1 INFO-VAX	Tue, 03 Apr 2001	Volume 2001 : Issue 186       Contents: Re: DCPS Newer than v1.3: EV7 and EV8 for the low end (was: VMS-Related: Affordable)> Re: EV7 and EV8 for the low end (was: VMS-Related: Affordable)> Re: EV7 and EV8 for the low end (was: VMS-Related: Affordable)> Re: EV7 and EV8 for the low end (was: VMS-Related: Affordable) RE: Free The OpenVMS Seven!  Re: Free The OpenVMS Seven!  Re: Free The OpenVMS Seven!  RE: Free The OpenVMS Seven!  RE: Free The OpenVMS Seven! / Re: Getting Apache/CSWS & Multinet to co-exist? $ GS160 froze due to a bad KGPSA-CX... HELP BACKING UP Oracle on VMS H Re: Hoff's Road Trips (was: Re: OpenVMS Technical Update Days - success!H Re: Hoff's Road Trips (was: Re: OpenVMS Technical Update Days - success! Hopf 6841 dtss time provider2 How can I tell if a boot is the result of a crash?= How to tell if a system is booting immediately after a crash. A Re: How to tell if a system is booting immediately after a crash. ' HP DAT drive on a OpenVMS -- Alpha box? + Re: HP DAT drive on a OpenVMS -- Alpha box?  Re: Is SUN crazy ???. Java SDK and StarTeam (this is a long shot...) Merging DCLTABLES.EXE  Re: Merging DCLTABLES.EXE  Re: Merging DCLTABLES.EXE  Re: Merging DCLTABLES.EXE  Re: Merging DCLTABLES.EXE  Re: Merging DCLTABLES.EXE  Re: Merging DCLTABLES.EXE 
 Re: modems Re: OpenVMS Educational Program  OT, but worth a look# Re: POSIX Streams, File Permissions 8 Re: Queue Manager gives me this error about once a month( Re: Seeking CD-R/CD-RW SCSI INQUIRY data( Re: Seeking CD-R/CD-RW SCSI INQUIRY data: Re: Seeking CD-R/CD-RW SCSI INQUIRY data / Yamaha CRW4416S: Re: Seeking CD-R/CD-RW SCSI INQUIRY data / Yamaha CRW4416S Re: Spice for OpenVMS Alpha? Re: TCP/IP 5.0A ECO Appears 
 Re: VAX in 1U 
 Re: VAX in 1U 
 Re: VAX in 1U 
 Re: VAX in 1U 
 Re: VAX in 1U  Re: VMS-Related: Affordable ) Re: When might SYS$SCHDWK not wake me up? " Re: [Q] Need Help on Backup Script" Re: [Q] Need Help on Backup Script) Re: [SOAPBOX] sys$specific or sys$common?   F ----------------------------------------------------------------------  % Date: Mon, 02 Apr 2001 17:28:09 -0600 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> ! Subject: Re: DCPS Newer than v1.3 ' Message-ID: <3AC90B09.5BD48EE8@fsi.net>    "John R. Lawson, Jr." wrote: > F > Well, you see ... I'm using the OpenVMS Hobbyist license, which mostM > assuredly entitles me to use the product.  The trouble is that the Hobbyist N > program never bothered to create media for the "over 100 non-revenue layered9 > products" that the Hobbyist program permits you to use.  > F > Any suggestions short of buying the ConDist (which I cannot afford)?  H I had asked the DFW guys about that, getting onto a year ago. Back then,G the scoop was that such was in the works. I was gonna just go ahead and H compile something based on the (then current) condist; however, that wasF about the time the (*AHEM*) hit the fan re: my professional situation. So, it just never happened.   B Once I get settled into my new digs later this season, I must just> undertake the project based on the premise that forgiveness is) frequently easier to get than permission.   H Meantime, e-mail me privately so I have your address and I'll see what IA can find that I can ZIP up and e-mail to you. My e-mail is in the F reply-to. How-to demangle it should be obvious, but if not, post again ...    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------   Date: 2 Apr 2001 21:24:03 -0500 9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) C Subject: EV7 and EV8 for the low end (was: VMS-Related: Affordable) 3 Message-ID: <BaOv4qwnkpty@eisner.encompasserve.org>   \ In article <87hf07w7hx.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:  F > Sad thing is, the EV7 and EV8 open up the chance for a VERY low costG > alpha system with more than enough performance to run tools and test.   ? I thought EV7 and EV8 were supposed to shine on large machines. = To me that means money wasted on small machines that will not  see the benefit.   ------------------------------  # Date: Tue, 03 Apr 2001 00:51:18 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>G Subject: Re: EV7 and EV8 for the low end (was: VMS-Related: Affordable) < Message-ID: <a89y6.28064$Wz.8257185@typhoon.ne.mediaone.net>  F "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message- news:BaOv4qwnkpty@eisner.encompasserve.org... > > In article <87hf07w7hx.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:  > H > > Sad thing is, the EV7 and EV8 open up the chance for a VERY low costI > > alpha system with more than enough performance to run tools and test.  > A > I thought EV7 and EV8 were supposed to shine on large machines. ? > To me that means money wasted on small machines that will not  > see the benefit.  J It may be cheaper to build a uni or dual EV7/EV8 box than is possible withI EV6n and predecessors due to the on-chip switching infrastructure (an EV7 K system could consist of the EV7 module, an I/O chip, and RAMBUS memory) but I the focus for these future chips is in fact large and Marvel-ous systems.    ------------------------------  % Date: Mon, 02 Apr 2001 17:52:22 -0700 ! From: Shane.F.Smith@Healthnet.com G Subject: Re: EV7 and EV8 for the low end (was: VMS-Related: Affordable) D Message-ID: <OF5D989FCB.CD9D6D0B-ON88256A23.0002D7FF@foundation.com>  K Looking at the articles on realworldtech, it seems to me the EV8 will shine J when there are multiple program threads to execute. Hopefully someone willI correct me if I understand wrong, or am remembering it wrong, but IIAR it K can handle simultaneous instruction building blocks "in flight" through the B chip from several different threads. If one thread is blocked, itsH instructions don't have to clear the pipe before another thread can takeE over. This allows cycles to be used that would otherwise be idle, and  increases overall throughput.   I If you've got only one thread running, but it doesn't block (duck, flying G pig!) it will fully utilize the processor. Two threads that don't block D will get timeslices just as they would on older chips. If you've gotD multiple threads that do block, the overall job will complete fasterI because the processor doesn't waste as much time waiting for instructions ? to be retired before going to the next available piece of work.   I A workstation would benefit from this when running multiple processes, or J multithreaded code. Hmmmm, Quake 3 is multithread enabled.... ;-) However,J I've read somewhere the chip price itself will way exceed anything but theJ highest end workstation prices, so it probably won't make it into low cost2 systems until EV9 or EV10 hit the streets. Sob....   Shane           J Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) on 04/02/2001 07:24:03 PM  E Please respond to Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)    To:   Info-VAX@Mvb.Saic.Com  cc:   D Subject:  EV7 and EV8 for the low end (was: VMS-Related: Affordable)    < In article <87hf07w7hx.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:   F > Sad thing is, the EV7 and EV8 open up the chance for a VERY low costG > alpha system with more than enough performance to run tools and test.   ? I thought EV7 and EV8 were supposed to shine on large machines. = To me that means money wasted on small machines that will not  see the benefit.   ------------------------------   Date: 3 Apr 2001 01:14:32 -0500 + From: young_r@encompasserve.org (Rob Young) G Subject: Re: EV7 and EV8 for the low end (was: VMS-Related: Affordable) 3 Message-ID: <mU8ll0f7gk2W@eisner.encompasserve.org>   h In article <OF5D989FCB.CD9D6D0B-ON88256A23.0002D7FF@foundation.com>, Shane.F.Smith@Healthnet.com writes: > M > Looking at the articles on realworldtech, it seems to me the EV8 will shine L > when there are multiple program threads to execute. Hopefully someone willK > correct me if I understand wrong, or am remembering it wrong, but IIAR it M > can handle simultaneous instruction building blocks "in flight" through the D > chip from several different threads. If one thread is blocked, itsJ > instructions don't have to clear the pipe before another thread can takeG > over. This allows cycles to be used that would otherwise be idle, and  > increases overall throughput.  >   A 	At running the risk of getting this slightly wrong.  It is a bit ? 	over and above this.  EV8 is 8-way superscalar.  To keep those A 	ALUs, busy EV8 adds SMT capability.  As described earlier:  "The E 	CPU appears as 4 CPUs and delivers the power of 2 CPUs" in realworld = 	apps.  EV8 will actually have or allow four program counters F 	(PC) , is 4-context theaded.  As for blocking and non-blocking , I'll? 	take a pass on that.  But regarding your speculation regarding  	Quake, etc.  The mantra is:  $ 			"It's the business apps, stupid!"  A 	With "stupid" directed to no one, a takeoff on "it's the economy ? 	stupid!" or better yet:  "it's the bandwidth stupid!"  Because + 	there is money in them thar business apps.   = 	This paper details the problem lies and what value SMT adds:   7 http://www.cs.washington.edu/research/smt/papers/os.pdf   & 	Looking at the conclusion on page 11:  ; 		5. Overall, operating system code causes poor instruction 7 		throughput on a superscalar.  This has a large impact ? 		for the Apache Web server, which achieves an IPC of only 1.1.  		< 		6.  SMT's latency tolerance is able to compensate for many: 		of the demands of operating system code.  When executing> 		Apache, SMT achieves a 4-fold improvement in throughput over8 		the superscalar, the highest relative gain of any SMT  		workload to date.    	