1 INFO-VAX	Thu, 18 May 2000	Volume 2000 : Issue 276       Contents:8 Accessing the assembler from Dec C++ on Alpha Tru64 Unix< Re: Accessing the assembler from Dec C++ on Alpha Tru64 Unix. Re: address munging revisited: the web version Re: big Fortran data file , Re: Compaq Bets Big on New Web Servers (WSJ), Re: Compaq Bets Big on New Web Servers (WSJ) Re: CPU Temperature from DCL Re: CPU Temperature from DCL7 Re: Creating a new user using a system service...HOW??? 7 Re: Creating a new user using a system service...HOW??? 7 Re: Creating a new user using a system service...HOW??? 7 Re: Creating a new user using a system service...HOW??? - RE: crystal clear vms filesystem from nt side - RE: crystal clear vms filesystem from nt side  Re: Debug problem ! Re: DECnet: work on a remote node  Re: DEFINE ? Re: EDT macros2 F$GETQUI: How do I maintain DISPLAY_QUEUE context?6 Re: F$GETQUI: How do I maintain DISPLAY_QUEUE context?= Geting Temperature from CA (was Re: CPU Temperature from DCL)  How To Strip <CR><LF>  Re: How To Strip <CR><LF>  I can no longer backup% Inserting a DATE (was Re: EDT macros)  Re: INSVIRMEM when linking Re: INSVIRMEM when linking Re: INSVIRMEM when linking Linux ==> Free VMS) Linux port question. $QIOW() to sendto(). 2 Re: Linux, Unix to VMS VT220 emulation - solution.# Re: Mapping HSJ Storage controllers # Re: Mapping HSJ Storage controllers  RE: Monitor Display on V7.2-1  RE: Monitor Display on V7.2-1  RE: Monitor Display on V7.2-1 8 Re: Open VMS connectivity problem- Urgent please respond- Re: qio completion time, after process exits? - Re: qio completion time, after process exits? = Re: Question about DFWDAYS and Hoff's presentation on VMS 7.3 0 samba printing problems and mailing list problem4 Re: samba printing problems and mailing list problem# Re: String Descriptors, BASIC and C / Re: turning off system messages on console port " Re: Utility to track device errors( Re: Vaxstation graphics to SVGA monitor?( Re: Vaxstation graphics to SVGA monitor?( Re: Vaxstation graphics to SVGA monitor?( Re: Vaxstation graphics to SVGA monitor?) Re: VMS and Ole for Process Control (OPC) ) RE: VMS and Ole for Process Control (OPC)  VMS File Modes Re: VMS File Modes Re: VMS File Modes Re: VMS File Modes Re: VMS File Modes Re: VMS File Modes Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop?  Re: VMS on the desktop? + Re: Where to start with motif programming ?  Re: Wildfire Announcement  Re: Wildfire Announcement  Re: Wildfire Announcement  Re: Wildfire Announcement  Re: Wildfire Announcement  Re: Wildfire Announcement = Re: Wildfire Announcement: Michael Capellas, can you say VMS? = Re: Wildfire Announcement: Michael Capellas, can you say VMS? = Re: Wildfire Announcement: Michael Capellas, can you say VMS? = Re: Wildfire Announcement: Michael Capellas, can you say VMS? = Re: Wildfire Announcement: Michael Capellas, can you say VMS?  Windows 98 Vs. Windows NT 4.0 . Re: Windows Front End to VT Based Applications. Re: Windows Front End to VT Based Applications  F ----------------------------------------------------------------------  # Date: Wed, 17 May 2000 21:11:37 GMT ' From: Jeffrey Ng <jeffng@dcs.qmw.ac.uk> A Subject: Accessing the assembler from Dec C++ on Alpha Tru64 Unix . Message-ID: <si62o9bq8ta83@corp.supernews.com>  G Does anyone know how to inline assembler instructions in Dec/Compaq C++  for Alpha Tru64 Unix, please?   H In gnu c++ on both Linux and Irix 4, I use the asm("instruction ": : "g"G (parms)) to access optimisation instructions for my scientific code. In H particular, I'm quite fond of the cache control 3DNow instructions on myF Athlon 550. Now, I'm trying to do the same for the dec alpha ev6 chipsE and their huge 512 byte cache lines. However, dec c++ will not accept 5 asm or __asm__. Any help will be greatly appreciated.    Cheers,  Jeffrey Ng.   F Ps: Please, also reply by email, after removing the nospam string from the email address.      ------------------------------  % Date: Wed, 17 May 2000 14:39:40 -0700 - From: "Dann Corbit" <dcorbit@solutionsiq.com> E Subject: Re: Accessing the assembler from Dec C++ on Alpha Tru64 Unix ( Message-ID: <AkEU4.1791$L82.1645@client>  4 "Jeffrey Ng" <jeffng@dcs.qmw.ac.uk> wrote in message( news:si62o9bq8ta83@corp.supernews.com...  G Why did you crosspost to the .x86 group?  Certainly you knew that Alpha G instructions are completely different.  That crosspost was so absurd, I  trimmed it.   J Why did you crosspost to news:comp.lang.c, which addresses the C language?L The DEC C++ compilers won't even compile C.  If you don't believe me, try to
 compile this:    #include <stdlib.h>  #include <string.h>  #include <stdio.h>   int main(void) {     char *c = malloc(20);	    if (c)     {        strcpy(c, "Hello, world");       puts(c);    }    else        puts("Sorry.");     return 0; }   < It won't even compile if you surround it with extern "C" { }   It's a C++ compiler *only*.   I > Does anyone know how to inline assembler instructions in Dec/Compaq C++  > for Alpha Tru64 Unix, please?  > J > In gnu c++ on both Linux and Irix 4, I use the asm("instruction ": : "g"I > (parms)) to access optimisation instructions for my scientific code. In J > particular, I'm quite fond of the cache control 3DNow instructions on myH > Athlon 550. Now, I'm trying to do the same for the dec alpha ev6 chipsG > and their huge 512 byte cache lines. However, dec c++ will not accept 7 > asm or __asm__. Any help will be greatly appreciated.  > 	 > Cheers, 
 > Jeffrey Ng.   I Compaq {formerly DEC} C++ comes with a nice Doc set, last I looked.  Have  you tried looking at it?  K news:comp.unix.ultrix was probably the logical place for your post (were it I not as dead as a doornail).  After all, Tru64 UNIX is the artist formerly  known as Ultrix.  ? Hence, if you have exhausted all other options, I would suggest K news:comp.unix.programmer as an absolute last resort to more sensible forms  of solution.  H > Ps: Please, also reply by email, after removing the nospam string from > the email address.   You post here, you read here.  --0 C-FAQ: http://www.eskimo.com/~scs/C-faq/top.html$  "The C-FAQ Book" ISBN 0-201-84519-91 C.A.P. Newsgroup   http://www.dejanews.com/~c_a_p I C.A.P. FAQ: ftp://38.168.214.175/pub/Chess%20Analysis%20Project%20FAQ.htm    ------------------------------  % Date: Wed, 17 May 2000 19:45:01 +0000  From: SysAdmin <djesys@fsi.net> 7 Subject: Re: address munging revisited: the web version ' Message-ID: <3922F6BD.6CF9B96D@fsi.net>    Wayne Sewell wrote:  [snip]E > It's in the nature of spam that there is absolutely no targeting to O > people who might actually be interested in the product.  They send alsolutely , > everything to every possible mail address.  @ Well, yes and no. Having once worked at the home of the NationalG Consumer Database, I can attest that a great deal of (S/390) processing D power is dedicated to the selection and targetting of mass mailings.H This technology is also being extended to commercial e-mail. At the sameC time, however, just about anyone can run a "spam-bot" on just about B anything connected directly to the internet and attempt to harvestG e-mail addresses, cookies and URLs - sorta like "dumpster diving", only ? the entire internet is being treated as if it were a dumpster.    E > Jake Blues: "Sell me your children!  How much for the little girl?"    I thought that was:   # Jake:	How much for the little girl?   . Restaurant patron - looks shocked and confused  , Jake:	Your women - I want to buy your women.  B Guess it's time to dig out "the boys" and let 'em rock once again!     David J. Dachtera    ------------------------------   Date: 15 May 2000 22:04:19 GMT* From: helbig@astro.rug.nl (Phillip Helbig)" Subject: Re: big Fortran data file. Message-ID: <8fps93$ipa$2@info.service.rug.nl>  ? In article <uezoprjyk1.fsf@altair.dfrc.nasa.gov>, Richard Maine % <maine@altair.dfrc.nasa.gov> writes:    ( Thanks very much for the detailed reply.  F >There are ways around this, but they involve extra work (or money forE >someone else to do the work).  Whether its worth the work depends on 4 >how much disk space you'd be talking about wasting.  I Half (discounting overheads).  The number of values of the last index is  I proportional to the value of the second-to-last index.  Thus, on average  " there are half the maximum number.  J >Sounds like what you need is basically an ISAM (indexed sequential accessF >method) file.  Anyway, that's what they were called back in the '60s.C >(This is not exactly new stuff).  Fortran doesn't provide built-in G >support for such a structure.  You can either throw something together ? >yourself or buy a 3rd party library that supports such things.   3 Again, interesting explanation.  I'll look into it.   I As I'm on a VMS machine, I'm sure there is some VMS file type which fits  B the bill, but I'd like to stick to a standard-conforming solution.   ------------------------------  # Date: Thu, 18 May 2000 03:46:08 GMT - From: young_r@eisner.decus.org (Robert Young) 5 Subject: Re: Compaq Bets Big on New Web Servers (WSJ) + Message-ID: <b62XN4HcS6VT@eisner.decus.org>   I In article <3922BDCD.DFC89F80@fsi.net>, SysAdmin <djesys@fsi.net> writes:  > itjck01@my-deja.com wrote: >>  F >> I think, if anyone is interested in a more accurate line item list, >> E*Trade is getting: >>  H >> GS160                           4 - 12proc./32gb                $5.4MH >> GS140                           3 - 8 proc./16gb                $2.7MH >> DS20                            10                              $726kH >> ES40                            1                               $177kH >> EV67 Boards                     12                              $576kH >> ESA 12000                       8 - 6 TB                        $5.2MH >> ES40 Cab.                       Inc.                            $127k >>  I >>                                                     Total       $14.9M  > @ > Do you happen to know if any of those will be running OpenVMS? >    	Yes.. of course.      	From the Press Release:  3 http://www.digital.com/hpc/news/news_pr160500a.html   L "Josh Levine, Chief Information Officer at E*Trade, said: "When we pioneeredM online investing years ago, we didn't know E*Trade today would be the world's M most visited online investing site. Our partnership with Compaq has helped us N to accommodate this business and infrastructure growth. From the beginning, weC have leveraged enterprise solutions from Compaq  particularly their N AlphaServers and OpenVMS clusters, along with ProLiant NT servers, storage andK mission-critical services  to enable the computing environment necessary to H meet the high demands of our growing business. The AlphaServer GS seriesN provides us with features we can't do without: reliability, performance, speed and scalability."    				Rob    ------------------------------  % Date: Wed, 17 May 2000 23:26:36 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> 5 Subject: Re: Compaq Bets Big on New Web Servers (WSJ) , Message-ID: <392362DD.904F7574@videotron.ca>   Robert Young wrote: 5 > http://www.digital.com/hpc/news/news_pr160500a.html  > N > "Josh Levine, Chief Information Officer at E*Trade, said: "When we pioneeredO > online investing years ago, we didn't know E*Trade today would be the world's O > most visited online investing site. Our partnership with Compaq has helped us P > to accommodate this business and infrastructure growth. From the beginning, weE > have leveraged enterprise solutions from Compaq  particularly their P > AlphaServers and OpenVMS clusters, along with ProLiant NT servers, storage and > mission-critical services     H Why didn't Compaq invite that guy to the wordwide conference, instead ofI inviting the Oracle guy who had absolutely no vested interest in Compaq's 	 success ?    ------------------------------  # Date: Wed, 17 May 2000 20:23:26 GMT 4 From: kaplow_r@eisner.decus.org.spamnot (Bob Kaplow)% Subject: Re: CPU Temperature from DCL + Message-ID: <hnBuwjU7c3c1@eisner.decus.org>   ] In article <8fshed$c7v$1@lead.zk3.dec.com>, "Paul A. Jacobi" <Paul.Jacobi@compaq.com> writes: K > One of the limitation of the GETSYI interface is that it does not provide  > any indicationM > of the sensor location.  This is one of the reason why the interface is not  > fully documented > N > For the ES40, the first three temperature values are from Thermal Zone 0, 1, > and 2,H > which are located somewhere inside the box.  The next two termperature > values are > from the 2 CPU modules.  > H > The exact location of the thermal zones and how the CPU temperature is
 > measuredE > is only available in system specifications which I am not aware are  > available to theG > public.  Also not published is the nominal temperature range for each  > platform.  > L > Temperature sensor placement varies widely on the various Alpha platforms, > so do I > not attempt to compare temperature values from various Alpha platforms.   I Perhaps you can explain something else about the temperature vector. I've F got an 8400 with 8 5/440 CPUs and a GS140 with 8 6/525 CPUs. Each onlyH returns 2 temperatures. They seem to correspond to CPU 3 and 4. WHy only two, and why these two?      MARSsho cpu   MARS, a AlphaServer 8400 5/440E Multiprocessing is ENABLED. Streamlined synchronization image loaded. 0 Minimum multiprocessing revision levels: CPU = 1   PRIMARY CPU = 00	 CPU sets: )    Active         00 01 02 03 04 05 06 07 )    Configure      00 01 02 03 04 05 06 07 )    Potential      00 01 02 03 04 05 06 07 )    Autostart      00 01 02 03 04 05 06 07     Failover       None' MARSsay f$getsyi("temperature_vector")   FFFFFFFFFFFFFFFFFFFFFF1B1AFFFFFF MARS      	Bob Kaplow	  E SPAM:	spamrecycle@ChooseYourmail.com	uce@ftc.gov	postmaster@127.0.0.1    ------------------------------   Date: 17 May 2000 18:49:07 GMT* From: helbig@astro.rug.nl (Phillip Helbig)% Subject: Re: CPU Temperature from DCL . Message-ID: <8fupj3$5v7$3@info.service.rug.nl>  + In article <hnBuwjU7c3c1@eisner.decus.org>, 7 kaplow_r@eisner.decus.org.spamnot (Bob Kaplow) writes:     >MARS, a AlphaServer 8400 5/440 	         ^ I This missing "n" (which is also missing from tcpip sh vers) will forever  ; remind us of the many VAXen which have populated the world.    ------------------------------   Date: 17 May 2000 18:30:36 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)@ Subject: Re: Creating a new user using a system service...HOW???6 Message-ID: <8fuogc$mem$2@mailint03.im.hou.compaq.com>  g In article <01BFC01A.8987BC00.paul.mccabe@cognotec.com>, Paul McCabe <paul.mccabe@cognotec.com> writes: L :Has anyone out there know of a way to create a new user in the UAF using a  :system service.  F   There is no such supported interface, unfortunately -- once the userF   record entry has been added by AUTHORIZE, you can do most everythingD   else via $setuai.  But there is no supported way -- other than viaD   AUTHORIZE -- to initially add a record for a new user into SYSUAF.  M :Our application currently creates users using a spawn to a command file and  L :running authorize.  It has worked perfectly for years.  Now our clients do I :not wish to give this process SYSPRV any more, they only want grpprv on   :these accounts.    A   You could install an image which spawns the procedure, and then C   spawn that.  (I hesitate to recommend the installation of a large A   and complex image with privileges, as that can lead to security @   problems if you are not very careful with the handling of the    image privileges.)  C   I might normally suggest the use of a subsystem identifier on the G   main image and the addition of an associated ACL on SYSUAF, but that  F   mechanism does not currently permit operations via subprocess.  (TheE   identifiers get stripped off when LOGINOUT runs down in the contextM   of the subprocess.)s  9 :We have tried using SETUAI but it doesn't seem to allow + :creation of users.  : A :Any ideas? (safe ones! - downtime of any kind is intolerable!!!)s  F   Please see the other discussion of security issues going on over in E   the "verifying username password" thread.  It's basically the same rF   as this question, though what you are doing is different -- the sameB   techniques mentioned over there can typically be used here, this9   includes the DECnet task-to-task technique I mentioned.g  F   You could also have an AUTHORIZE server process hanging around, and F   send user creation request messages (or even AUTHORIZE commands) to    it via a (protected) mailbox.y  B   I also cover some related material in the DCL book, and provide 0   examples of how to do similar sorts of things.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 17 May 2000 14:35:13 -0500n" From: Earl Lakia <lakia@ipact.com>@ Subject: Re: Creating a new user using a system service...HOW???) Message-ID: <3922F471.74E5108F@ipact.com>u  E You could create your own  user written system service and install ita with the additional privledge.   -earl    Paul McCabe wrote:  L > Has anyone out there know of a way to create a new user in the UAF using a > system service.o >oM > Our application currently creates users using a spawn to a command file and L > running authorize.  It has worked perfectly for years.  Now our clients doI > not wish to give this process SYSPRV any more, they only want grpprv on J > these accounts.  We have tried using SETUAI but it doesn't seem to allow > creation of users. >pB > Any ideas? (safe ones! - downtime of any kind is intolerable!!!) >s > Environment is Alpha 7.2 >l > Thanks a lot,c >r > Paul.    --
 Earl D. Lakiao0 Senior Staff Engineer         Web: www.ipact.com4 Snail Mail:                   Email: lakia@ipact.com
 IPACT Inc.1 260 S. Campbell St.           Phone: 219-464-7212s Valparaiso, IN 46383   ------------------------------  % Date: Wed, 17 May 2000 15:18:19 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> @ Subject: Re: Creating a new user using a system service...HOW???, Message-ID: <3922F078.2809A810@videotron.ca>   Arne Vajh=F8j wrote:< > AFAIK there are no supported way of creating a user from a
 > program.  J Yes. You can write programs that run as part of the all-in-1 (office serv= er)eH image (as a shareable image) and can then make use of the $UAI (ssysuaf)J dataset and can read, write and update fields in records. This is support= ed and documented.s   ------------------------------  % Date: Wed, 17 May 2000 20:51:37 +0000a From: SysAdmin <djesys@fsi.net> @ Subject: Re: Creating a new user using a system service...HOW???' Message-ID: <39230659.C12919D7@fsi.net>    JF Mezei wrote:c >  > Arne Vajhj wrote:> > > AFAIK there are no supported way of creating a user from a > > program. > N > Yes. You can write programs that run as part of the all-in-1 (office server)J > image (as a shareable image) and can then make use of the $UAI (ssysuaf)M > dataset and can read, write and update fields in records. This is supportedu > and documented.   H ...but EXTREMELY expensive!!!! After COBOL and Clustering, Office ServerE (p.k.a. "All-in-1") is one of Compaq most expensive licenses, even at- the per-user level.-   David J. Dachtera-   ------------------------------  % Date: Wed, 17 May 2000 14:14:55 -0400a) From: "Ebinger . Eric" <EEbinger@drc.com>o6 Subject: RE: crystal clear vms filesystem from nt sideB Message-ID: <7162F87E9EF4D311BA9900805FC1D3AE7A6138@and02.drc.com>   > -----Original Message-----@ > From: danny_erres@my-deja.com [mailto:danny_erres@my-deja.com]1 > hello everybody , i will start by saying that it+ > speak the "ls -la" language and not "showe0 > mount /all" (i think it goes like that) . i am   It doesn't.   2 > looking for a tool that will gain me the ability3 > to "see" a vms filesystem from NT and mount it (i . > am going to backup it) . what should i use ?0 > nfs , pathworks , samba or other tools i don't3 > really familiar with . one major thing is that asl1 > i know vms file attributes are very detailed som1 > is there any tool out there that will keep/have-0 > all the atrributes and protection although its
