA& news:vECz5.33$bb.4319@read2.inet.fi... > Hi,n >eL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind of-
 converter. >- > Could you help me. >2 > Arto > arto.tanninen@icl.fi >h >    ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200i" From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirect 0 Message-ID: <8qneg3$6iu$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, andgF not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in berichtf& news:vECz5.33$bb.4319@read2.inet.fi... > Hi,s >uL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind ofm
 converter. >r > Could you help me. >  > Arto > arto.tanninen@icl.fi >a >    ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200e" From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirect[0 Message-ID: <8qng8c$amm$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, and2F not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in berichte& news:vECz5.33$bb.4319@read2.inet.fi... > Hi,s >iL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind ofl
 converter. >  > Could you help me. >  > Arto > arto.tanninen@icl.fi >B >S   ------------------------------  # Date: Mon, 25 Sep 2000 13:01:56 GMTp, From: "Arto Tanninen" <arto.tanninen@icl.fi>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirect . Message-ID: <81Iz5.209$bb.19239@read2.inet.fi>  C Could I use BC27Z RapidPrint 200 to Dataproduct Line Printer Cable?    Arto  7 "Arto Tanninen" <arto.tanninen@icl.fi> wrote in messagen& news:vECz5.33$bb.4319@read2.inet.fi... > Hi,  >-L > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind of6
 converter. >9 > Could you help me. >e > Arto > arto.tanninen@icl.fi >s >g   ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200h" From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirectd0 Message-ID: <8qni0l$f27$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, and F not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in berichtr& news:vECz5.33$bb.4319@read2.inet.fi... > Hi,  >bL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind ofr
 converter. >c > Could you help me. >t > Arto > arto.tanninen@icl.fi >r >e   ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200e" From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirecta0 Message-ID: <8qnjot$jeb$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, and F not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in bericht & news:vECz5.33$bb.4319@read2.inet.fi... > Hi,f >oL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind of 
 converter. >f > Could you help me. >  > Arto > arto.tanninen@icl.fi >o >t   ------------------------------  % Date: Mon, 25 Sep 2000 11:25:40 +0200 " From: "Dirk Munk" <d.munk@kpn.com>O Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirects0 Message-ID: <8qnlh6$nl0$1@hdxl22.telecom.ptt.nl>  K The parallel connection on a DECserver 250 is a Dataproducts interface, andsF not a Centronics interface as normally seen on printers. So it is veryL unlikely that a HP JetDirect box can be connected to a LP29 or LP37 printer,B unless these printers are also equiped with a Centronics or Serial
 interface.  9 "Arto Tanninen" <arto.tanninen@icl.fi> schreef in berichto& news:vECz5.33$bb.4319@read2.inet.fi... > Hi,  >gL > I would like to connect Digital LP29 and LP37 to network via HP JetDirect.H > Printers are now connecting to DECserver 250 with cable BC27A. I'm nowC > looking for parallel cable DB25-DB37 or DB25-DB50 or some kind ofs
 converter. >n > Could you help me. >. > Arto > arto.tanninen@icl.fi >  >p   ------------------------------  % Date: Mon, 25 Sep 2000 14:39:50 +0000 - From: SysAdmin <djesys.nospam@fsi.net.mapson>oO Subject: Re: How to connect Digital line printer (LP29 or LP37) to HP JetDirecta. Message-ID: <39CF63B6.1F9A1DC3@fsi.net.mapson>   Arto Tanninen wrote: > E > Could I use BC27Z RapidPrint 200 to Dataproduct Line Printer Cable?e  < Well, the cable is not the issue. The issue is the interfaceF specification, and it varies between Centronics parallel, DataProductsE parallel and DEC-specific DataProducts parallel (the DEC interface is # not strictly "pure" Dataproducts").e  F The only server I am aware of which supports this with any reliability is the Emulex PS-2504-M.   David J. Dachteral   ------------------------------  , Date: Mon, 25 Sep 2000 11:23:08 +0107 (CEST)/ From: Tomasz Dryjanski <tdryjanski@hotmail.com>f" Subject: Migrating from 7.1 to 7.25 Message-ID: <F199T6xpqLvH4p44qQs00002ca4@hotmail.com>a  > We are preparing to migrate from 7.1 to 7.2. Do you know aboutE problems we may encounter? I mean obsolete console version, licensingi etc.  " This is the current configuration: AlphaServer 2100 4/275" VMS 7.1-1H1 with necessary patches UCX V.4.1, ECO 10a* Oracle (DEC) RDB 7.0-5 and Oracle DB 8.0.5% Decnet IV (V.6.2, as reported by NCP)  and others.o      Thanks in advance for any hints. T. D. I _________________________________________________________________________eI Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.T  D Share information about yourself, create your own public profile at  http://profiles.msn.com.   ------------------------------  % Date: Mon, 25 Sep 2000 14:51:28 +0000g- From: SysAdmin <djesys.nospam@fsi.net.mapson>a& Subject: Re: Migrating from 7.1 to 7.2. Message-ID: <39CF6670.FA3E81AF@fsi.net.mapson>   Tomasz Dryjanski wrote:m > @ > We are preparing to migrate from 7.1 to 7.2. Do you know aboutG > problems we may encounter? I mean obsolete console version, licensinga > etc. > $ > This is the current configuration: > AlphaServer 2100 4/275$ > VMS 7.1-1H1 with necessary patches > UCX V.4.1, ECO 10,, > Oracle (DEC) RDB 7.0-5 and Oracle DB 8.0.5' > Decnet IV (V.6.2, as reported by NCP)R
 > and others.d > " > Thanks in advance for any hints. > T. D.3  B Well, the Release Notes are going to be a good starting point, I'm sure...n   David J. Dachterag   ------------------------------   Date: 25 Sep 2000 15:20:32 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)& Subject: Re: Migrating from 7.1 to 7.26 Message-ID: <8qnqg0$4m0$1@mailint03.im.hou.compaq.com>  g In article <F199T6xpqLvH4p44qQs00002ca4@hotmail.com>, Tomasz Dryjanski <tdryjanski@hotmail.com> writes:e? :We are preparing to migrate from 7.1 to 7.2. Do you know aboutn :problems we may encounter?   E   The "any problems?" form of the open-ended question is exceedingly h!   difficult for anyone to answer.h  E   OpenVMS Alpha V7.1-2 and V7.2-1 are the usual targets for upgrades.h  D   There are ECO kits available for both OpenVMS Alpha V7.1-2 and forG   V7.2-1, and you will want to apply the mandatory kits for the releaseu)   as well as any other relevent ECO kits.g  G   New ECO kit format exists on OpenVMS Alpha V7.1-2 and later: no more hG   VMSINSTAL ECO kits are used for OpenVMS Alpha, all OpenVMS Alpha ECO     kits use PCSI.  0 :I mean obsolete console version, licensing etc.  D   You will find the upgrade prerequisite information in the manuals.3   I'd strongly encourage starting with the manuals.o  D   Typical customer licenses do not expire -- there are some licensesE   and some customers that have version-dependent or release-specific iC   or that have termination dates, but these are comparatively rare. G   (The CSLG licenses and Temporary Service PAKs (TSP) are probably the G*   most common of these "rare" licenses...)   :AlphaServer 2100 4/275b  A   Do you have enough physical memory for current and planned use?c  # :VMS 7.1-1H1 with necessary patchesm  H   Iif you really had all of the necessary patches for V7.1 or V7.1-1H*, (   you'd be at OpenVMS Alpha V7.1-2.  :-)   :UCX V.4.1, ECO 10  1   You will want to move to TCP/IP Services V5.0A.i  + :Oracle (DEC) RDB 7.0-5 and Oracle DB 8.0.5e  @   Oracle Rdb version looks good, donno about the Oracle version.  & :Decnet IV (V.6.2, as reported by NCP)  D   DECnet Phase IV is still available and -- for those customers thatF   require DECnet Phase IV support, Prior Version Support contracts areD   available.  If you decide to upgrade to DECnet-Plus (The DVNETEND,G   DVNETEXT, DVNETRTG and NET-APP-SUP* licenses will continue to work),  F   please read the DECnet-Plus documentation -- the upgrade of a simpleB   DECnet Phase IV configuration to a DECnet-Plus configuration is E   normally easy, but the operation and the management and the generalRH   requirements of DECnet-Plus are rather different from DECnet Phase IV.  ! :Thanks in advance for any hints..  H   Please read the upgrade manual, please read the new features manuals, F   and please read the release notes.  Current copies are available at:  '     http://www.openvms.compaq.com:8000/t  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Mon, 25 Sep 2000 12:11:20 GMTo- From: "Richard D. Piccard" <piccard@ohio.edu>a Subject: Re: NTP with UCX 4.2M( Message-ID: <39CF40E6.684035C1@ohio.edu>  I From what has been posted on comp.os.vms in the last year or so (NOT fromDH my own personal experience) the major CPU consumption when telnetting inA to a 5.0 system was not just if you came from Linux.  The bug wasr$ introduced in 5.0 and cured in 5.0A.  #                                 RDP      John Stott wrote:-  < > "Antony Wardle" <antony.wardle@nnnoospam.met.co.nz> wrote: > D > >You need 5.0a if you have any linux boxes that telnet to your vms@ > >hosts. We found that each telnet session would use 30% of theI > >cpu, therefore 3 telnets, would have your whole cpu all to themselves.  >D8 > I don't see this with UCX 4.2 on VMS 6.2 Vax or Alpha. > BTW, NTP works fine too. >D > --7 > John P. Stott                    jpstott@src.wisc.edus: > Synchrotron Radiation Center     http://www.src.wisc.edu6 > University of Wisconsin-Madison  http://www.wisc.edu   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  % Date: Mon, 25 Sep 2000 09:37:43 -0400.# From: Jim Agnew <agnew@hsc.vcu.edu>e  Subject: Re: Open File, or What?+ Message-ID: <39CF5527.CF541705@hsc.vcu.edu>   4 try backup/ignore=interlock orig.file newoutput.file  F did you spawn subprocesses in the debugger? is one of them holding it?  I the backup trick has worked for me in the past, but the force be with youP   JimP   "Douglas S. Meade" wrote:- > = > I was working yesterday on porting a legacy Fortran program < > from Lahey Fortran to Vax Fortran.  I was in the debugger,? > and decided to exit, but files were still open.  The debuggern8 > complained, but I exited anyway.  Now I have a problem8 > with one of the source files, UTIL.FOR.  I cannot copy< > it, edit it, or even type it, without hanging the console,: > at least for several minutes.  I doesn't even respond to? > ^T.  I've even tried shutting down the machine and rebooting, > > but the problem remains.  Does anyone have a clue what could< > be wrong with this file?  I've never seen this behavior on > any other OS.b > > > The machine in question is a Vaxstation 4000-500, with RF-73 > disks. >  > Doug Meade >o   ------------------------------  # Date: Mon, 25 Sep 2000 08:48:10 GMT22 From: a.koopman@markgraaf.nl.nospam (Adri Koopman)? Subject: Re: OpenVMS Systems as client to MS SQLserver  Server.9/ Message-ID: <39cf0fed.414906393@news.hccnet.nl>9  5 On Fri, 22 Sep 2000 11:38:19 -0400, "Fletcher Hearns"a <hearns@softapp.com> wrote:c  B >I have been asked to investigate making some VMS (and Tru65 Unix)I >application talk to a MS SQLserver Server box.  Basically I need the VMSaJ >(and TRU64 Unix box) to act as client to the MS box.  Does anyone know of. >ODBC Client software to allow this to happen? >. >Thanks in advance,e >i >e	 >Fletchers >C> Try getting (old) Sybase 4.9 client software and use TCP/IP asF networking protocol. You can use std DB-Library calls to access the MSB SQL server. If you stickt to basic SQL stuff and don't do anything fancy, things work fine.E I have some VMS app's working, originaly built against a SYBASE 4.9.2.D SQL server, but after adding the MS SQL Server to SYBASE:INTERFACES.E and changing DSQUERY to the MS SQL Server box, everything still worksd fine.p   Grtnx,     Adri Koopman Markgraaf B.V. a.koopman at markgraaf dot nlH   ------------------------------   Date: 25 Sep 2000 13:42:55 GMT* From: bdwheele@indiana.edu (Brian Wheeler)? Subject: Re: OpenVMS Systems as client to MS SQLserver  Server.m3 Message-ID: <8qnkov$o84$1@flotsam.uits.indiana.edu>   / In article <39cf0fed.414906393@news.hccnet.nl>, 5 	a.koopman@markgraaf.nl.nospam (Adri Koopman) writes: 7 > On Fri, 22 Sep 2000 11:38:19 -0400, "Fletcher Hearns". > <hearns@softapp.com> wrote:i > C >>I have been asked to investigate making some VMS (and Tru65 Unix)tJ >>application talk to a MS SQLserver Server box.  Basically I need the VMSK >>(and TRU64 Unix box) to act as client to the MS box.  Does anyone know ofa/ >>ODBC Client software to allow this to happen?l >> >>Thanks in advance, >> >>
 >>Fletcher >>@ > Try getting (old) Sybase 4.9 client software and use TCP/IP asH > networking protocol. You can use std DB-Library calls to access the MSD > SQL server. If you stickt to basic SQL stuff and don't do anything > fancy, things work fine.G > I have some VMS app's working, originaly built against a SYBASE 4.9.2CF > SQL server, but after adding the MS SQL Server to SYBASE:INTERFACES.G > and changing DSQUERY to the MS SQL Server box, everything still workst > fine.b >  > Grtnx, >  >  > Adri Koopman > Markgraaf B.V. > a.koopman at markgraaf dot nlt  K Also, there's a project called FreeTDS which is a GPL implementation of theiI TDS protocol which sybase and MS SQL use.  It has CTlib and DBlib as welliM as some preliminary ODBC library functions.  I've heard of some people tryingeC to build it on VMS with mixed results.  The project homepage is at:  	www.freetds.org  
 Brian Wheeler5 bdwheele@indiana.edu   ------------------------------  % Date: Mon, 25 Sep 2000 09:28:52 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.br  Subject: OpenVMS Undelete5L Message-ID: <OFF845C223.FEB80BAB-ON83256965.00415364@ep-bc.petrobras.com.br>  F The easiest  way to do that is .... a liitle bit stupid but works ....   $ CREATE/DIR DISK:[RECYCLE]h $ DEFINE RECYCLE DISK:[RECYCLE]v' $ DEL*ETE :== run sys$system:newdel.exe   . Where newdel.exe is a file rename program ....- Of course the DELETE parameters will not worke. but it can be useful for unpriviliged users...    
