INFO-VAX Fri, 20 Apr 2007 Volume 2007 : Issue 215 Contents: a complete free forum just for you Advanced Server and Active Directory RE: Advanced Server and Active Directory Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? Re: C++ Garbage Collector on VMS? RE: C++ Garbage Collector on VMS? Re: can you handle this? Re: can you handle this? Re: Dead clock, NTP and 1000 seconds Re: ES40 to ES45 Upgrade only $4995 ! If you live in California, get out now! Re: If you live in California, get out now! Re: If you live in California, get out now! Re: If you live in California, get out now! Re: If you live in California, get out now! Re: If you live in California, get out now! Re: If you live in California, get out now! Re: If you live in California, get out now! Re: If you live in California, get out now! Re: If you live in California, get out now! If you live in California, get out now! (Part 2) Re: If you live in California, get out now! (Part 2) Re: If you live in California, get out now! (Part 2) Re: If you live in California, get out now! (Part 2) Intel tips more details on Itanium roadmap Re: Intel tips more details on Itanium roadmap IT'S THAT TIME AGAIN - FREE DS10L OFFER Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER market resources Netbackup 5 VMS Client and the VMS backup image switch. Re: Netbackup 5 VMS Client and the VMS backup image switch. Re: Now I've seen everything Re: Now I've seen everything Re: OT: OpenVMS Re: OT: OpenVMS Re: OT: OpenVMS Re: OT: OpenVMS Re: OT: OpenVMS Re: OT: OpenVMS Re: OT: OpenVMS Re: OT: OpenVMS Re: OT: OpenVMS Re: Problem with BACKUP/LIST/JOU/SINCE/SELECT Re: Problem with BACKUP/LIST/JOU/SINCE/SELECT Re: Problem with BACKUP/LIST/JOU/SINCE/SELECT Re: Problem with BACKUP/LIST/JOU/SINCE/SELECT Re: Process Software: SocketException: function not implemented Re: Process Software: SocketException: function not implemented Re: Process Software: SocketException: function not implemented Running NFS client on Open VMS Re: Running NFS client on Open VMS Re: Running NFS client on Open VMS Re: Running NFS client on Open VMS Re: SSM ? Re: SSM ? [OpenVMS V8.3] MMOV ? Re: [OpenVMS V8.3] MMOV ? Re: [OpenVMS V8.3] MMOV ? Re: [OpenVMS V8.3] MMOV ? Re: [OpenVMS V8.3] MMOV ? [OpenVMS V8.3] SSM ? ---------------------------------------------------------------------- Date: 18 Apr 2007 18:45:44 -0700 From: ann Subject: a complete free forum just for you Message-ID: <1176947144.453445.306230@b75g2000hsg.googlegroups.com> Free Unlimited Storage Free Unlimited Bandwidth Free Free Subdomain Free More Skins Free No annoying pop-ups Free 99% Uptime Free Daily Backups Free Fast, Friendly Tech Support So,what's that? That's our free forum named http://support.forumgogo.com. We have the human_based management also provide you human_based forum service. You needn't,we won't ! This is a great site that is completely free, and can make some cool forums, here's an example of one, you can make your own topic, polls, and replies etc, and you can have an accout for every member. This next link is an example of a forum made from invisionfree. http://support.forumgogo.com ------------------------------ Date: Thu, 19 Apr 2007 18:18:22 -0500 From: John Subject: Advanced Server and Active Directory Message-ID: <4627F8BE.5030605@tx.rr.com> I am looking to install Advanced Server V7.3-B (or whatever is the latest version) and integrate into an Active Directory. Any suggestions as to how far should I integrate the Advanced Server into the Active Directory - keep it seperate, make it a part, etc.? ------------------------------ Date: Thu, 19 Apr 2007 22:46:02 -0400 From: "Main, Kerry" Subject: RE: Advanced Server and Active Directory Message-ID: > -----Original Message----- > From: John [mailto:norad869@tx.rr.com] > Sent: April 19, 2007 7:18 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Advanced Server and Active Directory >=20 > I am looking to install Advanced Server V7.3-B (or whatever is the > latest version) and integrate into an Active Directory. >=20 > Any suggestions as to how far should I integrate the Advanced Server > into the Active Directory - keep it seperate, make it a part, etc.? >=20 What is it that you are trying to do? - share files or share user authentication or ?? If it is user