1 INFO-VAX	Wed, 07 Jun 2000	Volume 2000 : Issue 316       Contents:' Re: Actual vs Theoretical Network Speed ' Re: Actual vs Theoretical Network Speed ' Re: Actual vs Theoretical Network Speed ' Re: Actual vs Theoretical Network Speed B Re: AlphaStation 200 firmware update problem - The Next GenerationB Re: AlphaStation 200 firmware update problem - The Next Generation Re: Amazon jumps ship 7 Availability Manager V1.4 and DECamds V7.2-1B available / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures / Re: C bashing (was Re: VMS File Caching Futures . Re: Charon VAX on PC emulator - Asynch DECNET? Comments on SW RA3000   Re: Console Monitoring Software.  RE: Console Monitoring Software.C Re: E-Commerce - Ariba, BroadVision & Commerce One Positions WANTED / FW: C bashing (was Re: VMS File Caching Futures 3 Re: FW: C bashing (was Re: VMS File Caching Futures 3 Re: FW: C bashing (was Re: VMS File Caching Futures 3 Re: FW: C bashing (was Re: VMS File Caching Futures  HELP! Motif Weirdness   Hobbyist vms 7.2 alpha questions! Re: I'm Trying to Model Something  Re: ICC programming examples ? Re: ICC programming examples ?; RE: is anyone able to get a DE500-BA working under 6.2 1h3? ! Re: Java graphing applets for VMS ! RE: Java graphing applets for VMS ! Re: Java graphing applets for VMS  Re: MMOV 2.2 and VMS 7.1 Re: motif debug problem  Re: motif debug problem  Re: MOZILLA M16 ?  Re: MOZILLA M16 ?  Re: MOZILLA M16 ?  Re: MOZILLA M16 ?  Re: MOZILLA M16 ?  Re: MOZILLA M16 ?  Re: MOZILLA M16 ?  Re: MOZILLA M16 ?  OpenVMS AND THE INTERNETA Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire) A Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire) A Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire) A Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire) A Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire) H Re: OpenVMS, SMTP, and email spam (was Re: I'm Trying to Model Something! Re: Porting VAX Macro code to AXP 3 Re: POS/credit card verification with OpenVMS Alpha 3 Re: POS/credit card verification with OpenVMS Alpha ' SMTP_RECV_MAIL from TCPIP$smtp_receiver  The DECUS Archives Re: vaxstation LED meanings?8 Re: What is the simplest way to network two VMS systems? Where is javadoc for VMS?  Re: Where is javadoc for VMS? P Windows NT hosted Uiltity that can read RMS index fles or VMS Backup formatted tP Re: Windows NT hosted Uiltity that can read RMS index fles or VMS Backup formattP Re: Windows NT hosted Uiltity that can read RMS index fles or VMS Backup formatt X25 and psi$k_reset   F ----------------------------------------------------------------------   Date: 6 Jun 2000 22:45:42 +0200 * From: eplan@kapsch.net (Peter LANGSTOEGER)0 Subject: Re: Actual vs Theoretical Network Speed* Message-ID: <393d62f6$1@news.kapsch.co.at>  V In article <393844E6.3F5F85F3@mmaz.com>, "Barry Treahy, Jr." <treahy@mmaz.com> writes:K >Actually I think I had a error in mine too, is not a true megabyte 2^21 or  >1048576 bytes?    And the next bug: 2^20 ;-)   --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888 < FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  # Date: Tue, 06 Jun 2000 22:41:01 GMT 3 From: "Gord Coulman" <nospam_gcoulman@ccinet.ab.ca> 0 Subject: Re: Actual vs Theoretical Network Speed8 Message-ID: <16f%4.409$vT6.141326@news1.telusplanet.net>  C Can anyone recommend a good utility to measure the actual bandwidth I available between two machines?  A Win9x utility is what I need (only one  OpenVMS machine).    Gord.    ------------------------------   Date: 7 Jun 2000 01:13:03 +0200 * From: eplan@kapsch.net (Peter LANGSTOEGER)0 Subject: Re: Actual vs Theoretical Network Speed( Message-ID: <393d857f@news.kapsch.co.at>  k In article <8hk06a$rvi$1@mailint03.im.hou.compaq.com>, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes: o >In article <16f%4.409$vT6.141326@news1.telusplanet.net>, "Gord Coulman" <nospam_gcoulman@ccinet.ab.ca> writes: E >:Can anyone recommend a good utility to measure the actual bandwidth K >:available between two machines?  A Win9x utility is what I need (only one  >:OpenVMS machine).  > ) >  Is "actual bandwidth" an oxymoron? :-)  > 5 >  Seriously, dts/dtr is one such tool, given DECnet.  > & >  The ping tool is another, given IP.  J Only if protocols are treated identical. And in more recent networks, ICMP> is not equally prioritized than say SAP over TCP or Voice, ...  F >  You could simply try whatever task it is that you need to perform, D >  as that will provide you with the effective end-to-end bandwidth.   YMMV   --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888 < FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------   Date: 6 Jun 2000 23:06:50 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)0 Subject: Re: Actual vs Theoretical Network Speed6 Message-ID: <8hk06a$rvi$1@mailint03.im.hou.compaq.com>  n In article <16f%4.409$vT6.141326@news1.telusplanet.net>, "Gord Coulman" <nospam_gcoulman@ccinet.ab.ca> writes:D :Can anyone recommend a good utility to measure the actual bandwidthJ :available between two machines?  A Win9x utility is what I need (only one :OpenVMS machine).  (   Is "actual bandwidth" an oxymoron? :-)  4   Seriously, dts/dtr is one such tool, given DECnet.  %   The ping tool is another, given IP.   E   You could simply try whatever task it is that you need to perform,  C   as that will provide you with the effective end-to-end bandwidth.   N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Tue, 06 Jun 2000 14:22:23 -0700 ! From: Shane.F.Smith@Healthnet.com K Subject: Re: AlphaStation 200 firmware update problem - The Next Generation C Message-ID: <OF429F99F2.B9242B17-ON882568F6.00756812@HEALTHNET.COM>   H Ok, here's the story so far. I was trying to load SRM on my AlphaStationG 200, but the screen kept blanking as the update utility started, and it D basically became a boat anchor until reboot. Terry Kennedy suggestedI changing the dip switches on my ZLXp-E2 display card to get it to display K properly. That worked, but all I got was a black screen with a block cursor K bottom left, and it was still a boat anchor. I tried hooking up a PC to the I serial port as per his other suggestion (And Dave Froble's), using Hoff's F pointers I got a standard PC null modem 9-pin cable, worked first timeJ @9600, and it turned out the utility really was communicating on port 1 asF the serial console. (Cue loud jubilant cheer.) Thanks very much Terry,1 Hoff, Wayne and Dave. Basic mission accomplished.   K So I updated the firmware using "Update SRM", and everything reported A OK. I I typed exit, and rebooted the machine. Here's where things started to go K wrong again. The output's below, it just keeps repeating with a short pause I between. Can anyone enlighten me as to where to go from here? The machine K has 24meg of RAM, surely that should be enough to get past the SRM startup? : Damn, but I wish I knew more about the hardware right now.   Shane   ! ff.fe.fd.fc.fb.fa.f9.f8.f7.f6.f5. P ef.df.ee.ed.ec.f4.eb.ea.e9.e8.e7.e6.insufficient dynamic memory for a request of
  236 bytes    PID       bytes  name -------- ---------- ---- 00000000      29248 ???? 00000001      28640 idle 00000002        800 dead_eater 00000003        800 poll 00000004        800 timer  00000005     130944 powerup  00000006       2880 tt_control 00000007        800 shell_0  00000008       1472 pka0_poll   , exception context saved starting at 001295E0   GPRs: -   0: 00000000 00045200  16: 00000000 0000001E -   1: 00000000 00111740  17: 00000000 00006000 -   2: 00000000 000E49D0  18: 00000000 EFEFEFC0 -   3: 00000000 000000EC  19: 00000000 000E4884 -   4: 00000000 00000003  20: 00000000 00000025 -   5: 00000000 00038218  21: 00000000 00000009 -   6: 00000000 00000020  22: 00000000 000D0EF8 -   7: 00000000 00000000  23: 00000000 00000000 -   8: 00000000 00000001  24: 00000000 00000001 -   9: 00000000 00000000  25: 00000000 00000001 -  10: 00000000 0012BF60  26: 00000000 0011083C -  11: 00000000 00000000  27: 00000000 00111750 -  12: 00000000 00000080  28: 00000000 5F0000FF -  13: 00000000 00000000  29: 00000000 00129720 -  14: 00000000 00000000  30: 00000000 00129720   15: 00000000 00000000   dump of active call frames:    PC  =  0011083C  PD  =  000E49D0  FP  =  00129720  SP  =  001295E0   B R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R29 saved starting at 00129728   R2  =  000ECCF0  R3  =  00127BE0  R4  =  00000000  R5  =  00129858  R6  =  00000000  R7  =  00126C80  R8  =  000D2FA0  R9  =  000D2F70  R10 =  00000024  R11 =  00000000  R12 =  00000000  R29 =  00129790    PC  =  0006F2BC  PD  =  000ECCF0  FP  =  00129790  SP  =  00129650   B R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R29 saved starting at 001297C0   R2  =  000ECD80  R3  =  00127BE0  R4  =  00129EE0  R5  =  00000000  R6  =  00000002  R7  =  00000000  R8  =  00000000  R9  =  00000000  R10 =  00000000  R11 =  00000000  R12 =  00000000  R29 =  00129840    PC  =  0006E6CC  PD  =  000ECD80  FP  =  00129840  SP  =  001296F0   * R2 R3 R4 R5 R29 saved starting at 00129958   R2  =  000EC0D0  R3  =  00126C80  R4  =  00000001  R5  =  00000002  R29 =  00129990    PC  =  0006E5BC  PD  =  000ECDA8  FP  =  00129990  SP  =  00129840    R29 saved starting at 00129998   R29 =  001299F0    PC  =  0006ABA0  PD  =  000EC0D0  FP  =  001299F0  SP  =  00129860   - R2 R3 R4 R5 R6 R29 saved starting at 00129A18    R2  =  000EB9B0  R3  =  001288A0  R4  =  00128A70  R5  =  00000000  R6  =  00000000  R29 =  00129A50    PC  =  00068300  PD  =  000EB9B0  FP  =  00129A50  SP  =  001298C0   ' R2 R3 R4 R29 saved starting at 00129A58    R2  =  00000000  R3  =  00000000  R4  =  00000000  R29 =  00000000    ------------------------------   Date: 6 Jun 2000 22:04:51 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)K Subject: Re: AlphaStation 200 firmware update problem - The Next Generation 6 Message-ID: <8hjsi3$qik$1@mailint03.im.hou.compaq.com>  g In article <OF429F99F2.B9242B17-ON882568F6.00756812@HEALTHNET.COM>, Shane.F.Smith@Healthnet.com writes: F :...The machine has 24meg of RAM, surely that should be enough to get  :past the SRM startup?... " :ff.fe.fd.fc.fb.fa.f9.f8.f7.f6.f5.% :ef.df.ee.ed.ec.f4.eb.ea.e9.e8.e7.e6. 7 :insufficient dynamic memory for a request of 236 bytes   F   Get more memory.  I'd tend to assume that 24 MB might not be enough,A   as 24 MB is well below the supported minimum for OpenVMS Alpha.   H   I'd also be looking for something very odd in how the memory SIMMs areI   configured, as the value 24 MB is itself weird -- the AlphaStation 200  I   has three banks of two SIMMS each, and I can think of no way to get to  H   24 MB using either the typical eight or sixteen megabyte SIMMs -- you G   could get to twenty-four with four megabyte SIMMs of course, but I'm  H   not aware of those being testing and configured with this box.  There D   are other, larger, SIMMs available, of course.  Bank 0 is the two F   adjacent SIMM slots nearest the middle of the motherboard, then the H   next two SIMM slots comprise Bank 1, then a slight separation and the G   two adjacent SIMM slots nearest the edge of the motherboard comprise  C   Bank 2.  Each slot in a bank must have the same memory capacity,  G   72 pin, parity, 70ns, SIMM.  Each bank can have a different capacity.   N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 6 Jun 2000 22:01:08 +0200 * From: eplan@kapsch.net (Peter LANGSTOEGER) Subject: Re: Amazon jumps ship* Message-ID: <393d5884$1@news.kapsch.co.at>  n In article <1UsKQmfVzA3G@tachxxsoftxxconsult>, wayne@tachysoft.xxx.044962.killspam.0138 (Wayne Sewell) writes:O >In article <3935c8d3@news.toast.net>, "Jeff Killeen" <Jeff@Killeen.cc> writes: O >> cbs.marketwatch.com/archive/20000531/news/current/amzn.htx?source=htx/http2_  >> mw  > N >Does this imply that amazon is currently running on vms?  I didn't know that.  J AFAIK, it ran U**X on Alphas and SPARCs (but I already proofed to be wrongK sometimes). So it doesn't really fit into COMP.OS.VMS except that we think, K everyone SHOULD run their apps on OpenVMS to gain the maximum advantage ;-)    --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888 < FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  $ Date: Tue, 6 Jun 2000 17:08:35 -0400& From: "AvailMan" <AvailMan@compaq.com>@ Subject: Availability Manager V1.4 and DECamds V7.2-1B available6 Message-ID: <8hjpkk$pe9$1@mailint03.im.hou.compaq.com>  :  Compaq Availability Manager Version 1.4 is now available,C with kits for OpenVMS VAX and Alpha, Windows NT/Windows 2000 Intel, F and Windows NT Alpha. The data analyzer client runs on these platformsC and monitors large numbers of OpenVMS nodes simultaneously (VAX and F Alpha) that have the data collector installed. Windows 2000 support is? new for this release. The details are in the documentation, and 4 everything you need to get started is in these kits.:  If you've been looking for a *FREE* tool (only an OpenVMS< license is required) to help manage your overall data centerD availability, Compaq's Availability Manager for OpenVMS provides theE kind of real-time monitoring and diagnostic capabilities you may find  very useful.>  Availability Manager evolved from the DECamds product that isD Motif-based and VMS-only. The new, easy to use Java GUI runs on bothE Windows and OpenVMS (Alpha only). Although functionally equivalent to G the latest release of DECamds, future enhancements are planned only for G the Availability Manager product. VAX users of the data analyzer client  will continue to need DECamds.B  Availability Manager Version 1.4 is being shipped coincident withH a release of DECamds, version 7.2-1B. Both fix a number of problems with9 earlier releases. The kits are available on our web site: 9  http://www.openvms.compaq.com/openvms/products/availman/ C Check it out and let us know what you think. Contact information is  on the web page.   ------------------------------  % Date: Tue, 06 Jun 2000 15:31:47 -0400 " From: Dan Sugalski <dan@sidhe.org>8 Subject: Re: C bashing (was Re: VMS File Caching Futures: Message-ID: <4.3.2.7.0.20000606130943.01d70a00@24.8.96.48>  - At 10:39 AM 6/6/00 -0500, Wayne Sewell wrote: I >In article <4.3.2.7.0.20000605150202.01c6da80@24.8.96.48>, Dan Sugalski   ><dan@sidhe.org> writes:3 > > At 11:59 AM 6/5/00 -0400, David A Froble wrote: N > >>Exactly!  This is what makes C inappropriate for applications development. > >a8 > > C was never designed as an app development language, >u> >Actually, it was never designed at all, as far as I can tell.  