 Fabio Cardosoi Analista de Suported   ------------------------------    Date: 25 Sep 2000 10:09:31 -0500, From: koehler@eisner.decus.org (Bob Koehler) Subject: Re: OpenVMS Undeletei+ Message-ID: <X7ZOCBeSUHMw@eisner.decus.org>9  x In article <OFF845C223.FEB80BAB-ON83256965.00415364@ep-bc.petrobras.com.br>, fabio_compaq@ep-bc.petrobras.com.br writes:H > The easiest  way to do that is .... a liitle bit stupid but works .... >  > $ CREATE/DIR DISK:[RECYCLE] ! > $ DEFINE RECYCLE DISK:[RECYCLE]-) > $ DEL*ETE :== run sys$system:newdel.exe2 >   H Which of course breaks all other delete commands, such as delete/symbol.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationo= Hubble Space Telescope Payload  | Federal Sector, Civil GroupeE  Flight Software Team           | please remove ".aspm" when replyingy   ------------------------------  % Date: Mon, 25 Sep 2000 10:40:40 -0300s) From: fabio_compaq@ep-bc.petrobras.com.brt Subject: Re: OpenVMS Undelete L Message-ID: <OFFAD78961.541340C7-ON83256965.004B07A4@ep-bc.petrobras.com.br>  
 Oops ! ! !   I am sorry, just to correcti  ) $ ERASE :=3D=3D run sys$system:newdel.exek   It=B4s only an idea ok ?  
 Fabio Cardosot                  = koehler@eisner.decus.org (Bob Koehler) em 25/09/2000 12:09:31-H                                                                        =     =20oH                                                                        =     =20 H                                                                        =     =20H    @                                                              =20@                                                              =20@                                                              =20@  Para:    Info-VAX@Mvb.Saic.Com                              =20@                                                              =20@  cc:      (bcc: Fabio dos Santos Cardoso/E-P-BC/Contratada)  =20@                                                              =20@                                                              =20@                                                              =20@  Assunto: Re: OpenVMS Undelete                               =20@                                                              =20           =     
 In articleA <OFF845C223.FEB80BAB-ON83256965.00415364@ep-bc.petrobras.com.br>,2+ fabio_compaq@ep-bc.petrobras.com.br writes:nH > The easiest  way to do that is .... a liitle bit stupid but works ...= .s >t > $ CREATE/DIR DISK:[RECYCLE]t! > $ DEFINE RECYCLE DISK:[RECYCLE] - > $ DEL*ETE :=3D=3D run sys$system:newdel.exeL >   H Which of course breaks all other delete commands, such as delete/symbol= .n  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationt= Hubble Space Telescope Payload  | Federal Sector, Civil Group E  Flight Software Team           | please remove ".aspm" when replying        =    ------------------------------  % Date: Mon, 25 Sep 2000 09:57:09 +0200 : From: Karl Rohwedder <extern.karl.rohwedder@volkswagen.de>% Subject: Re: OSU scripts under Apache - Message-ID: <39CF0555.6DAAEC44@volkswagen.de>    Hi,u  F we use the same CONF entry and it works fine. Do you have a proxy from the APACHE account like:   LOCAL:.DOE01P::APACHE$WWW@     HTTP_SERVER (D)>  D Does a NETWORK task gets started? Are entries in the NET$SERVER.LOG?     amartineau@my-deja.com wrote:e > I > I just installed Apache 1.3.12 on an Alpha server running VMS 7.2-1 andcH > Multinet V4.2 Rev A-X, Serving documents and standard CGIs works fine.I > But I need to migrate a bunch of OSU scripts. There is no documentationa9 > on how to configure this. The lines in httpd.conf are :e > F > # The /htbin directory invokes the OSUscript module, The DECnet task > stringD > # should be customized to invoke the OSU web server's scriptserver	 > object.7 > #m > <Location /htbin>a" >     SetHandler osuscript-handler- >     OSUscript 0::"0=WWWEXEC" www_root:[bin]e >     Order allow,deny >     Allow from all
 > </Location>- > F > It looks OK to me, so I did not change anything. When I try to run a  > script in www_root:[bin], likeJ > http://bla.com:8008/htbin/dcl_env_rm.com, I get a 500. There is no trace1 > of anything in the Apache logs or the OSU logs.h >  > Any idea anyone ? Thank you. >  > Alain Martineaui > F > P.S. I would rather stay with OSU, but the developers want servlets. > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.a   -- r  - mit freundlichen Gruessen | with best regardsu   Karl Rohwedder                C iT-Ingenieurteam     | Ellernbruch 11       | D-38112 Braunschweig hA Telefon: 0531/515521 | Telefax: 0531/515531 | Mobil: 0172/5434843iE  E-Mail: rohwedder@decus.decus.de           | iT-IngTeam@t-online.de n+          karl.rohwedder@it-ingenieurteam.de  DATEX-P: 4505018005::ROHWEDDER   ------------------------------  % Date: Mon, 25 Sep 2000 10:31:20 +0100 + From: "Tim Jackson" <tim.jackson@amsjv.com>i7 Subject: Re: Pascal question: Can OUTPUT be [VOLATILE]?a& Message-ID: <39cf1a98$1@pull.gecm.com>  2 "Jeremy Begg" <jeremy@vsm.com.au> wrote in message$ news:39CD98E8.A27C9029@vsm.com.au... > Hi,9 > C > I'm running COMPAQ PASCAL V5.7 on OpenVMS VAX & Alpha V7.2.  I am>
 working onH > a program which calls LIB$FILE_SCAN to locate files and do things with them.C> >  This routine invokes two procedures provided by my program. >aG > Not surprisingly, the compiler complains that my procedures should besG > declared [ASYNCHRONOUS].  If I do that, several other variables in my G > program must be declared [VOLATILE] (not really a problem).  However,c theserF > procedures use WRITELN to write to the predeclared text file OUTPUT, and the-< > compiler complains that OUTPUT needs to be [VOLATILE] too. >0, > I can't work out how to do this.  If I say >t( >     PROGRAM REVERSION (input, output);# >     VAR output : [VOLATILE] text;s >0H > the compiler complains about conflicting visibility for OUTPUT.  But I can'to > see what else I could do.? >m	 > Thanks,e >n
 > Jeremy BeggiC Try opening a new 'file' to SYS$OUTPUT and WRITELNing to that. i.e.0   PROGRAM REVERSION( newop);   VAR newop : [VOLATILE] text;  - [ASYNCHRONOUS, ....] PROCEDURE WHATEVER(...);n   BEGIN6   OPEN( newop, 'SYS$OUTPUT');t   REWRITE( newop);D   WRITELN( newop, 'This is sent to SYS$OUTPUT via a volatile file');   CLOSE( newop); END;   HTH_D ------------------ Purely Personal Opinion -------------------------D Tim Jackson                                    tim.jackson@amsjv.com Air Systems Groupp Alenia Marconi Systems Ltd.-   ------------------------------  % Date: Mon, 25 Sep 2000 15:52:22 +0100y2 From: Chris Sharman <Chris.Sharman@CCAgroup.co.uk>7 Subject: Re: Pascal question: Can OUTPUT be [VOLATILE]?F. Message-ID: <39CF74B6.17C0CCE3@CCAgroup.co.uk>   Jeremy Begg wrote: >  > Hi,r > N > I'm running COMPAQ PASCAL V5.7 on OpenVMS VAX & Alpha V7.2.  I am working onN > a program which calls LIB$FILE_SCAN to locate files and do things with them.> >  This routine invokes two procedures provided by my program. > G > Not surprisingly, the compiler complains that my procedures should be6E > declared [ASYNCHRONOUS].  If I do that, several other variables in n  D My preferred technique for this (when the procedures aren't _really_H asynchronous) is to use %immed when passing the procedure. This subvertsH all the checks, so you don't get winges about mismatched argument lists,H and don't have to declare the procedures asynchronous (just unbound). OfG course, if you do that for something which really is asynchronous (such4B as an ast) you're liable to get bitten at runtime sooner or later.   Chris    ------------------------------   Date: 25 Sep 2000 14:47:49 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) 3 Subject: Re: Porting Support (was Compaq VMS promo)i+ Message-ID: <8qnoil$9nk$1@info.cs.uofs.edu>l  + In article <VA.000000d7.17534fa1@sture.ch>, #  Paul Sture <paul@sture.ch> writes:a |> eK |> As someone who has tried to use StarOffice, I must add my 2 cents. It istP |> far from the panacea which folks here maybe think it is. I have tried variousX |> versions over the last 18 months, including the free CD from Sun, and that particularX |> version failed with being unable to import such a basic thing as tab delimited files.  P Well, just to keep things balanced, I have installed it on numerous RedHat LinuxQ and Windows95/98 systems without a hitch.  It isn't MS Office, but it offers most  of the functionality.e   |> tT |> I note that the Sun Linux version also failed the basic test of running accordingU |> to the installation instructions, although the installation procedure _was_ slick.e[ |> (Invoking it was not per the instructions - you have to go down an extra directory levelfS |> before typing "soffice" - far from trivial as I effectively had to do a completea_ |> disk search to find out where the thing was (and yes, a search of "soffice" produces all theyK |> files contained in a directory tree of that name), not friendly at all).M  L Yes, but as has been proposed, one of the requirements of the VMS port wouldK be documentation.  In my humble opinion, that means accurate and functionalRJ documentation.  Lack of good documentation is as much the bane of the free( software world as poor coding practices.   |> lV |> My most recent experience with StarOffice came with Suse Linux V7.0 (last weekend).b |> According to the literature, they (the way it's written, it sounds like Suse did it themselves)] |> have done extra work on import/export filters, but I cannot tell as yet, because the thingtR |> doesn't run after installation, stating that some shared libraries are missing.  N That's most likely a Linux thing rather than a StarOffice thing. even the mostN current configurations come in different and incompatable configurations.  One6 would linkt that would not be a problem with VMS.  :-)   |> eK |> I'm sorry, but if those currently distributing the thing cannot get the eJ |> installation right, there's a lot of hard work to be done. OK, I've gotP |> my 60 days free support from Suse to help me, but it's still "not there yet".  N Two points.  One.  What incentive do the current distributors have to care oneP way or the other??  Two. That's the reason for suggesting a grant to pay for theN porting.  Free is nice, but you get what you pay for.  The advantage here is aI major portion of the grunt work of development has been done, not only oneN StarOffice, but on a lot of other packages as well.  While not a trivial task,L porting to VMS is probably a lot less intensive than writing equivalent apps
 from scratch.e   |> iL |> > :My suggestion was for someone, Compaq (except they don't seem to care K |> > :much)... to put together a grant to provide some ... schools who are eO |> > :interested with some hardware and software and funding to hire promising a( |> > :students to actaully do the ports. |> > hI |> >   A good idea, though there is an associated cost beyond the obviousvL |> >   financials -- we would want to work with and support the effort, and K |> >   to track the progress of the porting work.  I'll suggest this option M |> >   as part of some other related (and currently unspecified) discussions :J |> >   pending or associated with this particular group of customers.  We M |> >   have already provided hardware and support to individuals and groups, PK |> >   assistance for the folks that have been porting particular software uC |> >   packages over to OpenVMS -- ample precedent, in other words.h  O Requiring periodic reports or even periodic public announcements and maybe even P code releases would seem to me to be well within the domain of whatever grant is made.  :   |> >    U |> Well, I'd rather see Compaq taking an interest in the WordPerfect office suite :-)rY |> Quite some risk of incurring Redmond's wrath, but potential income from day one on theg |> PC front...  M WordPerfect Office was available for VMS at one time.  But again, people seemuL to be zeroing in on a single application and now even a proprietary one.  MyK proposal was for taking the wealth of free software that is making Linux so L popular and putting that on VMS.  If all we're going to see is another $5000L application, we will still be discussing why no one wants to use VMS 5 yearsO from now.  It wasn't the availability of WordPerfect for Linux that contributedlB to it's popularity.  They were a johnny-come-lately in that arena.   |>  S |> I note that for some time, Corel have been struggling financially, and have justsU |> this week found a mystery investor. I sincerely wish it were Compaq, but only timeo
 |> will tell.e |> bK |> Whoever it turns out to be may be receptive to the idea of a VMS port....  I Which as i said above, is highly unlikely to change anything form what we  have right now.2   bill   -- aJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  # Date: Mon, 25 Sep 2000 10:03:17 GMTp% From: A.Greig@virgin.net (Alan Greig)b Subject: Re: Press Release2 Message-ID: <39cf22b0.1626576042@news.newsguy.com>  , On Fri, 22 Sep 2000 18:14:43 -0400, JF Mezei% <jfmezei.spamnot@videotron.ca> wrote:    >Alan Greig wrote:/ >> And we just received our balls this morning.n >o >I got mine when I was born.  ! But do your's flash when dropped?i     --
