1 INFO-VAX	Sat, 10 Aug 2002	Volume 2002 : Issue 438       Contents:# Apache execution of a VMS .COM file C Re: Arithmetic Data Corruption - Compaq TCP/IP Services 5.3 and NFS C Re: Arithmetic Data Corruption - Compaq TCP/IP Services 5.3 and NFS C Re: Arithmetic Data Corruption - Compaq TCP/IP Services 5.3 and NFS C Re: Arithmetic Data Corruption - Compaq TCP/IP Services 5.3 and NFS " Re: Automating EDIT/FDL ->OPTIMIZE" Re: Automating EDIT/FDL ->OPTIMIZE Can't kill process Re: Decserver 900TM  Re: Decserver 900TM  FORTRAN Alpha kit requested  Re: FTP> get *.* does not work Re: FTP> get *.* does not work/ Re: FTPed files from Mac to VMS start with <LF> ; HP's pulling of Compaq call-center contract puzzles Broward . Re: Kingston memory modules in AlphaServer 800? Re: OT - Stock prices (was: RE: HP-Compaq Merger Went Smoothly) . Re: OT Mac OS X woes, but advice needed please. Re: OT Mac OS X woes, but advice needed please. Re: OT Mac OS X woes, but advice needed please0 Rebuilds Was: device full: wrong error messages?0 Re: Register an Hobbyist layered product license0 Re: Register an Hobbyist layered product license2 Re: Samsung says alpha lives ... Gartner loses ...* Re: Second KLEZ from ATTBI Today - TROLL ! Re: snapshot# Re: Some Alpha news on The Inquirer  Re: test, ignore RE: uaf  Re: uaf  Re: uaf 3 Re: Where to get ancient Software distribution CD's   F ----------------------------------------------------------------------   Date: 10 Aug 02 18:39:15 +0200) From: p_sture@elias.decus.ch (Paul Sture) , Subject: Apache execution of a VMS .COM file) Message-ID: <crx2nH7Fs9NQ@elias.decus.ch>    Hi good folks,  M I am trying to get Apache on VMS to execute a bit of DCL, and it doesn't want  to play.   Here's the html   0 <form method="POST" action=/cgi-bin/dclproc.com>  & However, the error log shows up with :  F  malformed header from script. Bad header=%DCL-W-IVVERB, unrecognized $ co: /apache$root/cgi-bin/DCLPROC.COM  H It does the same when I feed Netscape with localhost/cgi-bin/dclprof.com  E Yet if I give Netscape localhost/cgi-bin/test-cgi-vms.com, that works 
 perfectly.  J Is there a way of declaring to Apache that a .com file such be executed asN such? Or is the <form method="POST" bit wrong for Apache (it was stolen from a WASD bit of code)    TIA.      __
 Paul Sture Switzerland    ------------------------------    Date: 09 Aug 2002 03:04:49 +0800, From: Paul Repacholi <prep@prep.synonet.com>L Subject: Re: Arithmetic Data Corruption - Compaq TCP/IP Services 5.3 and NFS- Message-ID: <87hei5894e.fsf@prep.synonet.com>   : "John Gemignani, Jr." <Paul.Non.Harvey@non-hp.com> writes:  F > The newer compiler generated code to use F0 and F1 as temporaries toE > perform certain types of arithmetic in place instead of calling OTS @ > support routines to do the work.  Alpha does not have a nativeE > integer divide instruction.  V5.1 had been specifically checked for 0 > any floating instructions and there were none.  D This would be a result of the EV6 and later being able to move from D Int to FP reg without having to go via memory? And thus slashing the, cost of a Int -> FP, FPdiv, -> Int sequence?     --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 09 Aug 2002 03:09:38 +0800, From: Paul Repacholi <prep@prep.synonet.com>L Subject: Re: Arithmetic Data Corruption - Compaq TCP/IP Services 5.3 and NFS- Message-ID: <878z3h88wd.fsf@prep.synonet.com>   ( John Reagan <john.reagan@hp.com> writes:  D > I don't speak for the TCPIP product, but my guesses are that is itE > new for V5.3 since that is when they seemed to have upgraded to the D > newer C compiler that exposed their build flaw.  Older C compilers? > would choose to use the OTS$DIV routines instead of using the A > floating point alternative.  It is not in the VAX version since C > we're talking about C-written drivers and using explicit floating " > registers, neither exist on VAX.  B John, the old Fortran Compilers on the 11s would put at the end of@ the listing whether or not any FP instructions where generated.  Perhaps this could be put back?    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Sat, 10 Aug 2002 16:30:21 GMT & From: John Reagan <john.reagan@hp.com>L Subject: Re: Arithmetic Data Corruption - Compaq TCP/IP Services 5.3 and NFS% Message-ID: <3D553DD1.9080500@hp.com>    Paul Repacholi wrote: < > "John Gemignani, Jr." <Paul.Non.Harvey@non-hp.com> writes: >  > F >>The newer compiler generated code to use F0 and F1 as temporaries toE >>perform certain types of arithmetic in place instead of calling OTS @ >>support routines to do the work.  Alpha does not have a nativeE >>integer divide instruction.  V5.1 had been specifically checked for 0 >>any floating instructions and there were none. >  > F > This would be a result of the EV6 and later being able to move from F > Int to FP reg without having to go via memory? And thus slashing the. > cost of a Int -> FP, FPdiv, -> Int sequence? >  >   H Only partially I think...  