J I'd disagree. You (and I) may not agree with the design decisions, but it  wasn't thrown together.f   > >and it's a prettyL > > pathetic one. It lacks the features you'd usually want in an applicationK > > development language, and its standard library's badly suited for it as I > > well. C's a low-level systems language, meant for writing compilers, 0 > deviceK > > drivers, OS kernel bits, and suchlike things, and it's not bad at that.  >0 >7I >I would amend that to say it's meant for writing such stuff *in a hackeraJ >fashion*.  (I am of course using the term "hacker" in the original sense;L >nothing to do with malicious intent.)   While C does have low level access H >for stuff, this is not to say that strongly-typed languages do not.  I J >have written compilers in Pascal and also kernel mode code.  You have to L >be careful with what you do and make sure that you don't reference any RTL J >routines, but this would apply to C as well.  I would say that Bliss can G >do *anything* that C can, and probably more, but will tend to prevent 0' >stupid typos from crashing the system.0  J Pascal (at least a useful implementation of itO) is newer than C. C is an I old language, relatively speaking, and reflects the systems available at 0J the time. It was meant to be a simple language that you could write tight O code it, and one whose compiler was trivial (relatively speaking) to implement.   K >I would say that using a real language is *more* important for kernel mode0M >stuff than for applications, because the cost of failure is so much greater, A >i.e. crashing the entire system instead of just the application.0  H Now, sure. But now we've got cycles and megabytes up the wazoo. Not the I case when C was written, certainly not on the systems that K&R had handy R when they wrote it.2  H >Speaking of compilers, they are no different from applications from theK >viewpoint of the system and the machine code.  They execute in user mode, 0M >read an input file, and create an output file.  You can write compilers in a + >strongly typed language as easily as in C.C  G Sure, as far as the system is concerned it's all the same, and C's not aI particularly great as a compiler language. (Oddly enough, except for the 0/ speed issue I'd rather write a compiler in perlD  N >C was intended for people who don't want the compiler to "get in the way" andK >prefer that it let them do anything, whether correct code is generated or  K >not. Strangely, they prefer to defer their errors to run time, where they  H >are much harder to diagnose.  Sure the compile is faster.  Good thing, G >because you have to compile so many more times to correct errors that 0H >would have been trapped in the first compile by Pascal or the Countess 
 >(see below).   J Now, yes. At the time C was created, no, since Pascal didn't exist. C's a L step up from assembly language, and a half-step down from most of the ALGOL J derivatives. For what it did, it beat ALGOL, COBOL, or Fortran. (And B, I " assume, though I've never used it)  G >Note that we are talking about differences in *compilation* speed, not-) >*execution* speed of the generated code..  G *Now*, yes. When C was created there *were* no optimizing compilers to  L speak of. They didn't come along for quite some time, like a decade or more.   >Depending on the compiler, a.M >strongly-typed language can generate code as good as or better than a sloppy K >one such as C.  Most of it depends on the back end.  Admittedly, it may be K >possible to generate slightly tighter code with a simpler (more primitive)oM >language such as C, but an industrial strength back end such as GEM makes upv >for this to a large extent.  < Unoptimized C's tighter, but optimized is likely less tight.  2 >Both Compaq C and Compaq Pascal use GEM on alpha.H >I would be surprised if there is that much difference in the generated $ >machine code for a given algorithm.  K I wouldn't. I'd expect the Pascal code to be faster. Courtesy of pointers, oI optimising C's a pain. (The Fortran compiler folks have commented on the  + ease of optimizing Fortran vs C before too)e   > >ItoN > > leans a bit too heavily on the speed side of the speed/safety tradeoff forG > > my tastes, but it was developed on hardware that is, at this point,o > > terribly slow. >2G >That's an excuse for the original sad state of C, but not for C today.a  H But the C of today is based on the C from the early '70s. It has to be, H otherwise it's a different language. (Probably a better one, but that's  neither here nor there)n   >The C4 >standards committee had a chance to fix this stuff.  I No, they didn't, unfortunately. Adding in some of the fixes (like a real eH string type) would've busted backwards compatibility. Politics played a A role too, of course, but it always does when people are involved.,   >Admittedly, they did6N >improve the language somewhat, since ANSI C will at least do *some* checking,F >unlike the K & R shit,  but C still lets a lot of stuff slip through.  I Yep, it does, and that's on purpose. FWIW, Dec C does check for a lot of pK stuff (like array bounds) if you ask it to, though there's a limit to what  A it can do when pointer casts are an intgral part of the language.o  J >Admittedly, there *is* a run time performance hit for stuff such as arrayK >bounds checking.  This would fall into the speed/safety tradeoff mentioned J >above.   Seems like 95% percent of the C run time errors I have found areN >related to exceeding a local stack-based string variable and wildwriting intoL >the stack.  At least in Pascal you have the option of checking array boundsN >during development, then you can turn it off when the program is released for >production.  J A /CHECK=BOUNDS on Dec C 6.0 and up will do runtime array bounds checking  at the expense of some speed.-  C > > Don't knock C for being a bad app language, since it isn't one.< >iF >I consider it a bad language for any purpose.  But I'm stuck with it.L >Unfortunately, it's been so long since I've used Pascal that it would take % >time to come back up to speed on it.   ' The original Pascal sucked too, though.l  K I'll admit that at this point I've found very few computer languages worth lJ much. I'm fond of perl and Forth, and Python looks reasonably nice though L at this point if a language makes me cater to the machine I don't much care J for it. I've got CPU power and memory to burn, and the languages can damn + well cater to *me*, thank you very much. :)    > >Knock theL > > twits that insist on using it as an app language--that's where the blame > > rightly lies.o >o >Agreed. >eK > > (I'm still not particularly fond of the language, mind, but its biggestiH > > problem is its mis-application, not in any fundamental design flaws) >eL >Disagreed.   It was fine for the hacker days when people were just playing > >with computers, but not for production systems in *any* role.  K Unix systems never were production machines, really. More CS academic toys t= that got pumped up with steroids and put into production use.   J >Admittedly, eunuchs systems don't have anything to replace it for kernel  >code, but vms does.  J Sure. Now all we need are programmers who can use a language other than C  with any facility. :(o  K I'd also like to see something developed other than BLISS. I'm sure we can a do better these days.    					Dan  I --------------------------------------"it's like this"-------------------w2 Dan Sugalski                          even samurai? dan@sidhe.org                         have teddy bears and even0;                                       teddy bears get drunk:   ------------------------------  + Date: Tue, 06 Jun 2000 21:26:09 +0000 (   )13 From: Christopher Smith <chriss@Mufasa.pubserv.com>08 Subject: Re: C bashing (was Re: VMS File Caching FuturesI Message-ID: <Pine.LNX.4.05.10006062120230.2476-100000@Mufasa.pubserv.com>1  ' On Tue, 6 Jun 2000, Dan Sugalski wrote:0  ) > The original Pascal sucked too, though.e  - The bytecode was an interesting idea, though.l  L > >Admittedly, eunuchs systems don't have anything to replace it for kernel  > >code, but vms does.  L > Sure. Now all we need are programmers who can use a language other than C  > with any facility. :(w  H I'd be perfectly happy to develop a large chunk of my code in pascal (orC modula-N, where N is some posative integer...), but unfortunately I:I haven't been able to talk my employer into springing for a compiler -- if H I can even find a decent compiler to run on this billybox trash that I'm writing apps for.. ;)e  M > I'd also like to see something developed other than BLISS. I'm sure we can F > do better these days.   H I'd like to see some of the better things that we've got get more use...I once everyone's using something sane, in a sane manner, then we can worry # about making new, better languages.u  H It's a real shortcoming of most "programmers" that they have no idea howE to pick the best language for a given task, and a shortcoming of mostuA users that they can't pick the right computer for a given task...h   That's life.   Regards,   Chris   O ===============================================================================l@ "My two cents"			(http://rootworks.com/twocentsworth.cgi?128562)= Christopher Smith(chriss@pubserv.com)			Prgramer^W ProgrammerW Prime Synergy of Champaign, IL.h% -------------------------------------:I "Where a calculator on the ENIAC is equipped with 18,000 vacuum tubes anduH weighs 30 tons, computers in the future may have only 1,000 vacuum tubes; and weigh only 1.5 tons." -- Popular Mechanics, March 1949  O -------------------------------------------------------------------------------    ------------------------------  % Date: Tue, 06 Jun 2000 17:25:40 -0400h- From: JF Mezei <jfmezei.spamnot@videotron.ca> 8 Subject: Re: C bashing (was Re: VMS File Caching Futures, Message-ID: <393D6C53.EDF84A37@videotron.ca>   Wayne Sewell wrote: ? > Actually, it was never designed at all, as far as I can tell.L  M C perhaps not. But ANSI-C yes. They put all sorts of pedantic restrictions on L it (such as the inability to move an unsigned char to a signed char and viceH versa without a complaint from the compiler (or the use of the /unsigned compiler switch).e  O > C was intended for people who don't want the compiler to "get in the way" andnP > prefer that it let them do anything, whether correct code is generated or not.  K But ANSI-C changed that by being much more pedantic at compile time. If youIL want to do your own thing, you really have to try hard by type casting stuff' and using memcpy instead of the = sign.    ------------------------------  " Date: Wed, 7 Jun 2000 00:03:51 GMT9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)L8 Subject: Re: C bashing (was Re: VMS File Caching Futures+ Message-ID: <t8iq+YD1JS$P@eisner.decus.org>    In article <Pine.LNX.4.05.10006062120230.2476-100000@Mufasa.pubserv.com>, Christopher Smith <chriss@Mufasa.pubserv.com> writes:   J > I'd be perfectly happy to develop a large chunk of my code in pascal (orE > modula-N, where N is some posative integer...), but unfortunately I K > haven't been able to talk my employer into springing for a compiler -- if-J > I can even find a decent compiler to run on this billybox trash that I'm > writing apps for.. ;):  * Try Delphi.  Inside it is based on Pascal.   ------------------------------  $ Date: Tue, 6 Jun 2000 16:25:10 -0700- From: "Dann Corbit" <dcorbit@solutionsiq.com>e8 Subject: Re: C bashing (was Re: VMS File Caching Futures% Message-ID: <vLf%4.390$tV4.42@client>   L OpenVMS is written mostly in C.  I think that qualifies as a piece of robust code.   E Crappy code comes in all flavors.  Fortran, Pascal, Ada, you name it.   F C programmers care just as much about correctness as any other sort of5 programmer, if not more [because they are forced to].E  I Bashing a language is like bashing a race of people or a religion.  "It's"@ different than what I am or what I like and therefore inferior."   Sheesh.9 --0 C-FAQ: http://www.eskimo.com/~scs/C-faq/top.html$  "The C-FAQ Book" ISBN 0-201-84519-91 C.A.P. Newsgroup   http://www.dejanews.com/~c_a_p:I C.A.P. FAQ: ftp://38.168.214.175/pub/Chess%20Analysis%20Project%20FAQ.htms   ------------------------------   Date: 7 Jun 2000 01:44:04 +0200t* From: eplan@kapsch.net (Peter LANGSTOEGER)8 Subject: Re: C bashing (was Re: VMS File Caching Futures( Message-ID: <393d8cc4@news.kapsch.co.at>  U In article <vLf%4.390$tV4.42@client>, "Dann Corbit" <dcorbit@solutionsiq.com> writes:T  >OpenVMS is written mostly in C.  2 Did you check it with the VMS source listing CDs ?F I got the impression by people who have this listings (not me, sorry), that this is BULLSHIT.  M >                                 I think that qualifies as a piece of robuste >code.  F No. Only newer pieces are written in C and they appear to be much more5 flaky than earlier VMS pieces (written in BLISS, ...)i  C Look at PATHWORKS. Over 200 system crashes the last 2 years here...   F >Crappy code comes in all flavors.  Fortran, Pascal, Ada, you name it. >SG >C programmers care just as much about correctness as any other sort ofi6 >programmer, if not more [because they are forced to]. >iJ >Bashing a language is like bashing a race of people or a religion.  "It'sA >different than what I am or what I like and therefore inferior."   H Maybe. But the percentage (not count) of crappy C programs is WAY higherD than with any other programming language. And so, I see it this way:  J The worst things win (like TCPIP vs OSI, VHS vs BETAMAX/Video2000, SGML vs ODA, M$ vs the rest, and so on)i   -- i< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888N< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  $ Date: Tue, 6 Jun 2000 16:56:47 -0700- From: "Dann Corbit" <dcorbit@solutionsiq.com>t8 Subject: Re: C bashing (was Re: VMS File Caching Futures& Message-ID: <6dg%4.393$tV4.185@client>  7 "Peter LANGSTOEGER" <eplan@kapsch.net> wrote in messagee" news:393d8cc4@news.kapsch.co.at...5 > In article <vLf%4.390$tV4.42@client>, "Dann Corbit"u! <dcorbit@solutionsiq.com> writes: " > >OpenVMS is written mostly in C. >v4 > Did you check it with the VMS source listing CDs ?H > I got the impression by people who have this listings (not me, sorry), > that this is BULLSHIT.  E Really.  The Alpha required a rewrite so that the old MACRO had to be- ported.s  H > >                                 I think that qualifies as a piece of robust > >code. >aH > No. Only newer pieces are written in C and they appear to be much more7 > flaky than earlier VMS pieces (written in BLISS, ...)  >uE > Look at PATHWORKS. Over 200 system crashes the last 2 years here...o  ! Not my fault.  I didn't write it.O ;-)6  H > >Crappy code comes in all flavors.  Fortran, Pascal, Ada, you name it. > > I > >C programmers care just as much about correctness as any other sort ofl8 > >programmer, if not more [because they are forced to]. > >oL > >Bashing a language is like bashing a race of people or a religion.  "It'sC > >different than what I am or what I like and therefore inferior."b >kJ > Maybe. But the percentage (not count) of crappy C programs is WAY higher+ > than with any other programming language.s  I Produce some statistics or admit that you are completely full of crap andaK just spouting without a grain of knowlege.  The Y2K problems (for instance)oH were mostly COBOL.  The greatest amount of defects is COBOL, but that isI largely because it is also the greatest volume of code.  I have a feelingaK you are a blowhard.  Legacy programmers fear C programmers because they aren  afraid of Unix (Tru64 included).  F Defects per KLOC is pretty much a universal constant unless incredibleK effort is made to ensure quality (E.g. space shuttle efforts, and things ofr