 > NTFS ??? > 3 > i am open for suggestion but keep in mind that myR1 > major need is to backup the vms filesystem withu- > NT tools without damaging the vms data (andn) > metadata) so i can restore it perfectlyr >   @ Sigh.  Would you consider restoring a Solaris (SPARC) filesystem? from NT?  So why would you think that you could do it with VMS?t  B Pathworks and Samba gives access to PORTIONS of VMS filesystems toC NT and support a subset of NTFS features (I don't believe that NTFSeF ACLs are supported).  However, as far as i know, neither product givesE access to the entire VMS filesystem.  Because Pathworks and Samba arecB providing a NTFS compatible view of the VMS files the VMS specificF features of the files (VMS ACLs, record attributes, etc) are hidden.  C Backing up the files from a NT system using NT tools will result in G the loss of information regarding these file attributes.  Restoring theaE files will result in files that may look correct from the NT side buts1 certainly will not be the same from the VMS side.x  C NFS is capable of serving VMS files but it is designed from a Unix nH perspective.  I doubt that any NT backup software will backup NFS based A files and if it did I would expect even more loss of information t7 regarding the file attibutes than with Pathworks/Samba.n  B Note that there is software available from Compaq that will backupC NT files from VMS.  If you are trying to come up with a single sitenF to backup all of your systems, centralizing on VMS is much more likely to be successfull.         Eric Ebinger   ------------------------------  % Date: Wed, 17 May 2000 16:28:48 -0400x# From: John Vottero <John@MVPSI.com>e6 Subject: RE: crystal clear vms filesystem from nt sideD Message-ID: <C15945A9D9EFCF11BA8B08002BBF1CCC0CD6EE@berry.mvpsi.com>  K Use VMS BACKUP to create a disk save set then ZIP to compress that save setoL then FTP the zipped save set to the NT system.  If you trust ZIP to maintainL VMS attributes, you could eliminate the BACKUP step but I wouldn't recommendL it.  If you have a fast network link you can eliminate the ZIP step but thenI you have to know how to repair BACKUP save sets which have been hacked byg FTP (that's not too hard).   > -----Original Message-----@ > From: danny_erres@my-deja.com [mailto:danny_erres@my-deja.com]' > Sent: Wednesday, May 17, 2000 1:09 PMa > To: Info-VAX@mvb.saic.com 4 > Subject: crystal clear vms filesystem from nt side >  > 1 > hello everybody , i will start by saying that i9+ > speak the "ls -la" language and not "show 0 > mount /all" (i think it goes like that) . i am2 > looking for a tool that will gain me the ability3 > to "see" a vms filesystem from NT and mount it (i . > am going to backup it) . what should i use ?0 > nfs , pathworks , samba or other tools i don't3 > really familiar with . one major thing is that asg1 > i know vms file attributes are very detailed sod1 > is there any tool out there that will keep/have 0 > all the atrributes and protection although its
 > NTFS ??? > 3 > i am open for suggestion but keep in mind that myp1 > major need is to backup the vms filesystem with - > NT tools without damaging the vms data (anda) > metadata) so i can restore it perfectlym >  > 	 > Regards  >  > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.  >    ------------------------------  # Date: Wed, 17 May 2000 19:30:33 GMTh2 From: robert_jm_barron@hotmail.com (Robert Barron) Subject: Re: Debug problem5 Message-ID: <3922f2fd.15653208@news.netvision.net.il>    >:K >: Here is my problem: when I try to debug the code the debugger of OpenVMSgL >: says that it doesn't find the source code: I'm executing the program from4 >: the directory where thera are the C and SC files. >lA >Advise the debugger to use the latest versions in your directory@ >instead by entering:  >a1 >  DBG> SET SOURCE drive:[your.directory] /LATESTr >aF I had a simliar problem with the 7.2 debugger not finding some sourcesA in character cell mode - moded to graphical mode and it found thex sources.  D BTW - the 7.x graphical debugger is MUCH better then the 6.x version   Robert.    ------------------------------    Date: 17 May 2000 23:58:00 -04004 From: "Robert Deininger" <rdeininger@mindspring.com>* Subject: Re: DECnet: work on a remote node* Message-ID: <B548E28A-1CCB7@165.247.46.24>  K On Tue, May 16, 2000 11:55 AM, Martin Vorlaender <martin@RADIOGAGA.HARZ.DE>I wrote:, >Phillip Helbig (helbig@astro.rug.nl) wrote:+ >: What does one need to do DECnet over IP?0 > % >The same TCP/IP stack on both sides.r  A No, these days it will even work between different tcp/ip stacks.   H You need DECnet plus, and one of the tcp/ip packages with the PWIPdriverD supported and enabled.  It works with multinet and ucx, I don't know about the others.o     ---------------------------s Robert Deininger rdeininger@mindspring.comu   ------------------------------  % Date: Wed, 17 May 2000 15:12:09 -0400-- From: JF Mezei <jfmezei.spamnot@videotron.ca>  Subject: Re: DEFINE ?n, Message-ID: <3922EF07.6CE23993@videotron.ca>   Dave Weatherall wrote:H > Occasionally, you do get the odd person who loves to RENAME everythingC > back to version ;1 just because 'it looks tidier', not knowing orcG > forgetting that various amounts of useful tracability information getr > lost, or become misleading.E  L There are many valid reasons to rename everything to ;1 (but you must ensure% to do a purge beforehand, of course).   L 1- Once source code in a directory is "frozen", you can rename everything to7 ;1 so that if any changes are made, it becomes obvious.   M 2- Log files, log files, log files. The main problems with VMS staying up for H so long is that pesky limit of 32764 in the version number. Most seriousL packages will act on this when a version limit becomes close to the limit by) renaming existing versions starting a1 ;1    ------------------------------  % Date: Mon, 15 May 2000 19:19:06 -0400 * From: David A Froble <davef@tsoft-inc.com> Subject: Re: EDT macrosn- Message-ID: <392085EA.2F766D73@tsoft-inc.com>o   Phillip Helbig wrote:t > H > OK, I STILL haven't gotten around to using TPU.  It's still on my listJ > of things to do, honest.  (My main reason for this is to be able to makeI > use of LSE, but folks keep saying it will have other benefits as well.)n > E > I've only found two real disadvantages with EDT---the 255 charactereH > limit, and the inability to save the file one is working on.  After myJ > .sig is a fix for the second problem---just use the .COM file instead of@ > EDT directly.  (The lack of an "undo" feature might be a thirdG > disadvantage; I'm still looking into something which manipulates .JOUt	 > files!)e > F > I have lots of macros, using no keypad mode (I think that's the one;I > INSERT; and stuff like that).  However, most of these are variations oncH > a theme and I really don't know that much about them.  I think in lineJ > mode---the stuff one can type in at the prompt.  If I have a sequence ofB > commands which will do the right thing in a .EDT file, how can I@ > transform these into a macro which I can call from within EDT?  O I think what you are asking is how to set up and use a macro while in characteroP (screen) mode.  One method is to first load the commands into alternate buffers,> and then define the macros as the buffer name. For an example:   $ t edtini.edt INCLUDE EDTCMD.CMD =M1 INCLUDE EDTCMD2.CMD =M2  INCLUDE M3.CMD =M3 INCLUDE M4.CMD =M4 INCLUDE M5.CMD =M5 INCLUDE M6.CMD =M6 INCLUDE M7.CMD =M7 DEFINE MACRO M1  DEFINE MACRO M2i DEFINE MACRO M3  DEFINE MACRO M4O DEFINE MACRO M5a DEFINE MACRO M6t DEFINE MACRO M7a =MAIN 0  $ t edtcmd.cmd SUB/DATE$(0%)/DATEL(D9%,4%)/%WH   SUB/DATE$(D9%)/DATEL(D9%,4%)/%WH SUB/CVT$$(/EDIT$(/%WHn SUB/1.00/1.02/%WHu SUB/31-Dec-99/31-Dec-2035/%WHi $r  N The files being included into alternative buffers contain EDT commands, as theN second example shows.  To use a macro, for example macro M3 defined above, theJ sequence is <PF1><7>M3<enter> where the 7 and enter are on the keypad.  AnO astute VAX Basic programmer would guess the example above was used in Y2K work.@   Dave   -- t4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596; 170 Grimplin Road               E-Mail: davef@tsoft-inc.coma Vanderbilt, PA  15486u   ------------------------------  % Date: Wed, 17 May 2000 23:21:17 -0400r* From: "Kevin P. Inscoe" <kevin@inscoe.org>; Subject: F$GETQUI: How do I maintain DISPLAY_QUEUE context?u* Message-ID: <392361AD.B08FAE6E@inscoe.org>  
 $  set ver@ $! This command procedure shows all queues and the jobs in them. $  TEMP = F$GETQUI("") $QLOOP:p5 $  QNAME = F$GETQUI("DISPLAY_QUEUE","QUEUE_NAME","*")t# $  IF QNAME .EQS. "" THEN goto exitt $  qbatch =m> F$GETQUI("DISPLAY_QUEUE","queue_batch",qname,"freeze_context") $ if qbatch .eqs. "TRUE"   $ then $       batch = 1, $       prt = 0  $ else $ endife $  WRITE SYS$OUTPUT ""  $ $  WRITE SYS$OUTPUT "QUEUE: ", QNAME QUEUE: DSN$BATCH $JLOOP:e9 $  jname = F$GETQUI("DISPLAY_JOB","JOB_name",,"all_jobs")4! %JBC-E-NOQUECTX, no queue contextU  = Why did the "FREEZE_CONTEXT" not work for me in line 9 above?C9 I want to determine batch entries from print entries as Io' have different items to check for each.u   TIAe   ~kevin -- D: Kevin P. Inscoe   Senior Unix System Engineer & Specialist: Deltona, FL       Itinerary at http://www.inscoe.org/where) e-mail:  kevin [at] inscoe [dot] org      @ http://www.inscoe.org                    28.9492N 81.1955W      : http://www.inscoe.org/comp     http://www.inscoe.org/radio: "No husband has ever been shot while doing the dishes" Unk   ------------------------------  % Date: Thu, 18 May 2000 12:30:33 +0800o- From: David B Sneddon <dbsneddon@bigpond.com>f? Subject: Re: F$GETQUI: How do I maintain DISPLAY_QUEUE context? + Message-ID: <392371E9.F4C9B22E@bigpond.com>a   "Kevin P. Inscoe" wrote: >  > $  set verB > $! This command procedure shows all queues and the jobs in them. > $  TEMP = F$GETQUI("")	 > $QLOOP:.7 > $  QNAME = F$GETQUI("DISPLAY_QUEUE","QUEUE_NAME","*")o  0 Don't you need the "WILDCARD" flag on this call?? $ qname = f$getqui("DISPLAY_QUEUE","QUEUE_NAME","*","WILDCARD")e  % > $  IF QNAME .EQS. "" THEN goto exit 
 > $  qbatch =e@ > F$GETQUI("DISPLAY_QUEUE","queue_batch",qname,"freeze_context")  1 and should this one be "WILDCARD,FREEZE_CONTEXT"?c   > $ if qbatch .eqs. "TRUE" > $ then > $       batch = 1g > $       prt = 0. > $ else	 > $ endif8 > $  WRITE SYS$OUTPUT "" > & > $  WRITE SYS$OUTPUT "QUEUE: ", QNAME > QUEUE: DSN$BATCH	 > $JLOOP:P; > $  jname = F$GETQUI("DISPLAY_JOB","JOB_name",,"all_jobs")u# > %JBC-E-NOQUECTX, no queue context  > ? > Why did the "FREEZE_CONTEXT" not work for me in line 9 above?o; > I want to determine batch entries from print entries as Iy) > have different items to check for each.s >  > TIAg >  > ~kevin > --< > Kevin P. Inscoe   Senior Unix System Engineer & Specialist< > Deltona, FL       Itinerary at http://www.inscoe.org/where& > e-mail:  kevin [at] inscoe [dot] org< > http://www.inscoe.org                    28.9492N 81.1955W< > http://www.inscoe.org/comp     http://www.inscoe.org/radio< > "No husband has ever been shot while doing the dishes" Unk     -- t Regards, Dave. I -------------------------------------------------------------------------wI David B Sneddon (dbs)  OpenVMS Systems Programmer   dbsneddon@bigpond.comeI DBS software at ...   http://www.users.bigpond.com/dbsneddon/software.htmrI "Life is what happens to you while you're busy making other plans" Lennoni   ------------------------------  % Date: Wed, 17 May 2000 17:37:08 -0400e- From: "Peter Weaver" <peter.weaver@stelco.ca>FF Subject: Geting Temperature from CA (was Re: CPU Temperature from DCL). Message-ID: <si648b628ta12@corp.supernews.com>  B Paul A. Jacobi wrote in message <8fshed$c7v$1@lead.zk3.dec.com>... >...F >Platform specific knowledge about the various sensors are included in they >CompaqsE >Analyze tool which can make a determination if the sensor values aree beyond >ans >acceptable range. >...  B How do you get this information from Compaq Analyze? I have playedE with CA as much as possible and I do not recall seeing anything aboute0 temperature in any output that I managed to get.  E We have two GS160's on order, will DIAGNOSE be supported on the GS160c or will we have to use CA?   ------------------------------  # Date: Wed, 17 May 2000 17:59:33 GMTd) From: "J.G. Peters" <jpeters@sctcorp.com>T Subject: How To Strip <CR><LF>0 Message-ID: <01bfc029$b414fe10$32341895@jpeters>  6 This is a continuation of the "Timeout Issues" thread., (By the way, thanks to those who responded.)  7 Nutshell: VPN software timing out telnet connections to-9 AlphaServer 2100 (OpenVMS 7.1/Multinet 4.2) at firewall.    2 Attempting to use $BRKTHRUW in a C program to send< a null or bell to a terminal in order to simulate "traffic."1 It works, but how do I suppress the <CR><LF> thate results in the line advancing?   Thanks, Joe-   ------------------------------   Date: 17 May 2000 19:06:12 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)" Subject: Re: How To Strip <CR><LF>6 Message-ID: <8fuqj4$nbb$1@mailint03.im.hou.compaq.com>  \ In article <01bfc029$b414fe10$32341895@jpeters>, "J.G. Peters" <jpeters@sctcorp.com> writes:3 :Attempting to use $BRKTHRUW in a C program to sendu= :a null or bell to a terminal in order to simulate "traffic."l2 :It works, but how do I suppress the <CR><LF> that :results in the line advancing?      carcon argument?  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Wed, 17 May 2000 21:39:48 GMTi* From: j-sivier@uiuc.edu (Jonathan Sivier ) Subject: I can no longer backup 7 Message-ID: <EkEU4.17149$nb2.349445@vixen.cso.uiuc.edu>   J    I've run into a frustrating problem.  I can no longer make tape backupsF on my VAX.  Everything was working fine until the end of March when weD had a disk crash.  During the repair process the tape drive was alsoF replaced.  Now whenever I try to do a backup a get an error saying theC expiration date of the tape is in the future and so it won't let meiD write to the tape.  This happens with new (never used) tapes as wellC as old ones.  It happens if I initialize the tape, with and withoutgH the /override=expiration argument and whether I put /override=expirationL in the mount command or not.  If I set the year to 1999 I can do the backup.I I suppose this could be a Y2K problem, though if that is the case why did I it work OK until the end of March.  The backup tape I restored to the new G disk from has the date 30-MAR-2000 on it's backup set.  Could this be aoH hardware problem caused by the new tape drive?  Why can't I override theI expiration date?  Why does this happen to brand-new tapes that have neverrD been written to?  Has anyone else ever run into a problem like this?      Thanks for any help.o   Jonathan  8 --------------------------------------------------------8 |  Jonathan Sivier      |  Flight Simulation Lab       |8 |  j-sivier@uiuc.edu    |  Beckman Institute           |8 |  Work: 217/244-1923   |  405 N. Mathews              |8 |  Home: 217/359-8225   |  Urbana, IL  61801           |8 --------------------------------------------------------8 |  Home page URL: http://www.uiuc.edu/ph/www/j-sivier  |8 --------------------------------------------------------   ------------------------------  % Date: Wed, 17 May 2000 18:05:03 -0400o- From: "Peter Weaver" <peter.weaver@stelco.ca>n. Subject: Inserting a DATE (was Re: EDT macros). Message-ID: <si65spak8ta75@corp.supernews.com>    Terry Marosites wrote in messageD <1137A4A23A51D311B2D600105A1D5213019AEE1A@seantexch.unitedad.com>...  B >Does anyone know how to define a gold key  C as  that will insert some aB >line line of text and the date and time  ( example: $! Changed by Terry on, >16-MAY-2000 14:40:08) .This has me stumped.    E Here is an EDT line that will define GOLD D to insert the date with a   two digit year (i.e. 17-MAY-00);  F define key gold d as "(i..^zdate-'..'CUTSR+'-'''+C2(D+C)2(+C)9(D+C))."  % Here is the GOLD C key you asked for;s  7 define key gold c as "(i$! Changed by Terry on^ZDate)."n  E I know I wrote the first line at a site running V4.7, I do not recallsD if the DATE command was available in prior versions of EDT. It works fine on AXP V7.2.i   -- Peter Weaver   ------------------------------   Date: 17 May 2000 18:53:47 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)# Subject: Re: INSVIRMEM when linkingo6 Message-ID: <8fuprr$n00$1@mailint03.im.hou.compaq.com>  t In article <8fs3v1$2bra$1@s2.feed.news.oleane.net>, "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr> writes:& :Ben on n'est pas sorti de l'auberge !  I   BabelFish is certainly not familiar with that particular French phrase.    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 17 May 2000 19:56:05 +0000r From: SysAdmin <djesys@fsi.net>e# Subject: Re: INSVIRMEM when linkingi' Message-ID: <3922F955.570639A9@fsi.net>    "Jean-Franois Marchal" wrote: [snip]F > The last image successfully linked was more than 50.000 blocks large  
 YIKE ! ! ! !    ) ...and I thought Netscape on VMS was bad!t  7 ...not to mention the DEC-C compiler (11344 blocks) ...t     David J. DachteraS   ------------------------------  % Date: Thu, 18 May 2000 08:20:17 +0800p- From: David B Sneddon <dbsneddon@bigpond.com> # Subject: Re: INSVIRMEM when linking + Message-ID: <39233741.E36FC435@bigpond.com>m   SysAdmin wrote:u >=20" > "Jean-Fran=E7ois Marchal" wrote: > [snip]F > > The last image successfully linked was more than 50.000 blocks la= rge< >=20 > YIKE ! ! ! ! >=20+ > ...and I thought Netscape on VMS was bad!  >=209 > ...not to mention the DEC-C compiler (11344 blocks) ...o >=20 > David J. Dachteraa   That's nothing...e  ' OLGA_OLC.EXE;1            191832/191844t   --=20  Regards, Dave.TF ---------------------------------------------------------------------= ----F David B Sneddon (dbs)  OpenVMS Systems Programmer   dbsneddon@bigpond= .comF DBS software at ...   http://www.users.bigpond.com/dbsneddon/software= .htmF "Life is what happens to you while you're busy making other plans" Le= nnon   ------------------------------  % Date: Wed, 17 May 2000 23:29:34 -0700d# From: Mark Tarka <markZERO@mcn.net>2 Subject: Linux ==> Free VMSm# Message-ID: <39238DCE.5108@mcn.net>r  , There was some guy at Cal Tech ranting about- a free VMS project, IIRC (not Vance...someonea	 saner :-)l   Is this an option (free VMS)?n  % I ask, because I'm taking a hard lookm" at Red Hat Linux right now (gotta   find a big enough hard drive :-)  " The box is an old DECstation 425c," a 486DX at a blistering 25 MHz....  ! Doesn't seem reasonable that it'd  run VMS...anyone?n    %   mark(0@mcn.net)  Dele the parens...g   ------------------------------  % Date: Wed, 17 May 2000 20:16:03 +0100e# From: Mark <Mark@mxnet.demon.co.uk>v2 Subject: Linux port question. $QIOW() to sendto().0 Message-ID: <3922EFF3.1FE2F9F@mxnet.demon.co.uk>   Hi,mF     I've been working on porting an network traffic generator app fromE an Alpha to Linux. The fundamental difference is that the Alpha (OpeniE VMS 7.1-1H1) uses a $QIO() to sned the ethernet packet to the driver.w5 Where as the Linux system uses the sendto() function.-  H Basically what I do is create and ethernet frame from scratch (ThereforeG in Linux I'm using SOCK_RAW). With the source and destination addresses2G etc. All of which is transmitted correctly (Using a Network analyser to E check the frames). However at about  1000 frames per second there are' constent collsions around 50%.  H I'm using a test LAN, therefore the is no other traffic on it apart fromH the Generator and the system under test. The test system just copies theF frames and sends it to the LAN. Therefore what ever the test generator generates this is doubled.  G Each frame is 146 bytes long. Which works out at about 22% (1000 framesoF per sec) band width used by the test gen and 22% from the system under6 test. Total 44% total bandwidth usaed on a 10Mbps LAN.  @ My question is: When I use the Alpha implementation there are noC collosion problems. It only occurs using Linux. Both boxes have DEC E tulip PCI cards. Does anyone have any suggestions as to what might be@F causing this? Does $QIO do anything different. I can't see why because it's really up to the card.:   Many thanksi   Mark   ------------------------------  # Date: Wed, 17 May 2000 19:19:25 GMT + From: "T.E.Dickey" <dickey@shell.clark.net> ; Subject: Re: Linux, Unix to VMS VT220 emulation - solution. < Message-ID: <1hCU4.26273$0o4.284421@iad-read.news.verio.net>  0 Erik Ahlefeldt <oahlefel@metz.une.edu.au> wrote:J >  There have been a few requests in this group for a good VT220 terminal L >  emulation when connecting to a VMS host from Linux or Unix. Having never J >  found a satisfactory emulation myself, I have taken and comprehensively  9 The XFree86 xterm supports ANSI color and VT220 emulationD There's an faq at3+ 	http://dickey.his.com/xterm/xterm.faq.htmlc 	ftp://dickey.his.com/xtermd   --  $ Thomas E. Dickey  <dickey@clark.net> http://dickey.his.coma ftp://dickey.his.com   ------------------------------  % Date: Wed, 17 May 2000 20:10:11 +0000n From: SysAdmin <djesys@fsi.net> , Subject: Re: Mapping HSJ Storage controllers' Message-ID: <3922FCA3.65777FC1@fsi.net>n   "Moore, John D" wrote: > 	 > Hi all,) > D > Don't mean to be a pain my first day on the list but I'm wondering@ > if anyone's come up with a good automated solution for mappingF > HSJ storage arrays ...  I've come up with one that works fairly well > (check it out here:t > * >    http://www.angelfire.com/oh4/longrun/  F One suggestion: use ZIP and UNZIP for OpenVMS instead of tar-balls forF OpenVMS files. .ZIP archives can be unpacked on OpenVMS, and - if madeB non-VMS specific, on UN*X, W/9x, W/3x, etc. "tar" and "gunzip" for: non-UN*X platforms are rather the exception than the norm.   David J. Dachtera3   Related Links:  * http://home.earthlink.net/~djesys/zip.html, http://home.earthlink.net/~djesys/unzip.html  5 (The InfoZip home page link on those pages is stale.)M   ------------------------------  % Date: Wed, 17 May 2000 14:13:53 -0700s From: James.F.Duff@FHS.com, Subject: Re: Mapping HSJ Storage controllersC Message-ID: <OF17DB1B88.2C3490D6-ON882568E2.00737014@HEALTHNET.COM>i   john_d_moore@reyrey.com writes:g >[snip]e9 >I've got 10 controllers and I don't feel like typing all > >the SHOW commands every time to trap the output.  This works,> >even if the coding is a little rough.  I'm disappointed there? >haven't been more takers ... guess everyone likes doing thingsf >the hard way.  @ I have 86 HSJs / HSZs, not counting the ones connected to the NTB servers.  Although I already have a parse that generates graphical: layouts, believe me, I grabbed a copy of your routines :-)   Jim. -- James.F.Duff@healthnet.com Pure personal opinionc   ------------------------------  % Date: Wed, 17 May 2000 15:19:14 -0700 1 From: "Farrell, Michael" <MFarrell@voltdelta.com>w& Subject: RE: Monitor Display on V7.2-1A Message-ID: <7DF45F22D904D31192EE00805F578DF207EED8@NY_EXCHANGE1>d  H My system manager tells me that all ECOs for V7.2 are applied, but he'llF double check tomorrow morning.  We do NOT have any fibre channel here.I Everything is clustered on either Fast-Wide SCSI or Fast Wide Ultra SCSI.NL We have two ES40s in the SCSI cluster, and an old DEC3000-600 that is LAVC'd