 Alan Greig   ------------------------------  # Date: Mon, 25 Sep 2000 10:10:35 GMTi% From: A.Greig@virgin.net (Alan Greig)m Subject: Re: Press Release2 Message-ID: <39cf23ab.1626827514@news.newsguy.com>  , On Fri, 22 Sep 2000 18:17:27 -0400, JF Mezei% <jfmezei.spamnot@videotron.ca> wrote:S  " >steven.reece@quintiles.com wrote:L >> Like Alan G., my ball arrived this morning but that's not the reason I've# >> got very little work done today.  >a> >If you only got one ball, shouldn't you be concerned ???? :-) >eG >Since they said "We're VMS and we've got ballS", should they be givinge >customers at least 2 balls ?.  D Well we did actually get two. Our NT Systems manager received one asE well much to his surprise. He suggested that Microsoft could send outpB their own but with the covering note saying: "One of our customers> recently reminded me how often you have to reboot NT so here's+ something to play with while your waiting".L     --
 Alan Greig   ------------------------------  # Date: Mon, 25 Sep 2000 10:12:10 GMTM% From: A.Greig@virgin.net (Alan Greig)i Subject: Re: Press Release2 Message-ID: <39cf24c6.1627110100@news.newsguy.com>  @ On 22 Sep 2000 23:28:03 GMT, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) wrote:i   >ol >In article <418E68E524A8D311ACCE00508B78866A768178@DENXCH>, Lorin Ricker <Lorin.Ricker@t-netix.com> writes: >:JF Mezei wrote:  >:I >:> Now, if you guys could simply drop that "open" bit to really signal anB >:> departure from Bob Palmer "kill VMS" policies, that would be a >:> major step forwards. >:K >:Bravo!  Hear, Hear!  Yes, good old, plain, proprietary, closed "VMS" that F >:works...  "Open..." just made it all seem so slick and presumptuous! >gF >  Those of us who lived through the prefixification dread the conceptC >  of deprefixification -- just the effort involved in changing thea, >  displays and the documentation boggles.   >DI >  As we have finally seen the original questions and the confusion that SI >  resulted from the "VMS to OpenVMS" discussions reduced, and we are nowaH >  most of the way through the Compaq-related product name changes, now G >  y'all want us to start up this renaming stuff all over again?  Did Im1 >  really understand that request correctly?  :-)4  * How about renaming VMS as "Linux Plus" ;-)   --
 Alan Greig   ------------------------------  % Date: Mon, 25 Sep 2000 17:25:47 +0100R- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>: Subject: Re: Press Release) Message-ID: <39CF7C8B.1A7AACF4@bbc.co.uk>e   Alan Greig wrote:g   >. > , > How about renaming VMS as "Linux Plus" ;-) >u  
 Nah, Linux ++l   :-)p   >i     --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukR  A I speak for myself only and my views in no way represent those of  MedAS or the BBC.2   ------------------------------   Date: 25 Sep 2000 06:28:17 GMT+ From: "Gerke Grashuis" <g.grashuis@kpn.com>s) Subject: Re: remote login using .com filei8 Message-ID: <01c026b9$c9f70bf0$8d4c15ac@HKTGN9911301604>  < Have a look at the RSH/REXEC commands from the TCP/IP suite.% Try $HELP TCP_IP_services R_Commands.s   Gerke.    7 krish <a.krishnamoorthy@digital.com> schreef in artikel-! <8qmnvp$njc@usenet.pa.dec.com>...- > Hello, > K >  I have two machine "A" and "B" at different destinations. Sitting at A iSI > want to run a .COM file , that will automatically execute a   .exe fileaE > residing at "B". One way of doing this is specifying a command line- insideA > the .COM with "command" decnet-nodename, username and password.e > 3 >  eg     $ dir 2.889"user password"::dka0:[000000]e > K > But i would like to know if there is any equivalent way of doing it usings > TCP/IP protocol. >  > Thanks in advance,	 >  -Krish  >  > --B > "Courage is the price that life extracts for granting peace with > yourself."Amelia Earhart >  >  >  >  >    ------------------------------    Date: 25 Sep 2000 10:06:29 -0500, From: koehler@eisner.decus.org (Bob Koehler)) Subject: Re: remote login using .com file + Message-ID: <R14EBcGD5tNN@eisner.decus.org>   Y In article <8qmnvp$njc@usenet.pa.dec.com>, "krish" <a.krishnamoorthy@digital.com> writes:  > Hello, > 3 >  eg     $ dir 2.889"user password"::dka0:[000000]  > K > But i would like to know if there is any equivalent way of doing it using  > TCP/IP protocol. >   &    Current versions of VMS will allow:  7    $ dir/ftp ip-node-name"user password"::dka0:[000000]   B    If the target system is not VMS, put the target path in quotes,    e.g.o  9    $dir/ftp ip-unix-node"user password"::"~auser/somedir"   F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation = Hubble Space Telescope Payload  | Federal Sector, Civil Group E  Flight Software Team           | please remove ".aspm" when replying    ------------------------------  % Date: Mon, 25 Sep 2000 09:30:59 +020065 From: Oswald Knoppers <Oswald.Knoppers@whitehouse.nl>-, Subject: Re: remote tailing of file with UCX- Message-ID: <39CEFF33.FEB48C9B@whitehouse.nl>o   tony_barratt@my-deja.com wrote:S   > $!1 > $ ty/cont/tail=20  dsa0:[oracle7.hnm]health.logT > $! > $! show time > $!exit  D You should add a 'define sys$output sys$net' command before the typeF command. In your example the output of the type command will end up in the logfile.   Regards,   Oswald   ------------------------------  % Date: Mon, 25 Sep 2000 11:47:01 +0100y/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk>t= Subject: re: Replacing system disk, will licenses still work? 7 Message-ID: <009F0A6E.C7F80B26.11@maxwell.ph.kcl.ac.uk>e   > Q > I want to cold-spare a system drive on a VAX4300.  I make a stand-alone backup cM > tape of the existing system image and blast it onto a replacement disk.  I  7 > would think that the licenses should still load, yes?    Yes.   > O > Under what hardware circumstances would the licenses not load, replacement ofEO > motherboard?  Network card?  From unix-land, many licenses like to tie to the . > physical network address.  Does VMS do this?  * None (well, maybe a motherboard *upgrade*)  L the only thing I know that can break your licenses is a hardware change thatK will turn the system into a different class of machine. Back in Q-bus days,uL *removing* the graphics board from a VAxstation turned it into a VAX server,N which wanted more expensive licenses. I don't think there are any "accidental"L hardware changes that can constitute an upgrade on the curent product range.  H Some third-party softwares don't use the LMF and do tie to the Ethernet  hardware address. (Bletch).o  B LMF is, thank goodness, a management aid not an enforcement tool.    	Yours,l
 		Nigel Arnots- 		NRA@MAXWELL.PH.KCL.AC.UK                   a  7 		"In the beginning there was nothing, which exploded."t   ------------------------------  % Date: Mon, 25 Sep 2000 17:14:09 +0100a- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>n= Subject: Re: Replacing system disk, will licenses still work?e) Message-ID: <39CF79D1.A1D6B9B3@bbc.co.uk>r   Murray Dawson wrote:  4 > On Fri, 22 Sep 2000 01:36:28 -0400, David A Froble > <davef@tsoft-inc.com> wrote: >h > >Charles Gilley wrote: > >>S > >> I want to cold-spare a system drive on a VAX4300.  I make a stand-alone backuphO > >> tape of the existing system image and blast it onto a replacement disk.  Ih: > >> would think that the licenses should still load, yes? > > . > >Even better, do an IMAGE copy disk to disk. >0G > Forgive my ignorance, but how would you go about doing an IMAGE copy?  >0  M Hmmm, if you are paranoid enough boot standalone backup, mount the source anda& destination drives on the system, then4 $ backup/image/verify source_disk: destination_disk:  2 All boot blocks, file aliases etc are kept intact.    --96 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uki  A I speak for myself only and my views in no way represent those ofm MedAS or the BBC.o   ------------------------------  % Date: Mon, 25 Sep 2000 14:57:16 +0200l0 From: Didier Morandi <Didier.Morandi@Easynet.fr>* Subject: restart an Alpha ETN42-CA machine* Message-ID: <39CF4BAC.72616030@Easynet.fr>  C I was given a few months ago an ALPHA desktop type ETN42-CA withoutiH terminal or anything else. No cable, no nothing. I decided these days to6 boot it. It is supposed to have VMS 6.something in it.  E As I don't have a VT or something similar to plug into, I'm trying toAF use a vanillia PC with W98 and the HyperTerminal program as the system3 console (is there a system console on a ETN42-CA?).o  G I have some difficulties to do the correct cabling. Should I use port APF and connect it to the COM1 port of the PC? If I should, what should beD the parameters? 8-0-N/XonXoff as usual? Should I use the modem cableG that I have on my PC (actually theone of my wife, I use an iBook :-) ofL a regular RS232/DB25 one?S  : All of this doesn't give any noise on the terminal screen.  F I also tried to connect the PC on the printer port, as before when oneG has a vaxstation without terminal :-). I selected the speed from 300 to  19200, no more success.r  @ Could someone tell me what is the regular hardware configurationD requested to use an ETN42-CA? (what is is, btw, a real Alpha or a PC with an ALPHA chip?)   Thanks,s   Didier (lost in PC land)   ------------------------------  % Date: Mon, 25 Sep 2000 11:54:45 -0400 2 From: rdeininger@mindspring.com (Robert Deininger). Subject: Re: restart an Alpha ETN42-CA machineL Message-ID: <rdeininger-2509001154450001@user-2ive73h.dialup.mindspring.com>  \ In article <39CF4BAC.72616030@Easynet.fr>, Didier Morandi <Didier.Morandi@Easynet.fr> wrote:  E > I was given a few months ago an ALPHA desktop type ETN42-CA withoutfJ > terminal or anything else. No cable, no nothing. I decided these days to8 > boot it. It is supposed to have VMS 6.something in it.  F "ETN42-CA" isn't a very recognizable system name.  It sounds like the E model number from the sticker on the back of the machine.  The "name" I of the system might be more familiar.  What's written on the front of theh system?m   -- w Robert Deininger rdeininger@mindspring.com)   ------------------------------  % Date: Mon, 25 Sep 2000 19:30:31 +0200-0 From: Didier Morandi <Didier.Morandi@Easynet.fr>. Subject: Re: restart an Alpha ETN42-CA machine* Message-ID: <39CF8BB7.B9B8F070@Easynet.fr>   Robert Deininger wrote:s > $ > What's written on the front of the	 > system?e  E haha, nothing but a sticker reading "Alpha generation". It looks like : this (to be read with courier :-) size is the one of a PC.  1    +--------------------------------------------+ 1  /                                            / ! 1 +--------------------------------------------+  !f? !(switch?)         ALPHA !   !F !---------![]! <--On/Off switchg1 ! [x]         Generation !   !L !  CD-ROM !  !  !e1 !------------------------!   !O !---------!  !  !a1 !     L1  L2             !   !P !         !  !  !w1 !(L1:disk icon)          !   !P !         !  !  !,1 !         (L2:bulb icon) !   !Y !         !  !  !t1 !                        !   !  !         !  !  !t1 !------------------------!   !  !         !  !  !l1 !------------------------!   !  !         !  !  !e1 !------------------------!   !  !         !  !  + 0 !------------------------!   !  !         !  ! /. +--------------------------------------------+  E Behind, I have the fan, the alim block, two male A&B ports (DB9), oneoG female C port (DB25), two female ADB ports (similar to the keyboard and E the mouse plugs of a PC), an Ethernet card multi-plugs (AUI, RJ45 andlE thinwire), an SCSI card and an audio card with a DB15 female plug andM. three audio plugs (input, and stereo output ?)   Better description?a   :-)f   D.   ------------------------------  % Date: Mon, 25 Sep 2000 16:18:06 +0000 $ From: Steve.Spires@yellowpages.co.uk Subject: RMS help required/ Message-ID: <00256965.00594C3A.00@quegw01.btyp>    cc:7 bcc:= Contact:   Tel: 3063  -  VSSG, 1st Floor, Bridge Street Plazas   RMS help requiredi    D Can someone point me in the right direction to analyse RMS problems,M specifically how to decode STS and STV values? I found mention of them in thetF RMS manual, but couldn't fathom where I should be looking for details.   TIAn   Steve Spires VMS System Manager BT/Yellow Pagest     [Information] -- PostMaster:D This transmission is intended solely for the addressee(s) and may beL confidential. If you are not the named addressee, or if the message has beenP addressed to you in error, you must not read, disclose, reproduce, distribute or use this transmission.  L Delivery of this message to any person other than the named addressee is notH intended in any way to waive confidentiality.  If you have received thisF transmission in error please contact the sender or delete the message.  