The newer versions of GEM will even move the H integers to the floating registers via the memory stack for the default D setting of /ARCHITECTURE (the default of GENERIC a pre-EV6 target). I With /ARCH=EV6 or greater, you will get the ITOFT and FTOIT instructions   as you predict.    -- John Reagan ' Compaq Pascal/{A|I}MACRO Project Leader  Hewlett-Packard Company    ------------------------------  # Date: Sat, 10 Aug 2002 16:40:01 GMT & From: John Reagan <john.reagan@hp.com>L Subject: Re: Arithmetic Data Corruption - Compaq TCP/IP Services 5.3 and NFS% Message-ID: <3D554016.1060902@hp.com>    Paul Repacholi wrote:    > D > John, the old Fortran Compilers on the 11s would put at the end ofB > the listing whether or not any FP instructions where generated. ! > Perhaps this could be put back?  >   I Interesting suggestion.  However, except for these rare cases, it really  H doesn't matter.  It isn't like you needed to SYSGEN a new RSX kernel or E make sure you had your FP11 floating board installed in your machine.   E While the actual work wouldn't be difficult (a counter in GEM of the  G number of floating instructions that got emitted into the object file;  G work to insert the counter into the listing at some point; and work by  B the user to then scan the listing file looking for the number and I verifying that it is 0), I don't think it is worth our time.  Given that  H   a user would have to explicitly visit their build procedures today to G determine which modules needed a 'search the listing and make sure the  F count is 0', the easier solution is to determine the same modules and B add the /INSTRUCTION=NOFLOAT qualifier.  That will stop the "free B floating instructions for integer divide" and if you use explicit E floating types in your C code, you'll get a warning message from the  E compiler saying that it conflicts with the /INSTRU=NOFLOAT qualifier  D (just don't compile /NOWARNINGS...  some here think it should be an 
 error level).    --   John Reagan ' Compaq Pascal/{A|I}MACRO Project Leader  Hewlett-Packard Company    ------------------------------    Date: 10 Aug 2002 00:03:46 +0800, From: Paul Repacholi <prep@prep.synonet.com>+ Subject: Re: Automating EDIT/FDL ->OPTIMIZE - Message-ID: <87ptwskoil.fsf@prep.synonet.com>   3 "David J. Dachtera" <djesys.nospam@fsi.net> writes:   A > In that case, you may need to go through a one-time exercise to B > manually adjust the bucket sizes and do CONVERT/FDL. After that,/ > automated optimisations should run just fine.   C If it is 17,000 files, perhaps changing the cluter size is a better  idea?    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Sat, 10 Aug 2002 00:40:00 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> + Subject: Re: Automating EDIT/FDL ->OPTIMIZE & Message-ID: <3D5465B4.26603EA@fsi.net>   "David J. Dachtera" wrote: >  > "David J. Dachtera" wrote: > >  > > Graham Burley wrote: > > >  > > > Joe wrote: > > > [snip]M > > > > Part of what I need is the bucket size optimized to drive cluser size  > > > > during the convert.  > > > [snip] > > > N > > > Edit the ANAL/RMS/FDL output and change the CLUSTER_SIZE before you feedE > > > this into EDIT/FDL/NOINTERACTIVE and you'll get the bucket size  > > > optimized. > > I > > I wonder what happens when you try this (cut from my earlier post and  > > edited): > > ' > > $ anal/rms &fsp/fdl/out='fname'.fdl A > > $ search 'fname'.fdl bucket/match=nor/out='fname'_nobktsz.fdl I > > $ edit/fdl/nointeractive/analysis='fname'_nobktsz.fdl 'fname"_opt.fdl  > > ) > > I'll try that and see what happens...  > < > O.k. Here's the results (and the corrected syntax, BTW)... >  > DJAS01::DDACHTERA$ @fdl_test( > $ fsp := sys$common:[sysexe]sysuaf.dat > $ fname := sysuaf $ > $ anal/rms &fsp/fdl/out=SYSUAF.fdl > $ search SYSUAF.fdl bucket# >         BUCKET_SIZE             3 # >         BUCKET_SIZE             3 = > $ search SYSUAF.fdl bucket/match=nor/out=SYSUAF_nobktsz.fdl 8 > $ edit/fdl/nointeractive/analysis=SYSUAF_nobktsz.fdl -' >         SYSUAF.fdl/out=SYSUAF_opt.fdl   > $ search SYSUAF_opt.fdl bucket# >         BUCKET_SIZE             4 # >         BUCKET_SIZE             4 # >         BUCKET_SIZE             4  > $ exit   Here's the "source" listing...  " DJAS01::DDACHTERA$ ty fdl_test.com& $ fsp := sys$common:[sysexe]sysuaf.dat $ fname := sysuaf # $ anal/rms &fsp/fdl/out='fname'.fdl  $ search 'fname'.fdl bucket = $ search 'fname'.fdl bucket/match=nor/out='fname'_nobktsz.fdl 7 $ edit/fdl/nointeractive/analysis='fname'_nobktsz.fdl -   	'fname'.fdl/out='fname'_opt.fdl $ search 'fname'_opt.fdl bucket  $ exit     --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 10 Aug 2002 06:43:28 -0700 # From: "Tom Linden" <tom@kednos.com>  Subject: Can't kill process 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIGEILFIAA.tom@kednos.com>   ? Running 7.1 on VAX have a small program that had been reported  = to cause an access violation by corrupting the stack.  Didn't @ see that, but when trying to kill the process, it won't go away.  ; The state is indicated as RWMPB (Modified page writer busy) ) STOP/ID does nothing.  Here's the program    BLOW: PROC OPTIONS(MAIN);  	DO WHILE (A('')=''); END;   	A: PROC(C) RETURNS(CHAR(*));          DCL C CHAR(*);         RETURN(C);	 	  END A; 