 to the ES40s.   H As I've said, this problem was noticed as far as back as V7.1.  I hadn'tI seen it for awhile, wasn't really looking, but it cropped up again today.A   Mike Farrell mfarrell@voltdelta.com   > -----Original Message-----6 > From:	itjck01@my-deja.com [SMTP:itjck01@my-deja.com]' > Sent:	Wednesday, May 17, 2000 1:22 PMn > To:	Info-VAX@MVB.SAIC.COMm( > Subject:	Re: Monitor Display on V7.2-1 > C > In article <7DF45F22D904D31192EE00805F578DF207EECA@NY_EXCHANGE1>,36 >   "Farrell, Michael" <MFarrell@voltdelta.com> wrote:I > > Below is a display from MONITOR.  Note that the Total Direct I/O ratekG > > is "9", but for the highest performing individual job on the system  > > the rate is "87"., > > H > > This system is running OpenVMS V7.2-1.  I have seen this on V7.1 and# > > higher rev levels sporadically.  > >-I > > Is this correct?  Am I misinterpreting the display.  Can someone shedK > > some light on this?s > >r
 > > Thanks > >T > > Mike Farrell > > mfarrell@voltdelta.com > >s > .D > <snip> > .F > C > Do you have any ECO/patches/updates applied to your OpenVMS 7.2-1s5 > system?  Are you using fibre attached disk storage?p > F > The reason I ask is, I believe one of the recent ECOs addressed this0 > problem, at least with fibre disk connections. >  > :) jck >  > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.    ------------------------------  # Date: Wed, 17 May 2000 23:11:44 GMTU From: itjck01@my-deja.com & Subject: RE: Monitor Display on V7.2-1) Message-ID: <8fv8v4$b05$1@nnrp1.deja.com>B  A Hmmmm... well, I have no fix or further ideas for the short term.h  E I know from prior discussion with Compaq folks, that MONITOR is being A looked at to be upgraded.  Perhaps if you open a service call andc3 document the problem/effect, that it may get fixed.    :) jck  A In article <7DF45F22D904D31192EE00805F578DF207EED8@NY_EXCHANGE1>,:4   "Farrell, Michael" <MFarrell@voltdelta.com> wrote:D > My system manager tells me that all ECOs for V7.2 are applied, but he'll H > double check tomorrow morning.  We do NOT have any fibre channel here.E > Everything is clustered on either Fast-Wide SCSI or Fast Wide Ultra- SCSI.-G > We have two ES40s in the SCSI cluster, and an old DEC3000-600 that isd LAVC'd > to the ES40s.: >:C > As I've said, this problem was noticed as far as back as V7.1.  I2 hadn'tD > seen it for awhile, wasn't really looking, but it cropped up again today. >m > Mike Farrell > mfarrell@voltdelta.com >n    & Sent via Deja.com http://www.deja.com/ Before you buy.G   ------------------------------  % Date: Wed, 17 May 2000 16:50:30 -0700l From: James.F.Duff@FHS.com& Subject: RE: Monitor Display on V7.2-1C Message-ID: <OF60D7F7AB.EC57C500-ON882568E2.00823671@HEALTHNET.COM>    itjck01@my-deja.com writes:"B >Hmmmm... well, I have no fix or further ideas for the short term. >DF >I know from prior discussion with Compaq folks, that MONITOR is beingB >looked at to be upgraded.  Perhaps if you open a service call and4 >document the problem/effect, that it may get fixed. >U >:) jckB >H [snip]  < If I understand the problem SYS$SHARE:XTI$XTILIB.EXE was not: installed as a known image after a reboot?  I don't have a4 vanilla 7.2 system but am looking at a 7.2-1 system. Should be the same.a  = Check SYS$MANAGER:VMS$IMAGES_MASTER.DAT.  The line to installh: the image should be in this file.  This is the master file6 that AUTOGEN uses to create SYS$MANAGER:VMSIMAGES.DAT.  5 Ensure the following line appears in the master file:     8 sys$share:xti$xtilib.exe/open/header/shared=address_data    + and run AUTOGEN though the genparams phase.w   This should fix your problem.n  H It may be worth while to extract the VMS$IMAGES_MASTER.DAT file from theE VMS installation kit and compare it to the one on your system in case- anything else is missing.   Regards, Jim. -- James.F.Duff@healthnet.com Pure personal opinionn   ------------------------------   Date: 15 May 2000 19:58:44 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)A Subject: Re: Open VMS connectivity problem- Urgent please respond-6 Message-ID: <8fpktk$jfo$1@mailint03.im.hou.compaq.com>  d In article <000a01bfbbff$2396c040$c78241ca@pol.pol.net.in>, Champs Onweb <champs@pol.net.in> writes:- :This is a multi-part message in MIME format.w  F   Please turn off MIME.  (Your entire message is being posted twice!) J   As I mentioned in an earlier reply, please pick a subject line relevent H   to the question posed.  Please omit the word "urgent", "please", etc.,H   from the title as it occupies characters that you could use to better I   identify the problem.  (If this is/was truely an urgent question, then pD   you will strongly want to consider direct contact with the Compaq I   Customer Support Center or some other formal or local/official product SG   support organization -- directly.)  Thanks!  (No offense is intended.-D   I am trying to help you learn how to obtain the answer(s) to your G   question(s) in the most expedious and appropriate fashion available.)D  I :The only difference with the application running on the Alpha versus theoH :VAX is how the executables are linked. The Alpha box requires somethingE :called SYMBOL VECTORS and different PSECT Attributes when linking...p :MG :So the question is, where does the extra connections come from and whyp0 :is doesn't occur on the VAX, but only on Alpha?  I   Please see the OpenVMS FAQ for information and a pointer to the OpenVMSeG   shareable image cookbook -- the cookbook (and the OpenVMS LINKER and -H   programming concepts manuals) include details on the requested OpenVMSJ   linking information. The Cookbook is probably the easiest spot to start,J   and I have a problem report and various updates to the antique C syntax I   used in the early chapters of the LINKER manual queued for some future eE   OpenVMS documentation release -- Ask The Wizard topic 4308 (when itdJ   eventually appears out on the website) will likely be relevent here, as A   that was a report of the problem with the LINKER documentation.-  E   As you do not indicate what database is in use in this application, E   the Oracle Rdb SQL precompiler requires some PSECT declarations in cH   a LINKER options file, as well, as the C compiler and Oracle Rdb have H   differing defaults.  I have found the following line -- when added to G   the LINKER options file -- quite helpful in resolving this particularI   "difference of opinion":  J PSECT=RDB$MESSAGE_VECTOR,QUAD,PIC,USR,OVR,REL,GBL,NOSHR,NOEXE,RD,WRT,NOVEC  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 17 May 2000 15:29:27 -0400w# From: Jim Agnew <agnew@hsc.vcu.edu>s6 Subject: Re: qio completion time, after process exits?+ Message-ID: <3922F317.45271322@hsc.vcu.edu>   ( looks like i was wrong, but i learned!!!   J.   Jim Agnew wrote: > 3 > Good question... I'd be interested in this also..t > < > my uneducated guess, knowing what little i know of VMS is: > + > (drum roll)   it will exit...  i think...t > D > no time to do an example a la Carl L., so that's my final answer.. >  > Jima >  > David Mathog wrote:t > >c, > > I couldn't find the answer for this one. > > L > > If a process issues a QIO for a write to disk (for instance) and doesn'tM > > wait for it to complete, can the process exit _before_ the QIO completes?sI > > Or will the process go into LEF or RWAST until all QIOs have cleared?n > >- > > Thanks,- > >- > > David Mathog! > > mathog@seqaxp.bio.caltech.edurB > > Manager, sequence analysis facility, biology division, Caltech   ------------------------------  % Date: Wed, 17 May 2000 15:45:43 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>I6 Subject: Re: qio completion time, after process exits?, Message-ID: <3922F6E3.FB36459E@videotron.ca>   David Mathog wrote:j > * > I couldn't find the answer for this one. > J > If a process issues a QIO for a write to disk (for instance) and doesn'tK > wait for it to complete, can the process exit _before_ the QIO completes?sG > Or will the process go into LEF or RWAST until all QIOs have cleared?n  M When you exit a process, I beleive that the image rundown does a $DEASSGN for9 all opened channels.  ) Now, the $DEASSGN systemn service states:   M "When you deassign a channel, any outstanding I/O requests on the channel areiJ cancelled. If a file is open on the specified channel, the file is closed. ...eP I/O channels assigned from user mode are automatically deassigned at image exit.     Now, for $CANCEL:AK "in general, this includes all I/O requests that are queued, as well as the) request currently in progress".a   further:N "If the I/O operation is a virtual I/O operation involving a disk or tape ACP,C the I/O cannot be cancelled, In the case of magnetic tape, however,m. cancellation may occur if the driver is hung."  E "Outstanding I/O requests are automatically cancelled at image exit".-  J So, my conclusion is that if the driver does not support $CANCEL, then theJ process will wait until the IO is complete (probably in one of those pesky$ RW___ state) until the io completes.  N That is what happens when you attempt to STOP a process that is doing a DIR onN a mounted tape. It will wait until whatever it had requested from the tape ACP% to complete before actually stopping.a   ------------------------------   Date: 17 May 2000 17:58:09 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)F Subject: Re: Question about DFWDAYS and Hoff's presentation on VMS 7.36 Message-ID: <8fumjh$lkt$1@mailint03.im.hou.compaq.com>  s In article <Pine.OSF.4.21.0005161458140.32400-100000@athena.csdco.com>, John Nebel <nebel@athena.csdco.com> writes:9 :f) :The OpenVMS V7.3 update is labeled "NDA"  :i2 :What has to be done in advance to gain admission?  I   No reasonable bribe will be refused.  :-)  Um, seriously, access to an wK   non-disclosure agreement (NDA) session typically involves signing a form iJ   that attests to your understanding of the status of the information you J   will hear, and details how you can (or cannot) disclose the information G   to others.  (Various companies likely already have applicable Compaq  &   NDAs in place, while others do not.)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 17 May 2000 20:38:09 GMT# From: system@niuhep.physics.niu.eduD9 Subject: samba printing problems and mailing list problem + Message-ID: <8fuvvh$67s$1@husk.cso.niu.edu>-   Hi,   =    Two problems, 1) I can't get on the samba-vms mailing list-, 2) I'm having problems setting up a printer.  0 1) I sent the following and recieved no response From:   NIUHEP::SYSTEM To:     LISTPROC@SAMBA.ORG CC:     SYSTEM Subj:a  " subscribe samba-vms Robert Morphis  9 I've tried it from their web server as well with no luck.a: I tried sending "help" to listproc@samba.org with no luck.	 Thoughts?t  ? 2)  The following is a message I tried to send directly to the t= samba-vms list with no luck (at least I didn't see it show upp9 on the archive and I didn't recieve any email responses).nC $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$d Hullo,A   I am trying to get samba 2.0.3 up and running.  Specifically I s want to serve a printer.     I am a complete novice at this.    VMS 7.2. ucx 5.0al  B Guest account pcfs$account has a default directory of user1:[pcfs]  , I'm not sure what the path should be set at.  A The files I try to print from a windows 95 computer end up in thek* directory specified by the path statement.  E I have included my smb.conf, the fake printcap file, and two snippets  from the log file.   I included the line   %       smbrun = /samba_root/bin/smbrun.  - in my smb.conf and still got the error below.e  M ============================================================================= " from SAMBA_ROOT:[000000]LOG.MUNU;1  K [2000/05/12 08:40:24, 0] DSA1:[KITS.SAMBA-2_0_3.SOURCE.LIB]SMBRUN.C;2:(104)T8   SMBRUN ERROR: Can't find smbrun. Installation problem?M =============================================================================e  # From SAMBA_ROOT:[000000]SMBRUN.COM;H  N $! The 'smbrun' command must be defined as "smbrun" in smb.conf. If you changeL $! this in the conf file, also change it here (there is no reason to do so).H $! Samba checks the existance via stat() and then issues a system() call@ $! with that string. Thus we define a symbol with the same name.M =============================================================================u         NIUHEP$ type [.lib]smb.conf)   [global] ;   workgroup = IfNa    workgroup = Physics    dead time = 104    map archive = noe    printing = bsd 4    printcap name = /samba_root/lib/dummyprintcap.dat    load printers = yes    status = yes F ;   print command = print %f/queue=%p/delete/passall/name="""""%s"""""E    print command = print %f/queue=%p/delete/passall/name="""""%f"""""y!    lprm command = delete/entry=%jc    security = sharet    guest only = yest    default service = default    create mode = 0777a    root directory = /sambaroot    debug level = 3      guest account = PCFS$ACCOUNTc ;   guest account = GUESTTD ;  This next option sets a separate log file for each client. Remove& ;  it if you want a combined log file.    log file = /samba_log/log.%m  ;   read raw = nof    read prediction = yes    socket options = TCP_NODELAYu! ;   socket options = SO_REUSEADDRt  F ;  You will need a world readable lock directory and "share modes=yes"C ;  if you want to support the file sharing modes for multiple usersS ;  of the same files. ;  lock directory = /usr/local/samba/var/locks   share modes = no   [homes]/    comment = Home Directories     browseable = no    read only = not    create mode = 0750t  
 [printers]    comment = All Printers-    path = /user1/pcfs/default     browseable = no    printable = yes ;   public = nou    public = yesg    guest ok = yesl    writable = no    create mode = 0700n  6 ======================================================# NIUHEP$ typ [.lib]dummyprintcap.dath sys$print|sys$print|sys$print|6 ======================================================   NIUHEP$ dir/sin [...]/siz    Directory SAMBA_ROOT:[000000]f   LOG.MUNU;1                99 LOG.SMB;1                 16   Total of 2 files, 115 blocks.s  ! Directory SAMBA_ROOT:[000000.VAR]n   LOG.NMB;1                214 LOG.SMB;1                  9 NMBD_STARTUP.LOG;1         4   Total of 4 files, 227 blocks.9  ' Directory SAMBA_ROOT:[000000.VAR.LOCKS]N   BROWSE.DAT;1               1 NMBD.PID;1                 0   From LOG.MUNU;1:  I [2000/05/12 10:28:37, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(148)n(   get_uai: for username = "PCFS$ACCOUNT"O [2000/05/12 10:28:37, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:(e.   Convert to UNIX: USER1:[PCFS] -> /user1/pcfsO [2000/05/12 10:28:37, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:(1   UNIX path: /user1/pcfsI [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(219) )   get_uai: $getuai returns UIC = 0098000Ep#            priv = 00108000 00000000mI [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(534) G   vms_setuid: uid = 0098000E, username = "PCFS$ACCOUNT", uic = 0098000EsI [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(582)n'   vms_setuid: privs = 00108000 00000000hM [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.LIB]DOSCALLS.C;1:(327)d   dos_ChDir to /user1/pcfsI [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.LIB]UTIL.C;2:(609)-   unix_clean_name []I [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.LIB]UTIL.C;2:(609)e   unix_clean_name [munu.652399]M  vms_stat: st = -1, mode = 075367           size = 419  vms_stat: st = -1, mode = 000053           size = -32121e, Open file "munu.652399"  ->  "./munu.652399"  ---flags: 40000A01, mode: 000600J [2000/05/12 10:28:38, 2] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]OPEN.C;3:(605)?   fortner opened file munu.652399 read=No write=Yes (numopen=1)nM [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(565)I   Transaction 6 of length 39M [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(402)n(   switch message SMBtdis (pid 553652399)M [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.LIB]DOSCALLS.C;1:(327)0   dos_ChDir to /samba_root/varI [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(502).   vms_setuid: uid = 00000000I [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(148)N"   get_uai: for username = "SYSTEM"O [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:(a>   Convert to UNIX: SYS$SYSROOT:[SYSMGR] -> /sys$sysroot/sysmgrO [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:(s    UNIX path: /sys$sysroot/sysmgrI [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(219) )   get_uai: $getuai returns UIC = 00010004d#            priv = FFFFFFFF FFFFFFFFlI [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(534)dA   vms_setuid: uid = 00010004, username = "SYSTEM", uic = 00010004cI [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(582)a'   vms_setuid: privs = FFFFFFFF FFFFFFFFII [2000/05/12 10:28:38, 0] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]UID.C;1:(285)s3   Warning: You appear to have a trapdoor gid system M [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]SERVICE.C;1:(512)-8   munu (131.156.85.44) closed connection to service IPC$O [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]CONNECTION.C;1:(40)v   Yielding connection to IPC$mM [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(565)    Transaction 7 of length 67M [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(402)s)   switch message SMBioctl (pid 553652399)cK [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]REPLY.C;2:(381)F   ignoring ioctlK [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]ERROR.C;1:(133) A   error packet at line 387 cmd=39 (SMBioctl) eclass=2 ecode=65535 M [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(565)t   Transaction 8 of length 4148M [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(402),)   switch message SMBwrite (pid 553652399) I [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(502)    vms_setuid: uid = 0098000EI [2000/05/12 10:28:38, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(148) (   get_uai: for username = "PCFS$ACCOUNT"O [2000/05/12 10:28:39, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:(-.   