0 INFO-VAX	Sun, 03 Feb 2002	Volume 2002 : Issue 65      Contents: Asynchonous clustering Bridgeworks & DCE  Re: curses on OpenVMS 5 Re: Ford picks Alpha, beats pant off of Sun, HP, IBM! 5 Re: Ford picks Alpha, beats pant off of Sun, HP, IBM!  Re: FREEWARE 5.0 RE: FREEWARE 5.0+ Re: Help: cluster crash when new node joins " Re: How to fix corrupted RMS file?+ Re: Itanium to be "Alpha Inside " says cNet  Re: LICENSE/CHARGE Re: NTP server with UCX 4.2  Pathworks/MS Network problem Purveyor runs VMS Perl! E Re: SGI solicits Alpha customers to trade in their "orphaned" systems $ Re: WTB:Freeware 4 or 5 CD for sale?  F ----------------------------------------------------------------------   Date: 2 Feb 2002 20:54:41 GMT ! From: William <wilby98@yahoo.com>  Subject: Asynchonous clustering * Message-ID: <a3hjmh$a81$1@news1.xs4all.nl>  3 I have just acquired VMS 5.5-2 on tape which I hope 6 to load into my MicroVAX II. My MVII is encased within4 the BA23 (rackmount) enclosure and will have an RD54& disk (159MB 5.25) and TK50 tape drive.  8 Although I will leave out DECwindows on the install I am8 still concerned about lack of disk space. Since no other3 hardware will fit in the enclosure I am considering 0 setting up a cluster with my VAXstation 3100 and0 MicroVAX 3100-30. Since the MVII lacks the DELQA0 ethernet card I would like to connect through an) asynchronous line. Will this work? If so, . will the "clustering" overcome the MVII's lack, of disk space and enable paging to the disks4 on the other platforms? Will the other CPU's kick in1 when the MVII's CPU gets overburdened? What other 0 features will clustering offer with this version of VMS?   1 Any comments are appreciated. Also please suggest / any "buzzwords" that I might run through Google - to lead me further into online documentation. + Some terms I have looked into: VMS, DECnet, 
 RMS, OpenCMU.   3 Also, here is a cool link I found while "Googling". - It relates to a VAX working in the Antarctic.   6 http://www.nerc-bas.ac.uk/public/uasd/sesame/data.html   Bill
 Amsterdam, NL    ------------------------------  % Date: Sat, 02 Feb 2002 22:13:05 +0100 B From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@mail.danbbs.dk> Subject: Bridgeworks & DCE. Message-ID: <3C5C5661.C456EBA2@mail.danbbs.dk>  : Am I the only one that thinks is a shame that Compaq makes6 such an interesting product like Bridgeworks available< for download but to actually use it you need DCE Application6 Development Kit, which I think is a very rare piece of8 software. If Bridgeworks were to be a success, then they8 should make DCE available for free via download as well.6 I have no problem about the C compiler being required.8 If you are doing any development, then you are likeky to? have the C compiler installed. But who has that DCE development " kit installed ? Not many I think !   Arne   ------------------------------   Date: 2 Feb 2002 17:26:35 -0800 0 From: chris_doran@postmaster.co.uk (Chris Doran) Subject: Re: curses on OpenVMS= Message-ID: <948f0720.0202021726.371bba06@posting.google.com>   q tony.cheung@asiayeah.com (Tony Cheung) wrote in message news:<f9dc0a5a.0202010803.4e858b9c@posting.google.com>...  > D > I am trying to create a text-console application on the OpenVMS. I > would be using C/C++.  > E > I've read about the curses implementations of OpenVMS and know that D > it's not 100% compatible with other curses. Is that still true for > OpenVMS v7.3?   A They did a major rewrite in V6.something, and claimed it was much , better, but I haven't used it enough to say.  E > Is there any other easier way to create a text-console with hilight E > menus, text-entry form? Or any other higher level libraries to use? ' > Any pointer will be much appreciated.   C As others have indicated, the SMG functions are the (free) standard D for VMS, and you can buy implementations for Windows and, IIRC, UnixD if you ever need to port. Once you get your head around the conceptsB of virtual keyboards, virtual displays, and pasteboards, SMG is asC easy as curses, and a lot more powerful. ISTR there's a fairly good D menus example in the manual (or maybe it was DSN), or I can send youB the basis of a SMG-based menu program I developed (too big to post here).   Chris    ------------------------------  * Date: Sat, 2 Feb 2002 21:50:12 +0000 (UTC), From: "Richard Maher" <maher_rj@hotmail.c0m>> Subject: Re: Ford picks Alpha, beats pant off of Sun, HP, IBM!1 Message-ID: <a3hmuj$kue$1@knossos.btinternet.com>    Hi,   6 I thought Ford UK was a big VMS site. No more I guess?   