 	END BLOW;     --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.381 / Virus Database: 214 - Release Date: 8/2/2002    ------------------------------  # Date: Fri, 09 Aug 2002 23:43:56 GMT . From: peter@langstoeger.at (Peter LANGSTOEGER) Subject: Re: Decserver 900TM5 Message-ID: <0tY49.139591$cU1.4042190@news.chello.at>   \ In article <1JW49.56$Xu3.15671@news2>, "news.alltel.net" <dbwoodw-nospam@alltel.net> writes:= >Anyone know where I can get the "WWENG2" load file for this?   , Normally from its owner: http://www.dnpg.com or by a fried ;-)    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atP A-1030 VIENNA  AUSTRIA              I'm looking for (a) Network _and_ VMS Job(s)   ------------------------------  # Date: Sat, 10 Aug 2002 00:18:39 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: Decserver 900TM' Message-ID: <3D5460B7.9B340E9F@fsi.net>    Peter LANGSTOEGER wrote: > ^ > In article <1JW49.56$Xu3.15671@news2>, "news.alltel.net" <dbwoodw-nospam@alltel.net> writes:? > >Anyone know where I can get the "WWENG2" load file for this?  > . > Normally from its owner: http://www.dnpg.com > or by a fried ;-)   @ "Fried"??!! But I stopped smoking that stuff AGES ago! I'm still	 fried??!!    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 09 Aug 2002 14:15:56 +0200 - From: Didier Morandi <Didier.Morandi@Free.fr> $ Subject: FORTRAN Alpha kit requested' Message-ID: <3D53B27B.C35E0170@Free.fr>   - (looks like my previous post did not show up)   N Where could I find an on-line distribution kit of the latest FORTRAN Alpha/VMS> kit? My Hobbyist PWS 600au CD-ROM drive is a non-VMS model :-(   Thanks,    D.   ------------------------------   Date: 10 Aug 02 11:29:12 +0200) From: p_sture@elias.decus.ch (Paul Sture) ' Subject: Re: FTP> get *.* does not work ) Message-ID: <1AaZ0Z+Po6$X@elias.decus.ch>   V In article <3D5382AD.843C72C@Free.fr>, Didier Morandi <Didier.Morandi@Free.fr> writes:   <snip>   > FTP> > FTP> get *.com' > 550 *.com: No such file or directory. - > %TCPIP-E-FTP_NOSUCHFILE, no such file *.com  > FTP> >  > Why?  1 For wildcard operations, try mget instead of get.    __
 Paul Sture Switzerland    ------------------------------  # Date: Sat, 10 Aug 2002 17:08:17 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network> ' Subject: Re: FTP> get *.* does not work ( Message-ID: <3D5543E3.10805@qsl.network>   Didier Morandi wrote:  > I run VMS 7.3 TCP/IP V5.1-15. / > I wish to copy files from my Mac to my Alpha. J > I started MPWS on the Mac and did connect to via FTP, providing my Mac's > username and password. > ; > I can do a SET DEF and a DIR and a GET one_file_only.COM. J > But if I do a FTP> get *.* or mget *.*or *.COM of whatever, I get this:  C Lower case commands to FTP use UNIX syntax, uppercase commands use   OpenVMS syntax.    IIRC: You can use:   GET *.* or mget *.*    but not:   get *.* or MGET *.*    -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------  % Date: Sat, 10 Aug 2002 11:08:59 -0400  From: MikeR <mike@rechtman.com> 8 Subject: Re: FTPed files from Mac to VMS start with <LF>6 Message-ID: <aj3a6q$18ejv8$1@ID-103225.news.dfncis.de>  # Did you set transfer mode to ASCII?   9 On Fri, 09 Aug 2002 04:55:31 -0400, Didier Morandi wrote:   J > Trying to transfer DCL command procedures from my iMac OS 9.2 to VMS 7.3B > gives files with lines starting with a <LF> (but the first one). >  > Someone has seen this before?  > 	 > Thanks,  >  > D.   ------------------------------    Date: 10 Aug 2002 00:03:32 -0700, From: webmaster@isitamerican.com (Webmaster)D Subject: HP's pulling of Compaq call-center contract puzzles Broward= Message-ID: <8e15fb52.0208092303.27806570@posting.google.com>   ; HP's pulling of Compaq call-center contract puzzles Broward C http://www.sun-sentinel.com/business/local/sfl-zanswer09aug09.story    By Doreen Hemlock  Business Writer  Posted August 9 2002       B News that computer giant Hewlett-Packard Co. is pulling a contractD from a Fort Lauderdale call center and shifting some of its business9 to lower-cost India prompted a mix of reactions Thursday.   D Economic development officials gave assurances that Florida's strong> call-center business continues to grow, while a Broward County0 commissioner suggested a boycott of HP products.  D Commissioner Sue Gunzburger said she'll ask the County Commission onE Tuesday to halt new purchases of HP printers for Broward, now that HP F is ending a contract with The Answer Group that could mean the loss of( up to 1,200 Broward jobs within 90 days.  