1 INFO-VAX	Fri, 04 Jun 2004	Volume 2004 : Issue 309       Contents:F Re: 3X-DEPVD-AA Permedia 2 graphics (was: re: Alpha screen resolution)F Re: 3X-DEPVD-AA Permedia 2 graphics (was: re: Alpha screen resolution)! Re: ACME password change CATCH-22  Alpha screen resolution   Re: Announcing WHOIS version 1.6 Re: CDE not coming up  Re: CDE not coming up  RE: CDE not coming up  Re: CDE not coming up  Re: Cluster disk mount problem Re: CSWS 2.0 + TCPware 5.6 ??? Re: CSWS 2.0 + TCPware 5.6 ??? Re: CSWS 2.0 + TCPware 5.6 ??? Re: CSWS 2.0 + TCPware 5.6 ??? Re: Curious about MOP file copy  Re: datatrieve
 few questions  Re: few questions 7 Re: HP developer resources for porting from unix to VMS # Re: LIB$INITIALIZE: simple example? # Re: LIB$INITIALIZE: simple example? $ Newbie, Parameters to an application( Re: Newbie, Parameters to an application( Re: Newbie, Parameters to an application Re: NFS how to Re: NFS how to Re: NFS how to Re: NFS how to Re: NFS how to" OpenVMS Primer for System Managers& Re: OpenVMS Primer for System Managers' Re: Print file to two seperate printers ' Re: Print file to two seperate printers ' Re: Print file to two seperate printers ' Re: Print file to two seperate printers ' Re: Print file to two seperate printers  Scripting SET PASSWORD Re: Scripting SET PASSWORD Re: Scripting SET PASSWORD Re: Scripting SET PASSWORD Re: Storing system files in CMS  Re: Storing system files in CMS # Re: Trying to do sftp via batch job # Re: Trying to do sftp via batch job   F ----------------------------------------------------------------------  # Date: Fri, 04 Jun 2004 17:38:33 GMT # From: hoff@hp.nospam (Hoff Hoffman) O Subject: Re: 3X-DEPVD-AA Permedia 2 graphics (was: re: Alpha screen resolution) 2 Message-ID: <ts2wc.2994$Xf3.1154@news.cpqcorp.net>  U In article <c9q4p4$q9r$1@sws1.ornl.gov>, "Bobby Coleman" <colemanrl@ornl.gov> writes: L :I'm currently setting up an Alpha station with OpenVMS 7.3 and am trying toJ :figure out how to set the display resolution and refresh rate of the cardM :(it is a Compaq combination video/ethernet card).  The system boots to a low . :color depth and screen resolution by default.  D   You will want to more specifically identify the graphics card whenG   asking these.  If you do not know what card it is, you can use either B   the SRM console or the SDA (ANALYZE/SYSTEM) command CLUE CONFIG.  F   I'm guessing that this cited card is the 3X-DEPVD-AA series PermediaC   2 graphics card; a GZ-class graphics device.  If so, here are the    available settings:   0      Resolution    Color Depths    Refresh Rates8      640 x  480    8, 16, 24       60, 65, 70, 75, 85 Hz8      800 x  600    8, 16, 24       60, 65, 70, 75, 85 Hz8     1024 x  768    8, 16, 24       60, 65, 70, 75, 85 Hz8     1152 x  870    8, 16, 24       60, 65, 70, 75, 85 Hz8     1152 x  900    8, 16, 24       60, 65, 70, 75, 85 Hz4     1280 x 1024    8, 16, 24       60, 65, 70, 75 Hz4     1600 x 1200    8, 16           60, 65, 70, 75 Hz4     1600 x 1280    8, 16           60, 65, 70, 75 Hz4     1920 x 1080    8, 16           60, 65, 70, 75 Hz4     1920 x 1200    8, 16           60, 65, 70, 75 Hz    F   This particular card configures using DECW$PRIVATE_SERVER_SETUP.COM.F   If you don't have this file present on your system or have a versionD   that was previously tailored, you can recreate it by copying over:  >     SYS$COMMON:[SYS$STARTUP]DECW$PRIVATE_SERVER_SETUP.TEMPLATE     to create a new:  9     SYS$COMMON:[SYS$STARTUP]DECW$PRIVATE_SERVER_SETUP.COM   G   Now you can configure the desired setting for the card.  For example, C   the following configures 1280 x 1024 with 24 bits depth at 75 Hz:   "     $ decw$xsize_in_pixels == 1280"     $ decw$ysize_in_pixels == 1024:     $ define/exec/system/nolog decw$server_pixel_depth  24:     $ define/exec/system/nolog decw$server_refresh_rate 75      Restart the DECwindows server:  %     @SYS$STARTUP:DECW$STARTUP RESTART   B   If this is *not* the 3X-DEPVD-AA, I'll need to know specifically"   which graphics card is involved.  F   The OpenVMS FAQ has some related information and contains DECwindowsI   troubleshooting details, though it does not contain controller-specific C   hard (eg: jumper, dial, switch) or soft (eg: see above) settings.   N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  $ Date: Fri, 4 Jun 2004 13:26:52 -0400* From: "Bobby Coleman" <colemanrl@ornl.gov>O Subject: Re: 3X-DEPVD-AA Permedia 2 graphics (was: re: Alpha screen resolution) ( Message-ID: <c9qbcv$5kj$1@sws1.ornl.gov>  J Thank you for the prompt reply! The video card is a 3X-DEPVZ-AA. AccordingI to the manual that came with it, the maximum display resolution is 1280 x K 1024. Is this correct and can the same procedure be used for setting it up? K Also... I tried the sda clue config but it didn't show the video card name. J It showed a few system parameters... but nothing about the video.  The SRM0 output shows a Permedia P2V graphics controller.   Bobby   0 "Hoff Hoffman" <hoff@hp.nospam> wrote in message, news:ts2wc.2994$Xf3.1154@news.cpqcorp.net...  : > In article <c9q4p4$q9r$1@sws1.ornl.gov>, "Bobby Coleman" <colemanrl@ornl.gov> writes:K > :I'm currently setting up an Alpha station with OpenVMS 7.3 and am trying  toL > :figure out how to set the display resolution and refresh rate of the cardK > :(it is a Compaq combination video/ethernet card).  The system boots to a  low 0 > :color depth and screen resolution by default. > F >   You will want to more specifically identify the graphics card whenI >   asking these.  If you do not know what card it is, you can use either D >   the SRM console or the SDA (ANALYZE/SYSTEM) command CLUE CONFIG. > H >   I'm guessing that this cited card is the 3X-DEPVD-AA series PermediaE >   2 graphics card; a GZ-class graphics device.  If so, here are the  >   available settings:  > 2 >      Resolution    Color Depths    Refresh Rates: >      640 x  480    8, 16, 24       60, 65, 70, 75, 85 Hz: >      800 x  600    8, 16, 24       60, 65, 70, 75, 85 Hz: >     1024 x  768    8, 16, 24       60, 65, 70, 75, 85 Hz: >     1152 x  870    8, 16, 24       60, 65, 70, 75, 85 Hz: >     1152 x  900    8, 16, 24       60, 65, 70, 75, 85 Hz6 >     1280 x 1024    8, 16, 24       60, 65, 70, 75 Hz6 >     1600 x 1200    8, 16           60, 65, 70, 75 Hz6 >     1600 x 1280    8, 16           60, 65, 70, 75 Hz6 >     1920 x 1080    8, 16           60, 65, 70, 75 Hz6 >     1920 x 1200    8, 16           60, 65, 70, 75 Hz >  > H >   This particular card configures using DECW$PRIVATE_SERVER_SETUP.COM.H >   If you don't have this file present on your system or have a versionF >   that was previously tailored, you can recreate it by copying over: > @ >     SYS$COMMON:[SYS$STARTUP]DECW$PRIVATE_SERVER_SETUP.TEMPLATE >  >   to create a new: > ; >     SYS$COMMON:[SYS$STARTUP]DECW$PRIVATE_SERVER_SETUP.COM  > I >   Now you can configure the desired setting for the card.  For example, E >   the following configures 1280 x 1024 with 24 bits depth at 75 Hz:  > $ >     $ decw$xsize_in_pixels == 1280$ >     $ decw$ysize_in_pixels == 1024< >     $ define/exec/system/nolog decw$server_pixel_depth  24< >     $ define/exec/system/nolog decw$server_refresh_rate 75 > " >   Restart the DECwindows server: > ' >     @SYS$STARTUP:DECW$STARTUP RESTART  > D >   If this is *not* the 3X-DEPVD-AA, I'll need to know specifically$ >   which graphics card is involved. > H >   The OpenVMS FAQ has some related information and contains DECwindowsK >   troubleshooting details, though it does not contain controller-specific E >   hard (eg: jumper, dial, switch) or soft (eg: see above) settings.  > ( >  ---------------------------- #include' <rtfaq.h> ----------------------------- 4 >     For additional, please see the OpenVMS FAQ --  www.hp.com/go/openvms/faq , >  --------------------------- pure personal# opinion --------------------------- G >         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com  >    ------------------------------  # Date: Fri, 04 Jun 2004 15:44:30 GMT ) From: "Rick Barry" <richard.barry@hp.com> * Subject: Re: ACME password change CATCH-222 Message-ID: <yN0wc.2975$h%2.1436@news.cpqcorp.net>  ' <danco@ns2.pebble.org> wrote in message + news:slrncbs1ru.3uk.danco@ns2.pebble.org... E > In article <zPmvc.2789$Lf1.525@news.cpqcorp.net>, Rick Barry wrote:  > I > Rick, any chance you could enhance (or "fix" by my way of thinking) the I > password change function code processing to permit the login type to be G > specified for appropriately privileged callers?  Without that change, G > suitably privileged detached server processes cannot use SYS$ACM() to F > perform password change requests on behalf of third parties.  And (I > think) that really stinks. >   L It's fixed on 7.3-2. You can specify the logon-type during a change-password password function.  
 Rick Barry OpenVMS System Software Group  Hewlett-Packard Company 
 Nashua, NH   ------------------------------  $ Date: Fri, 4 Jun 2004 11:33:56 -0400* From: "Bobby Coleman" <colemanrl@ornl.gov>  Subject: Alpha screen resolution( Message-ID: <c9q4p4$q9r$1@sws1.ornl.gov>   Hi,   K I'm currently setting up an Alpha station with OpenVMS 7.3 and am trying to I figure out how to set the display resolution and refresh rate of the card L (it is a Compaq combination video/ethernet card).  The system boots to a low- color depth and screen resolution by default.   