 Thank you.   ------------------------------  % Date: Mon, 25 Sep 2000 11:51:41 -0400e# From: Jim Agnew <agnew@hsc.vcu.edu>f Subject: Re: RMS help required+ Message-ID: <39CF748D.4AD8D54D@hsc.vcu.edu>n   Hey Steve!!!  7 you got to use the secondary error codes in the qio (w)So statement call...  not the status code, but the secondary code.  that should be in the qio call, or the IOSB...o  7       STATUS = SYS$QIOW( %VAL(WR_EF), %VAL(IO_CHANNEL),u@      *                  %VAL(WRITE_FUNCTION), %REF(WR_IOSB), , ,6      *          %REF(COMMAND), %VAL(COMMAND_LENGTH), ,+      *          %VAL(CARRIAGE_CONTROL), , )tD         IF (  .NOT.  STATUS  )       CALL LIB$SIGNAL( %VAL(STATUS) )L         IF (  .NOT.  WR_IOSB.IOSTAT) CALL LIB$SIGNAL( %VAL(WR_IOSB.IOSTAT) )Q       STATUS = LIB$WAIT( DATAPAUSE )            ! Let 9400 baud catch up with us.    any questions, lemme know..    Jim         % Steve.Spires@yellowpages.co.uk wrote:n >  > cc:d > bcc:? > Contact:   Tel: 3063  -  VSSG, 1st Floor, Bridge Street Plazac >  > RMS help requiredv > F > Can someone point me in the right direction to analyse RMS problems,O > specifically how to decode STS and STV values? I found mention of them in thecH > RMS manual, but couldn't fathom where I should be looking for details. >  > TIAr >  > Steve Spires > VMS System Manager > BT/Yellow Pagese >  > [Information] -- PostMaster:F > This transmission is intended solely for the addressee(s) and may beN > confidential. If you are not the named addressee, or if the message has beenR > addressed to you in error, you must not read, disclose, reproduce, distribute or > use this transmission. > N > Delivery of this message to any person other than the named addressee is notJ > intended in any way to waive confidentiality.  If you have received thisH > transmission in error please contact the sender or delete the message. >  > Thank you.   ------------------------------  % Date: Mon, 25 Sep 2000 17:54:38 +0100w- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>u Subject: Re: RMS help required) Message-ID: <39CF834D.89D438C5@bbc.co.uk>n   Hi Steve  & Did you try WRITE SYS$OUTPUT F$MESSAGE9 for the STS and STV values, certainly one is a VMS statusfD code, probably both are (its late, am too lazy to read manuals now).; Then, once you have the VMS message, (possibly with the aidr2 of SET MESSAGE) HELP/MESSAGE  will give you a more- complete descriuption. as I am sure you know.h  > You don't mention what context you are seeing these values. In1 a program you'd probably want to LIB$SIGNAL them.w  " What sort of RMS problems exactly?   Regardse  % Steve.Spires@yellowpages.co.uk wrote:v   > cc:c > bcc:? > Contact:   Tel: 3063  -  VSSG, 1st Floor, Bridge Street Plazay >  > RMS help requiredn >eF > Can someone point me in the right direction to analyse RMS problems,O > specifically how to decode STS and STV values? I found mention of them in thetH > RMS manual, but couldn't fathom where I should be looking for details. >n > TIAs >. > Steve Spires > VMS System Manager > BT/Yellow Pagesm    --a6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uk   A I speak for myself only and my views in no way represent those ofe MedAS or the BBC.i   ------------------------------   Date: 25 Sep 2000 13:47 -0400e From: hein@eps.zko.dec.c*m Subject: Re: RMS help required& Message-ID: <25SEP200013470972@miasys>  X In article <00256965.00594C3A.00@quegw01.btyp>, Steve.Spires@yellowpages.co.uk writes...  E >Can someone point me in the right direction to analyse RMS problems,vN >specifically how to decode STS and STV values? I found mention of them in theG >RMS manual, but couldn't fathom where I should be looking for details.h  Q http://www.openvms.compaq.com:8000/72final/4523/4523pro_001.html#status_code_test M http://www.openvms.compaq.com:8000/72final/4523/4523pro_002.html#types_errors   # Steve, What Language are you using?w  5 If you call RMS directly (typically from C or Macro),sn You can find the values in tge FAB$L_STS and _STV fields for file operations, RAB$L_STS for record operations.r The STS is the main RMS error code, the STV an underlying SYSTEM level code providing further detail if available.  M COBOL, BASIC and the likes MAP the RMS ERR field into generic language codes:w6 	COBOL: Error 30 = something whent wrong somewhere :-) 	BASIC: Error 11 = End of file> Most/all langauages have a funciton to retrieve the real McCoy 	COBOL:    MOVE RMS-STS TO STS.r 		    MOVE RMS-STV TO STV.(     		    MOVE RMS-FILENAME TO FILENAME. 		    DISPLAY FILENAME....% 	BASIC		value% = RMSSTATUS(1%, VALUE)e  , In the DEBUGGER you can use EXAM/COND status  B I tend to use DCL $EXIT for simple error code tyo text translation and F$MESSAGE for fancy stuff.3 Use DCL   HELP /MESS/FAC=RMS for basic explanationst   hth,
      Hein.  
 $ dir qwerfqw ! %DIRECT-W-NOFILES, no files foundh $ show symb $STATUSd   $STATUS == "%X10018290"r $ exit %x018290j %RMS-W-FNF, file not found $ help/mess/out=tt:/fac=rms fnf     FNF,  file not found   7   Facility:     RMS, OpenVMS Record Management Servicesh  2   Explanation:  The specified file does not exist.  H   User Action:  Check the file specification and verify that the device,E                 directory, file name, and file type are all specifiediM                 correctly. If a logical name is specified, verify the currentrL                 equivalence assigned to the logical name. If the equivalenceH                 is correct, verify that the correct volume is mounted onL                 the specified device and that the file was not inadvertently                 deleted.   ------------------------------   Date: 25 Sep 2000 14:12:42 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)- Subject: Seeking OpenVMS Freeware Submissionsa6 Message-ID: <8qnmgq$422$1@mailint03.im.hou.compaq.com>  @   re: Seeking new or updated submissions for OpenVMS Freeware V5    D   Do you have (or know of) useful new tools for OpenVMS, or updates,B   corrections, or new versions of existing OpenVMS software tools?  C   If you know of a new tool or a new tool version that you are NOT sC   responsible for, please let me know and I will work to obtain the0D   permission of the author(s) to include the software tool onto the D   Freeware distribution.  (But I need to know the update or the tool   exists, first.)   1   The OpenVMS Freeware V5 submission deadline is:i  ;      27 November 2000 (The Monday following the US Holiday)v  H   The earlier you can get the submission in ahead of the November date, I   the better and easier it will be.  (The later the submission, the less iJ   time I will have available to assist with and to deal with any transfer J   or kitting or configuration or documentation problems that might arise.)  B   Freeware submission information and guidelines are available at:  3     http://www.openvms.compaq.com/openvms/freeware/n  	   Thanks!p  	 	-- ps --   G   Most of (all of?) the software kits that have not been updated since fE   the OpenVMS Freeware V3 distribution will be removed from V5.  (An  0   encouragement to update older kits?  You bet!)  G   I have access to an FTP transfer area, where you can push submissions)G   to Compaq, or I can access FTP (anonymous or otherwise) sites to pulliG   kits, or kits can be accepted via various media and courier services..  J   Followups to this message have been set to: vmsnet.sdk.openvms.fieldtest  G   And apologies in advance: You will likely see copies of this message e<   posted a couple of more times between now and 27-Nov-2000.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 25 Sep 2000 15:37:53 +0100r, From: Ted Allwood <support@leva.leeds.ac.uk>1 Subject: re: Seeking OpenVMS Freeware Submissions 3 Message-ID: <009F0A8F.08499D87.23@leva.leeds.ac.uk>S   Stephen Hoffman writes  B >   re: Seeking new or updated submissions for OpenVMS Freeware V5 > E >   If you know of a new tool or a new tool version that you are NOT oE >   responsible for, please let me know and I will work to obtain thewF >   permission of the author(s) to include the software tool onto the F >   Freeware distribution.  (But I need to know the update or the tool >   exists, first.)/  C Dave Raggett's 'HTML Tidy' is a handy utility for fixing bloated or > broken html files - see http://www.w3.org/People/Raggett/tidy/ I'm using under VMS 7.1-2a  eA HTML Tidy expects to use stdin and stdout for file i/o so ideallyaB it could do with a DCL wrapper to make it a bit more VMS-friendly.     Regards, Ted0  K -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-sK Support@leva.leeds.ac.uk                                Tel:  0113 233 2167lG School of Mechanical Engineering,  University of Leeds,  Leeds  LS2 9JT0   ------------------------------  % Date: Mon, 25 Sep 2000 10:46:54 -0700l& From: bill robertson <wcr@pacbell.net>: Subject: Re: Son of BUT... (Re: Why are these files open?)+ Message-ID: <39CF8F8D.47AEC060@pacbell.net>    bill robertson wrote:a  0 > So, how do I get rid of those INSTALL entries?  R I received a number of posts in response to this, which I'll attempt to summarize:  8 "David J. Dachtera" <djesys.nospam@earthlink.net> writesI | "That I know of, the only way to clear these out is gonna be a reboot."t  U I can't say with absolute certainty, but I believe, that the system has been rebooted*7 several times since these version 1 files were present.X    `  "John E. Malmberg" <wb8tyw@qsl.net>: "Use the INSTALL utility to list out the |files.  Note thea exact syntax for the entry for MENU_SIGNON.EXE including the |angle brackets that are surroundingt the directory. | < |$install remove disk$dev1:<RLS_MTS_45.EXE>menu_signon.exe;1 |nL |The device, the directory in the same angle brackets as the install utility> |reports, and the exact version number seem to be significant.   andn  /  Chris Scheers <chris@applied-synergy.com> says-B |Try using INSTALL REMOVE to get rid of the reference to the file.   Well...w# (DEL and REMOVE w/ version numbers)j9 $ install del disk$dev1:<rls_mts_45.exe>menu_signon.exe;1 U %INSTALL-E-OPENIN, error opening DISK$DEV1:<RLS_MTS_45.EXE>MENU_SIGNON.EXE;1 as input2 -RMS-E-FNF, file not foundW %INSTALL-W-FAIL, failed to REMOVE entry for DISK$DEV1:<RLS_MTS_45.EXE>MENU_SIGNON.EXE;1o9 -INSTALL-E-ILLFILEVER, file version numbers not permittedo< $ install remove disk$dev1:<rls_mts_45.exe>menu_signon.exe;1U %INSTALL-E-OPENIN, error opening DISK$DEV1:<RLS_MTS_45.EXE>MENU_SIGNON.EXE;1 as inpute -RMS-E-FNF, file not foundW %INSTALL-W-FAIL, failed to REMOVE entry for DISK$DEV1:<RLS_MTS_45.EXE>MENU_SIGNON.EXE;1g9 -INSTALL-E-ILLFILEVER, file version numbers not permittedo   (Same w/out version numbers)7 $ install del disk$dev1:<rls_mts_45.exe>menu_signon.exekU %INSTALL-W-FAIL, failed to REMOVE entry for DISK$DEV1:<RLS_MTS_45.EXE>MENU_SIGNON.EXE / -INSTALL-E-NOKFEFND, Known File Entry not founda: $ install remove disk$dev1:<rls_mts_45.exe>menu_signon.exeU %INSTALL-W-FAIL, failed to REMOVE entry for DISK$DEV1:<RLS_MTS_45.EXE>MENU_SIGNON.EXE / -INSTALL-E-NOKFEFND, Known File Entry not foundp  6 Lastly, Tim Llewellyn <tim.llewellyn@bbc.co.uk> writesF |Are you SURE the disk is not mounted on another cluster member, which |also has the image installed?    System is not part of a cluster.  [ Let me thank you all again for your willingness to help.  Fortunately, the situation is noteW causing any harm, so this is sort of an exercise.  I remain willing to try (almost) anyt suggestions.   -billI   ------------------------------    Date: 25 Sep 2000 09:59:44 -0500, From: koehler@eisner.decus.org (Bob Koehler)# Subject: Re: SRAM Corruption Threade+ Message-ID: <+SWVqsHs0JvK@eisner.decus.org>   X In article <39CCF446.E14021C6@bellatlantic.net>, hg/jb <shsrms@bellatlantic.net> writes:F > Exactly my thought!  there had been some work on this in years past,H > and some speculation that the chip densities could show up the effects > of normal particle hits.  J    I don't believe it.  I didn't believe it when DEC Field Service blamed D    cosmic radiation for their VAX 8000 series memory problems, and I    don't believe it now.  H    Why?  Because I did the calculation back when I was a grad student inK    physics using cosmic rays as a test source for a detector I was buildingcG    and my SEL 32/55 reminded me of a detector.  (The memory boards weren/    mounted horizontally in an 19 inch chassis.)   E    We're still several orders of magnitude away from memory densities H    that should show cosmic ray problems, provided your system is located!    near the surface of the earth.r  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationn= Hubble Space Telescope Payload  | Federal Sector, Civil GroupeE  Flight Software Team           | please remove ".aspm" when replyingt   ------------------------------  # Date: Mon, 25 Sep 2000 14:10:09 GMTs5 From: "Larry D Bohan, Jr" <LBohan@dbc.spam_less..com>t# Subject: Re: SRAM Corruption Threade8 Message-ID: <otmusssu3icdjvebekmtbafuvnq149hbhb@4ax.com>  E On 25 Sep 2000 09:59:44 -0500, koehler@eisner.decus.org (Bob Koehler)f wrote:  Y >In article <39CCF446.E14021C6@bellatlantic.net>, hg/jb <shsrms@bellatlantic.net> writes:hG >> Exactly my thought!  there had been some work on this in years past,-I >> and some speculation that the chip densities could show up the effectsi >> of normal particle hits.E >nK >   I don't believe it.  I didn't believe it when DEC Field Service blamed .E >   cosmic radiation for their VAX 8000 series memory problems, and Io >   don't believe it now.m > I >   Why?  Because I did the calculation back when I was a grad student inhL >   physics using cosmic rays as a test source for a detector I was buildingH >   and my SEL 32/55 reminded me of a detector.  (The memory boards were0 >   mounted horizontally in an 19 inch chassis.) >-F >   We're still several orders of magnitude away from memory densitiesI >   that should show cosmic ray problems, provided your system is located-" >   near the surface of the earth. >1G >----------------------------------------------------------------------!@ >Bob Koehler                     | Computer Sciences Corporation> >Hubble Space Telescope Payload  | Federal Sector, Civil GroupF > Flight Software Team           | please remove ".aspm" when replying    C though, I once did see 7 systems (at one site) take soft/corrected !2 memory errors all at the same time.  (late 1980's)  D One wonders what the ultimate cause was; the stuff of urban legends.   ------------------------------    Date: 25 Sep 2000 13:06:11 -05002 From: malmberg@eisner.decus.org (John E. Malmberg)# Subject: Re: SRAM Corruption Threadh+ Message-ID: <ldKo$PYy6Xd3@eisner.decus.org>i  8 In article <otmusssu3icdjvebekmtbafuvnq149hbhb@4ax.com>,7 "Larry D Bohan, Jr" <LBohan@dbc.spam_less..com> writes:e >  dE > though, I once did see 7 systems (at one site) take soft/corrected o4 > memory errors all at the same time.  (late 1980's) > F > One wonders what the ultimate cause was; the stuff of urban legends.  1 I would tend to suspect an electromagnetic cause.   I Lightning does not have to hit a target to do significant damage.  It cannO induce briefly several thousand volts from an inductive coupling.  The damage Ie% have seen has been quite interesting.n    N At one site 10% of the CXY08's that were quite well electrically isolated fromO the rest of the world died as a result of a lightning storm passing by.  