0 INFO-VAX	Tue, 14 Jan 2003	Volume 2003 : Issue 27      Contents:! Re: assembly/disassembly on vms ? " Re: DCPS "green-bar" look printing Digital HardwareB Re: Fibre channel:  Is the INIT/ERASE function built into the HBA? Re: Freeware VDD64 GE HCT Tool Kit  Re: GE HCT Tool Kit  Help with VAX Basic  RE: Help with VAX Basic  Re: Help with VAX Basic  RE: Help with VAX Basic  Re: Help with VAX Basic  Re: Help with VAX Basic  Re: Help with VAX Basic F How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?J Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?J Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK?= Re: HSG-80 firmware, version 8.7-1F... anybody installed yet? = Re: HSG-80 firmware, version 8.7-1F... anybody installed yet? % Incremental restore problem VMS 7.3-1  Re: Microvax Floating point . Re: Most common OpenVMS versions in use today?. RE: Most common OpenVMS versions in use today?. Re: Most common OpenVMS versions in use today?. Re: Most common OpenVMS versions in use today?. Re: Most common OpenVMS versions in use today?. Re: Most common OpenVMS versions in use today? Re: Network install of VMS Re: Network install of VMS Re: Network install of VMS Re: Newbie to OpenVMS. Re: Newbie to OpenVMS. Re: Newbie to OpenVMS. Re: Newbie to OpenVMS. RE: Press Release from Fortel & Re: READ ONLY TRANSACTIONS ARE HANGING Reading file in DCL  RE: Reading file in DCL  Re: Reading file in DCL  Re: Reading file in DCL  Re: Reading file in DCL  Re: smtp to vmsmail  Re: smtp to vmsmail $ Re: Software distribution technology Suppressing SMTP headers Re: Suppressing SMTP headers* TCP/IP 5.0a -- port number range problem ?. Re: TCP/IP 5.0a -- port number range problem ?. RE: TCP/IP 5.0a -- port number range problem ? RE: trivia curiousity question Re: trivia curiousity question Re: trivia curiousity question Re: What does RWCLU mean? + Why doesn't SET BREAK %LABEL work any more? / Re: Why doesn't SET BREAK %LABEL work any more? / Re: Why doesn't SET BREAK %LABEL work any more? / Re: Why doesn't SET BREAK %LABEL work any more? / Re: Why doesn't SET BREAK %LABEL work any more? / Re: Why doesn't SET BREAK %LABEL work any more? / Re: Why doesn't SET BREAK %LABEL work any more? / Re: Why doesn't SET BREAK %LABEL work any more? / Re: Why doesn't SET BREAK %LABEL work any more?  Re: Wireless Ethernet  Re: Wireless Ethernet  Re: Wireless Ethernet  Re: Wireless Ethernet  RE: Wireless Ethernet  Re: Wireless Ethernet  RE: Wireless Ethernet   F ----------------------------------------------------------------------   Date: 14 Jan 2003 00:03:24 GMT# From: hoff@hp.nospam (Hoff Hoffman) * Subject: Re: assembly/disassembly on vms ?* Message-ID: <avvk4c$nob$1@web1.cup.hp.com>  R In article <1c7p1vghjn2m1nkgk0daajobf4q31rq2d2@4ax.com>, Dan <dan@vrx.net> writes:7 :I can't remember, but I know there used to be a way to D :take an EXE file on VMS (vax or alpha) and get an assembly language7 :dump something like dump /asm, but that doesn't exist.  : E :barring that, is there a list of opcodes/operands for vms assembler, * :like to convert a hex dump to assembler ?    J   Please check with your local legal and contract folks before proceeding,I   to ensure proper compliance with applicable software license agreements H   and particularly with any applicable local law(s), with any applicable   international treaties, etc.  J   One of the US laws that is potentially related to this activity is DMCA:  7     http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi? 9       dbname=105_cong_public_laws&docid=f:publ304.105.pdf   K   Local laws, details of contracts, and related agreements can and do vary.   J   I am not a lawyer.  I will also not (personally) debate nor defend DMCA.  K   Within OpenVMS, the Debugger and (on OpenVMS VAX) PATCH can be used.  For J   details of using PATCH-like capabilities that are available for OpenVMS $   Alpha, please see the OpenVMS FAQ.    N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Mon, 13 Jan 2003 20:55:55 -0500 + From: John Johnstone <jj_usenet2@yahoo.com> + Subject: Re: DCPS "green-bar" look printing ( Message-ID: <3E2327DB.638E65B@yahoo.com>  C Thanks for the help.  I didn't think to try the HELP.  If the PRINT G Parameters are in the documentation, I must have missed it.  If they're $ not there, they certainly should be.     John Reagan wrote: >  > Mark E. Levy wrote:  >  > >  > > Try $HELP PRINT_Parameter  > > G > > It appears that the parameter DATA_TYPE=LIST will do what you want.  > >  > > ML > >  > >  > A > You can also make LIST the default on a file-extension basis in 8 > SYS$LIBRARY:DCPS$FILE_EXTENSION_DATA_TYPE.DAT_DEFAULT. > H > Out of the box some of the compiler source extenions (.FOR,.PAS, etc.)E > default to LIST.  I always edit this file and set them all to ANSI.  >  > --
 > John Reagan ) > Compaq Pascal/{A|I}MACRO Project Leader  > Hewlett-Packard Company    ------------------------------  # Date: Tue, 14 Jan 2003 05:42:17 GMT / From: "Robin Schipper" <robinschip@hotmail.com>  Subject: Digital Hardware . Message-ID: <ZqNU9.127230$xm4.5880395@Flipper>   Hello,  2 We have some hardware witch we don't use any more.   any one interestted?   mail me at rschipper@tribase.nl   & The hardware is in Amsterdam (Holland)   ------------------------------  % Date: Mon, 13 Jan 2003 21:11:19 -0600 @ From: "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net>K Subject: Re: Fibre channel:  Is the INIT/ERASE function built into the HBA? 6 Message-ID: <3E237FD7.EDBA353B@earthlink.spamfree.net>   Scott Vieth wrote: >  > Hi:  > D > I was at work this weekend swapping out some 9GB and 18GB SBBs forF > 36GB SBBs.  Like any good sysadmin, I did an INIT/ERASE on the disks* > before I removed them from the ESA12000. > F > The INIT/ERASE operation takes a very long time (no suprise).  I hadF > three erase operations running at one time (against three six-memberE > RAID0 sets).  I watched the load on the system with MONI SYSTEM and E > MONI MODE.  To my surprise, the ES40 was basically idle.  No direct ' > I/O.  No CPU time consumed.  Nothing. 9 > But yet the activity lights on the disks were on solid.  > F > I fired-up the GUI app for managing the f.c. switches and saw a hugeE > amount of traffic (more than 25MB/sec) being received by the switch H > port connected to the ES40 and being transmitted to the port connected > to the HSG80.  > F > This is very odd:  The system is basically "dead".  MONI MODE showedG > an idle time of 397 (out of 400). Yet the HBA in the ES40 was sending  > traffic like mad.  > E > My question:  Is the "ERASE" part of INIT/ERASE built into the HBA? D > Does the OS tell the HBA to "Go erase that bad boy and let me know > when you are done."?  B Based on input from past postings, I'm inclined to believe that itF invokes a $QIO which causes some things to happen at the lower layers,3 tho I'm not quite sure what magic is afoot there...    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------   Date: 14 Jan 2003 01:09:43 GMT# From: hoff@hp.nospam (Hoff Hoffman)  Subject: Re: Freeware VDD64 * Message-ID: <avvo0n$nob$2@web1.cup.hp.com>  a In article <3E1EDEC2.8000109@maury-imprimeur.fr>, Tim Oakley <toakley@maury-imprimeur.fr> writes:   G :Have just updated to v7.3 and our optical disks which used to use the  J :PERCPETICS supplied 'WDD' drivers (OSMS i think) no longer work (in fact 2 :the system crashes when we load the WDD drivers). :  :HP France told me that... :either J :purchase new drivers from a company named USDESIGN who tookover the OSMS J :software from Compaq (this seems fair, but USDESIGN tell me that the new E :drivers will allow me to read data from my optical drives but *not*  5 :write to them (sounds like a Microsoft enhancement))   J   This compatability model is fairly common in the storage hardware space.H   The ability to read the media from older drives permits you to access I   older archives, while the removal of the write-related hardware and the H   associated logic for the older widget simplifies (and reduces the costI   and the risks to) the write-path of the device.  Various generations of G   DLT tape drives have had similar upward read media compatability, for G   instance, and the new Iomega Zip750 drives (are both read- and write- I   compatible with the previous-generation Zip250 drives and) can read but *   cannot write to the oldest Zip100 media.   :or J :try using the freeware VDD64 package which should allow me to access the " :optical disks as i did with OSMS. : J :Having downloaded the VDD package and read the help/user doc i don't see  :how VDD is going to help me.  : I :Has anyone used the freeware 'VDD64' (virtual device drivers) for their  F :optical drives or has anyone got a different solution to my problem ?      G   AFAIK, HP France is either confused, or some amount of information in C   the discussion was lost in the transition and/or the translation.   H   As you expect, VD64 and LD implement virtual disks, and these are not F   particularly useful for accessing foreign devices.  These mechanismsE   are usful for creating the master data copies that are then used as D   the source for block-oriented write operations to recordable disk I   media, particularly to CD disks -- this may have been either the source B   of the confusion, or potentially the intent of the suggestion.    G   If you can write to the RWZ-series devices using a block-oriented I/O J   sequential write mode (implemented with, for instance, a series of $qio I   or $qiow IO$_WRITELBLK calls), this suggestion from HP France does make F   good sense and will (probably) work.  (I'm not familiar with the RWZD   series device control interface, nor with the WDD device drivers.)  I   Current media for this task would be DVD-R, DVD+R, DVD-RW, DVD+RW, etc.   F   I have passed a reference to to the comments on the contents of the E   OpenVMS SPD (located elsewhere in the thread) along to the OpenVMS     Business Manager.     N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Mon, 13 Jan 2003 14:31:31 -0600 0 From: Earl Lakia <Lakia_n_o__s_p_a_m_@ipact.com> Subject: GE HCT Tool Kit1 Message-ID: <G8KdnW1IAqnwvL6jXTWcpQ@netnitco.net>   4 I have two customers using the HCT tool kit from GE.6 This product creates a socket connection to a GE Fanuc
 90/70 PLC.  1 After monitoring the TCP/IP traffic to the target 3 90/70 PLC, the number of calls are normal (a couple . a second).  However, the toolkit ends up doing- 4000 buffered I/O/second at one site and 6000 * buffered I/O/second at the other.  I can't2 figure out what it's doing under the sheets.  Also eats a lot of CPU resources.  3 Anyone else used this product? Anyone have a better , solution?  I can only assume that it must be. reading and writing to a mailbox or something.   -earl    ------------------------------   Date: 14 Jan 2003 01:41:38 GMT# From: hoff@hp.nospam (Hoff Hoffman)  Subject: Re: GE HCT Tool Kit* Message-ID: <avvpsi$nob$3@web1.cup.hp.com>  d In article <G8KdnW1IAqnwvL6jXTWcpQ@netnitco.net>, Earl Lakia <Lakia_n_o__s_p_a_m_@ipact.com> writes:5 :I have two customers using the HCT tool kit from GE. 7 :This product creates a socket connection to a GE Fanuc  :90/70 PLC.  : 2 :After monitoring the TCP/IP traffic to the target4 :90/70 PLC, the number of calls are normal (a couple/ :a second).  However, the toolkit ends up doing . :4000 buffered I/O/second at one site and 6000+ :buffered I/O/second at the other.  I can't 3 :figure out what it's doing under the sheets.  Also  :eats a lot of CPU resources.     C   This does imply the software has either lost its place; that the  F   host software has effectively become unsynchronized with the currentF   PLC communications and with the PLC data stream.  This situation is B   historically not particularly unusual when working with various #   models of PLCs, in my experience.   D   I don't specifically recall the 90/70, nor the GE Fanuc HCT stuff.    4 :Anyone else used this product? Anyone have a better- :solution?  I can only assume that it must be / :reading and writing to a mailbox or something.   D   Its been a few years since I've looked at similar configurations, E   but at least the older PLCs have traditionally had some of the most C   bizarre serial communications interfaces I can recall having ever E   seen -- schemes based on single-character I/O and home-grown serial H   protocols are not unusual.  Flow control (XON/XOFF in-band or hardwareG   flow control) and other common communications concepts (at least back G   when I was working regularly with PLCs) rather foreign to some of the G   PLC designers.  At least the first-generation network interfaces were G   home-grown network stacks and home-grown protocols, a situation which 3   did not bode well for future generations of PLCs.   C   The host software was accordingly, well, really crufty -- one PLC F   hardware engineer told me -- and entirely in earnest -- to stall theE   host system in a loop in the device driver, waiting for a character E   from the PLC by polling the device each pass through the loop.  And B   at least one was known to provide huge piles of data or a singleC   character, seemingly at a whim -- there was no way to predict, at G   least until you started seeing characters arrive.  And to keep it all B   interesting, various (older) PLCs were known to become entirely 0   catatonic on any incident of a protocol error.  D   I do hope and do expect that the newer PLCs and newer PLC softwareF   has improved.  (Factories tend to invest in a pile of similar PLCs, H   and will then keep the PLCs in operation for as long as the particularE   operation the PLCs are controlling is itself running -- and worse,  F   short of a direct lightning hit, some of the oldest and the worst ofF   the PLCs just don't die.   PLC upgrades and particularly wholely newG   PLCs are arguably thus somewhat unusual and somewhat of a luxury. :-)   H   If I were in your situation, I would check with the GE Fanuc folks andG   see if this is a known problem.  (I do not know if folks that support E   this software are still based in the Corporate Woods Office Park in E   Albany, NY, as it's been a few years.  I do not know if GE Fanuc is 6   presently supporting this PLC or this HCT, as well.)    N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------    Date: 13 Jan 2003 14:08:30 -08002 From: vibroplex@mindspring.com (Derek Cohn/WB0TUA) Subject: Help with VAX Basic< Message-ID: <d7d0c297.0301131408.f31e17a@posting.google.com>  