Convert to UNIX: USER1:[PCFS] -> /user1/pcfsO [2000/05/12 10:28:39, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:(s   UNIX path: /user1/pcfsI [2000/05/12 10:28:39, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(219)m)   get_uai: $getuai returns UIC = 0098000Ef#            priv = 00108000 00000000pI [2000/05/12 10:28:39, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(534)uG   vms_setuid: uid = 0098000E, username = "PCFS$ACCOUNT", uic = 0098000EfI [2000/05/12 10:28:39, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]PWD.C;71:(582)h'   vms_setuid: privs = 00108000 00000000tM [2000/05/12 10:28:39, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.LIB]DOSCALLS.C;1:(327)m   dos_ChDir to /user1/pcfsL [2000/05/12 10:28:44, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]REPLY.C;2:(2428)%   write fnum=5481 num=4096 wrote=4096nM [2000/05/12 10:28:44, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(565)M     [big snip ]n  L [2000/05/12 10:28:47, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]REPLY.C;2:(2428)%   write fnum=5481 num=2926 wrote=2926 M [2000/05/12 10:28:47, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(565)    Transaction 16 of length 45aM [2000/05/12 10:28:47, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(402) )   switch message SMBclose (pid 553652399)gO [2000/05/12 10:28:47, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:(a.   Convert to VMS: munu.652399 -> ./munu.652399O [2000/05/12 10:28:47, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:()   VMS File: []MUNU.652399nL [2000/05/12 10:28:47, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]REPLY.C;2:(2654)"   close fd=6 fnum=5481 (numopen=1)J [2000/05/12 10:28:47, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]OPEN.C;3:(155)O   fd_attempt_close fd = 6, dev = 3c7a30, inode = 56280, open_flags = 1, ref_couI vms_stat: st = 0, mode = 100602g           size = 19456O [2000/05/12 10:28:48, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:(e.   Convert to VMS: munu.652399 -> ./munu.652399O [2000/05/12 10:28:48, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.VMS]CVT_FILESPEC.C;105:(t   VMS File: []MUNU.652399m  vms_stat: st = -1, mode = 000127           size = 82 K [2000/05/12 10:28:48, 0] DSA1:[KITS.SAMBA-2_0_3.SOURCE.LIB]SMBRUN.C;2:(104)r8   SMBRUN ERROR: Can't find smbrun. Installation problem?O [2000/05/12 10:28:48, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.PRINTING]PRINTING.C;1:(1 O   Running the command `print []MUNU.652399/queue=sys$print/delete/passall/name=sK [2000/05/12 10:28:48, 2] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]CLOSE.C;1:(163)f2   PCFS$ACCOUNT closed file munu.652399 (numopen=0)M [2000/05/12 10:28:48, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(565)i   Transaction 17 of length 70dM [2000/05/12 10:28:48, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]PROCESS.C;1:(402) )   switch message SMBopenX (pid 553652399)iI [2000/05/12 10:28:48, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.LIB]UTIL.C;2:(609)F   unix_clean_name []I [2000/05/12 10:28:48, 3] DSA1:[KITS.SAMBA-2_0_3.SOURCE.LIB]UTIL.C;2:(609)    unix_clean_name [munu.a52399]a  vms_stat: st = -1, mode = 000127           size = 41d  vms_stat: st = -1, mode = 000053           size = -32121h, Open file "munu.a52399"  ->  "./munu.a52399"  ---flags: 40000A01, mode: 000600J [2000/05/12 10:28:48, 2] DSA1:[KITS.SAMBA-2_0_3.SOURCE.SMBD]OPEN.C;3:(605)?   fortner opened file munu.a52399 read=No write=Yes (numopen=1)W   Morphis@physics.niu.edu 8 Real Women change tires			abuse@uu.net postmaster@uu.net7 Real Men change diapers                 security@uu.net    ------------------------------  % Date: Wed, 17 May 2000 18:11:58 -0500u) From: "John E. Malmberg" <wb8tyw@qsl.net>S= Subject: Re: samba printing problems and mailing list problemR7 Message-ID: <0e3401bfc055$505bcd50$020a0a0a@xile.realm>h  ; Robert Morphis <system@niuhep.physics.niu.education> wrote:   , 1) I can't get on the samba-vms mailing list   > From:   NIUHEP::SYSTEM > To:     LISTPROC@SAMBA.ORG > CC:     SYSTEM > Subj:S >_$ > subscribe samba-vms Robert Morphis   You have the procedure correct.-  ; > I've tried it from their web server as well with no luck.-< > I tried sending "help" to listproc@samba.org with no luck. > Thoughts?c  I They may be having problems with the list server.  It seems to be workingwB today.  Although I have not receives a SAMBA digest in a few days.  J You may want to try again.  If it does not work, please E-mail me directlyK and I can post this as a problem to the various SAMBA lists I am subscribedq to.o  E Note that I do not have any control or influence on the SAMBA team org project.    @ > 2)  The following is a message I tried to send directly to the? > samba-vms list with no luck (at least I didn't see it show up ; > on the archive and I didn't recieve any email responses). E > $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$  > Hullo,B >   I am trying to get samba 2.0.3 up and running.  Specifically I > want to serve a printer. >_! > I am a complete novice at this.h >r > VMS 7.2. ucx 5.0aR >TD > Guest account pcfs$account has a default directory of user1:[pcfs] >g. > I'm not sure what the path should be set at. >sC > The files I try to print from a windows 95 computer end up in the-, > directory specified by the path statement. >4G > I have included my smb.conf, the fake printcap file, and two snippetso > from the log file. >r > I included the lineD >3' >       smbrun = /samba_root/bin/smbruna   Remove it or comment it out.  L There is a smbd_startup.com file in the distribution that is used to startup the smbd process.d  J It should be defining a logical smbrun samba_root:[bin]smbrun.com or whereL ever it is for that version.  (I am working on 2.0.6) and things are quite a bit different.  L The symbol smbrun = "@samba_root:[bin]smbrun.com" should also be set by this
 procedure.  L The logical makes the test for smbrun work, and the symbol makes the command work.s   -John/ wb8tyw@qsl.network   ------------------------------  # Date: Wed, 17 May 2000 23:33:34 GMTl From: rbanks_arel@my-deja.coms, Subject: Re: String Descriptors, BASIC and C) Message-ID: <8fva8d$coi$1@nnrp1.deja.com>L  2 In article <8fs2uh$qav$1@slb0.atl.mindspring.net>,5   "Randy Park" <rjpark@mindspring.com.nospaam> wrote:s >a9 > I've used BASIC a lot, and C a little bit.  Remember ini9 > BASIC strings can be dynamic or static and you can passe; > strings to subprograms either "by Descr" or "by ref".  Ifo8 > you pass strings by descriptor be sure your subprogram9 > uses the STR$xxx routines to manipulate the strings andl: > to allocate or deallocate them.  Changing the headers by: > yourself will get you into big trouble.  If passing them: > by reference you may also want to pass the length of the8 > string as a separate parameter so that your subprogram > doesn't write beyond the end.a > 9 > If you want to use the C routine as a "String function"A9 > in BASIC you need to add an additional parameter in the-7 > subprogram.  I forget whether the return value is the ( > first parameter or the last parameter. > ; > Also if you know what you are doing, by using the "OPTIONn; > SETUP=INACTIVE" compiler directive, you can significantlyo: > improve the performance of your BASIC subprograms to the: > point of approaching C language performance.  Using this9 > feature prohibits you from using dynamic strings, errora< > trapping, and I/O in the language.  It also changes GOSUBs8 > from a library routine to a single machine instruction	 > on VAX.- >c   Thanks for the info and help,p  H BASIC's default passing mechanism for strings is by descriptor so I will. continue to use descriptors in the C function.  B After playing around I've found that the return value is the firstC parameter.  Now the function works great (about 3 times faster thanD EDIT$).c  F By the way, I had a look at the BASIC help for OPTION and the INACTIVEF command can be used to turn certain processing options off, however in this case I won't need it.   Richard.    & Sent via Deja.com http://www.deja.com/ Before you buy.9   ------------------------------   Date: 17 May 2000 18:21:26 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)8 Subject: Re: turning off system messages on console port6 Message-ID: <8funv6$mem$1@mailint03.im.hou.compaq.com>  X In article <39227600.93E03012@digit.cg.yu>, Andrija Pantovic <panta@digit.cg.yu> writes:: :What would be the easiest way to turn off system messages, :appearing on system console of MicroVAX II.  B   Assuming the messages referenced are from OPCOM, please see the @   OpenVMS FAQ.  Section DCL4.  Or you could log into the console=   port directly on OPA0: and turn them off yourself using thet   REPLY command from DCL4.  A   If you are refering to message(s) other than those from OPCOM,  ?   we will need details and/or examples of these other messages.$   :SW is VMS v.4.7 :HW is MicroVAX II     Whoa, a true antique.m  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Thu, 18 May 2000 03:21:55 GMTl, From: TMcGrath@cendant.com.au (Tony McGrath)+ Subject: Re: Utility to track device errorsr4 Message-ID: <39235f8e.14987203@news.mel.aone.net.au>  1 On Mon, 15 May 2000 08:45:55 -0400, "AndradeJA-A" $ <andradeja@npt.nuwc.navy.mil> wrote:  J >Considering the fact that our cluster doesn't go down and considering the
 >fact thatM >device errors can accumulate in such a way as to make their analysis a minor=L >headache at best and overly time consuming at worst.  Does anyone know of aK >utility that regularly logs changes in SHOW ERROR and sends resulting datas >to. >an output file/email account. >=! >Jay Andrade - VMS System Manager=   Hi Jay,=- Here's what I've been using for many years...M@ It sits in a queue, runs every day, just before I get it at 9am.   Cheers from Oz,     Tony!  5 $!  CHECK_DEVICE.COM		Tony McGrath,  3-MAR-1993 10:03r< $   SAVE_VER = 'F$VERIFY( F$TRNLNM( "CHECK_DEVICE_VERIFY"))'I $!-----------------------------------------------------------------------wF $! Issue the "SHOW ERROR" command and send a mail message to me if any $! new errors are detected. I $!-----------------------------------------------------------------------r $   ON   ERROR   THEN GOTO EXIT1 $   ON CONTROL_Y THEN GOTO EXITt $! $   DEBUG     = 0 " $   SAY      := "WRITE SYS$OUTPUT"0 $   DSAY     := "IF DEBUG THEN WRITE SYS$OUTPUT"4 $   ERR_FILE  = "SYS$SCRATCH:SHOW_ERROR_OUTPUT.LIST"4 $   MAIL_FILE = "SYS$SCRATCH:SHOW_ERROR_OUTPUT.MAIL"4 $   DIFF_FILE = "SYS$SCRATCH:SHOW_ERROR_OUTPUT.DIFF"@ $   DIF$_SAMEFILE  = %X006C8009	! from SYS$MESSAGE:PRGDEVMSG.EXE $   DIF$_FILAREDIF = %X006C8013 @ $   SHOW$_NOERRORS = %X107880A1	! from SYS$MESSAGE:CLIUTLMSG.EXE/ $   THIS_FILE      = F$ENVIRONMENT("PROCEDURE")oF $   WHERE = F$ELEMENT(0,"]",F$PARSE(F$ENVIRONMENT("PROCEDURE"))) + "]", $   SAVE_LOCATION = F$ENVIRONMENT("DEFAULT") $!. $!   IF F$MODE() .EQS. "BATCH" THEN SET VERIFY $   SET DEFAULT 'WHERE'nI $!-----------------------------------------------------------------------t8 $! Check for new errors that weren't reported last time.C $! Do an initial check to see if there really is an error to reportrE $! If there is do it again sending the output to a file. Then compares. $! this file with a previous version to see if& $! (a) the is a new device in the list< $! (a) one of the counters for a device already reported has incrementedmI $!-----------------------------------------------------------------------  $   FOUND_ERROR = 0  $   SHOW ERROR" $   IF $STATUS .NE. SHOW$_NOERRORS $   THEN# $     SHOW ERROR /OUTPUT='ERR_FILE'= $     PURGE 'ERR_FILE' /KEEP=2< $     DIFFERENCES 'ERR_FILE' /OUTPUT='DIFF_FILE' ! /PARALLEL0 $     FOUND_ERROR = ($STATUS .NE. DIF$_SAMEFILE)	 $   ENDIF=I $!-----------------------------------------------------------------------[ $!I $!----------------------------------------------------------------------- ( $   IF .NOT. FOUND_ERROR THEN GOTO L1000 $!A $   CREATE 'MAIL_FILE' ! For user-friendly file/record attributes % $   OPEN /APPEND TMP_FILE 'MAIL_FILE'42 $   WRITE TMP_FILE "ERROR Report from ", THIS_FILE? $   WRITE TMP_FILE "The system has detected a new device error"P $   WRITE TMP_FILE ""  $   CLOSE TMP_FILE $!' $   APPEND /LOG 'DIFF_FILE' 'MAIL_FILE'_' $   APPEND /LOG  'ERR_FILE' 'MAIL_FILE': $! $   MAIL 'MAIL_FILE' TONY -23 	/SUBJECT="There is a device reporting errors !!" -V" 	/PERSONAL_NAME="CHECK_DEVICE.COM" $! $EXIT: $   SET DEFAULT 'SAVE_LOCATION'5= $   IF F$SEARCH(MAIL_FILE) .NES. "" THEN DELETE 'MAIL_FILE';*.= $   IF F$SEARCH(DIFF_FILE) .NES. "" THEN DELETE 'DIFF_FILE';*A% $   EXIT $STATUS+0*F$VERIFY(SAVE_VER):    C Tony McGrath, Systems Administrator, British Airways Executive Club0N Cendant Membership Services, 596 North Road, Ormond, Victoria, 3204, Australia- Phone : 61 3 9578 8233   Fax : 61 3 9578 6326a   ------------------------------   Date: 17 May 2000 18:59:06 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)1 Subject: Re: Vaxstation graphics to SVGA monitor?26 Message-ID: <8fuq5q$n00$3@mailint03.im.hou.compaq.com>  _ In article <39229943.4769D2BF@trailing-edge.com>, Tim Shoppa <shoppa@trailing-edge.com> writes:.A :A question that's a bit different than the typical "How do I useT :my old VR299 on a PC clone":;  @   FAQ.  MISC3.  Just added this detail into the current edition.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 17 May 2000 15:25:01 -0400N+ From: Tim Shoppa <shoppa@trailing-edge.com> 1 Subject: Re: Vaxstation graphics to SVGA monitor?81 Message-ID: <3922B9CD.24FF1D8C@trailing-edge.com>:   Hoff Hoffman wrote:  > a > In article <39229943.4769D2BF@trailing-edge.com>, Tim Shoppa <shoppa@trailing-edge.com> writes:mC > :A question that's a bit different than the typical "How do I useA > :my old VR299 on a PC clone":: > B >   FAQ.  MISC3.  Just added this detail into the current edition.  < Yeah, I saw that, and that's useful (especially the DEC partD number) but for me it's only a PAQ (partial answer to the question).B I also need to know which (if any) SVGA monitors will sync up withE various VAXstations, and these days most monitor manufacturers don't vD even provide the full sync range specs.  So I'm specifically lookingB for an answer like "a Yoyodyne GX-16 monitor works and you can buy0 new ones today from John Bigboote on Planet 10".   Tim.   ------------------------------  % Date: Wed, 17 May 2000 17:03:40 -0400t. From: "Kenneth Randell" <kenr@datametrics.com>1 Subject: Re: Vaxstation graphics to SVGA monitor?B+ Message-ID: <8fv1aa$cfu$1@bob.news.rcn.net>u   Hello --  C This may be of no use, but when I had to replace a VRT19-HA monitorDE (which I believe is an OEMed Sony GDM-xxxx or something like that), I E bought a Sony Multiscan 400PS monitor.  It has dual inputs (VGA and 5RE BNC connectors), and  supports sync-on-green as well.  I'm very happysJ with it.  You can get this (or at least you could about 8 months ago) fromB most any better-than-COMPUSA monitor place, such as NECX direct. I? have this one hooked up to a DEC 3000-300 at 1280x1024 with BNCC7 and a PWS 500au at 1280x1024 and am very happy with it.1   Ken RandellD    D Tim Shoppa wrote in message <39229943.4769D2BF@trailing-edge.com>...A >A question that's a bit different than the typical "How do I useB >my old VR299 on a PC clone":m >aC >Does anyone here have any experience hooking VAXstations (anythingBB >from VS2000's to 3100's with SPX or GPX graphics to 4000VLC's and >4000/60's)DI >to a modern multisync SVGA monitor?  It'd be nice if the monitor accepts3> >sync-on-green, but if not I'd be willing to build my own syncI >separator.  And obviously the multisync SVGA monitor will have to handle G >the scan rates from the Vaxstation(s) - but unfortunately I'm having aB >hardSH >time getting scan rate details on many modern commercial SVGA monitors.E >The specs usually list the supported resolutions, but not the actuald >scan rates. > A >Any leads?  Words from the wise?  I have too many customers with:C >dying VR160's, VR190's, VR290's, VR299's, etc., who really need to9A >keep their screens running reliably, and it'd be a big win if wevC >could just plug in a nice modern (maybe even generic) SVGA monitorAD >when their current monitor dies.  (Notice I said "when", not "if"!) >/ >-- C > Tim Shoppa                        Email: shoppa@trailing-edge.com2H > Trailing Edge Technology          WWW:   http://www.trailing-edge.com/7 > 7328 Bradley Blvd                 Voice: 301-767-5917_7 > Bethesda, MD, USA 20817           Fax:   301-767-59270   ------------------------------  # Date: Wed, 17 May 2000 23:12:10 GMTD* From: kuhrt@eisner.decus.org (Marty Kuhrt)1 Subject: Re: Vaxstation graphics to SVGA monitor?0+ Message-ID: <Q5Pqv7Adt532@eisner.decus.org>O  _ In article <39229943.4769D2BF@trailing-edge.com>, Tim Shoppa <shoppa@trailing-edge.com> writes:-B > A question that's a bit different than the typical "How do I use > my old VR299 on a PC clone": > D > Does anyone here have any experience hooking VAXstations (anythingC > from VS2000's to 3100's with SPX or GPX graphics to 4000VLC's and;2 > 4000/60's) to a modern multisync SVGA monitor?    ? I have my VAXstation 4060 plugged into a NEC MultiSync P750 and;B it works just fine.  The P750 has two inputs on the back, one set @ of barrel type connectors and one d-sub.  The d-sub goes to the > lone NT box I use to deal with email attachments via Outhouse.@ To switch between the two I simply push the button on the front./ Certainly cuts down on desk bloat for monitors.:   ------------------------------   Date: 15 May 2000 19:48:17 GMT) From: leslie@clio.rice.edu (Jerry Leslie)72 Subject: Re: VMS and Ole for Process Control (OPC)' Message-ID: <8fpka1$jbq$1@joe.rice.edu>    jlahman@LTVSteel.com wrote:K : Hi:B  ? : (and greatest) packages that will save programming efforts.  =B : To the last, they all incorporate Ole for Process Control (OPC) E : in their latest products.  (For further info, pls see www.opc.com).K   Try this site instead:     http://www.opcfoundation.org/o   OPC Foundation Homem    4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------  % Date: Thu, 18 May 2000 07:52:05 +0200t? From: "DECHAIZE Thierry (Dir INFRA)" <thierry.dechaize@sncf.fr> 2 Subject: RE: VMS and Ole for Process Control (OPC)@ Message-ID: <21EC2A9D83EED311A3EB0008C733892B032747@S70ERTBIA11>   Hoff wrote :C >	  The Compaq COM for OpenVMS V1.1 product and later (V1.1A is then currentm? >  release, with V1.1B nearing release) contain the necessary =2 functionality=20@ >  to support process control applications that are written in =
 accordance=20iD >  with the OLE for Process Control (OPC) specification, "OPC Data = Access< >  Automation Specification, V2.02", dated February 3, 1999.   	Thank's for this information.C 	And, when available the next release of Compaq COM for OpenVMS,=20n 	and what ara pre-requisites ?@ 	Where, on WEB, the specification of the OLE for Process Control (OPC) specification,E  	"OPC Data Access Automation Specification, V2.02"(dated February 3, 	 1999) aree 	available ?   	Thank's -----Message d'origine-----7# De: hoffman@xdelta.zko.dec.nospam ==& [mailto:hoffman@xdelta.zko.dec.nospam]  Date: mercredi 17 mai 2000 15:20 =C0: Info-VAX@mvb.saic.com0 Objet: Re: VMS and Ole for Process Control (OPC)      5 In article <392089C2.EFDD42F9@videotron.ca>, JF Mezeip& <jfmezei.spamnot@videotron.ca> writes: :jlahman@LTVSteel.com wrote:G :> According to Compaq's VMS Futures presentation, manufacturing is a =l keyc marketF :> for VMS.  However, what we are seeing is that to be a player in theH :> manufacturing market, a product or operating system must be able to = talk OPC.  