D "If they want to take jobs offshore, they should sell their printers- offshore -- not to Broward County," she said.   C "I would hope that others would feel the same way I do and consider0 boycotting the company."  F Gunzburger said in her 10 years as commissioner, she knows of no other? case when a local firm has lost so many jobs to another nation.p  > South Florida economic development leaders said they also wereA surprised: They knew of no other big loss of call-center business F offshore, and indeed, have seen the call-center industry expand -- not/ shrink -- even amid the current economic slump.   A "We don't see this as a trend," said high-tech consultant MichaelAD Langley, former president of the Broward Alliance and now an adviser@ to the state's development group, Enterprise Florida. "We're notC immune from companies losing contracts. And this is a good alert toiF make sure we monitor closely. But we haven't seen this as a retreating# industry, but one that is growing."e  B Broward County had an estimated 20,000 call center jobs last year,C according to the latest research from The Broward Alliance. And neweC call-centers have been setting up since then, including some in theoE travel industry, such as an $11 million Delta Air Lines center slatedt6 to employ more than 600 in Miramar starting next year.  E "We've been a favored location for call-centers," said the Alliance'sAE acting president Joan Goodrich, adding she's just received an inquiryaD from a company in the northeast U.S. interested in a Broward center.  E Miami-Dade has about 15,000 people working at call centers, includingAA nearly 6,000 employed by Fort Lauderdale-based Precision ResponseuD Corp. The number is rising, said Frank Nero, president of The Beacon( Council, the county's development group.  F "I haven't seen any trend of this business going offshore," said Nero,D noting companies are attracted to South Florida's multi-lingual workB force and relatively low labor, rental and tax rates compared with other competing U.S. locales.=  C http://www.sun-sentinel.com/business/local/sfl-zanswer09aug09.storyt  = Doreen Hemlock can be reached at dhemlock@sun-sentinel.com orf
 305-810-5009.I, Copyright  2002, South Florida Sun-Sentinel   ------------------------------    Date: 09 Aug 2002 23:50:17 +0800, From: Paul Repacholi <prep@prep.synonet.com>7 Subject: Re: Kingston memory modules in AlphaServer 800s- Message-ID: <87u1m4kp52.fsf@prep.synonet.com>   ' Jeremy Begg <jeremy@vsm.com.au> writes:-  D > I've recently acquired a refurbished AlphaServer 800.  It has only< > 128MB memory so I'd like to put a whole lot more in beforeF > commissioning it.  My preferred reseller has quoted $AUD2400 for 1GB5 > DEC memory or $AUD1300 for the Kingston equivalent..  vB > He will happily supply either brand but his technical staff haveC > warned me that they had a lot of problems with Kingston memory innA > AS800 systems a few years back: systems would crash with memory   > errors due to timing problems.   Ask them if it was on uV-IIs.a  eD > This is the first time I've heard a bad word said against Kingston= > but I don't doubt the sincerity of the comments made by theb6 > reseller.  Can anyone corroborate the error reports?  > Even DEC boards had problems, untill it was sorted fully. Part= timing and part loading on some of the drivers or reflections  as I remember.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------   Date: 9 Aug 2002 20:19:06 GMTh7 From: sy18889@rabmbit.famrp.cosm (Bradford J. Hamilton)SH Subject: Re: OT - Stock prices (was: RE: HP-Compaq Merger Went Smoothly)! Message-ID: <8Vii4ZtYJYKv@rabbit>a   Folks,   Very punny!b   :-)u  N Of course, I was trying to make a point that the *value* of "stock prices" was an unknown variable.   :-)r  H When I made my original comparison, I should have phrased it as follows:  + "stock price" .nes. "company value", etc...a  c In article <9EeR1WxhzMDN@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:nd > In article <3D542A41.F79C5ED@aaa.com>, Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> writes:. >> Are stock prices realy *string* variables ? > 4 > Some companies are just string shareholders along. -- c Bradford J. Hamilton& braMdhamAilPtoSn@aMtAtPbi.cSom		(home)& sMy1A88P89S@rabMbit.fAmPr.coSm		(work)  ; "All opinions that I express are my own, not my employer's"  "Lose the MAPS"    ------------------------------   Date: 10 Aug 02 11:01:12 +0200) From: p_sture@elias.decus.ch (Paul Sture)u7 Subject: Re: OT Mac OS X woes, but advice needed please-) Message-ID: <$RZfYJJpVjoa@elias.decus.ch>0  \ In article <080820020117554879%elliott@yrl.co.uk>, Elliott Roper <elliott@yrl.co.uk> writes:  F > OS X is the future. Or you could just send your cute little iBook toI > uncle and aunty Rotter's school for orphan computers (e-mail address as>	 > sender)  >>    I It could well be the future. I'll grant you I am suffering from a lack offK documentation. Phrases like "as described in the Mac OS X manual which cameoK with your system" grate somewhat when mine didn't come with any such thing.c  O >> What are your best recommendations on OS X backup strategies? (presently ftpr' >> to my Alpha before I lose work, thene >> off to tape)nF > That's OK. Shove all your Mac files through dropstuff and PGP first.I > When you get 'em back, all the type and creator metadata will be bettereB > preserved. And nobody on your VMS box will be able to sniff 'em.   