 Dear Friends,   B Believe it or not, there still is at least one application runningE that is written in VAX Basic.  I have recently learned a few commands C to support this application and I have a question.  I don't know if 3 there are any VAX Basic experts still out there....   @ I have written a program that accepts an input using the "input"B command.  When I run the executable image from the dollar sign, itE correctly asks for input and processes it accordingly.  However, when C I call my executable image from a DCL, the input command fails with  the following error:  ' %BAS-F-ENDFILDEV, End of file on device ? -BAS-I-ON_CHAFIL, on channel 0 for file SYS$INPUT:.; at user PC  00000000  -RMS-E-EOF, end of file detected -BAS-I-FROMOD, In module DEREK@ break on unhandled exception preceding SHARE$LIBRTL_CODE0+274244; %DEBUG-I-SOURCESCOPE, Source lines not available for .0\%PC <         Displaying source in a caller of the current routine  ) Here is the relevant portion of the code:   5 declare string action                                 5                                                       4 input "Do you want to Retrieve or Set [R,S]"; action  , I checked the VMS BASIC manual and saw this:   Remarks @ If you do not specify a channel, the default chnl-exp is #0 (theF controlling terminal). If a chnl-exp is specified, a file must be openE on that channel with ACCESS READ or MODIFY before the INPUT statement  can execute.  C However, I'm at a loss how to make this work.  Anyone have any idea > how to have an executable image running in a DCL accept input?  $ Thanks for any help you can give me!   Derek    ------------------------------  % Date: Mon, 13 Jan 2003 14:14:35 -0800 $ From: Shane Smith <ssmith@icius.com>  Subject: RE: Help with VAX Basic0 Message-ID: <01C2BB0E.2E1C5460@sulfer.icius.com>  H Believe me, there's a lot more than one BASIC app put there. What you'reG experiencing isn't a BASIC issue, it's how DCL works. When your command E file runs, it's accepting input from its command stream, SYS$COMMAND, F which is taking input from the command file itself. You need to divert% that to take input from the keyboard:   ) $ DEFINE/NOLOG/USER SYS$COMMAND SYS$INPUT  $ RUN whatever_you_want_to_run  H The /USER specifies a user-mode logical, which will be deleted when yourC executable finishes. This puts the original process logical back in 7 control, and input's taken form the command file again.   E Incidentally, if anyone can explain to me why the qualifier is /USER, F I'd love to know. DCL commands and qualifiers are usually descriptive, and this doesn't seem to be.   Shane    -----Original Message-----@ From: vibroplex@mindspring.com [mailto:vibroplex@mindspring.com]& Sent: Monday, January 13, 2003 2:09 PM To: Info-VAX@Mvb.Saic.Com  Subject: Help with VAX Basic    
 Dear Friends,   B Believe it or not, there still is at least one application runningE that is written in VAX Basic.  I have recently learned a few commands C to support this application and I have a question.  I don't know if 3 there are any VAX Basic experts still out there....   @ I have written a program that accepts an input using the "input"B command.  When I run the executable image from the dollar sign, itE correctly asks for input and processes it accordingly.  However, when C I call my executable image from a DCL, the input command fails with  the following error:  ' %BAS-F-ENDFILDEV, End of file on device ? -BAS-I-ON_CHAFIL, on channel 0 for file SYS$INPUT:.; at user PC  00000000  -RMS-E-EOF, end of file detected -BAS-I-FROMOD, In module DEREK@ break on unhandled exception preceding SHARE$LIBRTL_CODE0+274244; %DEBUG-I-SOURCESCOPE, Source lines not available for .0\%PC <         Displaying source in a caller of the current routine  ) Here is the relevant portion of the code:   5 declare string action                                 5                                                       4 input "Do you want to Retrieve or Set [R,S]"; action  , I checked the VMS BASIC manual and saw this:   Remarks @ If you do not specify a channel, the default chnl-exp is #0 (theF controlling terminal). If a chnl-exp is specified, a file must be openE on that channel with ACCESS READ or MODIFY before the INPUT statement  can execute.  C However, I'm at a loss how to make this work.  Anyone have any idea > how to have an executable image running in a DCL accept input?  $ Thanks for any help you can give me!   Derek    ------------------------------  % Date: Mon, 13 Jan 2003 23:29:02 +0100 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>   Subject: Re: Help with VAX Basic' Message-ID: <3E233DAE.94803E41@aaa.com>   ? With "run in a DCL" you mean a "run in a command file", right ?   * Anyway, your program reads from SYS$INPUT.  1 When you run your program "by hand", SYS$INPUT is  your terminal session.  8 When you run your program from a command file, SYS$INPUT is the command file itself.   = By defining SYS$INPUT to be SYS$COMMAND temporarily, your can C get the prompt at your terminal, even when run from a command file.    Something like :    $ DEF/USER SYS$INPUT SYS$COMMAND $ RUN <your-prog>   ? *Or* you can just put the answer directly in the command file :    $ RUN <your-prog>  R  $   9 But then, you must know the answer beforehand, of course.   ; This is nothing specific for VAX-Basic, it's DCL-basics :-)    Jan-Erik Sderholm.    Derek Cohn/WB0TUA wrote: >  > Dear Friends,  > B > I have written a program that accepts an input using the "input"D > command.  When I run the executable image from the dollar sign, itG > correctly asks for input and processes it accordingly.  However, when E > I call my executable image from a DCL, the input command fails with  > the following error: >    ------------------------------  % Date: Mon, 13 Jan 2003 14:46:48 -0800 $ From: Shane Smith <ssmith@icius.com>  Subject: RE: Help with VAX Basic0 Message-ID: <01C2BB12.9ABABD10@sulfer.icius.com>  F Jan-Eric is of course right. I got the SYS$COMMAND and SYS$INPUT mixedE up. Sorry for the red herring - really must notice next time that the . example I pulled up used ASSIGN, not DEFINE...   Shane    -----Original Message-----/ From: Jan-Erik S=F6derholm [mailto:aaa@aaa.com] & Sent: Monday, January 13, 2003 2:29 PM To: Info-VAX@Mvb.Saic.Com   Subject: Re: Help with VAX Basic    ? With "run in a DCL" you mean a "run in a command file", right ?   * Anyway, your program reads from SYS$INPUT.  1 When you run your program "by hand", SYS$INPUT is  your terminal session.  8 When you run your program from a command file, SYS$INPUT is the command file itself.   = By defining SYS$INPUT to be SYS$COMMAND temporarily, your can C get the prompt at your terminal, even when run from a command file.    Something like :    $ DEF/USER SYS$INPUT SYS$COMMAND $ RUN <your-prog>   ? *Or* you can just put the answer directly in the command file :    $ RUN <your-prog>  R  $   9 But then, you must know the answer beforehand, of course.   ; This is nothing specific for VAX-Basic, it's DCL-basics :-)    Jan-Erik S=F6derholm.    Derek Cohn/WB0TUA wrote: >=20 > Dear Friends,  >=20B > I have written a program that accepts an input using the "input"D > command.  When I run the executable image from the dollar sign, itG > correctly asks for input and processes it accordingly.  However, when E > I call my executable image from a DCL, the input command fails with  > the following error: >    ------------------------------  % Date: Mon, 13 Jan 2003 23:32:57 +0100 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>   Subject: Re: Help with VAX Basic' Message-ID: <3E233E99.BB8B9204@aaa.com>    The (full) options are realy :  4 /USER_MODE, the /SUPERVISOR_MODE, or /EXECUTIVE_MODE   which may make it clearer...  G (And, yes, I mixed up SYS$INPUT and SYS$COMMAND in my former answer...)    Jan-Erik Sderholm.    Shane Smith wrote: > G > Incidentally, if anyone can explain to me why the qualifier is /USER, H > I'd love to know. DCL commands and qualifiers are usually descriptive, > and this doesn't seem to be. >  > Shane  >    ------------------------------  % Date: Tue, 14 Jan 2003 00:18:37 +0100 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>   Subject: Re: Help with VAX Basic' Message-ID: <3E23494D.9234C600@aaa.com>   > Well, if *that* is the case, *I* *realy* made a fool of myself in my last post, not :-) :-)  A Honestly, I can never remember this one correctly. I just checked @ on a live VMS system, and it's "def/user sys$input sys$command". Phuu !!   
 Jan-Erik.    Shane Smith wrote: > H > Jan-Eric is of course right. I got the SYS$COMMAND and SYS$INPUT mixedG > up. Sorry for the red herring - really must notice next time that the 0 > example I pulled up used ASSIGN, not DEFINE... >  > Shane    ------------------------------  % Date: Mon, 13 Jan 2003 22:58:19 -0500 ( From: David Froble <davef@tsoft-inc.com>  Subject: Re: Help with VAX Basic, Message-ID: <3E238ADB.3070106@tsoft-inc.com>  Q Post your DCL code.  I'm not going to make a bunch of guesses.  What you want to   do should work.   ? Plenty of VAX Basic people and applications still in existance.    Dave     Derek Cohn/WB0TUA wrote:   > Dear Friends,  > D > Believe it or not, there still is at least one application runningG > that is written in VAX Basic.  I have recently learned a few commands E > to support this application and I have a question.  I don't know if 5 > there are any VAX Basic experts still out there....  > B > I have written a program that accepts an input using the "input"D > command.  When I run the executable image from the dollar sign, itG > correctly asks for input and processes it accordingly.  However, when E > I call my executable image from a DCL, the input command fails with  > the following error: > ) > %BAS-F-ENDFILDEV, End of file on device A > -BAS-I-ON_CHAFIL, on channel 0 for file SYS$INPUT:.; at user PC 
 > 00000000" > -RMS-E-EOF, end of file detected  > -BAS-I-FROMOD, In module DEREKB > break on unhandled exception preceding SHARE$LIBRTL_CODE0+274244= > %DEBUG-I-SOURCESCOPE, Source lines not available for .0\%PC > >         Displaying source in a caller of the current routine > + > Here is the relevant portion of the code:  > 7 > declare string action                                 7 >                                                       6 > input "Do you want to Retrieve or Set [R,S]"; action > . > I checked the VMS BASIC manual and saw this: > 	 > Remarks B > If you do not specify a channel, the default chnl-exp is #0 (theH > controlling terminal). If a chnl-exp is specified, a file must be openG > on that channel with ACCESS READ or MODIFY before the INPUT statement  > can execute. > E > However, I'm at a loss how to make this work.  Anyone have any idea @ > how to have an executable image running in a DCL accept input? > & > Thanks for any help you can give me! >  > Derek  >    ------------------------------    Date: 13 Jan 2003 17:05:43 -0800$ From: rcbryan@hotmail.com (RC Bryan)O Subject: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK? < Message-ID: <fbcf38dc.0301131705.b838357@posting.google.com>  D We have an existing application that uses QIOW to write data to disk as follows:       vmsCondCode = sys$qiow ( 2                     0,                   // no EFN=                     nChannel,            // Channel to device ?                     IO$_WRITEVBLK,       // write function code <                     Iosb,                // I/O status blockE                     0,                   // no AST completion routine =                     0,                   // no AST parameters <                     bufferAddr,          // P1 user's buffer?                     length,              // P2 length of buffer >                     StartPage,           // P3 = Record number4                     0,0,0);              // No P4-P6  E I cannot find a hard number anywhere that says what the max value can B be for length.  The program originally limited the I/O to size 32K7 chunks but I discovered that I can write bigger pieces.   + Is a limit of 64K reasonable for V7.1 on?   % Can I go with a much greater limit?   8 Is this documented anywhere that I have not found yet?  C Am I going to discover that my new SCSI disks can run fine with big 2 blocks but some old disk is going to have trouble?   Thanks in advance,	 /RC Bryan    ------------------------------  % Date: Mon, 13 Jan 2003 22:57:49 -0500 ' From: Stuart Fuller <stufuller@usa.net> S Subject: Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK? 0 Message-ID: <ur100b.i6f.ln@dadsys2.fuller.local>   RC Bryan wrote:   F > We have an existing application that uses QIOW to write data to disk
 > as follows:  >  >    vmsCondCode = sys$qiow ( 4 >                     0,                   // no EFN? >                     nChannel,            // Channel to device A >                     IO$_WRITEVBLK,       // write function code > >                     Iosb,                // I/O status blockG >                     0,                   // no AST completion routine ? >                     0,                   // no AST parameters > >                     bufferAddr,          // P1 user's bufferA >                     length,              // P2 length of buffer @ >                     StartPage,           // P3 = Record number6 >                     0,0,0);              // No P4-P6 > G > I cannot find a hard number anywhere that says what the max value can D > be for length.  The program originally limited the I/O to size 32K9 > chunks but I discovered that I can write bigger pieces.  > + > Is a limit of 64K reasonable for V7.1 on? % > Can I go with a much greater limit? 8 > Is this documented anywhere that I have not found yet?E > Am I going to discover that my new SCSI disks can run fine with big 4 > blocks but some old disk is going to have trouble?  L The I/O User's Guide documents the max I/O sizes for a variety of devices.  ( See section 2: the disk drivers section.   --             Stu    ------------------------------  % Date: Mon, 13 Jan 2003 23:34:18 -0500 ( From: David Froble <davef@tsoft-inc.com>S Subject: Re: How much disk I/O can I do with a sys$QIOW IO$_WRITEVBLK/IO$_READVBLK? , Message-ID: <3E23934A.5090902@tsoft-inc.com>   RC Bryan wrote:   F > We have an existing application that uses QIOW to write data to disk
 > as follows:  >  >    vmsCondCode = sys$qiow ( 4 >                     0,                   // no EFN? >                     nChannel,            // Channel to device A >                     IO$_WRITEVBLK,       // write function code > >                     Iosb,                // I/O status blockG >                     0,                   // no AST completion routine ? >                     0,                   // no AST parameters > >                     bufferAddr,          // P1 user's bufferA >                     length,              // P2 length of buffer @ >                     StartPage,           // P3 = Record number6 >                     0,0,0);              // No P4-P6    N I think the 'length' is an unsigned 16 bit integer, thus 65535 bytes.  I know J that when writing disk blocks, the limit is 127 blocks to a write or read.  