 Thank you, Bobby    ------------------------------  # Date: Fri, 04 Jun 2004 13:12:15 GMT 1 From: Michael Austin <maustin@firstdbasource.com> ) Subject: Re: Announcing WHOIS version 1.6 < Message-ID: <Py_vc.4450$DI3.3911@newssvr22.news.prodigy.com>   JF Mezei wrote:    > Michael Austin wrote:  >  >>>It is available through:  >>> * >>>http://www.vaxination.ca/vms/whois.html >>> ( >>>Comments/problem reports appreciated. >> >>It is now USABLE through- >>< http://www.firstdbasource.com/whois.php >  >  > @ > Do I get royaltyies for every use of my software on the net ?  > 
 > :-) :-) :-)   ; only if I get to charge for the use of my server per use...  :) :) ;)     > ? > It is rewarding to see one's software to be put to such uses.   % The PHP for this was so simple too...   	 M. Austin    ------------------------------  % Date: Fri, 04 Jun 2004 13:32:56 +0200 7 From: Robert Trawinski <robert.trawinski@softax.com.pl>  Subject: Re: CDE not coming up+ Message-ID: <c9pml8$a5o$1@bozon2.softax.pl>    Tom Linden wrote: C > Moved the system drive from PWS600 to an XP1000 and put back into C > cluster in place of the PWS600  The former had Diamond Multimedia F > controller and the latter Elsa Gloria.  Is the latter not recognized0 > by OpenVMS or is something else perhaps amiss? > --- ( > Outgoing mail is certified Virus Free.< > Checked by AVG anti-virus system (http://www.grisoft.com).A > Version: 6.0.680 / Virus Database: 442 - Release Date: 5/9/2004  >   # I worked on XP1000 with ElsaGloria.    Robert   ------------------------------  # Date: Fri, 04 Jun 2004 15:50:04 GMT # From: hoff@hp.nospam (Hoff Hoffman)  Subject: Re: CDE not coming up2 Message-ID: <MS0wc.2979$p43.1905@news.cpqcorp.net>  _ In article <NDEMLKKEBOIFBMJLCECIKEMKDFAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes: B :Moved the system drive from PWS600 to an XP1000 and put back intoB :cluster in place of the PWS600  The former had Diamond MultimediaE :controller and the latter Elsa Gloria.  Is the latter not recognized / :by OpenVMS or is something else perhaps amiss?   F   Tom's actually an audience plant here, and specifically phrasing hisI   question in a way that that allows me to reference the OpenVMS FAQ. :-) H   In particular, sections on the ELSA GLoria Synergy graphics controllerC   support, on DECwindows troubleshooting, and on the configuration  F   information that can be included when posting questions and that canF   particularly allow faster (and more targeted) answers to a question.F   Details such as the OpenVMS and DECwindows versions, obviously.  TomE   undoubtedly also knows that the current ECO kits for the particular B   OpenVMS release should be applied, but has craftily phrased his H   particular question to allow me to include an ECO reference, too.  :-)+   Thank you, Tom!  All good questions!  :-)   N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  $ Date: Fri, 4 Jun 2004 08:49:47 -0700# From: "Tom Linden" <tom@kednos.com>  Subject: RE: CDE not coming up9 Message-ID: <NDEMLKKEBOIFBMJLCECIAENKDFAA.tom@kednos.com>   E Always glad to help out:-)  Actually I did have a look, but maybe not G thoroughly enough.  I have two totally identical systems, XP1000's both A with two pci cards in same positions on both, the ELSA gloria and ; KZPBA-CY diff scsi controller.  The diff between the two is # ? FREJA  ? VMS V7.3     ? MEMBER  ? # ? ODIN   ? VMS V7.3-1   ? MEMBER  ?   B Now what is interesing  is that CDE comes up on FREJA but not ODIN   FREJA> sho dev g  . Device                  Device           Error.  Name                   Status           Count. GZA0:                   Online               0 ODIN> sho dev g - %SYSTEM-W-NOSUCHDEV, no such device available   @ The controller is good, because I can boot Redhat on same system       -----Original Message-----,   From: Hoff Hoffman [mailto:hoff@hp.nospam]%   Sent: Friday, June 04, 2004 8:50 AM    To: Info-VAX@Mvb.Saic.Com     Subject: Re: CDE not coming up      A   In article <NDEMLKKEBOIFBMJLCECIKEMKDFAA.tom@kednos.com>, "Tom  "   Linden" <tom@kednos.com> writes:D   :Moved the system drive from PWS600 to an XP1000 and put back intoD   :cluster in place of the PWS600  The former had Diamond MultimediaG   :controller and the latter Elsa Gloria.  Is the latter not recognized 1   :by OpenVMS or is something else perhaps amiss?    H     Tom's actually an audience plant here, and specifically phrasing hisK     question in a way that that allows me to reference the OpenVMS FAQ. :-) J     In particular, sections on the ELSA GLoria Synergy graphics controllerE     support, on DECwindows troubleshooting, and on the configuration  H     information that can be included when posting questions and that canH     particularly allow faster (and more targeted) answers to a question.H     Details such as the OpenVMS and DECwindows versions, obviously.  TomG     undoubtedly also knows that the current ECO kits for the particular D     OpenVMS release should be applied, but has craftily phrased his J     particular question to allow me to include an ECO reference, too.  :-)-     Thank you, Tom!  All good questions!  :-)    3    ---------------------------- #include <rtfaq.h>     ----------------------------- 4       For additional, please see the OpenVMS FAQ --    www.hp.com/go/openvms/faq 5    --------------------------- pure personal opinion     --------------------------- G           Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com       --- (   Incoming mail is certified Virus Free.<   Checked by AVG anti-virus system (http://www.grisoft.com).A   Version: 6.0.680 / Virus Database: 442 - Release Date: 5/9/2004     --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.680 / Virus Database: 442 - Release Date: 5/9/2004    ------------------------------  % Date: Fri, 04 Jun 2004 09:46:37 -0700 & From: Tom Crabtree <tccrab@sunset.net> Subject: Re: CDE not coming up, Message-ID: <c9q91c026k2@enews4.newsguy.com>  $ Insert OpenVMS disk into CDRom Drive3 @(CDRom Disk Drive identifier)[elsa.kit]install.com  reboot   Problem solved.    Tom Linden wrote: G > Always glad to help out:-)  Actually I did have a look, but maybe not I > thoroughly enough.  I have two totally identical systems, XP1000's both C > with two pci cards in same positions on both, the ELSA gloria and = > KZPBA-CY diff scsi controller.  The diff between the two is % > ? FREJA  ? VMS V7.3     ? MEMBER  ? % > ? ODIN   ? VMS V7.3-1   ? MEMBER  ?  > D > Now what is interesing  is that CDE comes up on FREJA but not ODIN >  > FREJA> sho dev g > 0 > Device                  Device           Error0 >  Name                   Status           Count0 > GZA0:                   Online               0 > ODIN> sho dev g / > %SYSTEM-W-NOSUCHDEV, no such device available  > B > The controller is good, because I can boot Redhat on same system >  >  >   -----Original Message-----. >   From: Hoff Hoffman [mailto:hoff@hp.nospam]' >   Sent: Friday, June 04, 2004 8:50 AM  >   To: Info-VAX@Mvb.Saic.Com " >   Subject: Re: CDE not coming up >    >   C >   In article <NDEMLKKEBOIFBMJLCECIKEMKDFAA.tom@kednos.com>, "Tom  $ >   Linden" <tom@kednos.com> writes:F >   :Moved the system drive from PWS600 to an XP1000 and put back intoF >   :cluster in place of the PWS600  The former had Diamond MultimediaI >   :controller and the latter Elsa Gloria.  Is the latter not recognized 3 >   :by OpenVMS or is something else perhaps amiss?  >   J >     Tom's actually an audience plant here, and specifically phrasing hisM >     question in a way that that allows me to reference the OpenVMS FAQ. :-) L >     In particular, sections on the ELSA GLoria Synergy graphics controllerG >     support, on DECwindows troubleshooting, and on the configuration  J >     information that can be included when posting questions and that canJ >     particularly allow faster (and more targeted) answers to a question.J >     Details such as the OpenVMS and DECwindows versions, obviously.  TomI >     undoubtedly also knows that the current ECO kits for the particular F >     OpenVMS release should be applied, but has craftily phrased his L >     particular question to allow me to include an ECO reference, too.  :-)/ >     Thank you, Tom!  All good questions!  :-)  >   5 >    ---------------------------- #include <rtfaq.h>  ! >   ----------------------------- 6 >       For additional, please see the OpenVMS FAQ --  >   www.hp.com/go/openvms/faq 7 >    --------------------------- pure personal opinion   >   --------------------------- I >           Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com  >    >   --- * >   Incoming mail is certified Virus Free.> >   Checked by AVG anti-virus system (http://www.grisoft.com).C >   Version: 6.0.680 / Virus Database: 442 - Release Date: 5/9/2004  >    > --- ( > Outgoing mail is certified Virus Free.< > Checked by AVG anti-virus system (http://www.grisoft.com).A > Version: 6.0.680 / Virus Database: 442 - Release Date: 5/9/2004  >    ------------------------------  % Date: Fri, 04 Jun 2004 08:49:11 +0100 * From: Nic Clews <sendspamhere@[127.0.0.1]>' Subject: Re: Cluster disk mount problem & Message-ID: <c9p9ld$el$1@lore.csc.com>   Alfred Falk wrote: > 1 > Paul Repacholi <prep@prep.synonet.com> wrote in ' > news:87d64hwss4.fsf@prep.synonet.com:  > / > > Alfred Falk <falk@arc.REMOVE.ab.ca> writes:  > > I > >> Everything seems okay now, but still the rebuild time I was seeing - 8 > >> measured in HOURS - suggests something wrong to me. > > B > > That sugests your network is having, or had, problems. Time to
 > > check it.  > D > What would we be looking for?  (Network manager thinks it's okay.)  8                                                   ^^^^^^  " Go into SDA on the running system:  
 $ ANAL/SYS   SDA> SH LAN/COU   A You'll have a few pages of counters for each protocol (SCA is the E cluster protocol) but you'll see IP, DECnet, whatever is running, PER H CARD if you have multiple network cards. I've not seen and I'm not awareH of detailed descriptions for the fields displayed, but I think the error' fields are relatively self explanatory.   A Keep two printouts of the counters, say an hour apart, and if the A "error" fields increase, hand the paperwork to your local notwork " disoperative, and tell 'em go fix!   --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------  $ Date: Fri, 4 Jun 2004 10:47:46 +0200- From: "Martin Vorlaender" <mv@pdv-systeme.de> ' Subject: Re: CSWS 2.0 + TCPware 5.6 ??? * Message-ID: <2iar9sFl3ua0U1@uni-berlin.de>  , "Neil Rieck" <n.rieck@sympatico.ca> wrote...L > Has anyone out there got Compaq Secure Web Server 2.0 to work with TCPwareK > 5.6 ?  When I tried to start the server only one log was written and here  > are the contents:  >  > file: APACHE$SWS.LOG;2 >  > $ Set NoOn1 > $ VERIFY = F$VERIFY(F$TRNLNM("SYLOGIN_VERIFY")) F > apache$httpd.exe;1: Could not determine the server's fully qualified domain& > name, using 127.0.0.1 for ServerNameE > (76)function not implemented : make_sock: could not bind to address  > 0.0.0.0:80/ > no listening sockets available, shutting down  > Unable to open logs 5 > APACHE$WWW job terminated at 3-JUN-2004 18:49:46.70   I Try and install DRIVERS_V562P020. I think SWS 2.0 (like Mozilla) uses BSD  4.4 + entry points (which get added by that ECO).    cu,    Martin --  >                         | Martin Vorlaender  |  OpenVMS rules!1  VMS is today what      | work: mv@pdv-systeme.de D  Microsoft wants        |   http://www.pdv-systeme.de/users/martinv/8  Windows NT 8.0 to be!  | home: martin@radiogaga.harz.de   ------------------------------   Date: 4 Jun 2004 06:16:36 -0700 ( From: bob@instantwhip.com (Bob Ceculski)' Subject: Re: CSWS 2.0 + TCPware 5.6 ??? = Message-ID: <d7791aa1.0406040516.5a759b1d@posting.google.com>   m "Neil Rieck" <n.rieck@sympatico.ca> wrote in message news:<VdOvc.113141$tb4.4434480@news20.bellglobal.com>...  > Hello, > L > Has anyone out there got Compaq Secure Web Server 2.0 to work with TCPwareK > 5.6 ?  When I tried to start the server only one log was written and here  > are the contents:  >  > file: APACHE$SWS.LOG;2 >  > $ Set NoOn1 > $ VERIFY = F$VERIFY(F$TRNLNM("SYLOGIN_VERIFY")) M > apache$httpd.exe;1: Could not determine the server's fully qualified domain & > name, using 127.0.0.1 for ServerNameE > (76)function not implemented : make_sock: could not bind to address  > 0.0.0.0:80/ > no listening sockets available, shutting down  > Unable to open logs 5 > APACHE$WWW job terminated at 3-JUN-2004 18:49:46.70  >  > * * * * *  > . > Removing 2.0 and reinstalling 1.3 does work.  B that is why we are still on Purveyor, and why we may port Purveyor@ to itanium ... I don't think the apache group tests with TCPware; and multinet, only ucx, and that is a potential problem ...    ------------------------------   Date: 4 Jun 2004 15:34:27 GMT , From: bill@gw5.cs.uofs.edu (Bill Gunshannon)' Subject: Re: CSWS 2.0 + TCPware 5.6 ??? * Message-ID: <2ibj42FktespU1@uni-berlin.de>  1 In article <6Z0wc.2981$X63.514@news.cpqcorp.net>, , 	"Rick Barry" <richard.barry@hp.com> writes:7 > "Bob Ceculski" <bob@instantwhip.com> wrote in message 9 > news:d7791aa1.0406040516.5a759b1d@posting.google.com...  >>[snip]E >> that is why we are still on Purveyor, and why we may port Purveyor C >> to itanium ... I don't think the apache group tests with TCPware > >> and multinet, only ucx, and that is a potential problem ... > N > People in this forum do not want to read "I don't think..." posts. They wantD > facts that will help them resolve whatever problem they're having. >   C Bob has never let facts get in his way before, why would you expect  it of him now.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  # Date: Fri, 04 Jun 2004 15:56:50 GMT ) From: "Rick Barry" <richard.barry@hp.com> ' Subject: Re: CSWS 2.0 + TCPware 5.6 ??? 1 Message-ID: <6Z0wc.2981$X63.514@news.cpqcorp.net>   5 "Bob Ceculski" <bob@instantwhip.com> wrote in message 7 news:d7791aa1.0406040516.5a759b1d@posting.google.com... 6 > "Neil Rieck" <n.rieck@sympatico.ca> wrote in message8 news:<VdOvc.113141$tb4.4434480@news20.bellglobal.com>... >[snip] D > that is why we are still on Purveyor, and why we may port PurveyorB > to itanium ... I don't think the apache group tests with TCPware= > and multinet, only ucx, and that is a potential problem ...   H OK, Bob, I've sat back quietly while you've made statements like this in6 the past, but I'm calling you on the carpet this time.  C We work closely with Process Software to ensure that Apache runs on G Multinet and TCPware before we ship. It's simply the right thing to do.   I Whatever is causing Neil's problem will be corrected. Neil should contact H Process Software to make sure he's up to the required rev level. If theyI can't resolve the problem, they'll contact us and we'll work with them to  resolve the problem.  L People in this forum do not want to read "I don't think..." posts. They wantB facts that will help them resolve whatever problem they're having.  
