1 INFO-VAX	Tue, 24 Dec 2002	Volume 2002 : Issue 709       Contents:1 Re: "VMS will be around long after we retire" ... 1 Re: "VMS will be around long after we retire" ... 1 Re: "VMS will be around long after we retire" ... ! RE: DECUS^H^H^H^H^HCUO membership 4 Re: Demise of VMS (and most other operating systems)& Re: Disk locked during file creation ?& Re: Disk locked during file creation ?& Re: Disk locked during file creation ?& Re: Disk locked during file creation ?" Re: drop modem line from decserver' Re: F$GETDVI ("AVL") in an IF statement ! RE: First Hammer performance test ! Re: First Hammer performance test ! Re: First Hammer performance test 9 Re: How to interpret extended File IDs - e.g. (86051,3,0) 9 Re: How to interpret extended File IDs - e.g. (86051,3,0) 5 HP-CUO Membership (Re: DECUS^H^H^H^H^HCUO membership) 6 Re: Is it possible to add a cluster node w/o decnet??? Link errors. Re: Link errors. Re: OpenVMS Branding Query Re: OpenVMS Branding Query Re: OpenVMS Branding Query Re: OpenVMS Branding Query Re: OpenVMS Branding Query- Re: OT: Whoa! Is Sun aiming at VMS's jugular? - Re: OT: Whoa! Is Sun aiming at VMS's jugular? - Re: OT: Whoa! Is Sun aiming at VMS's jugular? - Re: OT: Whoa! Is Sun aiming at VMS's jugular? - Re: OT: Whoa! Is Sun aiming at VMS's jugular? - Re: OT: Whoa! Is Sun aiming at VMS's jugular? . SAN recommendations mixing VMS, Solaris, Linux2 RE: SAN recommendations mixing VMS, Solaris, Linux# Semaphores and Shared memory on VMS  Re: Spiralog anyone? Re: VAX 7810's.  Re: VM web ring & spam. VMS has got apps ... 3300 and counting per HP!2 Re: VMS has got apps ... 3300 and counting per HP! Re: Vol Shadowing Question Re: Vol Shadowing Question  F ----------------------------------------------------------------------  % Date: Mon, 23 Dec 2002 15:19:08 -0400 ( From: JF Mezei <jfmezei@vl.videotron.ca>: Subject: Re: "VMS will be around long after we retire" .../ Message-ID: <3E0761AB.67300632@vl.videotron.ca>    "Main, Kerry" wrote:H > As I recall, that quote was intended to address new UNIX opportunitiesD > in general i.e. Tru64 vs. HP-UX.   The quote "and all new business > opportunities."   L As I recall, the quote was in the contexts of platforms. They would continueM to sell Alpha based systems to existing customers, but all new business would > initially go to PA-Risc and later Itanium when/if it is ready.  L Remember that they had 7 months to prepare this speech, unless Deutsche Bank9 forced them to change their tune at the very last minute.   K She could have said "In the short term, we will leverage all our assets and M pitch whatever platform will win us new business against our competitors, and E in the longer term, all our systems will be consolidated on Itanium."   M And when you combine this with Stallard's infamous memo, it was clear that at F some point in time HP really did think that it could conver ex Digital  customers onto HP-UX on Itanium.    N It doesn't take very long to review those speeches/documents and outline thoseK areas that would create concerns with existing customers, especially if you N are trying to make your "new" customers feel happy about being switched to HP.   ------------------------------  % Date: Mon, 23 Dec 2002 22:16:34 -0500 * From: "Bill Todd" <billtodd@metrocast.net>: Subject: Re: "VMS will be around long after we retire" ...2 Message-ID: <1cqcnQAuSpMRTJqjXTWc2Q@metrocast.net>  2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageL news:BE56C50EA024184DAF48F0B9A47F5CF402660C4B@kaoexc01.americas.cpqcorp.net. .. Bill,   C Imho (personal opinion based on what I have seen internally) I will F admit the quote you referenced can be interpreted in a number of ways.  F As I recall, that quote was intended to address new UNIX opportunitiesB in general i.e. Tru64 vs. HP-UX.   The quote "and all new businessF opportunities." was imho, not intended to mean that HP-UX would be the@ only platform of choice for all new business where OpenVMS, NSK,H Windows, Linux, and Novell OS's might be better alternatives - depending= on a number of different qualifiers and market opportunities.    *** L While your statement above is correct, it is more than somewhat misleading -H because while the roadmap statement I quoted indeed did not say anythingK about HP-UX being preferred over NSK, Windows, Linux, and Novell (nor did I L in any suggest that it had), it very specifically *did* say that PA-RISC wasG the target platform for *all* new RISC business and that Alpha would be J targeted only at the existing customer base (one could quibble that NSK isL RISC-based as well, but in the specific PA-RISC/Alpha context of the roadmapE section it didn't seem to be included and it was addressed separately G elsewhere).  And this was a hardware platform statement, not an OS (let F alone Unix-specific) statement, so VMS was clearly included (if no newL business is targeted at Alpha, it will be rather difficult to push VMS there for another 18 months or so).  ***   B Winning the business is what's important and that is what is being+ emphasized - regardless of the OS platform.    *** J Perhaps you'd care to cite a reference to that effect (that *specifically*I includes Tru64 and VMS):  the roadmap certainly says nothing of the kind, L and I suspect that carries somewhat more weight with the world at large than does your own personal opinion.  ***     The quote I provided in theH earlier thread is a recent example where HP lead with Tru64 UNIX and won
 the business.    *** I Once again, I must observe that the article you cited said nothing of the I kind:  it merely stated that Tru64 is what the customer wound up getting, ' not what HP may have 'lead' [sic] with.  ***   B With respect to OpenVMS and new business statements from senior HPF exec's, here is the recent Carly quote from the recent OpenVMS video -  D "OpenVMS has been recognized for its rock solid reliability for manyH years. It was built from the beginning to endure the test of time. HP isE proud to celebrate its 25th anniversary this year and I am especially G pleased that it continues to attract new Customers and new applications  in key market segments."   *** L In other words, she's pleased that HP has to beat off new VMS customers (andJ why wouldn't she be?), but, once again, says nothing about actually making any effort to *attract* them.    - bill   ------------------------------  % Date: Mon, 23 Dec 2002 22:13:16 -0500 ' From: Howard S Shubs <howard@shubs.net> : Subject: Re: "VMS will be around long after we retire" ...< Message-ID: <howard-0EF049.22131623122002@enews.newsguy.com>  0 In article <01C2AA6C.EA1D64E0@sulfer.icius.com>,&  Shane Smith <ssmith@icius.com> wrote:  % > WASH YOUR MOUTH OUT, CLIFF!!!!!!!!!  > J > This would be a major disaster. VMS has won our loyalty through its highI > quality. Can you imagine what it would be like three years after MS got  > their hands on it?   It would be NT 3.51.   --  4 Today, on Paper-view: The World Origami Championship   ------------------------------  % Date: Sun, 22 Dec 2002 15:45:51 +0000 9 From: "Patrick Caulfield" <patrick@NOSPAMtykepenguin.com> * Subject: RE: DECUS^H^H^H^H^HCUO membership@ Message-ID: <R6lN9.2239$1P6.293812@newsfep2-win.server.ntli.net>  6 On Fri, 20 Dec 2002 09:30:58 +0000, Shane Smith wrote:   >  > F > If anyone's really stuck with this and needs a US mailing address toG > join, drop me a note offline and we can come to some arrangement. VMS " > should be a basic human right...     Sounds like a good idea     /me mails a friend in California  G I can afford the 55 quid, its just that the price hike really annoys me F because all I'm paying for is for the UK chapter to send me flyers for" events I'm never going to attend*.  J If I'm going to have to pay that amount of money to run VMS as a hobby I'dM rather pay it to David Cathay at Montagar for his running of the hobbyist web  site.    Patrick   E * Not that I've anything against those events, I went to a lot when I  worked at a VMS shop.    ------------------------------  % Date: Mon, 23 Dec 2002 22:47:30 +0100 6 From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne@vajhoej.dk>= Subject: Re: Demise of VMS (and most other operating systems) ) Message-ID: <3E078472.7050107@vajhoej.dk>    Shane Smith wrote:  I > Better idea: Charge more maintenance for the non-VMS flavours and quote E > the difference in platform quality as the reason. Give examples and H > links. It's win-win. Either you make more money for a Windows sale, orJ > you get less grief and save someone from the Dark Side at the same time.   Excellent idea.   , If you own stock in IBM and SUN that is !!!!   