 print 127*512    65024  Ready   
 print 128*512    65536   ) The math seems to agree with that memory.   P Refering to the 'gray wall', the length will be rounded up to the next multiple K of 512, and 127 blocks is the largest transfer without exceeding 65535.  I  O couldn't find a specific statement about this, but the length part of a string  Q descriptor is also a 16 bit integer, and this seems to be a hard upper limit for  Q strings, buffers, and such on VMS.  Seems is the key word here, since I couldn't  ( find, in 5 minutes, anything definitive.  6 Note that quotas and such will affect what you can do.   Sorry I couldn't be more help.   Dave    G > I cannot find a hard number anywhere that says what the max value can D > be for length.  The program originally limited the I/O to size 32K9 > chunks but I discovered that I can write bigger pieces.  > - > Is a limit of 64K reasonable for V7.1 on?   ' > Can I go with a much greater limit?   : > Is this documented anywhere that I have not found yet?  E > Am I going to discover that my new SCSI disks can run fine with big 4 > blocks but some old disk is going to have trouble? >  > Thanks in advance, > /RC Bryan  >    ------------------------------  % Date: Mon, 13 Jan 2003 14:47:51 -0600 ( From: David Harrold <DHarrold@wi.rr.com>F Subject: Re: HSG-80 firmware, version 8.7-1F... anybody installed yet?8 Message-ID: <5b962vg5plom2fimkd4t7q7ou3gmhirocj@4ax.com>  M On Wed, 08 Jan 2003 09:39:08 -0600, David Harrold <DHarrold@wi.rr.com> wrote:   < >On Tue, 7 Jan 2003 15:55:20 -0500, koskaj@bender.com wrote: > N >>I was wondering if anyone out there has installed/upgraded to HSG-80 8.7-1F 8 >>firmware, and if there have been any issues with such. >> X >e   [snip]  N >I was also told that the documentation was required to correctly install this	 >release.e >a, >So, I would wait a while before installing. >n  L Well, I talked to one of my engineers late last week and he said part of theN issue was removing support for older FC HBAs, i.e the KGPSA-Bx adapters are noN longer support with ACS V8.7-1 (all variants).  I can't remember know what the/ other things were, but that's a big one for me.    Dave Harrold    N ..............................................................................N David Harrold                              E-Mail: David_Harrold at aurora.orgI Sr. Software Systems Engineer              Phone:          (414) 647-62048I                                            Pager:          (414) 941-4634yG Aurora Health Care                         Fax:          (414) 647-4999x 3031 W. Montana Street Milwaukee, WI 53215s   ------------------------------  # Date: Mon, 13 Jan 2003 23:09:35 GMTo! From: rob.buxton@wcc.spam.govt.nzoF Subject: Re: HSG-80 firmware, version 8.7-1F... anybody installed yet?% Message-ID: <3e234003.704682710@news>c  E The new release just arrived on my desk. like you, no instructions tow	 speak of.   ; On Thu, 9 Jan 2003 11:32:12 -0500, koskaj@bender.com wrote:   B >>From:	SMTP%"rob.buxton@wcc.spam.govt.nz"  8-JAN-2003 20:11:02.65 >>To:	Info-VAX@Mvb.Saic.Com  >>CC:	E >>Subj:	Re: HSG-80 firmware, version 8.7-1F... anybody installed yet?i >>= >>On Tue, 7 Jan 2003 15:55:20 -0500, koskaj@bender.com wrote:d >>O >>>I was wondering if anyone out there has installed/upgraded to HSG-80 8.7-1F o9 >>>firmware, and if there have been any issues with such.P >>>  >>P >>We're running 8.7F-0 on our SAN that has NT Servers attached with no problems.8 >>I've not seen 8.7F-1 which would be a Patched variant.E >>I've just checked the HP web Site and the patches there are for 8.6vG >>only and stop at 8.6n-8 which is not the latest as we were on 8.6F-10  >>before the upgrade.  >>D >>If there's an 8.7-1F as opposed to 8.7F-1 then I'm even more lost! >. >. >. >nQ >Well, on my firmware cards it reads HSG80 V8.71F SWKS ACS.  And on the shipping AM >package, it shows QB-6BUAA-SF for a part number and 871F for revision.  The  B >description field on shipping label also seems to indicate 8.71F. > L >Granted, if I look from the consoles of my HSG-80's, they currently report  >V87F-0. > K >I think I may try to upgrade on a test system from the 8.7F or V87F-0, to  M >V8.7-1F with the cards that I received and see what happens.  If I hose it, -J >well... what is test for anyway, along with service maintenance contract. ><grin>  >0 >:) jck  >John Koska. >Matthew Bender & Co., Inc.c! >A Member of the LexisNexis Group?   ------------------------------    Date: 13 Jan 2003 18:57:06 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett)>. Subject: Incremental restore problem VMS 7.3-1- Message-ID: <tL9QD11bohMz@malvm7.mala.bc.ca.>a  1 I seem to be having a strange problem with backuph
 on VMS 7.3-1.a  4  I restored an image backup to a disk, then tried to9 restore an incremental backup ( made with /since=backup )c7 on top of it. It all appears to have worked except that 8 directories which existed when the image backup was made6 but not when the incremental backup was done appear to7 have not been deleted ( at least not all of them ). Forl8 the directories that don't delete I get an error message6 "error deleting disk:[aaaaa]bbbb.dir;1xxxxxxxxxxx file: not found" (where the trailing xxxx... appears to be parts7 of longer filenames which presumably were previously inc4 the buffer ). This is followed by a message with the6 proper directory file name and a "directory not empty" error.  !  Anyone else seen this behaviour?r   ------------------------------    Date: 13 Jan 2003 16:11:13 -08007 From: jones.computer.srv@worldnet.att.net (Daryl Jones)u$ Subject: Re: Microvax Floating point= Message-ID: <8a646952.0301131611.183a746b@posting.google.com>   g JF Mezei <jfmezei.spamnot@vl.videotron.ca> wrote in message news:<3E206A2C.6CB41F78@vl.videotron.ca>...a > Theroretical question. > J > Floating point operations on Microvax machines are emulated by software, > correct ?- > K > *conceptually* and forgetting any need to be compatible with other vaxes, N > wouldn't it be possible to just change the shareable images (it is uvmrtl orL > something like that ?) so that they would use IEEE format floating point ? > O > Are are there still hardware limitations that truly tie the microvaxen to thet > vax floating point format ?R  	 JF Mezei,e   For the MVII and CVAX Systems:  E F,D, and G floating was implemented in the Floating Point Chip. The H & floating was implemented in Macrocode.   Daryl Jones    ------------------------------    Date: 13 Jan 2003 13:05:35 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 7 Subject: Re: Most common OpenVMS versions in use today?e3 Message-ID: <d887NnA8zT6u@eisner.encompasserve.org>O  c In article <3E22CFC3.417598AE@aaa.com>, Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> writes:  > What does "is game" mean ?	 > Thanks.l > Jan-Erik Sderholm.N >  >  > Bob Koehler wrote: >> oN >>    On VAXen I can tell you we've shut down our VMS 2.5 systems and upgradedG >>    our VMS 4.x systems to at least 5.x.  Anything after about 5.2 isn) >>    game.  On Alpha anything after 6.0.c  @   It means that if I look around a bit I can probably find it in   a production system.   ------------------------------  % Date: Mon, 13 Jan 2003 11:17:25 -0800 # From: "Tom Linden" <tom@kednos.com>a7 Subject: RE: Most common OpenVMS versions in use today?-9 Message-ID: <CIEJLCMNHNNDLLOOGNJIGEIBGGAA.tom@kednos.com>d   >-----Original Message-----rC >From: Bob Koehler [mailto:koehler@eisner.nospam.encompasserve.org]f( >Sent: Monday, January 13, 2003 11:06 AM >To: Info-VAX@Mvb.Saic.Com8 >Subject: Re: Most common OpenVMS versions in use today? >  >w1 >In article <3E22CFC3.417598AE@aaa.com>, Jan-Eriks3 >=?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> writes:I >> What does "is game" mean ?e
 >> Thanks. >> Jan-Erik Sderholm. >> >> >> Bob Koehler wrote:t >>>aB >>>    On VAXen I can tell you we've shut down our VMS 2.5 systems
 >and upgraded(H >>>    our VMS 4.x systems to at least 5.x.  Anything after about 5.2 is* >>>    game.  On Alpha anything after 6.0. >pA >  It means that if I look around a bit I can probably find it in  >  a production system.u  J It comes from the expression, "... is fair game" which probably relates to hunting.   >g >---' >Incoming mail is certified Virus Free.o; >Checked by AVG anti-virus system (http://www.grisoft.com). B >Version: 6.0.435 / Virus Database: 244 - Release Date: 12/30/2002 >d ---o& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.435 / Virus Database: 244 - Release Date: 12/30/2002r   ------------------------------  % Date: Mon, 13 Jan 2003 20:53:24 +0100l6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>7 Subject: Re: Most common OpenVMS versions in use today? ) Message-ID: <3E231934.7020901@vajhoej.dk>i   Otto wrote:t> > Could anybody please tell us which OpenVMS versions that are2 > commonly used in production environments today?  > Both VAX and Alpha?  > OpenVMS 5.x, 6.x, 7.x.  	 My guess:    4.7   : very few 5.5-2 : 10%i 6.2   : 20%  7.1   : 10%P 7.2-X : 40%w 7.3-X : 20%/   Arne   PS: Note the word "guess" !-   ------------------------------  # Date: Mon, 13 Jan 2003 23:22:34 GMTn! From: rob.buxton@wcc.spam.govt.nze7 Subject: Re: Most common OpenVMS versions in use today?E% Message-ID: <3e23496e.707094357@news>   E I'd probably reverse the numbers for 5.5-2 and 6.2, I'd agree with an3D other Poster that suggested those that moved from 5.5-2 to 6.x wouldC probably keep going. Although saying that I know of a couple of 6.2s( Systems. And yes, mines a guess as well.  3 We're currently a 7.2 (VAX) and 7.2-2 (Alpha) shop.r  B On Mon, 13 Jan 2003 20:53:24 +0100, =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk> wrote:   >Otto wrote:? >> Could anybody please tell us which OpenVMS versions that arep3 >> commonly used in production environments today?   >> Both VAX and Alpha? >> OpenVMS 5.x, 6.x, 7.x.m >l
