1 INFO-VAX	Sun, 18 May 2003	Volume 2003 : Issue 273       Contents:
 Re: <None>
 Re: <None>' Re: <None> ( Broken SMTP% when sending) ' Re: <None> ( Broken SMTP% when sending) ' Re: <None> ( Broken SMTP% when sending)  Another newbie question  Re: Another newbie question  Re: Another newbie question  Re: as2100 inside pictures* Re: Backspace on DECterm vs TPU (Xwindows)* Re: Backspace on DECterm vs TPU (Xwindows)* Re: Backspace on DECterm vs TPU (Xwindows)( Re: Damned power spike. Need new monitor( Re: Damned power spike. Need new monitor( Re: Damned power spike. Need new monitor( Re: Damned power spike. Need new monitor3 Re: DCL command implementation (was: WHICH command)   Re: Determining Disk Booted From DFU question Re: DFU question Re: Error message help needed  Re: Error message help needed  Re: Error message help needed  Re: Error message help needed  Re: Error message help needed  Re: Error message help needed : Re: getting data off an OpenVMS Alpha to Microsoft Windows2 Re: WHICH command (was: Error message help needed)2 RE: WHICH command (was: Error message help needed)2 RE: WHICH command (was: Error message help needed)2 RE: WHICH command (was: Error message help needed)2 RE: WHICH command (was: Error message help needed)2 RE: WHICH command (was: Error message help needed)2 Re: WHICH command (was: Error message help needed)2 Re: WHICH command (was: Error message help needed)  F ----------------------------------------------------------------------  # Date: Sat, 17 May 2003 17:46:09 GMT 1 From: Michael Austin <maustin@firstdbasource.com>  Subject: Re: <None> 2 Message-ID: <3EC67455.2A894F30@firstdbasource.com>   Paul Sture wrote:  > I > So, VMS mail doesn't seem to work to the outside world since yesterday. G > Never any problems before, but now it gives me this kind of response:  >  > MAIL> send/edit + > To:     somevalidaddress@validaddress.com  > CC:  > Subj:   Something $ > %MAIL-E-SENDABORT, no message sent+ > %TCPIP-E-SMTP_ABORT, SMTP session aborted  >  > Doing: >  > TCPIP STOP MAIL  > TCPIP START MAIL > 5 > appears to solve the problem, but only temporarily.  > " > Nope, even that no longer works. > 0 > Has anyone seen this, or has any ideas please? > -- > Paul Sture    0 What version of VMS/TCPIP+patches are you using?    Try increasing CLTPAGES to 1600.       --   Regards,  6 Michael Austin            OpenVMS User since June 19847 First DBA Source, Inc.    Registered Linux User #261163 7 Sr. Consultant            http://www.firstdbasource.com  816-373-0270 (Office)  816-728-3080 (Mobile)    ------------------------------   Date: 18 May 03 06:47:22 +0200) From: p_sture@elias.decus.ch (Paul Sture)  Subject: Re: <None> ) Message-ID: <q7OMJFRphYRm@elias.decus.ch>   f In article <%2uxa.915979$S_4.927055@rwcrnsc53>, brad@.gateway.2wire.net (Bradford J. Hamilton) writes:W > In article <gQB5sqlFzkKo@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes: I >>So, VMS mail doesn't seem to work to the outside world since yesterday. G >>Never any problems before, but now it gives me this kind of response:  >> >>MAIL> send/edit + >>To:     somevalidaddress@validaddress.com 
 >>CC:      >>Subj:   Something $ >>%MAIL-E-SENDABORT, no message sent+ >>%TCPIP-E-SMTP_ABORT, SMTP session aborted  >> >>Doing: >> >>TCPIP STOP MAIL  >>TCPIP START MAIL   >>5 >>appears to solve the problem, but only temporarily.  >>" >>Nope, even that no longer works. >>0 >>Has anyone seen this, or has any ideas please? > Q > I've never seen it, but I have an idea.  Any chance you are running out of disk 2 > space on the disk where SMTP or the queues live? >   F Approximately half a million free blocks on my system disk. I rebootedF and all seems OK so far, but I'd rather find out why. Did someone sendI me a malformed message which caused it to cease to function, for example?   K I have however, taken someone's offline advice and set TCPIP$SMTP_LOG_LEVEL F to 5 (done in SYS$STARTUP:TCPIP$SMTP_SYSTARTUP.COM), so we'll see what happens. --  
 Paul Sture   ------------------------------  % Date: Sat, 17 May 2003 14:19:51 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>0 Subject: Re: <None> ( Broken SMTP% when sending)) Message-ID: <3EC67D45.C4A72725@istop.com>    Paul Sture wrote: + > To:     somevalidaddress@validaddress.com  > CC:  > Subj:   Something $ > %MAIL-E-SENDABORT, no message sent+ > %TCPIP-E-SMTP_ABORT, SMTP session aborted      What if you try:  	 SEND/EDIT   L at what point does it fail ? before you get to the editor, or after you have, saved the message, expecting it to be sent ?  J The error message you have provided would normally be issued if you CTRL-CL while entering the message, or if you do not save the file using the editor.  K When you first invoke the smtp% transport for a message, a lot of files are K checked, so if you get past the first TO:, it works, even if the TCPIP STOP L MAIL has been issued. (the foreign image only queues the entries, and queingD work when the queue is stopped so you can still send emails to smtp%# destination if the mail is stopped.   K The first TO: also results in the TCPIP/SMTP control file being created. (I   believe in your mail directory).  L Without the /EDIT, after you have entered The subject, it will create a fileM in your sys$scratch: (default to sys$login). And then will write your message K line by ine to that file and it is only once you have entereted ctrl-Z that  the processing really begins.   M If I were you, I would use SET WATCH FILE  to check to see what MAIL tries to # do just before you get the message.    ------------------------------  # Date: Sat, 17 May 2003 21:28:36 GMT 1 From: Michael Austin <maustin@firstdbasource.com> 0 Subject: Re: <None> ( Broken SMTP% when sending)2 Message-ID: <3EC6A874.33FD607D@firstdbasource.com>   JF Mezei wrote:  >  > Paul Sture wrote: - > > To:     somevalidaddress@validaddress.com  > > CC:  > > Subj:   Something & > > %MAIL-E-SENDABORT, no message sent- > > %TCPIP-E-SMTP_ABORT, SMTP session aborted  >  > What if you try: >  > SEND/EDIT  > N > at what point does it fail ? before you get to the editor, or after you have. > saved the message, expecting it to be sent ? > L > The error message you have provided would normally be issued if you CTRL-CN > while entering the message, or if you do not save the file using the editor. > M > When you first invoke the smtp% transport for a message, a lot of files are M > checked, so if you get past the first TO:, it works, even if the TCPIP STOP N > MAIL has been issued. (the foreign image only queues the entries, and queingF > work when the queue is stopped so you can still send emails to smtp%% > destination if the mail is stopped.  > M > The first TO: also results in the TCPIP/SMTP control file being created. (I " > believe in your mail directory). > N > Without the /EDIT, after you have entered The subject, it will create a fileO > in your sys$scratch: (default to sys$login). And then will write your message M > line by ine to that file and it is only once you have entereted ctrl-Z that  > the processing really begins.  > O > If I were you, I would use SET WATCH FILE  to check to see what MAIL tries to % > do just before you get the message.    Jeff,   G The reason for my previous question regarding versions is a known issue F with STMP getting hung, and stopping/restarting mail and in some casesE increasing CTLPAGES will correct it.  You can search the latest TCPIP & patches for the fix to this problem.     --   Regards,  6 Michael Austin            OpenVMS User since June 19847 First DBA Source, Inc.    Registered Linux User #261163    ------------------------------   Date: 18 May 03 07:26:51 +0200) From: p_sture@elias.decus.ch (Paul Sture) 0 Subject: Re: <None> ( Broken SMTP% when sending)) Message-ID: <mhEda8+HLJPW@elias.decus.ch>   V In article <3EC67D45.C4A72725@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes: > Paul Sture wrote: , >> To:     somevalidaddress@validaddress.com >> CC: >> Subj:   Something% >> %MAIL-E-SENDABORT, no message sent , >> %TCPIP-E-SMTP_ABORT, SMTP session aborted >  >  > What if you try: >  > SEND/EDIT  >   E With a straight SEND, or SEND/EDIT, it failed after hitting return on  the subject line.    N > at what point does it fail ? before you get to the editor, or after you have. > saved the message, expecting it to be sent ? > L > The error message you have provided would normally be issued if you CTRL-CN > while entering the message, or if you do not save the file using the editor. > M > When you first invoke the smtp% transport for a message, a lot of files are M > checked, so if you get past the first TO:, it works, even if the TCPIP STOP N > MAIL has been issued. (the foreign image only queues the entries, and queingF > work when the queue is stopped so you can still send emails to smtp%% > destination if the mail is stopped.  > M > The first TO: also results in the TCPIP/SMTP control file being created. (I " > believe in your mail directory). > N > Without the /EDIT, after you have entered The subject, it will create a fileO > in your sys$scratch: (default to sys$login). And then will write your message M > line by ine to that file and it is only once you have entereted ctrl-Z that  > the processing really begins.  > O > If I were you, I would use SET WATCH FILE  to check to see what MAIL tries to % > do just before you get the message.   H SET WATCH is a fine idea, thanks. Unfortunately I haven't yet worked out how to reproduce the problem.    --  
 Paul Sture   ------------------------------    Date: 17 May 2003 16:59:15 -0500' From: "Glen G." <acme@aodontspamme.net>   Subject: Another newbie question1 Message-ID: <01c31cbf$15bc5060$2b8f4f3f@thegoodw>   G Thanks to everyone who helped with my problem running "hello.exe."  RUN  HELLO works : like a champ!  It's good to see such an active group here.  H Now I'd like to know -- where are the rest of the layered products?  The OpenVMS 7.3 CD-ROMJ which I received has about 15 products, but the OpenVMS Hobbyist site says that "over 100 of C Compaq's OpenVMS Layered Products are available for the Hobbyist."   Following this statement' is a list of, well, about 100 products.   I Where do I find them?  I'm particularly interested in the COBOL compiler.    Thanks again --    Glen 0/0       > -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----A http://www.newsfeeds.com - The #1 Newsgroup Service in the World! > -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----   ------------------------------  # Date: Sat, 17 May 2003 22:59:27 GMT 4 From: brad@.gateway.2wire.net (Bradford J. Hamilton)$ Subject: Re: Another newbie question. Message-ID: <j9zxa.879991$F1.110239@sccrnsc04>  [ In article <01c31cbf$15bc5060$2b8f4f3f@thegoodw>, "Glen G." <acme@aodontspamme.net> writes: H >Thanks to everyone who helped with my problem running "hello.exe."  RUN >HELLO works; >like a champ!  It's good to see such an active group here.  > I >Now I'd like to know -- where are the rest of the layered products?  The  >OpenVMS 7.3 CD-ROM K >which I received has about 15 products, but the OpenVMS Hobbyist site says  >that "over 100 ofD >Compaq's OpenVMS Layered Products are available for the Hobbyist."  >Following this statement ( >is a list of, well, about 100 products. > J >Where do I find them?  I'm particularly interested in the COBOL compiler.  M They can be found on a quarterly distributuion that HP sends to its customers E who pay for the service.  The distribution contains a number of CD's.   M If you know of a person whose company has one of these subscriptions, then it J is legal for you to "borrow" the subscription, and have access to the LP'sH mentioned in the Hobbyist site.  You probably need to develop a suitable "bribe" for your friend.	:-)  N I checked the Hobbyist site, and was surprised to find that the COBOL compilerN was not included on the VAX CD - it was certainly part of the (sold-out) Alpha CD.    >  >Thanks again -- >  >Glen  >0/0 >  >  > ? >-----= Posted via Newsfeeds.Com, Uncensored Usenet News =----- B >http://www.newsfeeds.com - The #1 Newsgroup Service in the World!? >-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----   A _________________________________________________________________ 0 Bradford J. Hamilton			"All opinions are my own"/ bMradAhamiPltSon@atMtAbi.cPoSm		"Lose the MAPS"    ------------------------------  % Date: Sat, 17 May 2003 22:55:18 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> $ Subject: Re: Another newbie question' Message-ID: <3EC70426.7BD7CCA4@fsi.net>    "Bradford J. Hamilton" wrote:  > ] > In article <01c31cbf$15bc5060$2b8f4f3f@thegoodw>, "Glen G." <acme@aodontspamme.net> writes: J > >Thanks to everyone who helped with my problem running "hello.exe."  RUN > >HELLO works= > >like a champ!  It's good to see such an active group here.  > > K > >Now I'd like to know -- where are the rest of the layered products?  The  > >OpenVMS 7.3 CD-ROM M > >which I received has about 15 products, but the OpenVMS Hobbyist site says  > >that "over 100 ofE > >Compaq's OpenVMS Layered Products are available for the Hobbyist."  > >Following this statement * > >is a list of, well, about 100 products. > > L > >Where do I find them?  I'm particularly interested in the COBOL compiler. > O > They can be found on a quarterly distributuion that HP sends to its customers G > who pay for the service.  The distribution contains a number of CD's.   H The name of the distro., BTW, is "Software Product Library" or "SPL" forG short. Sometimes referred to by it's old name, the "ConDist", short for  "Consolidated Distribution".  O > If you know of a person whose company has one of these subscriptions, then it G > is legal for you to "borrow" the subscription, and have access to the    ...distribution for the ...    > LP'sJ > mentioned in the Hobbyist site.  You probably need to develop a suitable% > "bribe" for your friend.        :-)  > P > I checked the Hobbyist site, and was surprised to find that the COBOL compilerP > was not included on the VAX CD - it was certainly part of the (sold-out) Alpha > CD.   F Since Alpha software occupies more storage "real estate" than it's VAXF counterparts, this is understandable. You can only cram so much onto aF 650MB CD-ROM. Of course, using a smaller clustersize (it was 3 for theB V7.2 Alpha CD, 2 is the minimum for an ODS-2(.0) disk that size, I believe) might help some.   H I could send you my totally unofficial and unapproved hobbyist's layeredG product distro., or you can look up the DCL code to produce your own on  my freeware site:   # http://www.djesys.com/freeware/vms/    Look for hbyspl010.zip.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 17 May 2003 18:33:39 -0400 1 From: Ed Wensell III <ewensell3@yahoo.commercial> # Subject: Re: as2100 inside pictures 0 Message-ID: <3EC6B8C3.7D31C2B7@yahoo.commercial>   Roland Barmettler wrote: > 