authentication integration, then you might want to consider the following product from Process Software: http://www.process.com/VMSauth/index.html=20 From the data sheet on the above page: " * Works with any LDAP v3 compliant server including Microsoft Active Directory and OpenLDAP" Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20 OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Wed, 18 Apr 2007 19:35:07 +0100 From: "John Wallace" Subject: Re: C++ Garbage Collector on VMS? Message-ID: <462664c2$0$8736$ed2619ec@ptn-nntp-reader02.plus.net> "John Santos" wrote in message news:MPG.208fdc054be334ee98974e@news.bellatlantic.net... > Garbage collectors do *not* (at least the ones I've ever dealt with) > find stuff in memory that was allocated, is no longer in use, but > wasn't deallocated. > Garbage collectors are *not* a cure for memory leaks. They are not > a cure for sloppy programming. > > What garbage collectors do is re-allocate memory, compressing > out the no-longer-used chunks, and pushing all the still-in-use > chunks together, in order to consolidate the free space. (Think > of a disk defragger, but intended for a memory pool rather than > for a disk file structure.) > Did you actually follow the link provided earlier, to http://www.hpl.hp.com/personal/Hans_Boehm/gc/ ? When you do, you will find it says: "The Boehm-Demers-Weiser conservative garbage collector can be used as a garbage collecting replacement for C malloc or C++ new. It allows you to allocate memory basically as you normally would, without explicitly deallocating memory that is no longer useful. The collector automatically recycles memory when it determines that it can no longer be otherwise accessed. A simple example of such a use is given here. ... Alternatively, the garbage collector may be used as a leak detector for C or C++ programs, though that is not its primary goal. " Is that the same kind of garbage collector as the "memory contiguity creator" to which you refer (which clearly has a role, but I'm just not sure the Boehm GC is the same thing)? Regards John ------------------------------ Date: Wed, 18 Apr 2007 19:25:50 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: C++ Garbage Collector on VMS? Message-ID: <4626a8d3$0$90265$14726298@news.sunsite.dk> Bob Koehler wrote: > In article , "Main, Kerry" writes: >> While the entire Java world uses garbage collection, I have never found >> anyone (yet) who could explain to me why a properly designed and >> debugged application needs to have a system level process running in the >> background cleaning up allocated memory and other app cleanup stuff on >> the system. > > It doesn't. That's just the Java approach. It's par tof why Java > can't be deterministic, even if compiled to native instructions using > the gnu Java compiler. > > But a different approach could have been to automatically free > memory, always at the same time, just as if one had written the > application in C and put the malloc and free calls in. Java seems > able to automatically do the equivalent of malloc/new at the right > time but not willing to do the same for free/delete. A different > lanuage/compiler/Java-implementation could have done it > determinstically. > > This would be easier for a language like Java, where String is an > instrinsic data type, than for C++ which can't prevent the use > of char*, but it also applies to objet creation and deletion when > strings aren't included. > > If Java can keep track of objects well enough to guarantee to the > garbage collector that it's OK to free them, then it could free them > at the point the guarantee is reached without a separate garbage > collector. That would at least have the benefit of making it > deterministic. I believe most systems that does not do GC the Java/.NET way uses reference counting (I know that COM and PHP use that). With reference counting it is easy to do it deterministic, but you can easily get memory leaks. Arne ------------------------------ Date: Wed, 18 Apr 2007 19:39:30 -0400 From: =?ISO-8859-15?Q?Arne_Vajh=F8j?