Regards  Richard Maher.   3 Bob Ceculski <bob@instantwhip.com> wrote in message 6 news:d7791aa1.0201181229.4382165@posting.google.com...$ > Alpha wins big again ... as usual! > ) > http://www.theinquirer.net/18010212.htm  >  > Alpha powers Land Rover cars >  > Tru64 cluster busters ( > By Mike Magee, 18/01/2002 12:29:13 BST > H > COMPAQ UK said it has won a deal to provide Ford subsidiary Land RoverH > with 18 AlphaServer ES40s and two ES45s for project management, system! > administration and maintenance. A > The AlphaServers will use Tru64 Unix, said Compaq, and the firm G > claimed it had won the deal because of benchmarks that beat the pants  > off HP, Sun, SGI and IBM.  > F > Richard George, business manager in charge of AlphaServers at CompaqF > UK, claimed that the systems were the industry's highest performers,A > and that Land Rover was also impressed with Tru64's clustering.  > G > The firm will use the machines and software to build its own computer B > assisted engineering (CAE) system for speeding up production and > design of the autos. > H > Fourteen of the AlphaServer ES40s will come with 2GB of memory and useH > 833MHz Alpha processors. The firm didn't say what the other four ES40s
 > will be.   ------------------------------  # Date: Sat, 02 Feb 2002 22:25:30 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>> Subject: Re: Ford picks Alpha, beats pant off of Sun, HP, IBM!= Message-ID: <uHZ68.24401$%h1.8580728@typhoon.ne.mediaone.net>   7 "Richard Maher" <maher_rj@hotmail.c0m> wrote in message + news:a3hmuj$kue$1@knossos.btinternet.com...  > Hi,  > 8 > I thought Ford UK was a big VMS site. No more I guess? >   < Dunno, but CAD and CAE and the like are typically UNIX apps.   ------------------------------   Date: 2 Feb 2002 23:24:36 GMT / From: Hans Bachner <Hans.Bachner@altavista.net>  Subject: Re: FREEWARE 5.0 ) Message-ID: <a3hq6p.31.1@bachner.priv.at>   " Tom Linden (tom@kednos.com) wrote:  F > I didn't get one of these with my 7.3 distribution.  Is there a siteH > from which it can be downloaded in its entirety?  The Compaq, AFAICT, ' > isn't suitable for such an endeavour.   F What's wrong with http://www.openvms.compaq.com/freeware/freeware50/ ?   --  9 ---------------- speaking only for myself --------------- 9 Hans Bachner              E-Mail: Hans.Bachner@compaq.com ) Compaq Computer Austria / Global Services    ------------------------------  $ Date: Sat, 2 Feb 2002 19:07:48 -0800# From: "Tom Linden" <tom@kednos.com>  Subject: RE: FREEWARE 5.0 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIIECGEBAA.tom@kednos.com>   @ What's wrong with it is  that I want to download a single image,   ok so I am lazy.       > -----Original Message-----8 > From: Hans Bachner [mailto:Hans.Bachner@altavista.net]+ > Sent: Saturday, February 02, 2002 3:25 PM  > To: Info-VAX@Mvb.Saic.Com  > Subject: Re: FREEWARE 5.0  >  > $ > Tom Linden (tom@kednos.com) wrote: > H > > I didn't get one of these with my 7.3 distribution.  Is there a siteJ > > from which it can be downloaded in its entirety?  The Compaq, AFAICT, ) > > isn't suitable for such an endeavour.  > H > What's wrong with http://www.openvms.compaq.com/freeware/freeware50/ ? >  > --  ; > ---------------- speaking only for myself --------------- ; > Hans Bachner              E-Mail: Hans.Bachner@compaq.com + > Compaq Computer Austria / Global Services  >    ------------------------------   Date: 2 Feb 2002 19:32:52 -0800 1 From: keithparris_NOSPAM@yahoo.com (Keith Parris) 4 Subject: Re: Help: cluster crash when new node joins< Message-ID: <cf15391e.0202021932.53a1b94@posting.google.com>  e nothome@spammers.are.scum (Malcolm Dunnett) wrote in message news:<6vlXGldfMCHs@malvm6.mala.bc.ca>... B > I have an NI based cluster with 4 systems in it. Two are runningE > VMS 7.1 and two are running VMS 7.2-1. Today I tried to add another F > system to the cluster ( running 7.2-2 ). It joined the cluster fine,E > but when it attempted to mount a disk which is served by one of the F > other nodes the new system and the one serving the disk both crashed > with an error of:  >  > CPU bugcheck codes: > >         CPU 00 -- LOCKMGRERR, Error detected by Lock Manager  B LOCKMGRERR (and the similar CNXMGRERR) are triggered when the LockF Manager (or Connection Manager, respectively) code detects an error asE it is parsing an incoming sequenced message.  