Thanks, I'll try that.   >>    >> Should I buy a VT520 instead?? > Yeah. Stay with teco. Who needs all that pointy-clicky stuff? > > Uncle rotter says "Send me the iBook, I'll send you a VT520"  A Catch me next time it crashes or loses data, I may be tempted :-)o   >> aO >> To be honest, if I am the only one having problems, then I'll put it down to-. >> hardware, and as it's still under warranty,A >> I can get an exchange. Gut feeling tells me otherwise however.: > I > Your's does not sound like a common problem. Mac OS X is very solid and3< > reliable compared to other toy computer operating systems. > D I am afraid that I am finding that iPhoto hangs if I try to anythingN other than simple import export stuff. On one occasion it wouldn't even let meH do a reboot. I have heard similar reports of iTunes and the DVD app too.  N I think a large part of my problem is the way this thing has been packaged. ItG has been aimed at the point and click crowd. No pipe character or curlyoJ brackets on the keyboard for instance. Yeah, _I_ can switch to US keyboardL layout to get those, but how many others are comfortable with that? But I'veK also permanently lost my French accented characters since I did that (SwissyC German keyboard here - both French and German accented characters).   F (Oh dear - a 20 year old struggle of US/UK folks not understanding theI requirements of keyboards designed for non-English languages is still not 
 resolved.)  I Also much of the help wants a permanent internet connection, which is nott exactly the point of a laptop.  t* > (And KLEZ can't touch it Terry.)....yet.  , Ohh. Doesn't he have an XP1000 to trade? -:) __
 Paul Sture Switzerlande   ------------------------------   Date: 10 Aug 02 11:19:40 +0200) From: p_sture@elias.decus.ch (Paul Sture)p7 Subject: Re: OT Mac OS X woes, but advice needed pleasea) Message-ID: <8UAFk5cVIsVR@elias.decus.ch>.  c In article <3D53C0E0.46056972@mindspring.com>, Atlant Schmidt <atlantnospam@mindspring.com> writes:e > Paul Sture wrote:a > i >> Totally off-topic I realise, but I know there are VMS pals out there who have experience of Mac stuff.p >>z >> My problem is that my 3 month old iBook, running on OS X regularly drops he odd file or two (or three ort four or maybee >> more)). It's there according to the index (Finder, whatever), but simply gives "cannot open file".i > - > You may send your iBook to me. Just as soonO/ > as I fix your problem, I'll send it back. :-)  > - > Meanwhile, I'd bet on "protections" myself.r- > There's certainly no substantial discussiona- > going on in the Mac world of any particular # > data lossage going on in MacOS/X.5 > J Thanks, I'll look into protections. Methinks there's a definite rat in theM system , which may date back to having to kill the initial installation via a J hardware reset because it refused to accept the fact that I didn't have itB hooked up to either phone line or internet connection at the time.  M Having been bitten in the past by M$ attempts to get me to sign up for MSN, aT3 ceratin amount of stubborness kicked in an that :-)e  > Sigh. Time to do the Windows thing and reinstall from scratch? __
 Paul Sture Switzerlandt   ------------------------------   Date: 10 Aug 02 18:18:31 +0200) From: p_sture@elias.decus.ch (Paul Sture)l7 Subject: Re: OT Mac OS X woes, but advice needed please ) Message-ID: <cQ8x6dXYovVs@elias.decus.ch>d  \ In article <080820020117554879%elliott@yrl.co.uk>, Elliott Roper <elliott@yrl.co.uk> writes:6 > In article <NTS8+u3dG8cL@elias.decus.ch>, Paul Sture! > <p_sture@elias.decus.ch> wrote:f > I > A little more detail may be required. What do you do to the file to geteG > "cannot open file"? ie what application are you opening it with? what  > file?d  L It just did it again. A .JPG file, not important, but it gave with "The fileM aussie_windaz.jpg could not be opened." Any other .JPG in the same view worksrM fine. I am not mucking around as several users (although I do that on  my VMSoK  box to test stuff out), so I don't really see how file ppermissions shouldf1 come into it. It appears it randomly loses stuff.S  O Not only that, but it just froze, and now it refuses point blank to speak to myk9 VMS machine: A message saying it simply couldn't find it.c  8 Oh, two reboots later, and it _does_ see my VMS machine.  J Do you now understand why I call it unreliable? I really cannot trust this system.    __