 Rick Barry OpenVMS System Software Group  Hewlett-Packard Company 
 Nashua, NH   ------------------------------  $ Date: Fri, 4 Jun 2004 18:52:55 +0200, From: "Hans Vlems" <hvlems.dotweg@zonnet.nl>( Subject: Re: Curious about MOP file copy* Message-ID: <2ibnsaFjppamU1@uni-berlin.de>   MOP won't pass routers.   / "Hal Kuff" <kuff@tessco.com> schreef in bericht ' news:c9o3fm$bv5@library2.airnews.net... J > True, but what I was looking for was a way to use MOP to actually copy aK > file from one system to another where decnet and ip are not available due = > the intervening routers/bridgy things but MOP will pass....  >  > 1 > "labadie" <labadie_g@decus.fr> wrote in message . > news:c9o0do$mdk$1@news-reader1.wanadoo.fr... > > Hal Kuff a crit :L > > >     I recall seeing a uility somewhere that could make a cient request > via G > > > MOP to have a MOP server send a file to the client... this was an 	 > openvms H > > > utility that would make the request and then store the file on the	 > OpenVMS I > > > server... it could be used where two systems are accessible via MOP  but  > not  > > > Decenet or IP....  > > >  > > > A > > mc ncp trigger... phys add ... via ... service password ... ?  >  >    ------------------------------  # Date: Fri, 04 Jun 2004 11:16:07 GMT . From: "Andrew Dodd" <Andrew.J.Dodd@zxcvhp.com> Subject: Re: datatrieve 2 Message-ID: <XRYvc.2957$_N2.2622@news.cpqcorp.net>  G Collections, whilst having some use, were almost always what killed the J performance of a running DTR system - leave aside the time to get started.F My understanding is that collections did not store the data but storedF pointers to the data. The pointer type depended on the underlying data- source. For RMS they would RFIDs or somesuch.   H This meant that every time a collection was read the underlying data was6 read. So if the domain was called yachts I often saw:-   Find all yachts  Find all with rig = "ketch" $ Find all with length greater than 10 Print   I Now with large files this took an age and used a lot of memory - the data / had been read three times in the above example.    To sit down and type  * Print all yachts with rig = "ketch" and...  I and have the results instantly coming out could be literally jaw dropping J for customers used to waiting many minutes. My best was 17 hours for a jobA down to less than 15 minutes - just took out all the collections.   
 Nostalgia.  & Andrew Dodd - now with Hewlwtt Packard   ------------------------------  % Date: Fri, 04 Jun 2004 19:18:45 +0200  From: gdb <blah1234@wp.pl> Subject: few questions2 Message-ID: <c9qau1$dub$1@portraits.wsisiz.edu.pl>   EHLO  H I'm trying to get work at a company which uses VMS(probably OpenVMS) as E a platform for it's software. I use Linux for over 4 years. Also had  A some minor contact with BSDs. Can you tell me what are the major  F differences between VMS and Linux? And more precise: Can I run VMS on I x86?, What are the .bck files which i downloaded as cd images...? and if  ; they are what software should I use to put them on the cds? D Sorry for the sky high lame level of this but I don't know anything + about VMS yet. Links will be appreciated :]    gdb    ------------------------------  % Date: Fri, 04 Jun 2004 10:46:29 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com>  Subject: Re: few questions' Message-ID: <40C0B575.8080404@MMaz.com>   
 gdb wrote:   > EHLO > G > I'm trying to get work at a company which uses VMS(probably OpenVMS)  # > as a platform for it's software.    G You may find yourself at a severe disadvantage unless this prospective  B employer doesn't want to pay well, because there exists a glut of 1 qualified VMS folks with decades of experience...   G > I use Linux for over 4 years. Also had some minor contact with BSDs.  H > Can you tell me what are the major differences between VMS and Linux?   G Those are all variations of Unix, VMS is not even remotely like Unix.   E Here is a site that has links to public systems that you can connect  H into to 'look' around but IMHO, you will not be able to get up to speed 7 as a qualified VMS system manager or coder, over-night:    http://www.vajhoej.dk/arne/vms/   + > And more precise: Can I run VMS on x86?,    B Yes, but only if you use a VAX emulator like SIMH or a commercial I emulator like CHARON-VAX, otherwise no, VMS only runs on VAX, Alpha, and   Itanic  > > What are the .bck files which i downloaded as cd images...?   F Those are backup savesets that, generally speaking, are only of value E and use on a running VMS system or a piece of VMS hardware running a   version of Standalone Backup.   D > and if they are what software should I use to put them on the cds?F > Sorry for the sky high lame level of this but I don't know anything - > about VMS yet. Links will be appreciated :]   H Welcome as a hobbyist interest, but you're over your head if you expect - to get a VMS related job anytime real soon...    Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------  $ Date: Fri, 4 Jun 2004 05:21:02 -0400# From: "John Smith" <a@nonymous.com> @ Subject: Re: HP developer resources for porting from unix to VMS, Message-ID: <hYKdnepGpOJio13d4p2dnA@igs.net>  A "Craig A. Berry" <craigberry@mac.com.spamfooler> wrote in message : news:260eeca58a800d656a4d053441c92a57@news.teranews.com...4 > > > "John Smith" <a@nonymous.com> wrote in message, > > > news:5cOdnTT-BKpWbSDdRVn-hw@igs.net... > > > > G > > > > Since the ISV is not about to let us look 'under the kimono' at  their code, I > > > > all I can hope to do is point them to all official and unofficial 	 resources L > > > > to help an application developer see the light and learn how to port that! > > > > piece of code to OpenVMS.  > G > Someone needs to look under the kimono, or at least have described to G > them what's there.  You've gotten lots of good general advice, but it H > may not be worth much without knowing something about what language orE > languages, architectural features, APIs, build tools, and auxiliary H > packages or libraries are involved.  You could tell them lots and lotsI > about mailboxes and global sections, for example, only to find out that C > there's no inter-process communication, or go on and on about the H > VMS-specific aspects of tuning large directories containing many smallF > files, only to find out that the documents are stored as blobs in an > RDBMS. > G > Forgive me if this has been mentioned already in this thread, but one I > thing you can encourage them to do is get a test drive account and just $ > try building and see what happens: > % > http://www.testdrive.hp.com/os/#vms  > I > > Another interesting potential solution we've been looking at possibly  taking< > > a stab at porting ourselves is at http://www.dspace.org/ > B > Since it runs on the Java platform, what exactlly do you mean by > "porting"?  . It's not what I mean...it's what we were told.6 They apparently have a component that is o/s specific.L Additional information is to be provided to us in the next few days by theirH developers and we will provide them with a list of porting resources for them to take a look at. H Of course their sales droids didn't know what OpenVMS was - I guess they? went to university after OpenVMS stopped being used in courses.        >gJ > > Notice that HP had more than a helping hand in this but apparently didC > > absolutely zero, nada, nil, nichts about making it OpenVMS-able  > > out-of-the-box.. >?H > Yes, HP was quite silly not to be promoting their competitor's product8 > in June, 2000.  Can't imagine what they were thinking.  4 Compaq, ChumHPaq, HP....what's the difference? SSDD.   ------------------------------  * Date: Fri, 4 Jun 2004 10:07:14 +0000 (UTC)3 From: "Richard Maher" <maher_rj@hotspamnotmail.com>n, Subject: Re: LIB$INITIALIZE: simple example?2 Message-ID: <c9phkh$rkd$1@hercules.btinternet.com>   Hi,   C Here is a small example I had lying around if it helps anyone else.m   Regards Richard Mahere   $ on warning then exit: $ if .not. f$privilege("cmkrnl,sysprv")  then goto no_priv9 $ if f$getsyi("arch_name") .nes. "Alpha" then goto no_vaxe $! $ create init_test_user.cobt identification division. program-id.    ef_rubbish. procedure division.n 00.      display "In Rubbish!".     exit program.t *  end program ef_rubbish.n identification division. program-id.    check_priv. *r data division. working-storage section.J 01  ss$_normal          pic 9(9)        comp    value external ss$_normal.- 01  sys_status          pic 9(9)        comp.- *-- 01  elev_priv           pic 9(18)       comp.r- 01  priv_index          pic 9(9)        comp.:- 01  privs_found         pic 9(9)        comp.a- 01  priv_result         pic 9(18)       comp.f- 01  priv_name_len       pic 9(4)        comp.q *1 01  priv_value_table. L     03  prv$m_cmkrnl    pic 9(18)       comp    value external prv$m_cmkrnl.L     03  prv$m_cmexec    pic 9(18)       comp    value external prv$m_cmexec.L     03  prv$m_sysnam    pic 9(18)       comp    value external prv$m_sysnam.L     03  prv$m_grpnam    pic 9(18)       comp    value external prv$m_grpnam.>     03  prv$m_allspool  pic 9(18)       comp    value external prv$m_allspool.      03  prv$m_impersonaten>                         pic 9(18)       comp    value external prv$m_impersonate.