This could be caused by E a something tromping on the packet in memory on the source node after E it's assembled but before it's sent (or an undetected memory hardware F error on the originating system), by a corruption of the packet withinF an adapter or on its way across the interconnect in such a way that anB ECC error is not generated, or by something stepping on the packetE after it is received but before it is parsed (or an undetected memory D error on the receiving system).  It could also be caused by a bug in3 the VMS code which assembles or parses the packets.   D In practice, bad hardware seems to be the most prevalent cause.  VMSB tries to trigger a crash on both the sending and receiving systemsE simultaneously so you can look at the bad packet in the dump and thus B get some info as to what sort of damage occurred to the packet and@ whether the damage seems to have occurred within the source node itself or subsequently. C ------------------------------------------------------------------- C Keith Parris | parris at encompasserve dot org | VMS consulting on: C Clusters, Disaster Tolerance, Internals, Performance, Storage & I/O    ------------------------------   Date: 2 Feb 2002 13:05 CST' From: carl@gerg.tamu.edu (Carl Perkins) + Subject: Re: How to fix corrupted RMS file? , Message-ID: <2FEB200213054815@gerg.tamu.edu>  . "Bill Todd" <billtodd@metrocast.net> writes...6 }"Jack Peacock" <peacock@simconv.com> wrote in message4 }news:4ED68.281237$Bw6.6868393@news.webusenet.com... }....  }  }> I don't think even , }> RAID hardware would have caught this one. } K }No, but RAID software would have, as long as the disks weren't on the same I }(failed) controller (and possibly even if they were, if the intermittent K }failures were sufficiently rare that only one copy of any given bucket was M }affected - though RMS almost certainly isn't smart enough to avoid having to 3 }choose the valid buckets manually in such a case).  }  }- bill   C The ideal avoidance mechanism would probably be shadowing to a disk H on a completely different node, requiring a cluster of at least 2 nodes.   --- Carl   ------------------------------   Date: 2 Feb 2002 13:38 CST' From: carl@gerg.tamu.edu (Carl Perkins) 4 Subject: Re: Itanium to be "Alpha Inside " says cNet, Message-ID: <2FEB200213382717@gerg.tamu.edu>  0 Paul Repacholi <prep@prep.synonet.com> writes...7 }"Terry C. Shannon" <terryshannon@mediaone.net> writes:  } E }> Yep. I doubt that superscalar, out of order execution, and related 1 }> Alpha RISC artifacts are compatible with EPIC.  } G }> On the other hand, on-chip switching and possibly SMT are candidates % }> for inclusion in future IPF chips.  } F }What part of EPIC don't you understand? it already IS SMT. For a very }small slow value of SMT...  }-- = }Paul Repacholi                               1 Crescent Rd.,    What makes you say that?  D The parallelism must be Explicitly (as in the E in EPIC) done by theA compiler. The compiler has no clue what other threads my be doing E at the time the code is run. It clearly can not make arrangements for F instructions in any thread(s) other than those in what it is compiling at the time.  E So it could, maybe, be consider SMT-like for threads all built by the C compiler at the same time, for the threads in a single application, F but only if the compiler can actually do it - it usually won't be ableE to, if it can at all, since the timing of the various threads will be F unpredictable making it impossible to explicity produce an instructionD stream for a thread that allows instructions from some other thread,D even if in the same application, to use some of the functional unitsC on the chip (and the other thread would have to have an instruction C stream that only used those specific units that the other thread is  leaving free for it to use).    B Even if the compiler and CPU can do this, which I seriously doubt,A I would not consider it to be SMT as SMT allows instructions from @ any thread on the system to be run simultaneously whether or not& they are part of the same application.   --- Carl   ------------------------------  % Date: Sat, 02 Feb 2002 22:24:46 -0700 $ From: Lee Y T Mah <lytmah@cha.ab.ca> Subject: Re: LICENSE/CHARGE ) Message-ID: <3C5CC99E.577D45DA@cha.ab.ca>    Here your are.  