 >My guess: >o >4.7   : very fewr >5.5-2 : 10% >6.2   : 20% >7.1   : 10% >7.2-X : 40% >7.3-X : 20% >o >Arnec >  >PS: Note the word "guess" ! >o   ------------------------------    Date: 13 Jan 2003 16:44:27 -0800, From: ian.burgess@start.com.au (Ian Burgess)7 Subject: Re: Most common OpenVMS versions in use today?o= Message-ID: <6b63fc08.0301131644.539cda3e@posting.google.com>e  / In production on VAX,  5.5-2 and 6.2 (one each)e    On Alpha, 6.2-1H1 (just one box)"           and 7.2-1 (cluster of 5)   Ian Burgess, University of Qld   ------------------------------  + Date: Tue, 14 Jan 2003 07:57:10 +0100 (MET)i9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>s7 Subject: Re: Most common OpenVMS versions in use today?o; Message-ID: <01KR7UPU635U8WW5NX@sysdev.deutsche-boerse.com>e  , > in MY opinion. 7.3 removes functionality,   * Please elabourate and/or add smileys.  :-)   ------------------------------  % Date: Mon, 13 Jan 2003 11:46:28 -0800n" From: GreyCloud <cumulus@mist.com># Subject: Re: Network install of VMSo( Message-ID: <3E231794.F3DB6D43@mist.com>   JF Mezei wrote:s >  > Phillip Helbig wrote: M > > > In the interim, I was hoping someone could shed a little light on how I N > > > will be able to get the OS installed on the VAX once I do have the CD in > > > my hands.e > M > If you do not have a companion VMS machine nearby, your only bet is to openpP > the VLC and crimp a scsi connector onto the ribbon that leads to the drive andM > then plug in a CDROM drive temporarily. If you want to start fresh, you cang- > then intall a brand new VMS onto the drive.i >   ; Inside the 4000 vlc, there isn't a ribbon cable per se, butp a wire bundle.  + > (that is what I did for my microvax 3100)W    7 That's what I did when I upgraded the small 120 Mb hard  drive to a 2 Gb hard drive.)6 But I just went and purchased a scsi ribbon cable with9 multiple connectors on it and made sure the scsi IDs wereu3 different... in this case the original drive ID wasR; changed.  The new drive had to have the same original ID asm the old drive.  < What I'm up against is acquiring a low cost scsi CD drive to7 plug into the rear scsi port.  If that can be done then-% upgrading won't be a problem anymore.-   ------------------------------  % Date: Mon, 13 Jan 2003 15:14:16 -0400F0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca># Subject: Re: Network install of VMS-/ Message-ID: <3E230FFE.364FFC91@vl.videotron.ca>m   GreyCloud wrote:> > What I'm up against is acquiring a low cost scsi CD drive to9 > plug into the rear scsi port.  If that can be done then_' > upgrading won't be a problem anymore.   J I took a cd drive from my mac to temporarily connect it to my 3100 (inside( with that extra ribbon cable connector).  K You may wish to look at computer supply places (datapro, black box etc) foroG the adaptor cable from the vax plug to the 50 min standard scsi (older)e! connectors that most cd roms use.e   ------------------------------  % Date: Mon, 13 Jan 2003 20:04:30 -05000  From: John Santos <JOHN@egh.com># Subject: Re: Network install of VMSi4 Message-ID: <1030113195630.631A-100000@Ives.egh.com>  $ On Mon, 13 Jan 2003, JF Mezei wrote:   > Phillip Helbig wrote:nL > > > In the interim, I was hoping someone could shed a little light on how=  IL > > > will be able to get the OS installed on the VAX once I do have the CD=  inr > > > my hands.m >=20 [...]n  L > Another option would be to use kermit to transfer the individual savesets=  overcL > a serial line ( VAXVMS072.*) to the vax and do an in-situ upgrade. You'd = haveL > to do your own cleanup first, removing all the lint that may have accumul= atedL > over the years. Place the savesets in the top directory of your system dr= ive/K > [00000]. When you vms instal, when prompted for device where the media iseD > located, you enter the disk drive name followed by [000000]   (eg: > DKA200:[000000] )n >=20K > Note: if you use kermit, you may have to fixup the file attributes of thenL > received files (SET FILE/ATTTRIB, or if on an older VMS version, seek out=  theL > FILE utility (I think it is on the freeware CD) which sets file attribute= s foru4 > you) until you can do BACKUP/LIST on the savesets.  H If you use a modern Kermit and "set file type labeled", it will preserveH attributes on a VMS->VMS transfer.  Even if you have to stage it throughI a Unix or windows (or any other system), if you tell the original sending.F Kermit on VMS to use labels, and the ultimate receiving Kermit (on theC destination VMS system) also to use labels, and do any intermediate I transfers (with Kermit, FTP or anything else) in binary, it will preserveg@ file attributes.  (The sending Kermit will put the VMS attributeB information in the labels, all the intermediate binary copies willE preserve the labels, and the receiving Kermit will process the labels E and use them to restore the attributes.)  Kermit has had this abilityF( for at least 5 years, maybe much longer.   From=20the VMS FAQ:i=          o  The Kermit file transfer package is available at:t-            o  http://www.columbia.edu/kermit/e(            o  ftp://kermit.columbia.edu/   --=20e John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Mon, 13 Jan 2003 21:07:43 -0600 @ From: "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net> Subject: Re: Newbie to OpenVMS. 6 Message-ID: <3E237EFF.23C987A7@earthlink.spamfree.net>   Amarendra GODBOLE wrote: >  > Greetings from India folks.r > B > I have started learning the OpenVMS OS, and being a newbie, I amD > facing many problems. First is, the editor help which shows me theE > keys E1, E2, and so on. What are the mappings of these keys on a PC  > keyboard ?  G That will depend largely upon the terminal program in use. Some of them(H try to maintain compatibility with the labels on the keycaps. Others tryE to maintain the function key location as if the user were at a DEC LK  keyboard (LK201, LK401, etc.).  E If your terminal program has any decent documentation, that may help,e+ though it may take a while to wade through.h  > > A pointer to a tutorial on the editor will be of great help.  5 Honestly, I've never seen one. As a sort of "primer":m   $ EDIT/EDT filespec   B ...brings you to a "*" prompt. Here, you can type C to enter "fullF screen mode". Just type on the screen, using the RETURN key to start aB new line, use the cursor keys to move around through the text, theD delete key (just above the RETURN (ENTER key on the main keypad)) toD delete. Avoid the numeric keypad until you find the HELP key. May beH SHIFT+F5 or NUM-LOCK, H. To exit the editor, CTRL+Z to return to the "*"@ prompt; then, type EXIT to save your work or QUIT to discard it.   $ EDIT/TPU filespec-  G ...brings you up in full screen mode right away. Use the cursor keys to G move around, the delete key to delete, the main keypad to type in text.m< CTRL+Z to exit. You may be prompted whether to save changes.  E > My aim of studying OpenVMS is to figure out the mappings of OpenVMS H > system services to those of Linux. Say, sys$getjpi. Now I need to findD > out all those Linux call(s), which might replace sys$getjpi. But IF > don't know the OpenVMS architecture, is internal workings, why there+ > is $ embedded in many if the names, etc..t  F Others have mentioned that the documentation set can be found on line. This URL will get you started:  " http://www.openvms.compaq.com/doc/  G You'll want to spend some time with the System Services manual. Some ofaG the correlations might not be obvious, but given time I think you might  figure a good chunk of it out.  G In general, I think it's best to consider VMS a super-set of Linux: VMSsF does everything that Linux and the various Gnu software that surroundsF it will do, and then some, although in some cases, the differences areC quite pronounced. For example, the shells vs. DCL. In some cases in F Linux-land, techniques specific to UN*X-like environments are the onlyC way to accomplish certain tasks, due to the fundamental differencesd between them and VMS.A  H > Second hurdle is the architecture of the system. I am not able to findE > out any good docs on the OpenVMS architecture, internals of the OS. D > Doc that will help me understand how OpenVMS really works, and not > just the command line usage.: > Any pointers regarding this will be greatly appreciated.  H Others have referred you to the Internals and Data Structures books. You  may also find this site helpful:   http://www.openvms.org/h  G Lastly, I'll refer you to some info. I posted on the 'web a while back:p  3 http://www.djesys.com/vms/freevms/index.html#mentorh   --   David J. Dachterar dba DJE Systemsm http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/v   ------------------------------   Date: 13 Jan 2003 20:50 CDTr' From: carl@gerg.tamu.edu (Carl Perkins)- Subject: Re: Newbie to OpenVMS.a- Message-ID: <13JAN200320503885@gerg.tamu.edu>n  , root@zworg.com (Amarendra GODBOLE) writes...A }I have started learning the OpenVMS OS, and being a newbie, I ameC }facing many problems. First is, the editor help which shows me theeD }keys E1, E2, and so on. What are the mappings of these keys on a PC }keyboard ?n  E The mapping depends entirely on how your terminal emulator is set up.t  = }A pointer to a tutorial on the editor will be of great help.t } D }My aim of studying OpenVMS is to figure out the mappings of OpenVMSG }system services to those of Linux. Say, sys$getjpi. Now I need to find C }out all those Linux call(s), which might replace sys$getjpi. But IuE }don't know the OpenVMS architecture, is internal workings, why therer* }is $ embedded in many if the names, etc..  C It is a convention that a lot of vendor supplied routines use a "$"nC to separate parts of names. User routines usually use "_" for this.eC The same is true for logical names and probably som e other places.p  G }Second hurdle is the architecture of the system. I am not able to find D }out any good docs on the OpenVMS architecture, internals of the OS.C }Doc that will help me understand how OpenVMS really works, and nott }just the command line usage.i9 }Any pointers regarding this will be greatly appreciated.a }--amarendra  8 You can find a rather large quantity of documentation at  " http://www.openvms.compaq.com/doc/  C The "Operating System Documentation" link there may be of interest.gG I would suggest going through the "OpenVMS Programming Concepts Manual"iF for the basics of how the programming stuff goes, along with a varietyB of examples. The details of the system service routines are in theJ "OpenVMS System Services Reference Manual" thnat you will also find there.   --- Carl   ------------------------------  % Date: Mon, 13 Jan 2003 21:31:03 -0600 @ From: "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net> Subject: Re: Newbie to OpenVMS.L6 Message-ID: <3E238477.2D56530B@earthlink.spamfree.net>   Amarendra GODBOLE wrote: > [snip]  > But I don't know ... why there+ > is $ embedded in many if the names, etc..n   Oh, yeah. Good point.   H It's generally considered that file names, logical names, routine names,C etc. containing a "$" are reserved to OpenVMS Engineering, and thatS- users should avoid using "$" in such places. d  B Generally, it is considered good form for users, ISVs, etc. to useD underscore ("_") instead of "$" in file/logical/symbol/object(/etc.) names...   -- p David J. Dachtera  dba DJE Systemso http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/A   ------------------------------    Date: 13 Jan 2003 22:09:27 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)i Subject: Re: Newbie to OpenVMS.-3 Message-ID: <4D3Gb3V4Mhux@eisner.encompasserve.org>0  y In article <3E238477.2D56530B@earthlink.spamfree.net>, "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net> writes:0 > Amarendra GODBOLE wrote:	 >> [snip]o! >> But I don't know ... why therei, >> is $ embedded in many if the names, etc.. >  > Oh, yeah. Good point.r > J > It's generally considered that file names, logical names, routine names,E > etc. containing a "$" are reserved to OpenVMS Engineering, and thatd/ > users should avoid using "$" in such places. y > D > Generally, it is considered good form for users, ISVs, etc. to useF > underscore ("_") instead of "$" in file/logical/symbol/object(/etc.)