 that ilk.)   > And so, I see it this way: >eL > The worst things win (like TCPIP vs OSI, VHS vs BETAMAX/Video2000, SGML vs! > ODA, M$ vs the rest, and so on)   I Worst things win if they are more cost effective.  People vote with theiraJ wallets.  Better sometimes costs more, for whatever reason.  That does notH mean that the people who chose something inferior were stupid.  In fact,K those who bought BetaMax instead of VHS, those who installed OSI instead ofeK TCP/IP, and those who invested in ODA are the ones with the cat in the bag.  --0 C-FAQ: http://www.eskimo.com/~scs/C-faq/top.html$  "The C-FAQ Book" ISBN 0-201-84519-91 C.A.P. Newsgroup   http://www.dejanews.com/~c_a_paI C.A.P. FAQ: ftp://38.168.214.175/pub/Chess%20Analysis%20Project%20FAQ.htm;   ------------------------------   Date: 6 Jun 2000 23:53:29 GMTv2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)8 Subject: Re: C bashing (was Re: VMS File Caching Futures6 Message-ID: <8hk2tp$t83$1@mailint03.im.hou.compaq.com>  U In article <vLf%4.390$tV4.42@client>, "Dann Corbit" <dcorbit@solutionsiq.com> writes:t  :OpenVMS is written mostly in C.  K   Hmmm.  OpenVMS is a mix of a variety of languages, and I don't think I'd lA   go as far as the use of "most" in the context of that sentence.c  M   While a fair percentage of the new OpenVMS code is written in in C, a very aH   sizeable chunk of OpenVMS is in Bliss, Macro32, and other languages.    L   I have a recollection that the last time I ran a module language count on O   the master packs, the big three (C, Macro, and Bliss) each comprised roughly sK   a third of the total number of modules.  (This count did not include the aN   "barrage" of other languages used within OpenVMS.  eg: PL/I, M64, SDL, DCL,    Fortran, C++, Ada, etc.)  F   That said, one of the salient differences around the use of C withinF   OpenVMS itself is the heavy use of descriptors (ASCID).  While ASCIZF   null-terminated strings are used, most of the system APIs use ASCID,G   with a few ASCIC and type-length-vector (itemlist) constructs around.oG   While ASCID and ASCIC are of course not a certain cure for the usual nI   sorts of buffer overruns seen, I've found more buffer-related problems rI   with ASCIZ constructs than with either ASCID or ASCIC.  And obviously, o8   most other C platforms do not use the ASCID construct.  F   There are also organizational software development norms, and these D   can make it easier or harder for bugs to exist -- bad code can be    written in any language.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 7 Jun 2000 00:32:11 GMTf2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)8 Subject: Re: C bashing (was Re: VMS File Caching Futures5 Message-ID: <8hk56b$eb$1@mailint03.im.hou.compaq.com>s  V In article <6dg%4.393$tV4.185@client>, "Dann Corbit" <dcorbit@solutionsiq.com> writes:8 :"Peter LANGSTOEGER" <eplan@kapsch.net> wrote in message# :news:393d8cc4@news.kapsch.co.at...e6 :> In article <vLf%4.390$tV4.42@client>, "Dann Corbit"" :<dcorbit@solutionsiq.com> writes:# :> >OpenVMS is written mostly in C.  :>5 :> Did you check it with the VMS source listing CDs ?*I :> I got the impression by people who have this listings (not me, sorry),e :> that this is BULLSHIT.o : F :Really.  The Alpha required a rewrite so that the old MACRO had to be :ported.  I   Um, the Macro32 compiler was used.  OpenVMS Alpha contains an extensiveoM   amount of VAX Macro32 code -- what was once processed by the VAX assembler  M   is now processed by an Alpha compiler.  The existing Macro32 code base was a   NOT rewritten "wholesale".  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Wed, 07 Jun 2000 01:17:10 GMTd= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) 8 Subject: Re: C bashing (was Re: VMS File Caching Futures0 Message-ID: <009EB385.0A8C9D0B@SendSpamHere.ORG>  V In article <6dg%4.393$tV4.185@client>, "Dann Corbit" <dcorbit@solutionsiq.com> writes:8 >"Peter LANGSTOEGER" <eplan@kapsch.net> wrote in message# >news:393d8cc4@news.kapsch.co.at...s6 >> In article <vLf%4.390$tV4.42@client>, "Dann Corbit"" ><dcorbit@solutionsiq.com> writes:# >> >OpenVMS is written mostly in C.. >>5 >> Did you check it with the VMS source listing CDs ?tI >> I got the impression by people who have this listings (not me, sorry),7 >> that this is BULLSHIT.s >tF >Really.  The Alpha required a rewrite so that the old MACRO had to be >ported.  F Sorry Dann but the truth is that the "old MACRO" that had to be portedG was done by making a MACRO compiler for the Alpha.  A few minor changes H to the sources and the MACRO32 assembler code runs as native Alpha code!  7 It seems to me that your 'solutionsiq' isn't very high!o  dJ >Produce some statistics or admit that you are completely full of crap andL >just spouting without a grain of knowlege.  The Y2K problems (for instance)I >were mostly COBOL.  The greatest amount of defects is COBOL, but that iscJ >largely because it is also the greatest volume of code.  I have a feelingL >you are a blowhard.  Legacy programmers fear C programmers because they are! >afraid of Unix (Tru64 included).,  G I fear C programmer because of the shit that they typically produce.  IcH code with C -- when I have too -- and take an extra effort to be certain that it is:9  
   1. readable B   2. not cluttered with function(function(function(function()))));9   3. not littered with the kludged/buggy C library calls    G >Defects per KLOC is pretty much a universal constant unless incredibleeL >effort is made to ensure quality (E.g. space shuttle efforts, and things of >that ilk.)E  F Clean your KLOK.  It really take very little effort to create quality E software.  It needs only a process and a desire to write quality s/w.g   --N VAXman- OpenVMS APE certification number: AAA-0001           VAXman@TMESIS.COM  L GNU Freeware -- What does the GNU *really* stand for?  Garbage!  Not Usable!   ------------------------------  + Date: Tue, 06 Jun 2000 21:21:38 -0400 (EDT)k" From: Dan Sugalski <dan@sidhe.org>8 Subject: Re: C bashing (was Re: VMS File Caching FuturesG Message-ID: <Pine.LNX.4.10.10006062117110.5810-100000@tuatha.sidhe.org>o  & On Tue, 6 Jun 2000, Dann Corbit wrote:  N > OpenVMS is written mostly in C.  I think that qualifies as a piece of robust > code.d  E Actually no, it's not. It's mostly in BLISS and Macro-32. Some of the ) newer (and, alas, buggier) stuff is in C.g  I C, unfortunately, doesn't mesh well with VMS at a system level any bettereD than it does at a user-code level, and C has a number of unfortunateH features that make it less safe as a system language than BLISS, amongstF other things. It's got the safety of Macro coupled with the feeling of$ safety of Pascal. A dangerous combo.  ,H > C programmers care just as much about correctness as any other sort of7 > programmer, if not more [because they are forced to].t  J Right, but being forced to care means that if you miss, you lose. ComputerF languages are supposed to help the programmer avoid making mistakes. CB requires the use of some constructs that are terribly error-prone.  eK > Bashing a language is like bashing a race of people or a religion.  "It'slB > different than what I am or what I like and therefore inferior."  I Believe me, I've plenty of experience in C. It's an OK language, but it's-G always second best (at best). For any application you can find a betterr	 language.    					Dan   ------------------------------  % Date: Tue, 06 Jun 2000 21:43:45 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>-8 Subject: Re: C bashing (was Re: VMS File Caching Futures, Message-ID: <393DA8C9.CD7CFA8C@videotron.ca>  & "Brian Schenkenberger, VAXman-" wrote:I > I fear C programmer because of the shit that they typically produce.  IaJ > code with C -- when I have too -- and take an extra effort to be certain
 > that it is:t    I code is as good as the programmer (or the tool that produces such code). ,  L Cobol more readable than C ? Depends who wrote it. Once used a tool that wasK supposed to simplify everything and generate perfect COBOL code. I ended uprL having to wait for the floor to be empty and stretch the listing to span theN whole floor and litterally had to run through the code to find out what it wasM doing due to all the weird paragraph names etc etc. Of course, there was CICSrJ code in there too (which goes through a pre-processor to be converted into weirder function calls).  L Any code in any language can be made unreadable. Some programmers think thatM fancy/complex C constructs result in faster programs, or impress their bossesrM with fancy routines. But once compiled with a good compiler, I am not sure itn makes much of a difference.o   ------------------------------  % Date: Tue, 06 Jun 2000 21:45:41 -0400s- From: JF Mezei <jfmezei.spamnot@videotron.ca>p8 Subject: Re: C bashing (was Re: VMS File Caching Futures+ Message-ID: <393DA93D.3428A0B@videotron.ca>t   Dan Sugalski wrote:aL > Right, but being forced to care means that if you miss, you lose. ComputerH > languages are supposed to help the programmer avoid making mistakes. CD > requires the use of some constructs that are terribly error-prone.   requires ? ? ? ? ? ? P  H Perhaps you mean "C code that can be ported to Unix" or "C code that wasW written on Windows". But I disagree that "C" by itself "requires" dangerous constructs.g   ------------------------------  + Date: Tue, 06 Jun 2000 23:15:23 -0400 (EDT)n" From: Dan Sugalski <dan@sidhe.org>8 Subject: Re: C bashing (was Re: VMS File Caching FuturesG Message-ID: <Pine.LNX.4.10.10006062310340.6118-100000@tuatha.sidhe.org>   # On Tue, 6 Jun 2000, JF Mezei wrote:    > Dan Sugalski wrote:pN > > Right, but being forced to care means that if you miss, you lose. ComputerJ > > languages are supposed to help the programmer avoid making mistakes. CF > > requires the use of some constructs that are terribly error-prone. >  > requires ? ? ? ? ? ?    H Close enough as to make no difference. The two big ones are pointers and9 null-terminated character arrays masquerading as strings.o  F > Perhaps you mean "C code that can be ported to Unix" or "C code thatG > was written on Windows". But I disagree that "C" by itself "requires"e > dangerous constructs.   H No, base C. It's certainly possible to work around it (witness both PerlJ and Python, both written in C and avoiding the pitfalls) but it requires aF lot of work and discipline. Try writing code of any complexity withoutH using pointers or null-terminated character arrays. It pretty much makes the language useless.2   				Dan1   ------------------------------   Date: 6 Jun 2000 22:49:23 CDTu= From: wayne@tachysoft.xxx.044962.killspam.0138 (Wayne Sewell)-8 Subject: Re: C bashing (was Re: VMS File Caching Futures. Message-ID: <TDQzqAIowPCd@tachxxsoftxxconsult>  \ In article <393D6C53.EDF84A37@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > Wayne Sewell wrote:m@ >> Actually, it was never designed at all, as far as I can tell. > O > C perhaps not. But ANSI-C yes. They put all sorts of pedantic restrictions onrN > it (such as the inability to move an unsigned char to a signed char and viceJ > versa without a complaint from the compiler (or the use of the /unsigned > compiler switch).- > P >> C was intended for people who don't want the compiler to "get in the way" andQ >> prefer that it let them do anything, whether correct code is generated or not.  > M > But ANSI-C changed that by being much more pedantic at compile time. If youtN > want to do your own thing, you really have to try hard by type casting stuff) > and using memcpy instead of the = sign.r  J I agree that the ANSI version is better.  I just think it should have gone farther.     -- uO ===============================================================================rM Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxe: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-)/O ===============================================================================0C Jake Blues: "Sell me your children!  How much for the little girl?"0   ------------------------------  # Date: Tue, 06 Jun 2000 21:52:40 GMTR2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>7 Subject: Re: Charon VAX on PC emulator - Asynch DECNET?.5 Message-ID: <Ioe%4.240$ru5.80198@typhoon.aracnet.com>c  & Alan Greig <agreig@my-deja.com> wrote:; > Now to beg/steal/borrow a libretto (the shirt pocket PC).   H Good luck.  I missed a chance to get one about a year and a half ago andK have regretted it since.  Not sure how well it would work for a Pocket VAX,a( but it would make a great pocket PDP-11!   			Zane    ------------------------------  $ Date: Tue, 6 Jun 2000 13:03:05 -05001 From: "Dave Gudewicz" <david.gudewicz@abbott.com>. Subject: Comments on SW RA30008 Message-ID: <8hje67$69g$1@fizban.fizban.pprd.abbott.com>  2 Anybody using the SW RA3000 Raid Array controller.  9 I just "found" one and wondered if its good, bad or ugly?p  K If given a choice between the 3000 and let's say a HSZ40/50, which might bes the better pick.  : I know one of the first answers will be "it depends".  ;-)   TIA,   Dave...t   ------------------------------  # Date: Tue, 06 Jun 2000 18:33:44 GMT + From: Staffan Malm <staffan.malm@chello.se>r) Subject: Re: Console Monitoring Software.a) Message-ID: <393D43A7.F47C408B@chello.se>S   Hi!v  ; As far as I know, the Console Manager is not dropped by CA. N There are two different packages, one is the point product (CommandIT) and the# other is bundled into UnicenterTNG. N The look and feel of the two is exactly the same as the old Polycenter Console Manager.G Both two can be integrated with Unicenter. CommandIT is integrated withA7 UnicenterTNG Framework and the other with UnicenterTNG. O This means that events that are "discovered" by CM will be sent to Unicenter as N SNMP traps and actions can be taken there (U**X or WinNT). The CM that we knowH of, proper console management, is still there, using VT or X interfaces.  N There is another product called Automation point that can do similar things as/ CM, and it runs on NT but that's another story.3   regardsk   Staffan.     Stuart Symonds wrote:$   > Hi,i >>M > Today I learnt that a product called TNG Console Manager developed first bytN > Digital and now by Computer Associates is being dropped. TNG Console ManagerN > is a fine product and from my understanding is in fairly wide spread use. CAL > have developed there own none VMS version that runs under NT but from whatN > I've heard is only half as capable as the older product and twice the price! >cM > I have heard of one VMS based product called Console Works by TECSYS, is itp! > any good, and are there others?e >i > Please reply via Email.L >, > Thanks in advance. >H	 > Stuart.    ------------------------------  % Date: Tue, 06 Jun 2000 20:45:26 -0400t+ From: "Main, Kerry" <Kerry.Main@compaq.com>l) Subject: RE: Console Monitoring Software. J Message-ID: <910612C07BCAD1119AF40000F86AF0D805284379@kaoexc4.kao.dec.com>   Stuart,o  I Most console manager products work on the basis of supporting any consoleg that outputs ascii text.  I The real issue is what do these console mgmt products support in terms ofdH message awareness ie. if a "-F-" type error happens, can the console pkgF escalate this via email, pager ect? These filter pkgs are OS specific.  E Here are a few alternative products as well (in no particular order):a  1 http://www.robomon.com/product_detail_robocen.htmh- http://www.tditx.com/tdicard/www/cwframe.htmlO  http://www.ki.com/products/clim/, http://www.globalmt.com/products/polycenter/K http://www.cai.com/products/commandit.htm (Console Manager replacement fromg CA)r   Regards,  