 > Hi Antti > 1 > I've got some pictures of my AS2100 on the Web: % > http://naboo.freestone.net/sysinfo/   H You're PHP interface to MONITOR is very nice. Something you put together or was it found on the web?    --  F Ed Wensell III - E-mail address is valid if you know the right bits to drop.   > There is no reason anyone would want a computer in their home.@ - Ken Olsen, President & founder of Digital Equipment Corp, 1977   ------------------------------   Date: 17 May 2003 17:57:05 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>3 Subject: Re: Backspace on DECterm vs TPU (Xwindows) + Message-ID: <ba5t5h0u13@enews1.newsguy.com>   + JF Mezei <jfmezei.spamnot@istop.com> wrote: O > On a Xserver (the MI-X on mac), I can run the DECTerm fine with the backspace " > key acting as the VMS backspace.  K > However, if I start a TPU window on the same Xserver, the use of the same 1 > backspace key generates the <CTRL-H> behaviour.    > Any idea why ?  H If all you need is a DECterm, you might want to try the dataComet demo. K I'll be buying a copy as soon as I get some other software I need more.  It L works better than anything else I've tried on the Mac for getting a terminal7 window on my VMS systems.  In fact it's almost perfect.   J I had pretty good luck using eXodus for displaying DECwindows, but I can'tJ remember what kinds of problems I ran into with the terminal.  Running MacA OS X, and Apple's new X-Windows software is pretty much hopeless.    			Zane    ------------------------------  % Date: Sat, 17 May 2003 14:43:59 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>3 Subject: Re: Backspace on DECterm vs TPU (Xwindows) ) Message-ID: <3EC682EB.FA4A41A8@istop.com>    "Zane H. Healy" wrote:I > If all you need is a DECterm, you might want to try the dataComet demo.   L The MI-X server works fine for decterm, mail etc. It is only with TPU that I7 have a problem where the backspace acts as the <ctrl-h>    ------------------------------   Date: 18 May 03 07:21:20 +0200) From: p_sture@elias.decus.ch (Paul Sture) 3 Subject: Re: Backspace on DECterm vs TPU (Xwindows) ) Message-ID: <dU5pvLOVuDgU@elias.decus.ch>   ` In article <ba5t5h0u13@enews1.newsguy.com>, "Zane H. Healy" <healyzh@shell1.aracnet.com> writes:- > JF Mezei <jfmezei.spamnot@istop.com> wrote: P >> On a Xserver (the MI-X on mac), I can run the DECTerm fine with the backspace# >> key acting as the VMS backspace.  > L >> However, if I start a TPU window on the same Xserver, the use of the same2 >> backspace key generates the <CTRL-H> behaviour. >  >> Any idea why ?  > J > If all you need is a DECterm, you might want to try the dataComet demo. M > I'll be buying a copy as soon as I get some other software I need more.  It N > works better than anything else I've tried on the Mac for getting a terminal9 > window on my VMS systems.  In fact it's almost perfect.  > L > I had pretty good luck using eXodus for displaying DECwindows, but I can'tL > remember what kinds of problems I ran into with the terminal.  Running MacC > OS X, and Apple's new X-Windows software is pretty much hopeless.   K Umm? I am typing this on my OS X system, using the Terminal utility, logged O into a VMS machine, and using TPU as the editor. It works fine for me, with the M exception of the accented keys, which have never worked as advertized in much O else but a few applications. Damn sloppy implementation for those of us who use I different keyboards IMNSHO, but the VT100 emulation works far better than  many others I have seen.  B I have not tried any version of X-Windows on this system, however.   --  
 Paul Sture   ------------------------------  # Date: Sat, 17 May 2003 18:14:05 GMT + From: "Joe Heimann" <heimann@ecs.umass.edu> 1 Subject: Re: Damned power spike. Need new monitor + Message-ID: <3ec67beb@news-1.oit.umass.edu>   " John Smith <a@nonymous.com> wrote:  / > "w_tom" <w_tom1@hotmail.com> wrote in message ' > news:3EC5613E.BDAF19E1@hotmail.com... ? >>   That plug-in UPS fails to tell the whole truth.  Such half  >> earth ground."    (much garbage snipped)  C > Did I leave out that we have rather large surge protection device F > attached to the main, after the meter but before power is fed to theH > transfer switch/distribution panel? And that our system is grounded toF > rather lengthy (100+ feet) interconnected large diameter copper rodsH > buried in the ground below the frost line? The generator also has someC > circuitry built-in to do surge protection. We use the point surge H > protection to save us from any variability/transients in the generatorC > output. And the local UPS'es handle over/under voltages and short C > interruptions quite well. We've never been down except when we've  > wanted to be down.  G Just ignore w_tom, he seems to have a search robot going on his usenet  G feed looking for any mention of spikes, UPS's and surge protectors.  He E then posts his usual polemic about surge protectors and getting whole F house protection, etc.  EE's and electricians I have shown it say halfF the info is good, the rest is at best poorly understood by the poster.   Joe Heimann    ------------------------------  % Date: Sat, 17 May 2003 23:28:20 +0100 ) From: Antonio Carlini <arcarlini@iee.org> 1 Subject: Re: Damned power spike. Need new monitor $ Message-ID: <3EC6B784.80603@iee.org>  O > Anyhow, I have found some wharehouse in montreal full of used equipment and I O > suspect I'll be able to get a nice deal. Ruins the weekend, but I may go this = > saturday. Perhaps a new definition for "hardware store" :-)   8 Given that it's Saturday evening now where you are, it's9 probably too late ... but did you check the internal fuse - that is likely to live in your monitor's psu?   8 All the monitors I've repaired that were completely dead9 (OK .. both of them!) just needed a new fuse. You need to 8 be careful to replace with an identical fuse (as well as; just being *careful* while inside the monitor) but it would 9 be a shame to lose a good monitor for the want of a fuse!    