 Paul Sture Switzerland    ------------------------------    Date: 09 Aug 2002 23:41:17 +0800, From: Paul Repacholi <prep@prep.synonet.com>9 Subject: Rebuilds Was: device full: wrong error messages? 0 Message-ID: <873ctom44i.fsf_-_@prep.synonet.com>  ( jlsue <jlsuexxxz@screaminet.com> writes:  C > On 7 AUG 2002 20:58:44 GMT, karcher@thuria.waisman.wisc.edu (Carli > Karcher) wrote:_  kA > >In a previous article, jlsue <jlsuexxxz@screaminet.com> wrote:f  D > >->Just FYI, DFU has a command that will take much less time to doE > >the ->rebuild.  That means that your users are "locked out" (hung)w* > >for a ->much, much shorter time period.  @ > >Aren't you thinking ov ANALYZE/DISK .vs. DFU's VERIFY here? IC > >wasn't aware that SET VOLUME/REBUILD affected user access to the 
 > >volume.  @ > Well, I must admit it's been awhile since I've had to use this7 > (since I don't do day-to-day administration anymore).o  iA > On volumes that have lots of stale info, the rebuild could takekF > awhile.  Now, I'm pretty sure the system hangs I/O while it does the= > rebuild (which is the reason we used MOUNT/NOREBUILD in ourrD > startups).  If you've never had rebuilds take very long, the hangs > may have gone unnoticed.  @ The rebuild holds the volume allocation lock, so there can be noB extends or truncates while it is happening. Think it also holds up? creates as well, but that may be because it blocks the new filer as 0 blocks in length :)  tA > But I am willing to accept that my memory is incorrect on this, C > since I have no way to test it out to prove one way or the other.u  B It also piles up the IOs pretty high on the drive, so IOs that can proceed are stuck in the queue..   -- 1< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Sat, 10 Aug 2002 00:11:43 GMT ( From: Alder <PGDEHMKOKIMD@spammotel.com>9 Subject: Re: Register an Hobbyist layered product licenseh, Message-ID: <3D545A3D.1090305@spammotel.com>   Craig A. Berry wrote:D- > In article <3d5420d0$1@obsidian.gov.bc.ca>,E. >  "Alder" <MUNDDGNTDYTV@spammotel.com> wrote: >  > K >>Can someone shed some light on why the Hobbyist kit would include invalidIK >>license PAKs for these components: LSE, MMS, SCA, and PCA, but valid onestI >>for CMS and DTM?  Is there another version of LSE/MMS/etc around that I  >>should try instead?m >  > D > I've installed and run DECset on various hobbyist systems with no F > trouble.  Did you both register and load *all* the licenses you got B > from Montagar?  MMS has an excellent freeware replacement (MMK, D > www.madgoat.com), but I don't think there are alternatives to the 	 > others.:  C Well I thought I had.  After all, VAXSET was showing up when I ran  G INSTALL LIST.  When I got the tip to run SHOW LICENSES the problem was  I blindingly obvious.  The VAXSET license needed to be LOADed.  Geeeeeeezz.P  / Thanks, Craig.  I hope this group is amused....n   Alder.   ------------------------------  # Date: Sat, 10 Aug 2002 16:46:36 GMTm- From: "John E. Malmberg" <wb8tyw@qsl.network>:9 Subject: Re: Register an Hobbyist layered product license@* Message-ID: <3D553EBC.3020704@qsl.network>   Doc.Cypher wrote:  >  dK > There is the possibility that it may work. However, it will not have been5# > tested with the v7.2 VMS version.   I There is a companion document to the OpenVMS S.P.D.  called an S.S.A. or dC Software Support Addendum that lists what software has been tested l' against a new operation system release.    -John, wb8tyw@qsl.network Personal Opinion OnlyM   ------------------------------  # Date: Sat, 10 Aug 2002 00:05:54 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>M; Subject: Re: Samsung says alpha lives ... Gartner loses ... ' Message-ID: <3D545DBA.FC6A2627@fsi.net>    Bob Lail wrote:  > L > Not the mention that Microway says Samsung denies they are reviving Alpha. > To bad !!! > . >     http://www.theinquirer.net/?article=4871  @ Hhmmm... 'Guess if it sounds to good to be true, it probably is.   -- o David J. Dachterao dba DJE Systemst http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/o   ------------------------------   Date: 10 Aug 02 11:06:53 +0200) From: p_sture@elias.decus.ch (Paul Sture)-3 Subject: Re: Second KLEZ from ATTBI Today - TROLL !i) Message-ID: <k+ly7UA6WhSb@elias.decus.ch>d  \ In article <aiti5i.268.1@jo.dyndns.org>, Franz-Josef Fornefeld <jo.fornefeld@gmx.de> writes: > Paul Winalski wrote: > F >> Eh??  If your email client doesn't execute attachments, how is Klez- >> going to start running in the first place?  > ' > HTML preview with JavaScript enabled.  >   > You do not even have to click.> > The virus/worm does not need to be in the attachment anyway. > 6 > So throw away any mailreader that does HTML preview.  $ Can you please tell Microsoft that?  __