Likewise, in this conclusion:  @ http://www.cs.washington.edu/research/smt/papers/smtdatabase.pdf  P "The conclusions from our simulation study are three-fold. First, while databaseM workloads have large footprints, there is substantial reuse that results in a M small, cacheable "critical" working set, if cache conflicts can be minimized. G Second, while cache conflicts are high, particularly in the per-process J private data region of the database, they  often result from poorly-suitedN mapping policies in the operating system. By selecting the appropriate mappingM policy, and using application-based address offsetting, conflict misses on an M 8-context SMT can be reduced to the point where they are roughly commensurate M with miss rates for a superscalar. Third, once cache conflicts are reduced to K the superscalar level, an SMT processor can achieve substantial performance M gains over a superscalar with similar resources. We show a 3-fold improvement K in utilization (IPC) for the OLTP workload and a 1.5-fold increase for DSS. K These performance gains occur because of SMTs ability to tolerate latencies K through fine-grained sharing of all processor resources among the executing N threads; this is demonstrated by our detailed measurements indicating that SMTJ achieves a striking improvement in memory-system utilization and fetch andH issue efficiency, when compared with the single-threaded superscalar. WeK believe that these results indicate that SMTs latency tolerance makes it an - excellent architecture for database servers."    ---   < 	All this to say SMT will keep those units busy and poor IPC6 	will benefit greatly. i.e. databases and Web servers.   				Rob    ------------------------------   Date: 2 Apr 2001 18:50:48 -0000 4 From: Doc.Cypher <Use-Author-Address-Header@[127.1]>$ Subject: RE: Free The OpenVMS Seven!6 Message-ID: <20010402185048.21166.qmail@nym.alias.net>  " -----BEGIN PGP SIGNED MESSAGE-----  A On Mon, 02 Apr 2001, Christopher Smith <csmith@amdocs.com> wrote:  >> -----Original Message----- 3 >> From: Christof Brass [mailto:brass@infopuls.com]  > > >> polemic. OO is/was a major step forward. If you don't fullyC >> understand the concept don't blame it. What makes you think that A >> you know "how computers actually operate"? Which computers? In ? >> what way is any other high level PL closer to "how computers A >> actually operate"? "However, C as a language has its merits.":  > J >I think that I understand what Doc was talking about here.  Let me try toM >explain.  I don't believe that he was talking about the whole OO (read that: G >"modular programming") concept as being against the way that computers G >operate, but rather I think he means that OO, as it's included in most M >modern languages introduces an amount of artificial complexity that may make 1 >the program's operating seem to be non-linear.    > E >You might interpret his argument as being for a simpler method of OO  >programming.    > I >I assume he's not against re-using his code once in a while, and writing $ >apps that are relatively "modular." > 	 >Regards,  >  >Chris   Well said Chris.  E I do prefer functional programming and state machine models to modern K objectification. The obsession with abstracting to miles away from hardware J seems to result in unreliable builds, and objects which are way beyond the size required.  E I suppose my dislike of OO stems from my assembler background. (Where # modularity and reuse are critical).      Doc.   -----BEGIN PGP SIGNATURE-----  Version: 2.6.2  @ iQEVAwUBOsey8sriC3SGiziTAQG3fAf/ZD9FcbEdO1SLAyTNST59tmBvy5S1NdOK@ brO6RW2Ppm7+/RnHVSP50Vx9gu02hjw6xOve0rJIUdtTzXGYZI5etYA5xf9iqRiQ@ GKD7GgwkexENVD6fmPIIwcQuDvzysMscBLGHn4ZZ00KuCQ+JjrCUfQM7DDRQbvnB@ YVM9ciH9qGXOzyM3k5h7n/yucMp8RKREqzSuxDvQO/vME/8RCMqpW4kfDOvRgaPN@ y1YYDL7ucruPa5ADa5SV5mPHYo/cPP6m0gn5voTNwEIovW/fIz7CiQCSFZFD3NtP8 VKd0QPDuwEjlJxu1v52Uv6bLkxqPlncwKhQHsjMnNpli1nLTVJwuLg== =OsKm  -----END PGP SIGNATURE-----    ------------------------------  % Date: Mon, 02 Apr 2001 23:54:26 +0000 ) From: Christof Brass <brass@infopuls.com> $ Subject: Re: Free The OpenVMS Seven!, Message-ID: <3AC91132.6D21296A@infopuls.com>   Christopher Smith wrote: >  > > -----Original Message-----4 > > From: Christof Brass [mailto:brass@infopuls.com] > : > > Using C today is like building cars as did Henry Ford. > J > Here I go again.  It occurs to me that this might not support your pointF > well.  After all the bodies of the cars built by Henry ford are muchL > stronger and more resilient than cars of today.  The internal workings areK > so simple as to actually allow one person to handle maintenance, _and_ in 7 > case of EMP, they'd still be perfectly functional. :)  > 
 > Regards, >  > Chris  > # > Christopher Smith, Perl Developer  > Amdocs - Champaign, IL >  > /usr/bin/perl -e 'A > print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");  > '   > I have to agree that these old cars had a lot of good features: compared to the new ones. But I still don't think that the; technique used to build them is competitive to the current. ; Image if those simple cars would been built with the modern @ production systems. OTOH it would be almost impossible to create6 these modern cars with the old production systems at a, reasonable cost and with reasonable quality.   ------------------------------  % Date: Tue, 03 Apr 2001 00:19:27 +0000 ) From: Christof Brass <brass@infopuls.com> $ Subject: Re: Free The OpenVMS Seven!, Message-ID: <3AC9170F.2B7AEAFF@infopuls.com>   "Doc.Cypher" wrote:  > $ > -----BEGIN PGP SIGNED MESSAGE----- > C > On Mon, 02 Apr 2001, Christopher Smith <csmith@amdocs.com> wrote:  > >> -----Original Message----- 5 > >> From: Christof Brass [mailto:brass@infopuls.com]  > > @ > >> polemic. OO is/was a major step forward. If you don't fullyE > >> understand the concept don't blame it. What makes you think that C > >> you know "how computers actually operate"? Which computers? In A > >> what way is any other high level PL closer to "how computers C > >> actually operate"? "However, C as a language has its merits.":  > > L > >I think that I understand what Doc was talking about here.  Let me try toO > >explain.  I don't believe that he was talking about the whole OO (read that:aI > >"modular programming") concept as being against the way that computers I > >operate, but rather I think he means that OO, as it's included in most O > >modern languages introduces an amount of artificial complexity that may make 1 > >the program's operating seem to be non-linear.n > >eG > >You might interpret his argument as being for a simpler method of OOe > >programming.  > >TK > >I assume he's not against re-using his code once in a while, and writingr& > >apps that are relatively "modular." > >  > >Regards,  > >  > >Chris >  > Well said Chris. > G > I do prefer functional programming and state machine models to moderneM > objectification. The obsession with abstracting to miles away from hardware L > seems to result in unreliable builds, and objects which are way beyond the > size required. > G > I suppose my dislike of OO stems from my assembler background. (Whered% > modularity and reuse are critical).P >  > Doc. >  > -----BEGIN PGP SIGNATURE-----N > Version: 2.6.2 > B > iQEVAwUBOsey8sriC3SGiziTAQG3fAf/ZD9FcbEdO1SLAyTNST59tmBvy5S1NdOKB > brO6RW2Ppm7+/RnHVSP50Vx9gu02hjw6xOve0rJIUdtTzXGYZI5etYA5xf9iqRiQB > GKD7GgwkexENVD6fmPIIwcQuDvzysMscBLGHn4ZZ00KuCQ+JjrCUfQM7DDRQbvnBB > YVM9ciH9qGXOzyM3k5h7n/yucMp8RKREqzSuxDvQO/vME/8RCMqpW4kfDOvRgaPNB > y1YYDL7ucruPa5ADa5SV5mPHYo/cPP6m0gn5voTNwEIovW/fIz7CiQCSFZFD3NtP: > VKd0QPDuwEjlJxu1v52Uv6bLkxqPlncwKhQHsjMnNpli1nLTVJwuLg== > =OsKmR > -----END PGP SIGNATURE-----e  @ Okay then: the objection came from abuse of oo. Occam's Razor is> the measure. If oo makes things more complicated than it's the< wrong technique for a given task. My experience was *always*< that the complex program systems I have to deal with (mostly: implemented with C or C++) were much more complicated than: necessary because oo wasn't used or not used correctly. If> you've *studied* the GoF book "Design Patterns" you would know< that design an oo framework is one of the hardest tasks. The? fact that it is very easy to use oo features in modern PLs very ? often leads to a designless oo system which nobody understands.y< A few weeks ago a prestige Internet banking project has been? canceled after about one year of implementation time with Java.:? Honestly I don't know the details but I know some of the people > who worked there, I will ask them and I expect that Java was a; major reason for failing because oo with Java isn't easy toe master.r  < Another important point is that most of the so called oo PLs7 only offer half way oo, e.g. C++ was called oo from ther@ beginning whereas it lacked (and still lacks) important features9 of the oo paradigm. Don't blame oo if the wrong oo fails.   ? But again: I insist in my question about "knowing how computers < really work". I'm sure that it is a misconception to avoid a; level of abstraction because it widens the distance between ; written source and machine instructions because it would ben> difficult to define the appropriate separation line to justify? the use of an abstraction given by the PL C but not Ada and not @ prefer assembly language. Assembly language BTW isn't that close? to what the processor really does if you have a modern RISC cpua? like Alpha which does a lot of runtime code optimisation on the ? CPU level like instruction reordering or parallel execution. If ? we take into account that the program doesn't run allone but is.= preempted and resumed later on it becomes really difficult tor image what the cpu really does.o< Moreover I additionally insist that lean oo isn't farer away? from the cpu than any other compiled or interpreted language asr> opposed to CPU specific assembly language. If you are using an: abstract PL (abstract in the sense that you don't know the? target CPU from the PL's point of view) you can't tell what the2> CPU really does unless you analyse the compiler/code generator9 and accept the constraint of  focusing only on one targetR processor family.   ? Abstraction is the right way to go to increase power, speed ande quality of programming.o   ------------------------------  % Date: Mon, 02 Apr 2001 19:55:42 -0500 + From: Christopher Smith <csmith@amdocs.com> $ Subject: RE: Free The OpenVMS Seven!L Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D1CF5@cmiexch1.cmi.itds.com>   > -----Original Message-----2 > From: Christof Brass [mailto:brass@infopuls.com]  @ > I have to agree that these old cars had a lot of good features< > compared to the new ones. But I still don't think that the= > technique used to build them is competitive to the current.e= > Image if those simple cars would been built with the modern0B > production systems. OTOH it would be almost impossible to create8 > these modern cars with the old production systems at a. > reasonable cost and with reasonable quality.  K Point taken, and as Steven Scheda(I certainly hope I spelled that properly)tK pointed out to me, old cars have their bad features as well.  I'm still not L sure how appropriate the metaphor was , but in all seriousness, I was making! a joke about the car metaphor. :)e   Regards,   Chris   ! Christopher Smith, Perl Developerm Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");7 '8   ------------------------------  % Date: Mon, 02 Apr 2001 20:37:35 -0500i+ From: Christopher Smith <csmith@amdocs.com>i$ Subject: RE: Free The OpenVMS Seven!L Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D1CF6@cmiexch1.cmi.itds.com>   > -----Original Message-----2 > From: Christof Brass [mailto:brass@infopuls.com]  B > Okay then: the objection came from abuse of oo. Occam's Razor is@ > the measure. If oo makes things more complicated than it's the> > wrong technique for a given task. My experience was *always*> > that the complex program systems I have to deal with (mostly< > implemented with C or C++) were much more complicated than< > necessary because oo wasn't used or not used correctly. If  L You can say that again.  I think that quite a bit (though not all -- I'm notL saying that "OO" languages can't be designed in a better way) of the problemL people have in OO languages is in the fact that their programmers understand: neither the language nor the machine for which they build.  @ > you've *studied* the GoF book "Design Patterns" you would know> > that design an oo framework is one of the hardest tasks. The  2 You'll also get no argument from me on this point.  > > Another important point is that most of the so called oo PLs9 > only offer half way oo, e.g. C++ was called oo from thebB > beginning whereas it lacked (and still lacks) important features; > of the oo paradigm. Don't blame oo if the wrong oo fails.i  I Bjarne Stroustrup has said on several occasions that C++ is not, strictly J speaking, an "object oriented" language. (Though is _possible_ to write OO0 code in C++)  People just won't listen to him.    A > But again: I insist in my question about "knowing how computersi> > really work". I'm sure that it is a misconception to avoid a= > level of abstraction because it widens the distance betweent= > written source and machine instructions because it would bea@ > difficult to define the appropriate separation line to justifyA > the use of an abstraction given by the PL C but not Ada and noto  G I think that he might be complaining about the lack of understanding in-H people who write the programs, and viewing that as a feature of the moreL abstract language.  Of course, it's not the languages fault exclusively, andH certainly abstraction is the right way to go in many cases, but somebodyI should ensure that new programmers get exposure to a variety of languagespA and paradigms so that their understanding is as good as possible.n  G In a sense, the argument should be for better education of programmers.-  J I personally see the same problems, but I believe the cause is that peopleJ are now unwilling to take the time to really understand the computers withL which they work.  There are a lot of interfaces which will give the end-userE completely wrong ideas about the underlying system (Macintosh desktopHD metaphor, anyone?  Perhaps even more sinister ... device abstractionK libraries.  I don't claim these things are bad -- I like them in many cases ? -- but they lead to misconception.), and a larger percentage ofwK "programmers" these days don't even think to challenge those ideas in theirt
 own minds.  > > Moreover I additionally insist that lean oo isn't farer awayA > from the cpu than any other compiled or interpreted language as @ > opposed to CPU specific assembly language. If you are using an< > abstract PL (abstract in the sense that you don't know theA > target CPU from the PL's point of view) you can't tell what the @ > CPU really does unless you analyse the compiler/code generator; > and accept the constraint of  focusing only on one targetc > processor family.b  E It depends on the OO language, and the language you compare with.  InUD general, though, you're right.  I wouldn't say that SmallTalk is any0 "further away from the CPU" than Pascal, really.  $ OO techniques only add to the set ofL "this-is-a-metaphor-and-is-not-exactly-how-the-computer-sees-this" facets of any high-level language.    J For instance, when you have a function which is buried in the hierarchy of# objects such that you call it thus:e  * MyObject.SomeInheretedClass.SomeFunction()   (oversimplification here...)L It is most likely compiled into just a plain jump.  The "hierarchy" there isI purely a linguistic device.  It shows which bits of previous work, if youh4 will, we've called upon in setting this "object" up.  H That's not a bad thing, and in fact, it's incredibly useful on occasion.L The problem comes when some person fails to realize that it's an "illusion,": just as macros, high-level flow control, and other things.  A > Abstraction is the right way to go to increase power, speed andg > quality of programming.4  G No argument there -- in projects sufficiently large. (Most of them are,vJ these days. :)  It's a concern of mine, though, that people don't "forget"G how to live without the more advanced constructs that are in high-levele languages.    L In other words:  It's always a good idea to pick the right tool for the job.E The only problem is that to know which is the right tool, you must bec' familiar with all of the tools at hand.   G One should never make the assumption that because this is the tool thatcJ everyone else uses for this job, it is the right tool, or even worse, that: because this is the only tool I know, it is the right one.   Regards,   Chrisg  ! Christopher Smith, Perl Developer  Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");h '   y   ------------------------------  % Date: Mon, 02 Apr 2001 11:29:42 -0700c0 From: Mark Berryman <Mark.Berryman@Mvb.Saic.Com>8 Subject: Re: Getting Apache/CSWS & Multinet to co-exist?, Message-ID: <3AC862A6.2919818D@Mvb.Saic.Com>   Andy Stoffel wrote:m > < > I'm looking for any advice/suggestions on how to correctly7 > configure Apache & Multinet (4.3A) to happily coexista > on a VMS 7.2-1 system. [remainder deleted]o   On the system using multinet:r  & $DEFINE/SYSTEM TCPIP$DEVICE UCX$DEVICE  C CSWS (Apache) expects to find the TCPIP$... logicals defined by thetG current version of TCPIP services for VMS.  Multinet still defines themt using the UCX$... namespace.  
 Mark Berrymane Mark.Berryman@Mvb.Saic.Com   ------------------------------  % Date: Mon, 02 Apr 2001 16:28:51 -0400 : From: "Koska, John C. (LNG-MBC)" <John.C.Koska@bender.com>- Subject: GS160 froze due to a bad KGPSA-CX...sK Message-ID: <3D35AD137AAAD411A6BA0008C7B1B12D6DBF2E@MBCALBEXC03.BENDER.COM>   5 I have a supported call open with Compaq, but I would 7 like to know if anyone else has experienced the problems that bit me today.  8 Environment:  GS160 Model 16 setup as monolithic SMP box8 with 12 KGPSA-CX cards distributed evenly in two master 6 PCI boxes and 2 slave PCI boxes (3 per box).  OpenVMS 4 Alpha 7.2-1H1 with latest fibre and other patches.  6 Console 5.8-251, PAL code 1.88-1. KGPSAs with patched 7 firmware from Fall 2000, degrading performance but mored1 reliable.  EMA12000 storage and 3 fibre switches t$ configured and wired for redundancy.  8 Problem:  Two bad KGPSA-CX (no green LEDs) caused system5 to hang, and worse... prevented system from init'ing o7 properly after power cycle.  (Console hangs on polling   of bad KGPSA cards)   5 Solution:  Remove failed KGPSA cards and power cycle.e  9 A Better Long Term Solution:  Hardware partition GS160 inl half, and cluster.  ; Concern/Worry:  I fear that even with hardware partitioned n= GS160, that if I had two KGPSA cards go bad again, each in a n4 hardware partition, I would be in same situation of < downtime. (Granted, I might stand a better chance of winning my state lottery.)  ' Anyone out there experienced similar?  m  = I am also posting to see if this is rare or somewhat common,  6 so as to drive forward with hardware partitioning and ! clustering versus not at my site.t  = Perhaps the above problem was just a fluke of partial failed C7 hardware, and the console being unable to deal with it.m  + Any thoughts, ideas to better my situation?t  : Compaq has told me that a newer version of the firmware is; available for the GS160, but it does not explicitly addressh9 the issue.  But then again, they indicate that it may fixi= it also.   I intend to upgrade firmware and see what happens.y   :) jck
 John Koska Matthew Bender & Co., Inc.