>     03  prv$m_diagnose  pic 9(18)       comp    value external prv$m_diagnose.VL     03  prv$m_log_io    pic 9(18)       comp    value external prv$m_log_io.K     03  prv$m_group     pic 9(18)       comp    value external prv$m_group.fL     03  prv$m_noacnt    pic 9(18)       comp    value external prv$m_noacnt.L     03  prv$m_prmceb    pic 9(18)       comp    value external prv$m_prmceb.L     03  prv$m_prmmbx    pic 9(18)       comp    value external prv$m_prmmbx.L     03  prv$m_pswapm    pic 9(18)       comp    value external prv$m_pswapm.L     03  prv$m_setpri    pic 9(18)       comp    value external prv$m_setpri.L     03  prv$m_setprv    pic 9(18)       comp    value external prv$m_setprv.L     03  prv$m_tmpmbx    pic 9(18)       comp    value external prv$m_tmpmbx.K     03  prv$m_world     pic 9(18)       comp    value external prv$m_world.IK     03  prv$m_mount     pic 9(18)       comp    value external prv$m_mount.oJ     03  prv$m_oper      pic 9(18)       comp    value external prv$m_oper.>     03  prv$m_exquota   pic 9(18)       comp    value external prv$m_exquota.L     03  prv$m_netmbx    pic 9(18)       comp    value external prv$m_netmbx.L     03  prv$m_volpro    pic 9(18)       comp    value external prv$m_volpro.L     03  prv$m_phy_io    pic 9(18)       comp    value external prv$m_phy_io.L     03  prv$m_bugchk    pic 9(18)       comp    value external prv$m_bugchk.L     03  prv$m_prmgbl    pic 9(18)       comp    value external prv$m_prmgbl.L     03  prv$m_sysgbl    pic 9(18)       comp    value external prv$m_sysgbl.L     03  prv$m_pfnmap    pic 9(18)       comp    value external prv$m_pfnmap.K     03  prv$m_shmem     pic 9(18)       comp    value external prv$m_shmem.AL     03  prv$m_sysprv    pic 9(18)       comp    value external prv$m_sysprv.L     03  prv$m_bypass    pic 9(18)       comp    value external prv$m_bypass.L     03  prv$m_syslck    pic 9(18)       comp    value external prv$m_syslck.K     03  prv$m_share     pic 9(18)       comp    value external prv$m_share.s>     03  prv$m_upgrade   pic 9(18)       comp    value external prv$m_upgrade.>     03  prv$m_downgrade pic 9(18)       comp    value external prv$m_downgrade.L     03  prv$m_grpprv    pic 9(18)       comp    value external prv$m_grpprv.>     03  prv$m_readall   pic 9(18)       comp    value external prv$m_readall.L     03  prv$m_import    pic 9(18)       comp    value external prv$m_import.K     03  prv$m_audit     pic 9(18)       comp    value external prv$m_audit.y>     03  prv$m_security  pic 9(18)       comp    value external prv$m_security.e *x9 01  priv_value_array                            redefinesn     priv_value_table.e:     03  priv_mask       pic 9(18)       comp    occurs 39. *i 01  priv_name_table.?     03  prv$m_cmkrnl    pic x(11)               value "CMKRNL".t?     03  prv$m_cmexec    pic x(11)               value "CMEXEC".i?     03  prv$m_sysnam    pic x(11)               value "SYSNAM". ?     03  prv$m_grpnam    pic x(11)               value "GRPNAM".aA     03  prv$m_allspool  pic x(11)               value "ALLSPOOL".n     03  prv$m_impersonateeD                         pic x(11)               value "IMPERSONATE".A     03  prv$m_diagnose  pic x(11)               value "DIAGNOSE". ?     03  prv$m_log_io    pic x(11)               value "LOG_IO".h>     03  prv$m_group     pic x(11)               value "GROUP".?     03  prv$m_noacnt    pic x(11)               value "NOACNT". ?     03  prv$m_prmceb    pic x(11)               value "PRMCEB".t?     03  prv$m_prmmbx    pic x(11)               value "PRMMBX".(?     03  prv$m_pswapm    pic x(11)               value "PSWAPM".h?     03  prv$m_setpri    pic x(11)               value "SETPRI".c?     03  prv$m_setprv    pic x(11)               value "SETPRV".a?     03  prv$m_tmpmbx    pic x(11)               value "TMPMBX".R>     03  prv$m_world     pic x(11)               value "WORLD".>     03  prv$m_mount     pic x(11)               value "MOUNT".=     03  prv$m_oper      pic x(11)               value "OPER". @     03  prv$m_exquota   pic x(11)               value "EXQUOTA".?     03  prv$m_netmbx    pic x(11)               value "NETMBX". ?     03  prv$m_volpro    pic x(11)               value "VOLPRO". ?     03  prv$m_phy_io    pic x(11)               value "PHY_IO".6?     03  prv$m_bugchk    pic x(11)               value "BUGCHK".,?     03  prv$m_prmgbl    pic x(11)               value "PRMGBL". ?     03  prv$m_sysgbl    pic x(11)               value "SYSGBL".E?     03  prv$m_pfnmap    pic x(11)               value "PFNMAP".t>     03  prv$m_shmem     pic x(11)               value "SHMEM".?     03  prv$m_sysprv    pic x(11)               value "SYSPRV".E?     03  prv$m_bypass    pic x(11)               value "BYPASS". ?     03  prv$m_syslck    pic x(11)               value "SYSLCK". >     03  prv$m_share     pic x(11)               value "SHARE".@     03  prv$m_upgrade   pic x(11)               value "UPGRADE".B     03  prv$m_downgrade pic x(11)               value "DOWNGRADE".?     03  prv$m_grpprv    pic x(11)               value "GRPPRV".$@     03  prv$m_readall   pic x(11)               value "READALL".?     03  prv$m_import    pic x(11)               value "IMPORT". >     03  prv$m_audit     pic x(11)               value "AUDIT".A     03  prv$m_security  pic x(11)               value "SECURITY".s *r9 01  priv_name_array                             redefinesf     priv_name_table.:     03  priv_name       pic x(11)               occurs 39. *t procedure division.i 00.      display "In Initialize".  6     call "get_priv" using elev_priv giving sys_status.A     if sys_status not = ss$_normal call "lib$stop" using by value  sys_status.d       if elev_priv not = zerosG         display "You BEWDY! I am now in the elevated state of Privilege 
 Nirvana with"o          display "(" no advancingD         perform varying priv_index from 1 by 1 until priv_index > 39               call "and_privs""                 using   elev_priv,.                         priv_mask (priv_index)#                         priv_resulta  &             if priv_result not = zeros                 call "str$trim"nK                     using    by descriptor priv_name (priv_index),priv_name  (priv_index)8                              by reference  priv_name_len$                 add 1 to privs_found"                 if privs_found > 1,                     display "," no advancing                 end-ifL                 display priv_name (priv_index)(1:priv_name_len) no advancing             end-if         end-performs         display ")"tK         display "Number of additional privileges available = ", privs_foundt with conversion.       exit program.k *  end program check_priv.. $! $ cobol/lis init_test_user.cob $! $ create get_priv.maro  E         .title  Get Priv - Another nail in the coffin of RTNUSENOTALLr         .ident  "V1.0"           $prvdef GLOBAL         $ssdef ;++4 ; Auto initialize- ;--          .extrn lib$initializel         .psect> lib$initialize,nopic,usr,con,rel,gbl,noshr,noexe,rd,nowrt,long           .address -L                  check_priv ; Call this routine when running up my shareable  C         .psect  _maher$data,pic,con,rel,lcl,noshr,noexe,rd,wrt,quad   	 auth_prv:a         .blkq           1t	 proc_prv:e         .blkq           1o	 imag_prv:          .blkq           1t max_prv:         .blkq           1-  	 jpi_list:-(         .word           8, jpi$_authpriv          .address        auth_prv         .long           0j(         .word           8, jpi$_procpriv          .address        proc_prv         .long           0 (         .word           8, jpi$_imagpriv          .address        imag_prv         .long           0i         .long           0>  	 jpi_iosb:t         .quad           0e  A         .psect  _maher$code,pic,con,rel,lcl,shr,exe,rd,nowrt,quad/  *         .sbttl  Derive elevated privileges  -         .call_entry     max_args=1,         ---                         home_args=false,    -s-                         quad_args=true,     - &                         label=get_priv            cmpb    (ap),#1          bgeq    10$           movzwl  #ss$_insfarg,r0          ret   10$:    $getjpiw_s -"                  itmlst=jpi_list,-                  iosb=jpi_iosb          blbc    r0,20$u          movl    jpi_iosb,r0          blbs    r0,30$@   20$:     ret   30$:          evax_or -*                  auth_prv,proc_prv,max_prv          evax_bic - (                  imag_prv,max_prv,@4(ap)            movzwl #ss$_normal,r0          ret  2         .sbttl  Bitwise and of privilege quadwords  -         .call_entry     max_args=3,         - -                         home_args=false,    -.-                         quad_args=true,     -w'                         label=and_privss            evax_and -e&                  @4(ap),@8(ap),@12(ap)            ret           .end   $!  $ macro/lis/enable=quad get_priv $! $ link/share=init_test_user.exem- init_test_user.obj,get_priv.obj,sys$input/optC  $ symbol_vector=(ef_rubbish=procedure)   gsmatch=lequal,1,0 $!1 $ user_spec = f$parse("init_test_user.exe") - ";"k4 $ define/system/exec/nolog init_test_user 'user_spec $! $ install := $install/commando $!4 $ if f$file_attributes("init_test_user.exe","known")> $ then        install replace init_test_user/open/header/share> $ else        install add     init_test_user/open/header/share $ endifn $!