C This is a Compaq AlphaServer ES45 Model 2, hardware model type 1976 I Type: A, Units Required: 100    (VAX/VMS Capacity or OpenVMS Unlimited or  Base) 4 Type: B, * Not Permitted *      (VAX/VMS F&A Server)9 Type: C, * Not Permitted *      (VAX/VMS Concurrent User) 5 Type: D, * Not Permitted *      (VAX/VMS Workstation) D Type: E, * Not Permitted *      (VAX/VMS System Integrated Products)6 Type: F, * Not Permitted *      (VAX Layered Products)* Type: G, * Not Permitted *      (Reserved)8 Type: H, Units Required: 1050   (Alpha Layered Products)2 Type: I, Units Required: 1050   (Layered Products)     Tom Linden wrote:   / > Is there a site which displays the results of  > $ SHOW LICENSE/CHARGE " > for all the differnet platforms? > H > In particular, I am interested in ES40,45. which if someone could post > would be much apreciated.    -- Lee   ; Lee Y T Mah                        Capital Health Authority ? Email: lytmah@cha.ab.ca            Information Systems, RAH CSC 4 Phone:  (780) 477-4725, 477-4233   10240 Kingsway NW? Fax:      (780) 491-5119, 491-5619    Edmonton, AB, CAN  T5H3V9    ------------------------------   Date: 3 Feb 2002 02:49:50 +0100 * From: eplan@kapsch.net (Peter LANGSTOEGER)$ Subject: Re: NTP server with UCX 4.2* Message-ID: <3c5c973e$1@news.kapsch.co.at>  k In article <1c0e37b1.0112120249.3200676a@posting.google.com>, pierre.bru@spotimage.fr (Bru, Pierre) writes: . >I already have the .CONF file witch contains: >  >server x.y.z.a  >master-clock n   ) I'd simply remove the 'master-clock' line   C >I do not use "peer" as I understand it as telling the other server E >that I am sure of the clock of my host and that I want to send it my - >time as a valuable source to verify its own.  > C >am I wrong ? do I need to include a "peer" line in my .CONF file ?    No.    --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111 2651; Network and OpenVMS system manager  Fax.    +43 1 81111 888 < KAPSCH AG      Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Sat, 02 Feb 2002 22:20:18 +0100 B From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@mail.danbbs.dk>% Subject: Pathworks/MS Network problem . Message-ID: <3C5C5812.69B68C60@mail.danbbs.dk>   Setup:.   VMS Alpha 7.2-1, PathWorks 6.0B, PDC ARNELAN3   Win9x, PathWorks 7.1A, logon workgroup not domain <   WinNT 4.0 SP6A, PathWorks 7.1A, logon workgroup not domain  ; I logon the Win9x systems just fine as user "arne" password / "foobar". No problems - drives get mapped fine.   E I login the WinNT system as local user "arne" with password "foobar". E No problems. But when I try to map the drives as user "\ARNELAN\arne" A then it does not work with password "foobar", but it do work with A password "FOOBAR". So everytime I logon the NT box I have to give @ the password in lowercase to NT and in uppercase for the shares.   And I do not understand why !    Any good ideas ?   Arne   ------------------------------   Date: 2 Feb 2002 11:53:50 -0800 ( From: bob@instantwhip.com (Bob Ceculski)  Subject: Purveyor runs VMS Perl!= Message-ID: <d7791aa1.0202021153.418832b4@posting.google.com>   F found this on this site ... I took an old perl script we had from when+ we brought our site in house and it worked!n  L Here is an example of how to use PERL from a CGI script. It uses three filesL TEST_CGI_VMS.HTML, PPERL.COM, and PERL_TEST.CGI. They were all placed in theK htbin directory. It has been tested with using PERL 5.005.02's standard CGIeL package built with DEC C V5.6-003 on OpenVMS Alpha V7.1-1H1.  The main trick was to use the following line:        $ perl <WWW_IN: 'perl_script  L This trick was learned from Process Software Tech Support. They did however,I seem to indicate that since PERL is not part of Purveyor, that this trick  is not technically supported.2  P ********************************************************************************   TEST_CGI_VMS.HTML:   <HEAD> <TITLE> Perl CGI test </TITLE> </HEAD>K <BODY> <h1>Perl test </h1>w9 <A HREF="http://www.xxx/PPERL.COM?PERL_TEST.CGI">Test</A>e <P>u </BODY>  </HTML>e  P ********************************************************************************  
 PPERL.COM:   $! Setup for PERLd9 $ define  PERL_ROOT DISK$xxx:[PERL5_005_02.] /trans=conc t# $ define  PERL5LIB  PERL_ROOT:[lib] 1 $ define  PERLSHR   PERL_ROOT:[000000]perlshr.exel, $ perl    :==  "$PERL_ROOT:[000000]perl.exe" $!= $! Assume the PERL scrip is in the current (HTBIN) directory. C $ THIS_PATH = F$Element (0, "]", F$Environment ("PROCEDURE")) + "]"W $ perl_script = THIS_PATH + p1 $! $ on ERROR THEN GOTO Cleanup $ perl <WWW_IN: 'perl_script $!