Arne   ------------------------------    Date: 23 Dec 2002 13:59:02 -08007 From: jones.computer.srv@worldnet.att.net (Daryl Jones) / Subject: Re: Disk locked during file creation ? = Message-ID: <8a646952.0212231359.3d9cfa27@posting.google.com>   g JF Mezei <jfmezei.spamnot@vl.videotron.ca> wrote in message news:<3E06460C.370760A9@vl.videotron.ca>... O > I am creating some 1.5 million block empty sequential files , and during that ; > time, I find that all other IO on the disk seems blocked.  > P > Are all disk IO operations actually locked during file creation ?  or was thisN > a case of the MSCP process just not getting enough CPU while another process > was busy creating the file ? > O > Filling a 10 gig drive ona microvax II is taking longer than expected :-( :-( 	 > ;-( :-(  >   	 JF Mezei,   C In case if you forgot, make sure the security on the disk is turned  off - initialize on creation.    Daryl Jones   $ > Also, while I have your attention: > N > I have a DILOG SQ739 controller which makes a scsi drive appear as MSCP RA81 > drive to VMS.  > N > In the case of real SCSI drives attached directly to VMS, the recommendationO > is to have automatic error correction/revectoring disabled and let VMS handle G > this.  In the case of my setup, would this still apply, with the SCSI P > controller translating the SCSI reports into MSCP reports, or should I let the > drive fix its own blocks ?   ------------------------------  % Date: Mon, 23 Dec 2002 18:09:08 -0500  From: Everhart <ge@gce.com> / Subject: Re: Disk locked during file creation ? + Message-ID: <au852u$d00$1@bob.news.rcn.net>   A I bet you have highwater marking on in that volume and are seeing > the time the XQP is busy zeroing the blocks. This probably has7 a volume lock and a high priority process accessing the C disk as fast as it can. Most of the time highwater does not do this B stop and zero operation, but when setting EOF to the end of a hugeD file, it will cause the file to be zeroed. The creating process' XQPE is of course busy. In fact it could be that the volume gets locked at F this time. Normally file creation does not have to lock the volume forI any long times though. Remember highwater marking is a security operation F and zeroing is required to be uninterrupted so that permissions of theI created file cannot be used to access old info on the blocks. That pretty E well requires that it block other process access until the blocks are F safely zeroed. Otherwise another process with the same user might haveD rights to read the file blocks with $qio (which does not respect RMS locks).  Daryl Jones wrote:i > JF Mezei <jfmezei.spamnot@vl.videotron.ca> wrote in message news:<3E06460C.370760A9@vl.videotron.ca>...  > O >>I am creating some 1.5 million block empty sequential files , and during that ; >>time, I find that all other IO on the disk seems blocked.  >>P >>Are all disk IO operations actually locked during file creation ?  or was thisN >>a case of the MSCP process just not getting enough CPU while another process >>was busy creating the file ? >>O >>Filling a 10 gig drive ona microvax II is taking longer than expected :-( :-( 	 >>;-( :-(  >> >  >  > JF Mezei,  > E > In case if you forgot, make sure the security on the disk is turned  > off - initialize on creation.  > 
 > Daryl Jones  >  > $ >>Also, while I have your attention: >>N >>I have a DILOG SQ739 controller which makes a scsi drive appear as MSCP RA81 >>drive to VMS.  >>N >>In the case of real SCSI drives attached directly to VMS, the recommendationO >>is to have automatic error correction/revectoring disabled and let VMS handle G >>this.  In the case of my setup, would this still apply, with the SCSI P >>controller translating the SCSI reports into MSCP reports, or should I let the >>drive fix its own blocks ?   ------------------------------  % Date: Mon, 23 Dec 2002 21:06:59 -0400 ( From: JF Mezei <jfmezei@vl.videotron.ca>/ Subject: Re: Disk locked during file creation ? / Message-ID: <3E07B31C.A4CE376D@vl.videotron.ca>    Everhart wrote:  > C > I bet you have highwater marking on in that volume and are seeing . > the time the XQP is busy zeroing the blocks.  L I owe you a chocolate bar. The volume was initialised with highwater marking by default.   H During day to day operations, would removing highwater offer significant performance improvement ?    ------------------------------  # Date: Tue, 24 Dec 2002 04:06:37 GMT 0 From: HARANGOZO CSABA   <spameater@spam.invalid>/ Subject: Re: Disk locked during file creation ? 4 Message-ID: <h3RN9.47$3q3.3902@nasal.pacific.net.au>  ) JF Mezei <jfmezei@vl.videotron.ca> wrote:  > Everhart wrote:  >>  D >> I bet you have highwater marking on in that volume and are seeing/ >> the time the XQP is busy zeroing the blocks.   N > I owe you a chocolate bar. The volume was initialised with highwater marking
 > by default.   J > During day to day operations, would removing highwater offer significant > performance improvement ?   3 	Well, depends on your definition of "significant". , 	Obviously, less overhead means improvement.A 	Taking into account older systems ( and disks and controllers ), A 	the improvement could be quite noticable. If site policy doesn't 3 	require it, turning it off is not a bad idea, IMO.    						Cheers,  Csaba  J  -------------------------------------------------------------------------H   CSABA I. HARANGOZO  |d|i|g|i|t|a|l|  csabah(at)zipworld(dot)com(dot)auJ  -------------------------------------------------------------------------;    EARTH::AUSTRALIA:[SYDNEY]HARANGOZO.CSABA;1, delete? [N]:    ------------------------------  % Date: Mon, 23 Dec 2002 22:53:42 -0600 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> + Subject: Re: drop modem line from decserver ' Message-ID: <3E07E856.79F13C5A@fsi.net>    JF Mezei wrote:  >  > "David J. Dachtera" wrote:H > > I believe he's looking for something automated. That is, in case theK > > user CTRL+\ out of SET HOST/DTE, he wants the server to drop DTR (or do 6 > > something else) so the modem will go back on-hook. > M > If the outbound line is set to be a "MODEM" with "DIALUP" and "HANGUP", and L > the terminal line has not been allocated, then getting out of SET HOST/DTEO > will result in VMS dropping DTR (or telling the terrminal server to drop DTR) 7 > which should result in the outbound modem hanging up.   . ...if the port is likewise set DIALUP ENABLED.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 23 Dec 2002 17:26:06 -0800. From: spamsink2001@yahoo.com (Alan E. Feldman)0 Subject: Re: F$GETDVI ("AVL") in an IF statement= Message-ID: <b096a4ee.0212231726.58d04af4@posting.google.com>   ` "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3E068939.25BCE6E7@fsi.net>... > JF Mezei wrote:  > >  > > Z wrote:@ > > > It fails big time because F$GETDVI causes an error instead > > > of returning "FALSE."  > > E > > But why does F$GETDVI cause an error instead of returning false ?  > H > Because you're confusing "exists" with "available". A device can existF > but not be available - see HELP SET DEVICE /AVAILABLE. A device thatI > does not exist cannot be examined to see whether or not it's available.   E Ah, but if you can't find it, it's certainly not available. It's also D not mounted. Now to determine whether it's a file-oriented device or% not, for that you have to examine it.   @ IOW: If it were available, you could examine it. Since you can'tE examine it, it isn't available. (!) (We are allowed to put logic into  lexical functions.)   E Even better is MNT, mounted. Is the device mounted? Well, we can list F all the mounted devices. If the one in question isn't listed, it isn't mounted, now is it?   F You can lock a sock in a safe to which you don't know the combination.B Is the sock available? NO! But you can't examine the sock. But you know it's unavailable.  D Now, please. We're *really* getting too deep into semantics here. ItF seems to me that DEC simply chose a simple way to do this: If a deviceF doesn't exist, other F$GETDVI functions for that device will return anE error. It's a choice the VMS engineers made. And this choice provides D more information from the F$GETDVI call. You can get true, false, orF an error. If you get an error, you know something is wrong. If you get= true or false, you know the requested status of the device in 	 question.   : > Remember: a device must exist before you can examine its > characteristics.  9 Yeah, but availability is a status, not a characteristic.   