 $ sql:==$sql$s $ sql    attach 'file mf_personnel';t   drop function ef_rubbish;i   create  function ef_rubbish ()          returns integer by value         language sql ;i          external name ef_rubbishC         location 'init_test_user' with all logical_name translationg         language cobol'         general parameter style variantw+         comment is 'Humiliate RTNUSENOTALL'l         BIND ON CLIENT SITEa         bind scope connect ;  commit;  exit;  $exitn $!	 $no_priv:oK $       write sys$output "Insufficient privilege. You need (CMKRNL,SYSPRV)"i $       exit 44u $no_vax:8 $       write sys$output "This code only works on alpha" $       exit 44r   ------------------------------  * Date: Fri, 4 Jun 2004 10:20:16 +0000 (UTC)3 From: "Richard Maher" <maher_rj@hotspamnotmail.com>-, Subject: Re: LIB$INITIALIZE: simple example?/ Message-ID: <c9pict$bdu$1@titan.btinternet.com>V  > And to put it into context here's the rant that went with it:-   ----- Original Message ----- From: Richard's Hotmail ' Sent: Wednesday, April 09, 2003 9:28 PMeG Subject: Gaping Security Holes Batman! *NOT* (Was: Rdb unable to verifyJ external routines)     > Hi,T >LF > As discussed, I've written a little test program (see attached) thatB > produces the following output when run from an account with onlyK > (TMPMBX,NETMBX) privileges on a server where RMU is installed with SYSPRVe
 > and CMKRNL.e >c$ > $ rmu/ver/rout [.test]mf_personnel > In InitializepE > You BEWDY! I am now in the elevated state of Privilege Nirvana with  > (CMKRNL,SYSPRV)e7 > Number of additional privileges available =         2? >SI > Clearly my external function inherits the privileges that RMU (or otherRI > mainline image) was installed with and can now run amok with everything A > SYSPRV and CMKRNL lets me do. But so what? RMU Verify (and more  importantly J > VMS) will insist on my shareable image being INSTALLed and pointed to byG > trusted (System/Exec) logical names. So given that you need CMKRNL tomK > install an image and SYSNAM privilege to create an Exec mode logical name@ inE > the System table, anyone who can satisfy Verify's prerequisites for. loadinguH > a shareable image can already do whatever they want on your system. IfH > they're willing to warrant that an external function can be trusted toH > behave itself when operating with elevated privileges then that's good8 > enough for me, *YOU*, and everyone else on the planet. > L > So if I'm in violent agreement with how RMU/VERIFY is handling things thenF > what's the problem? Well this whole thing got me thinking about whatL > RTNUSENOTALL was supposed to be for, and whether or not the right hand hasB > even the slightest inkling of what the left hand is doing at Rdb > engineering. >EL > .. External functions with privs is fine with verify but banned everywhere > elseG > .. Privileged sub-systems (Don't look at them and they might go away)nC > .. Cosi_authenticate finally fixed after years as a hackers dreamgH > .. Set Session Authorization. Atrophied and withering on the vine thru
 > neglect.K > .. Session User being used to check RUJ dir access instead of System Usere; > .. Rdb Auditing using System User instead of Session Userc; > .. Has anyone tested all the ACL stuff with Session User?i >uI > Why don't you get some sort of Security Tsar in and sort this mess out?- > I > Then I did a quick scan through old mails and found one dated 26/8/2000- withL > a rant on RTNUSENOTALL. (I've attached it at the end of this mail. This isJ > by no means the oldest mail on this subject; just the oldest I could get myK > hands on easily.) I wonder if whoever entered the Enhancement Request neegI > Bug #1679146 is happy with the turn around time? Three years! No that's L > THREE YEARS!!! My ego wants me to believe that Rdb engineering is draggingJ > its feet on this just to give me the shits, because surely even the mostJ > twisted of intellects must now concede that there is no place in Rdb for the.< > truly debilitating bit of redundancy that is RTNUSENOTALL? > K > Ah yes, things have to be prioritised, there are limited resources and we I > need to maximize our ROI and while we're quite happy to let you use oursI > software we'd much prefer it if you'd just piss-off and stop telling usD how L > to write it and as far as your business requirements go why don't you find > someone who gives a damn?l >s > Regards Richard Maheru >yI > PS: Note that there is a bug in my code in that for some strange reasony thetE > VALUE EXTERNAL clause maxes out with a longword value therefore any J > privilege greater than 2**31 (for example prv$m_security) can't be found byL > the attached code. Yes it's not very 64-bit friendly of the COBOL compilerK > to restrict this, but don't you think I've got enough on my plate without K > having to request yet another blatantly desirable enhancement? (Only this  > time against the COBOL team?)  >tK > It's only an example but if anyone is desperate for a working version letN me > know.  >S > ----- Original Message ----- > From: Richard's Hotmailt' > Sent: Monday, April 07, 2003 10:19 PM 5 > Subject: Re: Rdb unable to verify external routines  >  >m > > Hi John, > >lJ > > I had a quick look today. The first thing that I'd suggest is *always* > makeJ > > sure you can call the external routine from interactive SQL. The errorJ > > messages that SQL returns are a lot more comprehensive and useful than thew >eL > > terse comments from RMU/VERIFY. You may also want to get your developers > toJ > > knock up a simple routine that calls lib$find_image_symbol as this can > oftenhD > > reveal more secrets as to why VMS is having trouble activating a > shareable. > >eI > > Secondly, the help is pretty good and does point out that the logicale > names C > > involved should be both SYSTEM *and* EXEC and the image must bee
 installed.E > > (You don't have to install all 600 just the one with the externala routinet > in > > it)u > >,L > > The interesting thing here is RMU/VERIFY loaded my shareable even though ItI > > was an unprivileged (less than what RMU was installed with) user. I'ml > gonna L > > put a LIB$INITIALIZE routine in tomorrow for the hell of it and see if I > getwH > > (CMKRNL,SYSPRV) via RMU. Yes, *I* know it's not a problem. I've been	 > tellingi > > *you* that!u > >  > > Cheers Richard Maher.d > >p  > > ----- Original Message ----- > > From: Creed, Johne( > > Sent: Monday, April 07, 2003 3:18 AM7 > > Subject: RE: Rdb unable to verify external routines! > >  > >  > > > Richard, > > >-? > > > > > The logical FMSEXE used in the definition is a SYSTEM- > > > > logical, but not inlD > > > > > the EXEC table, and is defined as a search list could this > > > > be the problem?p > > > >nL > > > > I thinks so. They should be at EXEC mode. (As well as all subsequentE > > > > logicals) That's a pretty impressive search list for an imagei > > > > directory butrE > > > > I doubt if that's the problem. Are the executables INSTALLed?. > > >eI > > > We have around 600+ in-house written programs, but I do not believe  they > > arei > > > installed. > > >.  > > > > I find this fascinating! > > >pE > > > I do also, what is most interesting to me is that we had a lonee externalJ > > > function causing this same verification message when I first started > > workingrD > > > on Rdb and simply defining a logical with the same name as the externalL > > > function and then installing the image seemed to lick the problem, but > hasi; > > > not worked so far with these new external procedures.p > > >tG > > > The main difference I have found between the old and new externalE
 > > functions I > > > is the old one was located in SYS$LIBRARY and the others are in theS > search! > > > list of the FMSEXE logical.  > > >gK > > > I am probably going to have to open a TAR for this one since I am not I > > > totally up to speed on VMS & Rdb privileges with regards to images,  > > > installation, RMU, etc.  > > > L > > > Of course I will let the list in on the answer once it has been found. > > >  > > > Sincerely, > > > John E. Creed  > > >o > > >S >i7 > HISTORY..............................................e >  > Fish.g >E > Regards Richard Maher. > I > PS. Is Peter Jackson still around? Your ref RDB_80.NOTE Oracle Rdb - OnrF > Windows NT and OpenVMS (Aaah nostalgia) 17-FEB-1999 113.* Is there aL > parallel universe operating here or was I (and my/your customer) told (nayF > dismissed with) *BIND ON SERVER SITE*!!!! There be Dragons in imagesI > installed with privs! Abandon all hope ye who won't do as they're told!r >P# > > It isn't a security hole at alle >a5 > (How do you type the theme from the Twilight Zone?)e >rL > So it was "a truly debilitating bit of redundancy."? If not let me restate > the question:- > I >  "What problem is it exactly that Rdb is solving with this RTNUSENOTALLc > solution?" >tI > 17-Feb-1999 (and before) till 9-May-2001 to remove a couple of lines ofo > code. Outstanding! > H > Do we have an ETA on when this restriction will be removed? Surely row cache  > can spare a couple of bucks? >L > ----- Original Message ----- > From: Ian SmithL' > Sent: Wednesday, May 09, 2001 2:30 PMe7 > Subject: Re: Problem with SQL$FUNCTIONS not installeda >  >aG > > I think you mis-state the problem.  It isn't a security hole at all_ (i.e.o > thatG > > would imply doing things contrary to your security clearence).  Theg > request isB > > to relax the security checking when the external functions are
 installed. > >n > > IanV > >5 > > Richard Maher wrote: > >o	 > > > Hi,  > > >.K > > > Is 1679146 a Tar# or a Bug#? Can someone please post the details here  > for - > > > the benefit of the metalink challenged?e > > >SJ > > > Has the VMS security hole gone away? (As someone who regularly calls	 > lib$fispG > > > loaded user routines from an image installed with privs I am veryo > interested" > > > in this enhancement request) > > >  > > > Regards Richard Maherm > > >n" > > > ----- Original Message ----- > > > From: Richard Maher.- > > > Sent: Saturday, August 26, 2000 5:18 PM1; > > > Subject: RE: Problem with SQL$FUNCTIONS not installed  > > >n > > > > Hi Beth, > > > >$J > > > > I think it's worthwhile pointing out to the viewers the reason why it >mL > > > > "worked fine". Again, I have to imagine that you are running under aI > > > > privileged username. By privileged I mean that the UAF record for  the K > > > > username that you have logged-in to VMS under has either AuthorizedE orK > > > > Default privileges that are at least equal to, if not greater than,d > the F > > > > privileges that RMU.EXE was INSTALLed with at your site. In my > experience > > > itL > > > > is usual to INSTALL RMU with (SYSPRV,CMKRNL). Therefore I'm guessing > thatK > > > > the username that you were under has at least these privileges? (or  > you , > > > > dropped the /VIRTUAL on the UNLOAD?) > > > >. > > > > If you were to:- > > > >v > > > > $mc authorizerE > > > > UAF> modify your_username/priv=(noall,sysprv,tmpmbx,netmbx) /P* > > > > defpr=(noall,sysprv,tmpmbx,netmbx) > > > > UAF> exit. > > > > $set host 0- > > > >-K > > > > and login again and try to unload the same table with /VIRTUAL thenh > I'd J > > > > imagine that you would receive the RTNUSENOTALL error that I spoke of > > > > earlier. > > > >9F > > > > (When did we start needing SYSPRV just to run RMU? is there an > IdentifierF > > > > somewhere that says you can run RMU (Whether or not on a given