   And it can.m  <   The following is slightly paraphrased from some official = documentation:  C   The Compaq COM for OpenVMS V1.1 product and later (V1.1A is the =d currentw>   release, with V1.1B nearing release) contain the necessary = functionality=20?   to support process control applications that are written in = 
 accordance=20-C   with the OLE for Process Control (OPC) specification, "OPC Data =d AccessI   Automation Specification, V2.02", dated February 3, 1999. Compaq will =r  I   not be delivering any process control software, either components or=20tE   clients, but it will be delivering the necessary COM development=20eG   environment and runtime components that will allow process control=20 I   vendors or end users to develop applications that conform to the OPC=20t   specification.  F :Yep, and Compaq still states that international funds tranfers is a = key VMSoD :market with very strong market penetration, even though SWIFT has =	 announcedeC :that it will stop supporting its VMS product in a couple of years.s  F   I forwarded the (unrelated to OPC) SWIFT discussion (when it last=20:   surfaced) along to the relevent OpenVMS product manager.  2  --------------------------- pure personal opinion ---------------------------f3    Hoff (Stephen) Hoffman   OpenVMS Engineering   =m hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Wed, 17 May 2000 21:37:03 GMTc From: carends@evisions.com Subject: VMS File Modesh) Message-ID: <8fv3dt$4t3$1@nnrp1.deja.com>o   All,  C Forgive me for my ignorance but I am having a tough time overcomingpG what SHOULD be a simple problem IMO.  My company has developed a cross-,D platform product that sends a file that we have created to a printerE using whatever command the user would like to use.  Currently, in our   tests, this is just using PRINT.  7 The program is written in C and creates the file using:d    fopen("filename", "w"); This file is a binary PCL file.o  E When we send the file to the print queue using PRINT the queue cannotn9 understand the file and does not print it to the printer.l  F The problem I am having is that I do not know enough about VMS to knowF where the problem lies.  Do I need to create the file differently?  IsE there a flag to pass to PRINT to let it know that the file is Binary?-6 Is there a configuration problem with the print queue?  D I have attempted to create the file using different file modes like:#    fopen("filename", "w, recfm=A");eD ...among others but the file is never created so I obviously have no clue what I'm doing.  B Anyone who has sufficient knowledge of this problem please help meD out.  I am desperate and have exhausted my knowledge of VMS.  If youF could please send a copy of your response to my email address it would be appreciated.n  
 Thank you, Clayton M. Arendsp carends@evisions.com    & Sent via Deja.com http://www.deja.com/ Before you buy.r   ------------------------------  # Date: Wed, 17 May 2000 22:07:08 GMTb From: confusu@my-deja.comm Subject: Re: VMS File Modes ) Message-ID: <8fv560$704$1@nnrp1.deja.com>    All (who care),r  D Actually, we've taken the file that we've created using the program,F FTP'd it to our Windows machine (binary) and then FTP'd it back to the3 VMS system (binary).  It then printed successfully.   E So it seems obvious to me that I need to specify the correct filemode B (besides binary) when creating the file with fopen().  Can someone& please shed some light on this for me?  	 - Claytona    & Sent via Deja.com http://www.deja.com/ Before you buy.p   ------------------------------  # Date: Wed, 17 May 2000 22:23:53 GMTs From: Dirk Munk <munk@home.nl> Subject: Re: VMS File Modes ' Message-ID: <39231BF8.CEE10FD5@home.nl>a   carends@evisions.com wrote:b >  > All, > E > Forgive me for my ignorance but I am having a tough time overcoming I > what SHOULD be a simple problem IMO.  My company has developed a cross- F > platform product that sends a file that we have created to a printerG > using whatever command the user would like to use.  Currently, in our " > tests, this is just using PRINT. > 9 > The program is written in C and creates the file using:  >    fopen("filename", "w");! > This file is a binary PCL file.t > G > When we send the file to the print queue using PRINT the queue cannot.; > understand the file and does not print it to the printer.9  @ It would be helpfull if you can give us the exact error message.B My first shot would be that the most likely cause is not the queueH mechanisme, but the device/terminal setup for that printer. Most printerD ports are setup for ASCII data and need to be reconfigured for other types of data. c   > H > The problem I am having is that I do not know enough about VMS to knowH > where the problem lies.  Do I need to create the file differently?  IsG > there a flag to pass to PRINT to let it know that the file is Binary?o8 > Is there a configuration problem with the print queue? > F > I have attempted to create the file using different file modes like:% >    fopen("filename", "w, recfm=A"); F > ...among others but the file is never created so I obviously have no > clue what I'm doing. > D > Anyone who has sufficient knowledge of this problem please help meF > out.  I am desperate and have exhausted my knowledge of VMS.  If youH > could please send a copy of your response to my email address it would > be appreciated.- >  > Thank you, > Clayton M. Arends  > carends@evisions.com > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.y   ------------------------------  # Date: Wed, 17 May 2000 22:52:46 GMTt From: Dirk Munk <munk@home.nl> Subject: Re: VMS File Modesa' Message-ID: <392322BD.92014646@home.nl>n   confusu@my-deja.com wrote: >  > All (who care),d > F > Actually, we've taken the file that we've created using the program,H > FTP'd it to our Windows machine (binary) and then FTP'd it back to the5 > VMS system (binary).  It then printed successfully.u > G > So it seems obvious to me that I need to specify the correct filemodefD > (besides binary) when creating the file with fopen().  Can someone( > please shed some light on this for me?  E Most likely you will see that the original file was a stream-lf file, H and that the file that came back from the windows machine is a file withD fixed 512 byte records, and no record attributes (use dir/full). YouE might try to change the attributes of the original file with set file-F /attributes to a fixed 512 byte record size without record attributes,) and see what happens when you print it.  -     >  > - Clayton- > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.    ------------------------------  % Date: Wed, 17 May 2000 23:35:39 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>  Subject: Re: VMS File ModesO, Message-ID: <392364FC.4DFB6B5F@videotron.ca>   confusu@my-deja.com wrote:F > Actually, we've taken the file that we've created using the program,H > FTP'd it to our Windows machine (binary) and then FTP'd it back to the5 > VMS system (binary).  It then printed successfully.W  K 1- dir a DIR/FULL before you send the file to the foreign machine, and do aFK DIR/FULL of the file once it has been sent back to VMS. You can then adjust(J you fopen accordingly. Also, note whether each of the 2 file transfers are done in binary or text mode.  L 2-you did not mention what you were printing through. Is it through DCPS, orC just a vanilla print queue that makes no assumption on the content.n   ------------------------------  # Date: Thu, 18 May 2000 05:26:00 GMTa From: confusu@my-deja.comh Subject: Re: VMS File Modese) Message-ID: <8fvusr$2vt$1@nnrp1.deja.com>i   Dirk and JFo  C Thank you both for your replies.  I must say again that VMS is veryeF foreign to me.  The idea that there are so many different file formatsC at the OS level is an idea that is a tad cumbersome.  This does not D plague a UNIX or Windows-based environment.  I am quite used to (andA happy with) creating a file with "w" access and having it work noR matter what the circumstance.E  G I figured the Queue was having problems with the file format.  The best.F solution however would be to not have the user modify their queue.  WeE would rather correct the file-format than have each user create a new-2 queue just to handle our "incorrect" binary files.  C All examples I've ever seen on the newsgroups, regarding fopen() in-C VMS, have always used a simple "w" access to create files.  AnotherEC person on my team was able to track down the following solution.  I_7 would like to know if you have any further suggestions.R  0    fopen("filename", "w", "rat=none", "rfm=fix",&       "mrs=512", "ctx=stm", "mbc=20");  > Our files now print correctly.  The above code is bracketed in= #ifdef ... #endif for a VMS OS and will not be used for otherN environments (of course).     Thanks again for your responses,	 - Clayton     & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------   Date: 17 May 2000 18:07:56 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)  Subject: Re: VMS on the desktop?6 Message-ID: <8fun5s$lkt$2@mailint03.im.hou.compaq.com>  N In article <8fsn2q$bke@fidoii.CC.Lehigh.EDU>, "dls2" <dls2@Lehigh.EDU> writes: :I've never seen, or used, VMS.-  D   Acquire a used VAX or Alpha system and acquire a hobbyist license,   and try it for yourself.  1 :How does VMS fare as a desktop operating system?e  I   Quite nicely.  I use it daily.  If you are not using proprietary-format4I   files (eg: Microsoft file formats), OpenVMS is flexible and works well, 8   and has a variety of tools and applications available.   :From a sysop's perspective?  G   With OpenVMS, the "sysop" is more commonly called the System Manager..F   As for the answer: Rock solid.  (I will admit to having some limited7   level of experience as an OpenVMS system manager. :-)n   :From a user's perspective?   G   Donno.  Some years ago, I switched over from being known as a "user"  *   to being known simply as "trouble."  :-)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 17 May 2000 18:32:24 GMT* From: helbig@astro.rug.nl (Phillip Helbig)  Subject: Re: VMS on the desktop?. Message-ID: <8fuojo$5v7$1@info.service.rug.nl>  
 In article? <Pine.LNX.4.05.10005171548470.28598-100000@Mufasa.pubserv.com>, 6 Christopher Smith <chriss@Mufasa.pubserv.com> writes:   # >On 17 May 2000, Osmo Kujala wrote:h >uK >> It's sad, but we need Microsoft stuff. For example some documents is www & >> are only in Microsoft file formats. >nK >That's not nessecarilly true.  I've found some pretty creative ways aroundpJ >microsoft stuff, myself.  You should see my unix shell script that stripsB >all the garbage out of microsoft word documents, and prints theirH >plain-text content -- it works pretty well, and part of that plain-textI >content is the identifying information that microshaft's program embedds  >in the document!n >rH >I mention that specifically, because I think that the equivelant on VMS >wouldn't be difficult.a  8 http://www2.wku.edu/scripts/fileserv/fileserv.com?CATDOC   Package CATDOC  
    CATDOC.ZIPtG           Description: Extracts contents of MS-Word files as plain text #           Version: V3.4, 4-JUN-1998s/           Author: Victor Wagner, Hunter Goatley.           Architecture: VAX,AXPo           Size: 42 blocks            Language: Ch  * OK, it's a unix port, but it works on VMS.   ------------------------------   Date: 17 May 2000 18:45:06 GMT* From: helbig@astro.rug.nl (Phillip Helbig)  Subject: Re: VMS on the desktop?. Message-ID: <8fupbi$5v7$2@info.service.rug.nl>  B In article <8fuojo$5v7$1@info.service.rug.nl>, helbig@astro.rug.nl (Phillip Helbig) writes: l   >In article @ ><Pine.LNX.4.05.10005171548470.28598-100000@Mufasa.pubserv.com>,7 >Christopher Smith <chriss@Mufasa.pubserv.com> writes: X >a$ >>On 17 May 2000, Osmo Kujala wrote: >>L >>> It's sad, but we need Microsoft stuff. For example some documents is www' >>> are only in Microsoft file formats.i >>L >>That's not nessecarilly true.  I've found some pretty creative ways around  + >OK, it's a unix port, but it works on VMS.)  I This lets me read text files people send as base64 encoded .DOC files (I nF decode them with the C code below).  However, there is the more basic E problem that people think it's OK to send this stuff, and even worse W( that they expect to receive it from you.   #include <stdlib.h>0 #include <stdio.h> #include <ctype.h>   int decode_b64char( int ch ) {n     int  sextet;  =          if( ('A' <= ch) && (ch <= 'Z') )  sextet = ch - 'A'; D     else if( ('a' <= ch) && (ch <= 'z') )  sextet = (ch - 'a') + 26;D     else if( ('0' <= ch) && (ch <= '9') )  sextet = (ch - '0') + 52;7     else if( ch == '+' )                   sextet = 62;E7     else if( ch == '/' )                   sextet = 63;o
     else {K         fprintf( stderr, (isprint(ch) ? "Illegal base64 character '%c'\n" : M                                         "Illegal base64 character 0x%02x\n"),                   ch );         exit( EXIT_FAILURE );a     }r     return( sextet );t }      main( int argc, char *argv[] ) {      FILE *fin;     FILE *fout;      int  flushing = 0;     int  temp = 0;     int  bits = 0;     int  ch;       if( argc < 2 ) {C         fprintf( stderr, "Usage: b64decode <infile> [outfile]\n" );y         exit( EXIT_SUCCESS );      }a        fin = fopen( argv[1], "r" );     if( fin == NULL ) {          perror( "openin" );h         exit( EXIT_FAILURE );'     }        if( argc < 3 )         fout = stdout;
     else {&         fout = fopen( argv[2], "wb" );         if( fout == NULL ) {              perror( "openout" );             fclose( fin );!             exit( EXIT_FAILURE );u	         }u     }t  )     while( (ch = fgetc( fin )) != EOF ) {a         if( isspace(ch) )o             continue;            if( flushing ) {             if( ch != '=' ) { H                 fprintf( stderr, "Illegal character in EOF padding\n" );%                 exit( EXIT_FAILURE );i
             }r	         }          else {             if( ch == '=' )(                 flushing = 1;@             else {                 temp <<= 6; -                 temp |= decode_b64char( ch );e                 bits += 6;!                 if( bits >= 8 ) {l                     bits -= 8;;                     fputc( (0xff & (temp >> bits)), fout );.                 }s
             } 	         }w     }n       fclose( fout );n     fclose( fin ); }r   ------------------------------  % Date: Wed, 17 May 2000 14:59:34 -0400- From: "dls2" <dls2@Lehigh.EDU>  Subject: Re: VMS on the desktop?- Message-ID: <8fuq88$2ms@fidoii.CC.Lehigh.EDU>   2 "Nigel Arnot" <sysmgr@maxwell.ph.kcl.ac.uk> wrote:H > From a sysop's perspective its an absolute dream ... PROVIDED that all9 > the VMS workstations are clustered. If that's the case:  >.D > Adding another desktop is a one-off process taking an hour or two.8 > Per-user administration scales as the number of users.< > Per-system administration scales as the number of systems.8 > Per-disk administration scales as the number of disks.; > Any user can log into any system with no problems at all.FD > Everything works as per the documentation, which is comprehensive.  B Is there a central cluster server, or is everything decentralized?= For storage?  Where do files get saved?  Is there redundancy?P@ For control?  Where are user profiles, and cloning images, kept?' Does VMS even support cloning, or such?   F > The fly in the ointment is the cost of VMScluster licenses. And thatI > most users want applications on their desks that you can't get for VMS.a  C What is the cost of a clustering license?  Is the cost per cluster,VD or per node?  Is there more than one cluster license type?  Is thereA variance in the cost of a cluster license?  By machine types?  By: other criteria?e  6 Is VMS still as worth having, even when not clustered?     appreciatively,i Derrick Shearerk dls2@Lehigh.EDUH   ------------------------------  % Date: Wed, 17 May 2000 15:13:03 -0400m From: "dls2" <dls2@Lehigh.EDU>  Subject: Re: VMS on the desktop?- Message-ID: <8fur1i$l38@fidoii.CC.Lehigh.EDU>l  * "Ebinger . Eric" <EEbinger@drc.com> wrote:+ > > From: Kilgallen@eisner.decus.org.nospamd@ > > As with all operating systems, those who are only doing data> > > entry are better off if their GUI application activates onA > > login and never gives them a command line prompt.  The systemn >2A > This is NOT true.  If you are really doing data entry you don'tt@ > want a GUI at all.  (A menu system, yes, a GUI, no.)  Having aG > "you have mail" window pop up while you are typing in the data is noteI > my idea of fun.  Neither is switching from the keyboard to the mouse toa > the keyboard to the mouse...  8 A well designed GUI does not require one to use a mouse.9 Microsoft Windows is a very well designed GUI, in that ito> allows for complete navigation of its interface, by keyclicks,< alone.  Motif also has such functionality.  This is known as "explicit keyboard focus."  H > Yes, I spent a summer doing keypunching to put myself through college.D > Somehow, the idea of spending eight hours a day entering names andJ > address using a Windows applications makes me break out in a cold sweat.  6 Maybe if you learned to use the keyboard shortcuts, it+ might have been somewhat easier for you....-     Derrick Shearer  dls2@Lehigh.EDU    ------------------------------  + Date: Wed, 17 May 2000 20:32:08 +0000 (   )-3 From: Christopher Smith <chriss@Mufasa.pubserv.com>   Subject: Re: VMS on the desktop?J Message-ID: <Pine.LNX.4.05.10005172028081.28598-100000@Mufasa.pubserv.com>  % On 17 May 2000, Phillip Helbig wrote:g  K > This lets me read text files people send as base64 encoded .DOC files (I pH > decode them with the C code below).  However, there is the more basic G > problem that people think it's OK to send this stuff, and even worse i* > that they expect to receive it from you.  G True enough.  Allthough, if they really want to recieve .doc files, whye) not send them a plain ascii .doc file? :)   A I imagine that word would open it.... well, on the other hand...?o  I If it were up to me, people who need (and only people who _need_) to send E formatted text to somebody else would do it using HTML or postscript.r0 Plain ascii works just fine for everything else.  