 > names...  D I believe that ISVs using a registered facility name should be usingE a dollar sign.  That is the purpose of registering the facility name.,. (Facility codes are a different registration.)   ------------------------------  % Date: Mon, 13 Jan 2003 14:17:08 -0500r' From: "Main, Kerry" <Kerry.Main@hp.com>w& Subject: RE: Press Release from FortelT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4023D9B72@kaoexc01.americas.cpqcorp.net>   Sue - good stuff.r  ? For those who would like the external press pointer, reference:x. http://www.fortel.com/Press/Releases/Xuis.html   Regards   
 Kerry Main Senior Consultant- Hewlett-Packard (Canada) Co.! Consulting & Integration Services  Voice: 613-592-4660s Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)>     -----Original Message-----= From: Sue Skonetski [mailto:susan.skonetski@hp.nospam.com]=20e Sent: January 13, 2003 11:13 AMx To: Info-VAX@Mvb.Saic.Com " Subject: Press Release from Fortel    : Agreement Extends Best-of-breed Service Level Assurance toE European-based OpenVMS, UNIX/Linux, and Windows NT/2000 Installations   H Fremont-California-January 8, 2003-FORTEL Inc. (OTC BB: FRTLE) announcedH today a strategic agreement with XuiS, a privately owned British companyF that provides high quality software to manage OpenVMS, UNIX/Linux, and@ Windows NT/2000. Under the reseller agreement, XuiS will includeH FORTEL's SightLine as part of its product portfolio. This agreement willA allow XuiS's worldwide customer base to benefit from the improvedrH performance, greater efficiency, and increased return-on-investment that( SightLine brings to enterprise networks.  H XuiS cited customer demand for service level assurance software as beingA key to its decision to resell SightLine. "SightLine is a softwareb? solution that fits perfectly with our mission of providing only D best-of-breed products," said Tim Farrow, Managing Director of XuiS.B "Its addition to our portfolio enables us to offer our customers aD multi-platform performance monitoring and reporting tool that is the! most comprehensive in its field."   A FORTEL's SightLine collects and aggregates real-time information,e? end-to-end, independent of platforms, networks or applications,iG including information from external partners. SightLine can monitor the F critical path of an online transaction or request across heterogeneousC application components, analyzing the flow as a single entity. As a-8 result, it makes it possible to provide insight into the@ inter-relationships between all of the components involved in an enterprise network.l  G "We are looking forward to working with XuiS," said Terry Ewing, SeniorsF Vice President of Field Operations for FORTEL. "XuiS has well-deservedG credibility from an international customer base, many of whom have beentF with XuiS since its formation almost ten years ago. In addition, theirE software management focus on OpenVMS, UNIX/Linux, and Windows NT/2000-F fits very well with our target markets and will allow FORTEL to create new distribution channels."   F In addition to providing high quality software to manage OpenVMS, UNIXE and/or Windows NT/2000, XuiS also installs, configures and customizesnG software as well as trains administrators and provides customer supportr' on how to use the products effectively.t  
 ABOUT XuiS  G XuiS is a privately owned British registered company with internationaltG distribution outlets. The company specializes in providing high qualityTE software to manage OpenVMS, UNIX/Linux and Windows NT/2000. AlongsidetB the software, XuiS offers specialist professional services such asH consultancy and training as well as support at a number of levels, which* are tailored to the needs of the end-user.  E XuiS is headquartered in Surrey, UK. For more information about XuiS,f callH +44 (0) 1372 728881, email info@XuiS.com, or visit the company on the=20 +web atn
 www.XuiS.com.l   About FORTEL  H FORTEL provides the first real-time performance management solution thatC assures business end-to-end service-level goals. FORTEL's SightLinetA suite is based on analysis and correlation software that has beennB applied and tuned for more than 15 years in the systems management> performance arena by customers in finance and banking, defenseH management, manufacturing, retail services and government. FORTEL countsC among its customers many of the world's largest and most well knownnG organizations and enterprises. The Company is headquartered in Fremont,tE Calif., and can be contacted at (510) 440-9600 or by visiting its webi site at (www.fortel.com).c    FORTEL Public Relations Contact:   Emilio Dabul   Fusion Public Relations    New York NYB   212-651-4216   emilio.dabul@fusionpr.comU   FORTEL Contact:   # Steve Muxworthy, UK Country Managere   +44 (77) 3006 5342   steve.muxworthy@fortel.com   # # #e  B (c) 2003 FORTEL Inc. All rights reserved. FORTEL and SightLine areF trademarks of FORTEL Inc. All other product or company names mentioned@ in this release are trademarks of their respective corporations.  F Except for historical information contained in this press release, theD foregoing contains forward-looking statements that involve risks andH uncertainties. Actual results may differ materially from those indicatedB by such forward-looking statements based on a variety of risks andD uncertainties, including risks and uncertainties relating to factorsD such as our ability to achieve the intended benefits of the resellerE agreement with XuiS, the timely and successful development and market A acceptance of our products, services and features, our ability to H successfully attract and retain customers, our profitability targets andB financial metrics, the number, size and duration of contracts, andD software prices, activities of competitors, demand for our products,G volume of business and general economic conditions. FORTEL is currentlyQE not in compliance with the filing requirements of the U.S. Securities3C and Exchange Commission (the "SEC") and the risks and uncertaintiesIA above include potential adverse consequences relating to FORTEL's C noncompliance with the SEC's filing requirements and its ability to C become compliant with such requirements in the future. The guidance ? contained in this press release is based on limited information H available to the Company today, which is subject to change. Although theC guidance provided in this press release may change after today, the : Company undertakes no obligation to revise or update theseA forward-looking statements after today. Further information about H potential factors that could affect our financial results is included inH our Quarterly Reports on Form 10-Q and most recent Annual Report on Form> 10-K/A, which have been filed with the Securities and Exchange Commission.    ------------------------------  % Date: Mon, 13 Jan 2003 17:47:48 -0500b From: G <gdixon@ctri.com>w/ Subject: Re: READ ONLY TRANSACTIONS ARE HANGINGb8 Message-ID: <isf62v83d4diehedb0hq12rdldnn13r744@4ax.com>  C RMU/SHOW STATISTICS should allow you to see what process is causingt the stalls.a  F In general, there could be a lot of things behind the scenes adding toB the problem.  For instance, one of your programmers not explicitlyC starting a transaction, or using a DECLARE TRANSACTION instead of a E SET TRANSACTION.  Or other maintenance things like; how often are theI indexes rebuilt?  E Check to see if you have an Rdb support contract and if so, give them. a call.    Grant   @ On 13 Jan 2003 07:02:00 -0800, rallykarro@hotmail.com (karolina) wrote:   >Hi  >f@ >I have a problem with ORACLE RDB on a OpenVMS. The version I am: >running is Rdb V7.0-64. Below is the problem I am having. >lE >One process continously writes data into a table with commit between, >each/4 >write, the transaction is declared with read write.G >2 programs are continously reading the actual row added using a unique  >index( >and transaction declare with read only.G >When running in a cluster with one primary and one standby instance ofu >theD >reading programs (4 instances reads the same record), sometimes the	 >write in22 >the first process is hanging up to 10-20 seconds.@ >In RMU/ANALYZE it seems that the reading processes has opened a >PROTECTED READP- >instead of SHARED READ which is the default.e >a7 >Why does my transaction hang for that long doing this?g& >Is there any RDB setup I can look at?@ >I can't see that I am having the same problem on a single node. >  >Please help me! >s >regards >.	 >karolinaa   ------------------------------   Date: 14 Jan 2003 03:48:06 GMT- From: WADE@DAX.ASUB.ARKNET.EDU (Wade Fincher)v Subject: Reading file in DCL5 Message-ID: <slrnb26a5d.7gp.WADE@DAX.ASUB.ARKNET.EDU>   C Does anyone have any ideas on how I might go about reading the lastr> 30 bytes of a binary file and assigning the data to a logical?   -- -    F  ___ ___   _______   ______     _______     __________________________G |   Y   | |   _   | |   _  \   |   _   |   /      Wade Fincher        \:H |.  |   | |.  l   | |.  |   \  |.  l___|  | wade@sisko.asub.arknet.edu |H |. / \  | |.  _   | |.  |    \ |.  __)_   | Director of Computer Serv. |H |:      | |:  |   | |:  l    / |:  l   |  | Arkansas State University  |H |::.|:. | |::.|:. | |::.. . /  |::.. . |  |       Beebe Branch         |G `---^---' `---^---' `------'   `-------'   \__________________________/n  9        Any opinions expressed above are simply the result ;      of random bit changes caused by cosmic rays and should :       not be considered a sign of any form of intelligence   ------------------------------  % Date: Mon, 13 Jan 2003 18:59:34 -0800 $ From: Shane Smith <ssmith@icius.com>  Subject: RE: Reading file in DCL0 Message-ID: <01C2BB35.FEBF35C0@sulfer.icius.com>  H Does it have to be DCL? Could you run a small image? If so, it shouldn't be too hard.   Shaneb   -----Original Message-----@ From: WADE@DAX.ASUB.ARKNET.EDU [mailto:WADE@DAX.ASUB.ARKNET.EDU]& Sent: Monday, January 13, 2003 7:48 PM To: Info-VAX@Mvb.Saic.Comi Subject: Reading file in DCL    C Does anyone have any ideas on how I might go about reading the last*> 30 bytes of a binary file and assigning the data to a logical?   -- g    F  ___ ___   _______   ______     _______     __________________________G |   Y   | |   _   | |   _  \   |   _   |   /      Wade Fincher        \ H |.  |   | |.  l   | |.  |   \  |.  l___|  | wade@sisko.asub.arknet.edu |H |. / \  | |.  _   | |.  |    \ |.  __)_   | Director of Computer Serv. |H |:      | |:  |   | |:  l    / |:  l   |  | Arkansas State University  |H |::.|:. | |::.|:. | |::.. . /  |::.. . |  |       Beebe Branch         |G `---^---' `---^---' `------'   `-------'   \__________________________/t  9        Any opinions expressed above are simply the resultw;      of random bit changes caused by cosmic rays and should :       not be considered a sign of any form of intelligence   ------------------------------  % Date: Mon, 13 Jan 2003 21:25:38 -0600s@ From: "David J. Dachtera" <djesys.nospam@earthlink.spamfree.net>  Subject: Re: Reading file in DCL6 Message-ID: <3E238332.D739BD15@earthlink.spamfree.net>   Wade Fincher wrote:u > E > Does anyone have any ideas on how I might go about reading the last @ > 30 bytes of a binary file and assigning the data to a logical?  " Assuming "binary" means Fixed_512:   $ OPEN/READ INP filespec $READ_LOOP:@ $ READ/END=EOF_INP INP P9G $ GOTO READ_LOOP	 $EOF_INP:. $ CLOSE INPt $ P9_LEN = F$LENGTH( P9 )9' $ VALUE = F$EXTR( P9_LEN - 30, 30, P9 )  $ DEFINE lnm "''VALUE'"3   ...ought to do it. l  ; Not the most "elegant", I grant you, but should do the job.-   -- o David J. Dachterat dba DJE Systemso http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/e   ------------------------------  % Date: Tue, 14 Jan 2003 00:23:21 -0400o0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>  Subject: Re: Reading file in DCL/ Message-ID: <3E2390B7.12517023@vl.videotron.ca>d   Wade Fincher wrote:  > E > Does anyone have any ideas on how I might go about reading the lastr@ > 30 bytes of a binary file and assigning the data to a logical?   An inefficient method:  ) $OPEN/READ MYDATA binary_chocolate.recipew $LOOP:! $READ/END=ENDLOOP mydata mysymbolH $LAST_LINE = mysymbol. $GOTO LOOP:. $!	 $ENDLOOP:.! $THE_LENGTH = F$LENGTH(last_line). $start = the_length - 30( $myvalue = F$EXTRACT(start,30,last_line) $! $DEFINE CHOCOLATE "''myvalue'"  ( The above is far from foolproof though.   K This may sound silly, but is there a way to define a logical whose value is4N that of a symbol without symbol substitution happening at thd DCL level ?  (soD that any quotes in the symbol wouldn't mess things up for instance).   ------------------------------  % Date: Tue, 14 Jan 2003 00:48:43 -0400.0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>  Subject: Re: Reading file in DCL/ Message-ID: <3E2396A7.E5AC00AA@vl.videotron.ca>e   "David J. Dachtera" wrote: > $ READ/END=EOF_INP INP P9w > $ GOTO READ_LOOP > $EOF_INP:3