 > database?))o > > > >gI > > > > My question still stands. "What problem is it exactly that Rdb isa	 > solvingeD > > > > with this RTNUSENOTALL solution?" I believe it to be a truly > debilitating > > > > bit of redundancy. > > > >iI > > > > On a pre-emtive note, before anyone is churlish enough to suggestn that > allTI > > > > Rdb users learn how to write User Written System Services, a taskn	 > which IoE > > > > for one find quite difficult, could I point out that the sameoI > > > > suggestion/recommendation could be put to the authors of RMU withk > equal $ > > > > validity (or lack thereof!). > > > >n > > > > Regards Richard Maherd > > > >n > > > > >From: "Seefelt, Beth"> > > > > >Subject: RE: Problem with SQL$FUNCTIONS not installed. > > > > >Date: Fri, 25 Aug 2000 20:07:22 -0400	 > > > > >:	 > > > > >  > > > > >Richard,<	 > > > > >rL > > > > >You were absolutely right.  I removed the /priv and it worked fine.	 > > > > >s > > > > >Thanks!	 > > > > >k
 > > > > >Bethw	 > > > > >o# > > > > >-----Original Message-----5 > > > > >From: Richard Maher/ > > > > >Sent: Friday, August 25, 2000 11:46 AM > > > > > >Subject: RE: Problem with SQL$FUNCTIONS not installed	 > > > > >p > > > > >Hi Beth,e	 > > > > >oH > > > > >I believe Jerry was referring to the RTNUSENOTALL error which I	 > suspectr	 > > > youi2 > > > > >would receive if you were to INSTALL> add5 > > > > >sys$library:sql$functions.exe/open/hdr/sharey	 > > > > >vE > > > > >Without your install command I can only imagine that you arer enteringD > > > > >something like /PRIV which is not needed and is why you are	 receivingt > thedL > > > > >IMGTRACED error. (I have no idea why Rdb engineering would ship any > imagelH > > > > >that has been linked without the /NOTRACE qualifier (except for > debugginguK > > > > >purposes) but I don't think it is relevant and anyway, mine is nota to > > > > >reasonw > > > > >why.)	 > > > > >rJ > > > > >What I will assert is that the RTNIMAGE_UNINSTALLED (coupled with > onlyJ > > > > >checking system/executive_mode logical names, regardless of "with allrK > > > > >logical name translation") is the *only* check(s) that needs to be- > > > > >performedK > > > > >when activating an external function that has been invoked from andK > > > > >executable image that has been installed with privileges! (that iss to
 > > > say,J > > > > >*greater* privileges than the user invoking the image is entitled to > via 6 > > > > >Authorized and Default privileges in the UAF)	 > > > > >lC > > > > >You need CMKRNL to INSTALL an image and SYSNAM to define ay
 > system/exec G > > > > >logical name. If someone has these privileges then they can dod
 > anything > > > on
 > > > > >aG > > > > >system, so if an image is INSTALLed and is only accessable via$
 > SYS/EXECL > > > > >logical names then surely the image is *trusted* and therefore safe > to bee > > > > >called?	 > > > > >qF > > > > >For example, I have an external function that will take out a system > wideL > > > > >$ENQW therefore the driving image (because I don't want to give theH > > > > >privilege to every user and every image that user will execute) will > beI > > > > >INSTALLed with SYSLCK privilege. This external function is to bep	 > invoked H > > > > >from a trigger so that when a row is updated a lock is promoted which- > > > will, K > > > > >potentially, deliver a BLAST to a monitor process that is watching> > anycE > > > > >changes made to that row. This function *cannot* have SERVER  BINDINGiL > > > > >(regardless of my personal hatred of server binding and superfluousD > > > > >processes) because the MAIN image and the external function
 > communicateh
 > > > > >withdH > > > > >each other via EXTERNAL PSECTs. (IE: COBOL EXTERNAL variables).	 > > > > >tK > > > > >In summary, I have specifically and expressly given image A SYSLCKa > > > > >privilegeL > > > > >solely  so that it can be inherited by external function B (and any > otherIJ > > > > >external function that is called during image A execution as long as > it > > > isJ > > > > >INSTALLed and the image is accessed via a SYS/EXEC. Why won't RdbJ > > > > >engineering let me do this? VMS does! (IE: lib$find_image_symbol)	 > > > > >eF > > > > >These questions were posed (informally to the list server and
 > formally	 > > > vialL > > > > >Oracle support last year) to Rdb engineering and received responses > likeF > > > > >"Well, if you don't care about system security then just give
 > everyoneI > > > > >SYSPRV!" or "We don't do these things for laughs Peter, I'm sure- Rdb-: > > > > >security could be subverted. Use SERVER BINDING!"	 > > > > >3G > > > > >Now, I don't profess to know evrything, but after 18 years VMSs > > > development 
 > > > > >IK > > > > >do consider myself to be capable of reasoned debate on the subjectr > andoD > > > > >simply can't understand why Rdb insists on this extra check > > > > >(*restriction*).h > > > > >Please advise.r	 > > > > >e > > > > >Regards Richard Maher.:	 > > > > >V# > > > > >-----Original Message-----  > > > > >From: Jerry Alan Braga.. > > > > >Sent: Friday, August 25, 2000 5:23 AM> > > > > >Subject: Re: Problem with SQL$FUNCTIONS not installed	 > > > > >o	 > > > > >y! > > > > >We had the same problem.e	 > > > > >dJ > > > > >The fix is to change the SQL_FUNCTIONS.SQL and add the quailifier of > BIND > > > > >ONy? > > > > >SERVER SITE in the creation of the function/procedure.s	 > > > > >CJ > > > > >I am not at work right now so I cannot remember the exact syntax,	 > look uppD > > > > >the help on BIND ON SERVER SITE under the create function / > procedure.	 > > > > >uJ > > > > >If you can't find it I will be glad to send you my version of the	 > script.w	 > > > > >e% > > > > >----- Original Message -----  > > > > >From: "Seefelt, Beth"0 > > > > >Sent: Thursday, August 24, 2000 9:20 PM: > > > > >Subject: Problem with SQL$FUNCTIONS not installed	 > > > > >f	 > > > > >t > > > > > >L) > > > > > > Has anyone seen this before -  > > > > > >eK > > > > > > I created a view of a table and I'm trying to do an RMU /UNLOADt > > > /VIRTUAL > > > > >ofl) > > > > > > the view and get this error -t > > > > > >p> > > > > > > %RMU-E-OUTFILDEL, Fatal error, output file deletedE > > > > > > -RDB-E-EXTFUN_FAIL, external routine failed to compile or  execute  > > > > > > successfullyB > > > > > > -RDMS-E-INVRTNUSE, routine SIGN can not be used, image% > > > > > > "SQL$FUNCTIONS" not activ  > > > > > > ated@ > > > > > > -RDMS-E-RTNIMAGE_UNINST, routine image not installed0 > > > > > > -RDMS-I-TEXT, Error activating image7 > > > > > > DSA0:[SYS0.SYSCOMMON.][SYSLIB]SQL$FUNCTIONSeH > > > > > > .EXE;5, Shareable images must be installed to run privileged image  > > > > > >h > > > > > >hJ > > > > > > I tried to install SQL$FUNCTIONS but its linked with traceback so > I 	 > > > geta
 > > > > >thisu
 > > > > > > -s > > > > > >e7 > > > > > > %INSTALL-E-FAIL, failed to CREATE entry foro0 > > > > > > DISK$AXPVMSSYS:<SYS0.SYSCOMMON.SYSLI > > > > > > B>SQL$FUNCTIONS.EXErJ > > > > > > -INSTALL-E-IMGTRACED, /TRACEBACK image cannot be a privileged, > > > > > > execute_only, or > > > > > > resident image > > > > > >PL > > > > > > The table contains a computed field that uses the SIGN function. > Is > > > > >there$ > > > > > > any way I can unload it? > > > > > >- > > > > > > TIA, > > > > > >y > > > > > > Beth   ------------------------------   Date: 4 Jun 2004 01:30:34 -0700D% From: ethiobite@yahoo.com (sebastien)C- Subject: Newbie, Parameters to an applicationw= Message-ID: <bfd5f3ca.0406040030.364ec1b9@posting.google.com>r   Hello,  " I done a simple application in Ada and passed it some parameters,  + $ define application disk5:[runtime]app.exe ? $ app_with_params = "$application ""lowercase"" ""UPPERCASE"" "a $ app_with_params    That works fine, now  3 1. What is the reason of the $ on the second line ?r  # 2. Why is it not necessary to type v    $ run app_with_params    on the third line ?  6 3. What is the difference in VMS between :=, = and :==   Thanks,   
 Sbastien.   ------------------------------  $ Date: Fri, 4 Jun 2004 10:55:24 +0200- From: "Martin Vorlaender" <mv@pdv-systeme.de>h1 Subject: Re: Newbie, Parameters to an application * Message-ID: <2iaro2Fkja8tU1@uni-berlin.de>  * "sebastien" <ethiobite@yahoo.com> wrote...$ > I done a simple application in Ada  > and passed it some parameters, >e- > $ define application disk5:[runtime]app.exedA > $ app_with_params = "$application ""lowercase"" ""UPPERCASE"" "o  ; You can join those two lines into one (if you don't use thes logical name otherwise):  J $ app_with_params = "$disk5:[runtime]app.exe ""lowercase"" ""UPPERCASE"" "   > $ app_with_paramsh >h > That works fine, now >n5 > 1. What is the reason of the $ on the second line ?t  ? Syntax. A symbol with a leading $ is interpreted as a so-calledw< foreign command by DCL if it appears as the first token on a@ command line. The image following the $ is then executed by DCL.  $ > 2. Why is it not necessary to type >    $ run app_with_params >    on the third line ?  ; No need to. And besides, RUN would claim any parameters andr: qualifiers following the image name - which tends to throw@ an error messsage if they were really meant for the application.  8 > 3. What is the difference in VMS between :=, = and :==  : See the documentation at http://www.hp.com/go/openvms/doc/0 (also for "foreign command") or the online HELP.   cu,i   Martin -- d;    OpenVMS @ 25      | Martin Vorlaender  |  OpenVMS rules!y.                      | work: mv@pdv-systeme.deA    Still exceeding   |   http://www.pdv-systeme.de/users/martinv/u5    expectations      | home: martin@radiogaga.harz.des   ------------------------------  # Date: Fri, 04 Jun 2004 16:05:26 GMTu# From: hoff@hp.nospam (Hoff Hoffman)o1 Subject: Re: Newbie, Parameters to an application 1 Message-ID: <a51wc.2983$p43.304@news.cpqcorp.net>.  e In article <bfd5f3ca.0406040030.364ec1b9@posting.google.com>, ethiobite@yahoo.com (sebastien) writes:T  B :I done a simple application in Ada and passed it some parameters,  :   Please see the OpenVMS FAQ, and particularly the section  ,     "How do I run a program with arguments?"     The FAQ is available at:  %     http://www.hp.com/go/openvms/faq/r  D   I'd generally recommend the text version of the FAQ, as it is both&   easy to download and easy to search.    , :$ define application disk5:[runtime]app.exe@ :$ app_with_params = "$application ""lowercase"" ""UPPERCASE"" " :$ app_with_params :r :That works fine, now  :p4 :1. What is the reason of the $ on the second line ?  A   That is one of the two documented mechanisms for implementing a    foreign command.     You could also use:   %     $ app :== $disk5:[runtime]app.exeo!     $ app "lowercase" "UPPERCASE"1  B   Oh, and the quote characters are delimeters, but are doubled-up C   because you are entering a string and (in this case) wish to haveo4   the quote character itself included in the string.    $ :2. Why is it not necessary to type  :   $ run app_with_params  :   on the third line ?   A   That is one of the documented mechanisms for a foreign command.o  7 :3. What is the difference in VMS between :=, = and :==   A   Local symbol string assignment, local symbol string (if quoted)y=   or numeric assignment, and global symbol string assigment.  A   Respectively.  The most recent OpenVMS DCL dictionary has addedg4   entries for these constructs, akin to the @ entry.  =     OpenVMS operating system and other product documentation:   '       http://www.hp.com/go/openvms/doc/        The OpenVMS DCL Dictionary:.  >       http://h71000.www7.hp.com/doc/732FINAL/9996/9996PRO.HTML  B   You will want to review the OpenVMS User's Guide, as well as the?   DCL dictionary.  Those documents cover these topics and also  A   provide you with assistance in differentiating and in using DCL($   symbols and OpenVMS logical names.  F   From the documentation, you will learn various things.  For example,G   symbols are implemented in the command interpreter, and logical namesnD   are implemented deep in the I/O subsystem.  (That distinction willF   eventually interest you, even if you do not currently understand whyF   quite yet.  It is key to understanding the contexts in which you can/   use symbols and logical names, for instance.)$  A   Other sections of the OpenVMS FAQ that will become of interest:   +     How can I substitute symbols in a PIPE?u3     Use of RUN/DETACH, LOGINOUT, and logical names? =     Modular Programming, Facility Prefixes and Symbol Naming?_-     Why doesn't DCL symbol substitution work?l     How can I undelete a file?  E   I can also recommend my Writing Real Programs in DCL Book, too. :-)$   -- x  N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.coml   ------------------------------   Date: 3 Jun 2004 23:21:14 -0700$% From: Bart.Zorn@xs4all.nl (Bart Zorn)u Subject: Re: NFS how to = Message-ID: <a98cd882.0406032221.76f7d0cc@posting.google.com>p  u Michael Austin <maustin@firstdbasource.com> wrote in message news:<vzRvc.4368$eb1.4261@newssvr22.news.prodigy.com>...(" > VAXman- @SendSpamHere.ORG wrote:  > > Anyone using TCPIP services? > > " > > Whatever happened to examples? > > L > > How do I setup, for example, my system disk so I can NFS mount it on my  > > powerbook? > >  h > @tcpip$config  >     Services >        enable portmapper >        enable nfs  >  > tcpip  >     help createx >             directory  >             container  >          add proxy   And don't forget:o      tcpip      map      set conf map8      add exporta  	 Bart Zornp   ------------------------------  $ Date: Fri, 4 Jun 2004 11:01:57 +0200- From: "Martin Vorlaender" <mv@pdv-systeme.de>  Subject: Re: NFS how to * Message-ID: <2ias4cFkrv6oU1@uni-berlin.de>   <healyzh@aracnet.com> wrote... > VAXman- wrote:  > > Anyone using TCPIP services? >8" > > Whatever happened to examples? >eK > > How do I setup, for example, my system disk so I can NFS mount it on my  > > powerbook? > 9 > It's a bit of a pain.  Appletalk is easier to setup :^)  > H > Here are my notes that I got from somewhere (here probably) on setting > up NFS...  > = > To add NFS exports, or machines/users that can access them: = > > > ADD MAP "<unix_Style_dir_name>" <VMS_disk_or_directory> 0 > > > ADD export "<unix_Style_dir_name>" /host=*D > > > ADD proxy <vms_username> /uid=<vms_uid> /gid=<vms_gid> /host=* > > > /nfs=(outgoing,incoming)   To add to that:   ; I beleive modern UCX's add a TCPIP$NOBODY account and proxy ? (with UID/GID -2,-2) by default. If it doesn't, do it manually,  as it is needed.   cu,    Martin --  B    Emacs would be a great   | Martin Vorlaender  |  OpenVMS rules!5    operating system,        | work: mv@pdv-systeme.de H    if only it came with     |   http://www.pdv-systeme.de/users/martinv/<    a decent editor...       | home: martin@radiogaga.harz.de   ------------------------------  % Date: Fri, 04 Jun 2004 11:43:33 +0200u* From: Paul Sture <nospam@sture.homeip.net> Subject: Re: NFS how toS* Message-ID: <2iaui6FkijpvU1@uni-berlin.de>   VAXman- wrote: > Anyone using TCPIP services? >   > Whatever happened to examples? >B  ! There are examples in TCPIP HELP.1  J > How do I setup, for example, my system disk so I can NFS mount it on my  > powerbook? >     @ I've been wanting to do this as well. Here's what I have so far.  F I enabled NFS server and PORTMAPPER as suggested here, and to test it  out locally, NFS client too.    " TCPIP> map "/alpha" sys$sysdevice:- TCPIP> set config map "/alpha" sys$sysdevice:   ? The /host=* in the export command can also be a list of clients  e.g. /host=("mac",mynode)(  ' TCPIP> add export "/alpha/xfer" /host=*   5 The UIC for username STURE on my Alpha is [30,1], so:   + TCPIP> add proxy sture /uid=30/gid=1/host=*   I I've just realised that I missed out the /nfs=(outgoing,incoming) as per A* Zane's suggestion. I'll look at that next.  J TCPIP> mount dnfs1: "/alpha/xfer" /host=mynode /system /path="/alpha/xfer"B %TCPIP$DNFSMOUNT-S-MOUNTED, /alpha/xfer mounted on _DNFS1:[000000] TCPIP>
 $ sh dev dnfs$  F Device                  Device           Error    Volume         Free 	 Trans Mnt1H   Name                   Status           Count     Label        Blocks 	 Count Cnt . DNFS0:          (DINA)  Online               0F DNFS1:          (DINA)  Mounted              0  /alpha/xfer    unknown   $ dir dnfs1:[000000]   Directory DNFS1:[000000]   X.TXT;1   - Now, as described at http://tinyurl.com/ywp4d    $ tcpip . TCPIP> add proxy tcpip$nfs /uid=0/gid=1/host=*5 TCPIP> add proxy tcpip$nobody /uid=-2 /gid=-2 /host=*    On the Mac:   ( As described at http://tinyurl.com/2up6c    > mkdir alpha;  > sudo mount -o "-P" mynode:/alpha/xfer /Users/sture/alphaa  > ls alpha  ls: alpha: Permission denied   and to dismount it:t  "  > sudo umount /Users/sturep/alpha  E OK, I've tried su to use root and still get Permission denied, but I a  _can_ see the disk from the Mac:    > dflI Filesystem              512-blocks     Used    Avail Capacity  Mounted on @ /dev/disk0s5              39066000 22316600 16358744    58%    /C devfs                          180      180        0   100%    /devfC fdesc                            2        2        0   100%    /dev D <volfs>                       1024     1024        0   100%    /.volG automount -nsl [325]             0        0        0   100%    /Network < automount -fstab [331]           0        0        0   100%  /automount/Servers< automount -static [331]          0        0        0   100%  /automount/static)< dina:/alpha/xfer          17773500 14897118  2876382    84%  /Users/sture/alpha  / And both the total and free blocks are correct.    Other comments:   1 o - The mount survives a logout/login on the Mac.   C o - In Finder, the alpha subdirectory shows up with an icon with angG      arrow, which I think is an alias, because clicking on it gives the 
      message:   I         The alias "alpha" could not be opened, becauase the original item          could not be found.a  > o - In Finder -> Network I can see mac.local but nothing else.  4 That's it for the moment I'm afraid, as lunch calls.   ------------------------------  # Date: Fri, 04 Jun 2004 10:57:22 GMT " From:   VAXman-  @SendSpamHere.ORG Subject: Re: NFS how to 0 Message-ID: <00A32D89.B3816670@SendSpamHere.ORG>  W In article <2iaui6FkijpvU1@uni-berlin.de>, Paul Sture <nospam@sture.homeip.net> writes:  >VAXman- wrote:a >> Anyone using TCPIP services?y >> e! >> Whatever happened to examples?, >> >," >There are examples in TCPIP HELP. >hK >> How do I setup, for example, my system disk so I can NFS mount it on my m
 >> powerbook?  >>   > A >I've been wanting to do this as well. Here's what I have so far.  > G >I enabled NFS server and PORTMAPPER as suggested here, and to test it   >out locally, NFS client too.w >  > # >TCPIP> map "/alpha" sys$sysdevice:e. >TCPIP> set config map "/alpha" sys$sysdevice: > @ >The /host=* in the export command can also be a list of clients >e.g. /host=("mac",mynode) >i( >TCPIP> add export "/alpha/xfer" /host=* > 6 >The UIC for username STURE on my Alpha is [30,1], so: >,, >TCPIP> add proxy sture /uid=30/gid=1/host=* >tJ >I've just realised that I missed out the /nfs=(outgoing,incoming) as per + >Zane's suggestion. I'll look at that next., > K >TCPIP> mount dnfs1: "/alpha/xfer" /host=mynode /system /path="/alpha/xfer"aC >%TCPIP$DNFSMOUNT-S-MOUNTED, /alpha/xfer mounted on _DNFS1:[000000]  >TCPIP>  >$ sh dev dnfs > G >Device                  Device           Error    Volume         Free  
 >Trans MntI >  Name                   Status           Count     Label        Blocks  
 >Count Cnt/ >DNFS0:          (DINA)  Online               0_G >DNFS1:          (DINA)  Mounted              0  /alpha/xfer    unknown  >  >$ dir dnfs1:[000000]_ >, >Directory DNFS1:[000000]s >n >X.TXT;1 > . >Now, as described at http://tinyurl.com/ywp4d >e >$ tcpip/ >TCPIP> add proxy tcpip$nfs /uid=0/gid=1/host=* 6 >TCPIP> add proxy tcpip$nobody /uid=-2 /gid=-2 /host=* >  >On the Mac: >d) >As described at http://tinyurl.com/2up6c  >e > > mkdir alpha < > > sudo mount -o "-P" mynode:/alpha/xfer /Users/sture/alpha > > ls alpha >ls: alpha: Permission deniedi  K This is where I am as well.  I can't seem to figure out what "permissions" j need to be set to allow access.y     -- oB http://www.legacy-2000.com  for the *best* OpenVMS system securityC                             solutions that others only claim to be.e -- eK VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM"            r5   "Well my son, life is like a beanstalk, isn't it?" _   ------------------------------  % Date: Fri, 04 Jun 2004 12:16:33 +0100e* From: Nic Clews <sendspamhere@[127.0.0.1]> Subject: Re: NFS how toe' Message-ID: <c9plq9$46m$1@lore.csc.com>   ! VAXman-, @SendSpamHere.ORG wrote:  > Y > In article <2iaui6FkijpvU1@uni-berlin.de>, Paul Sture <nospam@sture.homeip.net> writes:  > >VAXman- wrote:s ...r> > > > sudo mount -o "-P" mynode:/alpha/xfer /Users/sture/alpha > > > ls alpha > >ls: alpha: Permission deniedo > L > This is where I am as well.  I can't seem to figure out what "permissions"! > need to be set to allow access.e    ! Maybe you need to be using PCNFS.$  D The issue is [I think] that you need a successful proxy mapping fromE your powerbook to the proxy you've added in TCPIP. The logging on the"A VMS side of NFS operations isn't great. VMS is expecting to see a47 UID/GID pair from the system trying to mount the share.h   -- o? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com$   ------------------------------   Date: 4 Jun 2004 05:02:42 -0700t3 From: pa@it.singer-friedlander.com (Piyush Avichal) + Subject: OpenVMS Primer for System Managersr= Message-ID: <eb55ac2a.0406040402.5e2e08e3@posting.google.com>O  F Anyone know whats wrong with the OpenVMS Primer for System Managers at. http://h71000.www7.hp.com/wbt/pc/welcome.htm ? Most of the pages are blank.   Has anyone used this recently?   Thanks,w   Piyush   ------------------------------  # Date: Fri, 04 Jun 2004 16:12:01 GMTR# From: hoff@hp.nospam (Hoff Hoffman)o/ Subject: Re: OpenVMS Primer for System ManagersE2 Message-ID: <lb1wc.2984$p43.2855@news.cpqcorp.net>  s In article <eb55ac2a.0406040402.5e2e08e3@posting.google.com>, pa@it.singer-friedlander.com (Piyush Avichal) writes:cG :Anyone know whats wrong with the OpenVMS Primer for System Managers atn0 :http://h71000.www7.hp.com/wbt/pc/welcome.htm...  I   Thanks for the report!  Yes, that looks to be broken.  I've passed yournH   problem report along to the OpenVMS webmaster.  (I'd encourage future H   reports be sent directly to the OpenVMS webmaster via the contact linkD   there, as problem reports here might not be seen.)  Again, thanks!  N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqyN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.comt   ------------------------------  % Date: Fri, 04 Jun 2004 07:47:25 +0200 2 From: Karl Rohwedder <emil.mustermann@t-online.de>0 Subject: Re: Print file to two seperate printers* Message-ID: <c9p2d4$j3i14@doiweb4.b2x.vwg>   Stuart Norris wrote: > Dear Readers,a > F > I have program that produces a report in FORTRAN and prints the fileG > from inside the application and then deletes the file.  