It hasi not been repeated since.  M It is quite possible a spike got through the power system that just happen toyN match a resonant circuit for that model of memory with just enough power.  And# probably a hard thing to reproduce.e      B Electromagnetic is more commonly considered Radio Frequency or RF.  H The rules for making an RF safe design tighten up considerably at modern processor and cache speeds.o  O If a circuit is exposed to a frequency it is resonant on, it does not take much " energy at all to induce a current.  E And the resonant frequency can be based on current temperature of the  components.h  N And the humidity, and other external factors can make the paths appear random.  I If a UHF or Microwave RF source is causing an equipment malfunction, theniN either moving the equipment to a less likely path, or changing the temperatureN would appear to cure the problem.  I would expect moving the equipment to have' more of an effect than the temperature.       N Most equipment is designed to prevent such interference from occuring.  But itF is not perfect.  A worse problem is when the interference is generatedM internally.  Or the internal suppression is just good enough under laboratory O conditions when coupled with the shielding, but not good enough for a few sites(N out in the field.  If the failure occurs as infrequently as a few times a day,J it can be almost impossible to find a Radio Frequency Interference source.      M It is interesting to watch a VT Terminal or SVGA Monitor screen convulse whenMN a standard low power 800 Mhz walkie-talkie is keyed up next to it.  But I also5 note that the data circuitry has never been affected.v    L It makes you wonder about where some of the higher power transmitters may be- hidden in relationship to your computer room.   G In a large building, in order to provide good coverage, there are smallnM cell-phone relays and pager service relays typically installed.  And if therenM are two-way radio's in use, typically they are now very close to a cell phone M in operation.  These repeaters all have more power than an the walkie-talkie.d    N I know that in order to be sold, that equipment must meet certain RF shieldingH rules, and the ones for the datacenter are not as tight as for the home.  N And that same sheilding also protects the electronics on the inside to a greatN degree.  I do not know if the manufacturers of datacenter class computers testN their systems against being interfered by cell phone and other UHF / MicrowaveM repeaters.  Let alone all of the other types of transmitters that are around.b       -John  wb8tyw@qsl.network   ------------------------------  % Date: Mon, 25 Sep 2000 17:37:08 +0100s- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>t# Subject: Re: SRAM Corruption Thread ) Message-ID: <39CF7F34.2D9BB2C6@bbc.co.uk>b   Bob Koehler wrote:  Z > In article <39CCF446.E14021C6@bellatlantic.net>, hg/jb <shsrms@bellatlantic.net> writes:H > > Exactly my thought!  there had been some work on this in years past,J > > and some speculation that the chip densities could show up the effects > > of normal particle hits. >sK >    I don't believe it.  I didn't believe it when DEC Field Service blamed,F >    cosmic radiation for their VAX 8000 series memory problems, and I >    don't believe it now. >eJ >    Why?  Because I did the calculation back when I was a grad student inM >    physics using cosmic rays as a test source for a detector I was building I >    and my SEL 32/55 reminded me of a detector.  (The memory boards werem1 >    mounted horizontally in an 19 inch chassis.), >eG >    We're still several orders of magnitude away from memory densities J >    that should show cosmic ray problems, provided your system is located# >    near the surface of the earth.y >e  J Yup, but there's always a slim chance that something highly energetic from+ the tail of the distribution hits your RAM.r  B But I thought this was why we all feel safe, having ECC RAM in our VAX and Alpha systems?  > Or, is this thread about effects that ECC RAM does not handle?   Regardss   --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukn  A I speak for myself only and my views in no way represent those oft MedAS or the BBC.m   ------------------------------  % Date: Mon, 25 Sep 2000 17:39:09 +0100e- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>t# Subject: Re: SRAM Corruption Threada) Message-ID: <39CF7FAD.180D433A@bbc.co.uk>t   "Larry D Bohan, Jr" wrote:  E >  though, I once did see 7 systems (at one site) take soft/correctedr4 > memory errors all at the same time.  (late 1980's) >m  J  now, what was the name of that supernova they detected in the underground) neutrino decay expts? That was late 80's.e     --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uko  A I speak for myself only and my views in no way represent those of  MedAS or the BBC..   ------------------------------  % Date: Mon, 25 Sep 2000 18:00:26 +0100S- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>B# Subject: Re: SRAM Corruption Thread ) Message-ID: <39CF84AA.3A1B4FC5@bbc.co.uk>d   "John E. Malmberg" wrote:$  Q >  And that same sheilding also protects the electronics on the inside to a great,P > degree.  I do not know if the manufacturers of datacenter class computers testP > their systems against being interfered by cell phone and other UHF / MicrowaveO > repeaters.  Let alone all of the other types of transmitters that are around., >e  S An ex-DEC now-Compaq engineer I've known for over a decade told me he only ever had   H problems with the very early analogue phones, he is in and out of serverG rooms all day taking calls on his mobile and never had an issue with an & unexpected crash with a digital phone.  I Of course, if you are REALLY paranoid, you force people to turn off their K mobiles before entering the server room, but this is a real PITA especially J if you have a crisis that requires console interaction and no phone in the
 machine room.s   --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uki  A I speak for myself only and my views in no way represent those of. MedAS or the BBC.    ------------------------------  # Date: Mon, 25 Sep 2000 17:30:15 GMT - From: "Richard D. Piccard" <piccard@ohio.edu>p# Subject: Re: SRAM Corruption Thread ' Message-ID: <39CF8B9E.9DBCE2A@ohio.edu>   [ Another scenario besides cosmis rays is trace radioactive contamination of the packaging ofoY the integrated circuit.  Ceramics are probably more likely than plastics to contain heavyu[ metals.  A few-MeV alpha particle smashing directly into the chip though a distance of lessh> than 1 mm of air might be expected to cause some consequences.  V Many years ago I worked with a sensitive light detector (image intensifier tube) whoseX input was coupled to a fiber-optics blank on whose surface was a ZnS radiation detector.Q The surface of the fiber-optics blank got flat by being polished.  We were deeplyt[ suspicious that the polishing rouge had trace radioactive contamination, because there werepQ more or less random point flashes of very high energy that we could find no othera explanation for.  #                                 RDP/     Bob Koehler wrote:  Z > In article <39CCF446.E14021C6@bellatlantic.net>, hg/jb <shsrms@bellatlantic.net> writes:H > > Exactly my thought!  there had been some work on this in years past,J > > and some speculation that the chip densities could show up the effects > > of normal particle hits. >uK >    I don't believe it.  I didn't believe it when DEC Field Service blamedoF >    cosmic radiation for their VAX 8000 series memory problems, and I >    don't believe it now. > J >    Why?  Because I did the calculation back when I was a grad student inM >    physics using cosmic rays as a test source for a detector I was buildingrI >    and my SEL 32/55 reminded me of a detector.  (The memory boards weree1 >    mounted horizontally in an 19 inch chassis.)  >eG >    We're still several orders of magnitude away from memory densities-J >    that should show cosmic ray problems, provided your system is located# >    near the surface of the earth.  >jH > ----------------------------------------------------------------------A > Bob Koehler                     | Computer Sciences Corporationk? > Hubble Space Telescope Payload  | Federal Sector, Civil GroupfG >  Flight Software Team           | please remove ".aspm" when replyingp   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  % Date: Mon, 25 Sep 2000 18:49:25 +0100g* From: "Richard Brodie" <R.Brodie@rl.ac.uk># Subject: Re: SRAM Corruption Threadd+ Message-ID: <8qo374$uhq@newton.cc.rl.ac.uk>h  ^ "Tim Llewellyn" <tim.llewellyn@bbc.co.uk> wrote in message news:39CF7F34.2D9BB2C6@bbc.co.uk...  D > But I thought this was why we all feel safe, having ECC RAM in our > VAX and Alpha systems?  ; Unfortunately the ECC part is the main memory DRAM. SRAM isi? typically used for cache, and may or may not be ECC throughout.-B And empirically, they do seem to be sensitive to stray high energy particles :(   ------------------------------  # Date: Mon, 25 Sep 2000 10:48:08 GMTe= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)o' Subject: Re: Suggested VMS new feature!N0 Message-ID: <009F0A45.06DF6A66@SendSpamHere.ORG>  X In article <jgh3yuTJwCt2@eisner.decus.org>, young_r@eisner.decus.org (Rob Young) writes:y >In article <Fbyz5.3201$s76.243319@bgtnsc06-news.ops.worldnet.att.net>, "John Nixon" <jorlnixon@worldnet.att.net> writes: M >> I don't have any problem with third party products.  I push for them and Ic* >> wish there were more available for VMS. >> nL >> However, DELETE is a native VMS command, and I would not feel comfortableI >> buying or writing a replacement for it.  Who knows what may occur in a|D >> future version of VMS that will invalidate my new DELETE command. >> dM >> I am glad to see that my original post garnered such response, but I dont'hI >> really understand all the criticism of it.  To me it seems like a veryoB >> logical extension that would be very beneficial to many people. >> - >e >.: >	Maybe they were too harsh.  Brian Schenkenberger pointedA >	out that he has delete/confirm as a global.  I take a different-  I A permanent global at that!  It cannot be inadvertently deleted restoringe+ the non-confirmational version of DELETE!  i  K This makes VMSINSTAL, DECW$STARTUP RESTART (used every freakin' time I have K to use NetScape) and myriad other procedures cough up confirmation messages.K but at least I know that it is still defined.  VMSINSTAL can be a real PITAa3 but I work around that by spawning off VMSINSTAL.  i     --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM.            lO city, n., 1. a place where trees are cut down and streets are named after them.s   ------------------------------  # Date: Mon, 25 Sep 2000 10:35:01 GMTB% From: A.Greig@virgin.net (Alan Greig)I' Subject: Re: Suggested VMS new feature!g2 Message-ID: <39cf2762.1627777660@news.newsguy.com>  E On Fri, 22 Sep 2000 22:45:06 -0400, "Bill Todd" <billtodd@foo.mv.com>o wrote:  K >My dim recollection is that TOPS-10 Delete (as opposed to Expunge - a termvK >which at least in me elicits a more appropriately respectful reaction thanoG >'Delete/MakeItSo') was accompanied by automatic space reclamation when H >necessary.  With quotas enabled, such a facility would presumably startI >reclaiming space as necessary (preferably oldest-deleted-first) to avoid  >exceeding quota.w  ? That was TOPS-20 not TOPS-10. A deleted file continued to count @ against a users disk quota. By defaullt the LOGOUT command would@ EXPUNGE deleted files but you could override this. If the systemD detected that it was crtically low on disk space a broadcast messageE was sent warning "Starting expunge of deleted files in 60 seconds". AyC user could also specifically expunge deleted files with the expungeC5 command  or delete/expunge a specific file in one go.e  F TOPS-20 also better implemented quotas in that there was no need for a? program to retry a request in order to use the overdraft quota.AF Instead you could automatically use up to your logged in quota but had* to tidy up or expunge before logging out.   E This was just another little feature that I think VMS should have hadw from the start.a   --
 Alan Greig   ------------------------------  # Date: Mon, 25 Sep 2000 11:04:35 GMT % From: A.Greig@virgin.net (Alan Greig)t' Subject: Re: Suggested VMS new feature!h2 Message-ID: <39cf2ae0.1628671595@news.newsguy.com>  @ On Mon, 25 Sep 2000 00:08:13 GMT, system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) wrote:e    I >The UNDELETE request seems to come from the single-use PeeCee mindset.  ,H >What do you do if one morning all the users on the system login and, inI >a pre-coffee stupor, delete their LOGIN.COM file and request the operatocI >to restore it?  Which LOGIN.COM of several hundred in the WASTEBASKET dor >you restore and to whom???   D That's why I woudn't  have a wastebasket folder but rather just flagC the file for delete and have them hidden to normal file operations.d= The user could then just issus the hypothetical VMS command $n6 DIR/DELETED command and $ UNDELETE the approprate one.     --