F I consider it common (maybe it's not common) curtesy to work out whichF formats your intended recipient can read before sending anything aside from human-readable ascii text.n   Regards,   Chriss  O =============================================================================== @ "My two cents"			(http://rootworks.com/twocentsworth.cgi?128562)= Christopher Smith(chriss@pubserv.com)			Prgramer^W Programmer  Prime Synergy of Champaign, IL. % -------------------------------------aI "Where a calculator on the ENIAC is equipped with 18,000 vacuum tubes anddH weighs 30 tons, computers in the future may have only 1,000 vacuum tubes; and weigh only 1.5 tons." -- Popular Mechanics, March 1949 -O -------------------------------------------------------------------------------.   ------------------------------  % Date: Wed, 17 May 2000 20:18:49 +0000m From: SysAdmin <djesys@fsi.net>y  Subject: Re: VMS on the desktop?' Message-ID: <3922FEA9.4DA4BDEF@fsi.net>u   dls2 wrote:n: > A well designed GUI does not require one to use a mouse.1 > Microsoft Windows is a very well designed GUI, b  8 No, it isn't. Remember those words: "well designed GUI".  d > in that it@ > allows for complete navigation of its interface, by keyclicks,> > alone.  Motif also has such functionality.  This is known as > "explicit keyboard focus."  C ...but depends upon the program author to provide support for them.   8 > Maybe if you learned to use the keyboard shortcuts, it- > might have been somewhat easier for you....a  H I known of many places - even in MS Orifice where you are "trapped", and" cannot navigate except by mousing.  < So far, Linux + KDE + Navigator poses such challenges, also.   David J. Dachtera    ------------------------------  % Date: Wed, 17 May 2000 20:49:28 +0000s From: SysAdmin <djesys@fsi.net>l  Subject: Re: VMS on the desktop?' Message-ID: <392305D8.418472EA@fsi.net>    dls2 wrote:s > D > Is there a central cluster server, or is everything decentralized?  E There is no such OpenVMS concept as a "cluster server", although HSCsi6 and HSJs are sometimes called by that name (misnomer).   > For storage?  1 Local or common disk / tape storage is available.    > Where do files get saved?   # The usual places: disk and/or tape.h   > Is there redundancy?  F HSx controllers usually support RAID. As well, local (host-based) RAID options are available.  B > For control?  Where are user profiles, and cloning images, kept?  F Well, not sure what you might mean by "cloning images". However, in anE OpenVMS cluster, the User Authorization File is usually accessible byfF all of the cluster nodes. Sometimes, nodes have local UAFs. It depends- on the needs of the site / application / etc.l  ) > Does VMS even support cloning, or such?t  C Can you be a bit more descriptive / specific? Are you talking abouteC disk-image "clones"? The OpenVMS term would be "Physical BACKUP" oreD "logical (or virtual) disk image container file". You use the BACKUPD command to produce a /PHYSICAL saveset of a disk volume. To create aH disk image container file, MOUNT the disk /FOREIGN and COPY it to a fileG (COPY terminates with an error; however, the resulting file is complete- and intact).  ,E > What is the cost of a clustering license?  Is the cost per cluster, F > or per node?  Is there more than one cluster license type?  Is thereC > variance in the cost of a cluster license?  By machine types?  Bye > other criteria?V  H In a nutshell, the answers to these question are: Exhorbitant. Per node. Yes. Yes. Yes. Yes.   H The cost of OpenVMS and clustering licenses is kept artificially high byF the powers that be within Compaq. Not sure if  this reflects a lack ofG confidence in the product or lack of confidence among the OpenVMS folks H at Compaq as to whether either the product or they have what it takes toF handle a larger user base. Unofficial numbers put the world-wide countE of OpenVMS sites somewhere around one half-million. This has been theaG best guesstimate (though it has been steadily declining) for some yearsaE now, so it seems to be their "comfort zone". The net effect of "price D support" has been to spell ultimate death for the product. Compaq isG largely in denial about this and, if approached, will vigorously defendtC its self-destructive position. (Psych. students will recognize thisoH syndrome from their studies. Maybe your prof. will give you extra creditC if you submit this posting with a correct diagnosis along with youre final exam.)  H I have been a staunch supporter of Affordable OpenVMS for some time now,@ and will remain so even after VMS's untimely demise, which seems unavoidable at this point.  :8 > Is VMS still as worth having, even when not clustered?  H Sure, if you got the bux. Depending what you want it for, OpenVMS can be@ a very stable, very powerful web and e-commerce platform. On the< otherhand, so can Linux which can be had for a virtual song.  F For commercial purposes, you'll need MUCHO DINERO for OpenVMS, and MASE DINERO if you want to cluster. Don't be confused by talk of the "CSA"hD (a.k.a. "partner" or "ISV") program: this is for developers, not endC users. Developers and business partners get something of break. End  users get the royal shaft.  E For non-commercial purposes, the OpenVMS Hobbyist program is probably- what you want.  G Now, if you're talking about sharing services such as ODBC and file and E print services, then you may need to refine the paradigm a bit. ThesehD are common concepts in the "Windows desktop" world. OpenVMS providesD file and print services (though in forms you may not recognize or beE familiar with). Database services take rather a different form in thel OpenVMS world.   David J. Dachterae   ------------------------------  % Date: Wed, 17 May 2000 16:55:55 -0400n From: "dls2" <dls2@Lehigh.EDU>  Subject: Re: VMS on the desktop?- Message-ID: <8fv12e$fau@fidoii.CC.Lehigh.EDU>m  3 "Richard B. Gilbert" <DRAGON@compuserve.com> wrote:h  > Message text written by "dls2"? > >On the desktop, is it likely to be perceived any differentlye; > >than a Un*x workstation?  Other than stability, does VMS 8 > >have any benefits when used as a desktop workstation? >iI >         Fewer applications are "windowed".  Most people open a bunch ofeC > DECterm windows and use the command line interface.   Many of theoD > applications that are windowed have both a windows interface and a > character cell interface.m  7 Then for the desktop, doesn't it make more sense to use 6 terminals, rather than actual VMS machines?  Graphical terminals, or character-based.  $ For standalone desktop VMS machines?# For clustered desktop VMS machines?y   > >Why VMS?  <snip>L >         VMS is generally much more secure than Unix or Windows.  It's muchK > more difficult to "hack into" VMS than Unix or Windows.   There have beenzI > problems in the past and there will doubtless be problems in the future  but5H > they were, and we expect will be, much less numerous than the problems withI > other operating systems.  Digital, now Compaq, fixes such problems when H > they are identified; it's not necessary to write a Robert Morris style worm$ > or a "Backend 2000" to get action.  7 There are many more Unix and Windows installations thani8 there are VMS installations.  Is VMS really more secure?  1 OpenVMS VAX and Alpha Version 6.1 is rated C2....n/ SEVMS VAX and Alpha Version 6.1 is rated B1....a. http://www.radium.ncsc.mil/tpep/epl/index.html  & OpenVMS is at version 7.x, however....  9 >         I could go on but I'm too lazy to write a book!:  : You're not the only one who's mentioned to me the need for9 writing a book to answer the question of "Why VMS?"... ;ps     appreciatively,  Derrick Shearera dls2@Lehigh.EDUd   ------------------------------  % Date: Wed, 17 May 2000 16:31:37 -0400  From: "dls2" <dls2@Lehigh.EDU>  Subject: Re: VMS on the desktop?- Message-ID: <8fuvks$fns@fidoii.CC.Lehigh.EDU>   " "SysAdmin" <djesys@fsi.net> wrote:
 > dls2 wrote: < > > A well designed GUI does not require one to use a mouse.2 > > Microsoft Windows is a very well designed GUI, >:: > No, it isn't. Remember those words: "well designed GUI". >e > > in that itB > > allows for complete navigation of its interface, by keyclicks,@ > > alone.  Motif also has such functionality.  This is known as > > "explicit keyboard focus." > E > ...but depends upon the program author to provide support for them.s  ; Yes, but remember my wording of "allows for."  "Allows for"o does not mean "implements."u  : > > Maybe if you learned to use the keyboard shortcuts, it/ > > might have been somewhat easier for you....s >aJ > I known of many places - even in MS Orifice where you are "trapped", and$ > cannot navigate except by mousing. >r> > So far, Linux + KDE + Navigator poses such challenges, also.  % Is VMS any better off in this regard?      Derrick Shearern dls2@Lehigh.EDUc   ------------------------------   Date: 17 May 2000 21:46:54 GMT* From: helbig@astro.rug.nl (Phillip Helbig)  Subject: Re: VMS on the desktop?. Message-ID: <8fv40e$976$1@info.service.rug.nl>  G In article <392305D8.418472EA@fsi.net>, SysAdmin <djesys@fsi.netwrites:,  J > The cost of OpenVMS and clustering licenses is kept artificially high byH > the powers that be within Compaq. Not sure if  this reflects a lack ofI > confidence in the product or lack of confidence among the OpenVMS folks-J > at Compaq as to whether either the product or they have what it takes to > handle a larger user base. k  H Some people claim it is too expensive, but I think this is a non-issue. G If you're a hobbyist, it's free.  If you're in an academic environment,lH the cost is nominal (and probably paid for out or a budget which doesn'tG impact you directly).  If you're commercial then, yes, it is expensive,>E but you get what you pay for and money to pay good programmers has tosD come from somewhere.  I don't see any reason to complain since fundsE spent on software are such a small fraction of any company's budget. p  I As I've said before, there IS a need for a low-cost license to cover the  D grey area between hobby and commercialism, especially for folks who D won't mind paying big bucks as soon as their venture takes off.  It G would be nice to have some official feedback on this, especially since rF it was discussed here recently and, as far as I could make out, there ( were no real objections to it by anyone.   ------------------------------  + Date: Wed, 17 May 2000 22:15:21 +0000 (   ) 3 From: Christopher Smith <chriss@Mufasa.pubserv.com>   Subject: Re: VMS on the desktop?J Message-ID: <Pine.LNX.4.05.10005172208120.28598-100000@Mufasa.pubserv.com>    On Wed, 17 May 2000, dls2 wrote:  9 > Then for the desktop, doesn't it make more sense to use 8 > terminals, rather than actual VMS machines?  Graphical  > terminals, or character-based.  J Maybe even in a broader context.  I certainly appreciate the advantages ofD being able to run programs from a central location, and for a lot ofH "desktop" purposes, a dumb terminal(xterm or otherwise) would work well.    9 > There are many more Unix and Windows installations thanu: > there are VMS installations.  Is VMS really more secure?  F Having some experience on all the above mentioned platforms, I can say! that, in my professional opinion:x   Windows has no security.  G Unix might have some security depending on which day of the week it is,r and who's your administrator.   E VMS has security.  Sometimes the security is better (depending on the G system manager), but it always seems to manage more security than otherLA systems that are common on small-to-mid sized machines, at least.   H I don't think that this has much to do with the number of installations.  J Of course, it's impossible to make an idiot-proof system, but again, in myE opinion, VMS is a closer-to-idiot-proof system than most others where  security is concerned.   Regards,   ChrisF  O =============================================================================== @ "My two cents"			(http://rootworks.com/twocentsworth.cgi?128562)= Christopher Smith(chriss@pubserv.com)			Prgramer^W Programmer  Prime Synergy of Champaign, IL.h% ------------------------------------- I "Where a calculator on the ENIAC is equipped with 18,000 vacuum tubes and H weighs 30 tons, computers in the future may have only 1,000 vacuum tubes; and weigh only 1.5 tons." -- Popular Mechanics, March 1949  O -------------------------------------------------------------------------------1   ------------------------------  % Date: Wed, 17 May 2000 17:55:59 -0400k From: "dls2" <dls2@Lehigh.EDU>  Subject: Re: VMS on the desktop?- Message-ID: <8fv4j2$cii@fidoii.CC.Lehigh.EDU>m  " "SysAdmin" <djesys@fsi.net> wrote:
 > dls2 wrote:I > >hF > > Is there a central cluster server, or is everything decentralized? >dG > There is no such OpenVMS concept as a "cluster server", although HSCss8 > and HSJs are sometimes called by that name (misnomer).  2 Aren't HSC (CI), HSJ (CI), HSD (DSSI), HSZ (SCSI),4 & HSG (FC) just designations for device controllers?  ) Do the names have any particular meaning?o' What are CI and DSSI?  Do they see much & use relative to SCSI and FibreChannel?   > > For storage? >n3 > Local or common disk / tape storage is available.    OK.r   > > Where do files get saved?I >h% > The usual places: disk and/or tape.r    & Usually locally, or remotely, or both?   > > Is there redundancy? >oH > HSx controllers usually support RAID. As well, local (host-based) RAID > options are available.  . Are large numbers of diskless VMS workstations0 commonly, served by small numbers of VMS servers2 with large storage capacity, a common arrangement?  + Would the workstations likely be clustered?e& Would the servers likely be clustered?  / Would the workstations and servers be clustereds" together, or clustered seperately?  D > > For control?  Where are user profiles, and cloning images, kept? > H > Well, not sure what you might mean by "cloning images". However, in anG > OpenVMS cluster, the User Authorization File is usually accessible bygH > all of the cluster nodes. Sometimes, nodes have local UAFs. It depends/ > on the needs of the site / application / etc.g  4 The UAF for a cluster would have to be accessible to all cluster nodes.  2 But what do you mean that "*sometimes*, nodes have5 local UAFs?"  Do you mean that cluster nodes may have 0 node specific UAFs?  Or do you mean that cluster4 UAFs are generally not on any cluster nodes?  Aren't8 all machines of a cluster refered to as "nodes?"  If so,7 then is there any particular node which maintains a UAF 7 accessible to all other nodes, or do all nodes maintainm1 copies of a single UAF, or some other arangement?h  + > > Does VMS even support cloning, or such?i >oE > Can you be a bit more descriptive / specific? Are you talking abouthE > disk-image "clones"? The OpenVMS term would be "Physical BACKUP" or F > "logical (or virtual) disk image container file". You use the BACKUPF > command to produce a /PHYSICAL saveset of a disk volume. To create aJ > disk image container file, MOUNT the disk /FOREIGN and COPY it to a fileI > (COPY terminates with an error; however, the resulting file is completen > and intact).  4 Yes, I am talking about disk images; having a single4 disk, or boot, image propagated to multiple machines: of a single configuration, with, or without, local drives.  2 Does VMS support boot serving for machines without2 local disks?  Does VMS support serving disk images( at bootup for machines with local disks?  G > > What is the cost of a clustering license?  Is the cost per cluster,nH > > or per node?  Is there more than one cluster license type?  Is thereE > > variance in the cost of a cluster license?  By machine types?  Byd > > other criteria?h > J > In a nutshell, the answers to these question are: Exhorbitant. Per node. > Yes. Yes. Yes. Yes.t   Ouch!e  J > The cost of OpenVMS and clustering licenses is kept artificially high byH > the powers that be within Compaq. Not sure if  this reflects a lack ofI > confidence in the product or lack of confidence among the OpenVMS folkseJ > at Compaq as to whether either the product or they have what it takes toH > handle a larger user base. Unofficial numbers put the world-wide countG > of OpenVMS sites somewhere around one half-million. This has been thesI > best guesstimate (though it has been steadily declining) for some years G > now, so it seems to be their "comfort zone". The net effect of "priceeF > support" has been to spell ultimate death for the product. Compaq isI > largely in denial about this and, if approached, will vigorously defendtE > its self-destructive position. (Psych. students will recognize thisPJ > syndrome from their studies. Maybe your prof. will give you extra creditE > if you submit this posting with a correct diagnosis along with youru > final exam.)  - Who's guesstimate is the half-million figure?l  J > I have been a staunch supporter of Affordable OpenVMS for some time now,B > and will remain so even after VMS's untimely demise, which seems > unavoidable at this point.  9 When is VMS's untimely demise expected?  Hasn't it always0$ been predicted as sooner than it is?   <snip>I > Now, if you're talking about sharing services such as ODBC and file andoG > print services, then you may need to refine the paradigm a bit. ThesesF > are common concepts in the "Windows desktop" world. OpenVMS providesF > file and print services (though in forms you may not recognize or beG > familiar with). Database services take rather a different form in the  > OpenVMS world.  ; Are the OpenVMS file and print services similar to those ofu Un*x, or different, still?   What about database services?b     appreciatively,s Derrick Shearero dls2@Lehigh.EDUh   ------------------------------  % Date: Wed, 17 May 2000 18:28:36 -0400n' From: "Bill Todd" <billtodd@foo.mv.com>   Subject: Re: VMS on the desktop?( Message-ID: <8fv69l$68p$1@pyrite.mv.net>  L SysAdmin <djesys@fsi.net> wrote in message news:392305D8.418472EA@fsi.net...