D Also, suppose you utilize F$GETDVI("DKA983","EXISTS") and the device= doesn't exist. How can you then examine it to determine if it ? exists?!!! You can't! Because it doesn't exist! So this lexical D function should either return TRUE, or an error about not being able@ to examine the device to determine whether it exists because the8 lexical function can't find it because it doesn't exist.  ? Again I have turned logic on its head!!! Highly illogical. Hah!   E (Where's Mr. Spock when you need him? ~"Why, thank you doctor.~") (~" B = approx. quote -- I can't swear I got it right -- from the "Court> Martial" Episode, IIRC. Mr. Spock: the ultimate straight man.)   > > Seems like aR > > big bug to me, especially when "EXISTS" does function properly and won't cause( > > an error if the drive doesn't exist. > 0 > Documented / expected behavior, in both cases.  B Correct, except for cases like F$GETDVI("NODEX::DKA200:","EXISTS") which returns a warning.   R > > I guess I can understand the issyue of what do yo return when someone requestsR > > the number of free blocks on a device that is not known.  But shouldn't one be2 > > able to test for AVL and get a true or false ? > * > Is your personal kramblefratz available?  C No. If it were, I could produce it. Since I can't produce it, it is B not available. And I already know I don't have one. Similarly, theD system knows it doesn't have the nonexistent device, and so can justE as easily determine it to be unavailable. Logic is not forbidden from  Lexical Function programs.  @ > How can you know that until you know whether or not you have a > kramblefratz?   E Well, F$GETDVI could have been designed to check for existence if the E device is not found, report it as unavailable. But it wasn't designed 	 that way.    > Quibbledunk? Same issue. > 7 > ...or would also expect SHOW DEVICE QBP0: to display:  >  > $ SHOW DEVICE QBP0:/FULL > B > Unknown device nodename$QBP0:, device type unknown, offline, not3 > mounted, not shareable, not available to cluster,   D Excellent!!! This is truly an excellent argument in favor of the theE choice the VMS engineers made. *I*, for one, certainly would not want C to see such a message for a non-existent device. Good work. This is  the best part of your post.   I > How can you examine a DCB that not exist? How can a non-existant device 
 > have a DCB?   E You can't, but you can use reason, logic, what have you, to determine  some things as explained above.   ) NTL, the choice DEC made is okay with me.    Disclaimer: JMHO  &-)  Alan E. Feldman    ------------------------------  % Date: Mon, 23 Dec 2002 15:27:44 -0800 $ From: Shane Smith <ssmith@icius.com>* Subject: RE: First Hammer performance test0 Message-ID: <01C2AA97.D92C7BF0@sulfer.icius.com>  5 Just found an English translation of the C't article:   ) http://www.heise.de/ct/english/02/26/018/    Enjoy.   Shane    -----Original Message-----% From: Dirk Munk [mailto:munk@home.nl] * Sent: Saturday, December 21, 2002 12:46 AM To: Info-VAX@Mvb.Saic.Com & Subject: First Hammer performance test    F The German magazine CT has issued a first comparisation test between aA 1.2 GHz Hammer, a 1.2 GHz Athlon XP and a 2.2 GHz Pentium 4 (real  clock speeds).  C All tests were done in 32 bit mode, Later tests will use the 64 bit  mode too of course.   E The Hammer was a lot faster than the Athlon XP, and a bit slower than A the Pentium. Using AMD's normal way of branding CPU's, the Hammer B could be branded as a 2000+ processor (=comparible to a 2 GHz P4).  D The memory access in particular was extremely fast, also much faster as with the P4.   B Keeping in mind that this Hammer was just a prototype, the editorsF expect that the entry model 2 GHz Hammer can be branded as a 3300+ CPUF when it hits the market in Q1/Q2 2003. And since the Hammer also has aC Alpha EV7 like possibilty for interprocessor communication, I can't E wait to see the first tests for multi-cpu servers and workstations.     E Somewhere in the next year I will buy myself a new PC. At that time I D will have the choice between pure 32 bit CPU's, a CPU that is a veryE fast 32 bit CPU and (most likely) a (very?) fast 64 bit CPU (so I can = run 32 bit and 64 bit windooz as well as 64 bit Linux), and a F (expensive) CPU that is a very slow 32 bit CPU, and a fast 64 bit CPU.  F Three guesses what I'm going to buy if the price is somewhat ok. Now IC just wonder what the new 'industry standard' CPU is going to be for D what is now known as the Wintel market. Perhaps in two years we will) talk about the WinAmd or WinAth market ?     ------------------------------    Date: 23 Dec 2002 18:58:16 -0800( From: bob@instantwhip.com (Bob Ceculski)* Subject: Re: First Hammer performance test= Message-ID: <d7791aa1.0212231858.39b8a032@posting.google.com>   ^ Dirk Munk <munk@home.nl> wrote in message news:<fc680vs3ulrssfthkirarejoj73lcci54e@4ax.com>...H > The German magazine CT has issued a first comparisation test between aC > 1.2 GHz Hammer, a 1.2 GHz Athlon XP and a 2.2 GHz Pentium 4 (real  > clock speeds). > E > All tests were done in 32 bit mode, Later tests will use the 64 bit  > mode too of course.  > G > The Hammer was a lot faster than the Athlon XP, and a bit slower than C > the Pentium. Using AMD's normal way of branding CPU's, the Hammer D > could be branded as a 2000+ processor (=comparible to a 2 GHz P4). > F > The memory access in particular was extremely fast, also much faster > as with the P4.  > D > Keeping in mind that this Hammer was just a prototype, the editorsH > expect that the entry model 2 GHz Hammer can be branded as a 3300+ CPUH > when it hits the market in Q1/Q2 2003. And since the Hammer also has aE > Alpha EV7 like possibilty for interprocessor communication, I can't G > wait to see the first tests for multi-cpu servers and workstations.    > G > Somewhere in the next year I will buy myself a new PC. At that time I F > will have the choice between pure 32 bit CPU's, a CPU that is a veryG > fast 32 bit CPU and (most likely) a (very?) fast 64 bit CPU (so I can ? > run 32 bit and 64 bit windooz as well as 64 bit Linux), and a H > (expensive) CPU that is a very slow 32 bit CPU, and a fast 64 bit CPU. > H > Three guesses what I'm going to buy if the price is somewhat ok. Now IE > just wonder what the new 'industry standard' CPU is going to be for F > what is now known as the Wintel market. Perhaps in two years we will* > talk about the WinAmd or WinAth market ?  ? and I am waiting to see what lawsuits hp files againset amd ... = some alpha people are there and most likely stole alpha stuff @ just like intel did, only Palmer was paid off to let them do it,> but Carly will be different ... hp should collect some hefty $ for patent violations ...    ------------------------------  % Date: Mon, 23 Dec 2002 22:49:02 -0400 ( From: JF Mezei <jfmezei@vl.videotron.ca>* Subject: Re: First Hammer performance test/ Message-ID: <3E07CB1D.31E5D7BF@vl.videotron.ca>    Bob Ceculski wrote: @ > but Carly will be different ... hp should collect some hefty $ > for patent violations ...   K I am not sure that HP will have much weight on such an issue. Consider that N HP/Compaq abandonned Alpha development and gave away its designs to Intel. NotJ sure a judge would see how AMD's use of Alpha technologies would impair HP" since HP has stopped making chips.  J And Intel can't sue AMD sicne Intel doesn't own the rights to Alpha (yet).   ------------------------------   Date: 24 Dec 02 07:08:03 +0100) From: p_sture@elias.decus.ch (Paul Sture) B Subject: Re: How to interpret extended File IDs - e.g. (86051,3,0)) Message-ID: <HxFrxhB2Q1jP@elias.decus.ch>   c In article <BHeoDRsPegJj@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: W > In article <akgVE9fgNQ+V@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes: C >> I have a utility which scans disks to check for file protection. A >> It just broke, when it encountered a file with ID (86051,3,0).  > @ > I suppose you won't use this as an excuse to buy a license for' > LJK/Security to do the same thing :-)  >    :-)    D >> macro FIB$W_FID_RVN = 8,0,16,0 %;       !  RELATIVE VOLUME NUMBER< >> macro FIB$B_FID_RVN = 8,0,8,0 %;        !  SHORT FORM RVNB >> macro FIB$B_FID_NMX = 9,0,8,0 %;        !  EXTENDED FILE NUMBER > F >> The question is how to interpret that. Is there some flag somewhere6 >> which says that the extended file number is in use? > D > I believe FIB$B_FID_NMX is always used for ODS-2/5 disks, and that: > an ODS-2/5 volume set cannot have more than 255 members. > L > When dealing with ODS disks, LJK/Security unconditionally uses expressions > of the form: > < > 	( UNSIGNED_LONGWORD ( FIB.FID.FID_RVN.FID_NMX ) * 65536 ). > 	    + UNSIGNED_LONGWORD ( FIB.FID.FID_NUM ) > 1 Many thanks for the that. Definitely appreciated.    --  