 Alan Greig   ------------------------------    Date: 25 Sep 2000 10:31:32 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) ' Subject: Re: Suggested VMS new feature!g+ Message-ID: <QctzSzE17AZ1@eisner.decus.org>i  Z In article <39cf2ae0.1628671595@news.newsguy.com>, A.Greig@virgin.net (Alan Greig) writes:B > On Mon, 25 Sep 2000 00:08:13 GMT, system@SendSpamHere.ORG (Brian! > Schenkenberger, VAXman-) wrote:a >  > J >>The UNDELETE request seems to come from the single-use PeeCee mindset.  I >>What do you do if one morning all the users on the system login and, in J >>a pre-coffee stupor, delete their LOGIN.COM file and request the operatoJ >>to restore it?  Which LOGIN.COM of several hundred in the WASTEBASKET do >>you restore and to whom??? > F > That's why I woudn't  have a wastebasket folder but rather just flagE > the file for delete and have them hidden to normal file operations. ? > The user could then just issus the hypothetical VMS command $28 > DIR/DELETED command and $ UNDELETE the approprate one.  O And if a new file with that name and version had been created in the meantime ?c   ------------------------------  % Date: Mon, 25 Sep 2000 14:28:38 +0000M- From: SysAdmin <djesys.nospam@fsi.net.mapson>i' Subject: Re: Suggested VMS new feature!f. Message-ID: <39CF6116.5AC4E953@fsi.net.mapson>   "Richard B. Gilbert" wrote:i > : > Message text written by INTERNET:system@SendSpamHere.ORGI > >The UNDELETE request seems to come from the single-use PeeCee mindset.nI > What do you do if one morning all the users on the system login and, indJ > a pre-coffee stupor, delete their LOGIN.COM file and request the operatoJ > to restore it?  Which LOGIN.COM of several hundred in the WASTEBASKET do > you restore and to whom??? > <n > ( > Don't these files still have an owner?  > The owner UIC may or may not have a relationship to the sourceG directory. Multiple users - each with a different login directory - can  share the same UIC.t   David J. Dachterae   ------------------------------   Date: 25 Sep 2000 15:17:18 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)' Subject: Re: Suggested VMS new feature!i, Message-ID: <8qnq9u$luq@gap.cco.caltech.edu>  Z In article <39CED2FC.CED5A82D@tsoft-inc.com>, David A Froble <davef@tsoft-inc.com> writes: >eP >Sorry, I guess I wasn't clear enough.  In the 'wastebasket', the saved filename7 >would include the directory it came from, for example:  >  >	DISK0:[DFE.TEMP]LOGIN.COM;34 >d >would be named: >d >	DFE_TEMP___LOGIN.COM_34;1r >o  K If you're willing to accept that every disk (or logical equivalent) should s6 have it's own wastebasket, then you could have instead  )   DISK0:[WASTEBASKET.DFE.TEMP]login.com;1m  H with no need for name munging and the date of the file would reflect theK time of deletion.  The version number changes to 1 because it is "the first G deleted version of this file".  If the same file, with the same versioneH number was deleted you would then get the ;2 variant in the wastebasket.K (And one might expect something like /versions=10 to be in effect for these=J directories.)  If you decided that you needed to recover the "-3" deleted - version this would find the correct file withi   $ undelete login.com;-3   G However, it would probably only be safe to restore it as login.com;35,  + since the ;31 might already exist (again!).-  F That each disk has it's own wastebasket doesn't mean that the undeleteI utility can't provide a way to view the ensemble in some way.   I'd also -' expect, for a well integrated facility:T   $ dir/delete/since="-5-"  I to show me all files living in the wastebasket, deleted from the current h directory, in the last 5 days.   Regards,   David Mathog mathog@seqaxp.bio.caltech.eduo? Manager, sequence analysis facility, biology division, Caltech l   ------------------------------  % Date: Mon, 25 Sep 2000 11:26:05 -0400t2 From: rdeininger@mindspring.com (Robert Deininger)' Subject: Re: Suggested VMS new feature!-L Message-ID: <rdeininger-2509001126050001@user-2ive73h.dialup.mindspring.com>  Y In article <39cf2ae0.1628671595@news.newsguy.com>, A.Greig@virgin.net (Alan Greig) wrote:.    F > That's why I woudn't  have a wastebasket folder but rather just flagE > the file for delete and have them hidden to normal file operations..? > The user could then just issus the hypothetical VMS command $o8 > DIR/DELETED command and $ UNDELETE the approprate one.  G This is the best of the ideas I have seen so far.  Alas, I doubt anyone K but VMS engineering can implement it.  "Hidden from normal file operations"oM implies some nontrivial additions to RMS.  If the default behavior is "ignore I hidden files", some utilities will probably need to, by default, overrideDQ the RMS defaults.  Backup comes to mind immediately, but there are probably more.n  J Support for remote (DECnet) RMS might need to be expanded as well.  If youD want to DIR/DELETED node::[a.b], the "hidden" attribute will have to be known to FAL.  I None of this is impossible, but the VMS engineers might not have too mucht free time right now.    F As for home-grown solutions, if deleted files are to be "hidden" in a  "trash" directory, I think the directory should belong to the individual user instead of being shared.  There would have to be a directory (or theC possibility of creating one) on each disk.  You could use a logicalO@ name username_TRASH_DIR, which the delete-to-trash command wouldL translate to find the trash directory for the user.   The logical name couldD be defined either by the system manager or the individual user.  TheO delete-to-trash command renames the files to the trash directory on the currentKM disk, and the undelete command brings them back.  The files always belong theu7 the original owner, are charged against his quota, etc.n  J The files in the trash should remember where they came from.  Perhaps thisM information could be saved in an application-specific ACL.  One tricky point:EI if the user renames the files from the trash to somewhere else, they neede% to "forget" their original directory.e  M (I use the Macintosh terminology "Trash", rather than the me-too, stolen-ideaE0 name "Wastebasket" prefered by the Gates-beast.)   --   Robert Deininger rdeininger@mindspring.comR   ------------------------------  % Date: Mon, 25 Sep 2000 17:11:34 +0100m- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>e' Subject: Re: Suggested VMS new feature! ) Message-ID: <39CF7936.DEBBC27E@bbc.co.uk>a  & "Brian Schenkenberger, VAXman-" wrote:  I >  The UNDELETE request seems to come from the single-use PeeCee mindset. I > What do you do if one morning all the users on the system login and, in0J > a pre-coffee stupor, delete their LOGIN.COM file and request the operatoJ > to restore it?  Which LOGIN.COM of several hundred in the WASTEBASKET do > you restore and to whom??? >/  D Hmmm, this is VMS, not a single user M$ system, one would assume theL mythical integrated undelete utility would be designed to cope with multiple users.  J Then again, I do backups and am careful with delete and friends, just like I am with a sharp tool.n   RegardsS --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukt  A I speak for myself only and my views in no way represent those ofr MedAS or the BBC.    ------------------------------  % Date: Mon, 25 Sep 2000 15:03:43 +0100m0 From: andrew harrison <andrew.nospam@uk.sun.com>* Subject: Re: Sun Hardware problems persist* Message-ID: <39CF5B3F.1B79D18D@uk.sun.com>   "Main, Kerry" wrote: >  > Andrew, Andrew ..c > ( > Feeling a bit warm these days are we ? >  > :-)- >  > re: your quote on eTrade ..- > M > >>> But eTrade did not have one outage associated with this issue, they hado* > a number each of which lasted hours. >>> > N > Yep, a long time ago (Feb 99 timeframe) they went through some configurationK > issues over the total space of about 4-5 days. No question. No vendor can0, > say they have NEVER had hardware problems. > A No Kerry that is not true, the eTrade cluster hung on a number ofm
 occasions,B this may or may not have been caused by configuration issues. The C actual cause was rumoured to be a Compaq engineer making a mistake 1B upgrading the cluster and the cluster getting wedged. But this was@ only rumour. What is not disputed is that the cluster hung when  it apparently should not have.  B It may well have been configuration issues that caused the clusterH to hang, however given that the general concensus on this group is that > clusters don't hang I am sure you can see that this unexpected# outage was very damaging to Compaq.      > Nope.w > M > Are a good deal of them ? Don't know, but press articles do offer some very J > interesting insight. As I recall, a number of the posted articles quotedN > eBay spokespeople as stating cpu hardware or database server issues were the
 > problem. > = How do you know, you see the whole point about the whole eBays? FUD storm is that the people who accused Sun of being the causes< of eBays problems have done so without any facts to back up ? their case. In the only case where any details have emerged the-8 famous 22 hour outage Sun was shown not to be at fault.   > People like you say eBay has had an outage, eBay runs on Sun's) therefore the Sun's caused eBays outage.    @ This line of reasoning ignores the fact that a significant part D of eBays infrastructure does not run on Sun's, Web, CGI etc servers @ are actually hosted on NT and that in a very complex e-Commerce A stack like eBays there are lots of other possibilities that could   cause them to lose their system.  ; And the Press articles do not offer an interesting insight r; unless you consider publishing articles that turned out to e< be totally factually incorrect as being somehow insightfull.  ; I don't need to remind you that all the articles published L< at the time of the 22 hour outage which suggested that there: was some Sun supplied software of hardware problem at the 8 root of the failure were incorrect, so much for insight.  J > Anyway, this subject of the cache issues has likely started to bore many
 > readers. > N > The bottom line is that, imho, when faced with hardware (or software) issuesM > that impact many Customers, it is the responsibility of the vendor to admito8 > it with a formal statement, then fix it, then move on. > D So will Compaq admit that eTrade's outage was caused by Compaq ????? ????????   regardse Andrew Harrisoni Enterprise IT Architectn   ------------------------------  % Date: Mon, 25 Sep 2000 17:25:37 +0100s0 From: andrew harrison <andrew.nospam@uk.sun.com>* Subject: Re: Sun Hardware problems persist* Message-ID: <39CF7C81.C43D3012@uk.sun.com>   David A Froble wrote:n >  > andrew harrison wrote: > >e > > David A Froble wrote:eR > > > Trying to put a bit of constraint on things.  Really people, there should beQ > > > some limits on 'Andrew kicking'.  After all, he's just doing his job, righteO > > > Andrew? :-)  None of the rest of us would like it if people started goingnV > > > further than a bit of subtle suggestion when beating us up about our job.  Let's: > > > try to avoid some of the rather explicit stuff, huh? > > >h > >e7 > > Thanks but don't worry, I take the view that peoplea8 > > who start the personal abuse flame war do so because7 > > they know conciously of unconciously that they havel8 > > lost the argument or that it is not going as well as > > they hoped.  > R > So, I can assume, since you do not deny it, that spreading FUD on comp.os.vms IS > your job?c >   8 Sorry David, I origionally got my reputation for OpenVMS; bashing because of my attacking Robs ludicrous claims over l@ Spiralog and other OpenVMS Digital technologies that ultimately + have not delivered on their origional hype.o  ; I objected at the time and sadly either for good technical a6 reasons (Spiralog/WildFire) or for good market reasons: (OpenVMS in general/Galaxies) my FUD as you called it has ) been vastly more accurate than the hype. l  > Then came 22 hour eBay outage which one of the major anti-Sun B FUDsters and spin doctors has now admitted was all the injudicious? cutting and pasting of CNET articles and wasn't actually caused 8 by Sun HW or SW. I objected strongly at the time and got8 flamed for my objections. Lets just say that the flamers were at best badly led.p  < So sorry if you call this FUD then you really are in trouble8 you see in my book FUD is untrue, and making a statement: that is true cannot be FUD. Seems to me that you should be< directing your criticisms at the real FUD merchants on this < group who are the people who let their desire to boost Alpha( and OpenVMS get in the way of the facts.  ; Just as an example consider the claim that eBay were gaggeda> by Sun to stop them revealing the real causes of their outage.  : It is now common knowledge that the outage was not caused = by Sun HW or SW and that in fact Sun had more to gain by the h< real cause of the outage being revealed. However Sun did not@ reveal what the cause was because we had a customer and partners to protect.r  ? Contrast this with the eTrade outage, just as serious as eBay, i< but no one has ever suggested that eTrade themselves or any < of the software vendors were responsible for the outage. The? general opinion is that it was caused by OpenVMS cluster SW andy? a mistake by a Compaq engineer, but this information has never r> been made public, despite the fact that Compaq has no partners= that will look silly if the information as to the exact causeo is made public.i  8 So where is the URL detailing the reason for the eTrade 4 outage so that Compaq customers can be sure it does 6 no happen to them ? shoudn't everone know ? why don't 7 they ? could it be that eTrade are under NDA and Compaq-5 don't want to admit that they messed up ? is there a r3 conspiracy going on here ?????? Perhaps Compaq onlyg4 agreed to fix the eTrade system if they didn't tell  anyone why it failed.   6 Now this of course is FUD because I don't have a clue 5 if any of the suggestions are correct but it is just T* the same as the FUD directed at Sun :):):)   Regardsa Andrew Harrisonc Enterprise IT Architectr   ------------------------------  % Date: Mon, 25 Sep 2000 10:04:11 -0400s- From: "Peter Weaver" <peter.weaver@stelco.ca>sV Subject: Re: Top "F" Reasons OpenVMS isn't going to Die  (WAS:Re: [VMS] VAX is  dying)/ Message-ID: <ssumsi6r1irh29@corp.supernews.com>q  . "Matt Morley" <Matt@MPCM.com> wrote in message" news:39CC1985.2CE7499F@MPCM.com... >...E > away with it, until I understand it. But not knowing it, I going to ? > figure out hardware... and cheap at that (us college studentsc struggle > for hardware). >...  A There are always VAX and Alpha machines on e-Bay, but if you want6A really cheap hardware, then visit www.charon-vax.com and downloadwD their VAX emulator. Order the CDROM from DECUS and get your hobbyist licenses and away you go.m   ------------------------------  + Date: Mon, 25 Sep 2000 15:15:12 +0000 (   )r3 From: Christopher Smith <chriss@Mufasa.pubserv.com>uY Subject: Re: Top "F" Reasons OpenVMS isn't going to Die  (WAS:Re: [VMS] VAX isdying)dyingtH Message-ID: <Pine.LNX.4.05.10009251512410.942-100000@Mufasa.pubserv.com>  - On Fri, 22 Sep 2000, David J. Dachtera wrote:-  H > How do we convince "them" to stop "preaching to the choir" and instead  > "put the word in the streets"?  E That gives an image of somebody who's poofed their hair up and shaved." "VMS" in the back of their head...  3 On the other hand, it is early monday morning here.e  J Seriously, though, we probably want the word in the executive lounge where2 all the pointy-hairs and bean counters can see it.   Regards,   Chris   O ===============================================================================u@ "My two cents"			(http://rootworks.com/twocentsworth.cgi?128562)= Christopher Smith(chriss@pubserv.com)			Prgramer^W Programmert Prime Synergy of Champaign, IL.k% -------------------------------------eI "Where a calculator on the ENIAC is equipped with 18,000 vacuum tubes anddH weighs 30 tons, computers in the future may have only 1,000 vacuum tubes; and weigh only 1.5 tons." -- Popular Mechanics, March 1949 -O -------------------------------------------------------------------------------r   ------------------------------  % Date: Mon, 25 Sep 2000 08:08:42 +0100e< From: John Macallister <j.macallister1@physics.oxford.ac.uk>5 Subject: UNDELET (Was Re: Suggested VMS new feature!).7 Message-ID: <000925080842.ae4c@nplvms.physics.ox.ac.uk>G  R The UNDELETE recycle bin in VMS could simply be a parallel directory tree on disk.P For example [000000] would be the tree of normal files and [777777] might be theR  recyle bin files. It would of necessity have to be limited i.e. a certain amount P  of disk space would be allocated to the whole recycle bin. For many small filesO  there would be good UNDELETE recovery over a reasonable time span just as with Q  the PC but for very large files or massive *.*.* accidents the recycle bin wouldeL  only be effective for immediate recovery if the whole disk were potentially  available to the recycle bin.  M Most people have become accustomed to having an UNDELETE feature and VMS must 1  provide it if it's to have a longer term future.r   John   ------------------------------  # Date: Mon, 25 Sep 2000 14:00:02 GMT0/ From: "John Nixon" <jorlnixon@worldnet.att.net>@9 Subject: Re: UNDELET (Was Re: Suggested VMS new feature!) F Message-ID: <CTIz5.6461$tl2.498418@bgtnsc07-news.ops.worldnet.att.net>  I "John Macallister" <j.macallister1@physics.oxford.ac.uk> wrote in messagee1 news:000925080842.ae4c@nplvms.physics.ox.ac.uk...    snipJ > Most people have become accustomed to having an UNDELETE feature and VMS must3 >  provide it if it's to have a longer term future.e >d > John >e >e end snip  K Thanks John.  I was beginning to think I was the only "pee-cee" nuckle-heade9 around that thought this was a desirable feature for VMS.a   ------------------------------  % Date: Mon, 25 Sep 2000 14:50:07 +0000r- From: SysAdmin <djesys.nospam@fsi.net.mapson>f9 Subject: Re: UNDELET (Was Re: Suggested VMS new feature!) . Message-ID: <39CF661F.357825EC@fsi.net.mapson>   John Nixon wrote:5 > K > "John Macallister" <j.macallister1@physics.oxford.ac.uk> wrote in messagee3 > news:000925080842.ae4c@nplvms.physics.ox.ac.uk...c >  > snipL > > Most people have become accustomed to having an UNDELETE feature and VMS > must5 > >  provide it if it's to have a longer term future.  > >e > > John > >  > >a