= Subject: Re: C++ Garbage Collector on VMS? Message-ID: <4626ac07$0$90269$14726298@news.sunsite.dk> John Santos wrote: > I suppose that observation would be relevant if it had anything to do > with what a garbage collector does. > > Garbage collectors do *not* (at least the ones I've ever dealt with) > find stuff in memory that was allocated, is no longer in use, but > wasn't deallocated. (Maybe in LISP they do, it's been ages since > I dealt with LISP, but that's the exception, not the rule.) > Garbage collectors are *not* a cure for memory leaks. They are not > a cure for sloppy programming. > > What garbage collectors do is re-allocate memory, compressing > out the no-longer-used chunks, and pushing all the still-in-use > chunks together, in order to consolidate the free space. (Think > of a disk defragger, but intended for a memory pool rather than > for a disk file structure.) Java, .NET, PHP etc. uses garbage collection in the sense Kerry described (to get automatically deallocate memory). Arne ------------------------------ Date: Wed, 18 Apr 2007 16:44:45 -0700 From: "Tom Linden" Subject: Re: C++ Garbage Collector on VMS? Message-ID: On Wed, 18 Apr 2007 16:25:50 -0700, Arne Vajhøj wrote: > Bob Koehler wrote: >> In article >> , >> "Main, Kerry" writes: >>> While the entire Java world uses garbage collection, I have never found >>> anyone (yet) who could explain to me why a properly designed and >>> debugged application needs to have a system level process running in >>> the >>> background cleaning up allocated memory and other app cleanup stuff on >>> the system. >> It doesn't. That's just the Java approach. It's par tof why Java >> can't be deterministic, even if compiled to native instructions using >> the gnu Java compiler. >> But a different approach could have been to automatically free >> memory, always at the same time, just as if one had written the >> application in C and put the malloc and free calls in. Java seems >> able to automatically do the equivalent of malloc/new at the right >> time but not willing to do the same for free/delete. A different >> lanuage/compiler/Java-implementation could have done it >> determinstically. >> This would be easier for a language like Java, where String is an >> instrinsic data type, than for C++ which can't prevent the use of >> char*, but it also applies to objet creation and deletion when >> strings aren't included. >> If Java can keep track of objects well enough to guarantee to the >> garbage collector that it's OK to free them, then it could free them >> at the point the guarantee is reached without a separate garbage >> collector. That would at least have the benefit of making it >> deterministic. > > I believe most systems that does not do GC the Java/.NET way uses > reference counting (I know that COM and PHP use that). > > With reference counting it is easy to do it deterministic, but > you can easily get memory leaks. I believe the first use of ref counts was in a paper by Freiburghouse on the Multics PL/I compiler, ca. 1967. Of course, it didn't have many leaks :-) > > Arne -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Wed, 18 Apr 2007 20:18:57 -0400 From: =?ISO-8859-15?Q?Arne_Vajh=F8j?= Subject: Re: C++ Garbage Collector on VMS? Message-ID: <4626b544$0$90264$14726298@news.sunsite.dk> Tom Linden wrote: > I believe the first use of ref counts was in a paper by Freiburghouse on > the Multics PL/I compiler, ca. 1967. Of course, it didn't have many > leaks :-) If it only uses reference counting it will be definition have memory leaks. Some reference counting systems supplement reference counting with a garbage collector to get rid of the special cases. Arne ------------------------------ Date: Thu, 19 Apr 2007 08:44:20 +0800 From: "Richard Maher" Subject: Re: C++ Garbage Collector on VMS? Message-ID: Hi, As an aside, VMS Personae use reference counting so that when you delete them they don't actually die until everyone (including all outstanding $qio i/o) has ceased interest in them. I can't remember looking at that bit of the source listings but the I can only assume it is the $qio completion code (and the completion code of *all* other utilities that could possibly increment the reference counts) that checks to see if final clean-up of the persona structures is required. A few extra instructions for *every* i/o? Why not? Isn't CPU fast and cheap now? Surely VMS engineering could've come up with a Valkyrie Persona Grabber process but instead chose