 > dls2 wrote:    ...0  G > > What is the cost of a clustering license?  Is the cost per cluster, H > > or per node?  Is there more than one cluster license type?  Is thereE > > variance in the cost of a cluster license?  By machine types?  Byh > > other criteria?S >n@ > In a nutshell, the answers to these question are: Exhorbitant.  6 Well, that may depend upon one's bases for comparison.  L First, the entry-level per-node cluster license price just dropped from $12KH to $6K, with a crippled (effectively, workstation-only, not suitable forK providing services to other nodes) version at $1.5K.  So Compaq *is* moving K in the right direction, though even these prices aren't chicken feed, givenp? that the entry-level hardware cost is about $3.5K (for a DS10).e  H Now, if this license applies to any system in the 'workstation' tier, itI also applies to DS20s (starting at over $10K hardware, I think) and ES40snH (starting at $20+K for hardware, I think), so for those systems it seemsJ pretty reasonable, given what you get (especially what you get compared toK competing alternatives that term themselves 'clusters').  So one path couldBJ be to try to convince Compaq that it needs a special entry-level tier that, doesn't include these upper-low-end systems.  G But even for a DS10, the cluster license price compared to the hardwareRL price isn't out of line with industry practice.  I recently read a note thatJ indicated the per-node price to *upgrade* a Novell cluster node to the newF release cost $5K, and the per-node cost-of-entry to a Win2K cluster isI likely largely reflected in the $2700 price difference between W2K ServeriE (no clustering) and W2K Advanced Server (2-node failover clustering -eL whoopee!).  And the Linux 'cluster' variants, at least as of the most recentL information I've read, don't really address availability all that well (saveI perhaps for read-only redundant web-server-style stuff), so the fact thatd" they're free doesn't really count.   - bill  
  Per node. > Yes. Yes. Yes. Yes.  >tJ > The cost of OpenVMS and clustering licenses is kept artificially high byH > the powers that be within Compaq. Not sure if  this reflects a lack ofI > confidence in the product or lack of confidence among the OpenVMS folks J > at Compaq as to whether either the product or they have what it takes toH > handle a larger user base. Unofficial numbers put the world-wide countG > of OpenVMS sites somewhere around one half-million. This has been thefI > best guesstimate (though it has been steadily declining) for some years.G > now, so it seems to be their "comfort zone". The net effect of "pricerF > support" has been to spell ultimate death for the product. Compaq isI > largely in denial about this and, if approached, will vigorously defendeE > its self-destructive position. (Psych. students will recognize thiseJ > syndrome from their studies. Maybe your prof. will give you extra creditE > if you submit this posting with a correct diagnosis along with yourt > final exam.) >hJ > I have been a staunch supporter of Affordable OpenVMS for some time now,B > and will remain so even after VMS's untimely demise, which seems > unavoidable at this point. >i: > > Is VMS still as worth having, even when not clustered? >EJ > Sure, if you got the bux. Depending what you want it for, OpenVMS can beB > a very stable, very powerful web and e-commerce platform. On the> > otherhand, so can Linux which can be had for a virtual song. >rH > For commercial purposes, you'll need MUCHO DINERO for OpenVMS, and MASG > DINERO if you want to cluster. Don't be confused by talk of the "CSA"aF > (a.k.a. "partner" or "ISV") program: this is for developers, not endE > users. Developers and business partners get something of break. Endi > users get the royal shaft. >vG > For non-commercial purposes, the OpenVMS Hobbyist program is probablye > what you want. >aI > Now, if you're talking about sharing services such as ODBC and file and G > print services, then you may need to refine the paradigm a bit. These1F > are common concepts in the "Windows desktop" world. OpenVMS providesF > file and print services (though in forms you may not recognize or beG > familiar with). Database services take rather a different form in the" > OpenVMS world. >i > David J. Dachterac   ------------------------------  # Date: Thu, 18 May 2000 01:55:27 GMT,9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)k  Subject: Re: VMS on the desktop?+ Message-ID: <BljPmNpSkSQz@eisner.decus.org>b  N In article <8fur1i$l38@fidoii.CC.Lehigh.EDU>, "dls2" <dls2@Lehigh.EDU> writes:, > "Ebinger . Eric" <EEbinger@drc.com> wrote:, >> > From: Kilgallen@eisner.decus.org.nospamA >> > As with all operating systems, those who are only doing datae? >> > entry are better off if their GUI application activates onoB >> > login and never gives them a command line prompt.  The system >>B >> This is NOT true.  If you are really doing data entry you don'tA >> want a GUI at all.  (A menu system, yes, a GUI, no.)  Having adH >> "you have mail" window pop up while you are typing in the data is notJ >> my idea of fun.  Neither is switching from the keyboard to the mouse to >> the keyboard to the mouse...  > : > A well designed GUI does not require one to use a mouse.; > Microsoft Windows is a very well designed GUI, in that it @ > allows for complete navigation of its interface, by keyclicks,> > alone.  Motif also has such functionality.  This is known as > "explicit keyboard focus." > I >> Yes, I spent a summer doing keypunching to put myself through college.nE >> Somehow, the idea of spending eight hours a day entering names andsK >> address using a Windows applications makes me break out in a cold sweat.p > 8 > Maybe if you learned to use the keyboard shortcuts, it- > might have been somewhat easier for you....n  @ No, an efficient GUI-based data entry program should not requireB keyboard shortcuts -- just the natural keys relevant to the domain@ of the input data.  Maybe carriage-return means enter the record? and tab means go to the next field.  The power of the GUI is toh@ display clearly to the user status indications regarding what is	 going on..  > Certainly no mousing should be involved in the core data entry> business, although I could see using the mouse once an hour or@ so, perhaps to switch from entering orders to entering invoices, or whatever.   ------------------------------  % Date: Wed, 17 May 2000 22:11:03 -0400l# From: sol gongola <sol@adldata.com>e  Subject: Re: VMS on the desktop?& Message-ID: <39235137.601@adldata.com>   dls2 wrote:  > 5 > "Richard B. Gilbert" <DRAGON@compuserve.com> wrote:h" > > Message text written by "dls2"A > > >On the desktop, is it likely to be perceived any differentlyn= > > >than a Un*x workstation?  Other than stability, does VMSe: > > >have any benefits when used as a desktop workstation? > >.K > >         Fewer applications are "windowed".  Most people open a bunch ofdE > > DECterm windows and use the command line interface.   Many of the F > > applications that are windowed have both a windows interface and a > > character cell interface.n > 9 > Then for the desktop, doesn't it make more sense to use78 > terminals, rather than actual VMS machines?  Graphical  > terminals, or character-based. >k& > For standalone desktop VMS machines?% > For clustered desktop VMS machines?8 >    not really. F A single workstation display functioning as a x terminal (or x server)F allows me to have many terminal sessions at the same time. I can have F sessions open from any accessible VMS system as well as sessions open G from other unix systems on my network. To get the same capability with rG terminals, i'd have to surround myself with up to 10 terminal displays k5 and keyboards, each with a cable running to a system.t  
 > > >Why VMS?e > <snip>N > >         VMS is generally much more secure than Unix or Windows.  It's muchM > > more difficult to "hack into" VMS than Unix or Windows.   There have beenlO > > problems in the past and there will doubtless be problems in the future butfO > > they were, and we expect will be, much less numerous than the problems withnK > > other operating systems.  Digital, now Compaq, fixes such problems when O > > they are identified; it's not necessary to write a Robert Morris style worm & > > or a "Backend 2000" to get action. > 9 > There are many more Unix and Windows installations thand: > there are VMS installations.  Is VMS really more secure? > 3 > OpenVMS VAX and Alpha Version 6.1 is rated C2....e1 > SEVMS VAX and Alpha Version 6.1 is rated B1.... 0 > http://www.radium.ncsc.mil/tpep/epl/index.html > ( > OpenVMS is at version 7.x, however.... > ; > >         I could go on but I'm too lazy to write a book!r > < > You're not the only one who's mentioned to me the need for; > writing a book to answer the question of "Why VMS?"... ;p.  ? Its not the 'need' to write a book. It is the large quantity of:' information that could be put in it. :)k >  sol gongolas   ------------------------------  % Date: Wed, 17 May 2000 22:12:10 -0400o, From: taterskins@patriot.net (Ramon L. Tate)  Subject: Re: VMS on the desktop?D Message-ID: <taterskins-ya023480001705002212100001@news.patriot.net>  M In article <8fuq88$2ms@fidoii.CC.Lehigh.EDU>, "dls2" <dls2@Lehigh.EDU> wrote:o   [snip]  D > Is there a central cluster server, or is everything decentralized?? > For storage?  Where do files get saved?  Is there redundancy? B > For control?  Where are user profiles, and cloning images, kept?) > Does VMS even support cloning, or such?o [snip] . > appreciatively,  > Derrick Shearerd > dls2@Lehigh.EDU   D Storage may be housed in individual nodes or in separate boxes (e.g.G HSCxx); this includes disks as well as tapes. Storage units may (or mayiE not) be served (i.e. made "visible") to any or all cluster members asmK desired. The master UAF will reside on a disk that is served to all cluster J members, enabling users to be logged into any available node transparentlyH (or only to specific ones - the system administrator has full control of< this). There are also NFS implementations available for VMS.  I Disk backups may be done to either disk or tape and can be done in such a=K way as to create copies that can be used to fully restore any failed drive,oH including a "system" disk, i.e. the one that contains the master clusterB information base. Of course, disk shadowing ("mirroring") and RAIDB redundancy options are available (extra cost) as well as striping.  B Application images can be made available on multiple cluster nodesH ("cloning"?) and their availability managed centrally in such a way thatK users can get access to them even in the event of the failure of a specific-H node. A particularly powerful capability is the generic queue. Print andF batch jobs can be targetted to a generic queue which in turn feeds anyG number of specific queues managed by specific nodes. This provides loadsG balancing and a degree of protection from single-node or queue failure.r  	 Etc......r   --  
 Ramon L. Tatee	 Casa Maa?= taterskins@patriot.net   "Skin" that "tater" before replying!    ------------------------------  % Date: Wed, 17 May 2000 23:33:53 -0400  From: "dls2" <dls2@Lehigh.EDU>  Subject: Re: VMS on the desktop?- Message-ID: <8fvock$kqm@fidoii.CC.Lehigh.EDU>e  & "sol gongola" <sol@adldata.com> wrote:
 > dls2 wrote:I7 > > "Richard B. Gilbert" <DRAGON@compuserve.com> wrote:e$ > > > Message text written by "dls2"C > > > >On the desktop, is it likely to be perceived any differentlyl? > > > >than a Un*x workstation?  Other than stability, does VMSs< > > > >have any benefits when used as a desktop workstation? > > > J > > >         Fewer applications are "windowed".  Most people open a bunch ofG > > > DECterm windows and use the command line interface.   Many of themH > > > applications that are windowed have both a windows interface and a > > > character cell interface.h > >k; > > Then for the desktop, doesn't it make more sense to usel: > > terminals, rather than actual VMS machines?  Graphical" > > terminals, or character-based. > > ( > > For standalone desktop VMS machines?' > > For clustered desktop VMS machines?  >e
 > not really.    Why "not really?"e  H > A single workstation display functioning as a x terminal (or x server)G > allows me to have many terminal sessions at the same time. I can have G > sessions open from any accessible VMS system as well as sessions open H > from other unix systems on my network. To get the same capability withH > terminals, i'd have to surround myself with up to 10 terminal displays7 > and keyboards, each with a cable running to a system.o  6 Terminals can have the same graphical and multisession7 capabilities as Un*x or VMS workstations; individually,a: rather than the 10 or so equivalent terminals you suggest.  2 By example, NCD HMXpro24s provide such capability.  5 I now think that standalone VMS machines are not verye6 good as desktop machines.  They lack the graphical and; audial capability of other offerings, such as Irix, or evenr Windows.  1 Is VMS any better for the desktop when clustered?o     Derrick ShearerU dls2@Lehigh.EDUl   ------------------------------  # Date: Thu, 18 May 2000 04:08:03 GMTo7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>a  Subject: Re: VMS on the desktop?- Message-ID: <39236D77.AF314E9F@earthlink.net>C   dls2 wrote:w4 > Aren't HSC (CI), HSJ (CI), HSD (DSSI), HSZ (SCSI),6 > & HSG (FC) just designations for device controllers?  = As applied to these devices, the term "cluster controller" ish? inappropriate. They "control" disk arrays, and can be used in a ' clustered or non-clustered environment.n  w+ > Do the names have any particular meaning?g  H Originally, "H.S.C." stood for "Hierarchical Storage Controller". As forG the others, the third character now designates the transport (physical) B medium: HSJ is CI, HSD is DSSI, HSZ is SCSI, HSG is "glass" (read: fibre).o  ) > What are CI and DSSI?  Do they see muchB( > use relative to SCSI and FibreChannel?  E Well, DSSI is officially "deprecated", to use a UN*Xly term. It was asC sort of DEC-proprietary, SCSI-like interface. Throughput is roughlye? equivalent to SCSI-1, according to Hoff. CI is a co-axial loop,qD dual-path arrangement with automatic failover should one path (TX/RXD pair of coax. cables) become unusable. In practice, it actually usesB both paths, if I'm not too badly mistaken. I'll assume that you'reA familiar with SCSI and FC/fabric (note that OpenVMS supports onlyVE fabric, not FCAL, again if my info. is correct - and it may very wellg not be).  h > > > Where do files get saved?n > >r' > > The usual places: disk and/or tape.e > ( > Usually locally, or remotely, or both?   Could be either, really. u  0 > Are large numbers of diskless VMS workstations2 > commonly, served by small numbers of VMS servers4 > with large storage capacity, a common arrangement?  G I don't know if "large numbers" would be appropriate to say. Obviously,rH some of the OpenVMS engineering folk would be very likely to see such anE arrangement. In practice, it would be not common, but not unheard of.e  l- > Would the workstations likely be clustered?u  D With each other? You do sometimes see this, though the usefulness isG debatable since the cluster communications medium for both SCS and MSCP., traffic would be the ethernet (10/100Mbits).  ( > Would the servers likely be clustered?1 > Would the workstations and servers be clusteredt$ > together, or clustered seperately?  C The servers with each other, probably. With the workstations, you'dt likely see it both ways.  t6 > The UAF for a cluster would have to be accessible to > all cluster nodes. > 4 > But what do you mean that "*sometimes*, nodes have7 > local UAFs?"  Do you mean that cluster nodes may haveh > node specific UAFs?   H Yes. This is sometimes necessary, for certain reasons. Normally, cluster nodes will share a common UAF.  6 > Yes, I am talking about disk images; having a single6 > disk, or boot, image propagated to multiple machines< > of a single configuration, with, or without, local drives.   Again, possible.  :4 > Does VMS support boot serving for machines without > local disks?    E Yes. There is also a discontinued device known as an InfoServer which. serves a similar purpose.   & > Does VMS support serving disk images* > at bootup for machines with local disks?  G Well, I suppose you could. However, with system disks reaching the 4 toiE 8 GB range, I can't imagine a workstation having enough memory for antF 8GB ramdisk AND a running OpenVMS system - unless you're talking aboutH some "magic" outside of my experience (very possible and highly likely).  e/ > Who's guesstimate is the half-million figure?o  H Official numbers from (formerly Digital, now Compaq) place the number of/ OpenVMS sites world-wide in that range, +/- 5%.v  i; > When is VMS's untimely demise expected?  Hasn't it always & > been predicted as sooner than it is?  E In general, yes. However, unless some serious marketing happens soon,1D OpenVMS will reach Compaq's goal of the "high-end e-commerce server"D market - because (almost) no one else can afford to either buy it or support it.0  G With the exception of certain large sites here in the Chicago, IL (USA)s> market, OpenVMS will be all but a memory inside of 18 months -F experience speaking here; witness: all of the "small" sites of which IB am aware will have abandonded OVMS within 24 months, including theG number 1 health care org. in Metro Chgo., which I really don't consider E "small" (three-node 8400 cluster with 1.5TB of disk). The biggest VMSrB VAR/reseller in the area (V.A.C.S.) reports near-zero sales of newA Alpha/VMS systems so far in the calendar year. The used market isu- somewhat more active, but hardly encouraging.   F So, IMNSHO, it's fairly safe to say that while total death for OVMS isB not on the visible horizon, within five years OVMS will become the "PICK" of the 21st century.a  v= > Are the OpenVMS file and print services similar to those of  > Un*x, or different, still?   See http://www.samba.org/t  a > What about database services?   lC There exists a product called CONNX, and others, which provide ODBChF access to RMS files. I believe similar facilities may exist for Oracle@ and/or RDB, but these are not within my sphere of daily contact.   -- e David J. Dachteral dba DJE Systems " http://home.earthlink.net/~djesys/  : Unofficial Affordable OpenVMS Home Page and Message Board:+ http://home.earthlink.net/~djesys/vms/soho/    ------------------------------  % Date: Wed, 17 May 2000 22:21:27 +0200 . From: "Marc Van Dyck" <marc.vandyck@skynet.be>4 Subject: Re: Where to start with motif programming ?* Message-ID: <8fuusv$f8f$1@news0.skynet.be>  I Well, I found the examples in DECW$EXAMPLES, and an almost complete Motif H documentation shelf, on the layered products documentation CDs. The only problem"H is that this shelf does not contain the OSF/Motif programming reference, only theK programmer's guide and the style guide. So I'm left without any descriptionp of, for(J example, the arguments of the API routines. Anyone where I can look around forr? this reference ? Also, do you know if the Pascal sources of the  DECW$MOTIF.PEN* environment file are available somewhere ?  