 > end snip > M > Thanks John.  I was beginning to think I was the only "pee-cee" nuckle-headO; > around that thought this was a desirable feature for VMS.3  1 OH, NO! I agree that is *IS* a desirable feature.s  H The logistics of it in a multi-user environment, however, make it rather, a more serious issue than just "we need it".  C Y'know, just thinking about it, I'm curious now: How does WhineDozeeG handle the concept of the "recycle Bin" in a system with more than justc the C: drive for read/write?  9 For example: if I send D:\MYFILE.DOC into the recycle bins@ (C:\WINDOWS\RECYCLED\), does the file still live on D:, or is it transferred to C:?   ...or on NT:  H If Administrator sends a file to the Recycle bin, can David (with Admin. priv.) restore it?  E What about (Same scenario) when the file originally "lived" on the D:i (hard disc) drive (partition)?  ; How 'bout the multi-user, multi-partition question on W/9x?s  F Hhmmm... I wonder how KDE on Linux handles this. I have a Trash bin on= my desktop, and "/" and "/home" are on separate partitions...    David J. Dachtera    ------------------------------    Date: 25 Sep 2000 13:29:12 -05002 From: malmberg@eisner.decus.org (John E. Malmberg)9 Subject: Re: UNDELET (Was Re: Suggested VMS new feature!) + Message-ID: <kb3vM2gcKuVl@eisner.decus.org>d  . In article <39CF661F.357825EC@fsi.net.mapson>,/ SysAdmin <djesys.nospam@fsi.net.mapson> writes:t > E > Y'know, just thinking about it, I'm curious now: How does WhineDozeoI > handle the concept of the "recycle Bin" in a system with more than justn > the C: drive for read/write? > ; > For example: if I send D:\MYFILE.DOC into the recycle biniB > (C:\WINDOWS\RECYCLED\), does the file still live on D:, or is it > transferred to C:? >  > ...or on NT:  G Maybe.  The recycle bin is stored in special folders, and can reside onhD multiple disks as needed.  I have not looked into how optimal it is.  pJ > If Administrator sends a file to the Recycle bin, can David (with Admin. > priv.) restore it?  K The recycle bin appears to be a per-user folder.  Administrator can not seee1 David's and David can not see the Administrators.n  L Since it appears to be a directory structure that is just hidden away, thereH may be a backdoor way for a user to undelete a file that a user deletes.     G > What about (Same scenario) when the file originally "lived" on the D:   > (hard disc) drive (partition)?  O Windows NT, to provide you with the illusion of speed hides a deleted file fromeE you and really deletes it later, as it feels that it needs the space.   L Since portions of the waste basket can be on multple drives, it appears that? windows handles it automagically decides where to put the file.   O Again, I have not checked to see if the logic is documented, or how it actuallye works.  i= > How 'bout the multi-user, multi-partition question on W/9x?f  $ You are repeating earlier questions.    cH > Hhmmm... I wonder how KDE on Linux handles this. I have a Trash bin on? > my desktop, and "/" and "/home" are on separate partitions...a   I do not know.    O Please note that the Windows "Wastebasket" only applies to local drives and noteI to network served files.  If a file is deleted over a network it is gone.n  I [Unless it is on a VMS system serving the files and running Glen's Safetye
 product :-) ]i    h -John. wb8tyw@qsl.network   ------------------------------    Date: 25 Sep 2000 03:09:26 -0400/ From: jordan@lisa.gemair.com (Jordan Henderson)e Subject: Re: VMS and UNIX * Message-ID: <8qmtn6$rpo$1@lisa.gemair.com>  / In article <39CE43EA.FFE17A2E@nc.prestige.net>,a0 Michael Austin  <maustin@nc.prestige.net> wrote:- >This is a multi-part message in MIME format.e' >--------------F801FB42F1D3BEDDF5995DE2s+ >Content-Type: text/plain; charset=us-asciie  >Content-Transfer-Encoding: 7bit >b >t >  >Jack Patteeuw wrote:r >d= >> so what is the differences between chocolate and vanilla ?a >> >,7 >about the same the difference between OpenVMS and UNIXe >l  9 I find this ironic in light of a piece of trivia I heard s the other day:  @   Q.  What food containing vanilla accounts for the most vanilla       consumption?     A.  Chocolater  A Similary, VMS is nearly a superset of UNIX (with some notable andlC important exceptions), but I know of no UNIX that is close to beingt a superset of VMS.  A If Compaq could just clean up those notable exceeptions, it wouldr be a great thing...T     >> >> Jack  >> >> Beg Comp User wrote:  >> >S >> > What is the difference between VMS and UNIX?  What are the advantages of each?n	 >> > TIA.i >w' >--------------F801FB42F1D3BEDDF5995DE2s. >Content-Type: text/x-vcard; charset=us-ascii; > name="maustin.vcf"  >Content-Transfer-Encoding: 7bit- >Content-Description: Card for Michael Austinf! >Content-Disposition: attachment;  > filename="maustin.vcf" > 
 >begin:vcard m >n:Austin;Michael  >tel;work:704-947-1089 >x-mozilla-html:FALSE  >org:Michael Austin, Inc >adr:;;;;;; , >email;internet:michaelaustininc@hotmail.com >title:President >x-mozilla-cpt:;0  >fn:Michael E. Austin1
 >end:vcard > ) >--------------F801FB42F1D3BEDDF5995DE2--. >>   -Jordan Hendersonr jordan@greenapple.com    ------------------------------   Date: 25 Sep 2000 14:43:34 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: VMS and UNIX-6 Message-ID: <8qnoam$4av$1@mailint03.im.hou.compaq.com>  f In article <20000923153236.16269.00000927@ng-cg1.aol.com>, begcompuser@aol.com (Beg Comp User) writes:P :What is the difference between VMS and UNIX?  What are the advantages of each?  :TIA.   %   Poster: Please see the OpenVMS FAQ.5     The FAQ is available at: a  "     http://www.openvms.compaq.com/  4   And let's please NOT start this thread up again.    :   Pre-emptive note: this thread is going into my killfile.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 25 Sep 2000 16:33:46 +0530e, From: "krish" <a.krishnamoorthy@digital.com> Subject: Volume sete* Message-ID: <8qnbfc$2cf@usenet.pa.dec.com>   Hello,  $  I am facing some peculiar problems.  3  I created a volume set using the following commandr  E     $ MOUNT/BIND=<volume set name>/SYSTEM DKB300:,DKB500  TEST0,TEST1   I  After creating this i am getting the rvn of dkb300 as 1 and dkb500 as 2.d5 But when i create files on dkb500 the fid shows up as    (14,1,1).h  I  Doesn't the third field in the fid stand for the relative volume number?b  L  I am also not able to backup the files from this disk. When i give a backupG command of a file existing in this disk it is giving me a nofiles foundo error.  %  Could somebody clarify this for me..o    Thanks in advance.a   -Krish   --@ "Courage is the price that life extracts for granting peace with yourself."Amelia Earhart   ------------------------------  % Date: Mon, 25 Sep 2000 08:05:41 -0400r( From: Jonas Lindholm <jlindholm@rcn.com> Subject: Re: Volume setn' Message-ID: <39CF3F95.2FB5A10B@rcn.com>g   Hi!t  N When you specify dkb500 as disk VMS see it is a volume set and decide on which disk to put the file.nP Use the qualifier /VOLUME=relative-volume-number if you need to control on which  disk the file should be created.  = You backup files from the volume set, not from a single disk.iC You should see a volume set as one disk, not as two separate disks.r   /Jonas Lindholmt     krish wrote:   > Hello, > & >  I am facing some peculiar problems. >i5 >  I created a volume set using the following commandf >eG >     $ MOUNT/BIND=<volume set name>/SYSTEM DKB300:,DKB500  TEST0,TEST1e > K >  After creating this i am getting the rvn of dkb300 as 1 and dkb500 as 2.i7 > But when i create files on dkb500 the fid shows up ask
 >   (14,1,1).b > K >  Doesn't the third field in the fid stand for the relative volume number?r >oN >  I am also not able to backup the files from this disk. When i give a backupI > command of a file existing in this disk it is giving me a nofiles found. > error. > ' >  Could somebody clarify this for me..s >b >  Thanks in advance.?
 >   -Krish >n > --B > "Courage is the price that life extracts for granting peace with > yourself."Amelia Earhart   ------------------------------  # Date: Mon, 25 Sep 2000 14:04:42 GMT:8 From: Veli =?iso-8859-1?Q?K=F6rkk=F6?= <korkko@decus.fi> Subject: Re: Volume seta( Message-ID: <39CF5AC2.5B3A5A77@decus.fi>  B The idea with volume sets is to do ALL I/O via the the rvn 1. I.e.= in your case all file creation, deletion and updating/reading @ should happen via DKB300:. You are not supposed to bypass DKB300> by yourself going directly to the other rvn. Just create a few$ files via DKB300 and check the fids.  = Volume set is almost always operated via rvn 1. Most notable r7 exception is /IMAGE mode restore of volumeset where youc must specify all rvns.   _veli    krish wrote: >  > Hello, > & >  I am facing some peculiar problems. > 5 >  I created a volume set using the following commands > G >     $ MOUNT/BIND=<volume set name>/SYSTEM DKB300:,DKB500  TEST0,TEST1o > K >  After creating this i am getting the rvn of dkb300 as 1 and dkb500 as 2.t7 > But when i create files on dkb500 the fid shows up as 
 >   (14,1,1).I > K >  Doesn't the third field in the fid stand for the relative volume number?l > N >  I am also not able to backup the files from this disk. When i give a backupI > command of a file existing in this disk it is giving me a nofiles founde > error. > ' >  Could somebody clarify this for me..o >  >  Thanks in advance.k
 >   -Krish >  > --B > "Courage is the price that life extracts for granting peace with > yourself."Amelia Earhart   ------------------------------  % Date: Mon, 25 Sep 2000 15:08:06 +0100d* From: "Richard Brodie" <R.Brodie@rl.ac.uk> Subject: Re: Volume seta+ Message-ID: <8qnm85$pgq@newton.cc.rl.ac.uk>u  \ "krish" <a.krishnamoorthy@digital.com> wrote in message news:8qnbfc$2cf@usenet.pa.dec.com... > Hello, >t& >  I am facing some peculiar problems. >s5 >  I created a volume set using the following commande >oG >     $ MOUNT/BIND=<volume set name>/SYSTEM DKB300:,DKB500  TEST0,TEST1o >eK >  After creating this i am getting the rvn of dkb300 as 1 and dkb500 as 2.UB >  But when i create files on dkb500 the fid shows up as (14,1,1).  K This is an 'it hurts' when I do that question. Once you have a bound volumecG mounted, don't refer to explicitly to a volume other than the root one.h   ------------------------------  % Date: Mon, 25 Sep 2000 20:05:44 +0530e, From: "krish" <a.krishnamoorthy@digital.com> Subject: Re: Volume set * Message-ID: <8qno0g$95q@usenet.pa.dec.com>  
 Greetings,  "  Thanks for enlighting this to me.     Have one more question !!   H   Well if i have a very important file in a member(not root) of a volumeH set, and i want to take a backup of that file alone, do i have to take a  backup of the whole volume set ?  K I created some files using /volume qualifier in the create command, and theS+ files got created on the respective volume.eL But when i gave a backup with the root volume as the it backed up only files6 belonging to the root volume and not the other volume.     Thanks in advance. -Krish   ------------------------------   Date: 25 Sep 2000 15:44:08 GMT* From: helbig@astro.rug.nl (Phillip Helbig) Subject: Re: Volume setp. Message-ID: <8qnrs8$b0l$1@info.service.rug.nl>  - In article <39CF5AC2.5B3A5A77@decus.fi>, Velid6 =?iso-8859-1?Q?K=F6rkk=F6?= <korkko@decus.fi> writes:   C >The idea with volume sets is to do ALL I/O via the the rvn 1. I.e.t> >in your case all file creation, deletion and updating/readingA >should happen via DKB300:. You are not supposed to bypass DKB300a? >by yourself going directly to the other rvn. Just create a fewh% >files via DKB300 and check the fids.a  C Of course, it makes even more sense than it usually does (which is -H already a lot) to have a /TRANC=(TERM,CONC) logical point to the volume  set and use that exclusively.:   ------------------------------  # Date: Mon, 25 Sep 2000 13:40:37 GMTy0 From: "Mike Flaherty" <mflaherty2@earthlink.net>1 Subject: Why is it so hard to port things to VMS?aF Message-ID: <pBIz5.18980$nk3.922732@newsread03.prod.itd.earthlink.net>  
 Hello All,  J I have a general question regarding VMS and I know the folks here can helpI me with it.  I have been a VMS administrator for 10 years but I know veryXG little about programming other than DCL / UNIX shell scripting - I alsoeJ administer UNIX systems.  I am a big supporter of the open source movementG and I use open source solutions whenever possible such as Apache, Perl,-= Samba, and C++ (needed to compile the other listed products).   K When Compaq announced they were going to port Apache to VMS, I was phsyched"L because they said they would also support it under the VMS support contract.G I called to see when it was expected out and I was told sometime in thenH spring (this year).  I waited and waited and waited while our developersF kept asking me when it would be available.  Finally,  I noticed that aI production version was released on 8/28 also with a production version ofsG Perl (BONUS!!!).  I assume the delay was due to porting issues.  In therL interim they recommended the beta version but warned me that it was prone to crashing the system.  K I also know that the folks who are trying to port Samba are struggling with H their own challenges too.  I used to be on their distribution list but II kept getting too many messages regarding bug reports so I removed myself.h  G Please don't get me wrong.  I'm not trying to put down VMS, Compaq, the L Apache or Samba teams.  In fact I can't wait to load these new products ontoG my systems.  However, I am just curious as to what it is about VMS that E makes it harder to port code to it.  Is it just as hard to port thesey products to Windows?  L For example, from what I understand, Apache and Perl are written in C++.  IfH there is a C++ compiler on VMS, then why won't the products compile likeL they do on any UNIX variant?  I know that if I took the straight source codeI from any of these products, they would easily load on Solaris, HP-UX, andgH probably even Compaq True64.  Is this because the source code is writtenF specifically for the UNIX platforms?  In that case it wouldn't be very