Antonio      --     --   --------------- - Antonio Carlini             arcarlini@iee.org    ------------------------------  % Date: Sat, 17 May 2003 19:09:36 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>1 Subject: Re: Damned power spike. Need new monitor ) Message-ID: <3EC6C12F.B2BB4AFA@istop.com>    Antonio Carlini wrote:: > Given that it's Saturday evening now where you are, it's; > probably too late ... but did you check the internal fuse / > that is likely to live in your monitor's psu?   M I replaced the externally accessible fuse (not obvious how to get to it), but I I lack the tool to open the monitor makes it impossible for me to open it C without seriously damaging the casing. (the screws require a female L screwdriver that is thin enough to fit inside the narrow cavity/hole. OK notM as bad as the original Macintosh computers, but definitely more "proprietary" A than all others DEC monitors/terminals I have opened in the past.    ------------------------------  % Date: Sat, 17 May 2003 19:13:27 -0400   From: w_tom <w_tom1@hotmail.com>1 Subject: Re: Damned power spike. Need new monitor + Message-ID: <3EC6C217.7F85FD40@hotmail.com>   ;   That "surge protection device attached to the main, after 1 the meter but before power is fed to the transfer 5 switch/distribution panel" is effective 'whole house' = protector.  Those large diameter copper rods buried below the > frost line would be effective surge protection.  You have what; most facilities don't even know is required - serious surge 6 protection.  Also that generator should contain proper8 equipment making point of use protectors just a waste of; money.  Any surge that should approach electronic equipment < would be well below existing thresholds of properly designed< equipment.  For ATX power supplies, that would be 1000 volts< differential mode.  No backup generator should generate 1000= volt transients on 120 VAC distribution.  Therefore 'point of 0 use' protectors only enrich their manufacturers.  6   If something is a source of transients, then why put5 transient suppressers everywhere else?  Instead quash > transients at the source.  If your gas driven generator system> is properly designed (and it sounds as if it was), then it has2 required protection - at the potential source of a; transients.  Specifications should confirm this.  Transient : are quashed at that backup generator making 'point of use'& protectors unnecessary and redundant.    John Smith wrote: C > Did I leave out that we have rather large surge protection device F > attached to the main, after the meter but before power is fed to theH > transfer switch/distribution panel? And that our system is grounded toF > rather lengthy (100+ feet) interconnected large diameter copper rodsH > buried in the ground below the frost line? The generator also has someC > circuitry built-in to do surge protection. We use the point surge H > protection to save us from any variability/transients in the generatorC > output. And the local UPS'es handle over/under voltages and short C > interruptions quite well. We've never been down except when we've  > wanted to be down.   ------------------------------    Date: 17 May 2003 17:17:28 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) < Subject: Re: DCL command implementation (was: WHICH command)3 Message-ID: <5V0yC+vl1gz3@eisner.encompasserve.org>   _ In article <CIEJLCMNHNNDLLOOGNJIGEGOHDAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:   G >>>>In the case of the second and third commands, the executable is not H >>>>SHOW, but SETSHOSECUR.EXE and SETSHOSERVER.EXE respectively. NeitherF >>>>is found by reference to the PATH variable list, but rather by the >>>>DCLTABLES data.  >>> 7 >>> I suspected there was more to it.  Not very pretty.  >>F >>Ah, but it does not have to be pretty; it just has to work properly,! >>and it has done that for years.  >>= >>The attempt to make all operating systems appear to be Unix  >>is what is not pretty. > M > Actually, I was thinking more of the grammatical ambiguity and the need for I > special casing in the DCl parser, but then that is not uncommon.  There  > would J > seem to be some overlap between the parsing performed by DCL and that by > SHOW. L > My preference would be to minimize interfaces and keep things as simple as, > possible. De gustibus non est disputandem.  C But for reliability, they have aggregated into one image everything C concerned with SET and SHOW of Security items and another image all % that does the same for Server images.   A Looking at DCLTABLES data it might not look pretty but looking at @ the source of the images themselves it is much more orderly than the alternative.   ------------------------------  # Date: Sun, 18 May 2003 01:20:00 GMT 1 From: Michael Austin <maustin@firstdbasource.com> ) Subject: Re: Determining Disk Booted From 2 Message-ID: <3EC6DEAF.E2E5AB8C@firstdbasource.com>   John Travell wrote:  > 0 > <rob.buxton@wcc.spam.govt.nz> wrote in message  > news:3ebeebd8.16619117@news...5 > > On Fri, 9 May 2003 23:28:04 +0100, "John Travell"   > > <john@travell.uk.net> wrote:H > > I think this will just show you which is considered to be the ShadowC > > Set Master, but not necessarily the device that was booted off.  > > H > > If you've split and reformed the shadow set since the last boot thenC > > the "Master" will simply be the device that was not dismounted.  > > C > > Some of the f$getdvi lexicals can be used to look at the Device G > > Members, but again, these just relate to the shadow sets and do not ! > > give you what you were after.  > > N > Comes of trying to answer before I have read the question at least twice :-( >  > >  > > > @ > > >"Jack Trachtman" <Jack.Trachtman@vmmc.org> wrote in message; > > >news:69d784c4.0305091349.f99ded2@posting.google.com...  > > >> [VMS V7.3]  > > >>< > > >> All our VMS systems have their system disks shadowed. > > >>= > > >> How can I find out which disk of the pair was actually  > > >> used to boot from?  > > >> >  > $ ana/sys * > SDA> show device/address=@sys$ar_bootucbC > works in V7.3-1, I cannot verify that it works in older versions.  >  > -- > John Travell" > VMS crashdump expertise for hire > john@travell.uk.net  > http://www.travell.uk.net/ >  > --- ( > Outgoing mail is certified Virus Free.< > Checked by AVG anti-virus system (http://www.grisoft.com).C > Version: 6.0.478 / Virus Database: 275 - Release Date: 06/05/2003    works in 7.2-1h1   --   Regards,  6 Michael Austin            OpenVMS User since June 19847 First DBA Source, Inc.    Registered Linux User #261163    ------------------------------  % Date: Sat, 17 May 2003 21:32:11 -0500 ( From: brandon@dalsemi.com (John Brandon) Subject: DFU question 1 Message-ID: <03051721321139@dscis6-0.dalsemi.com>   O Since you guys introduced me to the freeware of DFU I have some basic questions 3 about the package.  Specifically the DEFRAG option.   9 Has anyone had problems with it?  Namely file corruption?   9 Run it against all file types (*.*.*) or typically ASCII?   J Are there any drives or file types that would not be a good idea to run it) againts?  System drive?  Pathworks?  etc.    Thanks!        