 Paul Sture Switzerland)   ------------------------------    Date: 09 Aug 2002 23:44:29 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: snapshotP- Message-ID: <87y9bgkpeq.fsf@prep.synonet.com>e  1 Phil Tregoning <Philip.Tregoning@esa.int> writes:l  + > "David Lee" <phongle@kornet.net> wrote in % > news:aiqva7$rn9$1@news1.kornet.net:2   F > > Is there a snapshot feature in VMS 7.1.2?  or some kind of utility. > > that allow you to take a snap at a screen?  F > XV will let you grap a portion of an X display and save it to a file3 > in the format of you choice. You can get it from:a  s= >   http://www.openvms.compaq.com/freeware/freeware40/xv310a/   C Speaking of XV, and jpegs libs stuff. I sometimes get XV or DISPLAY.F from Image Magic give an error about `undefined SOF 0x2C' or the like.C Linux with the SAME jpeg library does not do this. Anyone know why?H   -- e< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.y@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 09 Aug 2002 03:28:16 +0800, From: Paul Repacholi <prep@prep.synonet.com>, Subject: Re: Some Alpha news on The Inquirer- Message-ID: <874re5881b.fsf@prep.synonet.com>i  , "Bill Todd" <billtodd@metrocast.net> writes:  5 > "Jan-Erik Sderholm" <aaa@aaa.com> wrote in messaged# > news:3D5174C4.2D520E1F@aaa.com...e& > > "Samsung revisits Alpha processor", > > http://www.theinquirer.net/?article=4842   ? > Hoo, hah!  Hadn't seen that one.  Having 1.75 MB 12 ns. - 9.6dD > ns. cache (at 1 - 1.25 GHz) right on the chip like EV7 will surelyC > shouldn't hurt.  I've not been able to find information about theiC > latency of the current EV68 off-chip L2 cache, but the latency ofeB > the PA-RISC off-chip L2 is reportedly 40 ns. which suggests thatE > off-chip cache may cease to be attractive when EV7 comes along withbE > its 75 ns. latency to local main memory but might remain attractivee, > in some situations for EV69 (as L3 cache).  c8 > I first read the comment about 'dual CPU systems' as aB > dual-core-on-a-chip reference, but on second glance suspect it'sE > not: it really would be a bit early for anyone outside of Compaq tocF > have come up with that level of enhancement.  I'm actually impressedE > that even the on-die L2 appeared, though it could well already have.# > been in the works before 6/25/01.i  B It was anounced at Hot Chips or one of them last year, or the yearB before. Seems you can put Humptey back together! Samsung also haveB their own version of the EV8. I suspect they are looking at a veryA large neibour who are continuing with their `Red Flag' 64-bit CPUr project.  F > In any event, it's great to see that Samsung seems to have the ballsD > that cHumPaq lacked.  While one can't expect anything like the EV8@ > core out of them, or even the EV7 on-chip routers unless their> > agreement includes complete access to the EV7 technology, itE > wouldn't be impossible to imagine future products with at least dief< > shrinks and perhaps dual cores and/or on-chip memory glue.   See above. -   -- -< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.o@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Sat, 10 Aug 2002 16:48:42 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network>o Subject: Re: test, ignore * Message-ID: <3D553F4D.7040102@qsl.network>   Didier Morandi wrote:l	 > dummy 5t  0 Please use the test newsgroups for this, thanks.   -John  wb8tyw@qsl.network Personal Opinion Onlyi   ------------------------------  , Date: Sat, 10 Aug 2002 11:23:28 +0200 (CEST): From: "Gotfryd Smolik, VMS lists" <gotfryd@stanpol.com.pl> Subject: RE: uafJ Message-ID: <Pine.LNX.4.21.0208101116520.21990-100000@irys.stanpol.com.pl>  ( On Fri, 9 Aug 2002, A Bonaveidogo wrote:   >+D. >+ >+b) is the current config >+logicals define properly...   <  Will little disagre: as we say you have multiple definition; of SYSUAF, also two (EXEC and SUPER modes) in system table..=  That creates multitude of possiblites with name translation:n4 supose both SYSUAF definition are identical but some7 translation for name "$1$DKA1" exists somewhere only inW supervisor mode...=  Please - delete all unneeded logicals (the *only* needed is t4 system table executive mode one), after that restart4 security server (SET SERVER SECURITY_SERVER/RESTART)  and report if something hanges !    Regards - Gotfryd   -- aE =====================================================================tF $ ON F$ERROR("LANGUAGE","ENGLISH","IN_MESSAGE").GT.F$ERROR("NORMAL") - 		THEN EXCUSE/OBJECT=MEw. $!                        