 "open" right?u   Just thought I would ask,s Mike   ------------------------------    Date: 25 Sep 2000 12:31:37 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) 5 Subject: Re: Why is it so hard to port things to VMS?s+ Message-ID: <5H7zEY5FwSvN@eisner.decus.org>   y In article <pBIz5.18980$nk3.922732@newsread03.prod.itd.earthlink.net>, "Mike Flaherty" <mflaherty2@earthlink.net> writes:l  I > Please don't get me wrong.  I'm not trying to put down VMS, Compaq, theaN > Apache or Samba teams.  In fact I can't wait to load these new products ontoI > my systems.  However, I am just curious as to what it is about VMS thatuG > makes it harder to port code to it.  Is it just as hard to port these1 > products to Windows? > N > For example, from what I understand, Apache and Perl are written in C++.  IfJ > there is a C++ compiler on VMS, then why won't the products compile likeN > they do on any UNIX variant?  I know that if I took the straight source codeK > from any of these products, they would easily load on Solaris, HP-UX, andpJ > probably even Compaq True64.  Is this because the source code is writtenH > specifically for the UNIX platforms?  In that case it wouldn't be very > "open" right?J  K Look at the archives of this newsgroup (Deja.com or elsewhere if your localaJ server has expired them) for the past three months for comments about thisF by David Mathog and Brian Schenkenberger.  They have discussed certain> shortcomings in allegedly "portable" source from Unix systems.  E Certainly things would also be difficult if one tried to port betweenoJ HP-MPE and OS/400, but in the case of "just C" that "works on Unix" peopleK (authors of the original in particular) have a false sense of portability).    ------------------------------   Date: 25 Sep 2000 15:37:32 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)5 Subject: Re: Why is it so hard to port things to VMS? 6 Message-ID: <8qnrfs$4po$1@mailint03.im.hou.compaq.com>  y In article <pBIz5.18980$nk3.922732@newsread03.prod.itd.earthlink.net>, "Mike Flaherty" <mflaherty2@earthlink.net> writes:   D :I assume the delay [porting Apache -srh] was due to porting issues.  H   More centrally: the available engineering staff and other conflicting !   higher-priority work schedules.b  L :I also know that the folks who are trying to port Samba are struggling with :their own challenges too. h  -   Without details, it is difficult to answer.m  K   This might be as simple as the Samba folks lacking a set of contacts for sK   questions.  Or it might be a case of non-portable assumptions within the fK   code -- more than a few source code packages have been implemented using )   platform-specific extensions.s  I   In my experience in the general area of Samba-like tools, dealing with iI   seemingly minor and apparently innocuous issues such as the usual sortshJ   of differences in the various file types available among platforms, and J   finding the protocol-level details for non-published protocols tends to J   involve far more time and effort than do most of the "obvious" problems.  E   The more complex a tool is -- tasks as process creation and controlaF   and parallelism and performance, for instance -- the more difficult F   the code can be to (cleanly) port.  Or you end up with a rther nasty=   trade-off: a good and fast port, but somewhat incompatible.n  L   Some tools are themselves trivial to port, but they require three or four G   other prerequisite tools -- stuff that you have to port first, beforet   you can get to what you want.D  H :Please don't get me wrong.  I'm not trying to put down VMS, Compaq, theM :Apache or Samba teams.  In fact I can't wait to load these new products onto H :my systems.  However, I am just curious as to what it is about VMS thatF :makes it harder to port code to it.  Is it just as hard to port these :products to Windows?l  K   I have seen code that is trivially easy to port, and I've seen code that lM   is basically easier to rewrite than it would be to port.  It varies widely.   M :For example, from what I understand, Apache and Perl are written in C++.  IfcI :there is a C++ compiler on VMS, then why won't the products compile likenM :they do on any UNIX variant?  I know that if I took the straight source code J :from any of these products, they would easily load on Solaris, HP-UX, andI :probably even Compaq True64.  Is this because the source code is writteneG :specifically for the UNIX platforms?  In that case it wouldn't be very  :"open" right?  H   "Open" really only effectively means that you can see the source code.I   Most any programmer can crank source code tightly into OpenVMS or into hE   any other particular platform, or can make the code fairly portablei    across a variety of platforms.  I   It occurs to me that a "port-o-meter" source code portability analysis  H   tool might be useful, but I really don't have the cycles to create it.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 25 Sep 2000 15:44:25 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)5 Subject: Re: Why is it so hard to port things to VMS? , Message-ID: <8qnrsp$luq@gap.cco.caltech.edu>  y In article <pBIz5.18980$nk3.922732@newsread03.prod.itd.earthlink.net>, "Mike Flaherty" <mflaherty2@earthlink.net> writes:V >aH >Please don't get me wrong.  I'm not trying to put down VMS, Compaq, theM >Apache or Samba teams.  In fact I can't wait to load these new products onto:H >my systems.  However, I am just curious as to what it is about VMS thatF >makes it harder to port code to it.  Is it just as hard to port these >products to Windows?t  L Umm, depends.  If you're willing to accept a port to the Cygwin environment K that should be relatively easy to accomplish since it's very unixy.  If youVL want a true "native" Windows port without using that environment you'd have 6 to do at least as much work as you would for OpenVMS.    >tM >For example, from what I understand, Apache and Perl are written in C++.  If I >there is a C++ compiler on VMS, then why won't the products compile like, >they do on any UNIX variant?c  J C++ is a bad example - the standard has only recently settled down enough L for it to be considered at all a portable language.  It is more accurate to K consider most of the extant C++ code as "gnu CC" specific. See for instanceWH the SPEC web site, where they discuss somewhere or other how they pickedG the programs in the test set, and why only 1 is written in C++.  (Circa"3 about 2 months ago, when I last visited that site.)e    1 >  I know that if I took the straight source codeoJ >from any of these products, they would easily load on Solaris, HP-UX, andI >probably even Compaq True64.  Is this because the source code is writtens& >specifically for the UNIX platforms?   I Yes.  Such packages typically contain numerous platform specific functioneJ calls and implicit Unix assumptions scattered throughout the code.  (A fewK common examples: /dir/subdir/foo.blat.tar.gz is a valid filename ; creating-I a file with the same name as an existing file deletes the existing file ;HK system("any command") is valid (it NEVER is) ; /dev/null exists and behaves. in a certain manner.)  a    G They also now tend to be packaged with "autoconf", which is a tool thateF makes porting across different unices somewhat easier - at the cost ofI making porting to anything else much more difficult.  (It depends on "sh"FI scripts - about the least portable language that this could possibly have  been implemented in.)   / >In that case it wouldn't be very "open" right?e  E Well, not very portable anyway.  Open in that sense usually refers topK access to the code - not any sort of guarantee that it will run on multiple  platforms. e   Regards,   David Mathog mathog@seqaxp.bio.caltech.edu-? Manager, sequence analysis facility, biology division, Caltech     ------------------------------   Date: 25 Sep 2000 15:48:50 GMT* From: helbig@astro.rug.nl (Phillip Helbig)5 Subject: Re: Why is it so hard to port things to VMS?m. Message-ID: <8qns52$b0l$2@info.service.rug.nl>  F In article <pBIz5.18980$nk3.922732@newsread03.prod.itd.earthlink.net>,3 "Mike Flaherty" <mflaherty2@earthlink.net> writes: b  N > For example, from what I understand, Apache and Perl are written in C++.  IfJ > there is a C++ compiler on VMS, then why won't the products compile like  > they do on any UNIX variant?    H I'm not a C++ user, but a colleague thinks that the Compaq C++ compiler I is better than any of the 5 or 6 others he's tried.  The compiler is not r the problem.    0 > I know that if I took the straight source codeK > from any of these products, they would easily load on Solaris, HP-UX, andrJ > probably even Compaq True64.  Is this because the source code is written( > specifically for the UNIX platforms?    F That's my guess.  Little things like file names with two dots, bigger ' things like calling system routines....   " > In that case it wouldn't be very > "open" right?   - There are lies, damned lies and open systems.e   ------------------------------  % Date: Mon, 25 Sep 2000 12:05:48 -0400a2 From: rdeininger@mindspring.com (Robert Deininger)5 Subject: Re: Why is it so hard to port things to VMS?dL Message-ID: <rdeininger-2509001205480001@user-2ive73h.dialup.mindspring.com>  x In article <pBIz5.18980$nk3.922732@newsread03.prod.itd.earthlink.net>, "Mike Flaherty" <mflaherty2@earthlink.net> wrote:   <snip>  N > For example, from what I understand, Apache and Perl are written in C++.  IfJ > there is a C++ compiler on VMS, then why won't the products compile likeN > they do on any UNIX variant?  I know that if I took the straight source codeK > from any of these products, they would easily load on Solaris, HP-UX, andlJ > probably even Compaq True64.  Is this because the source code is writtenH > specifically for the UNIX platforms?  In that case it wouldn't be very > "open" right?   I Bingo!  "Open-source" in practice means "fairly portable among unixes" inCH many cases.  A lot of these folks have never programmed off of unix, andD their code (and makefiles etc.) are FILLED with unix-specific stuff.  F Good programs would have all their nonportable stuff isolated in a fewK modules and hidden behind a small API.  You'd have to replace those moduleshF when porting to a new platform, but the bulk of the code would remain 
 unchanged.  I Alas, that is NOT how most unix folks seem to program.  Platform-specificaM code is woven thoughout the whole application.  Variants for Solaris, HP, andJA all the others are pouned in using C preprocessor #if directives.t  M The worst problems seem to come in C and C++ programs.  OTOH, I've dealt withsE Fortran and Fortran 90 programs from unix that were painless to port.   N The sad fact is, crappy code is hard to port, and there is lots of crappy code
 out there.   --   Robert Deininger rdeininger@mindspring.comD   ------------------------------   Date: 25 Sep 2000 16:08:36 GMT/ From: Thomas Dickey <dickey@saltmine.radix.net>15 Subject: Re: Why is it so hard to port things to VMS? * Message-ID: <8qnta4$snv$2@news1.Radix.Net>  3 Robert Deininger <rdeininger@mindspring.com> wrote:iz > In article <pBIz5.18980$nk3.922732@newsread03.prod.itd.earthlink.net>, "Mike Flaherty" <mflaherty2@earthlink.net> wrote:  J > many cases.  A lot of these folks have never programmed off of unix, andF > their code (and makefiles etc.) are FILLED with unix-specific stuff.  F ...indeed: but turn it around - what programs are developed on VMS and then ported to other platforms?    -- -= Thomas E. Dickey <dickey@radix.net> <dickey@herndon4.his.com>0 http://dickey.his.comi ftp://dickey.his.com   ------------------------------  % Date: Mon, 25 Sep 2000 13:49:06 -0400 8 From: "Brian Hetrick" <bhetrick@notinnedmeats.iname.com>5 Subject: Re: Why is it so hard to port things to VMS?a6 Message-ID: <8qo30k$688$1@mailint03.im.hou.compaq.com>   "Robert Deininger" wrote ...@ > Bingo!  "Open-source" in practice means "fairly portable amongD > unixes" in many cases.  A lot of these folks have never programmedB > off of unix, and their code (and makefiles etc.) are FILLED with > unix-specific stuff. >.D > Good programs would have all their nonportable stuff isolated in aC > few modules and hidden behind a small API.  You'd have to replace C > those modules when porting to a new platform, but the bulk of the  > code would remain unchanged. > C > Alas, that is NOT how most unix folks seem to program.  Platform- C > specific code is woven thoughout the whole application.  Variantsg; > for Solaris, HP, and all the others are pouned in using Ca > preprocessor #if directives.   A hearty "amen" here.e  E The most troubling part of this, to me, is that it is _not_ difficultaC to build truly portable, or at least easily migrated, applications.uF However, most of the people building UNIX applications think they haveB conquered the universe if they have a single-source implementationC that works on all of Sun, OpenBSD, and Linux -- even if half of thee! lines are under #if conditionals.c  F I think much of the difficulty (and much of why FORTRAN and COBOL codeD generally "just goes") is that C is a low-enough level language thatD it is easy to code an implementation rather than an abstraction.  InD UNIX, for example, you just write a structure to the file descriptorD of the pipe rather than calling a "put into IPC queue" routine, withE predictable results for those systems for which an IPC queue is not aaD file system object.  The pain of interacting with the system at thatF level is higher in higher level languages, which leads to modularizingB the pain out of the application, which leads to an easily replaced "system adaptation" layer.  C C code that uses the standard library quite often "just goes" -- ofaD course it does what it did before, as it is standards compliant, butE it often has meaningful semantics in the new environment as well.  It ? seems, though, that many UNIX programmers do not understand thet@ difference between section 2 and section 3 man pages.  Oh, well.  D > The sad fact is, crappy code is hard to port, and there is lots of > crappy code out there.  D Nonportable code is hard to port.  Nonportable code is "crappy" onlyC when migration is one of your goals.  When the goal is "scratch the A developer's itch" (where the impetus for Open Source comes from),g8 insisting that the goals should include "portability" isA problematical.  Granted, it would be convenient if, say, most GNUC> software had been intended to be multi-platform (as opposed toC multi-UNIX-and-UNIX-like-platform), but then you start getting intoeE the problems of "least common denominator" design and implementation.c  F Besides, if applications portability were common, code migration shops would go out of business.w   ------------------------------  # Date: Mon, 25 Sep 2000 09:31:12 GMTw% From: A.Greig@virgin.net (Alan Greig) ( Subject: Re: [JAVA] Adobe Acrobat Viewer2 Message-ID: <39cf1a9e.1624510092@news.newsguy.com>  @ On Sat, 23 Sep 2000 13:44:30 +0200, "Hans Vlems" <hvlems@iae.nl> wrote:   >Peter,d >eM >your description of local management thinking and behaviour is quite similaroG >to what I see at Fuji. We run over 50 vms systems in the factories buth >they're= >all considered obsolete and replacements will be unix boxes.eF >Why: VMS is legacy, obsolete and DEC said so. Nigh impossible to turn >that idea around.    : Sounds like this might be a case for Rich Marcello to do a presentation.    --
 Alan Greig   ------------------------------   End of INFO-VAX 2000.537 ************************