 1275 Broadway- Albany, NY  122042 USAQ 518-487-3255 JKoska@bender.comF  ; "I post personal opinion only, and all the disclaimers one O8 could imagine apply.  That includes, I speak for myself 6 only and my views in no way represent my employer(s)."   ------------------------------  $ Date: Mon, 2 Apr 2001 14:57:43 -0500% From: "steve smith" <ssmith@lbpc.com>P& Subject: HELP BACKING UP Oracle on VMS& Message-ID: <BR4y6.144$F96.819@client>  G yes, we have oracle8 running on a alphavms system. I didn't do it, just1J inherited it. The help is needed in getting rman to create hot backup fileG on a bound volume, if I omit the ".d%d.p%p' from the rman command I geth error message about block size   -- Steve Smithg Manager Technical Services Information Technology Law Bulletin Publishing Co.a (312)644-7807 Ext.137  (312)644-4355 Fax  ndelise@LBPC.com http://www.lawbulletin.com   ------------------------------  # Date: Mon, 02 Apr 2001 17:54:47 GMTr2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)Q Subject: Re: Hoff's Road Trips (was: Re: OpenVMS Technical Update Days - success!y2 Message-ID: <H13y6.447$fB6.11958@news.cpqcorp.net>  M In article <9a2sdg$bdh$1@news.IAEhv.nl>, "Hans Vlems" <hvlems@iae.nl> writes:hH :Well I do hope we'll get a similar result in Lisbon, early May when the# :European Decus event will be held.\B :I noticed Steve Hoffman's name among those that present sessions.  F   AFAIK, I will attending the London OpenVMS Technical Update in May, F   but I will likely NOT be in attendence at the upcoming Lisbon event.  B   The London OpenVMS Technical Update event is scheduled the week    after the Lisbon event.   F   Local OpenVMS management was concerned by the two consecutive weeks D   disappearing from my engineering schedule due to participation in F   London, Lisbon, and other (likely) European-area activities the weekI   of 13-19 May.  Through June, I will be attending the London and Dallas j?   Technical Update events, and was at the DC event last week.  i  H   Why Lisbon?  My presentation load was to be the "lightest" at Lisbon, H   and my engineering schedule is...  (The Lisbon travel schedule change G   was made just  last week.)  I do not know who will be presenting the r-   OpenVMS technical update session at Lisbon.a  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 02 Apr 2001 16:20:27 -0400t- From: JF Mezei <jfmezei.spamnot@videotron.ca>eQ Subject: Re: Hoff's Road Trips (was: Re: OpenVMS Technical Update Days - success!b+ Message-ID: <3AC8DF0A.A016D76@videotron.ca>a   Hoff Hoffman wrote:hG >   Local OpenVMS management was concerned by the two consecutive weekssE >   disappearing from my engineering schedule due to participation ineH >   London, Lisbon, and other (likely) European-area activities the week >   of 13-19 May.l  K How does Local VMS management ensure that you actually return when they let  you out of the building ?n  L Sound to me like athletes and ballet dancers in the golden days of the USSR:L they were allowed out only with an KGB escort that would make sure that theyM returned, and never allowed out more than was sufficient to make the contest.n   :-) :-) :-)    ------------------------------  $ Date: Mon, 2 Apr 2001 17:17:40 +0200- From: "Ruzsinszky Attila" <aruzsi@mailbox.hu>u% Subject: Hopf 6841 dtss time providers, Message-ID: <9aa4ud$3lo$1@athena.euroweb.hu>   Hi,e  : Who uses that and can help me to configure and find error?   Thanks:: Ruzsi    ------------------------------  % Date: Mon, 02 Apr 2001 05:57:13 -0500r. From: Lyndon Bartels <lbartels@pressenter.com>; Subject: How can I tell if a boot is the result of a crash?u. Message-ID: <3AC814B9.213BB681@pressenter.com>  F One of the things I'd like to accomplish during a system startup is to% determine how my computer went down. K  D I'd like to be able to determine (during system startup) whether the> previous "downing" was the result of a crash, power-outage, orB what-ever. I'd like to know this, so I can then send an email to aE pager, or something. But, I don't want to to anything if the previousc shutdown was a graceful one.  A Is there a way, by doing some tests at startup to determine this?     4 Hmmmm.. Oh yeah. Alpha VMS, v7.2-1 (Almost all ECOs)     Thanks in advance,   Lyndon     -- 'G My opinions are mine and mine alone. They seldom align with those of myn	 employer.r   ------------------------------  % Date: Mon, 02 Apr 2001 19:55:53 -0500t. From: Lyndon Bartels <lbartels@pressenter.com>F Subject: How to tell if a system is booting immediately after a crash.. Message-ID: <3AC8D949.281B9C29@pressenter.com>  F I want, during a startup, if the system has gone done as a result of a( crash or power outage or... whatever....  G Why? So if the system crashes and reboots in the middle of the night, Is. can tell the computer to page me or something.  B I was thinking of having syshtdown.com write out a data file. ThenB during subsequent boot, systartup_vms look for the file, if found,F delete all versions of the file, and peacefully exit. If not found, so my alarm procedures.    	 Thoughts?n         Thanks in advance,   Lyndon   -- sG My opinions are mine and mine alone. They seldom align with those of myi	 employer.    ------------------------------  % Date: Mon, 02 Apr 2001 23:16:54 -0400r- From: JF Mezei <jfmezei.spamnot@videotron.ca>tJ Subject: Re: How to tell if a system is booting immediately after a crash., Message-ID: <3AC940A3.22012703@videotron.ca>   Lyndon Bartels wrote:r > H > I want, during a startup, if the system has gone done as a result of a* > crash or power outage or... whatever....  N I just indiscriminatly send myself an email whenever the system boots. It actsN as log as well in the SYSTEM account so you can see when the system was booted because I keep those messages.   ------------------------------  $ Date: Tue, 3 Apr 2001 01:00:57 +0200! From: "kalev-" <you.wish@duh.net>e0 Subject: HP DAT drive on a OpenVMS -- Alpha box?- Message-ID: <9ab05l$mrn$1@cubacola.tninet.se>     Will a HP SureStore DAT24e work? HW: AlphaStation "255" OS: OpenVMS v6.1   Anyone have any tips?a) Compaq said "dont as us -- ask HP" ..hmm.e jk   ------------------------------  # Date: Mon, 02 Apr 2001 23:40:58 GMTy2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)4 Subject: Re: HP DAT drive on a OpenVMS -- Alpha box?2 Message-ID: <e68y6.466$fB6.13446@news.cpqcorp.net>  Q In article <9ab05l$mrn$1@cubacola.tninet.se>, "kalev-" <you.wish@duh.net> writes:-! :Will a HP SureStore DAT24e work?p     The OpenVMS FAQ section:  D MISC21.   Third-party disk/tape/controllers/SCSI/widgets on OpenVMS?     will be of interest.   :HW: AlphaStation "255"9  C   AlphaStation 255 4/233 or AlphaStation 255 4/300 series, an AlphahE   EV4-class microprocessor, probably with a fast-narrow SCSI storage e   configuration.   :OS: OpenVMS v6.1      Interesting.    E   AFAIK, the AlphaStation 255 is not supported on OpenVMS Alpha V6.1, -   and requires a later OpenVMS Alpha release.s   :Anyone have any tips?* :Compaq said "dont as us -- ask HP" ..hmm.     It might, it might not.b  G   You will want to upgrade to OpenVMS Alpha V6.2 or (better) to V7.1-2 eD or later if you want to play mix-n-match on your SCSI chain.  These D   OpenVMS releases are far better at adapting to and operating with +   various third-party SCSI storage widgets.   7   And again, please see the OpenVMS FAQ section MISC21.   I   Question SUPP5 in the FAQ is quite similar, save that is the "graphics d'   variant" of this particular question.e  H   Be aware that the media lifetime on DDS is often specified as roughly K   2000 head passes, and a single typical BACKUP pass will perform multiple -H   head passes over the start of the tape cartridge.  (For contrast, DLT :   media is often specified at a half-million head passes.)  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 2 Apr 2001 21:20:21 +0100s* From: eplan@kapsch.net (Peter LANGSTOEGER) Subject: Re: Is SUN crazy ???r* Message-ID: <3ac8df05$1@news.kapsch.co.at>  ] In article <3AC8AFD0.180DCAF1@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes: + >fabio_compaq@ep-bc.petrobras.com.br wrote:s >> tK >> http://www.sun.com/2001-0327/release/;$sessionid$UELFXTYAABK4TAMTA1LU3NQi >h >You need to consider the date.e  > And the poster need to consider the date too. It's 30-Mar-2001' And don't tell us this was intended ;-)g   -- r< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888n< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  $ Date: Mon, 2 Apr 2001 22:18:19 +0100$ From: "Simon" <anybody@anywhere.com>7 Subject: Java SDK and StarTeam (this is a long shot...)eC Message-ID: <w26y6.13819$u93.1779877@news6-win.server.ntlworld.com>e  - I'm trying to get the StarTeam source controlo+ client working on VMS 7.2 (Alpha). In order . to do this i have to install the Java runtime.  / I've installed the tcp/ip services, Applied thea. threading patch, then installed the JDK 1.1.8.+ It all seems to work OK, but when I run thew/ java binary I can't seem to specify the installe root ie:u3 java -Dinstall.root=DKA100:[000000.vms$common.java]i -classpath starteam.jar stcmdf  , gives me an error about install.root.  There+ is no documentation on how the install.rootg( should be specified on VMS, and StarBase' tech support don't know of anyone usinga" StarTeam on VMS.  Can anyone who's' managed to do this please get in touch.a) I'm pretty new to VMS so I may be missinga something obvious.   thanks for any help,   -Simon   email:  smh@REMOVE_THIS@dr.com   ------------------------------  % Date: Mon, 02 Apr 2001 17:38:33 -0400n- From: JF Mezei <jfmezei.spamnot@videotron.ca>n Subject: Merging DCLTABLES.EXE, Message-ID: <3AC8F154.86FC2913@videotron.ca>  D If one has two separate nodes with different software that are beingI integrated into a homogenous cluster, what is the best way to merge the 2  DCLTABLES.EXE ?   L Must one make a manual list of commands present on node 2 but not on node 1,F use VERB to extract them and then add them to node 1's DCLTABLES.EXE ?  K What about subcommands that might be present on node 2 but not node 1 (somen) subset of the SET command for instance) ?-   ------------------------------  % Date: Mon, 02 Apr 2001 17:54:42 -0400s  From: norm.raphael@jamesbury.com" Subject: Re: Merging DCLTABLES.EXE4 Message-ID: <C2256A22.007780DB.00@jklh21.valmet.com>  K Looks like you want to extract all commands with VERB from each and compareaL them one to one where they match, then decide which one to use.  The rest of yourM procedure looks right.  Now you know one reason why remerging system disks ispP not a supported activity.  