 Kerry Main Senior Consultant,
 Compaq Canadax Professional Servicese Voice : 613-592-4660 FAX   : 819-772-7036 Email : kerry.main@compaq.comt       -----Original Message-----8 From: Stuart Symonds [mailto:Stuart.Symonds@t-online.de]$ Sent: Tuesday, June 06, 2000 7:55 AM To: Info-VAX@Mvb.Saic.Com % Subject: Console Monitoring Software.      Hi,   K Today I learnt that a product called TNG Console Manager developed first byPL Digital and now by Computer Associates is being dropped. TNG Console ManagerL is a fine product and from my understanding is in fairly wide spread use. CAJ have developed there own none VMS version that runs under NT but from whatL I've heard is only half as capable as the older product and twice the price!  K I have heard of one VMS based product called Console Works by TECSYS, is it  any good, and are there others?.   Please reply via Email..   Thanks in advance.   Stuart.    ------------------------------   Date: 6 Jun 2000 23:05:02 +0200-* From: eplan@kapsch.net (Peter LANGSTOEGER)L Subject: Re: E-Commerce - Ariba, BroadVision & Commerce One Positions WANTED* Message-ID: <393d677e$1@news.kapsch.co.at>  R In article <8hguaa$g2a$1@nnrp1.deja.com>, elyse-marissa@sap-consulting.com writes:D >One of our clients has an immediate need for E-Commerce consultantsG >with experience in ARIBA, BroadVision or Commerce One. I have openings * >from Consultant to Senior Manager levels. > A >Requirements: 1 Full Implementation with experience using ARIBA,  >BroadVision or Commerce One.V >cG >ONLY CURRENT H1-B Visa, Green Card Holders and US Citizens NEED APPLY., >c0 >Compensation: 90K to 225K - depending on level. >o% >Openings in every region of the USA.g >eG >For more information please call Elyse at SAPient ES, Inc. at 215-345-c( >8100 or email a copy of your resume to:" >elyse-marissa@sap-consulting.com. >t >Hoping to be of Service,  >SAPient ES, Inc.u  7 Please, quick, which of this products runs on OpenVMS ?C   -- ,< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888r< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  % Date: Tue, 06 Jun 2000 14:02:19 -0400h! From: Dan Allen <dallen@nist.gov>o8 Subject: FW: C bashing (was Re: VMS File Caching Futures: Message-ID: <NEBBIALHDHJMJINPGMOAOEGGCDAA.dallen@nist.gov>  < C as invented by K&R was merely a (slightly more) convenient! packaging/abstraction of Macro-11aG with an I/O library. It was intended for "system programming" purposes,- whatever that might J have meant to you ca. 1980, if you were alive and programming in that time frame.   Not opinion, just fact.    Dan    -----Original Message-----D From: Wayne Sewell [mailto:wayne@tachysoft.xxx.044962.killspam.0138]% Sent: Tuesday, June 06, 2000 11:40 AMs To: Info-VAX@Mvb.Saic.Comt4 Subject: C bashing (was Re: VMS File Caching Futures    G In article <4.3.2.7.0.20000605150202.01c6da80@24.8.96.48>, Dan SugalskiE <dan@sidhe.org> writes:t1 > At 11:59 AM 6/5/00 -0400, David A Froble wrote: L >>Exactly!  This is what makes C inappropriate for applications development. >>6 > C was never designed as an app development language,  = Actually, it was never designed at all, as far as I can tell.i   >and it's a prettyJ > pathetic one. It lacks the features you'd usually want in an applicationI > development language, and its standard library's badly suited for it as.F > well. C's a low-level systems language, meant for writing compilers, deviceI > drivers, OS kernel bits, and suchlike things, and it's not bad at that.r    H I would amend that to say it's meant for writing such stuff *in a hackerI fashion*.  (I am of course using the term "hacker" in the original sense;sJ nothing to do with malicious intent.)   While C does have low level access foriG stuff, this is not to say that strongly-typed languages do not.  I have  written,K compilers in Pascal and also kernel mode code.  You have to be careful witheH what you do and make sure that you don't reference any RTL routines, but thisJ would apply to C as well.  I would say that Bliss can do *anything* that C can,J and probably more, but will tend to prevent stupid typos from crashing the system.n  J I would say that using a real language is *more* important for kernel modeL stuff than for applications, because the cost of failure is so much greater,@ i.e. crashing the entire system instead of just the application.    G Speaking of compilers, they are no different from applications from theiI viewpoint of the system and the machine code.  They execute in user mode,h readG an input file, and create an output file.  You can write compilers in at* strongly typed language as easily as in C.  I C was intended for people who don't want the compiler to "get in the way"  andeI prefer that it let them do anything, whether correct code is generated or! not.H Strangely, they prefer to defer their errors to run time, where they are muchI harder to diagnose.  Sure the compile is faster.  Good thing, because youo haveL to compile so many more times to correct errors that would have been trapped in8 the first compile by Pascal or the Countess (see below).  F Note that we are talking about differences in *compilation* speed, notF *execution* speed of the generated code.  Depending on the compiler, aL strongly-typed language can generate code as good as or better than a sloppyJ one such as C.  Most of it depends on the back end.  Admittedly, it may beJ possible to generate slightly tighter code with a simpler (more primitive)L language such as C, but an industrial strength back end such as GEM makes upG for this to a large extent.  Both Compaq C and Compaq Pascal use GEM oni alpha.F I would be surprised if there is that much difference in the generated machine  code for a given algorithm.        >ItlL > leans a bit too heavily on the speed side of the speed/safety tradeoff forE > my tastes, but it was developed on hardware that is, at this point,A > terribly slow.  K That's an excuse for the original sad state of C, but not for C today.  Theh CdI standards committee had a chance to fix this stuff.  Admittedly, they did-C improve the language somewhat, since ANSI C will at least do *some*a	 checking,<E unlike the K & R shit,  but C still lets a lot of stuff slip through.s    I Admittedly, there *is* a run time performance hit for stuff such as arraynJ bounds checking.  This would fall into the speed/safety tradeoff mentionedI above.   Seems like 95% percent of the C run time errors I have found areoH related to exceeding a local stack-based string variable and wildwriting intoK the stack.  At least in Pascal you have the option of checking array boundsvI during development, then you can turn it off when the program is releasedf fora production.t     >eA > Don't knock C for being a bad app language, since it isn't one.s  E I consider it a bad language for any purpose.  But I'm stuck with it.cJ Unfortunately, it's been so long since I've used Pascal that it would take time to come back up to speed on it.   
 >Knock theJ > twits that insist on using it as an app language--that's where the blame > rightly lies.o   Agreed.r  I > (I'm still not particularly fond of the language, mind, but its biggestuF > problem is its mis-application, not in any fundamental design flaws)  J Disagreed.   It was fine for the hacker days when people were just playing withF computers, but not for production systems in *any* role.   Admittedly, eunuchstH systems don't have anything to replace it for kernel code, but vms does.    J The Countess referred to above is Augusta Ada Byron, Countess of Lovelace,5 daughter of Lord Byron.  Also known as just Ada.  :-)0   --L ============================================================================ ===(8 Wayne Sewell, Tachyon Software Consulting  (281)812-0738 wayne@tachysoft.xxx 8 http://www.tachysoft.xxx/www/tachyon.html and wayne.htmlK change .xxx to .com in addresses above, assuming you are not a spambot  :-)wL ============================================================================ ===iC Jake Blues: "Sell me your children!  How much for the little girl?"    ------------------------------  # Date: Tue, 06 Jun 2000 18:29:31 GMTt= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)L< Subject: Re: FW: C bashing (was Re: VMS File Caching Futures0 Message-ID: <009EB34C.17D2DA34@SendSpamHere.ORG>  ^ In article <NEBBIALHDHJMJINPGMOAOEGGCDAA.dallen@nist.gov>, Dan Allen <dallen@nist.gov> writes: >h >m= >C as invented by K&R was merely a (slightly more) convenienth" >packaging/abstraction of Macro-11H >with an I/O library. It was intended for "system programming" purposes, >whatever that mightK >have meant to you ca. 1980, if you were alive and programming in that times >frame.e >t >Not opinion, just fact. >i >Dan  H I was alive and programming in 1980 using Fortran and Macro32 on VMS.  CI _was_ alive at that time but it was alive mostly in the 'land of eunuchs'5 where it ought to have stayed.   Not opinion, just fact. ;)   --N VAXman- OpenVMS APE certification number: AAA-0001           VAXman@TMESIS.COM  L GNU Freeware -- What does the GNU *really* stand for?  Garbage!  Not Usable!   ------------------------------  % Date: Tue, 06 Jun 2000 14:42:22 -0400p" From: Dan Sugalski <dan@sidhe.org>< Subject: Re: FW: C bashing (was Re: VMS File Caching Futures: Message-ID: <4.3.2.7.0.20000606144050.01d21d20@24.8.96.48>  * At 02:02 PM 6/6/00 -0400, Dan Allen wrote:  = >C as invented by K&R was merely a (slightly more) convenientrL >packaging/abstraction of Macro-11 with an I/O library. It was intended for J >"system programming" purposes, whatever that might have meant to you ca. < >1980, if you were alive and programming in that time frame.  L ca. 1970, IIRC. Certainly quite a ways before '80. I think,but I'm not 100% J sure, that it predated the PDP-11. (I'm thinking the PDP-8, though I have  no proof at hand)i   					Dan  I --------------------------------------"it's like this"-------------------a2 Dan Sugalski                          even samurai? dan@sidhe.org                         have teddy bears and even ;                                       teddy bears get drunkr   ------------------------------  $ Date: Tue, 6 Jun 2000 17:08:51 -07005 From: "cstranslations" <cstranslations@email.msn.com>i< Subject: Re: FW: C bashing (was Re: VMS File Caching Futures) Message-ID: <#jQKISB0$GA.347@cpmsnbbsa08>   H Brian Schenkenberger, VAXman- <system@SendSpamHere.ORG> wrote in message* news:009EB34C.17D2DA34@SendSpamHere.ORG...F > In article <NEBBIALHDHJMJINPGMOAOEGGCDAA.dallen@nist.gov>, Dan Allen <dallen@nist.gov>o  ? > >C as invented by K&R was merely a (slightly more) conveniente$ > >packaging/abstraction of Macro-11J > >with an I/O library. It was intended for "system programming" purposes, > >whatever that mightH > >have meant to you ca. 1980, if you were alive and programming in that time	 > >frame.e  L According to the preface (to both the first and second edition) which is/areJ included in the 2nd edition of K&R's "The C Programming Language" (a very,G very, small book as compared to, oh, say, the phone book sized tomes ondG Visual Basic and Visual C++ over there on the book case collecting dust I which in spite of their size are absolutely useless - save for door stopsmI and collecting dust, but then . . . ) the first edition of K&R was out in H 1978. It was based on B which Thompson wrote in 1970. That would seem to2 place its appearance some where in the early '70s.  3 > I was alive and programming in 1980 using Fortranm  . Well - there's no accounting for taste :-) :-)   Joe    ------------------------------  # Date: Tue, 06 Jun 2000 22:42:12 GMT.) From: Jeff Kenward <jkenward@my-deja.com>r Subject: HELP! Motif Weirdness) Message-ID: <8hjuo0$p6r$1@nnrp1.deja.com>   @ We have an application that is doing a lot of xmcreates (variousE widgets) and carting them off to a PC running Exceed (same problem in= Excursion).=  G It seems to get to a point where the application hangs during a call touH one of the xmCreates.  It seems like a resource limit issue because if IF tell it to not create the current set of widgets, it just hangs on the	 next set.a  D I only have this problem when the transport is TCPIP.  It works fine when I'm using DECNET.  E Anyone got any ideas which resource limit I might be hitting or otherr ideas?   Thanks!-   -Jeff- jeffatwork@kenward.org    & Sent via Deja.com http://www.deja.com/ Before you buy.l   ------------------------------  # Date: Wed, 07 Jun 2000 00:42:02 GMT  From: Dan <dan@vrx.net>5) Subject: Hobbyist vms 7.2 alpha questions<' Message-ID: <393D99D9.50D75BF9@vrx.net>s  > I tried installing tcpip but the system said I already had UCX