John Brandon VMS Systems Administrator  Dallas Semiconductor first.last@dalsemi.com 972.371.4172 wk    ------------------------------  % Date: Sat, 17 May 2003 23:11:43 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: DFU question ' Message-ID: <3EC707FF.774B80F0@fsi.net>    John Brandon wrote:  > Q > Since you guys introduced me to the freeware of DFU I have some basic questions 5 > about the package.  Specifically the DEFRAG option.  > ; > Has anyone had problems with it?  Namely file corruption?   D Not to date. However, remember that DFU *IS* unsupported freeware...  ; > Run it against all file types (*.*.*) or typically ASCII?   F I know that "*.*.*" is legal instead of "*.*;*", but careful: on ODS-5A with extended filespec processing enabled, it may not do what you F expect: return all versions of all matching filespec.'s. In such case,D it may only return filespec.'s containing two periods. Dunno, but itF seems reasonable to expect. I've not had the chance to play with ODS-5 yet.  F To answer your question, however, the filetype extension and/or actualG content is meaningless at the filesystem level (it may be meaningful at F the RMS level, but not ODS); so, yes, content type can be ignored from this perspective.   L > Are there any drives or file types that would not be a good idea to run it+ > againts?  System drive?  Pathworks?  etc.   C Well, understand that DFU DEFRAG will defragment (either totally or D /BESTTRY) *FILES*, but not freespace. In fact, doing so will tend to@ aggravate freespace fragmentation. You may aid performance usingH existing files; however, newly created files will be more likely to have1 many fragments if not created CONTIGUOUS_BESTTRY.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 17 May 2003 12:36:00 -0500 ( From: Michael Rice <marice@whiteice.com>& Subject: Re: Error message help needed. Message-ID: <vccso1daqtr82@corp.supernews.com>  - On 5/17/2003 6:14 AM, Robert Deininger wrote: H > In article <CIEJLCMNHNNDLLOOGNJIIEFLHDAA.tom@kednos.com>, "Tom Linden" > <tom@kednos.com> wrote:  >  >  >>It would be nice to say  >> >>$ which hello  >  > . > Well, I can "say" that to my VMS system now. > I > You must be thinking of a "which" command from some other environment.  1 > Care to explain to the rest of us what it does?   B The 'which' command searches the 'path' (and aliases) to find the G particular instance of an executable that would be executed.  So, if I  5 were to type 'which tar' on my SGI, it would respond  I '/usr/freeware/bin/tar', even though there is also a 'tar' executable in  E '/sbin/tar'.  The reason being that '/usr/freeware/bin' comes before  F '/sbin' in my 'path' environment variable.  This can be very handy if ' you don't know where a program resides.   I All well and good for Unix-like environments, but since VMS doesn't have  D the concept of an executable search path, it doesn't really matter. 3 However, if it did, 'which authorize' would return  G 'SYS$SYSROOT:[SYSEXE]AUTHORIZE.EXE', regardless of the current default  
 directory.   ------------------------------   Date: 17 May 2003 15:07 CDT ' From: carl@gerg.tamu.edu (Carl Perkins) & Subject: Re: Error message help needed- Message-ID: <17MAY200315075650@gerg.tamu.edu>   , Michael Rice <marice@whiteice.com> writes.... }On 5/17/2003 6:14 AM, Robert Deininger wrote:I }> In article <CIEJLCMNHNNDLLOOGNJIIEFLHDAA.tom@kednos.com>, "Tom Linden"  }> <tom@kednos.com> wrote: }>   }>   }>>It would be nice to say }>>  }>>$ which hello }>   }>  / }> Well, I can "say" that to my VMS system now.  }>  J }> You must be thinking of a "which" command from some other environment. 2 }> Care to explain to the rest of us what it does? } C }The 'which' command searches the 'path' (and aliases) to find the  H }particular instance of an executable that would be executed.  So, if I 6 }were to type 'which tar' on my SGI, it would respond J }'/usr/freeware/bin/tar', even though there is also a 'tar' executable in F }'/sbin/tar'.  The reason being that '/usr/freeware/bin' comes before G }'/sbin' in my 'path' environment variable.  This can be very handy if C( }you don't know where a program resides. } J }All well and good for Unix-like environments, but since VMS doesn't have E }the concept of an executable search path, it doesn't really matter. t4 }However, if it did, 'which authorize' would return H }'SYS$SYSROOT:[SYSEXE]AUTHORIZE.EXE', regardless of the current default  }directory.D  G Not quite. Recent versions of VMS do have a search path for executablesa (and command files).  M A roughly equivalent command on VMS would first check to see if the parametereI was a symbol, if so it would return the value of the symbol. If not, thenmM would come the hard part - it would have to check to see if the parameter wasiO in the process's command tables, if so it would tell you what image the commandeK tables indicate would be run (in some cases, like when different qualifiers M cause different images to be run, this could be difficult to do in a sensiblepK manner unless you parse the parameter in the same way DCL does to determinenJ what parameters, if any, were specified and what image that works out to).E If it wasn't in the command tables, you would, at long last, check togI see if the path logical DCL$PATH is defined - if it is not, then there isCI no such command, if it is then you'd do an F$SEARCH("DCL$PATH:''p1'.EXE")eH or equivalent in whatever programming language you are using to find outG what .EXE it would run (you may need logic to check to see if P1 has an I extender already) and display the result if there is one, otherwise you'daD need to do the same but for .COM to see if that gives you any resultA (displaying it if there is on, or reporting that there is no suchs command if not).   Or something along those lines.    --- Carl   ------------------------------    Date: 17 May 2003 15:54:00 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)d& Subject: Re: Error message help needed3 Message-ID: <31BcMEmTmyRm@eisner.encompasserve.org>T  W In article <17MAY200315075650@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl Perkins) writes:   O > A roughly equivalent command on VMS would first check to see if the parameterlK > was a symbol, if so it would return the value of the symbol. If not, thenaO > would come the hard part - it would have to check to see if the parameter wasCQ > in the process's command tables, if so it would tell you what image the commandnM > tables indicate would be run (in some cases, like when different qualifierssO > cause different images to be run, this could be difficult to do in a sensible-M > manner unless you parse the parameter in the same way DCL does to determine2L > what parameters, if any, were specified and what image that works out to).G > If it wasn't in the command tables, you would, at long last, check to@K > see if the path logical DCL$PATH is defined - if it is not, then there iscK > no such command, if it is then you'd do an F$SEARCH("DCL$PATH:''p1'.EXE") J > or equivalent in whatever programming language you are using to find outI > what .EXE it would run (you may need logic to check to see if P1 has an K > extender already) and display the result if there is one, otherwise you'd F > need to do the same but for .COM to see if that gives you any resultC > (displaying it if there is on, or reporting that there is no suchM > command if not).  @ The easy way for VMS Development to do that would be by adding a9 mode to the image activator just to support this command.e   ------------------------------    Date: 17 May 2003 16:41:45 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)s& Subject: Re: Error message help needed3 Message-ID: <ESJwFxFc9cOu@eisner.encompasserve.org>e  c In article <31BcMEmTmyRm@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:nY > In article <17MAY200315075650@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl Perkins) writes:e > P >> A roughly equivalent command on VMS would first check to see if the parameterL >> was a symbol, if so it would return the value of the symbol. If not, thenP >> would come the hard part - it would have to check to see if the parameter wasR >> in the process's command tables, if so it would tell you what image the commandN >> tables indicate would be run (in some cases, like when different qualifiersP >> cause different images to be run, this could be difficult to do in a sensibleN >> manner unless you parse the parameter in the same way DCL does to determineM >> what parameters, if any, were specified and what image that works out to).oH >> If it wasn't in the command tables, you would, at long last, check toL >> see if the path logical DCL$PATH is defined - if it is not, then there isL >> no such command, if it is then you'd do an F$SEARCH("DCL$PATH:''p1'.EXE")K >> or equivalent in whatever programming language you are using to find out J >> what .EXE it would run (you may need logic to check to see if P1 has anL >> extender already) and display the result if there is one, otherwise you'dG >> need to do the same but for .COM to see if that gives you any resulttD >> (displaying it if there is on, or reporting that there is no such >> command if not).n > B > The easy way for VMS Development to do that would be by adding a; > mode to the image activator just to support this command.t  B Of course the .COM processing would have to be done by DCL itself.   ------------------------------  % Date: Sat, 17 May 2003 20:55:00 -0500n1 From: "David J. Dachtera" <djesys.nospam@fsi.net> & Subject: Re: Error message help needed' Message-ID: <3EC6E7F4.9D734539@fsi.net>    Michael Rice wrote:t > [snip]J > All well and good for Unix-like environments, but since VMS doesn't haveE > the concept of an executable search path, it doesn't really matter.o   Well, actually, yes it has.)  E The logical name DCL$PATH (which is optional - it can be setup at the E system-wide level (NEVER recommended, IMO), group (less of a bugaboo, F but still only recommended with great caution), job (a little better),1 process (probably most common usage), etc. level.    Example:% DJAS01::DDACHTERA$ sh log dcl$path/fuw6    "DCL$PATH" [super] = "USER_COM:" (LNM$JOB_80D19A80)         = "USER_IMG:"e; 1  "USER_COM" [exec] = "USER_ROOT:[EXE]" (LNM$JOB_80D19A80)sA 1  "USER_IMG" [exec] = "USER_ROOT:[EXE.ALPHA]" (LNM$JOB_80D19A80)l  F There is a *VERY* large caveat associated with usage of this, however.E See http://www.djesys.com/vms/support/dclinter/sld010.htm and the two  slides which follow.  C As other posters have surmised, DCL decides (apparently) whether to D invoke the procedure if a .COM is found before a .EXE, or invoke the1 image activator if a .EXE is found before a .COM.   / The big problem is a malformed command like so:i   $ dirdka0:* A %DCL-W-NOLBLS, label ignored - use only within command proceduresn  \DIRDKA0:\   H ...where upon the first file found matching * is executed. That could beH benign, or inconvenient or it could be catastrophic, depending upon whatG it finds first. In my case, it first finds $.COM which contains nothinga more than an EXIT statement.  4 > However, if it did, 'which authorize' would returnH > 'SYS$SYSROOT:[SYSEXE]AUTHORIZE.EXE', regardless of the current default > directory.  B ...unless DCL$PATH is defined, in which case the behavior would beA different, assuming which.c is programmed to emulate DCL's searchS	 behavior.   H According to the following web page, DCL$PATH (referred to as "Automatic2 Foreign Commands") first appeared in OpenVMS V6.2:  A http://h71000.www7.hp.com/openvms/os/openvms-release-history.htmlt    It's also in the OpenVMS FAQ at:9 http://h71000.www7.hp.com/wizard/faq/vmsfaq_008.html#dcl11   --   David J. Dachtera- dba DJE Systemse http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/S   ------------------------------   Date: 18 May 03 06:37:11 +0200) From: p_sture@elias.decus.ch (Paul Sture)M& Subject: Re: Error message help needed) Message-ID: <JSh2R8uW79j5@elias.decus.ch>d  U In article <00A1FFFD.C5EE27FC@SendSpamHere.ORG>,   VAXman-  @SendSpamHere.ORG writes:sW > In article <uxHIIaquqQJK@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes:D >>In article <rdeininger-1705030714150001@user-uinj476.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes: J >>> In article <CIEJLCMNHNNDLLOOGNJIIEFLHDAA.tom@kednos.com>, "Tom Linden" >>> <tom@kednos.com> wrote:  >>>  >>>>It would be nice to sayi >>>> >>>>$ which hellou >>> 0 >>> Well, I can "say" that to my VMS system now. >>> K >>> You must be thinking of a "which" command from some other environment. n3 >>> Care to explain to the rest of us what it does?s >>G >>From unix land, and if you dont' understand this, I cannot blame you:s >>P >>WHICH(1)                System General Commands Manual                WHICH(1) >> >>NAMEO >>     which - locate a program file including aliases and paths (csh(1)) only)  >>
 >>SYNOPSIS >>     which [name] ...  >>
 >>DESCRIPTIONfN >>     which takes a list of names and looks for the files which would be exe-O >>     cuted had these names been given as commands.  Each argument is expandedaN >>     if it is aliased, and searched for along the user's path.  Both aliases6 >>     and path are taken from the user's .cshrc file. >> >>FILES 2 >>     ~/.cshrc  source of aliases and path values >>
 >>DIAGNOSTICSFN >>     A diagnostic is given for names which are aliased to more than a singleM >>     word, or if an executable file with the argument name was not found ino >>     the path. >> >>BUGSJ >>     Must be executed by a csh(1), or some other shell which knows about >>     aliases.m >>	 >>HISTORYm, >>     The which command appeared in 3.0BSD. >>P >>3rd Berkeley Distribution       April 23, 1991       3rd Berkeley Distribution >>(END)  >>   >> >>-- n >>Paul Sture >  > , > Yeah... that really clear it up for me. %P > 1 thought you'd all like that clear description :-)   D This is what it tells me, in action (stupidly long prompt containing# my system and username edited out):   
 % which logino login: shell built-in command.