 Paul Sture Switzerland    ------------------------------   Date: 24 Dec 02 07:31:31 +0100) From: p_sture@elias.decus.ch (Paul Sture) B Subject: Re: How to interpret extended File IDs - e.g. (86051,3,0)) Message-ID: <PIgNcVTu6uCz@elias.decus.ch>   U In article <00A18DD9.F5ACCC8C@SendSpamHere.ORG>,   VAXman-  @SendSpamHere.ORG writes: W > In article <akgVE9fgNQ+V@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes:  >>A >>From STARLET.REQ (Alpha V7.3-1), the following shows the latest E >>layout, with the last byte of the 3 word array being used to extend  >>the first word.  >>4 >>literal FIB$S_FID = 6;                  !  FILE ID8 >>macro FIB$W_FID_NUM = 4,0,16,0 %;       !  FILE NUMBERA >>macro FIB$W_FID_SEQ = 6,0,16,0 %;       !  FILE SEQUENCE NUMBER C >>macro FIB$W_FID_RVN = 8,0,16,0 %;       !  RELATIVE VOLUME NUMBER ; >>macro FIB$B_FID_RVN = 8,0,8,0 %;        !  SHORT FORM RVN A >>macro FIB$B_FID_NMX = 9,0,8,0 %;        !  EXTENDED FILE NUMBER M >>macro FIB$W_FID_DIRNUM = 4,0,16,0 %;    !       Directory number of File-Id J >>macro FIB$L_FID_RECNUM = 6,0,32,0 %;    !       Record number of File-ID >>macro FIB$W_DID = 10,0,0,0 %;  >>H >>It appears that the "extended file number" field was introduced around >>the era of VMS  V7.1.  > " > You've been asleep at the wheel. > A More apt an analogy than you might realise. Before my present job A I thought I would be lucky to ever sit behind a VMS wheel again.   > ( >>The question is how to interpret that. >  > +--------+--------+--------+ > |   NMX  |       NUM       | > +--------+--------+--------+ >  > Creating a 24 bit value. >   ! Concise and to the point. Thanks.  >  --  
 Paul Sture Switzerland    ------------------------------  % Date: Tue, 24 Dec 2002 02:00:38 +0000 1 From: Steve Reece <SYSTEM@ipl.demon.co.nospam.uk> > Subject: HP-CUO Membership (Re: DECUS^H^H^H^H^HCUO membership)6 Message-ID: <3E07BFC5.3239660D@ipl.demon.co.nospam.uk>  5 I've been volunteered to put a few words to this one.   @ First, I understand that there has been some discussion off-lineH regarding the subscription year where a member of HP-CUO (the user groupE formerly known as CUO-UK and known before that as DECUS UK) joined in ; July and has now received a renewal for January to January.   H What's important to remember overall is that the user group is providingA the users with new and improved services which we hope will be of  benefit to you.   D This week, you should receive a copy of Ping, the technical magazineD published by the HP/Works user group.  This is being sent out in theG hope that HP-CUO members will find it useful and educational.  I have a ; vested interest here in that I have an article in there....   D A copy of HP User, the magazine from the HP-CUA user group should be8 finding its way to 'our' members in the near future too.  F I don't believe that there has been a significant increase in personalH membership levels for quite some time.  During that time, the user groupG has had to become self sufficient and able to pay its own way.  I guess F that the most similar change in the past was the transition from beingD free to charging a membership.  This change was made because fundingC from Digital was not forthcoming and we, as a user group, wanted to F continue to provide a service to our members.  As such, members had toH start paying for the privilege of membership rather than having it given* to them by Digital funding the user group.  B As well as magazines and the hobbyist license program for OpenVMS,E HP-CUO members get regular mailings telling you about the events that @ are planned and which you may attend.  Again, there is a fee forD attending these events as they have to turn in a surplus in order toH continue the planning of subsequent meetings.  That a user does not wishA to attend any events is up to the user.  We try to put on a broad @ program of events covering as many of the operating environmentsE supported/shipped by HP as possible.  You say that you used to attend E events when you were in a big VMS shop Patrick.  Surely it is in your C interest to see that HP-CUO run as many events as possible in areas > other than pure VMS so that we might interest you in coming to
 something?  B With the merging of three user groups into one, it is possible forA HP-CUO members to attend meetings being organised by HP/Works and G HP-CUA, in addition to those organised by HP-CUO.  The other two groups  have a reciprocal arrangement.  F Equipment has been made available in the past via the CEPS program and< it is hoped that this facility will be made available again.  D With discounts on HP training being offered by the 'classic' HP userE groups, I would suspect that we will be able to accrue these benefits H once the format of the new training programmes have been settled and the: interactions between the user groups have been sorted out.  E One thing confuses me in all of the discussion on this thread though:   @ If you want to go to an event, why not go to a local (i.e. UK orE European) event and make your opinions known.  If you only go to a US H event then you'll continue going to the US and not see that changes have occurred in the UK. E If you want a service that is offered to you locally then why not try F and support the local group and get that service from them?  That way,A it's more likely that the UK group will hear your irritations and ' comments and can help to sort them out.   E If you tell us what you like and what you want, we've got a chance of F satisfying your needs in terms of being the user group that you want. E If you keep quiet and don't tell us, we'll never know and we probably  won't meet all of your needs.    Steve.     Patrick Caulfield wrote: > 8 > On Fri, 20 Dec 2002 09:30:58 +0000, Shane Smith wrote: >  > >  > > H > > If anyone's really stuck with this and needs a US mailing address toI > > join, drop me a note offline and we can come to some arrangement. VMS $ > > should be a basic human right... >  > Sounds like a good idea  > " > /me mails a friend in California > I > I can afford the 55 quid, its just that the price hike really annoys me H > because all I'm paying for is for the UK chapter to send me flyers for$ > events I'm never going to attend*. > L > If I'm going to have to pay that amount of money to run VMS as a hobby I'dO > rather pay it to David Cathay at Montagar for his running of the hobbyist webV > site.4 > 	 > Patricke > G > * Not that I've anything against those events, I went to a lot when Ii > worked at a VMS shop.e   --  G "A shadow fell over her face; clear, as if the composure were rent like E a veil.  And her lips parted, but only with a short intake of breath.tA Then she said, 'Well, then you are right.  Indeed, we are even.'" % 		Louis, "Interview with the Vampire"    ------------------------------  % Date: Tue, 24 Dec 2002 01:36:17 +0000 1 From: Steve Reece <SYSTEM@ipl.demon.co.nospam.uk>A? Subject: Re: Is it possible to add a cluster node w/o decnet??? 6 Message-ID: <3E07BA11.532FD8A4@ipl.demon.co.nospam.uk>  H IIRC Jim, VMS 5.5-2 would expect to be running DECnet within the clusterF to be able to do things like use monitor across cluster nodes.  DECnetC management capabilities would be required in order to use satellitelD booting as the LANCP interface enhancements to deal with MOP did not appear until about v6.2.  D Similarly, I would doubt that CONFIG_CLUSTER_LAN.COM would have beenG seen before v6.2 as it would only be there once LANCP was being shippedr! with MOP management capabilities.V   This is from memory so E&OE.   Steve.   Jim Agnew wrote: > A > subject says it all... the vms 5.5-2 clusterconfig.com seems toaI > want/need decnet running, but with the advent of the new cslg, it looksA< > like decnet is gone (or i had too much Christmas cheer...) > B > So before twiddling with this, has anyone been there, done that? >  > jiml   -- nG "A shadow fell over her face; clear, as if the composure were rent likenE a veil.  And her lips parted, but only with a short intake of breath. A Then she said, 'Well, then you are right.  Indeed, we are even.'"-% 		Louis, "Interview with the Vampire"-   ------------------------------  % Date: Mon, 23 Dec 2002 22:30:08 -0600z. From: Lyndon Bartels <lbartels@pressenter.com> Subject: Link errors.e. Message-ID: <3E078E70.74C7F14E@pressenter.com>   Alpha VMS 7.3-1, (Latest ECOs.)  CC 6.5      5 On my port of PostgreSQL.... I'm getting link errors:p  = %LINK-W-USEUNDEF, undefined symbol DECC$GXSNPRINTF referencedl3 %LINK-W-USEUNDEF, undefined symbol ERRNO referencede    2 I included the seemingly appropriate header files.  % I feel like I'm missing something....t       Thanks in advance,   Lyndon     -- nG My opinions are mine and mine alone. They seldom align with those of myt	 employer.d    H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.   ------------------------------  % Date: Tue, 24 Dec 2002 00:38:05 -0400 ( From: JF Mezei <jfmezei@vl.videotron.ca> Subject: Re: Link errors.m/ Message-ID: <3E07E4A5.417F175B@vl.videotron.ca>m   Lyndon Bartels wrote: ? > %LINK-W-USEUNDEF, undefined symbol DECC$GXSNPRINTF referencedm5 > %LINK-W-USEUNDEF, undefined symbol ERRNO referencedy    ) Have no idea of what GXSNPRINTF is about.t  5 But for errno, here is what the older doc has to say:n? (and looking at errno.h, it seems to be defined as a CMA* call.t   9.3  Error Codes (<errno.h>)  ;    The<errno.h> header file defines several macros used forw    error reporting.y    	    Macrosi      EDOM.	    ERANGE-    =        Error codes that can be stored inerrno. They expand tor@        integral constant expressions with unique nonzero values.        Variable or Macro      errno    =        An external variable or a macro that expands to a mod-K>        ifiable lvalue with typeint, depending on the operating        system.  <        Theerrno variable is used for holding implementation-A        defined error codes from library routines. All error codest?        are positive integers. The value oferrno is 0 at programr>        startup, but is never set to 0 by any library function.B        Therefore,errno should be set to 0 before calling a library-        function and then inspected afterward.n   ------------------------------  % Date: Mon, 23 Dec 2002 22:45:55 +0100e6 From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne@vajhoej.dk># Subject: Re: OpenVMS Branding Querya) Message-ID: <3E078413.7070508@vajhoej.dk>o   JF Mezei wrote:*   > Hans Vlems wrote:t9 >>The DECWindows logo changed to Compaq with VAX/VMS 7.3.e > I > Isn't that ironic ... Didn't 7.3 get released at a time where Curly hadP= > already announced his intentions to sell himself to Carly ?i    
 Sounds rigth.t    B And very funny from a general perspective, but very understandable= from a software development perspective. I assume there are a 9 quite some time from code freeze to release to customers.i   Arne   ------------------------------  % Date: Tue, 24 Dec 2002 01:11:13 +0000t1 From: Steve Reece <SYSTEM@ipl.demon.co.nospam.uk>n# Subject: Re: OpenVMS Branding Queryh6 Message-ID: <3E07B431.40BAAB37@ipl.demon.co.nospam.uk>  H OpenVMS v7.2-1 (for Alpha only, obviously) carries the Compaq branding. E The release date is alleged to be July 1999 by the covering letter inu one of my sets.   H I have a feeling (though it's not confirmed as I don't have a set of 7.2G OS disks to hand) that the 7.2 distributions were actually the first to  carry Compaq branding.   Steve.     Doug Quebbeman wrote:  >  > Say, > 7 > At what version does OpenVMS cease carrying a Digitals0 > branding and start to carry a COMPAQ branding? > 	 > Thanks,w	 > -doug qt   -- dG "A shadow fell over her face; clear, as if the composure were rent likeoE a veil.  And her lips parted, but only with a short intake of breath.6A Then she said, 'Well, then you are right.  Indeed, we are even.'"t% 		Louis, "Interview with the Vampire"o   ------------------------------  + Date: Tue, 24 Dec 2002 01:18:02 +0000 (UTC) 3 From: vance@alumni.caltech.edu (Vance R. Haemmerle)i# Subject: Re: OpenVMS Branding Querys+ Message-ID: <au8cka$drt$1@naig.caltech.edu>h  5 In article <au7l71$57o0v$1@ID-143435.news.dfncis.de>,O! Hans Vlems <hvlems@iae.nl> wrote:r8 >The DECWindows logo changed to Compaq with VAX/VMS 7.3.    H   Are you sure that change wasn't associated with DecWindows Motif 1.2.6 and not specifically VMS 7.3?p    H >"Doug Quebbeman" <dhquebbeman@theestopinalgroup.com> schreef in bericht8 >news:7368b27e.0212230416.4b6e2d11@posting.google.com... >> Say,e >>8 >> At what version does OpenVMS cease carrying a Digital1 >> branding and start to carry a COMPAQ branding?a    F    VAX/VMS 7.2 still says "Copyright Digital Equipment Corporation..."C when it starts up.  I don't know about Alpha 7.2, 7.2-1, 7.2-2 etc.S   -- Vance Haemmerle  vance@alumni.caltech.edu   ------------------------------  % Date: Mon, 23 Dec 2002 21:08:27 -04000( From: JF Mezei <jfmezei@vl.videotron.ca># Subject: Re: OpenVMS Branding Querye/ Message-ID: <3E07B374.D31A8535@vl.videotron.ca>O   "Vance R. Haemmerle" wrote: H >    VAX/VMS 7.2 still says "Copyright Digital Equipment Corporation..."  P As it should continue to do. Pointless to change the name every couple of years.   ------------------------------   Date: 24 Dec 2002 04:23:16 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com># Subject: Re: OpenVMS Branding Query2+ Message-ID: <au8nfk0u2v@enews3.newsguy.com>p  ) JF Mezei <jfmezei@vl.videotron.ca> wrote:D: >> The DECWindows logo changed to Compaq with VAX/VMS 7.3.  I > Isn't that ironic ... Didn't 7.3 get released at a time where Curly hadn= > already announced his intentions to sell himself to Carly ?a  L IIRC, 7.2 came out right after the announcement that Compaq was buying DEC. @ I remember being surprised that the DECwindows logo for 7.2 said d|i|g|i|t|a|l.   		Zane   ------------------------------  # Date: Sun, 22 Dec 2002 19:41:48 GMTt- From: bdc@world.std.com (Brian 'Jarai' Chase) 6 Subject: Re: OT: Whoa! Is Sun aiming at VMS's jugular?& Message-ID: <H7JDDo.Ju1@world.std.com>  6 In article <atfjnu$13lnoo$1@ID-135708.news.dfncis.de>,) Bill Gunshannon <bill@cs.uofs.edu> wrote:v/ > In article <T_AK9.347688$QZ.51138@sccrnsc02>,y/ > 	"Mark E. Levy" <levy@sysman-inc.com> writes:n  O > > David, I keep trying to tell you but you won't listen. You don't understand- > > the underlying hardware. a > I > I' am hardly an expert on PC hardware (I don't particularly like it andeJ > really wish the 68K family had one the original position as the heart of@ > the IBM PC.) but I have a problem with some of these comments.  I I despise PCs as well, but Mark's claims of Microsoft Windows instability I being because of shoddy commodity hardware are greatly exaggerated.  I do J agree that the engineering and quality of enterprise class systems made by? IBM, Sun, and HP are going to make them more reliable.  But thetH instability of Windows is about 99.9% the fault of the operating system,H and maybe 0.1% the fault of lesser hardware.  At my place of employment,G we run Windows NT, Windows 2000, and Linux on identical PC platforms.  :H The Windows systems are lucky to see uptimes of more than a few weeks.  H The Linux systems tend to see uptimes of many months.  The #1 reason forG system reboots on our Windows boxen is due to OS crashes; the #1 reasontG for system reboots on Linux boxen are due to OS upgrades.  My main *BSDiD box has an uptime of nearly a year without the slightest hiccup--I'mF already two OS release versions behind on it; it seems a shame ruin my uptime with an OS upgrade.  O > > 1. Bank-switched memory, and no real memory management. Let's not even talk?+ > > about that memory hole between 640K-1M.I > J > Let's.  To the best of my knowledge, the hole was deliberately put thereL > when memory first went above 640K in order to be backwards compatable withL > device that mapped their own on board memory right after the original 640KL > PC address space (like the VideoBlaster!)  I am almost certain that all orK > at least most modern boxes that even support that option have the abilityo > to turn it off.a  E I'm also not really sure what Mark means by the IA32 architecture nottJ having real memory management.  Certainly all the Unix variants I've used,J both free and commercial, make extensive use of the IA32 memory managementM facilities.  These are outlined in the _80386 Programmer's Reference Manual_:-K <http://www7.informatik.uni-erlangen.de/~msdoerfe/embedded/386html/c05.htm>j  P > > 2. Polled interrupt structure vs. VAX/Alpha vectored interrupts (this is oneO > > reason why VMS can recognize new devices on-the-fly and NT needs to go thru   > > "add new hardware" bullshit) > L > Can't say how much of a problem the difference in interupt structure wouldN > be, but blaming it for the shortcomings of NT isn't the answer.  Under LinuxG > (and probably the BSD's as well) new devices can be found on the fly.   H Yeah.  My impression with item 2 above is that Mark hasn't any clue whatC he's talking about with how devices are detected and interrupts arelG serviced on commodity IA32 platforms--at least not on operating systemsgH like Unix.  Maybe those are problems with Windows, I don't know, but theE limitations he mentioned mentioned certainly aren't a function of theo underlying hardware.  L > > 3. Lack of processor modes (Kernel, Exec, Super & User). This is vitallyM > > important to VMS' security. This is the single most difficult obstacle toa