 installed.1 I don't have a ucx$config anywhere on the system.r. I don't have tcp$config anywhere on the system7 I can't figure out the product name for decnet phase ivPB I can't do a product remove of ucx because it says it's not there.H I do a product show product * and it doesn't list ucx, decnet, or tcpip.  F How can I use the 7.2 hobbyist CD to upgrade my VMS without wiping out all my data?   Dan.   ------------------------------   Date: 6 Jun 2000 21:21:36 GMT"# From: system@niuhep.physics.niu.edu * Subject: Re: I'm Trying to Model Something+ Message-ID: <8hjq10$l13$1@husk.cso.niu.edu>h  $ simpler <brad@chem.wisc.edu> writes: >Can you help me?i >e >Here's a problem: > E >Let's say there are M open mail relay systems and N active spammers.: >4 >M >> Nr > H >Do you agree with me that, until M is on the order of N,  reducing M by8 >blocking open relay will have no effect on spam volume? > F >I've tried to say that in news.admin.net-abuse.email, but they aren't$ >buying it.  Am I missing something?  J You need to take into account P, the total number of possible relay sites,% i.e the number of sites running smtp.a  @ When M/P is close to one, it is effortless to find a relay site.I If M/P is close to 1/1000 then life becomes significantly more difficult,d even though M>>N.  n Morphis@physics.niu.edu 8 Real Women change tires			abuse@uu.net postmaster@uu.net7 Real Men change diapers                 security@uu.netr   ------------------------------  # Date: Tue, 06 Jun 2000 18:22:26 GMTr= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) ' Subject: Re: ICC programming examples ?g0 Message-ID: <009EB34B.1A08E929@SendSpamHere.ORG>  \ In article <393D1704.A039268C@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:F >Are there any ICC (intra cluster communications) programming examples( >available ? (server and client sides) ? >GN >Also, what was the reasoning behind the selction of a non QIO based interfaceN >? It would have made things much simpler/versatile for programs to have a QION >interface which would have enabled a single program to handle mailbox, decnet+ >and ICC connections with much shared code.   H The ICC services are using the SCS communications layer and if you were G familiar with the internal routines which provide SYSAP communications,HG you'd better appreciate/understand why the ICC services exist in their 2
 present form.      --N VAXman- OpenVMS APE certification number: AAA-0001           VAXman@TMESIS.COM  L GNU Freeware -- What does the GNU *really* stand for?  Garbage!  Not Usable!   ------------------------------   Date: 6 Jun 2000 18:58:06 CDTo= From: wayne@tachysoft.xxx.044962.killspam.0138 (Wayne Sewell)i' Subject: Re: ICC programming examples ?p. Message-ID: <vHCkkrG3ZGUA@tachxxsoftxxconsult>  \ In article <393D1704.A039268C@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:G > Are there any ICC (intra cluster communications) programming examplesp) > available ? (server and client sides) ?m > O > Also, what was the reasoning behind the selction of a non QIO based interfacewO > ? It would have made things much simpler/versatile for programs to have a QIOwO > interface which would have enabled a single program to handle mailbox, decnetd, > and ICC connections with much shared code.    O Personally, I like the high level ICC routines.  I especially like the way theyeI are geared to client/server, i.e. the client performs a single transceivesL operation which handles both the request and the response.  I don't mind theO differing methods for i/o.  You already have that between decnet and tcp.  EvenaL though they both can use qio, it's not the *same* qio.   Okay, the writevblkN and readvblk qios may be the same, but the setup qios are quite different.   IN have a generic communications module in my RTL.  The application programs callL the same routines for communications, irregardless of the transport (decnet,F icc, tcp).  Therefore, even though different methods are used for each0 transport, they can be found in only one module.   Wayne  -- SO ===============================================================================iM Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxx:: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-)eO ===============================================================================oC Jake Blues: "Sell me your children!  How much for the little girl?"    ------------------------------  % Date: Tue, 06 Jun 2000 21:05:55 -0400o+ From: "Main, Kerry" <Kerry.Main@COMPAQ.COM>fD Subject: RE: is anyone able to get a DE500-BA working under 6.2 1h3?J Message-ID: <910612C07BCAD1119AF40000F86AF0D80528437A@kaoexc4.kao.dec.com>  L Yep, as long as you have the patches applied, a DE500-BA works fine with 6.2H 1H3 of OpenVMS. We had a Customer running with a pile of these adapters.* They have since upgrade to OpenVMS V7.2-1.  L Biggie watchout is to not use autonegotiate. There appears to be a number ofG differences in terms of defining exactly what the right way to do this.fJ Typical problems are that the link appears to be working, but very slowly.  E Solution is to physically set the connections at both the console and & network switch end. 100Mb, Full Duplex   >> SET EWA0_MODE FASTFD-   Regards,    
 Kerry Main Senior Consultant,
 Compaq Canadan Professional Servicesn Voice : 613-592-4660 FAX   : 819-772-7036 Email : kerry.main@compaq.com        -----Original Message-----0 From: The Youngun [mailto:ftg_dts@earthlink.net]$ Sent: Tuesday, June 06, 2000 9:09 AM To: Info-VAX@Mvb.Saic.Com-@ Subject: is anyone able to get a DE500-BA working under 6.2 1h3?    E Is anyone able to get a DE500-BA working under Open VMS 6.2 1H3 on anF AlphaServer 800?  4 Per COMPAQ, the following patches have been applied:   alpclusio01R	 alpcliu03  alplan05	 alpboot12v   Thanks in advance, The Youngunc   ------------------------------  % Date: Tue, 06 Jun 2000 14:40:13 -0400s" From: Dan Sugalski <dan@sidhe.org>* Subject: Re: Java graphing applets for VMS: Message-ID: <4.3.2.7.0.20000606143932.01d81e50@24.8.96.48>  5 At 01:56 PM 6/6/00 -0400, jlahman@LTVSteel.com wrote:[  K >We want to present production data to the end users in a graphical format 1I >via a browser.  There are graphical java applets available for Unix and C	 >Windoze.r= >However, I have not seen any graphical java applets for VMS.8  J You're running the applets in a browser on a PC. It doesn't *matter* what 1 OS the machine running the webserver's running...i   					Dan  I --------------------------------------"it's like this"-------------------e2 Dan Sugalski                          even samurai? dan@sidhe.org                         have teddy bears and even ;                                       teddy bears get drunk    ------------------------------  % Date: Tue, 06 Jun 2000 14:30:52 -0400 # From: John Vottero <John@MVPSI.com>e* Subject: RE: Java graphing applets for VMS: Message-ID: <C15945A9D9EFCF11BA8B08002BBF1CCC0CD719@BERRY>  ; Wait a minute, what happened to "Write once, run anywhere"?g   > -----Original Message-----: > From: jlahman@LTVSteel.com [mailto:jlahman@LTVSteel.com]& > Sent: Tuesday, June 06, 2000 1:57 PM > To: Info-VAX@Mvb.Saic.Como( > Subject: Java graphing applets for VMS >  > 	 > Hi all:- > ; > We want to present production data to the end users in a n > graphical format via a@ > browser.  There are graphical java applets available for Unix  > and Windoze.> > However, I have not seen any graphical java applets for VMS. > = > Has anybody ported any of these graphical java appliets to t > VMS? or are there anyp+ > graphical java applets available for VMS?l >  > Thanks >  >  >    ------------------------------  " Date: Tue, 6 Jun 2000 21:54:53 GMT3 From: cornelius@eisner.decus.org (George Cornelius).* Subject: Re: Java graphing applets for VMS+ Message-ID: <KIsKUcG2L7BZ@eisner.decus.org>t  U In article <852568F6.0062A658.00@notesnta.LTVSteel.com>, jlahman@LTVSteel.com writes: 	 > Hi all:  > Q > We want to present production data to the end users in a graphical format via acL > browser.  There are graphical java applets available for Unix and Windoze.> > However, I have not seen any graphical java applets for VMS. > R > Has anybody ported any of these graphical java appliets to VMS? or are there any+ > graphical java applets available for VMS?u  ? I thought the whole point of Java was that it could be platform > independent.  As long as an applet did not rely on any special= features of its environment one would think it would run justh< as well under OpenVMS Alpha, and would do so without so much as a recompile.i  ? If there are graphical differences between a Windows and a Unix @ environment, I would expect the Unix one, being X11 based, would4 be roughly the same as a VMS DECWindows environment.  @ All of this is speculation, but in theory that is supposed to be the way it works.    --8 George Cornelius              cornelius@eisner.decus.org0                               cornelius@mayo.edu   ------------------------------   Date: 6 Jun 2000 18:52:58 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)! Subject: Re: MMOV 2.2 and VMS 7.1t6 Message-ID: <8hjhaa$m0l$1@mailint03.im.hou.compaq.com>  x In article <7a1+fuLOFdUc@gaelic>, pmoreau@cenaath.cena.dgac.fr (Patrick MOREAU, CENA Athis, Tel: 01.69.57.64.40) writes:N :Multimedia Services for OpenVMS is available on the June CD's with support of2 :Ensoniq card and DSx0 workstations and servers !! :oN :The minimum OVMS version is 7.1-2, but is there a chance to see it working onL :7.1H1 ? (My Home Alpha 255 runs 7.1-1H1 with a good stability and if I can  :avoid an update ...)<  D   I would strongly encourage an upgrade to (at least) OpenVMS Alpha E   V7.1-2 for any system currently running V7.1, V7.1-1H1 or V7.1-1H2.aE   PCSI is used for all ECO and update operations starting in V7.1-2, oC   and we also collected up a large variety of ECOs and rolled them sD   all into V7.1-2, and we have subsequently made various UPDATE kitsC   available for V7.1-2.  All of which make the use of OpenVMS Alpha "   V7.1-2 preferable over V7.1-1H1.  D   I would strongly discourage trying to use an OpenVMS Alpha V7.1-2 *   kit on an earlier OpenVMS Alpha release.  C   All EV6- and EV67-class boxes are also first supported on V7.1-2.i  D   The AlphaStation 255 also runs rock-solid on OpenVMS Alpha V7.2-1,8   with the current GRAPHICS and UPDATE ECO kits applied.  0   Bottom line: are you feeling lucky today?  :-)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 6 Jun 2000 18:25:34 GMTo- From: "Phil Tregoning" <ptregoni@esoc.esa.de>.  Subject: Re: motif debug problem= Message-ID: <01bfcfe3$1201d8e0$4b53b083@ptregoni.esoc.esa.de>o  8 Dave Weatherall <djweath@attglobal.net> wrote in article, <DTiotGxQ0bj6-pn2-mvGQR9OYzOYY@localhost>...5 > On Sun, 5 Jun 3900 23:20:04, manser@decus.de wrote:o > H > ====================================================================== > > here are my questions  > >  > >  > > 1) what is going wrong ?: > > 2) how can i use the normal (not the motif) debugger ? > >  > > thanks for any help8 > 5 > I can't answer Q.1 but Q.2 should be something likew >  >      define dbg$decw$input ""e  >      define dbg$decw$output "" >   / I think this is because the debugger can't find=. VMSDEBUGUIL.UID. Your directory listing showed/ it was present, so maybe the DECwindows logicalr( is incorrect. This should find the file:  * $ DIR DECW$SYSTEM_DEFAULTS:VMSDEBUGUIL.UID  5 If it doesn't, find out why not (DECW$SYSTEM_DEFAULTSi/ should include SYS$LIBRARY in its search path).b   Phil   ------------------------------  % Date: Tue, 06 Jun 2000 17:45:21 +0200r2 From: martin@RADIOGAGA.HARZ.DE (Martin Vorlaender)  Subject: Re: motif debug problem; Message-ID: <393d1c91.524144494f47414741@radiogaga.harz.de>a   manser@decus.de wrote: : I have the following problem :-8 : i am running decc v5.6, decw motif 1.2-3, vax vms v6.2 :m$ : Here are the outputs of my session :v : $cc/debug xxx. : $link/debug xxxa
 : $run xxxP : X Toolkit Warning: I18NOpenFile: Couldn't open file vmsdebuguil - MrmNOT_FOUND : MrmOpenHierarchy ERROR ...s : 1) what is going wrong ?  O I've no idea. Normally UID files are supposed to reside in DECW$SYSTEM_DEFAULTS  or DECW$USER_DEFAULTS.  8 : 2) how can i use the normal (not the motif) debugger ?     $ DEFINE DBG$DECW$DISPLAY " "    cu,    Martin --D                        |  Martin Vorlaender  |  VMS & WNT programmer1   OpenVMS: When you    |  work: mv@pdv-systeme.de H   KNOW where you want  |        http://www.pdv-systeme.de/users/martinv/8   to go today.         |  home: martin@radiogaga.harz.de   ------------------------------   Date: 7 Jun 2000 01:08:53 +0200t* From: eplan@kapsch.net (Peter LANGSTOEGER) Subject: Re: MOZILLA M16 ?* Message-ID: <393d8485$1@news.kapsch.co.at>  W In article <3934F606.FC204858@theblakes.com>, Colin Blake <colin@theblakes.com> writes:  >Peter LANGSTOEGER wrote:h6 >> I know that M16 and M17 are not out, yet. But why ? >rX >M16 is meant to be the "feature complete" milestone. The dates were pushed out to allowY >time for everything to come in. The tree closed last week and is currently going through 8 >"stabilization". Shouldn't be more than a few days now.  - What is "few" ? Much more than 7 it seems ;-)-  J >> And the milestone list did get updated the last weeks more than once... > U >If the milestone list were up to date, M16 would have shipped over two weeks ago :-)r  2 Seems that the milestone plan got updated again...F M16 should now be out ("on the wire") since 2 days and in fact is not.  L Do you know, if M16 has then/now a working AUTOPROXY support (on U**X/VMS) ?  O Do you know, when GTK+ will be available as an extra (PCSI) kit for all of us ?@   TIAm   -- G< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888,< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------   Date: 6 Jun 2000 23:34:57 GMTe2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: MOZILLA M16 ?6 Message-ID: <8hk1r1$ss6$1@mailint03.im.hou.compaq.com>  W In article <393d8485$1@news.kapsch.co.at>, eplan@kapsch.net (Peter LANGSTOEGER) writes:   D   OpenVMS Engineering is porting the Mozilla browser code base over !   to OpenVMS.  Key word: porting.e  3 :Seems that the milestone plan got updated again...   8   Um, Mozilla schedules are nothing if not flexible. :-)  G :M16 should now be out ("on the wire") since 2 days and in fact is not.m  H   So?  (Without intended that remark to appear rude.)  Put another way, E   M16 will be out when it is out -- the Mozilla schedules are usually-G   best/rough estimates, and are definitely not a useful substitute for  .   