 % which su /usr/bin/suy    H Really, incredibly useful. To someone somewhere, sometime, someplace :-)  K I imagine it comes from the mentality that makes folks need to search theirl6 disk hierarchy to find where the hell their files are.  J Me? I prefer something obviously old fashioned, called naming conventions,E but why the hell bother with them when you can boast of a 2GHz PC anddG impress the shit out of a layman by issuing a command which is designed-, to look gibberish to all but the select few?  9 Sorry, *nix fans, but I've got have a go occasionally :-)    -- l
 Paul Sture   ------------------------------    Date: 17 May 2003 15:39:50 -0700' From: danthemanthatjams@yahoo.com (Dan)$C Subject: Re: getting data off an OpenVMS Alpha to Microsoft Windowsk< Message-ID: <a5de6471.0305171439.293d8d3@posting.google.com>  D If the network is not working, you could try Kermit to transfer over your serial connection.k  % Maybe Kermit is installed on the box.y  
 Good Luck,   Dan    ------------------------------    Date: 17 May 2003 12:53:44 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)f; Subject: Re: WHICH command (was: Error message help needed)b3 Message-ID: <KrFNxgpJ4w0U@eisner.encompasserve.org>w  Y In article <vccso1daqtr82@corp.supernews.com>, Michael Rice <marice@whiteice.com> writes:L  D > The 'which' command searches the 'path' (and aliases) to find the I > particular instance of an executable that would be executed.  So, if I  7 > were to type 'which tar' on my SGI, it would respond  K > '/usr/freeware/bin/tar', even though there is also a 'tar' executable in bG > '/sbin/tar'.  The reason being that '/usr/freeware/bin' comes before YH > '/sbin' in my 'path' environment variable.  This can be very handy if ) > you don't know where a program resides.d > K > All well and good for Unix-like environments, but since VMS doesn't have lF > the concept of an executable search path, it doesn't really matter.   # I believe VMS has that, these days.N  5 > However, if it did, 'which authorize' would return .I > 'SYS$SYSROOT:[SYSEXE]AUTHORIZE.EXE', regardless of the current default 0 > directory.  	 And would    	WHICH SHOW    return one answer while_   	WHICH SHOW SECURITY     and_ 	WHICH SHOW SERVER   return other answers ?   Wouldw   	WHICH SHOW STATUS   return the stringh   	SYS$COMMON:[SYSEXE]DCL.EXE-   ?-   ------------------------------  % Date: Sat, 17 May 2003 11:31:06 -0700d# From: "Tom Linden" <tom@kednos.com>A; Subject: RE: WHICH command (was: Error message help needed)t9 Message-ID: <CIEJLCMNHNNDLLOOGNJICEGLHDAA.tom@kednos.com>3   >-----Original Message----->5 >From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net] & >Sent: Saturday, May 17, 2003 10:54 AM >To: Info-VAX@Mvb.Saic.Com< >Subject: Re: WHICH command (was: Error message help needed) >r > < >In article <vccso1daqtr82@corp.supernews.com>, Michael Rice ><marice@whiteice.com> writes: >eD >> The 'which' command searches the 'path' (and aliases) to find theI >> particular instance of an executable that would be executed.  So, if Ie7 >> were to type 'which tar' on my SGI, it would respondgK >> '/usr/freeware/bin/tar', even though there is also a 'tar' executable indG >> '/sbin/tar'.  The reason being that '/usr/freeware/bin' comes before H >> '/sbin' in my 'path' environment variable.  This can be very handy if* >> you don't know where a program resides. >>K >> All well and good for Unix-like environments, but since VMS doesn't havePF >> the concept of an executable search path, it doesn't really matter. >,$ >I believe VMS has that, these days. > 5 >> However, if it did, 'which authorize' would returnnI >> 'SYS$SYSROOT:[SYSEXE]AUTHORIZE.EXE', regardless of the current defaultt
 >> directory., >d
 >And would >  >	WHICH SHOW >b >return one answer while >  >	WHICH SHOW SECURITYy >    and >	WHICH SHOW SERVER   A They should probably return a diagnostic.  The executable is SHOWhA and WHICH is only returning where the shell first finds it on theo specified PATH variable list.h >A >return other answers ?a >o >Would >- >	WHICH SHOW STATUS  >c >return the string >e >	SYS$COMMON:[SYSEXE]DCL.EXE >  >? >X >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com).aA >Version: 6.0.480 / Virus Database: 276 - Release Date: 5/12/2003  >  ---S& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.480 / Virus Database: 276 - Release Date: 5/12/2003   ------------------------------    Date: 17 May 2003 14:21:29 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)0; Subject: RE: WHICH command (was: Error message help needed)w3 Message-ID: <G$8Hkq+GSr+U@eisner.encompasserve.org>P  _ In article <CIEJLCMNHNNDLLOOGNJICEGLHDAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:1 >  >  >>-----Original Message-----6 >>From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]   >>And woulds >>
 >>	WHICH SHOWD >> >>return one answer whilea >> >>	WHICH SHOW SECURITY	 >>    and  >>	WHICH SHOW SERVER > C > They should probably return a diagnostic.  The executable is SHOW C > and WHICH is only returning where the shell first finds it on theh > specified PATH variable list.  >- >>return other answers ?  C In the case of the second and third commands, the executable is not D SHOW, but SETSHOSECUR.EXE and SETSHOSERVER.EXE respectively. NeitherB is found by reference to the PATH variable list, but rather by the DCLTABLES data.a   >>Would2 >> >>	WHICH SHOW STATUS >> >>return the stringe >> >>	SYS$COMMON:[SYSEXE]DCL.EXEs   ------------------------------  % Date: Sat, 17 May 2003 12:55:56 -0700c# From: "Tom Linden" <tom@kednos.com>o; Subject: RE: WHICH command (was: Error message help needed)h9 Message-ID: <CIEJLCMNHNNDLLOOGNJIOEGMHDAA.tom@kednos.com>    >-----Original Message-----t5 >From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]m& >Sent: Saturday, May 17, 2003 12:21 PM >To: Info-VAX@Mvb.Saic.Com< >Subject: RE: WHICH command (was: Error message help needed) >  >a@ >In article <CIEJLCMNHNNDLLOOGNJICEGLHDAA.tom@kednos.com>, "Tom ! >Linden" <tom@kednos.com> writes:t >> a >> t >>>-----Original Message----- 7 >>>From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]t >' >>>And would >>>  >>>	WHICH SHOW >>>u >>>return one answer while >>>C >>>	WHICH SHOW SECURITYe
 >>>    and >>>	WHICH SHOW SERVERn >> tD >> They should probably return a diagnostic.  The executable is SHOWD >> and WHICH is only returning where the shell first finds it on the  >> specified PATH variable list. >> >>>return other answers ?e >.D >In the case of the second and third commands, the executable is notE >SHOW, but SETSHOSECUR.EXE and SETSHOSERVER.EXE respectively. Neither:C >is found by reference to the PATH variable list, but rather by the  >DCLTABLES data.  4 I suspected there was more to it.  Not very pretty.    >  >>>Would >>>p >>>	WHICH SHOW STATUSa >>>t >>>return the string >>>- >>>	SYS$COMMON:[SYSEXE]DCL.EXE >8 >---' >Incoming mail is certified Virus Free.u; >Checked by AVG anti-virus system (http://www.grisoft.com).(A >Version: 6.0.480 / Virus Database: 276 - Release Date: 5/12/2003e >a ---u& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.480 / Virus Database: 276 - Release Date: 5/12/2003   ------------------------------    Date: 17 May 2003 15:18:57 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)o; Subject: RE: WHICH command (was: Error message help needed)g3 Message-ID: <ARL1mKldcKuT@eisner.encompasserve.org>   _ In article <CIEJLCMNHNNDLLOOGNJIOEGMHDAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:g  6 >>From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]  A >>In article <CIEJLCMNHNNDLLOOGNJICEGLHDAA.tom@kednos.com>, "Tom t" >>Linden" <tom@kednos.com> writes:  8 >>>>From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net] >>