 > > overcome.a > J > Except that the fact that the VAX can be successfully emulated (with allJ > it's modes) and run VMS means at worst you would have needed to add someI > level of hardware abstraction between VMS and the real hardware, but it M > should have been doable.  Or are you saying that all these business runningdE > VMS on CHARON emulators commercially are all hackable VMS systems??p  H And more significantly, Mark is mistaken about the lack of this facilityH in the IA32 architecture.  Starting with the 80386, Intel have supported@ four privilege levels.  Taken from section 6.3.1.3 of the _80386 Programmer's Reference Manual_:   J ->  The concept of privilege is implemented by assigning a value from zeroF ->  to three to key objects recognized by the processor. This value isF ->  called the privilege level. The value zero represents the greatestD ->  privilege, the value three represents the least privilege. [...] -> MI ->  Figure 6-2  shows how these levels of privilege can be interpreted asdF ->  rings of protection. The center is for the segments containing the? ->  most critical software, usually the kernel of the operatingeG ->  system. Outer rings are for the segments of less critical software.a  6 The full reference manual is available online as HTML:K <http://www7.informatik.uni-erlangen.de/~msdoerfe/embedded/386html/toc.htm>o  L > > Please, stop whining about why VMS Engineering or management won't do an. > > IA32 port and open your eyes to the facts. > D > While I agree with the fact that it is likely way too late for theC > practical consideration of an IA32 port, I disagree with the ideahE > that it was ever impossible for engineering reasons.  Management is. > another story.  E I'd also agree that it seems like, technically, IA32 would be able totH support a port of OpenVMS.  The architecture is really ugly, though.  MyG guess is that the biggest headache would be having to juggle the measlyMJ handful of registers, nearly all of them with strange and peculiar specialH behaviors.  One thing that the VAX, the Alpha, and now IA64 all have areF lots of truly general purpose registers.  No doubt VMS was designed toH make use of these forunate resources, so trying to go back and shoe-hornI it into IA32, which has half the registers of the VAX with maybe a fourthlI of their usefulness, would be a pain.  I think it'd be a bad idea at this > point.  It's too late; there are bigger things to worry about.  G My sympathies are with the OpenVMS engineers.  Given that the Alpha wastI killed and they've been left with IA64 as their only option, I don't envynI their situation in the slightest.  I do certainly hope, that during theireD IA64 porting efforts, they're making ever effort to retool the code,A making it as portable as possible.  Tying your code closely to an-K underlying hardware architecture, at some point, comes back to haunt you.  BF When the architecture dies, it'll drag your OS down like a boat anchor> around its neck.  Portability is your life raft to the future.   -brian.E -- :F --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ -----J Computers are useless.  They can only give you answers.  -- Pablo Picasso.   ------------------------------  # Date: Sun, 22 Dec 2002 20:23:54 GMTe- From: bdc@world.std.com (Brian 'Jarai' Chase) 6 Subject: Re: OT: Whoa! Is Sun aiming at VMS's jugular?& Message-ID: <H7JFBu.JrH@world.std.com>  / In article <3DFB816C.6BF54F0C@vl.videotron.ca>,f2 JF Mezei  <jfmezei.spamnot@vl.videotron.ca> wrote: > "Mark E. Levy" wrote:o  N > > the things that VMS needs (can you say PALcode?). The IA32 just *does not*( > > have the necessary hardware support, > E > Are you 100% sure of this ? I realise that for the early batches ofiI > 8086s, this may have been true. but what about current 8086 models such  > as Pentium 4 ? >lO > > 1. Bank-switched memory, and no real memory management. Let's not even talkB+ > > about that memory hole between 640K-1M.t > G > Didn't that all go away when they introduced the pentium 3 version ofpE > the 8086 with true 32 bit architecture and virtual memory support ?   I The year would've been 1986, when the 80386 was introduced by Intel.  TheuG hole between 640K-1M is a legacy of the original IBM PC design.  In the2F worst case, you can just ignore that address range of 384KB of memory.  C There were a lot of memory managment kludges for MS-DOS and earlierMH versions of Windows, like the brain damaged Expanded and Extended memoryJ systems.  Again, these had nothing to do with the inherent capabilities ofB the IA32 architecture, but rather the lack of Microsoft taking any@ initiative to make full use of the 80386, outside of their earlyG (abandoned) efforts to co-develop OS/2 with IBM.  It took them nearly a J decade, with the introduction of Windows 95, before they even began to tapG the potential of IA32 in their software.  (One can only wonder how long ) it'll take them to do the same for IA64).a  J During the late 1980s, there were a number of commercial Unix vendors withE AT&T SysV derived offerings on for the 80386 (ESIX, Interactive Unix,iE etc.).  Then in 1991, Linus Torvalds released his initial versions ofnH Linux.  Those systems made use of the virtual memory facilities of IA32,F and I'm pretty sure that the first IBM OS/2 version for the 386 did as well (circa 1988-1989?).  H If people on c.o.v actually think these things are only recent additionsI to Intel processors, and the PC world, it's no wonder you've been letting'8 the WinTel world eat your lunch!  Know your competition!   -brian.e -- cF --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ -----J Computers are useless.  They can only give you answers.  -- Pablo Picasso.   ------------------------------  # Date: Tue, 24 Dec 2002 02:21:07 GMTiL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")6 Subject: Re: OT: Whoa! Is Sun aiming at VMS's jugular?6 Message-ID: <00A18E38.86A165BC@SSRL.SLAC.STANFORD.EDU>  X In article <SACN9.2547$I64.1400@news.get2net.dk>, "Dr. Dweeb" <Dweeb@nospam.com> writes:M >There was a BLISS-IA32 compiler for years.  COMPAQ killed it along with lotstJ >of other stuff during the preliminary phases of the kill Alpha procedure.  C Which managed to whack the NT version of Rdb as collateral damage. sK (Although I really thought the Rdb group could have afforded to pick up oneiG guy to maintain the IA32 Bliss compiler, they didn't seem to think so.)    -- Alane  O ===============================================================================x0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056uM  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025CO ===============================================================================    ------------------------------  # Date: Tue, 24 Dec 2002 02:30:56 GMTyL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")6 Subject: Re: OT: Whoa! Is Sun aiming at VMS's jugular?6 Message-ID: <00A18E39.E53ADE2E@SSRL.SLAC.STANFORD.EDU>  O In article <8EA11405E59BD611BA7100104B93C260AB248A@exdel01.del.mgsl.com>, Kesavn0 Tadimeti <Kesav_Tadimeti@KeaneIndia.com> writes:   >Hi all,K >Some of you folks may have heard of a real time OS called QNX that runs ontH >intel Pentium class processors. In some production installations VMS is >being replaced by QNX.   H >So I guess it should be possible to have a reliable secure OS along theM >likes of VMS on Intel. As some one pointed out, I guess porting VMS to IntelvE >is difficult since BLISS, MACRO compilers need to be written first. s  L I don't think the existence of QNX (about which I've only heard good things,J incidentally, and which has had what seems like very sensible pricing for L a long time) tells us anything about the possibility of porting VMS to IA32.I The various BSDs seem to be pretty reliable and fairly secure (especially?M NetBSD) too.  But this has a lot to with having been designed for portability  originally.C  J It's easier to design something for a pariticular hardware platform, or toC port something that was designed to be portable, than it is to takelJ something designed for a particular platform and put it on a significantlyG different one.  While work was done in the Alpha port and more is being K done in the Itanium port to make VMS less hardware-specific, I suspect that = work isn't going to backport to 32-bit machines very readily.    -- Alant    O ===============================================================================t0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056eM  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   940250O ===============================================================================e   ------------------------------  % Date: Mon, 23 Dec 2002 22:44:18 -0400s( From: JF Mezei <jfmezei@vl.videotron.ca>6 Subject: Re: OT: Whoa! Is Sun aiming at VMS's jugular?/ Message-ID: <3E07CA02.B46B0150@vl.videotron.ca>   * Alan Winston - SSRL Admin Cmptg Mgr wrote:I > different one.  While work was done in the Alpha port and more is beingfM > done in the Itanium port to make VMS less hardware-specific, I suspect thatb? > work isn't going to backport to 32-bit machines very readily.u  N But Hammer will support 64 bits and so far appears to have far greater chances( of being commodity than that IA64 thing.   ------------------------------  # Date: Tue, 24 Dec 2002 03:48:08 GMThL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")6 Subject: Re: OT: Whoa! Is Sun aiming at VMS's jugular?6 Message-ID: <00A18E44.AE2650AF@SSRL.SLAC.STANFORD.EDU>  Z In article <3E07CA02.B46B0150@vl.videotron.ca>, JF Mezei <jfmezei@vl.videotron.ca> writes:+ >Alan Winston - SSRL Admin Cmptg Mgr wrote:aJ >> different one.  While work was done in the Alpha port and more is beingN >> done in the Itanium port to make VMS less hardware-specific, I suspect that@ >> work isn't going to backport to 32-bit machines very readily. >rO >But Hammer will support 64 bits and so far appears to have far greater chancesA) >of being commodity than that IA64 thing.n  M I'm not arguing against a Hammer port; I was specifically arguing against therI claim that QNX running on IA32 meant that VMS could (easily) run on IA32.   K I'd hope that the additional excision of platform dependencies done in the 6; course of the Itanium port would make a Hammer port easier.e   -- Alant    O ===============================================================================t0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056-M  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025bO ===============================================================================2   ------------------------------   Date: 23 DEC 2002 19:35:18 GMT4 From: karcher@thuria.waisman.wisc.edu (Carl Karcher)7 Subject: SAN recommendations mixing VMS, Solaris, Linux 6 Message-ID: <23DEC02.19351880@thuria.waisman.wisc.edu>  G We would like to implement a SAN (4 TB initially with growth to 100 TB)t with the following goals:   ; 1. Support the following platforms as hosts within the SAN: E    Sun/Solaris, Sgi/Irix [1], Linux, Windows 2000 [2] and HP Openvms.   B 2. Support a SAN shared file system between multiple heterogeneousH    clients, Solaris/SPARC and Linux/IA32 currently, as well as serve theC    same files out using NFS and CIFS. The filesystem should supportrF    large volumes and files, and be dynamically expandable. The abilityD    to add failover clustering and Hierarchical Storage Management is    also highly desirable.   .    Possible shared file system candidates are::    SGI CXFS (would require supporting SGI/Irix on the SAN)    ADIC StorNext FSi    Sun QFS + Tivoli SANergyh   Notes:  D    [1] Sgi/Irix is only necessary if CXFS is chosen as a SAN shared         file system.y  .    [2] Windows is not essential but desirable.  G I've been hoping HP's EVA would fit the bill however it doesn't support G SGI. So we are looking for a shared unix file system that the EVA wouldiJ support between Solaris and Linux. As HP is closed this week I'm wondering/ if anyone else has suggestions/recommendations.h   --G -- Carl Karcher, Waisman Computing Services, Waisman Center, UW-Madisonn9 --                  karcher.nomorespam@waisman.wisc.edu      ------------------------------  % Date: Tue, 24 Dec 2002 00:31:22 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> ; Subject: RE: SAN recommendations mixing VMS, Solaris, Linux T Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660C4F@kaoexc01.americas.cpqcorp.net>   Carl,   1 You might be interested in this SAN design guide:lH ftp://ftp.compaq.com/pub/products/storageworks/techdoc/san/AA-RMPNE-TE.p df  & Reference pg. 6-13 (note SGI included)C "ACS works in a heterogeneous environment that includes Tru64 UNIX,nC OpenVMS, Microsoft Windows NT and Windows 2000, Novell NetWare, Sun1C Solaris, HP-UX, SGI IRIX, IBM AIX, Linux x86, and Linux Alpha. Thisl, application is at the storage system level."   Regards,  
 Kerry Main Senior Consultants Hewlett-Packard (Canada) Co.! Consulting & Integration Services  Voice: 613-592-4660d Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)        -----Original Message-----> From: Carl Karcher [mailto:karcher@thuria.waisman.wisc.edu]=20 Sent: December 23, 2002 2:35 PMs To: Info-VAX@Mvb.Saic.Com-7 Subject: SAN recommendations mixing VMS, Solaris, Linux     G We would like to implement a SAN (4 TB initially with growth to 100 TB)e with the following goals:,  ; 1. Support the following platforms as hosts within the SAN:<E    Sun/Solaris, Sgi/Irix [1], Linux, Windows 2000 [2] and HP Openvms.l  B 2. Support a SAN shared file system between multiple heterogeneousH    clients, Solaris/SPARC and Linux/IA32 currently, as well as serve theC    same files out using NFS and CIFS. The filesystem should supporteF    large volumes and files, and be dynamically expandable. The abilityD    to add failover clustering and Hierarchical Storage Management is    also highly desirable.F  .    Possible shared file system candidates are::    SGI CXFS (would require supporting SGI/Irix on the SAN)    ADIC StorNext FSF    Sun QFS + Tivoli SANergy    Notes:  F    [1] Sgi/Irix is only necessary if CXFS is chosen as a SAN shared=20        file system.>  .    [2] Windows is not essential but desirable.  G I've been hoping HP's EVA would fit the bill however it doesn't supporttG SGI. So we are looking for a shared unix file system that the EVA wouldd@ support between Solaris and Linux. As HP is closed this week I'm9 wondering if anyone else has suggestions/recommendations.D   --G -- Carl Karcher, Waisman Computing Services, Waisman Center, UW-Madisona; --                  karcher.nomorespam@waisman.wisc.edu =20l   ------------------------------  % Date: Mon, 23 Dec 2002 22:27:37 -06007. From: Lyndon Bartels <lbartels@pressenter.com>, Subject: Semaphores and Shared memory on VMS. Message-ID: <3E078DD9.7F3A7449@pressenter.com>  - I'm working on the port of PostgreSQL to VMS.9  C It needs Semaphore operations, as well as Shared memory operations.n5 (sem_wait, sem_init, sem_close, shmctl, shmget, etc.)e  G Has anybody down a "port", "copy", "clone" or whatever you want to calll
 it to VMS?  8 I've been looking around and haven't found anything yet.     Thanks in advance,     Lyndon   --  G My opinions are mine and mine alone. They seldom align with those of my_	 employer.,    H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.   ------------------------------  # Date: Sun, 22 Dec 2002 15:34:06 GMT_+ From: Ryan Moore <rmoore@rmoore.dyndns.org>> Subject: Re: Spiralog anyone?;= Message-ID: <Pine.LNX.4.50.0212220728460.15041-100000@jaipur>0  * On Sun, 22 Dec 2002, Chris Clifford wrote:N > What was the reason that Spiralog was dropped? I've never used it and I seemK > to remember hearing about some problems when the filesystem filled up butmH > was it really so bad that it had to be discontinued or are there other  > reasons for its disappearance?  C Apparently there were many reasons Spiralog was discontinued.  BothhH technical reasons and "internal"/political reasons.  I remember posts toH this news group within the last 2 years that talked about those reasons.7 Maybe something like Google can search the NG archives.   J But I remember one of the "official" reasons was that the performance justJ wasn't as good as expected.  One of the points of the log-based filesystemI was to increase performance.  And I guess if you were sending data at theaE disk such you did a lot of writes and little reads, you might get the A performance increase.  But I guess the engineers felt the overall - performance wasn't as good as they had hoped.    -Ryanu   ------------------------------  % Date: Tue, 24 Dec 2002 02:10:55 +0000r1 From: Steve Reece <SYSTEM@ipl.demon.co.nospam.uk>p Subject: Re: VAX 7810's.6 Message-ID: <3E07C22F.1A9699C2@ipl.demon.co.nospam.uk>   Leigh,  H Is there some kind of timeout occurring, in which traffic is first beingF headlined for the card that is not connected, that fails so it goes to the card that is connected...?   Steve.   