a clock (or sometimes even a calendar).  :-)  M :Do you know, if M16 has then/now a working AUTOPROXY support (on U**X/VMS) ?p  F   Consider checking the Mozilla website or checking directly with the D   Mozilla folks themselves -- and the OpenVMS port is based off the G   Linux port, so you can watch to see what that port has, and what its     current plans are, etc.   P :Do you know, when GTK+ will be available as an extra (PCSI) kit for all of us ?  I   There is a discussion going around releasing the GTK stuff for OpenVMS  K   via Freeware or elseware, but I am not immediately aware of a particular iF   decision as yet.  (Nor am I aware of any schedule for reaching this I   particular decision.  :-)  And if the GTK stuff will actually be going .?   onto the Freeware, I'll likely hear about it sooner or later.   F   As it is quite clear that you want the autoproxy stuff working, you D   could always seek to join the development team that is working on H   Mozilla itself, and get the autoproxy stuff working -- this is simply L   the way that the Mozilla code base is developed and debugged and enhanced.  J   Once the particular feature(s) are available and working in the Mozilla I   code base, OpenVMS Engineering will endeavor to get it all ported over  F   to OpenVMS, and operating in a fashion compatible with the behaviour<   of the original Linux port.  Keywords: ported, compatible.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Tue, 06 Jun 2000 23:39:29 GMTi' From: Colin Blake <colin@theblakes.com>y Subject: Re: MOZILLA M16 ?- Message-ID: <393D8C09.56134160@theblakes.com>r   Peter LANGSTOEGER wrote:  Y > In article <3934F606.FC204858@theblakes.com>, Colin Blake <colin@theblakes.com> writes:n > >Peter LANGSTOEGER wrote:w8 > >> I know that M16 and M17 are not out, yet. But why ? > > Z > >M16 is meant to be the "feature complete" milestone. The dates were pushed out to allow[ > >time for everything to come in. The tree closed last week and is currently going throughe: > >"stabilization". Shouldn't be more than a few days now. >t/ > What is "few" ? Much more than 7 it seems ;-)h  ] Everything I was reading lead me to believe it was real close. What can I say, except that wee are closer now :-)  L > >> And the milestone list did get updated the last weeks more than once... > >uW > >If the milestone list were up to date, M16 would have shipped over two weeks ago :-)y >m4 > Seems that the milestone plan got updated again...H > M16 should now be out ("on the wire") since 2 days and in fact is not.  = Yep, this is what makes estimated release dates difficult :-)   N > Do you know, if M16 has then/now a working AUTOPROXY support (on U**X/VMS) ?  < It doesn't look like it. The only autoproxy bug I could findE (http://bugzilla.mozilla.org/show_bug.cgi?id=16773) is marked as M20.a  Q > Do you know, when GTK+ will be available as an extra (PCSI) kit for all of us ?g  X You can use the shareable images from a Mozilla kit now. Up to M15 is GTK+ 1.2.0. M16 is\ 1.2.7 (or maybe 1.2.8). Details on getting the source code available are being ironed out as* we speak. Are you in desperate need of it?   ------------------------------   Date: 7 Jun 2000 02:04:37 +0200'* From: eplan@kapsch.net (Peter LANGSTOEGER) Subject: Re: MOZILLA M16 ?* Message-ID: <393d9195$1@news.kapsch.co.at>  k In article <8hk1r1$ss6$1@mailint03.im.hou.compaq.com>, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:rX >In article <393d8485$1@news.kapsch.co.at>, eplan@kapsch.net (Peter LANGSTOEGER) writes: > E >  OpenVMS Engineering is porting the Mozilla browser code base over :" >  to OpenVMS.  Key word: porting.  ( Thanks. That was really news for me. NOT  4 >:Seems that the milestone plan got updated again... >j9 >  Um, Mozilla schedules are nothing if not flexible. :-)i   Ok. Now I know it, too.)B Isn't good to have a date, when to check again, instead of readingA COMP.OS.VMS all day and waiting if someone posts "it's out now" ?   N >:Do you know, if M16 has then/now a working AUTOPROXY support (on U**X/VMS) ? >oG >  Consider checking the Mozilla website or checking directly with the nE >  Mozilla folks themselves -- and the OpenVMS port is based off the  H >  Linux port, so you can watch to see what that port has, and what its  >  current plans are, etc.  K This is hard work. I already did it twice. And asking some guys involved in M porting and testing is way easier ;-) And, solution was promised to be M15....  Q >:Do you know, when GTK+ will be available as an extra (PCSI) kit for all of us ?  >sJ >  There is a discussion going around releasing the GTK stuff for OpenVMS L >  via Freeware or elseware, but I am not immediately aware of a particular G >  decision as yet.  (Nor am I aware of any schedule for reaching this oJ >  particular decision.  :-)  And if the GTK stuff will actually be going @ >  onto the Freeware, I'll likely hear about it sooner or later.   Ok.f  G >  As it is quite clear that you want the autoproxy stuff working, you =E >  could always seek to join the development team that is working on oI >  Mozilla itself, and get the autoproxy stuff working -- this is simply tM >  the way that the Mozilla code base is developed and debugged and enhanced.   M Me. A C programmer ? Are you dreaming ? FORTRAN, BASIC, PASCAL, ASM51, ASM85,=M ASM86, PLM51, PLM80, PLM86, DCL and some bits of MODULA, but no C, C++, JAVA,mO Visual Basic, Visual Studio, Delphi. My programmer times have been years ago...   K >  Once the particular feature(s) are available and working in the Mozilla :J >  code base, OpenVMS Engineering will endeavor to get it all ported over G >  to OpenVMS, and operating in a fashion compatible with the behavioura= >  of the original Linux port.  Keywords: ported, compatible.   H Which is the point. Is the porting effort started after the milestone is: released or during the whole milestone development phase ?   -- o< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888s< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------   Date: 7 Jun 2000 02:19:39 +0200i* From: eplan@kapsch.net (Peter LANGSTOEGER) Subject: Re: MOZILLA M16 ?* Message-ID: <393d951b$1@news.kapsch.co.at>  W In article <393D8C09.56134160@theblakes.com>, Colin Blake <colin@theblakes.com> writes:t >Peter LANGSTOEGER wrote:.O >> Do you know, if M16 has then/now a working AUTOPROXY support (on U**X/VMS) ?F >l= >It doesn't look like it. The only autoproxy bug I could findpF >(http://bugzilla.mozilla.org/show_bug.cgi?id=16773) is marked as M20.  M It was promised for M15, then M17, then on 16-May-2000 as "to be fixed by the1K end of the week" and now for M20. The original bug entry seems to be 20145,g5 but there are a lot of duplicates (more than 15) now.. And M20 is 2001, I think.0  N So, I'm still stuck with NETSCAPE V3 for some another months. And I got beatenL for telling, that "not porting NETSCAPE V4 to VMS is a bad decision" then...  R >> Do you know, when GTK+ will be available as an extra (PCSI) kit for all of us ? > Y >You can use the shareable images from a Mozilla kit now. Up to M15 is GTK+ 1.2.0. M16 iss] >1.2.7 (or maybe 1.2.8). Details on getting the source code available are being ironed out as-+ >we speak. Are you in desperate need of it?-  J No, but I like the idea of products (games ?) based on GTK+ to soon pop up- on VMS, too. Maybe only a wishful thinking...s   -- e< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-8880< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  # Date: Wed, 07 Jun 2000 00:43:23 GMT 2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com> Subject: Re: MOZILLA M16 ?5 Message-ID: <LUg%4.243$ru5.81382@typhoon.aracnet.com>n  + Peter LANGSTOEGER <eplan@kapsch.net> wrote:wL > No, but I like the idea of products (games ?) based on GTK+ to soon pop up/ > on VMS, too. Maybe only a wishful thinking...t  J I've no idea what it would take, but what would be *really* cool is a portL of gnumeric, and AbiWord.  I believe both are GTK based apps.  I'm seriously/ impressed by what little I've seen of Gnumeric.-   			Zanee   ------------------------------  + Date: Tue, 06 Jun 2000 21:16:59 -0400 (EDT)s" From: Dan Sugalski <dan@sidhe.org> Subject: Re: MOZILLA M16 ?G Message-ID: <Pine.LNX.4.10.10006062115370.5810-100000@tuatha.sidhe.org>.  & On Tue, 6 Jun 2000, Colin Blake wrote:  G > > Do you know, when GTK+ will be available as an extra (PCSI) kit forn > > all of us ?a > G > You can use the shareable images from a Mozilla kit now. Up to M15 is C > GTK+ 1.2.0. M16 is 1.2.7 (or maybe 1.2.8). Details on getting theMD > source code available are being ironed out as we speak. Are you in > desperate need of it?r  E For the interested, I just saw an announcement on the freeciv mailingJI list. Someone did a port of freeciv (a Civilization/Civ II clone) to VMS,nI using the GTK stuff in Mozilla as a base. Appparently it's up and runningd% and took basically no effort to port.,   					Dan   ------------------------------  + Date: Tue, 06 Jun 2000 21:24:55 -0400 (EDT)e" From: Dan Sugalski <dan@sidhe.org> Subject: Re: MOZILLA M16 ?G Message-ID: <Pine.LNX.4.10.10006062122060.5810-100000@tuatha.sidhe.org>o  ' On 7 Jun 2000, Peter LANGSTOEGER wrote:n  m > In article <8hk1r1$ss6$1@mailint03.im.hou.compaq.com>, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:gZ > >In article <393d8485$1@news.kapsch.co.at>, eplan@kapsch.net (Peter LANGSTOEGER) writes:; > >  Um, Mozilla schedules are nothing if not flexible. :-)a >  > Ok. Now I know it, too.MD > Isn't good to have a date, when to check again, instead of readingC > COMP.OS.VMS all day and waiting if someone posts "it's out now" ?.  : Peter, Colin and Hoff have both been excessively un-blunt.  E Don't wait for an announcement here. Watch the mozilla website, sincerH that's where things get announced. It's not Colin's fault that M16 isn'tI out for *any* platform, and grumbling about not having a VMS version of a? non-existant release is silly.  F Any bugs in Mozilla should *also* be filed with the mozilla folks, notD mentioned here. If autoproxy doesn't work, then file a bug report at; mozilla.org. Nobody can fix it if they don't know about it.    					Dan   ------------------------------  % Date: Tue, 06 Jun 2000 15:51:32 -0400o) From: Pam Chester <Pam.Chester@bhusa.com>e! Subject: OpenVMS AND THE INTERNET D Message-ID: <7141F6498C56D311A3A100508B2CCBCE5D9637@imail.bhusa.com>  < More on the book OpenVMS AND THE INTERNET and Digital Press:  L This title was under contract at Digital Press. It was cancelled sometime inI late 1998 or 1999, by mutual agreement--as I remember, it was actually atlJ the author's suggestion. He felt that the OpenVMS products and initiativesI he'd planned to discuss were not going to be developed after all, or were L being taken in other directions, and that he wouldn't be able to produce the  book he had originally proposed.  K It takes a while to get the book pulled from Amazon, B&N, etc. once it's inn@ their systems--Sorry if there was poor communication about that.  L Do you feel there's a market for a book on OpenVMS and apache, or some other OpenVMS and Internet topics?  K Digital Press aims to serve the OpenVMS market, so we're eager to know whate! kinds of books you'd like to see.l  D Please don't hesitate to contact us if there's a book you want us to publish.   Pam Chester: Associate Editor
 Digital Press  225 Wildwood Avenue= Woburn, MA  01801= 781-904-2603 pam.chester@bhusa.com    ------------------------------  # Date: Tue, 06 Jun 2000 18:38:02 GMTh1 From: "Mark D. Jilson" <jilly@clarityconnect.com> J Subject: Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire)2 Message-ID: <393D44ED.4290C1B5@clarityconnect.com>  E I guess I'm confused by what you call 'normal support'.  To me normaluF support includes being able to elevate an issue to an engineering teamD to get a fix for the issue.  PVS allows that on old version.  If youG haven't purchased PVS then you  may not ask for an issue to be elevatedsE to an engineering team, doesn't mean the CSC won't talk to you if youYE have still purchased a CSC support contract for remedial and advisoryeF services.  If a version doesn't have a PVS option then you cannot evenG purchase the ability to ask that an issue be elevated to an engineering2D team but you may still purchase a CSC support contract.  I routinelyG handle VMS calls for V4.n and even some V3.n.  This is what is known as E BE or Best Effort.  We see what we can do via our knowledge bases but H the CSC won't spend a lot of time on an issue.  I know of no VMS supportH center within Compaq that will refuse a VMS call for a BE version of VMS1 if the customer has a valid CSC support contract.m   Tim Llewellyn wrote: >  > "Mark D. Jilson" wrote:F > f > > Please see http://www.openvms.digital.com/extsup.html for a look atsupport plans that have been in  > > place since well before Y2K. > \ > Sure Mark, I am well aware of that page and what it says. I was asking for more details on] > the UK only specific stuff that Steve posted about. It turns out this only relates to which/R > versions are supported under prior version support rather than "normal support". > 	 > Regards- >  --e8 > Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project2 > MedAS at the BBC, Whiteladies Road, Bristol, UK.C > Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukh > C > I speak for myself only and my views in no way represent those ofG > MedAS or the BBC.    -- 6D Jilly	- Working from Home in the Chemung River Valley - Lockwood, NY0 	- jilly@clarityconnect.com			- Brett Bodine fan. 	- Mark.Jilson@Compaq.com			- since 1975 or so, 	- http://www.jilly.baka.com               -   ------------------------------  % Date: Tue, 06 Jun 2000 17:21:03 -0400i- From: JF Mezei <jfmezei.spamnot@videotron.ca>AJ Subject: Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire), Message-ID: <393D6B3D.C03E33CC@videotron.ca>   David Mathog wrote:wF > Right, but there's no way to avoid it.  As somebody else pointed outG > before (sorry, don't recall who, but I agree) a true Stream file type B > without the requirement for an LF every 64k would eliminate most > of these problemso    J SET file foo.bar/ATRRIB=(RFM=UDF,RAT=NONE) is as close to a stream file asD you'll get. It should not be converted/translated in any way by RMS.  L If there is a problem in that the C RTL, by default, creates a text file. IfM you intend to have binary unformatted data, then create the file accordingly.    ------------------------------  $ Date: Tue, 6 Jun 2000 20:47:01 -0400' From: "Bill Todd" <billtodd@foo.mv.com>iJ Subject: Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire)( Message-ID: <8hk5sg$29i$1@pyrite.mv.net>  5 Rob Young <young_r@eisner.decus.org> wrote in messagea% news:hDa+dY7YNfQk@eisner.decus.org...ML > In article <8hi52f$qbf@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edu (David Mathog) writes: >, > >nG > > OpenVMS still makes sense for huge applications, where IO has to bei tweakedzL > > out by hand or transaction integrity is required, or uptime requirementsL > > are severe.  But for general computing, not  only is OpenVMS overpriced, > > it's now also underpowered.s > > 9 > > Really, really, sorry to be able to say all that :-(.m > >  >  > David, >e? > I understand your frustration but no offense intended but youb@ > live in a very small world.  According to Shannon (if I recallB > correctly) , 90% of VMS nodes out there are in clusters.  Again,B > I may be misremembering that figure but gotta take a chance.  If? > so, we can expect that most of the storage those clusters are ? > accessing is/are via controllers (HSJ, HSZ , etc.. know wherehB > I'm going with this).  So turn on write back caching at the unit= > level and if you are worried about data integrity make sure : > you shadow across separate controllers in separate cabs.  K "If you are worried about data integrity"?  Well, if you have no redundancysL in your storage (no shadowing/mirroring, RAID-5, etc.), I guess you can justF choose to consider a write-back cache failure to be a disk failure andD restore from backups.  But if you think availability of your data isH sufficiently valuable to configure in any storage redundancy, you'd damn? well better not use write-back caching without making it stable   (battery-backed) *and* mirrored.  L So for installations of the calibre you're suggesting above, I'd suggest youK remove the phrase 'if you are worried about data integrity' and just assertkH that the extra (mirrored battery-backed memory) hardware overhead is theD price you pay for getting competitive performance out of VMS without massaging its defaults.   H Though not quite equal performance, since writes to the write-back cacheD (and reads from the controller cache, for that matter) take around aK millisecond (your own observation, IIRC) vs. well under 100 microseconds to-L the Unix system cache:  you'll cut your default-settings performance penaltyK from 10-to-1 or worse down to maybe around 2:1 (at least for writes; if you G can use VIOC for reads, they may be closer to equal), which changes the 1 perception from 'damn slow' to merely 'sluggish'.R  K Put as good a face as you can on it, it doesn't make VMS look great in thisT area.r   > B > Linux is great and Linux is fast, yada yada.   But Linux doesn't@ > have 4 or 5 nodes in a shared disk cluster.  Shared disk?  Who > needs shared disk....t  D Good question:  most installations don't, at all.  While shared-diskC configurations are the *only* way to address some (extreme) scalinggE situations well, those situations aren't that common.  Lacking such a K scaling problem, if you can successfully partition your data such that it'swG usually local to the node that needs it (the common case, especially in K read-mostly web-server environments), your performance will often be better)H than in a shared-disk cluster where you *haven't* paid attention to suchG partitioning, because in the latter case the nodes will be caching dataiJ redundantly (using up memory they could be using more effectively in otherL ways) and fighting each other over the occasional need to update it (and XFC? won't eliminate these problems, though it may ameliorate them).l   >  > And on and on we go...   Until you understand it.   - bill   >n > Yeee HAAAAAA!!!!!e >t > Robi >i   ------------------------------  % Date: Tue, 06 Jun 2000 21:04:34 -0500u* From: Keith Brown <kbrown780@usfamily.net>J Subject: Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire), Message-ID: <393DADB2.32E81C7A@usfamily.net>   Bill Todd wrote: >5 > M > "If you are worried about data integrity"?  Well, if you have no redundancyaN > in your storage (no shadowing/mirroring, RAID-5, etc.), I guess you can justH > choose to consider a write-back cache failure to be a disk failure andF > restore from backups.  But if you think availability of your data isJ > sufficiently valuable to configure in any storage redundancy, you'd damnA > well better not use write-back caching without making it stablet" > (battery-backed) *and* mirrored. >   ? Bill,  I would venture to say that most if not (near all) of usV8 with systems that support large external disk arrays use@ mirrored cache with battery backup and RAID arrays. As I've said2 before we do it on our DU as well as VMS systems.    > J > Though not quite equal performance, since writes to the write-back cacheF > (and reads from the controller cache, for that matter) take around aM > millisecond (your own observation, IIRC) vs. well under 100 microseconds to"N > the Unix system cache:  you'll cut your default-settings performance penaltyM > from 10-to-1 or worse down to maybe around 2:1 (at least for writes; if yourI > can use VIOC for reads, they may be closer to equal), which changes thes3 > perception from 'damn slow' to merely 'sluggish'.w >   ? We have installed a couple of VMS systems with external storage 5 in the last year (a DS20 and ES40) which run database @ applications (Oracle and a proprietary system).  "damn slow" and? "sluggish" are not words that have been used to describe them. a? I'm not sure where your impressions come from on this topic but > I do have to ask, have you used one of these systems lately? I< don't mean this as a slam, but sometimes perceptions are not= correct.  As I said earlier, we run Oracle on DU systems alsou? and we don't see the 10-to-1 difference in performance to referG: to. Frankly, performance is very similar on the DU and VMS	 machines.r     --   Keith Browne kbrown780@usfamily.net   ------------------------------  $ Date: Wed, 7 Jun 2000 00:33:16 -0400' From: "Bill Todd" <billtodd@foo.mv.com>hJ Subject: Re: OpenVMS commentaries (was Re: Gartner commentary on Wildfire)( Message-ID: <8hkj4q$n50$1@pyrite.mv.net>  5 Keith Brown <kbrown780@usfamily.net> wrote in message & news:393DADB2.32E81C7A@usfamily.net... > Bill Todd wrote: > >  > >,D > > "If you are worried about data integrity"?  Well, if you have no
 redundancyK > > in your storage (no shadowing/mirroring, RAID-5, etc.), I guess you cana justJ > > choose to consider a write-back cache failure to be a disk failure andH > > restore from backups.  But if you think availability of your data isL > > sufficiently valuable to configure in any storage redundancy, you'd damnC > > well better not use write-back caching without making it stable $ > > (battery-backed) *and* mirrored. > >r >tA > Bill,  I would venture to say that most if not (near all) of use: > with systems that support large external disk arrays useB > mirrored cache with battery backup and RAID arrays. As I've said3 > before we do it on our DU as well as VMS systems.   G My comment was based on Ron's implication that mirroring the write-backiD cache could be considered optional in the kind of environment he wasC describing.  As I said and you suggest, it effectively isn't, whichWI effectively drives up the cost of using VMS by the additional cost of the-L HSx controllers and their stable, mirrored caches compared with much simplerE controllers and no on-board cache that can achieve the same or bettera= default performance for applications in the Unix environment.    >n > >lL > > Though not quite equal performance, since writes to the write-back cacheH > > (and reads from the controller cache, for that matter) take around aL > > millisecond (your own observation, IIRC) vs. well under 100 microseconds toH > > the Unix system cache:  you'll cut your default-settings performance penalty-K > > from 10-to-1 or worse down to maybe around 2:1 (at least for writes; ifn yougK > > can use VIOC for reads, they may be closer to equal), which changes then5 > > perception from 'damn slow' to merely 'sluggish'.  > >N >CA > We have installed a couple of VMS systems with external storage 7 > in the last year (a DS20 and ES40) which run database.B > applications (Oracle and a proprietary system).  "damn slow" and@ > "sluggish" are not words that have been used to describe them.  B Are VMS bigots entirely incapable of following the details of this discussion?o  J Parse the paragraph to which you responded again.  Look especially for theI phrase 'default-settings'.  *That's* what this entire discussion has been?L about:  the fact that applications using default file system settings in theB VMS environment are at such a performance disadvantage compared toH applications using default file system settings in the Unix environment.  D Add hardware write-back cache to VMS, and (as I said) its *relative*K performance in this respect improves from 'damn slow' to 'merely sluggish': K this is not to say that it's 'sluggish' in an absolute sense, just relativetH to its Unix competition (without the hardware assist but under otherwiseF equivalent conditions).  (And my sketchy analysis above overlooked the> advantage that Unix's automatic access-pattern recognition andF multi-buffered read ahead confers even when VMS uses cache hardware toK compensate for its default algorithmic inefficiencies, so VIOC may not makea7 VMS read performance as close to equal as I suggested.)i  A > I'm not sure where your impressions come from on this topic but @ > I do have to ask, have you used one of these systems lately? I> > don't mean this as a slam, but sometimes perceptions are not? > correct.  As I said earlier, we run Oracle on DU systems also-A > and we don't see the 10-to-1 difference in performance to refera< > to. Frankly, performance is very similar on the DU and VMS > machines.0  H Of course it is - not that that has anything to do with the matter underE discussion.  Where Oracle uses the respective file systems at all, itaI doesn't use anything like default settings and mechanisms to access them.fF Expect performance to be about the same in both envrironments, becauseJ Oracle uses as little of the file system as possible in both environments.F Unlike the applications we're talking about, which are at the opposite extreme of (non-)optimization.   - bill   >  >l > --
 > Keith Browni > kbrown780@usfamily.net   ------------------------------   Date: 6 Jun 2000 22:25:21 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)Q Subject: Re: OpenVMS, SMTP, and email spam (was Re: I'm Trying to Model Something 6 Message-ID: <8hjtoh$qve$1@mailint03.im.hou.compaq.com>  Q In article <8hjq10$l13$1@husk.cso.niu.edu>, system@niuhep.physics.niu.edu writes:2K :You need to take into account P, the total number of possible relay sites,n& :i.e the number of sites running smtp. :.A :When M/P is close to one, it is effortless to find a relay site.sJ :If M/P is close to 1/1000 then life becomes significantly more difficult, :even though M>>N.    H   This clearly assumes that a member of set N (the spammers) would have G   particular difficulty in creating or acquiring T (a tool) to detect M J   (open SMTP relays), and USO (unsubstantiated personal opinion) indicatesI   this is likely BA (bogus assumption).  For one corollary, consider the  J   various WD (war dialers) that are available.  Resolving for M is likely C   little more difficult than finding modems, given the TI (typical oF   inclination) and HxN (history of N), and particularly given M/t-subVJ   (rate of potential new "victim" M arriving over unit time t) vs M/t-subCI   (rate of potential "closures" to M arriving over unit time t).  So long H   as M/t-subV exceeds M/t-subC, the efforts of N will likely continue to   be successful.   	---  H   We could also have this very same discussion using only English -- andJ   not something approaching "geekish" -- of course, but that's not nearly    as much fun. :-)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Tue, 06 Jun 2000 18:16:32 GMT-= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) * Subject: Re: Porting VAX Macro code to AXP0 Message-ID: <009EB34A.4765DAE6@SendSpamHere.ORG>  Z In article <393D1A5A.B88F8826@tsoft-inc.com>, David A Froble <davef@tsoft-inc.com> writes: >Ron Maisch wrote: >>{...snip...} pO >The above is just the idea, you may need indirect addressing.  The key is thatt6 >'(AP)+' won't work in the compiled Alpha environment.  K The (AP)+ will work in the Alpha environment.  However, you must understandhL what the compiler will do when it sees this construct.  Short and simple, itL will not reference the arguments nor the argument list if it was homed.  In-L stead, the appearance of (AP)+ will be flagged as a modification to the arg-K ument pointer and the compiler will substitute register R12 and then do theeL equivalent of the reference and post-increment.  So, how do you maintain theK (AP)+ in your code?  Simple!  Move the address of the 'homed' argument listlJ to register R12.  I developed two macros (.START_AP_JACKET/.END_AP_JACKET)H which encapsulates the code using the (AP)+ construct (or any code whichJ attempts to modify the value of AP) within a .JSB_ENTRY.  The macro moves L the address of (AP) to R12 and then invokes the .JSB_ENTRY encapsulated codeL within the limits of the two macros.  This is very similar to the $LOCK_PAGEL and $UNLOCK_PAGE macros used to delimit and lock down code to be executed at