GS@stanpol.zabrze.plE =====================================================================a   ------------------------------  , Date: Sat, 10 Aug 2002 11:16:16 +0200 (CEST): From: "Gotfryd Smolik, VMS lists" <gotfryd@stanpol.com.pl> Subject: Re: uafJ Message-ID: <Pine.LNX.4.21.0208101056580.21990-100000@irys.stanpol.com.pl>  / On Thu, 8 Aug 2002, Kenneth H. Fairfield wrote:@   >+A Bonaveidogo wrote: >+ >+> Ken. >+>sK >+> Both nodes use their own systems disk... dkb1 can be accessed from both / >+> node... mounted with the same volume label.t [...]mF >+    Whether this is the root problme or not, it's important not onlyE >+to DEFINE/SYSTEM/EXEC SYSUAF (and RIGHTSLIST, etc.) in SYS$MANAGER:-C >+SYLOGICALS.COM, but to _mount_ $1$DKB1: in SYLOGICALS as well, so@D >+that any process starting later in the system startup sequence has >+access to those files.    Agree.J   [...]aE >+    You say this problem just came up recently.  Did it occur afteroC >+one or the other system rebooted?  If not, I think we're grabbingc< >+at straws and need to look elsewhere for the root cause...    Agree with the question :)e [...]aA >+P.S. The supervisor mode definition of SYSUAF in your Job tablelA >+     will be ignored.  If you want ot override the System tablelA >+     definition, the logical still needs to be defined /EXEC int9 >+     whatever table you define it in for it to be used.g    Disagree !r  ?  The problem has something with Davids question: both DIRECTORYt7 and AUTHORIZE are "normal" (means: user mode) programm;a: that means: but supervisor and non-system-table definition< are in use. But executive/kernel routines requires EXECUTIVE7 definition, and AFAIR SETP0 honours system table *only*n; (what has some logic: when user logs in, then loginout mustt8 access SYSUAF & co *without* (before !) the group number; is known and before job table even exists - then make sensel% to use the same way for SET PASSWORD.s7  Summary: without big affort anyone can force AUTHORIZEs= to access differrent SYSUAF than LOGINOUT (and SET PASSWORD).t6  Really - best auditing info from SET PASSWORD can say, what is wrong with multiple SYSUAF logicals.  2  My second suspiction is with logical name for the1 device name: because we do not know, that $1$DKB1c1 logical name doesnt't exists, and the translationb. for SYSUAF is *not* prevented with /TRANS=TERM1 *nor* with undersore prefix - it is possible that + someone defines on one node something like:   ) $ DEF/SYS/EXEC $1$dka1 $1$dka2:/tran=concn  2  Have no time to check and exclude the possiblity:. *probably* auditing saves the real device name0 (and we seen it identical for AUTHORIZE access).    Regards - Gotfryd   -- aE =====================================================================sF $ ON F$ERROR("LANGUAGE","ENGLISH","IN_MESSAGE").GT.F$ERROR("NORMAL") - 		THEN EXCUSE/OBJECT=MEm. $!                        GS@stanpol.zabrze.plE =====================================================================w   ------------------------------    Date: 09 Aug 2002 23:18:28 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: uaf- Message-ID: <877kj0m56j.fsf@prep.synonet.com>n  ) Howard S Shubs <howard@shubs.net> writes:e  I > In article <BFBEDDF2CFEDD411917400508BF3A6FFCBD6F1@exchsvr.FSC.COM.FJ>,h* >  A Bonaveidogo <Asena@fsc.com.fj> wrote:   % > > both node use the same rightslist.   < > And you're absolutely and totally certain, having actuallyD > inspected, that both nodes are definitely using the same UAF file?+ > The file names wouldn't happen to both bel= > SYS$SPECIFIC:[SYSEXE]SYSUAF.DAT;1 or some such, would they?s  F I am suss of this as well. If the disk with sysuaf is not mounted whenF the security server is started, what happens? Or if the SYSUAF file isD not accessable for some reason. Does it wait for the file to appear?# Or does it fall back to sys$system?   * Do a $SHOW DEV $1$DKB0:/FILE/OUT=FOO.lis  # $SHOW DEV $1$DKA0:/FILE/OUT=FOO.lisnK $SHOW DEV $1$DKC0:/FILE/OUT=FOO.lis (or what ever your system disks are...)>  ) then $SEARCH sysuaf foo.lis on each node.i   -- h< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.b@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Sat, 10 Aug 2002 00:15:55 GMTh1 From: "David J. Dachtera" <djesys.nospam@fsi.net>m< Subject: Re: Where to get ancient Software distribution CD's' Message-ID: <3D54600F.76BE5480@fsi.net>c   Dann Corbit wrote: > F > I want to get a set of old ConDIST CD's (now called Software ProductL > Library).  I tried DEC/COMPAQ/HP/"Flavor-of-the-day" and also Ebay, but no > soap.  > K > Where should I look to find a set of (for instance) "OpenVMS VAX Softwareo > Product Library: March 1996" > K > I have the opposite problem of what is usual -- we have the license filesA, > for the needed products, but not the CD's! >  > Any ideas?  H If you can give some clues as to what you need, I could try to borrow anH old CONDIST and collect what you need on to a single a CD. The output of LICENSE LIST may suffice.t  A E-mail me privately if you'd prefer to take this off-line. How tol' de-mung the reply-to should be obvious.f   --   David J. Dachteral dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/?   ------------------------------   End of INFO-VAX 2002.438 ************************