You will also want to watch out for sharables  which,N when copied, will also have to be put into appropriate libraries (you probably knowM the commands better than I do.  Those were the two biggest gotchas when I didp thisP several years (and versions) ago.  Watch out for VMSIMAGES.DAT on some products, too.        6 jfmezei.spamnot@videotron.ca on 04/02/2001 05:38:33 PM  . Please respond to jfmezei.spamnot@videotron.ca   To:   Info-VAX@mvb.saic.comi4 cc:    (bcc: Norm Raphael/Worcester/Neles-Jamesbury) Subject:  Merging DCLTABLES.EXEh        D If one has two separate nodes with different software that are beingI integrated into a homogenous cluster, what is the best way to merge the 2  DCLTABLES.EXE ?l  L Must one make a manual list of commands present on node 2 but not on node 1,F use VERB to extract them and then add them to node 1's DCLTABLES.EXE ?  K What about subcommands that might be present on node 2 but not node 1 (somer) subset of the SET command for instance) ?e   ------------------------------  # Date: Mon, 02 Apr 2001 21:59:51 GMTr2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)" Subject: Re: Merging DCLTABLES.EXE2 Message-ID: <rD6y6.464$fB6.13168@news.cpqcorp.net>  \ In article <3AC8F154.86FC2913@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:7 :...what is the best way to merge the 2 DCLTABLES.EXE ?e : M :Must one make a manual list of commands present on node 2 but not on node 1,aG :use VERB to extract them and then add them to node 1's DCLTABLES.EXE ?  : L :What about subcommands that might be present on node 2 but not node 1 (some* :subset of the SET command for instance) ?  B   VERB and DIFFs, most likely.  In combination with a little DCL, -   you will end up with (only) the exceptions.t  B   I don't know of a specific tool to merge the two DCL tables, notA   that I would particularly trust an automated tool to merge the ,(   two DCL environments correctly anyway.  C   Alternatively, you could use this as an opportunity to clean out "C   the cruft built up on the system disks, and to learn what layered,   products are really in use.r    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 02 Apr 2001 18:27:31 -0400M- From: JF Mezei <jfmezei.spamnot@videotron.ca>t" Subject: Re: Merging DCLTABLES.EXE, Message-ID: <3AC8FCCA.967CEE8C@videotron.ca>  ! norm.raphael@jamesbury.com wrote:eM > Looks like you want to extract all commands with VERB from each and compare B > them one to one where they match, then decide which one to use.   _ Looks like I want to petition Hoff to write a SET COMMAND/MERGE/AUTOMATIC/SOURCE=file/DEST=filea  G :-) :-) :-) :-)g  R > not a supported activity.  You will also want to watch out for sharables  which,B > when copied, will also have to be put into appropriate libraries  N There is a nifty SYS$UPDATE:DECW$GET_IMAGE_VERSION.COM  which is of great helpN in deciding which image should overwrite which (which has the higher version).L What I find intersting is that it comes with decwindows and not vanilla VMS.      R > several years (and versions) ago.  Watch out for VMSIMAGES.DAT on some products, > too.    P What exactly is VMSIMAGES.DAT and sources/logic is used by autogen to build it ?   ------------------------------  # Date: Mon, 02 Apr 2001 23:20:28 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)" Subject: Re: Merging DCLTABLES.EXE2 Message-ID: <0P7y6.465$fB6.13347@news.cpqcorp.net>  \ In article <3AC8FCCA.967CEE8C@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:" :norm.raphael@jamesbury.com wrote:N :> Looks like you want to extract all commands with VERB from each and compareC :> them one to one where they match, then decide which one to use. r :g` :Looks like I want to petition Hoff to write a SET COMMAND/MERGE/AUTOMATIC/SOURCE=file/DEST=file ::-) :-) :-) :-)    G   Remember to write your petition on the back of a cashier's check withsG   numeric value of indeterminate magnitude.  Pending receipt of a check,I   inscribed with with a value of sufficient magnitude, well, I have some u   other work in the queue.    O :There is a nifty SYS$UPDATE:DECW$GET_IMAGE_VERSION.COM  which is of great helpeO :in deciding which image should overwrite which (which has the higher version).oM :What I find intersting is that it comes with decwindows and not vanilla VMS.6    /   That tool does not do what you think it does.u  +   The PCSI generation mechanism does do it.t    J :> several years (and versions) ago.  Watch out for VMSIMAGES.DAT on some  :> products, too.& :.M :What exactly is VMSIMAGES.DAT and sources/logic is used by autogen to build o :it ?I    B   That is the list of images that should be installed when OpenVMSC   bootstraps.  AFAIK, AUTOGEN is the only producer of VMSIMAGES.DAT D   files.  I am aware of cases were folks did not run an AUTOGEN whenG   required, and thus the contents of VMSIMAGES.DAT were old and crufty.P      N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 3 Apr 2001 01:58:55 +0200m) From: maulis@ludens.elte.hu (Adam Maulis)o" Subject: Re: Merging DCLTABLES.EXE! Message-ID: <F0PT3hxPJ72N@ludens>/  \ In article <3AC8FCCA.967CEE8C@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: [...]o > D > What exactly is VMSIMAGES.DAT and sources/logic is used by autogen > to build it ?S    E VMSIMAGES.DAT is used by the startup: use this with standard input ofpD install utility. So VMSIMAGES.DAT containts all the install commands' what is must execute in every startup. e: (for exactly: sys$startup:VMS$INITIAL-050_VMS.COM do that)    J VMSIMAGES.DAT is created by autogen from sys$manager:vms$images_master.dat    E   You may want to modify the sys$manager:vms$images_master.dat itselfcI   rather than put a dozen of "install replace ..." lines to systartup_vmsgJ   and calculating global pages/sections requirements by hand. But you have<   to know that every VMS upgrade or patch may overwrite your   vms$images_master.dat.      I (There is 2 o'clock AM so I cannot correct my English language errors :-)i    I Adam Maulis              maulis@ludens elte hu         VMS system managereH  .......................................................................H  VMS Competence Center                             VMS Szakertoi KozpontH  Eotvos Lorand University                  Eotvos Lorand TudomanyegyetemH  Budapest, Hungary                                              BudapestH  =======================================================================   ------------------------------  # Date: Tue, 03 Apr 2001 01:16:14 GMT-) From: rob.buxton@wcc.govt.nz (Rob Buxton)-" Subject: Re: Merging DCLTABLES.EXE0 Message-ID: <3ac922ff.18712176@news.wcc.govt.nz>  C We went through a very similar  exercise. I built a new System Disk-
 from scratch. B This involved identifying and then installing all required layered	 products.-  * The advantage is a nice clean System Disk.  E You could then use verb to list  the different DCLTABLES and identify F the differences.  Then determine if you still use those products. OnceD complete there should be no need to merge any files. You also end up@ with a System Disk that only contains what you have installed, a- benefit if you're not the original Sys Admin.g   Rob., On Mon, 02 Apr 2001 17:38:33 -0400, JF Mezei% <jfmezei.spamnot@videotron.ca> wrote:r  E >If one has two separate nodes with different software that are beingnJ >integrated into a homogenous cluster, what is the best way to merge the 2 >DCLTABLES.EXE ? >aM >Must one make a manual list of commands present on node 2 but not on node 1,uG >use VERB to extract them and then add them to node 1's DCLTABLES.EXE ?e > L >What about subcommands that might be present on node 2 but not node 1 (some* >subset of the SET command for instance) ?   ------------------------------  " Date: Tue, 3 Apr 2001 00:08:11 GMT7 From: moroney@world.std.spaamtrap.com (Michael Moroney)9 Subject: Re: modemsr& Message-ID: <GB6wDn.D99@world.std.com>  0 Lyndon Bartels <lbartels@pressenter.com> writes:  A >Has anyone done any work with modems connected directly to Alpha:C >workstations? I want to add on to my 500au for dialing in to work,:2 >either via SLIP, PPP, or to a terminal server....  H >Any suggestions on hardware, and software would be greatly appreciated.  @ I've installed a couple of modems in Alphastations with success.E Pretty much any ISA non-PnP modem should work (mine were both cheapietF generics)  Note: PCI modems will not work (no VMS driver) and Plug andE Pray modems will not work either.  I think most modems now are PnP soiI you'll have to get an older one with jumpers.  Set it to COM3 or COM4 anduG IRQ 5.  The page at http://www.tmesis.com/modem/ mentioned by VAXman istF pretty good.  One thing:  There is a PPP autologin option (forget whatB it's called) that VMS does not support.  If your ISP uses this youH can not use PPP on this ISP.  VMS unceremoniously hangs up as soon as it& connects if your ISP has this feature!  A If you need more help I can get the config info and post it here.s   -Mikee   ------------------------------  % Date: Mon, 02 Apr 2001 23:40:46 +0000s) From: Christof Brass <brass@infopuls.com>e( Subject: Re: OpenVMS Educational Program, Message-ID: <3AC90DFE.27710E7E@infopuls.com>   Christopher Smith wrote: >  > > -----Original Message-----4 > > From: Christof Brass [mailto:brass@infopuls.com] > % > > steven.reece@quintiles.com wrote:  > P > > > There is no justification for VMS to be uncontaminated and architecturallyO > > > pure.  TCP/IP on OpenVMS may not be my first choice of network transport,,8 > > > but in the present day and age it is a Good Thing. > C > > Why do you relate architecture and implementation of protocols?i > I > I feel compelled to comment here.  I understand that you're speaking of D > "system level" architecture here, but in a broader sense, anythingN > distributed with the system, whether or not it is normal non-privileged, andL > completely optional code, has had design considerations made for it. (Even> > if only the layout of the CD from which the OS is installed) > M > So I guess what I'm saying is that UCX, for instance, is a component of VMS.N > as it's distributed today, and therefore figures in (in however small a way)L > with the architecture of the system.  That the system may function withoutK > this product doesn't necessarily mean that it's not a part of the system.-M > The system may function without SCSI drivers too, I'm sure. :)  Do we counttI > them, then, because they're installed automatically, or because they'rerN > system level code?  Or do we exclude them, and say that anything with is not > the "kernel" will not count? > M > (Those aren't rhetorical questions -- I am curious about where you draw the-E > line.  What is part of the "architecture," and what is incidental?)d > 