 elevated IPL.   P >I believe that Brian may have some tools to allow a different solution for this >problem.  (VAXMAN@TEMSIS.COM)$                    ^^^^^^--- TMESIS.   ;)   --N VAXman- OpenVMS APE certification number: AAA-0001           VAXman@TMESIS.COM  L GNU Freeware -- What does the GNU *really* stand for?  Garbage!  Not Usable!   ------------------------------   Date: 6 Jun 2000 20:22:24 +0200 * From: eplan@kapsch.net (Peter LANGSTOEGER)< Subject: Re: POS/credit card verification with OpenVMS Alpha* Message-ID: <393d4160$1@news.kapsch.co.at>  E In article <8hgghn$4hq$1@nnrp1.deja.com>, itjck01@my-deja.com writes:o@ >In article <ejTZ4.17117$XX4.273703@news-east.usenetserver.com>,/ >  "Ruslan R. Laishev" <laishev@dls.net> wrote:F@ >>     This product run under VMS ? Can you provide any URL(s) ? >  >www.paylinx.com   Platform requirements:   Microsoft Windows NT Server  Pentium 133MHz/Digital Alpha 2GB Hard Drive ...t  L Did you understand the question above (and in the first posting in thread) ?   -- t< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888l< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  # Date: Tue, 06 Jun 2000 20:47:03 GMTR From: itjck01@my-deja.com < Subject: Re: POS/credit card verification with OpenVMS Alpha) Message-ID: <8hjo01$jr2$1@nnrp1.deja.com>:  * In article <393d4160$1@news.kapsch.co.at>,   eplan@kapsch.net wrote: G > In article <8hgghn$4hq$1@nnrp1.deja.com>, itjck01@my-deja.com writes:iB > >In article <ejTZ4.17117$XX4.273703@news-east.usenetserver.com>,1 > >  "Ruslan R. Laishev" <laishev@dls.net> wrote:CB > >>     This product run under VMS ? Can you provide any URL(s) ? > >o > >www.paylinx.com >e > Platform requirements: >, > Microsoft Windows NT Server  > Pentium 133MHz/Digital Alpha > 2GB Hard Drive > ...n > D > Did you understand the question above (and in the first posting in > thread) ?R  G I believe PAYLINX new 3.1 version (EnterpriZ) will run on OpenVMS Alpha E in addition to the above platform, or so I have been told.  We are int< the process of getting a tape/CD of their software so we can! demonstrate that it indeed works.h  > Our contact at Paylinx is Philip Panto, a district manager forB Paylinx.  I don''t have his phone number handy.  But I am sure the) folks at Paylinx can help in that regard.t   :) jck   --$ Free personal opinion is what I post    & Sent via Deja.com http://www.deja.com/ Before you buy.e   ------------------------------  # Date: Tue, 06 Jun 2000 17:43:48 GMT ( From: sol gongola <sol@mail.adldata.com>0 Subject: SMTP_RECV_MAIL from TCPIP$smtp_receiver) Message-ID: <8hjd8g$asd$1@nnrp1.deja.com>e  C I sometimes see this message in an smtp log file. It doesn't follown> the standard vms message format and i did not see it mentioned in the tcpip manuals:r  + SMTP_RECV_MAIL: remote host not in hosts DBi    & Sent via Deja.com http://www.deja.com/ Before you buy.l   ------------------------------   Date: 6 Jun 2000 20:21:35 GMT , From: bill@cs.scranton.edu (Bill Gunshannon) Subject: The DECUS Archives , Message-ID: <8hjmgg$1cb4$1@info.cs.uofs.edu>  ; It's nice that the DECUS Archives are now on a VMS machine.t; Now, if someone would just change all the "/pub" references 9 on the web page to "[ANONYMOUS]" we'll be in business!!  o   :-)o   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   i   ------------------------------  # Date: Wed, 07 Jun 2000 03:12:24 GMTy From: Dan <dan@vrx.net>n% Subject: Re: vaxstation LED meanings?R& Message-ID: <393DBD14.795517B@vrx.net>   nope. she's dead.e' tried the scsi termination idea. no go.a same problem. very sad.    Dan.   Antonio Carlini wrote:  B > In article <393CF482.76DA7F74@vrx.net>, Dan <dan@vrx.net> wrote: > >how about 7,2,1 ?- > >I wish someone would post the whole thing.n >t8 > The maintenance guide gives a whole bunch of stuff but >         1000.xxxxr > means "TEST xxxx failed" sos >         1000.0110eK > means test 6 failed which means one (or both?) of your SCSI busses has ane/ > error - are both of them properly terminated?e >a > xxxx codes are:  >         1111 - MONOt >         1110 - CLK >         1101 - NVR >         1100 - DZh >         1011 - MEM >         1001 - FPt >         1000 - ITc$ >         0111, 0110 - SCSI (A or B) >         0101 - SYS >         0100 - GFX >         0001 - Ethernet,M > (0011 & 0010 seem to either never be used or are reserved for option cards)y > O > >I pulled the scsi/fdi adapter out, unplugged all the drives and scsi cables,2E > >even removed half the ram (it was fully loaded) and still no dice.l > >same thing. >c. > No SCSI terminator will give you this error.! > Put the RAM back - it looks OK!  >eL > >does video test patterns (I'm using it as a console not via seral network! > > connection), the S3 switch isa5 > >down. but no display at all after the video tests.a > P > You should begin to see the KA43 banner - but the tests do take a *long* time;O > multiple minutes. It is very easy to decide that anything taking this long toaP > say hello is broken. It isn't broken. Try leaving it for ten minutes (with theI > monitor brightness and contrast set so that the monitor will definitelyi > display something :-) )w >sN > >according to a web page I was at 3 and 2 are both "reserved digital tests", > > contact digital. but I'd >r > You have a support contract? > A > >Maybe save me from a $100US cost of a replacement motherboard.  >h > Cheaper on ebay no?  >t > t > a" > drives but you cannot leave out > external). >g	 > Antoniot > K > Antonio Carlini                            Mail: carlini@true.lkg.dec.comt% > DECnet-Plus for OpenVMS EngineeringC8 > COMPAQ                                     Reading, UK   ------------------------------   Date: 7 Jun 2000 00:20:00 +0200 * From: eplan@kapsch.net (Peter LANGSTOEGER)A Subject: Re: What is the simplest way to network two VMS systems?0* Message-ID: <393d7910$1@news.kapsch.co.at>  m In article <01bfcfd8$60718080$4b53b083@ptregoni.esoc.esa.de>, "Phil Tregoning" <ptregoni@esoc.esa.de> writes:B9 >JF Mezei <jfmezei.spamnot@videotron.ca> wrote in article $ ><393B366F.9F097B69@videotron.ca>...
 >> Dan wrote:h >> > rI >> > I installed DEC Net Plus instead of phase 4, compaq likes to suggests >this." >> > is this actually a good idea? >>  K >> DECNET-PLUS is overly complex and a pain to manage. Avoid it unless yourh >> network requires it.d > L >Lots of people seem to say this in comp.os.vms. I've been using DECnet/plus< >for a few years now, and quite like it. Am I just perverse?  J Nope. I use DECnet Phase 5 since the DECnet VAX Extensions V5.4 and I findF it very good. I can't remember the last crash I got because of DECnet5K (sure, there have been some), must be more than 5 years ago. The last yearssF my only VMS crashes were cause by PATHWORKS (>90%) and bad VMS patches (eg. F11X).i   For me,r  B 1) DECdns was an improvement, because the DECnet node database got7 centralized (instead of node-specific binary files) ands  M 2) DECdts was an improvement, because the VMS systems got timesynchronizationx? with Daylight Saving Time support and VMS got timezone support.l  D 3) DECnet5 was an improvement, because you got a GAP client for freeD (so all my cluster nodes became able to access X.25 and not only oneD node we used to have an **REALLY EXPENSIVE** X.25 license then) and   N 4) you got DECnet over IP which is COPY/DELETE/DIR/PURGE/RENAME instead of FTP? use of F$SEARCH, F$FILE_ATTRIBUTES over an IP only backbone andc  J 5) NCL was an improvement, because you got a well structured language toolJ to also manage your other CMIP systems like the DECnis (which then offeredF the GAP server and the X.25 link moved from an VAX 8600 to the DECnis)    I Only the NET$CONFIGURE.COM is a little bit clumsy (eg. if you upgrade the I DECnet-Plus, you loose DECnet application definitions from the NCL files,tJ and the supported way is to reinstall [!!] all layered products which makeD changes to DECnet application definitions. So, I'm used to change myI NCL config startup files with an editor and COPY them from node to node).7  I And DECdns is not as easy as most of us want/need to work with (just likeRL the successor: OSF/DCE Cell Directory Server) but is a really powerful tool,K if you need it. But for a DECnet only config, keep with LOCAL and DOMAIN !!   F Ph5 _is_ different to Ph4 and I can understand why people hate changesJ (I'm now old enough to understand it), but it is worth to change (at quiet0 times - are there any in VMSland ? - of course).    F To make it short, it was not a bad decision to got to Ph5. It may lostL some/most of it advantages over the years, but it is still better than Ph4 !   --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888b< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  # Date: Tue, 06 Jun 2000 20:13:20 GMTy) From: Mr. Poologic <poologic@my-deja.com>t" Subject: Where is javadoc for VMS?) Message-ID: <8hjm0f$i2l$1@nnrp1.deja.com>n  9 javadoc does not available on my VMS system 7.2-1 system.e  5 Why is it not there? Where can I find a VMS javadoc?o   --- http://members.aol.com/tadamsmar/poologic.htmo+ Learn how to win your college hoops tourneyr office pools    & Sent via Deja.com http://www.deja.com/ Before you buy.e   ------------------------------   Date: 7 Jun 2000 00:27:23 +0200v* From: eplan@kapsch.net (Peter LANGSTOEGER)& Subject: Re: Where is javadoc for VMS?* Message-ID: <393d7acb$1@news.kapsch.co.at>  U In article <8hjm0f$i2l$1@nnrp1.deja.com>, Mr. Poologic <poologic@my-deja.com> writes:R: >javadoc does not available on my VMS system 7.2-1 system.   What is JAVADOC ?u  5 >Why is it not there? Where can I find a VMS javadoc?1  O Is it part of JAVA ? Did you install JAVA (means JDK or JRE, but JDK is better)CF on your OpenVMS Alpha system ? Did you start JAVA during VMS startup ?   Is JAVA/DOC what you want ?    -- s< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888i< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  # Date: Tue, 06 Jun 2000 21:50:35 GMTe2 From: "Doug Stephenson" <Doug.Stephenson@sbpa.com>Y Subject: Windows NT hosted Uiltity that can read RMS index fles or VMS Backup formatted ta$ Message-ID: <Lme%4.60$vh.19201@news>  H We are attempting to move at 10 G of data in RMS indexed data files on aL nightly basis from an Alpha to an SQL database on NT (Win2K).  One method isF to convert the files to sequential, move them and then load them.  The; problem is that we do not have the time on a nightly basis.tK If we had a utility that could read VMS Backup and also understand RMS dataeF structures we could perform the conversion on the NT machine.  The VMSD machine is for production and must be available as much as possible.  H We have considered ODBC, but it takes too much time.  Using FTP requiresL that the files not be open by any VMS application (and they still need to be( converted from Indexed for SQL to load).   Anyone know of any solutions?u   ------------------------------  $ Date: Tue, 6 Jun 2000 16:41:15 -0700- From: "Dann Corbit" <dcorbit@solutionsiq.com>nY Subject: Re: Windows NT hosted Uiltity that can read RMS index fles or VMS Backup formattr& Message-ID: <A_f%4.391$tV4.388@client>  = "Doug Stephenson" <Doug.Stephenson@sbpa.com> wrote in messagel news:Lme%4.60$vh.19201@news...J > We are attempting to move at 10 G of data in RMS indexed data files on aK > nightly basis from an Alpha to an SQL database on NT (Win2K).  One method  isH > to convert the files to sequential, move them and then load them.  The= > problem is that we do not have the time on a nightly basis. H > If we had a utility that could read VMS Backup and also understand RMS dataH > structures we could perform the conversion on the NT machine.  The VMSF > machine is for production and must be available as much as possible. >lJ > We have considered ODBC, but it takes too much time.  Using FTP requiresK > that the files not be open by any VMS application (and they still need to  be* > converted from Indexed for SQL to load). >i > Anyone know of any solutions?e  I Why not link the tables and leave the data where it is?   SQL performancetK against RMS files will generally whup the tar out of SQL Server if it is on F a high-end alpha.  Since you are talking about ten gigs, I am guessing! (hoping) that's where it resides.i  L If you must transfer the data, why do it at night?  Have (instead) a processL that transfers the data continuously.  It could run 24x7x365.25, just moving records from the VAX to the PC.eL If you should be so fortunate as to have a timestamp in the data, that couldJ simplify the process enormously, since you would only have to transfer the updates.  I Is it possibly to modify the structure of the records to add a timestamp?aI Naturally, it would also require an update to the legacy code that writestD the file, but it would certainly reduce the I/O problem tremendouslyH (Unless, of course, you have already done as I have suggested and the 10 gigs is the delta!). ;-)l   Just some ideas. --0 C-FAQ: http://www.eskimo.com/~scs/C-faq/top.html$  "The C-FAQ Book" ISBN 0-201-84519-91 C.A.P. Newsgroup   http://www.dejanews.com/~c_a_plI C.A.P. FAQ: ftp://38.168.214.175/pub/Chess%20Analysis%20Project%20FAQ.htm    ------------------------------  $ Date: Wed, 7 Jun 2000 10:46:17 +1000/ From: "Phil Howell" <howellp@snowyhydro.com.au>pY Subject: Re: Windows NT hosted Uiltity that can read RMS index fles or VMS Backup formatt-2 Message-ID: <wXg%4.10364$N4.377210@ozemail.com.au>  1 Have a look at the connx product from solutionsiqnC ODBC access using a dictionary can be slow but it also has a remote-? procedure call interface and rms like client routines which areH9 significantly faster (10Gb is still a lot of data though)iL However, if you have an odbc server on the vms end and your rms files have aH reasonable "table-like" structure then you can attach them from your sqlK server database and they will appear to be "local", read-only access to theeA rms data should not interfere with any processing at the vms end.' Phil; Doug Stephenson <Doug.Stephenson@sbpa.com> wrote in messager news:Lme%4.60$vh.19201@news...J > We are attempting to move at 10 G of data in RMS indexed data files on aK > nightly basis from an Alpha to an SQL database on NT (Win2K).  One methodn isH > to convert the files to sequential, move them and then load them.  The= > problem is that we do not have the time on a nightly basis.lH > If we had a utility that could read VMS Backup and also understand RMS dataH > structures we could perform the conversion on the NT machine.  The VMSF > machine is for production and must be available as much as possible. > J > We have considered ODBC, but it takes too much time.  Using FTP requiresK > that the files not be open by any VMS application (and they still need toP be* > converted from Indexed for SQL to load). >s > Anyone know of any solutions?  >  >B   ------------------------------  # Date: Wed, 07 Jun 2000 03:29:28 GMTe# From: moi_is_me <moi_is_me@usa.net>L Subject: X25 and psi$k_reset) Message-ID: <8hkfi8$58g$1@nnrp1.deja.com>v  
 Greetings,  A    Please excuse my ignorance, but i have no manuels to hand, ando#    have not worked with X25 before..  =    I have been asked to port some X25 code to TCP. ThroughoutpC    the offending code, are calls to qiow to "reset the circuit".ie. C    psi$k_reset (p4) The QIOW call itself, is wrapped within a timert  "    Q ... Was does psi$k_reset do ?  C    Q ... IF there is any data (in or out) currently within a system 8          buffer, can psi$k_reset cause any loss of data.  6    Q ... Is there an equivalent Socket function call ?   TIAn Fred    & Sent via Deja.com http://www.deja.com/ Before you buy.o   ------------------------------   End of INFO-VAX 2000.316 ************************