I assume with ( > the open / close / dispose statements. > 1 > The program prints to a queue called a$print.  l > A > But I would like to print two copies of the report on differentx0 > locations to queues - say b$print and c$print. > D > How do I set up the a$print so that it feeds to two seperate queue> > b$print and c$print without modifiy the FORTRAN application? > ? > Is there a way to do it using GENERIC and EXECUTION queues.  e > 5 > Both b$print and c$print are DCPS execution queues.s >  > Thanks >  > StuartI Perhaps you can use the EXECSYMB symbiont, which allows you to write youra) own DCL procedure to be used as symbiont. 1 You can then do anything you want with your jobs.I   ------------------------------   Date: 4 Jun 2004 07:41:14 -0600l From: briggs@encompasserve.org0 Subject: Re: Print file to two seperate printers3 Message-ID: <VMsBnxSGUSrZ@eisner.encompasserve.org>g  o In article <51262235.0406032002.27b8ef9f@posting.google.com>, stuie_norris@yahoo.com.au (Stuart Norris) writes:e > Dear Readers,g > F > I have program that produces a report in FORTRAN and prints the fileG > from inside the application and then deletes the file.  I assume with ( > the open / close / dispose statements. > 1 > The program prints to a queue called a$print.    > A > But I would like to print two copies of the report on differentw0 > locations to queues - say b$print and c$print. > D > How do I set up the a$print so that it feeds to two seperate queue> > b$print and c$print without modifiy the FORTRAN application? > ? > Is there a way to do it using GENERIC and EXECUTION queues.  e > 5 > Both b$print and c$print are DCPS execution queues.a  A Get a copy of EXECSYMB and create a virtual queue, "a$print" that = executes by submitting the print file to the two real queues,  "b$print" and "c$print".   	John Briggs   ------------------------------   Date: 4 Jun 2004 08:16:20 -0600i- From: Kilgallen@SpamCop.net (Larry Kilgallen)l0 Subject: Re: Print file to two seperate printers3 Message-ID: <7Cz1ZGaWZ0Dc@eisner.encompasserve.org>e  o In article <51262235.0406032002.27b8ef9f@posting.google.com>, stuie_norris@yahoo.com.au (Stuart Norris) writes:e > Dear Readers,  > F > I have program that produces a report in FORTRAN and prints the fileG > from inside the application and then deletes the file.  I assume with:( > the open / close / dispose statements. > 1 > The program prints to a queue called a$print.    > A > But I would like to print two copies of the report on different 0 > locations to queues - say b$print and c$print. > D > How do I set up the a$print so that it feeds to two seperate queue> > b$print and c$print without modifiy the FORTRAN application?  E If it were just a single other queue you could stop the A$PRINT queue G and have a periodic batch job change the /COPIES= count on A$PRINT jobsh* to 2 and then requeue the jobs to B$PRINT.   ------------------------------   Date: 4 Jun 2004 08:17:31 -0600p- From: Kilgallen@SpamCop.net (Larry Kilgallen)i0 Subject: Re: Print file to two seperate printers3 Message-ID: <6IDKEtk2k3Ec@eisner.encompasserve.org>Y  o In article <51262235.0406032002.27b8ef9f@posting.google.com>, stuie_norris@yahoo.com.au (Stuart Norris) writes:n > Dear Readers,M > F > I have program that produces a report in FORTRAN and prints the fileG > from inside the application and then deletes the file.  I assume withT( > the open / close / dispose statements. > 1 > The program prints to a queue called a$print.  Y > A > But I would like to print two copies of the report on differentu0 > locations to queues - say b$print and c$print. > D > How do I set up the a$print so that it feeds to two seperate queue> > b$print and c$print without modifiy the FORTRAN application?  D Have you tried setting /RETAIN on the A$PRINT queue (or is that only for batch queues?).n   ------------------------------  $ Date: Fri, 4 Jun 2004 11:52:49 -0400< From: "Peter Weaver" <WeaverConsultingServices@sympatico.ca>0 Subject: Re: Print file to two seperate printers* Message-ID: <2ibk6jFl9tfrU1@uni-berlin.de>   Larry Kilgallen wrote: >...F > Have you tried setting /RETAIN on the A$PRINT queue (or is that only > for batch queues?).h  H /RETAIN will work on device queues and will be an easy solution for thisB problem. Set the regular queue to /RETAIN=ALL and have a .COM that@ checks for retained jobs and requeue them to the second printer.  G In the late 80's I had a system that ran Ultimate (a Pick emulator) andxE it sent jobs directly to the printer with a /DELETE. I had to set allrF the printers to /RETAIN=ALL since Ultimate reports could take hours toG run and then the printer would jam (or more often someone would pick upiG an extra sheet of paper with their report) so my operator and I  had toh often requeue jobs.r   -- s Peter Weaver Weaver Consulting Services Inc.u Canadian VAR for CHARON-VAXt www.weaverconsulting.ca    ------------------------------  $ Date: Fri, 4 Jun 2004 10:44:54 -06008 From: "Michael D. Ober" <obermd-.@.-alum-mit-edu-nospam> Subject: Scripting SET PASSWORDo0 Message-ID: <aG1wc.15$5Z4.19921@news.uswest.net>  I Our users log in to captive DCL menu scripts.  I need to be able to allowlI them to execute the SET PASSWORD command in this environment.  Any ideas?n   Thanks,t
 Mike Ober.   ------------------------------  # Date: Fri, 04 Jun 2004 16:54:59 GMT 1 From: Michael Austin <maustin@firstdbasource.com>.# Subject: Re: Scripting SET PASSWORDh; Message-ID: <DP1wc.2526$ut6.689@newssvr23.news.prodigy.com>    Michael D. Ober wrote:  K > Our users log in to captive DCL menu scripts.  I need to be able to allowaK > them to execute the SET PASSWORD command in this environment.  Any ideas?i > 	 > Thanks,B > Mike Ober. >  >    $ if menu_select .eqs.  n  $  theneF $      set password !! user will be prompted to enter old and new pwd. $      goto start_menu $  endif   Michael Austin.    ------------------------------  $ Date: Fri, 4 Jun 2004 13:20:04 -0400 From: norm.raphael@metso.com# Subject: Re: Scripting SET PASSWORD Q Message-ID: <OF8DFF609B.683A73AB-ON85256EA9.005F033B-85256EA9.005F6502@metso.com>o  F "Michael D. Ober" <obermd-.@.-alum-mit-edu-nospam> wrote on 06/04/2004 12:44:54 PM:  K > Our users log in to captive DCL menu scripts.  I need to be able to alloweK > them to execute the SET PASSWORD command in this environment.  Any ideas?g >l	 > Thanks,- > Mike Ober. >m >L5 Have your menu invoke this (###### is your own rule):R  , $TYPE DISK$SOMETHING:[SOMEDIR]SETPASSWD.COM;   $     !  $   CREATE SYS$OUTPUT   G     Enter your old password, then the new password, then retype the new D     password.  Please make sure your new password is at least ######
 characters	     long.r    % $   DEFINE/USER SYS$INPUT SYS$COMMANDe $   SET PASSWORD $   EXIT   ------------------------------  # Date: Fri, 04 Jun 2004 17:04:02 GMT11 From: Michael Austin <maustin@firstdbasource.com>t# Subject: Re: Scripting SET PASSWORD < Message-ID: <6Y1wc.2530$Dv6.1547@newssvr23.news.prodigy.com>   Michael Austin wrote:m > Michael D. Ober wrote: > L >> Our users log in to captive DCL menu scripts.  I need to be able to allowL >> them to execute the SET PASSWORD command in this environment.  Any ideas? >>
 >> Thanks,
 >> Mike Ober.- >> >> >  > $ if menu_select .eqs.  nh	 > $  then H > $      set password !! user will be prompted to enter old and new pwd. > $      goto start_menu
 > $  endif >  > Michael Austin.    duh,.. that won't work...s   ------------------------------   Date: 4 Jun 2004 07:14:55 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)g( Subject: Re: Storing system files in CMS3 Message-ID: <Wmm2dxne5Ws$@eisner.encompasserve.org>   N In article <slrncbsl25.3uk.danco@ns2.pebble.org>, danco@ns2.pebble.org writes:  C > I'd like to know when UNIX will start supporting VMS style namingdD > conventions.  The character that really seems to bother all unicesC > I've used is the dollar sign "$".  I gather UNIX file names can'ti# > have dollar characters in them.  s  G    UNIX file names certainly can have $ in them.  In act, they can haverF    any byte in them.  I've had lots of files on UNIX systems with $ in    them.  A    What UNIX does with $ is another thing.  Most shells use it to>    delimit substitution.  G    But it's more fund to have a file actually nameed *.c in a directory)    full of C source files.   ------------------------------   Date: 4 Jun 2004 09:20:18 -0700h- From: soccer13player@yahoo.com (Nom de Plume) ( Subject: Re: Storing system files in CMS= Message-ID: <f401eb7f.0406040820.23525362@posting.google.com>    Martin Kirby wrote:n > Nom de Plume wrote:u@ > > Does CMS support complete UNIX style naming conventions yet? > E > It has full support for ODS-5 style element names. It does not havey( > support for case sensitive file names. >  c > > J > > One really great thing I would like added to CMS is the replacement ofE > > __GENERATION__ within a file like I have available from sccs/rcs.  > >  > H > Not sure what this means. CMS V4.2 (I think) added the ability to do aA > REPLACE/INSERT which inserts the new generation into class(es).i >   C So, it still does not fully support UNIX-style naming.  It would ben& nice if VMS engineering can get to it.  < No, one can add a symbolic constant to a file something likeF __generation__ (I do not remember the exact symbol) and when you fetch< the file from rcs/sccs it is automatically replaced with the generation number.   JMOD   ------------------------------  % Date: Fri, 04 Jun 2004 08:13:44 +0200  From: Dirk Munk <munk@home.nl>, Subject: Re: Trying to do sftp via batch job2 Message-ID: <c9p3uu$mdu$1@news1.tilbu1.nb.home.nl>   James T Horn wrote: & > When trying to run a batch job with: > F > $sftp "-b" /rucs1/ucs/data/cbme01f_get1.txt username@sftp.domain.com >  > I get an error:,$ > FATAL: OpenVMS$tty_gettty() failed8 > %TCPIP-F-SSH_FATAL, non-specific fatal error condition > O > This same command "$ sftp..." works fine when run from a terminal. Any ideas? ( Yes, that is a well known silly problem.P These SSH utilities normally have those wonderfull progress lines, ( 20% done - M 40% done etc.) In order to do that the utility needs (vt100) cursor control. oP When the utility starts, it checks if the terminal supports cursor control. And ? gueass what? A batch terminal does not have cursor control.....    ------------------------------   Date: 4 Jun 04 06:18:35 PSTa From: mckinneyj@cpva.saic.com , Subject: Re: Trying to do sftp via batch job( Message-ID: <y9YbHABwO$tl@cpva.saic.com>  = In article <843706dc.0406031329.597aa809@posting.google.com>, %  horn@shsu.edu (James T Horn) writes: & > When trying to run a batch job with: > F > $sftp "-b" /rucs1/ucs/data/cbme01f_get1.txt username@sftp.domain.com >  > I get an error:e$ > FATAL: OpenVMS$tty_gettty() failed8 > %TCPIP-F-SSH_FATAL, non-specific fatal error condition > O > This same command "$ sftp..." works fine when run from a terminal. Any ideas?s    F In instances where a utility must be associated with a terminal, and ID need it to run in batch, I use a homegrown PTD$ application that I'mB not currently in a position to share. However, there is an exampleG (that I've never used) of such an app, written in C, that you can fetchn from  0 http://mvb.saic.com/freeware/vmslt03a/vu/ptd.zip  ! and probably adapt to your needs.I   ------------------------------   End of INFO-VAX 2004.309 ************************