 > Regards, >  > Chrisf > # > Christopher Smith, Perl DeveloperV > Amdocs - Champaign, IL >  > /usr/bin/perl -e 'A > print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");k > '"  9 The difference between architecture and implementation ofk8 features ("of features" was left out unintentionally but3 obviously misleading) is that the architecture onlyi= defines/describes how features are to be implemented, how then? APIs are to be designed and how these components work together.s> As a simple example let's look at a new VMS command. It should> have a clear distinction between qualifiers and parameters and; if it offers functions which are already present with other6< commands its qualifiers to control them should have the same: name. The schema how to do it is the architectural aspect.< Although I'm *not* a VMS systems programmer I try to give an: example more close to the TCP/IP topic. The VMS API to use@ TCP/IP services should be similar to other I/O and communication; related APIs already present on VMS. This could include the 6 necessity to combine some operations which are offered> separately on other OSs like UNIX (which should avoided) where@ the idea of small building blocks more often than not leads to a> sort of API where it's difficult to get things done correctly.   ------------------------------  % Date: Mon, 02 Apr 2001 17:57:46 -0700 ! From: Shane.F.Smith@Healthnet.comf Subject: OT, but worth a lookvD Message-ID: <OFF9E1E0EF.FC7AA13C-ON88256A23.00054A0B@foundation.com>  F OK, matey-peeps. If you need a chuckle head on over to www.hardocp.comJ (don't worry, it's not pr0n, it's an overclockers' website with a sense ofK humour). There's a picture that will probably make it to many people's cube-K wall, near the top just below "It was one of those moments when you saw thenI door wide open to totally mess with someone's head." It's a thumbnail, so-+ you'll have to click it to see it properly.2   Enjoy.   Shanea  G P.S. This is the site I first saw "All your base are belong to us!" on,D3 about three months before it hit the news services.F   ------------------------------  % Date: Mon, 02 Apr 2001 23:51:06 +0000 ) From: Christof Brass <brass@infopuls.com>e, Subject: Re: POSIX Streams, File Permissions, Message-ID: <3AC9106A.E1880B19@infopuls.com>   andrew harrison wrote: >  > Christof Brass wrote:  > >n > > andrew harrison wrote: > > >1 > > > Christof Brass wrote:i   [SNIP]  > > > IP is a widely used standard and at the moment there is no= > > promising way to avoid it. Besides that it is possible to-0 > > implement an IP stack according to the RFCs. > >@ > 8 > Of course it is but the UCX stack came from UNIX where9 > it was implimented according to the RFC's. The point is 7 > it origionated from UNIX so you need to rip it out ifo7 > you are to deliver on your architectural purity rant.   > I really don't understand the problem. You can use whatever OS; you want to develop helpful ideas. The ideas may or may not = influenced by the OS. The fact that VMS has been designed andh? implemented because the engineers at DEC knew after 10 years ofi@ bad experience with UNIX (which should be avoided) what mistakes; to avoid led to a clean and well thought through design. Ofn: course they didn't implement all forthcoming standards and= functions. The important thing is how a feature is integratedt? and that its API is according to the rules/standards of the OS.s  = [All the examples of standards/protocols snipped: CDE, Motif,n  DECthreads, PERL, Tcl/Tk, CORBA]  > If a lot of protocols have firstly been implemented not on VMS? than this reflects mainly the fact that the market share of VMSI in these areas isn't that big.   ------------------------------   Date: 2 Apr 2001 20:46:27 +0100c* From: eplan@kapsch.net (Peter LANGSTOEGER)A Subject: Re: Queue Manager gives me this error about once a month@* Message-ID: <3ac8d713$1@news.kapsch.co.at>  | In article <300320011313123700%spencer@pageweavers.spaamfree.com>, David Spencer <spencer@pageweavers.spaamfree.com> writes:G >As described in the header, I get the following message on the consoleT$ >(somewhat abbreviated for clarity): >s >$9 >%%%%%%%%%%%  OPCOM  30-MAR-2001 11:12:25.67  %%%%%%%%%%% # >Message from user SYSTEM on XXXXXXe  >%QMAN-E-WRITEERR, error writingB >$2$DKA0:[SYS0.SYSCOMMON.][SYSEXE]SYS$QUEUE_MANAGER.QMAN$JOURNAL;1' >-RMS-F-RMV, ACP remove function failed-' >-SYSTEM-F-DATACHECK, write check errore >%QMAN-E-OPENERR, error openingsB >$2$DKA0:[SYS0.SYSCOMMON.][SYSEXE]SYS$QUEUE_MANAGER.QMAN$JOURNAL;10 >-RMS-E-FEX, file already exists, not superseded+ >-SYSTEM-W-DUPFILENAME, duplicate file namec >aA >I have other systems that never display this message. I've triedt@ >and tried to figure out what's different with this one but with	 >no luck.   F I'd start with the access rights for delete to this file (for SYSTEM).G The queue manager does obviously replace the file by deleting it first,eJ then creating it newly and then doing a content check (which fails because6 the delete already failed). But I may be wrong also...   -- h< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888e< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  # Date: Mon, 02 Apr 2001 18:13:09 GMTB2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)1 Subject: Re: Seeking CD-R/CD-RW SCSI INQUIRY data 2 Message-ID: <Vi3y6.450$fB6.11888@news.cpqcorp.net>  R In article <GB1v9u.IHC@spcuna.spc.edu>, Terry Kennedy <terry@gate.tmk.com> writes:C :In comp.os.vms Hoff Hoffman <hoffman@xdelta.zko.dec.nospam> wrote:e :>     Yamaha CDR100 :>     Yamaha CDR400 :tG :  FWIW, these are both incredibly obsolete widgets. I've owned both ofo  :'em, as well as the CDRW4260S.   D   Any computer-related widget that is currently or soon shipping is    (by definition) obsolete...f  G   With high-volume widgets such as CD-R, the product obsolescence, um, 1$   shipping cycle is yet short(er)...  C :After that, I started buying internal IDE burners (TDK VeloCD) as ># :they were much faster/cheaper/etc.   F   I haven't yet looked at any IDE CD-R widgets...  (And DQDRIVER only /   recently provided IO$_DIAGNOSE support, too.)   G   If somebody sends me an IDE CD-R widget -- though my stable of Alpha bF   host hardware currently only includes SCSI connections -- I'll look 0   at the capabilities of the IDE CD-R drive. :-)  D :  If this has something to do with supporting burners under VMS, it2 :would be really cool to support the IDE VeloCD's.  I   Supporting burners is an entertaining and exceedingly difficult topic, uJ   and formal CD-R support is definitely not something that I would expect    to see added anytime soon.  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  + Date: Mon, 02 Apr 2001 21:41:00 -0500 (CDT)  From: sms@antinode.org1 Subject: Re: Seeking CD-R/CD-RW SCSI INQUIRY data ) Message-ID: <01040221410051@antinode.org>o  - From: "Paul A. Jacobi" <pjacobi@mediaone.net>f  C > The reporting of LUNs is different from SCSI-1 and SCSI-2.  [...]g. > This is a bug in the device firmware.  SinceG > Windows does not normally use SCSI LUNs, this feature of the firmwaredN > usually doesn't get tested.  You probably won't have any luck convincing theI > manufacture to change the firmware, since it work with Windows and they $ > probably won't care about any O.S.  F    Certainly, if _I_ ask them, the effect would be vanishingly small. F If an appropriate representative of Compaq Computer Corporation asked,C and offered some technical assistance, the effect might be greater.:  J > You will also need to verify if the CD-ROM drive supports 512-byte blockM > mode.  Some drives claim to implement 512-byte block mode, but implement it  > incorrectly.  #    And this test would be done how?   H ------------------------------------------------------------------------  C    Steven M. Schweda               (+1) 651-699-9818  (voice, home)sC    382 South Warwick Street        (+1) 763-781-0308  (voice, work)aG    Saint Paul  MN  55105-2547      (+1) 763-781-0309  (facsimile, work)H9    sms@antinode.org                sms@provis.com  (work)D   ------------------------------  % Date: Mon, 02 Apr 2001 19:50:06 -0500i. From: Lyndon Bartels <lbartels@pressenter.com>C Subject: Re: Seeking CD-R/CD-RW SCSI INQUIRY data / Yamaha CRW4416Sa. Message-ID: <3AC8D7EE.1F56E5B0@pressenter.com>  < For What's it's worth, here's the info from my Yamaha 6416S.   Alpha PWS 500au, AXP VMS V7.2-1      $ show dev dkb200/full  < Disk HOTROD$DKB200:, device type YAMAHA CRW6416S, is online,
 file-oriented F     device, shareable, available to cluster, error logging is enabled.  2     Error count                    0    Operations completed                  0F     Owner process                 ""    Owner UIC                      [SYSTEM];     Owner process ID        00000000    Dev Prot           E S:RWPL,O:RWPL,G:R,Wf6     Reference count                0    Default buffer size                 512  