 > $ CLOSE INP, > $ P9_LEN = F$LENGTH( P9 )     L Question: when you reach end of file, and the READ statement branches to theM /END=label, is there a garantee that the target symbol remains unchanged (eg:i) still contains the last succesful read) ?:    N Also, in a fixed 512 file, will the last read yield the right number of bytes,K or just 512 bytes (with badding from true end of file to fill 512 bytes). ?    ------------------------------  % Date: Mon, 13 Jan 2003 20:13:57 +0100s2 From: rossbach.dieter@t-online.de (Dieter Robach) Subject: Re: smtp to vmsmail. Message-ID: <avv35l$jo$06$1@news.t-online.com>  1 >Do you have the VMS73_SYS-V0500 patch installed?o  E No, I'm runnning 7.3-1 with VMS731_sys-V200, but the description for u VMS73_SYS-V0500fJ fits exactly to my problem. I'll double the ctlpages and see what happens.   Thanks., Dieter    If so, read the ECOL >summary warning about SMTP mail delivery hangs after installing this patch.G >A workaround is given which says to increase CTLPAGES.  There's also ap2 >command procedure included to clear the messages.  P Yes, the procedure helps, I made a similar one to cleanup my environment during   all the tests today.    ------------------------------  # Date: Mon, 13 Jan 2003 23:30:15 GMT 0 From: Domain Administrator <alphase@pacbell.net> Subject: Re: smtp to vmsmail+ Message-ID: <3E234C76.CC9534B8@pacbell.net>e  3 This sounds similar to a problem I've been having. hG Using VMS 7.2 (no patches) TCPIP 5.1 (no patches). All works fine for a-H few days, then Incoming mail sits in the files in [TCPIP$SMTP]. 3 queuesF running but nothing gets delivered (I'm only using the POP service forB final delivery). I run alanlize mail/repair and every entry shows:5 %TCPIP-I-ANA_NOENTR, internal consistency check error1H Some mail then gets delivered then it usually stops again! The msg above$ ALWAYS shows up during analyze mail.E Sometimes I have to manually delete a number of files in order to gett things started again.n
 Any ideas?     "Dieter Robach" wrote:t > F > I am running VMS 7.3-1 with the lastest patches and TCPIP 5.1 ECO 1. > % > I have a strange problem with smtp:t >  > Outgoing mail works fine.aP > Incoming mail is delivered to vmsmail but the que entry in the smtp processingN > queue stays in "processing" mode and does not finish. The final announcementJ > about new mail via vmsmail is not send, but the mail is in the "newmail"% > folder, when you open your mailbox.o >  > What's that? > 	 > Regardse >  > Dieter   --     Have VMS, Will Travel- Wire paladin (at alphaseDOTcom)3
 San Franciscoo   ------------------------------  # Date: Mon, 13 Jan 2003 23:10:57 GMTl0 From: Domain Administrator <alphase@pacbell.net>- Subject: Re: Software distribution technology.+ Message-ID: <3E2347EF.7BBC418A@pacbell.net>s    I just finished a similar setup.  1. Customer fills out web form. E 2. User's email & generated password is added to a file NOT under thew web document directoryA 2a. Email is sent back to him with details on how to download therF software including a userid & generated password (NOT a VMS user/pass)G 2b. At the same time, a directory is created under the primary documentr1 directory on a VMS machine allowing http access. dH 3. User access via a web form filling in userid & passw mailed to him. ID check the entry against the file I updated in step 2. At that time IB also add an entry to the access.log file in that dir, so I can see) exactly when and how often he downloaded.yE 3a. A Webpage is dynamically created pointing to the user's directory5E w/o the ".html" giving him access to any of the file's I put there inu directory fashion.4 4. Browser requests download of any file clicked on.  A I actually copy the files to the directory and a separate processaH deletes the files & directory after a given time. I do not use the aliasH (set file/ent) because the zip files I create are unique and contain the user's unique license.  A You can try this by downloading an evaluation version of the Java = product for free at http://alphase.com/aex/Announcement.html.a  E I also set up a version for anonymous ftp, since C/Net requires a oneaF step download process for evaluation software, but of course all I canA capture about that is their IP address and date/time of download.3   -- 2   Have VMS, Will Travele Wire paladin (at alphaseDOTcom):
 San Franciscoe     JF Mezei wrote:t > * > Just a trial balloon, comments please... > O > Customer fills out web form. Email is sent back to him with details on how torP > download the sofware. At the same time, an account is created on a VMS machineN > allowing FTP access. In his SYS$LOGIN, alias entries to the software package8 > he is allowed to download are added (set file /enter). > O > This way, he can come back to download the software again if needed. If a new N > version comes out, one can add a new entry to the sys$logins to point to theM > new version of the software (giving him access tothe old and new versions).( >  > How workable would that be ? > N > My goal would be to better track who uses my software (it would be freeware)H > so that I could also send email announcements of new versions etc etc. > L > Would makin software available via non-anonymous FTP be a big problems forH > some folks whose corporate infrastructure may allow HTTP but not FTP ?   ------------------------------  # Date: Tue, 14 Jan 2003 04:45:22 GMTV/ From: "Curtis Justus" <donotspamme@foo.bar.edu>a! Subject: Suppressing SMTP headersm: Message-ID: <CBMU9.36679$Gm2.522453@twister.rdc-kc.rr.com>   Hello,  F I am trying to suppress the SMTP header information from incoming SMTPL e-mail messages.  I have configured my options on my SMTP service so that it* reads "NOHEADERS," yet they still show up.  ; What I would like to have happen is to have the subject andnL sender/recipients show up in VAXMail, and that is it.  Does anybody know how- I can get rid of the SMTP header information?r   I am running UCX V4.2-21.e   Thanks in advance, cj   ------------------------------  + Date: Tue, 14 Jan 2003 07:53:53 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>-% Subject: Re: Suppressing SMTP headerso; Message-ID: <01KR7UK8MNQE8WW5NX@sysdev.deutsche-boerse.com>a  H > I am trying to suppress the SMTP header information from incoming SMTPF > e-mail messages.  I have configured my options on my SMTP service so5 > that it reads "NOHEADERS," yet they still show up. s > = > What I would like to have happen is to have the subject and J > sender/recipients show up in VAXMail, and that is it.  Does anybody know4 > how I can get rid of the SMTP header information?   ) Right, from tcpip help set conf smtp/opt:.  (          o  Header control. Specify one:  ?             HEADERS       Headers printed at bottom of messagesa)             NOHEADERS     Headers omitteda<             TOP_HEADERS   Headers printed at top of messages;             NOTOP_        Resets TOP_HEADERS to the defaultn             HEADERS5  E You DO have to restart SMTP, though (I don't think it's necessary to 2 restart all of TCPIP).   ------------------------------  % Date: Mon, 13 Jan 2003 15:54:07 -0500o, From: "Wayne W. Scott" <wayne.scott@bea.com>3 Subject: TCP/IP 5.0a -- port number range problem ?.$ Message-ID: <3E23276F.70107@bea.com>  	 Hi Folks:v  P I am researching a problem for a customer who is running TCP/IP 5.0a on OpenVMS K V 7.1-2; hardware is an HP Alphaserver ES40 with 4 CPUs and 4 GB of memory.j  K I am looking for evidence of a problem in which port numbers above 64k are QN assigned.  There is a lot of ECO data to sift through.  If anyone is familiar O with such a scenario and can point me to documentation, it would be a big help.s  
 Thank you, Wayne Scottn   ------------------------------  # Date: Mon, 13 Jan 2003 22:19:16 GMTe+ From: Ryan Moore <rmoore@rmoore.dyndns.org>y7 Subject: Re: TCP/IP 5.0a -- port number range problem ?-= Message-ID: <Pine.LNX.4.50.0301131418130.22210-100000@jaipur>0  * On Mon, 13 Jan 2003, Wayne W. Scott wrote:Q > I am researching a problem for a customer who is running TCP/IP 5.0a on OpenVMSiM > V 7.1-2; hardware is an HP Alphaserver ES40 with 4 CPUs and 4 GB of memory.o >cL > I am looking for evidence of a problem in which port numbers above 64k areO > assigned.  There is a lot of ECO data to sift through.  If anyone is familiaraQ > with such a scenario and can point me to documentation, it would be a big help.i  J I'm not sure what you mean by "assigned".  But port numbers in TCP and UDPI are 16-bit values.  They can't have a value greater than 65535.  I assumer that's what you mean by "64k".   -Ryane   ------------------------------  % Date: Mon, 13 Jan 2003 14:41:00 -0800 $ From: Shane Smith <ssmith@icius.com>7 Subject: RE: TCP/IP 5.0a -- port number range problem ? 0 Message-ID: <01C2BB11.DBFE9D10@sulfer.icius.com>   Wayne W. Scott wrote:iI > I am researching a problem for a customer who is running TCP/IP 5.0a onr OpenVMSrM > V 7.1-2; hardware is an HP Alphaserver ES40 with 4 CPUs and 4 GB of memory.h > L > I am looking for evidence of a problem in which port numbers above 64k areF > assigned.  There is a lot of ECO data to sift through.  If anyone is familiarK > with such a scenario and can point me to documentation, it would be a big  help.   H I get the feeling something's missing here, Wayne. IIRC port numbers areF two-byte unsigned words, and they can run up to 65535. If you're usingH the right datatype, they /can't/ go over that number. What language, and* what's returning this out-of-range number?  E Of course, if you're getting numbers above 64k decimal, as in 64,000,p it's operating normally.   Shanev   ------------------------------  % Date: Mon, 13 Jan 2003 10:54:49 -0800 # From: "Tom Linden" <tom@kednos.com>i' Subject: RE: trivia curiousity questiona9 Message-ID: <CIEJLCMNHNNDLLOOGNJIOEHPGGAA.tom@kednos.com>u   Also look atF  http://h18003.www1.hp.com/products/quickspecs/soc_archives/80755.html   >-----Original Message-----o >From: Dan [mailto:dan@vrx.net]r' >Sent: Monday, January 13, 2003 9:18 AMR >To: Info-VAX@Mvb.Saic.Com$ >Subject: trivia curiousity question >e >g@ >what was the last Vax (not alpha) to be manufactured, and when?E >also, what speed was this model, how much do they roughly go for nowe= >(used/fleabay/etc) or are they not generally available used?o >hC >I've been teading some website model history and many of them seem C >to contradict each other. Some mention the vax4000, others mentionv@ >some other models as being the last. I once found a really nice? >website listing the models, dates, and speed (compared to 750) 0 >even with pictures, but I've lost the bookmark. > F >To me the Vax is still my favorite machine, of course, I grew up withG >them, literally. Started programming on a PDP-10 way back in 1976 whentG >I was 8 years old. There is something special about these beasts, even0> >though some of todays newer systems must dwarf them in power. >uD >I still remember running a computer store (with 8 terminals) off an? >old pdp-8, which even the best pc's couldn't do (at the time).I >uF >Seems sad to me that Intel paid of Compaq to wipe out Digital becauseD >of issues over their stealing chip technology. And now that HP owns, >Compaq, one has to wonder about the future. >t> >Also, does anyone know when the copyrights or patents on DECsG >technlogy expire? It should be some year soon. I got to thinking about C >this because I saw Peter Pan (Disney) is now public domain. Yes, Im9 >know it's unrelated, but the era aughta be pretty close.t >sE >It's been a strange journey, from the days of DEC to the days of theh >c64/amiga, and finally the PC.e >g >Dan.a >i >---' >Incoming mail is certified Virus Free.t; >Checked by AVG anti-virus system (http://www.grisoft.com).yB >Version: 6.0.435 / Virus Database: 244 - Release Date: 12/30/2002 >s ---l& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.435 / Virus Database: 244 - Release Date: 12/30/2002t   ------------------------------  % Date: Mon, 13 Jan 2003 15:08:19 -0500. From: Dan <dan@vrx.net>o' Subject: Re: trivia curiousity questiono8 Message-ID: <h4762vgedquabj3r0rvm8fslq15q3akacg@4ax.com>   empty document, I'll try again  2 On Mon, 13 Jan 2003 10:32:25 -0700, "Jill Forrest"! <chihuajill@earthlink.net> wrote:e >Dan,k >a* >Is this the website you were looking for: >eC >http://research.microsoft.com/~gbell/Digital/timeline/tmlnhome.htmB > + >It has a nice timeline of Digital history.  >  >Jills   ------------------------------  % Date: Mon, 13 Jan 2003 15:08:47 -0500b From: Dan <dan@vrx.net>r' Subject: Re: trivia curiousity questiona8 Message-ID: <m5762vsl3i9cfohul7oaajn22pd8e39gug@4ax.com>   hrm not it, but nice ;)e  B On Mon, 13 Jan 2003 10:54:49 -0800, "Tom Linden" <tom@kednos.com>  wrote:  