 TIA again,   -- Marc  (Marc.Vandyck@skynet.be)  8 Tim Llewellyn <tim.llewellyn@bbc.co.uk> wrote in message# news:39211EFF.12F28FCC@bbc.co.uk...i6 > There are some examples in DECW$EXAMPLES. The latest= > V7.2 doc CD only seems to have the Motif Installation Guidee; > and Release Notes. Maybe the old Xlib programming manuals,@ > (there was one for a C binding and one for a VMS style bindingA > I seem to remember) are not there anymore, possibly they are oni  > the layered products docs CDs. >    ------------------------------  # Date: Wed, 17 May 2000 19:55:36 GMTb4 From: kaplow_r@eisner.decus.org.spamnot (Bob Kaplow)" Subject: Re: Wildfire Announcement+ Message-ID: <tGh5IgHrWmi8@eisner.decus.org>   m In article <3921725A.86779C78@merlin.arc.nasa.gov>, "Arthur E. Ragosta" <ragosta@merlin.arc.nasa.gov> writes:s > It's finally on the website! > I > Although you would think that they would know better by now (based uponf > all ofJ > the grousing in this forum) to not introduce their latest machines with: > J > "The new Alphaserver GS80 is the industry's fastest RISC Unix server..." > E > It does mention that it runs VMS at the bottom of the page... Geez.t >  > Prices!  I need prices!! > G > Can anyone who attended/listened to the announcement tell me what the- > entry- > price is for the GS80. > 	 > thanks.0  G I must say I was overall disappointed at the announcement. Gon't get me I wrong, I'm excited about WildFire, and my employer is one of the 237 withgJ orders already in, with delivery next month. But the two hour announcementG was remarkably "content free" IMHO. The technical content wasjust aboutlL ZERO. If I hadn't known about Wildfire for over 2.5 years now, read "ShannonH Knows Compaq", and picked up other tidbits over the past couple years, IF certainly wouldn't have learned anything about what exactly the new GS  series is from the announcement.  E I hear there was a press packet for the folks in New York. Nothing inuK Chicago except a one page letter from Capellas that can be found on the weby> site.We did get lunch, the highlight being the Chocolate cake!  H To answer your question, entry level GS80s start under $100K [I was justI thinking today that the breakthrough PDP-5 and PDP-8 series was the firstsI computer under $100K], and the typical system config (without storage etc L IIRC) is around $400-600K. First out will be the GS160 model 16 in June. TheL GS160 model 8 will ship in July, followed byt the GS320 and finally the GS80; in late Q3. They expect to ship $1B by the end of the year.   G A few numbers: The GS Series is expected to have 5-6 year life with 24XaG performance growth: 8X from expanding from 32 to 256 CPUs, and a 3X CPU G improvement. For comparison the existing TurboLaser has seen a 3.6X CPUeK speedup (plus whatever the EV67 upgrade gives), but no CPU increase in justr
 over 4 years.t  K Oracle announced a record 150K TPS benchmark that will be published as soonfL as it is audited. I don't know if it was the same benchmark, but they quoted 11,200 simultaneous users.  I [The web page with VAX and Alpha performance numbers has moved yet again,iJ and still has nothing for any EV67 system inclusing the new GS boxes. What, is there is almost a year old]                H Capellas stated that the analysts are wrong, and server consolidation isL real. WildFire is expected to be a big player in this market. Early adoptersF of E-business have hit performance walls and need this system and it's! "unprecedented growth potential".e  K Capellas did mention VMS several times. In looking at the list of customers"F on the GS web page, 8 are VMS and 9 are Unix. Sadly, ours will be UNIX replacing VMS :-(i  G Bill Heil was asked about the partitioning capability, and answered the J question for Unix, and had to be reminded that OpenVMS Galaxy had software partitioning too.   I Gary Bloom, VP of Oracle was asked whether all the systems he was talkingoK about were Compaq. He admitted that internally Oracle uses a mix of Compaq, 3 HP, and Sun, and that they are pretty evenly split.i  K There was one humorous comment, from Gary: What is the difference between asL traditional company and a ".COM" company? Traditional companies have revenue
 and earnings.e    F Tasteless question of the week:  I wonder if Compaq will be shipping a6 WildFire to Los Alamos any time in the near future :-)     	Bob Kaplow	  E SPAM:	spamrecycle@ChooseYourmail.com	uce@ftc.gov	postmaster@127.0.0.1    ------------------------------  # Date: Wed, 17 May 2000 19:59:48 GMTa4 From: kaplow_r@eisner.decus.org.spamnot (Bob Kaplow)" Subject: Re: Wildfire Announcement+ Message-ID: <$qJdk7EgRRGs@eisner.decus.org>,  a In article <3921b59a.725415552@News.Service.Emory.Edu>, kdbeasl@emory.edu (Kevin Beasley) writes:iH > 2)  The GS80 won't be available until late Q3, but the GS160 and GS320H > will be shipping in June.  The price range looks like $100K - $1M (US, > and very roughly).  7 I heard only the GS160 model 16 will ship this quarter.t  @ > 5) There were many more Compaq employees there than customers.  L Yup, especially if you discount the VARs that were present. As an ex-DECcie, it was liek a reunion.  D > 6) The satellite feed stayed up through the entire program and the > food was great.o  K At least in Chicago, there were a couple audio drops. Probably a 747 flying E over the dish :-) But not bad. The Chocolate cake at the Ritz was the  definite highlight.s   	Bob Kaplow	  E SPAM:	spamrecycle@ChooseYourmail.com	uce@ftc.gov	postmaster@127.0.0.1M   ------------------------------  % Date: Wed, 17 May 2000 15:07:49 -0400t- From: JF Mezei <jfmezei.spamnot@videotron.ca> " Subject: Re: Wildfire Announcement, Message-ID: <3922EE03.B40916EF@videotron.ca>  ! norm.raphael@jamesbury.com wrote: D > swithcing backplane (with a bigger cabinet, IIRC) is required. TheG > existing quad modules can be moved, however.  Another nice feature isgF > that in different quads different versions of the CPU can run so theI > processors in any existing quads do not have to be swapped, but can be.0  M Another aspect of Capellas missing an opportunity. While he did allude to the?N 20-fold increase in power and how existing wildfire customers would be able toM benefit from that, he did not give it in eloquent wording such as yours whichi0 makes it obvious to the press what it all means.  J Also, when asked if this would result in lwoer sales, Capellas should haveM been quote direct: "Wildfire computers are not disposable machines. By making?D these machines fully upgradable for quite some time, our customers'sD investments will last a long time, and we will be building long term" relationships with our customers."  L (or include some other wording to let the media know that Compaq's customersK are less likely to switch vendors because Compaq protects their investments M instead of making the machises obsolete as fast as possible. Remember that aseL an enterprise computing company, our revenus do not come solely from sellingJ as many boxes as possible, but servicing and maintaining our customers' ITM infrastructure is a major part of our business, so looking only at the numberoD of units shipped doesn't provide an indication of our profitability.   ------------------------------  % Date: Wed, 17 May 2000 15:29:57 -0400s- From: JF Mezei <jfmezei.spamnot@videotron.ca>t" Subject: Re: Wildfire Announcement, Message-ID: <3922F332.A3B4B387@videotron.ca>   Art Rice wrote:eH > Recently I have heard many lay people decribe HP9000 and SUN E10000 asH > mainframes.  They are not but under the really old definition of beingE > the "main" computer within the corporation you can see where peoplee > make that mistake.  O Was the VAX9000 a mainframe ? If so, wouldn't the GS series be mainframes too ?u  G > The article does state that Linux would run natively on the S/390 andnH > then goes on to describe them as "virtual" servers (presumably running
 > on MVS?)  I I doubt very much that LINUX could run as a task in MVS. It would also betL quite interesting to see how IBM would have implemented a terminal interfaceM to LINUX since the 390 is so "3270" oriented (block mode instead of character1 cell mode).:  M My guess is that they are using the 390's partitioning abilities which also aeJ CPU to be shared amongst multiple OS instances. (think of it as a hardware implementation of VM).  F > On point number two I agree totally.  The Linux instances should fit6 > right in with the UNIX average of 35 hours downtime.  I Not so sure. It all depends on how much effort IBM will put on the devicer drivers etc etc. e  H Considering that the press only knows about LINUX (and has no idea aboutK True64), I have a feeling that the IBM announcement will make a much bigger G splash because it will feture the word "LINUX" which gives the media anh instant orgasm.t  E Here is a question: Could Compaq  legally and ethically port its unixeJ clustering to Linux but only on ALPHA linux ? If so, perhaps Compaq shouldL have really announced it was going to port TRue64 to Linux and from then on,N only has VMS as a proprietary for the robustness, and sell Linux for midrange,L and Microsoft softwrae for those who like to practice their <atl><ctrl><del> typing skills.    L I think that Compaq's wildfire announcement could have had more media weightN if they had featured Linux more prominently in it. Just as "open systems" wereK the "soupe du jour" in Palmer's days, "LINUX" is the current soupe du jour.n   ------------------------------  % Date: Wed, 17 May 2000 16:27:32 -0400o  From: norm.raphael@jamesbury.com" Subject: Re: Wildfire Announcement4 Message-ID: <C22568E2.006F8C7D.00@jklh21.valmet.com>   /m   /To:  Info-VAX@mvb.saic.coma /cc:# /Subject: Re: Wildfire Announcementi /u /yH /I must say I was overall disappointed at the announcement. Gon't get meJ /wrong, I'm excited about WildFire, and my employer is one of the 237 withK /orders already in, with delivery next month. But the two hour announcementaH /was remarkably "content free" IMHO. The technical content wasjust aboutM /ZERO. If I hadn't known about Wildfire for over 2.5 years now, read "ShannonlI /Knows Compaq", and picked up other tidbits over the past couple years, I0G /certainly wouldn't have learned anything about what exactly the new GSu! /series is from the announcement.s /-  L I agree completely. Good numbers on (x24) over life, but no appreciation for! (technically) what it can do now.S   /[SNIP]  /iH /Bill Heil was asked about the partitioning capability, and answered theK /question for Unix, and had to be reminded that OpenVMS Galaxy had softwareb /partitioning too. / L Here, too.  They seemed to know that it supported OpenVMS, but seemed not to) know what OpenVMS was or what it was for.r   /[SNIP]a /C /    Bob Kaplowr /h9 /SPAM:    spamrecycle@ChooseYourmail.com      uce@ftc.gov( postmaster@127.0.0.1 /s   ------------------------------  # Date: Wed, 17 May 2000 23:44:48 GMTa0 From: "Terry C. Shannon" <shannon@world.std.com>" Subject: Re: Wildfire Announcement& Message-ID: <Fuq9yw.A7B@world.std.com>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3922F332.A3B4B387@videotron.ca... > Art Rice wrote:oJ > > Recently I have heard many lay people decribe HP9000 and SUN E10000 asJ > > mainframes.  They are not but under the really old definition of beingG > > the "main" computer within the corporation you can see where peoplel > > make that mistake. >iK > Was the VAX9000 a mainframe ? If so, wouldn't the GS series be mainframesL too ?   H Well, Bob Glorioso called that box a "mainframe." Outside of 454 unlucky% turkeys, er, buyers, nobody else did.   K The VAX 9000 was the end result of a pissing match between Bob Glorioso andlH Bill Demmer. Bill advocated the PRISM architecture in an SMP environmentL (which ultimately came to fruition as Calypso, which outsold the VAX 9000 byK a factor of 20 or better). Glorioso won the battle, but helped Digital loseo the war.   And so it goes...e   cheers,n  G terry "would you like vector processors with that burger, sir?" shannonr   ------------------------------  % Date: Wed, 17 May 2000 14:56:07 -0400t- From: JF Mezei <jfmezei.spamnot@videotron.ca>?F Subject: Re: Wildfire Announcement: Michael Capellas, can you say VMS?, Message-ID: <3922EB45.8970A2D7@videotron.ca>   Nigel Arnot wrote:H > I think we should wait and see how the story develops in the technicalL > press, where Compaq briefings might be read, where Compaq can legitimatelyJ > smooch the journalists, and where Compaq can reasonably influence thingsI > with its advertizing. that's where we'll first notice a change of tune,e > should there be one.  N I disagree. Those in the know, are, and have been, fully aware of wildfire forM quite some time. It is not as if wildfire is new or a surprise. Yesterday was M the official PUBLIC unveiling, aimed at financial analysts/press in New York, @ and it *should* have impressed the hell out of those wall streetH analysts/media so that they can report to the masses that Compaq has got  something new and very powerful.  K I was expecting Capellas to sell the wildfire with fireworks and a big bangvM that would have let the analysts with awe and ispiration. "Wow, Compaq really & does have something new and hot here".  I That press conference could have very well been mistaken for the officialuM unveiling of their 32 processor 8086 machine. I didn't see anything that madesN Wildfire stand out from the rest. All I heard was a repeated statement that atK the core of the internet, you'll need large fault tolerant machines. (whiche* neither True64 nor VMS are at this point).  L Capellas missed a HUGE opportunity to push the alpha architecture when askedK the question about itanium/ia64.  "While others may be struggling to removeaK the bugs from their first 64 bit chip, the Alpha chip has been around for 8nE years, spans from the desktop to the datacentre providing scalabilitydJ unmatched by any other vendor. Compaq may not have marketed its leadershipN very well but it will keep it, and the introduction of Wildfire raises the bar< quite high at a time when others are just learning to jump."  N It doesn't take an expert to come up with sentences like the above, yet, it isB as if Capellas was under orders from Andy Grove not to push Alpha.   ------------------------------  # Date: Wed, 17 May 2000 22:17:18 GMTa  From: james_e_becker@my-deja.comF Subject: Re: Wildfire Announcement: Michael Capellas, can you say VMS?) Message-ID: <8fv5ov$7kg$1@nnrp1.deja.com>n  0 In article <009EA2E9.A5D3C4EC@SendSpamHere.ORG>,    system@SendSpamHere.ORG wrote:' [much snippage for the sake of brevity]tF > Well you've gotten lots of goodies at Compaq expense but you ALREADYF > know about Galaxy, VMS and ... dare I say it for fear of a trademarkG > infringement... Wildfire!  Here was a chance for "free" publicity andoF > and Capellas blew it.  The viewers of that segment probably think itE > is just another hot box to run monopoly$chlock's schlockwarez.  CaniG > freebie key-chains make up for the lost 10 minutes of TV air time?  Io > doubt it.h  : You bet. And even though I knew what this announcement was8 going to be about, it wasn't because of the hype, it was0 only because of what amounts to scuttlebutt from4 comp.os.vms, DECUServe, and people I know at Compaq.  8 Compaq does its most fervent preaching to the converted.8 (Hmmm, I suppose that's what I'm doing in this message.): Capellas keeps acknowledging that marketing has been weak,8 but we've yet to see a strong, public message about VMS.  
 Jim Becker    & Sent via Deja.com http://www.deja.com/ Before you buy..   ------------------------------  # Date: Wed, 17 May 2000 23:51:32 GMT 0 From: "Terry C. Shannon" <shannon@world.std.com>F Subject: Re: Wildfire Announcement: Michael Capellas, can you say VMS?& Message-ID: <FuqAA4.BBH@world.std.com>  - <james_e_becker@my-deja.com> wrote in messagea# news:8fv5ov$7kg$1@nnrp1.deja.com.../2 > In article <009EA2E9.A5D3C4EC@SendSpamHere.ORG>," >   system@SendSpamHere.ORG wrote:) > [much snippage for the sake of brevity]mH > > Well you've gotten lots of goodies at Compaq expense but you ALREADYH > > know about Galaxy, VMS and ... dare I say it for fear of a trademarkI > > infringement... Wildfire!  Here was a chance for "free" publicity and H > > and Capellas blew it.  The viewers of that segment probably think itG > > is just another hot box to run monopoly$chlock's schlockwarez.  CaneI > > freebie key-chains make up for the lost 10 minutes of TV air time?  IM
 > > doubt it.o >u< > You bet. And even though I knew what this announcement was: > going to be about, it wasn't because of the hype, it was2 > only because of what amounts to scuttlebutt from6 > comp.os.vms, DECUServe, and people I know at Compaq. >.: > Compaq does its most fervent preaching to the converted.: > (Hmmm, I suppose that's what I'm doing in this message.)< > Capellas keeps acknowledging that marketing has been weak,: > but we've yet to see a strong, public message about VMS.  I Apparently you were not in NYC yesterday. Rich Marcello spent a half hour K talking to customers (nope, not the press/analyst folks, who probably can't : even spell VMS) about OpenVMS. Did a pretty good job, too.  F Oh, by the way, WildFire is NT-capable (the microcode is there, a betaK release booted, etc), but it ain't gonna happen. That was decided back last H August 23. Microsoft dearly wanted to showcase NT on a 32-way enterpriseK server, but Compaq sorta kinda got sick and tired of paying $150M per annum  in fealty to MisterBill.  K And yes, that's what the AlphaNT program cost the Q on an annualized basis.e   ------------------------------  % Date: Wed, 17 May 2000 20:44:41 -0400o- From: JF Mezei <jfmezei.spamnot@videotron.ca>wF Subject: Re: Wildfire Announcement: Michael Capellas, can you say VMS?, Message-ID: <39233CF4.1AD39927@videotron.ca>  ! james_e_becker@my-deja.com wrote: : > Compaq does its most fervent preaching to the converted.  I But the converted are very weary of the lack of visibility of VMS and areuI concerned that the lack of visibility means less applications which measnh continued decline of VMS.e  < > Capellas keeps acknowledging that marketing has been weak,: > but we've yet to see a strong, public message about VMS.    L A corporation the size of Compaq which publicly admits to a problem but doesI not take steps to correct the problem is bound to go the way of the dodo.t   ------------------------------  # Date: Thu, 18 May 2000 00:44:37 GMTP= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)cF Subject: Re: Wildfire Announcement: Michael Capellas, can you say VMS?0 Message-ID: <009EA3C9.2DEEC529@SendSpamHere.ORG>  Y In article <FuqAA4.BBH@world.std.com>, "Terry C. Shannon" <shannon@world.std.com> writes:p
 >{...snip...} J >Apparently you were not in NYC yesterday. Rich Marcello spent a half hourL >talking to customers (nope, not the press/analyst folks, who probably can't; >even spell VMS) about OpenVMS. Did a pretty good job, too.g  M Reality check.  Torquemada didn't invade the Vatican to convert the faithful!   M Compaq needs to have an Alpha and VMS inquisition to convert the billyheathenS sinners.  G >Oh, by the way, WildFire is NT-capable (the microcode is there, a betatL >release booted, etc), but it ain't gonna happen. That was decided back lastI >August 23. Microsoft dearly wanted to showcase NT on a 32-way enterpriseaL >server, but Compaq sorta kinda got sick and tired of paying $150M per annum >in fealty to MisterBill.  > L >And yes, that's what the AlphaNT program cost the Q on an annualized basis.  K Well, if they want to piss away '$150M per annum', I can bet that would buyEK a whole lot of TV air-time sound bites.  Just enough to utter the words VMSe
 and Alpha.   --N VAXman- OpenVMS APE certification number: AAA-0001           VAXman@TMESIS.COM  L GNU Freeware -- What does the GNU *really* stand for?  Garbage!  Not Usable!   ------------------------------  # Date: Thu, 18 May 2000 03:55:32 GMTt' From: "Joe Thomas" <jtonthepc@home.com>a& Subject: Windows 98 Vs. Windows NT 4.0> Message-ID: <UQJU4.58625$k5.1613325@news1.frmt1.sfba.home.com>  K What is the key differences in the Windows 98 system and the Windows NT? ItdG would just be for home use not for a business. I know it costs more butyK still. Is there any hardware you need, and is there any specifications that  need to be made?  2 If you have an Answer Please Reply To This Address jtonthepc@home.com  A (For safety reasons please type this numnber on the subject line:I
 939-877-0021)a   ------------------------------  % Date: Thu, 18 May 2000 12:27:25 +1000)/ From: "Phil Howell" <howellp@snowyhydro.com.au>r7 Subject: Re: Windows Front End to VT Based Applications-3 Message-ID: <kyIU4.35786$PL4.718414@ozemail.com.au>   K It is possible to use a terminal emulator with scripting capabilities to dorJ this, but it can be tedious to set up depending on the consistency of your application.E One package that we use on alpha/vms (but also runs on unix) suppliesiL scripts for a product called Wintegrate. In this case every screen in the VTK application has a screen id in a fixed position. The relevant client script + is triggered when the VT screen id changes.l. I believe Reflection can do something similar. Phil8 Stuart, Ed <Ed.Stuart@austinenergy.com> wrote in messageL news:CB874B506A79D1118FBC006097306B8903120218@ohms.electric.ci.austin.tx.us. ..I > Hello all;  we've been introduced to a product from Seagull that puts atD > Windows GUI front end on IBM 3270/5270 based applications.  To the	 mainframerK > or AS/400 the session looks like a telnet terminal session, to the clienteH > the application has a Windows GUI.  We also have some VMS applications thatL > are being run via a terminal emulator and would like to use a product likeL > the one from Seagull to put a Windows front end on these applications.  DoH > you know of such a product for VMS systems?  More detailed information aboutF@ > the Seagull product can be found at http://www.seagullsw.com/. >k > Ed Stuart ' > Manager, Systems and Desktop Services ! > Information Technology Services  > City of Austin, Austin Energyg > Ed.Stuart@austinenergy.com >r, > "Glittering prizes and endless compromises0 > shatter the illusion of integrity" - Neil Pert >dD > *Please apply a generous amount of all the usual disclaimers here* >e >    ------------------------------  % Date: Wed, 17 May 2000 23:42:51 -0400t% From: JM <vmswiz@geonospamcities.com>a7 Subject: Re: Windows Front End to VT Based ApplicationshO Message-ID: <486F94117A555E88.4C3DBB6189C0EC89.E2CA707ADE689E12@lp.airnews.net>s   Stuart, Ed wrote:r > I > Hello all;  we've been introduced to a product from Seagull that puts as? > Windows GUI front end on IBM 3270/5270 based applications. Do N > you know of such a product for VMS systems?  More detailed information about@ > the Seagull product can be found at http://www.seagullsw.com/.  5 Probably lower level than you are looking for, but...p  F Check out "Apptrieve" from WRQ. It's a screen-scraper-type application@ except it generates an ODBC database from the data it gathers inD real-time...you create a "data model" based on keystrokes and screenF data you want it to "scrape". Updates to data happen the same way. YouE still have to write a GUI application in your favorite PeeCee tool or H WRQ's own SuperNova 4GL, but the application code can be left unchanged.  D The WRQ folks told us that they plan to port the server component of1 this tool from NT to VMS sometime in the future. s  C WRQ is the company that makes the Reflections VT terminal emulator.    http://www.wrq.com/D  ' We've only just begun using it, YMMV...i   			*JM*    ------------------------------   End of INFO-VAX 2000.276 ************************