 $ etc dkb200:a $! $! SCSI_INFO V1A2t $!E $!      Copyright (c) 1995 by DIGITAL Equipment Corporation, Maynard,t Mass.tB $!      All Rights Reserved. Unpublished rights reserved under the	 copyrightv" $!      laws of the United States. $! $! Issuing INQUIRY QIOWm $!# $!   SCSI Inquiry Data (58 bytes) :  $!= $!      000:  05 80 02 02 27 00 00 10 59 41 4D 41 48 41 20 20i= $!      010:  43 52 57 36 34 31 36 53 20 20 20 20 20 20 20 20 = $!      020:  31 2E 30 62 33 30 2F 30 35 2F 39 39 00 00 00 00r+ $!      030:  00 00 00 00 00 00 00 00 00 00a $!' $!      Peripheral Device Type : CD-ROM  $!% $!      Peripheral Device Qualifier :S $!F $!              Peripheral device type supported & connected to target $!& $!      Vendor Identification : YAMAHA $!( $!      Product Identification: CRW6416S $!$ $!      Product Revision Level: 1.0b       -- -G My opinions are mine and mine alone. They seldom align with those of myo	 employer.n   ------------------------------  % Date: Tue, 03 Apr 2001 00:14:57 -0400-. From: Chuck McCrobie <mccrobie@cablespeed.com>C Subject: Re: Seeking CD-R/CD-RW SCSI INQUIRY data / Yamaha CRW4416SE. Message-ID: <3AC94E41.D8DE5937@cablespeed.com>  H Our SCSI bus analyer reported a CHECK CONDITION on a MODE SELECT.  Don'tF remember the details though.  It seemed that DKDRIVER attempts to MODE- SELECT something that the drive doesn't like.A   Eberhard Heuser-Hofmann wrote: > @ > >From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) > >> Yamaha CRW4416S:o > G > >   Coincidentally, I just hooked mine back up to my AlpSta 200 4/233MK > >(VMS V7.2-1), after consolidating the two old disks onto one newer disk.sK > >With the shorter SCSI cables, it no longer completely paralyzes the SCSIr > >bus.e >  > >> $! Issuing INQUIRY QIOW >   > >   Mine's slightly different: > & > >$!   SCSI Inquiry Data (58 bytes) : > >$!p@ > >$!      000:  05 80 02 02 27 00 00 10 59 41 4D 41 48 41 20 20@ > >$!      010:  43 52 57 34 34 31 36 53 20 20 20 20 20 20 20 20@ > >$!      020:  31 2E 30 68 30 39 2F 30 39 2F 39 39 00 00 00 00. > >$!      030:  00 00 00 00 00 00 00 00 00 00 > >[...]' > >$!      Product Revision Level: 1.0he > B > >   Even with the shorter cables, it still appears at every LUN: >  > >alp $ show device dka > S > >Device                  Device           Error    Volume         Free  Trans Mnt S > > Name                   Status           Count     Label        Blocks Count Cnt S > >ALP$DKA0:               Mounted              2  VMS072ALP     15133685   545   1=1 > >ALP$DKA400:             Online               0u1 > >ALP$DKA500:             Online               0g > 1 > >Device                  Device           ErrorE1 > > Name                   Status           Count 1 > >ALP$DKA501:             Offline              1 1 > >ALP$DKA502:             Offline              1d1 > >ALP$DKA503:             Offline              1 1 > >ALP$DKA504:             Offline              1D1 > >ALP$DKA505:             Offline              1e1 > >ALP$DKA506:             Offline              1d1 > >ALP$DKA507:             Offline              1p > G > >   Is there an approved way to suppress the extras, or should I stope. > >caring, or am I the only one who gets them? > K > >   Also, running CDRECORD 1.8.1 (slightly modified), I did a "-dummy" of H > >about a full disk worth and got about 170 errors on PKA0:.  Normal or > >not?  > K > >   Is "%MOUNT-F-FORMAT, invalid media format" the best I can expect whennG > >trying to mount a normal CD-ROM on the thing?  That's with the blockM > >size jumper installed.o > I > >   I haven't tried it lately, but as I recall, booting from it failed,  > >too.t > E > >   Comments from anyone with relevant experience would be welcome.  > ' > Two reasons may cause these problems:e3 > 1.) Firmware of the drive (Upgrade it with a PC).-3 > 2.) scsi-driver-problems (get the newest driver).T > 
 > eberhard > H > PS: I am able to mount a burned CD with my Yamaha 4260S (OpenVMS 7.1).Q > ===============================================================================H >  > Dr. Eberhard Heuser-Hofmann  > Univ. Konstanz > Fakultaet fuer Chemie? > Universitaets-Strasse 10 > D-78464 Konstanz	 > Germanyt0 > Phone: +49-7531-88-2026, FAX: +49-7531-88-3139, > email: vaxinf@chclu.chemie.uni-konstanz.de > N > ============================================================================ > Q > ===============================================================================e >  > Dr. Eberhard Heuser-Hofmannr > Univ. Konstanz > Fakultaet fuer Chemie  > Universitaets-Strasse 10 > D-78464 Konstanz	 > Germany 0 > Phone: +49-7531-88-2026, FAX: +49-7531-88-3139, > email: vaxinf@chclu.chemie.uni-konstanz.de > Q > ===============================================================================s   -- a --g   ------------------------------  % Date: Mon, 02 Apr 2001 19:58:59 -0700P, From: Mark Berryman <berryman@karfoogle.com>% Subject: Re: Spice for OpenVMS Alpha? , Message-ID: <3AC93C73.2050207@karfoogle.com>   andrew harrison wrote:   > Tim Llewellyn wrote: >  >> Bill Gunshannon wrote:- >> - >> -- >>> In article <3AC49C9F.B90D2FF4@bbc.co.uk>,44 >>>  Tim Llewellyn <tim.llewellyn@bbc.co.uk> writes: >>> |>C >>> |> and btw I have beeing doing some research in comp.unix.* :-)p >>> 
 >>> And??? >>>  >>> bill >>  H >> well, they don't seem to have a high opinion of Solaris security over >> in comp.security.unix.n >> e >  > / > I am not sure what the point of your post is.o > . > Posters on this group do have a high opinion1 > of OpenVMS security. This high opinion however c4 > appears to be based on a lack of information about/ > the actual risks to OpenVMS fostered by what y. > is a culture of security through obscurity. 8 > The CERT advisories for OpenVMS are a perfect example 
 > of this. > 5 > Can you say with certainty that the differences you 2 > see in each newsgroup are anything other than a . > manifestation of the difference in approach. > 1 > Various posters have claimed on this newsgroup n0 > that OpenVMS pretty much clean as far as CERT 3 > advisories are concerned only to find that Compaqo, > had not posted responses when OpenVMS was  > vunerable. > 3 > Various posters have also claimed that OpenVMS ise3 > immune to the stack overflow exploits attributed i0 > to UNIX and NT only to find that this is also  > not the case.s  G As desperate as Andrew is for OpenVMS to turn out to be no more secure h? than Solaris, or as desperate as he appears to be to show some wE conspiracy of silence within VMS engineering I don't think the facts a
 bear him out.   G For instance, part of my job is to verify what attacks our systems may tD be vulnerable to - said systems including VMS, NT, and various Unix F flavors.  Not once, *not once*, have the VMS systems I've tested been G vulnerable to any of the DoS or exploit attacks that are documented on iG CERT (and I've tested pretty much all of them).  Solaris, on the other  I hand, has been vulnerable to most of them (I don't have my list in front "2 of me but there were one or two that it survived).  F Andrew can wish and allege all he wants; for me, I'll believe my test  results.  
 Mark Berrymani  G P.S. Andrew, you also need to remember that IP is a very small part of eE the security model of VMS.  You don't earn, and maintain, a security pL reputation of the magnitude of VMS without something pretty solid behind it.   ------------------------------   Date: 2 Apr 2001 22:24:47 +0100a* From: eplan@kapsch.net (Peter LANGSTOEGER)$ Subject: Re: TCP/IP 5.0A ECO Appears* Message-ID: <3ac8ee1f$1@news.kapsch.co.at>  Z In article <998egq.ac.1@hans.myfqdn.de>, Hans.Bachner@altavista.net (Hans Bachner) writes:M >The long awaited VMS72_PCSI ECO for OpenVMS VAX is now officially available P  ! I did read the announcement, too.iK But, after two more weeks, it is still NOT available in the FTP server areai! and NOT downloadable via DSNlink.c1 So, is it really available or do I need glasses ?y" Or should I get paranoic instead ?  H >(I'm sure your local support organisations have provided a provisional  >version of this kit earlier).  6 Nope. They wanted me to sign a beta tester agreement !+ Some kind soul here did help me instead ;-)7  M >                                The kit does not show up on the patches web eM >site yet, but should soon be available there. Until then, your local Compaq -# >support can provide the final kit.o  G But only if I have a support contract, which I unfortunately don't havewI any longer since a year now (for VAX; or soon won't have for Alpha, too).i   Nice try, but alas   -- l< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888d< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Mon, 02 Apr 2001 19:18:00 +0100e  From: steven.reece@quintiles.com Subject: Re: VAX in 1UH Message-ID: <OFCA095551.D5559542-ON80256A22.0062DD41@qedi.quintiles.com>  H Since Compaq no longer manufacture VAXen, I think you're probably out ofH luck.  However, depending upon which models of the VAX 4000 and MicroVAX 3100 you have.........  G Have you thought about migrating to, say, VAX 4000-108 or MicroVAX 3100uK model 98 systems and use Storageworks disks and controllers on them?  WouldsK the maintenance on such a setup be less than on the three systems and R400X. that you have now?  I Three of the things that need to be considered as systems get older are :i$ - how much does it cost to maintain? - how much does it cost to run?t1 - how much is the floorspace that it's taking up?i  H In the case of old systems (like, say, a VAX 11-780 with lots of vintageK disks around) the floorspace is expensive, the maintenance is expensive andnJ the power it consumes is expensive.  One can gain significant benefits allB round if one purchases something like a VAX 4000-100 system with aG Storageworks shelf or two and use this to hold the data and support the # applications instead of the 11-780.   H If the combined VUP rating of the three systems you have is, say, 18 VUPH and you use two rackmount systems with smaller footprints and get 20 VUPI then you may have provided additional processing power for your users tooo3 (all things being equal, which they rarely are :-))k  B There are, of course, other considerations in the average business environment....e   Steve.   Fabio wrote:? >>>An interesting idea nowadays would be Compaq launching a new- (definitive)D VAX / MicroVAX in a 1U cabinet to maintain the legacy (???) systems.E I have three VAXes here which I cannot "shutdown" because they have aa0 legacy application. And they are ocupping space:   2 (two) VAX 4000 1 (one) MicroVAX 31007 1 (one)  R400X <<<s   ------------------------------  # Date: Mon, 02 Apr 2001 19:14:25 GMT:2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: VAX in 1U2 Message-ID: <lc4y6.456$fB6.12264@news.cpqcorp.net>  x In article <OFFA46A003.2E542EBB-ON03256A22.005E5BD6@ep-bc.petrobras.com.br>, fabio_compaq@ep-bc.petrobras.com.br writes:J :An interesting idea nowadays would be Compaq launching a new (definitive)E :VAX / MicroVAX in a 1U cabinet to maintain the legacy (???) systems.r  H   DIGITAL discontinued all VAX microprocessor production quite some timeF   ago and the last VAX microprocessors were fabbed well before the fabG   itself was sold to Intel, and the last of the (new) VAX systems using E   the existing stockpile of VAX microprocessors was then incorporateduA   into processor modules and shipped out over the last few years.t  H   The chances of seeing a 1U-class computer system with an embedded VAX %   microprocessor are rather unlikely.   F :I have three VAXes here which I cannot "shutdown" because they have a1 :legacy application. And they are ocupping space:) :u :2 (two) VAX 4000s :1 (one) MicroVAX 3100 :1 (one)  R400X   G   There are a gazillion systems with similar names -- due to the terse  I   model naming provided, I cannot tell if you have the smallest VAX 4000 hH   series systems or if you have one of the largest and could potentiallyJ   use an existing (smaller) VAX 4000 series.  