 >Also look attG > http://h18003.www1.hp.com/products/quickspecs/soc_archives/80755.htmlo >s   ------------------------------    Date: 13 Jan 2003 15:46:28 -08007 From: jones.computer.srv@worldnet.att.net (Daryl Jones)a" Subject: Re: What does RWCLU mean?= Message-ID: <8a646952.0301131546.2cfc65d3@posting.google.com>i  Q Saminda C F Lam <ccminda@ust.hk> wrote in message news:<3E1E9FE6.60809@ust.hk>...DD > What does RWCLU mean?  The manual says "Resource Wait for Cluster J > Transition".  I found processes in RWCLU state on an Alpha cluster.  Is A > it indicative of some resource problem?  If so, what resources?t >  > Thanks in advance,	 > Samindat   Saminda,  A The lock manager uses RWSCS to stall executions of a process on atE VMScluster node when the lock manager must wait for a response from aeF remote system that has information about the particular lock resource.  F A process that issues any lock requests on any node of a VMScluster inF transition (that is, while a node is being added or removed) is placed9 into RWCLU wait until the cluster memebership stabilizes.T   Daryl Joneso   ------------------------------  + Date: Mon, 13 Jan 2003 22:43:07 +0000 (UTC)f* From: bleau@umtof.umd.edu (Lawrence Bleau)4 Subject: Why doesn't SET BREAK %LABEL work any more?0 Message-ID: <avvfdr$4sj$1@grapevine.wam.umd.edu>  E I'm running OpenVMS AXP 7.1-2, Alpha Fortran 7.2, and have a question D about the debugger.  I remember, years ago, be the SET BREAK command8 accepting a Fortran statement label as an argument; e.g.   	SET BREAK %LABEL 10  A This is still documented (in DEBUG's HELP), and DEBUG accepts the + syntax, but now gives me the error message:m  @ %DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol table  A I compiled the program with /DEBUG=ALL to force it to include alln  symbols, but that had no effect.  F When did Fortran stop including statement labels in its symbol table? < If it didn't, how do I get this to work?  Sample code below.   $ CREATE TEST.FORe
 	INTEGER K 	ACCEPT *,Ku 10	TYPE *,Kd 	STOPx 	END $ FORTRAN/DEBUG=ALL/NOOPT TEST $ LINK/DEBUG TEST 
 $ RUN TEST DBG>SET BREAK %LABEL 10r@ %DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol table     Lawrence Bleau University of Maryland" Physics Dept., Space Physics Group 301-405-6223 bleau@umtof.umd.edue   ------------------------------  + Date: Mon, 13 Jan 2003 23:04:16 +0000 (UTC) * From: bleau@umtof.umd.edu (Lawrence Bleau)8 Subject: Re: Why doesn't SET BREAK %LABEL work any more?0 Message-ID: <avvglf$4sj$3@grapevine.wam.umd.edu>  f In article <bag62vkuda67odhf1vu26jrhmshrevj3pu@4ax.com>, John Laird <john@laird-towers.org.uk> writes:K >In your example, you have no reference to the label 10, so it has probablyfM >been (over-) optimised away(*)  What happens if you actually refer to it, noe >matter how pointlessly ?i  K No change.  I added a  GOTO 10  statment after the TYPE in the test programl6 and recompiled and I still get the same error message.   Lawrence Bleau University of Maryland" Physics Dept., Space Physics Group 301-405-6223 bleau@umtof.umd.educ   ------------------------------  % Date: Mon, 13 Jan 2003 23:28:50 +0000 + From: John Laird <john@laird-towers.org.uk>.8 Subject: Re: Why doesn't SET BREAK %LABEL work any more?8 Message-ID: <qoi62v0efdvilcbqc3o98tmih143njnudu@4ax.com>  G On Mon, 13 Jan 2003 23:04:16 +0000 (UTC), bleau@umtof.umd.edu (Lawrenceh
 Bleau) wrote:C  g >In article <bag62vkuda67odhf1vu26jrhmshrevj3pu@4ax.com>, John Laird <john@laird-towers.org.uk> writes: L >>In your example, you have no reference to the label 10, so it has probablyN >>been (over-) optimised away(*)  What happens if you actually refer to it, no >>matter how pointlessly ? >fL >No change.  I added a  GOTO 10  statment after the TYPE in the test program7 >and recompiled and I still get the same error message.   H One for the compiler group, then.  (I don't think Steve Lionel reads covL regularly.)  My experience is they've usually seen and fixed any bug we everE reported, so it's a question of finding or waiting on the appropriatel update.d     	Johno   ------------------------------  # Date: Mon, 13 Jan 2003 23:54:09 GMT + From: Ryan Moore <rmoore@rmoore.dyndns.org>h8 Subject: Re: Why doesn't SET BREAK %LABEL work any more?= Message-ID: <Pine.LNX.4.50.0301131549550.22253-100000@jaipur>   C I don't know how this works in Fortran, but in C, I always have theaE compiler create a .LIS file that contains line numbers of the code inoI question and then do a SET BREAK %LINE command.  Can Fortran do somethinge similar?  * On Mon, 13 Jan 2003, Lawrence Bleau wrote:h > In article <bag62vkuda67odhf1vu26jrhmshrevj3pu@4ax.com>, John Laird <john@laird-towers.org.uk> writes:M > >In your example, you have no reference to the label 10, so it has probablyeO > >been (over-) optimised away(*)  What happens if you actually refer to it, nol > >matter how pointlessly ?  >-M > No change.  I added a  GOTO 10  statment after the TYPE in the test programv8 > and recompiled and I still get the same error message.   Here's a bogus example:S  .                                 Source Listing> 13-JAN-2003 23:51:54  DEC C V6.0-001                    Page 1  8 15-FEB-2002 23:42:15  DISK$DEVELOPMENT:[RMOORE.TMP]B.C;1  "               1 #include <stdio.h>             874e             875 main()       1     876 {a9       1     877    unsigned long  Count, Whole, Fraction;r       1     878sE       1     879    for (Count = 0; Count < (unsigned long)0xffffffff;e Count++)       2     880    {.       2     881       Whole = Count / 1000000;9       2     882       Fraction = (Count % 1000000) / 100;.0       2     883       if ((Count % 100000) == 0)9       2     884          fprintf (stdout, "%d %d %d\a\n",d Count,Whole,Fraction);       1     885    }       1     886 }3    4 So I'd do a SET BREAK %LINE 882 to stop on the line:  # Fraction = (Count % 1000000) / 100;O  9 The first column shows nested levels of statement blocks.0   ------------------------------  % Date: Mon, 13 Jan 2003 22:47:19 +0000r+ From: John Laird <john@laird-towers.org.uk>t8 Subject: Re: Why doesn't SET BREAK %LABEL work any more?8 Message-ID: <bag62vkuda67odhf1vu26jrhmshrevj3pu@4ax.com>  G On Mon, 13 Jan 2003 22:43:07 +0000 (UTC), bleau@umtof.umd.edu (Lawrenceh
 Bleau) wrote:l  F >I'm running OpenVMS AXP 7.1-2, Alpha Fortran 7.2, and have a questionE >about the debugger.  I remember, years ago, be the SET BREAK commande9 >accepting a Fortran statement label as an argument; e.g., >V >	SET BREAK %LABEL 10s >eB >This is still documented (in DEBUG's HELP), and DEBUG accepts the, >syntax, but now gives me the error message: >lA >%DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol tableg >eB >I compiled the program with /DEBUG=ALL to force it to include all! >symbols, but that had no effect.- > G >When did Fortran stop including statement labels in its symbol table? @= >If it didn't, how do I get this to work?  Sample code below.B >C >$ CREATE TEST.FOR >	INTEGER K7 >	ACCEPT *,K >10	TYPE *,K >	STOP >	ENDo >$ FORTRAN/DEBUG=ALL/NOOPT TEST  >$ LINK/DEBUG TEST >$ RUN TESTr >DBG>SET BREAK %LABEL 10A >%DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol tableC  J In your example, you have no reference to the label 10, so it has probablyL been (over-) optimised away(*)  What happens if you actually refer to it, no matter how pointlessly ?  3 (*)  Notwithstanding the argument about /NOOPT ;-))      	Johnk   ------------------------------  + Date: Tue, 14 Jan 2003 00:21:49 +0000 (UTC)o* From: bleau@umtof.umd.edu (Lawrence Bleau)8 Subject: Re: Why doesn't SET BREAK %LABEL work any more?0 Message-ID: <avvl6t$6tu$1@grapevine.wam.umd.edu>  I >One for the compiler group, then.  (I don't think Steve Lionel reads cov M >regularly.)  My experience is they've usually seen and fixed any bug we everiF >reported, so it's a question of finding or waiting on the appropriate >update.  B I just received an email from Henry Juengst, who gave me this tip:  ? >  There are many things wrong with the Alpha Fortran compiler.o2 >Have a look at the OpenVMS.org Bug Reports forum. >sG >The good news is that your example works correctly with Compaq Fortran,K >V7.4B-1684-46B5P. The debugger find label 10 and stops at the break point.o   Thanks, Henry!   Lawrence Bleau University of Maryland" Physics Dept., Space Physics Group 301-405-6223 bleau@umtof.umd.edun   ------------------------------  % Date: Mon, 13 Jan 2003 23:09:02 -0500g( From: David Froble <davef@tsoft-inc.com>8 Subject: Re: Why doesn't SET BREAK %LABEL work any more?, Message-ID: <3E238D5E.2060704@tsoft-inc.com>   John Laird wrote:m  I > On Mon, 13 Jan 2003 22:43:07 +0000 (UTC), bleau@umtof.umd.edu (Lawrencel > Bleau) wrote:S >  > G >>I'm running OpenVMS AXP 7.1-2, Alpha Fortran 7.2, and have a question F >>about the debugger.  I remember, years ago, be the SET BREAK command: >>accepting a Fortran statement label as an argument; e.g. >> >>	SET BREAK %LABEL 10 >>C >>This is still documented (in DEBUG's HELP), and DEBUG accepts the - >>syntax, but now gives me the error message:s >>B >>%DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol table >>C >>I compiled the program with /DEBUG=ALL to force it to include allr" >>symbols, but that had no effect. >>H >>When did Fortran stop including statement labels in its symbol table? > >>If it didn't, how do I get this to work?  Sample code below. >> >>$ CREATE TEST.FORX >>	INTEGER K
 >>	ACCEPT *,Kt
 >>10	TYPE *,K  >>	STOPo >>	END  >>$ FORTRAN/DEBUG=ALL/NOOPT TEST >>$ LINK/DEBUG TEST  >>$ RUN TEST >>DBG>SET BREAK %LABEL 10 B >>%DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol table >> > L > In your example, you have no reference to the label 10, so it has probablyN > been (over-) optimised away(*)  What happens if you actually refer to it, no > matter how pointlessly ? > 5 > (*)  Notwithstanding the argument about /NOOPT ;-))c    Q When compiling for debugging, ALWAYS use the /NOOPT switch.  Don't even think of w/ trying to understand what the GEM backend does.p  Q What the debugger is looking for is a line number, not a label.  You do not have dM 10 lines in your example program.  I got bit by this one years ago.  Yeah, I  < know what the doc says, but it doesn't work that way for me.  L Try SEARCH 0 TYPE, then set the break for the line # the debugger shows you.   Dave   ------------------------------    Date: 13 Jan 2003 21:52:21 -0800, From: ian.burgess@start.com.au (Ian Burgess)8 Subject: Re: Why doesn't SET BREAK %LABEL work any more?< Message-ID: <6b63fc08.0301132152.752d0f9@posting.google.com>   (Lawrence Bleau) wrote eG > I'm running OpenVMS AXP 7.1-2, Alpha Fortran 7.2, and have a questionbF > about the debugger.  I remember, years ago, be the SET BREAK command: > accepting a Fortran statement label as an argument; e.g. >  > 	SET BREAK %LABEL 10 > C > This is still documented (in DEBUG's HELP), and DEBUG accepts thea- > syntax, but now gives me the error message:r > B > %DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol table > C > I compiled the program with /DEBUG=ALL to force it to include allc" > symbols, but that had no effect. > H > When did Fortran stop including statement labels in its symbol table? > > If it didn't, how do I get this to work?  Sample code below. >  > $ CREATE TEST.FORb > 	INTEGER K
 > 	ACCEPT *,Kn
 > 10	TYPE *,K  > 	STOP, > 	END  > $ FORTRAN/DEBUG=ALL/NOOPT TEST > $ LINK/DEBUG TESTo > $ RUN TEST > DBG>SET BREAK %LABEL 10.B > %DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol table >  >  > Lawrence Bleau  ' No real solution, but a bit of light...r5 I added    PROGRAM  TEST and changed the label to 123t   The symbol is there, as ".123" e DBG> sho sym *123e label TEST\.123n  - but I can't find a syntax that is accepted...e  ! DBG> help Built_in_Symbols %labelu   DEBUGi     Built_in_Symbols  
     %LABEL  I          %LABEL n is the debugger syntax for referring to label n in youriH          program. This is intended for languages like FORTRAN which haveF          numeric program labels. You can qualify the label with a path/          name specifying the containing module.m            Example:   ,          DBG> SET BREAK MODULENAME\%LABEL 10  E          The old syntax of %LABEL MODULENAME\n is no longer accepted.s     DBG> set break TEST\%label 123F %DEBUG-E-NOSYMBOL, symbol 'TEST\%LABEL 123' is not in the symbol table  	 Hmmm...??a  
 Ian Burgess, t University of Queensland   ------------------------------  % Date: Tue, 14 Jan 2003 06:31:11 +0000t7 From: David B Sneddon - bigpond <dbsneddon@bigpond.com>n8 Subject: Re: Why doesn't SET BREAK %LABEL work any more?* Message-ID: <3E23AEAF.1000304@bigpond.com>   Ian Burgess wrote: > (Lawrence Bleau) wrote S > G >>I'm running OpenVMS AXP 7.1-2, Alpha Fortran 7.2, and have a questioniF >>about the debugger.  I remember, years ago, be the SET BREAK command: >>accepting a Fortran statement label as an argument; e.g. >> >>	SET BREAK %LABEL 10 >>C >>This is still documented (in DEBUG's HELP), and DEBUG accepts ther- >>syntax, but now gives me the error message:t >>B >>%DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol table >>C >>I compiled the program with /DEBUG=ALL to force it to include alle" >>symbols, but that had no effect. >>H >>When did Fortran stop including statement labels in its symbol table? > >>If it didn't, how do I get this to work?  Sample code below. >> >>$ CREATE TEST.FOR- >>	INTEGER K