 $ Cleanup: $   deass PERLSHRb $   deass PERL_ROOTt $   deass perl5lib $   delete /symbol /global perld $ exit ! PPERL.COM mains  P ********************************************************************************   PERL_TEST.CGI:   #!/usr/local/bin/perl -w use CGI qw(:standard); my $favorite = param("flavor");l0 print header, start_html("Ice Cream"), h1("Hi"); if ($favorite) {1    print p("Your favorite flavor is $favorite.");0 }  else {  print hr, start_form;@  print p("please select a flavor: ",textfield("flavor","mint"));  print end_form, hr; }6 print end_html();>   ------------------------------  # Date: Sat, 02 Feb 2002 21:36:37 GMTn1 From: Ed Wensell III <ewensell3@yahoo.commercial>oN Subject: Re: SGI solicits Alpha customers to trade in their "orphaned" systems0 Message-ID: <3C5C5B9B.EDFD9276@yahoo.commercial>   "Main, Kerry" wrote: > F > mmm... re: SGI loyalty to its users - amazing how marketeers like to > rewrite history eh ? > 0 > >>> SGI introduces the concept of loyalty. <<< > & > Reference - articles from the past : >    Let's add fuel to the fire...   2 http://www.theregister.co.uk/content/54/23708.html? "Exclusive Silicon Graphics Inc has transferred much of its 3D 6) graphics patents portfolio to Microsoft.". ...hA "SGI insisted at the time these are "non core" technologies, but 2E sources close to the Mountain View are emphatic that these represent yD the bulk of SGI's 3D intellectual property assets, a view confirmed ( by documents disclosed to The Register."  H So, if they gave up a bulk of their 3D, what's left? Just expensive PCs?   -- e Ed Wensell IIIE NetBSD/Alpha at home - Solaris/SPARC at work - OpenVMS in a past lifetA E-mail address is valid if you know the appropriate bits to drop.    wibble?m   ------------------------------   Date: 2 Feb 2002 13:52 CST' From: carl@gerg.tamu.edu (Carl Perkins) - Subject: Re: WTB:Freeware 4 or 5 CD for sale?c, Message-ID: <2FEB200213521323@gerg.tamu.edu>  / kuhrt@encompasserve.org (Marty Kuhrt) writes...nV }In article <3C4EDE15.1D5708C8@caltech.edu>, David Mathog <mathog@caltech.edu> writes: }> John Forkosh wrote: }>> F }>> I purchased Freeware 5 CD from them back on Sept 4, 2001, for $25.J }>> Call 1-800-DIGITAL (or 1-800-ATCOMPAQ, I can't recall which I dialed),, }>> and ask for QA-6KZAA-H8.  Worked for me. }>  : }> One of my bigger pet peeves.  Since no telephone button; }> is labeled with a Q one cannot dial ATCOMPAQ.  I wonder o( }> if they own all numbers ATCOMPA[0-9]? }>   } D }Would it matter?  Does the phone company pay any attention to what C }you type after the 7th digit after area code?  You'd get the same RB }effect if you typed ATCOMPA as if you typed ATCOMPAQWEHAVENOCLUE  }wouldn't you?   Not necessarily.  D The phone company may not care, but the internal phone system might.F It is not uncommon to have additional digits treated as the extension,D to dial a specific phone (like, say, our phone system: to get to theA phone on my desk rather than the receptionist you add the 3 digit. number that is my extension).e  B But this was all hashed out weeks ago. Somebody is way behind with the newsgroup (or InfoVAX).M   --- Carl   ------------------------------   End of INFO-VAX 2002.065 ************************