Leigh Bowden wrote:w > N > Route is identical i.e. one is unplugged and the other plugged in its place. >  > -- > Leigh G. Bowden.% > http://www.bowdenfamily.fsnet.co.ukm# > LGBowden#bowdenfamily.fsnet.co.ukn > +44 161 477 2526 >  > Opinions are mine alone.   -- cG "A shadow fell over her face; clear, as if the composure were rent likegE a veil.  And her lips parted, but only with a short intake of breath..A Then she said, 'Well, then you are right.  Indeed, we are even.'"e% 		Louis, "Interview with the Vampire"e   ------------------------------  % Date: Sun, 22 Dec 2002 16:23:24 +0100 ) From: Bart Zorn <B.Zorn@xs4all.nospam.nl>t Subject: Re: VM web ring & spama4 Message-ID: <3e05d8ec$0$148$e4fe514c@news.xs4all.nl>   Michael Unger wrote:; > "Jerome H. Fine" <jhfineb9rv@b9rvnospamcompsys.to> wrote:d >  >  >>>Michael Austin wrote: >>- >>>would you mind sharing that piece of code?u >> >>Jerome Fine replies: >>> >>This set up requires no code at all, but I do have to change: >>my identity whenever I post to a news group.  I am using4 >>Windows 98 SE/Netscape 4.78 for e-mail.  Sometimes= >>I forget to change my identity, so my actual e-mail addresst; >>is sent.  Overall, it has greatly reduced spam.  And whena? >>it gets too annoying, I just choose 4 more random characters.s >>: >>If everyone would do something similar and as simple, so= >>much of the spam would be impossible to deliver that within : >>a year, spam would almost disappear due to the fact that< >>harvesting e-mail addresses would no longer be useful.  In= >>fact, a coordinated effort and changing e-mails your e-maile= >>address every 2 months (with a default available that can't = >>easily be harvested) might eliminate spam rather quickly if  >>99% could not be delivered.i >> >>[...]n >  > @ > Not sure about that. I once got spam with a huge amount of BCCI > addresses, among them being "g.g@127.0.0.1". (The mail program does not/F > know about "BCC" headers and took that as plain text; so the conceptH > of "blind copies" to hide the recipients doesn't seem to be fool-proof > .. :-) > 	 > Michaelh  G Slightly OT, but the address "g.g@127.0.0.1" requires some elaboration.g  H My knowledge about this comes from the time that I was managing OpenVMS  systems with PMDF.  H The address 127.0.0.1 will return a host not found error to the sender. F The reason is, that 127.0.0.1 is not valid in the context of a RFC822 H address. However, the address "postmaster@[127.0.0.1]" is valid (and it F MUST be valid on all e-mail systems!) and the result will be that the G postmaster of the system from which the message was send, will get the sG message and thus will be notified about the ongoing spamming. Might be  # even more annoying for the spammer!u    	 Bart Zorn    ------------------------------    Date: 23 Dec 2002 18:50:12 -0800( From: bob@instantwhip.com (Bob Ceculski)7 Subject: VMS has got apps ... 3300 and counting per HP!h= Message-ID: <d7791aa1.0212231850.7cef5f62@posting.google.com>l  < this is from skhpc on the openvms.org site ... for those who say vms has no apps ...   F During the presentations by Mark Gorham, Sue Skonetskiwho was awardedF a gold- and-ruby pin for her ceaseless work in creating and ramping upD the OpenVMS Ambassador contingent symposium attendees got an earful? about the future of the OS. As previously noted, the OS port is:E proceeding according to plan, and it's likely that a production-ready A VMS-on-IPF release will be available sooner rather than later. IneA fact, a VMS V8.0 release will be coming your way Pretty Soon Now. E While the VMS V8.0 release targets a close circle of ISVsthe current C target is ~30 early adoptersthe current strategy is to release VMSoE V8.2 in ~2004 as a full production release, followed by VMS V8.4 as al: completely ready for prime time release which should debutA concurrently with with VMS-capable HP hardware. Separately, HP isdE devoting significant attention to ISV recruitment and retention, withb= a key focus on an ISV Fast Track program which aims to have 5tD third-party apps up and running on Day One. And hardly a day goes byD without another ISV bellying up to the VMS-on-IPF bar. Work proceedsC apace on software license trade-in programs; HP has committed to anrD attractive software license trade-in program, establishing terms andD conditions for third party providers remains a work in progress. TheD current OpenVMS apps base is ~3,300 apps, 100 of which were added inF the past year. An additional ~15 apps are being added to the portfolio
 each week.   ------------------------------  % Date: Mon, 23 Dec 2002 22:56:18 -0600o1 From: "David J. Dachtera" <djesys.nospam@fsi.net>r; Subject: Re: VMS has got apps ... 3300 and counting per HP!e' Message-ID: <3E07E8F2.2BC11A72@fsi.net>t  # Bob Ceculski wrote (quoting skhpc):i > [snip] > ThesF > current OpenVMS apps base is ~3,300 apps, 100 of which were added inH > the past year. An additional ~15 apps are being added to the portfolio > each week.  8 ...but nobody knows about them due to stealth marketing.   -- c David J. Dachterah dba DJE Systemsm http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 22 Dec 2002 16:33:46 +0100t5 From: "Chris Clifford" <chris.clifford@openvms.co.uk>A# Subject: Re: Vol Shadowing Question * Message-ID: <3e05db7b@news.swissonline.ch>  / "Koloth" <koloth@telocity.com> wrote in message?% news:3E042510.5090008@telocity.com...e8 > We have a disaster tolerant configuration using hsg80sK > we opted for a system disk for each site and have a shadowed common disk.eH > It took some fun to more the sys$errorlog, operator logs, UAF, startup4 > files. and other stuff over there.  if you are notE > in a disaster tolerant config why are you not just have one commongb > system disk and shadow it?  I Sounds like the shadowset consists of MSCP-served disks utilising gigabit K ethernet though so you could not have one common system disk and expect thegI systems to boot in all circumstances. See the Volume Shadowing manual forn details.  F Best to go for two system disks and move certain files which should beH consistent across the cluster to a common shadowed disk with MSCP-served5 members, e.g. UAF, RIGHTSLIST, licence database, etc.n   - Chris.   ------------------------------    Date: 23 Dec 2002 14:28:49 -08007 From: jones.computer.srv@worldnet.att.net (Daryl Jones) # Subject: Re: Vol Shadowing Questionr= Message-ID: <8a646952.0212231428.63764c65@posting.google.com>e  Q Dan Moore <dmoore@sosu.edu> wrote in message news:<3E03F886.C92DB021@sosu.edu>...r > Greetings! > 3 >     I would like a sanity check for our shadowingeJ > strategy/configuration. We have been using a starnd alone system for theG > last few years and now we are moving to a 2 node cluster connected byaH > gig Ethernet Fiber. Both are Alpha Systems running OpenVMS 7.3-1. Each: > Alpha contains one shadow set member for each data disk. > H >    We are forced to use  a sperate system disk on each node because ofF > the fiber connections. I would like any user to be able to log on toE > either system and have the exact same access to the same data. This I > requires that the system common files be available to each system, evenn' > if the other system is not available.  > J >   One thought is to mount  the system disk of our primary system on eachJ > node as a shadowset member. Then point the other  Alpha at the shawdowedJ > system disk of the primary system. This way, each Alpha will have a copyI > of the system files (And the primary system will have a shadowed systemtH > disk). We tested this and it seems to work well; however, I know thereH > could be some down sides on this strategy depending upon what fails in > what sequence. > C >   Another thougth is to move these file to one of the data disks,-I > therefore avoiding the issues that can occor with shadowed system diskse > when things go wrong.= > % >   Any advice or insight is welcome.a > 	 > Thanks,e >  > Dan3 >  >  >    Dan,  : A strategy I have employed ove the years is the following:  = 1) Make the System disk a read-only disk and controller basedh	 shadowed..C 2) Create a second so-called System disk with the modifiable systemn! files - controller based shadowedt> 3) Create application disk for all software startup software - controller based shadowing.s  @ Now this is the part I'm not sure of? (separate license,network, etc...)t  @ 4) CPU shadow (not controller based) the system disk (read-only)$ 5) CPU shadow the second system disk# 6) CPU shadow the application disk.m  3 Be careful of the merge vs copy performacne issues.:   I hope this helps.   Daryl Jones-   ------------------------------   End of INFO-VAX 2002.709 ************************