 >>	ACCEPT *,K1
 >>10	TYPE *,K  >>	STOPg >>	END  >>$ FORTRAN/DEBUG=ALL/NOOPT TEST >>$ LINK/DEBUG TESTN >>$ RUN TEST >>DBG>SET BREAK %LABEL 10lB >>%DEBUG-E-NOSYMBOL, symbol '%LABEL 10' is not in the symbol table >> >> >>Lawrence Bleau >  > ) > No real solution, but a bit of light...h7 > I added    PROGRAM  TEST and changed the label to 123n > ! > The symbol is there, as ".123" t > DBG> sho sym *123c > label TEST\.123C > / > but I can't find a syntax that is accepted...o > # > DBG> help Built_in_Symbols %labelr >  > DEBUGe >  >   Built_in_Symbols >  >     %LABEL > K >          %LABEL n is the debugger syntax for referring to label n in your:J >          program. This is intended for languages like FORTRAN which haveH >          numeric program labels. You can qualify the label with a path1 >          name specifying the containing module.e >  >          Example:e > . >          DBG> SET BREAK MODULENAME\%LABEL 10 > G >          The old syntax of %LABEL MODULENAME\n is no longer accepted.l >  >   > DBG> set break TEST\%label 123H > %DEBUG-E-NOSYMBOL, symbol 'TEST\%LABEL 123' is not in the symbol table >  > Hmmm...??  >  > Ian Burgess, b > University of Queensland >   H When you RUN TEST, the modulename is identified as TEST$MAIN, *not* TEST( because there is no PROGRAM statement...   Regards, Dave.Z -- sI David B Sneddon (dbs)    VMS Systems Programmer     dbsneddon@bigpond.com I Sneddo's quick guide ...          http://www.users.bigpond.com/dbsneddon/hI DBS freeware at ...   http://www.users.bigpond.com/dbsneddon/software.htmtI "Life is what happens to you while you're busy making other plans" Lennons   ------------------------------  % Date: Mon, 13 Jan 2003 20:58:44 +0100n" From: "Hans Vlems" <hvlems@iae.nl> Subject: Re: Wireless Ethernet5 Message-ID: <avv5pr$k1ktq$1@ID-143435.news.dfncis.de>g  ? "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> schreef in bericht ) news:3E225FD4.2F9631CF@vl.videotron.ca...oH > Out of curiosity, how does authentication work for wireless ethernet ? >_I > Is wireless ethernet simply a bridge between two ethers ? Or is there a / > hub/router that manages the wireless access ?   K The equipment that I was familiar with were just bridges, one port happened|F to connect to a directional antenna. The bridge may be configured to aJ certain extent. It could be configured to use encryption, to block certain& protocols, to restrict multicasts etc.  F > Does the DHCP server on VMS have the ability to server those users ?  E If DHCP is allowed across the wireless link, there's no reason why it K shouldn't work. The wireless connection in its most simple configuration issJ a transparent ethernet pipe, albeit at a lower speed. The stuff is sold asF 11 Mb/s but that implies that both directions are more or less equally loaded.m  L > Someone told me that WiFi will make it big in internet cafs etc where youI > could buy a prepaid card to gain access to the lan/internet by entering   > username/password on the card. >r= > Are there mechanisms in those wireless devices to do that ?.   Sorry, never seen that..  E > Or would access control exist solely on the ethernet address of thebK > laptop/device ? (eg: enter that address in the DHCP server's database ando only> > MAC adresses in the DHCP server are granted a DHCP response)  K That would be the solution. Or put each wireless bridge behind an IP router_I and controll IP addresses that way: a masl of 255.255.255.253 would allow|> just two addresses: one for the client and one for the router.   Hans   ------------------------------  % Date: Mon, 13 Jan 2003 12:51:28 -0800 0 From: Mark Berryman <Mark.Berryman@Mvb.Saic.Com> Subject: Re: Wireless Ethernet, Message-ID: <3E22B650.4965E196@Mvb.Saic.Com>   JF Mezei wrote:| > H > Out of curiosity, how does authentication work for wireless ethernet ? > I > Is wireless ethernet simply a bridge between two ethers ? Or is there at/ > hub/router that manages the wireless access ?c  B Both.  Although, from the context of your message, I doubt you areF referring to the former.  Wireless bridges, which connect two ethernetG segments via a wireless medium (radio or laser) have existed for years.f  G Today, however, when one refers to wireless ethernet, one is most ofteneE talking about an access point which physically plugs into an EthernetaH LAN and provides the means for a client with a wireless NIC to appear asE if connected to that ethernet segment.  Access-points are not routersiE (although it is possible for a product to combine the two functions);IF they are simply a node on the network that appears to have several MAC1 addresses behind it (as a hub or a switch would).    > F > Does the DHCP server on VMS have the ability to server those users ?  B Yes, or any other DHCP server.  Other nodes on the network have noC direct way of knowing that these users are connected via a wirelesse: connection.  They appear as any other user on the network.   > L > Someone told me that WiFi will make it big in internet cafs etc where youI > could buy a prepaid card to gain access to the lan/internet by entering9  > username/password on the card.  H Wrong tense; they have already made it big.  Many internet cafes, coffee! shops, etc. are doing this today.    > = > Are there mechanisms in those wireless devices to do that ?m  G Yes.  Basically the same way that a network switch can be configured toLG authenticate a user before allowing any of that user's packets onto the." network.  See the 802.1X standard.   > E > Or would access control exist solely on the ethernet address of theFP > laptop/device ? (eg: enter that address in the DHCP server's database and only> > MAC adresses in the DHCP server are granted a DHCP response)  < This method can be used as well but see the previous answer.  
 Mark Berryman(   ------------------------------  # Date: Tue, 14 Jan 2003 01:13:09 GMT 1 From: Michael Austin <maustin@firstdbasource.com>a Subject: Re: Wireless Ethernet2 Message-ID: <3E2363FD.8BCFD203@firstdbasource.com>   JF Mezei wrote:D > H > Out of curiosity, how does authentication work for wireless ethernet ?  G Every wireless device can work in unsecure mode or secure mode.  If youLG do not provide a WEP (wireless Encryption Protocol) Key (40-128 bit) onoF a WEP-enabled WAP (Wireless Access Point) then the WAP device will notE "service" the connection.   There are technologies that use differentt& WEP and/or authenticaion mechanisms.     > I > Is wireless ethernet simply a bridge between two ethers ? Or is there a1/ > hub/router that manages the wireless access ?l   Can be.  And can be secured. > F > Does the DHCP server on VMS have the ability to server those users ? > L > Someone told me that WiFi will make it big in internet cafs etc where youI > could buy a prepaid card to gain access to the lan/internet by enteringi  > username/password on the card. > = > Are there mechanisms in those wireless devices to do that ?f   Yes. > E > Or would access control exist solely on the ethernet address of thenP > laptop/device ? (eg: enter that address in the DHCP server's database and only> > MAC adresses in the DHCP server are granted a DHCP response)  F If you do not provide a WEP key in your router/WAP device, then anyoneG with a Wireless card can access the internet through your connection -- D and it would look like the data was transferred to you.  Haven't youC seen the stories of the network security guy driving around a majoraD metropolitain area and gaining access to many very large corporationG infrastructures?  BTW, it is very diffiecult to guess a WEP key that ispH 128 bits in length.  Not impossible, but will take you a very long time.   -- l Regards,  6 Michael Austin            OpenVMS User since June 19847 First DBA Source, Inc.    Registered Linux User #261163r   ------------------------------  % Date: Mon, 13 Jan 2003 18:18:39 -0700l% From: Dan O'Reilly <dano@process.com>I Subject: Re: Wireless EthernetB Message-ID: <5.1.0.14.2.20030113181744.00bb3ca0@raptor.psccos.com>  , At 06:13 PM 1/13/2003, Michael Austin wrote:G >If you do not provide a WEP key in your router/WAP device, then anyonerH >with a Wireless card can access the internet through your connection --8 >and it would look like the data was transferred to you.  J Not strictly true.  Some WAP's, such as the Linksys I use, can also filter on allowable MAC addresses.g   ------J +-------------------------------+----------------------------------------+J | Dan O'Reilly                  |  "There are 10 types of people in this |J | Principal Engineer            |   world: those who understand binary   |J | Process Software              |   and those who don't."                |J | http://www.process.com        |                                        |J +-------------------------------+----------------------------------------+   ------------------------------  % Date: Mon, 13 Jan 2003 17:16:10 -0800f$ From: Shane Smith <ssmith@icius.com> Subject: RE: Wireless Ethernet0 Message-ID: <01C2BB27.A134B5A0@sulfer.icius.com>   <SNIP> Michael Austin wrote:eH > If you do not provide a WEP key in your router/WAP device, then anyoneI > with a Wireless card can access the internet through your connection --aF > and it would look like the data was transferred to you.  Haven't youE > seen the stories of the network security guy driving around a majordF > metropolitain area and gaining access to many very large corporationI > infrastructures?  BTW, it is very diffiecult to guess a WEP key that isnJ > 128 bits in length.  Not impossible, but will take you a very long time.  G OK, now this is the part I'm curious about, especially since I'm in thenB process of installing some wireless gear. How is this key used? ItG surely it isn't transmitted, or it would be easy to "sniff" it with thee@ right gear. So it must be used to encrypt something. Is there anC additional header packet with wireless, containing an agreed phraseoF that's encrypted with the key? If it doesn't decode to the right text, the packet is refused?  E Incidentally, a while back someone went one further than the securitynF guy in the car. Apparently you get even more connections when flying aC light airplane over the city. The lengths some people will go to tox prove a point!   Shane    ------------------------------  % Date: Mon, 13 Jan 2003 21:26:23 -0400d0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca> Subject: Re: Wireless Ethernet/ Message-ID: <3E236720.A6BAA2CC@vl.videotron.ca>3   Michael Austin wrote:lI > Every wireless device can work in unsecure mode or secure mode.  If youoI > do not provide a WEP (wireless Encryption Protocol) Key (40-128 bit) onoH > a WEP-enabled WAP (Wireless Access Point) then the WAP device will not > "service" the connection.   M How does one provide a WEP key ? Does one need a proprietary application that,0 talks to the wireless ethernet card's hardware ?  F For instance, *IF* I had a VAX laptop with PCcard slot with a wirelessK ethernet card in it, would the card come with some useless windows drivers,u" rendering my card useless on VMS ?   ------------------------------  % Date: Mon, 13 Jan 2003 18:25:43 -0800t$ From: Shane Smith <ssmith@icius.com> Subject: RE: Wireless Ethernet0 Message-ID: <01C2BB31.4AB961D0@sulfer.icius.com>  D The software that comes with the device will provide a mechanism forG setting the WEP. My Linksys WAP/Hub/Router lets you set the WEP throughnF a web interface, the card I haven't installed yet has a setting in the& driver, according to the instructions.  F I think you are (ahem) stuffed if you want the PC card in the VAX. YouH /might/ be able to program the WEP through a PC laptop and have it stillB work when it goes in the VAX, it all depends on where the WEP getsE stored. You just have to hope it's NVRAM on the card, not on the diskr with the drivers.-   Shanea   -----Original Message-----7 From: JF Mezei [mailto:jfmezei.spamnot@vl.videotron.ca]t& Sent: Monday, January 13, 2003 5:26 PM To: Info-VAX@Mvb.Saic.Com  Subject: Re: Wireless Ethernet     Michael Austin wrote: I > Every wireless device can work in unsecure mode or secure mode.  If youaI > do not provide a WEP (wireless Encryption Protocol) Key (40-128 bit) on-H > a WEP-enabled WAP (Wireless Access Point) then the WAP device will not > "service" the connection.   H How does one provide a WEP key ? Does one need a proprietary application that0 talks to the wireless ethernet card's hardware ?  F For instance, *IF* I had a VAX laptop with PCcard slot with a wirelessB ethernet card in it, would the card come with some useless windows drivers," rendering my card useless on VMS ?   ------------------------------   End of INFO-VAX 2003.027 ************************