Your R400X storage enclosureH   could definitely be configured into a smaller footprint, as it likely G   contains drives that are both huge (physical size) and tiny (storage p!   capacity) by current standards.C  G   You might well have a chance to reduce the size of this configurationmG   and stay on a VAX platform, in other words.  (This is not something I1G   would recommend, but VAX systems such as the VAXstation 4000 model 9x '   series are both fast and very small.)l  J   That said, Compaq (and DIGITAL before it) has been actively encouraging I   an upgrade from VAX to Alpha for many years now, and the end of the VAXsH   line was publicized -- though the publicity and the announcements wereJ   clearly (and unfortunately) not visible in your area.  You could likely J   easily stuff all of this existing application into an AlphaServer DS10L.  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 2 Apr 2001 22:07:35 +0200s) From: maulis@ludens.elte.hu (Adam Maulis)r Subject: Re: VAX in 1U! Message-ID: <Ifj3KwrXOTQL@ludens>   x In article <OFFA46A003.2E542EBB-ON03256A22.005E5BD6@ep-bc.petrobras.com.br>, fabio_compaq@ep-bc.petrobras.com.br writes:K > An interesting idea nowadays would be Compaq launching a new (definitive)iF > VAX / MicroVAX in a 1U cabinet to maintain the legacy (???) systems.  , 1, by an 1U height intel box from any vendor 2, by charon vax license to it+ 3, enjoy the performance and the small sizeo+ 4, of course you cannot enjoy the stability-   --  3 the vaxstation 4000/vlc has a size of 1 u or smilar J the vaxstation 4000/vlc has a greater cpu power than many vax4000 systems       	 > Regards0 >  > FC    I Adam Maulis              maulis@ludens elte hu         VMS system managertH  .......................................................................H  VMS Competence Center                             VMS Szakertoi KozpontH  Eotvos Lorand University                  Eotvos Lorand TudomanyegyetemH  Budapest, Hungary                                              BudapestH  =======================================================================   ------------------------------   Date: 2 Apr 2001 21:23:50 +0200s& From: Cthulhu <cthulhu@kadath.deep.it> Subject: Re: VAX in 1U) Message-ID: <9aajk6$15a$1@kadath.deep.it>a  * fabio_compaq@ep-bc.petrobras.com.br wrote:   > 1 (one)  R400X  
 What is this?D   	discoveringly,  	   Cthulhuk   -- e  G        Ph'nglui mglw'nafh Cthulhu http://www.rlyeh.it wgah'nagl fhtgan! # 		        <cthulhu at rlyeh dot it>d   ------------------------------  $ Date: Mon, 2 Apr 2001 23:32:42 +02004 From: "Peter Ljungberg" <peter.p.ljungberg@telia.se> Subject: Re: VAX in 1U, Message-ID: <9aar8i$i0t$1@news.han.telia.se>  8 I think my "VAXstation 4000 VLC" is in 1U, or isn't it ?   /P.Lj   6 <fabio_compaq@ep-bc.petrobras.com.br> wrote in messageF news:OFFA46A003.2E542EBB-ON03256A22.005E5BD6@ep-bc.petrobras.com.br...K > An interesting idea nowadays would be Compaq launching a new (definitive)pF > VAX / MicroVAX in a 1U cabinet to maintain the legacy (???) systems.G > I have three VAXes here which I cannot "shutdown" because they have as2 > legacy application. And they are ocupping space: >l > 2 (two) VAX 4000 > 1 (one) MicroVAX 3100v > 1 (one)  R400X >H >O >K	 > Regardsc >y > FC >    ------------------------------  % Date: Mon, 02 Apr 2001 23:25:04 +0000l) From: Christof Brass <brass@infopuls.com>o$ Subject: Re: VMS-Related: Affordable, Message-ID: <3AC90A50.B6200D72@infopuls.com>   "David J. Dachtera" wrote:  = Continue, at least for about one year if you can afford it. I : think the next 12 to 24 month will tell if there is a turn> around in the user base and the preception of what VMS is good= for. To reverse the trend low cost systems and students using  them will be essential.5   ------------------------------  % Date: Mon, 02 Apr 2001 21:47:52 +0100s+ From: "antonio.carlini" <arcarlini@iee.org> 2 Subject: Re: When might SYS$SCHDWK not wake me up?' Message-ID: <3AC8E578.D427A870@iee.org>n   Jan Vorbrueggen wrote: > 3 > Chris Scheers <chris@applied-synergy.com> writes:4 > L > > However, it is possible that your "process" code is calling some routineJ > > that internally does a non-blocking system call, then does a $HIBER toK > > wait for an AST.  In this case, if the code has been written correctly,!J > > the $HIBER will eat your pending wakeup, the code will realize that itJ > > shouldn't be awake yet, and it will go back $HIBER.  At some point theJ > > AST will fire, waking the code, which continues from the $HIBER.  ThenK > > "process" will continue until you get to your $HIBER, which will reallymK > > hibernate because the $SCHDWK has already fired.  (A routine that callse+ > > the C RTL "sleep" can also cause this.)c > M > This seems to me a very plausible explanation. It also means that doing thedL > $SCHDWK and $HIBER without processing in between should solve the problem.A > But using a repeating $SCHDWK is programming with a safety net.t  , I don't see why the AST routine is mentioned& above -- unless it does something like% schedule another wakeup or cancel ther- pending one, it won't affect anything. If you@+ were in a particular $HIBER before the AST,n! you'll still be in it afterwards.   & In addition a non-blocking system call* that then needs to wait for I/O completion+ would use either an AST and some signallingn+ mechanism or $SYNCH ... $HIBER is not a waye I would choose to do it.  ( It would be useful to see the details of% the loop that this process lives in -s$ without that there's a great deal of room for guesswork.)  & One thing that occurred to me is that + it is possible to intercept user mode calls ( to system services. Hijacking $HIBER and% keeping a log in a ring buffer of the $ last N PCs to invoke it might settle% the discussion ... if it's always the # main $HIBER then the above scenariot" is unlikely. Similarly, monitoring. calls to $CANWAK (not $CANTIM) may be useful, ( just in case a miscoded call is throwing away a $SCHDWK.      AntonioD   --     ---------------l- Antonio Carlini             arcarlini@iee.org    ------------------------------  % Date: Mon, 02 Apr 2001 14:55:05 -0400D- From: "Richard D. Piccard" <piccard@ohio.edu>7+ Subject: Re: [Q] Need Help on Backup Script ' Message-ID: <3AC8CB09.7050406@ohio.edu>P  H $ SUBMIT/NOPRINT/NOTIFY is my usual approach.  It amuses me to think of  the un-negated command as    $ SUBMIT/PRINT/TIFY    				RDP      David B Sneddon wrote:  5 > At 09:24 PM 21/03/01 -0500, Ram Rajadhyaksha wrote:  > M >> I am attempting to make a backup script that will backup 3 volumes. When I L >> commit it to batch (SUBMIT MYBACKUPSCRIPT.COM), it always fails after theH >> second volume for some reason. The first time was probably because itF >> encounted an open file and I had the ON ERROR set to branch. Other  >> times the0 >> job simply terminates and I have no idea why. >> nE >> Here's my script (drive names changed to protect the innocent :-):D >> 1- >> $ WRITE SYS$OUTPUT "Starting Full Backup."a >> $ SHOW TIME >> $ ALLOCATE MKAxxx:0 >> $ INIT MKAxxx: MYLABELr >> $ DEALLOCATE MKAxxx:" >> $ ! >> $ ! Mount Taped >> $ ! >> $ MOUNT/FOR MKAxxx:L >> $ BACKUP/NOREWIND/LOG/LABEL=MYLABEL DKA1:[000000...]*.*;* MKAxxx:DKA1.SAVL >> $ BACKUP/NOREWIND/LOG/LABEL=MYLABEL DKA2:[000000...]*.*;* MKAxxx:DKA2.SAVL >> $ BACKUP/NOREWIND/LOG/LABEL=MYLABEL DKA3:[000000...]*.*;* MKAxxx:DKA3.SAV >> $ DISMOUNT/UNLOAD MKAxxx:C >> $ SHOW TIME$OUTPUT "Full Backup Script Done. Check log to verifya >> completeness."g	 >> $ EXITh >> )L >> And the script terminates after the BACKUP command finishes on disk 2 at: >>  4 >> BACKUP-S-COPIED, COPIED DKA2:[000000]VOLSET.SYS;1 >> S: >> Disk 2 happens to be the live OpenVMS system disk, btw. >> = >> = >> Second question:  >> rM >> After this thing fails, the machine prints out the WHOLE log to the systemrI >> printer. When you backup 100000 files, this is not an environmentally 2 >> happyI >> thing. Is there a way to turn that damn feature off and have it simply=% >> record everything to a log file???= >> =
 >> Thanks. >> = >> --  >> Ram Rajadhyaksha= >> DLZ Corporation >> www.dlzcorp.com >  >  > I > In the DBS-DCL package (from the address below) is a GENERIC_BACKUP.COMeN > procedure which will do what you want.  It will locate a tape device, unlessJ > you define one for it, it will find all your disks (assuming you want to9 > backup all of them) and do an image backup of each one.=I > (You will also need the DBS-LIBRARY ZIP file for the program that finds-I > all the disks and creates the required logicals.  We use this on all ofaF > our systems here.  You can also specify multiple tape devices shouldG > the need arise.  It handles all the necessary errors and creates listi3 > files of each save set and keeps a history file.)>/ > Just do @GENERIC_BACKUP and it does the rest.o > Hope this helps. >  >  > 
 > Regards, > Dave.w     -- >B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  % Date: Mon, 02 Apr 2001 16:39:48 -0600s1 From: "David J. Dachtera" <djesys.nospam@fsi.net>b+ Subject: Re: [Q] Need Help on Backup Script ' Message-ID: <3AC8FFB4.2AF4B3BE@fsi.net>y   "Richard D. Piccard" wrote:r > I > $ SUBMIT/NOPRINT/NOTIFY is my usual approach.  It amuses me to think off > the un-negated command ash >  > $ SUBMIT/PRINT/TIFYn   (Tongue firmly in cheek)  H I would say that the default behavior, /TIFY is to not let you know whenD a job has completed. Only the negated form - /NOTIFY - is available.1 I've never tried the double-negative - /NONOTIFY.i   :-)c   -- d David J. DachteraC dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.a   ------------------------------  % Date: Mon, 02 Apr 2001 15:01:53 -0400t- From: "Richard D. Piccard" <piccard@ohio.edu> 2 Subject: Re: [SOAPBOX] sys$specific or sys$common?' Message-ID: <3AC8CCA1.5040805@ohio.edu>s  I Just for the record, that was true of **all** [Micro]VMS 4.7 --> VMS 5.0  A upgrades where the V4.7 system was not already in a cluster.  It eB aggravated me enough that I wrote an article about it:  "VMS V5.0 E Upgrade: 'Gotcha' for NON-cluster Systems."  DECUS U.S. Chapter SIGs yB Newsletters, Vol. 4, No. 1 (October, 1988), pages VAX-8 and VAX-9.   				RDPc     JF Mezei wrote:l  P > When I upgrated from MicroVMS 4.7 to VMS 5.0, all of a sudden, I was presentedM > with that really weird directory structure with SYS$COMMON and SYS$SPECIFICl" > which did not exist in MicroVMS. > % > Everything was put into SYS$COMMON.d > P > So, it is clear to me that even outside of a cluster environment, files should5 > be put in SYS$COMMON unless there is a good reason.n     --  B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------   End of INFO-VAX 2001.186 ************************