 >>>>And wouldu >>>> >>>>	WHICH SHOWd >>>> >>>>return one answer whilei >>>> >>>>	WHICH SHOW SECURITY >>>>    and  >>>>	WHICH SHOW SERVER >>> E >>> They should probably return a diagnostic.  The executable is SHOWgE >>> and WHICH is only returning where the shell first finds it on theS! >>> specified PATH variable list.o >>>  >>>>return other answers ? >>E >>In the case of the second and third commands, the executable is notmF >>SHOW, but SETSHOSECUR.EXE and SETSHOSERVER.EXE respectively. NeitherD >>is found by reference to the PATH variable list, but rather by the >>DCLTABLES data.  > 6 > I suspected there was more to it.  Not very pretty.   D Ah, but it does not have to be pretty; it just has to work properly, and it has done that for years.1  ; The attempt to make all operating systems appear to be Unixs is what is not pretty.   ------------------------------  % Date: Sat, 17 May 2003 14:41:53 -0700e# From: "Tom Linden" <tom@kednos.com>r; Subject: RE: WHICH command (was: Error message help needed)e9 Message-ID: <CIEJLCMNHNNDLLOOGNJIGEGOHDAA.tom@kednos.com>    >-----Original Message----- 5 >From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net])% >Sent: Saturday, May 17, 2003 1:19 PMe >To: Info-VAX@Mvb.Saic.Com< >Subject: RE: WHICH command (was: Error message help needed) >  > ? >In article <CIEJLCMNHNNDLLOOGNJIOEGMHDAA.tom@kednos.com>, "Tom ! >Linden" <tom@kednos.com> writes:0 >f7 >>>From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]i >hA >>>In article <CIEJLCMNHNNDLLOOGNJICEGLHDAA.tom@kednos.com>, "Tomi# >>>Linden" <tom@kednos.com> writes:W >s9 >>>>>From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]i >>>t >>>>>And would >>>>>t >>>>>	WHICH SHOW >>>>>a >>>>>return one answer while >>>>>n >>>>>	WHICH SHOW SECURITYu >>>>>    and >>>>>	WHICH SHOW SERVERw >>>>F >>>> They should probably return a diagnostic.  The executable is SHOWF >>>> and WHICH is only returning where the shell first finds it on the" >>>> specified PATH variable list. >>>> >>>>>return other answers ?r >>>oF >>>In the case of the second and third commands, the executable is notG >>>SHOW, but SETSHOSECUR.EXE and SETSHOSERVER.EXE respectively. NeitheruE >>>is found by reference to the PATH variable list, but rather by the- >>>DCLTABLES data. >>6 >> I suspected there was more to it.  Not very pretty. >oE >Ah, but it does not have to be pretty; it just has to work properly,3  >and it has done that for years. > < >The attempt to make all operating systems appear to be Unix >is what is not pretty.   K Actually, I was thinking more of the grammatical ambiguity and the need forhG special casing in the DCl parser, but then that is not uncommon.  Therey wouldgH seem to be some overlap between the parsing performed by DCL and that by SHOW.sJ My preference would be to minimize interfaces and keep things as simple as* possible. De gustibus non est disputandem.     >t >---' >Incoming mail is certified Virus Free.c; >Checked by AVG anti-virus system (http://www.grisoft.com).iA >Version: 6.0.480 / Virus Database: 276 - Release Date: 5/12/2003  >e ---o& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.480 / Virus Database: 276 - Release Date: 5/12/2003   ------------------------------  % Date: Sat, 17 May 2003 20:59:46 -0500s1 From: "David J. Dachtera" <djesys.nospam@fsi.net> ; Subject: Re: WHICH command (was: Error message help needed)i' Message-ID: <3EC6E912.E86F08C2@fsi.net>d   Larry Kilgallen wrote: > [ > In article <vccso1daqtr82@corp.supernews.com>, Michael Rice <marice@whiteice.com> writes:i > E > > The 'which' command searches the 'path' (and aliases) to find thevJ > > particular instance of an executable that would be executed.  So, if I8 > > were to type 'which tar' on my SGI, it would respondL > > '/usr/freeware/bin/tar', even though there is also a 'tar' executable inH > > '/sbin/tar'.  The reason being that '/usr/freeware/bin' comes beforeI > > '/sbin' in my 'path' environment variable.  This can be very handy ifs+ > > you don't know where a program resides.- > >-L > > All well and good for Unix-like environments, but since VMS doesn't haveG > > the concept of an executable search path, it doesn't really matter.e > % > I believe VMS has that, these days.  > 6 > > However, if it did, 'which authorize' would returnJ > > 'SYS$SYSROOT:[SYSEXE]AUTHORIZE.EXE', regardless of the current default > > directory. >  > And would  >  >         WHICH SHOW >  > return one answer whileo >  >         WHICH SHOW SECURITYv	 >     andt >         WHICH SHOW SERVERf >  > return other answers ? >  > Wouldi >  >         WHICH SHOW STATUSt >  > return the string  > $ >         SYS$COMMON:[SYSEXE]DCL.EXE >  > ?s  @ That would likely require a VMS-only version of "which" that hadD knowledge of DCLTABLES and how to decode it, similar to the way VERB does.n  G Of course, it would still need to support DCL$PATH in cases like ZIP oroF UNZIP which might be found that way. Likewise for any other object notC found in the process's command table that might found via DCL$PATH.u   -- r David J. Dachterag dba DJE Systemsr http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/r   ------------------------------   Date: 18 May 03 07:05:33 +0200) From: p_sture@elias.decus.ch (Paul Sture)r; Subject: Re: WHICH command (was: Error message help needed)e) Message-ID: <OJ6a$$j2RVth@elias.decus.ch>   c In article <KrFNxgpJ4w0U@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: [ > In article <vccso1daqtr82@corp.supernews.com>, Michael Rice <marice@whiteice.com> writes:t > E >> The 'which' command searches the 'path' (and aliases) to find the fJ >> particular instance of an executable that would be executed.  So, if I 8 >> were to type 'which tar' on my SGI, it would respond L >> '/usr/freeware/bin/tar', even though there is also a 'tar' executable in H >> '/sbin/tar'.  The reason being that '/usr/freeware/bin' comes before I >> '/sbin' in my 'path' environment variable.  This can be very handy if p* >> you don't know where a program resides. >> eL >> All well and good for Unix-like environments, but since VMS doesn't have G >> the concept of an executable search path, it doesn't really matter. M > % > I believe VMS has that, these days.m >   D Sorry, but I cannot resist this one. Quite apart from the relativelyA recent DCL$PATH, VMS has had the idea of search paths since V4.0, D which I first saw approximately 18 years ago. Not specifically aimedI at executables, for the nitpickers out there, but search paths indeed :-)m   -- h
 Paul Sture   ------------------------------   End of INFO-VAX 2003.273 ************************