to do the job properly? Could one not also apply the same "Procedural languages are past their sell-buy date and cannot meet modern business and RAD requirements" arguments of ten years ago to the almost ancient Javas and C++s of today? Didn't Garbage Collectors come out because these languages were just too bloody SLOW to clean-up after themselves in-line? Isn't it like Rdb's Row-Ca$h where you hide all the real work and overhead behind the smoke and mirrors of another process/thread? As someone else has already pointed out, if these GCs are so artificially bloody intelligent and can workout what's live and what's not through counting, or tracing, or waving a dead chicken over their head, then why doesn't the User's thread (in the abscence of timely coder instruction) clean it up there and then? Anyway, I say over-engineering! Maybe we need an Object Working-Set for each class? Keep 3 A-blobs around 'cos that's what this run-unit normally needs but run up 10 B-widget constructors to begin with. I know, let's take RMS out of the process and put it in a seperate server, let's tell Rdb to go the 'Orrible route and pass the SQL to a seperate server process, Let's throw VMS out the door all together :-( I can understand that all of "those people" who program in Certain languages, are used to having stack-unwinding save their arses when they exit a routine (and who simply cannot fathom what I am talking about when I refer to *compile-time* initialization of local variables) are clearly very grateful that some nanny-state is there to blow their little noses and tuck their shirts in! But the rest of us have already got a heap-manager and it's call lib$*_vm. Cheers Richard Maher PS. Also, I personally don't look upon that fact that you don't have access to a pointer in these languages as a bonus. But if there are no pointers then your garbage can't be pointed at I suppose. PPS. Why hasn't VMS engineering come up with an EXEC-mode-safe version of lib$*_vm yet??? "Exciting new functionality in 8.3" my arse! Though I bet there was much adolescent giggling around the coffee urn at the naming of f$cunits :-( I put forward a reasonable, easy to retrofit in existing lib$ API, and low impact, solution for 2 heaps inner and outer (the 1:mode looked too hard/costly) and no one was able to debunk it, and yet no one seems interested in pursuing it. Is there any "real" VMS engineer that would be willing to look at this suggestion and then pay me the respect to justify any functional/performance/security reasons why it cannot be implemented? Shouldn't take too long should it? "Tom Linden" wrote in message news:op.tqz6wvsitte90l@hyrrokkin... > On Wed, 18 Apr 2007 16:25:50 -0700, Arne Vajhøj wrote: > > > Bob Koehler wrote: > >> In article > >> , > >> "Main, Kerry" writes: > >>> While the entire Java world uses garbage collection, I have never found > >>> anyone (yet) who could explain to me why a properly designed and > >>> debugged application needs to have a system level process running in > >>> the > >>> background cleaning up allocated memory and other app cleanup stuff on > >>> the system. > >> It doesn't. That's just the Java approach. It's par tof why Java > >> can't be deterministic, even if compiled to native instructions using > >> the gnu Java compiler. > >> But a different approach could have been to automatically free > >> memory, always at the same time, just as if one had written the > >> application in C and put the malloc and free calls in. Java seems > >> able to automatically do the equivalent of malloc/new at the right > >> time but not willing to do the same for free/delete. A different > >> lanuage/compiler/Java-implementation could have done it > >> determinstically. > >> This would be easier for a language like Java, where String is an > >> instrinsic data type, than for C++ which can't prevent the use of > >> char*, but it also applies to objet creation and deletion when > >> strings aren't included. > >> If Java can keep track of objects well enough to guarantee to the > >> garbage collector that it's OK to free them, then it could free them > >> at the point the guarantee is reached without a separate garbage > >> collector. That would at least have the benefit of making it > >> deterministic. > > > > I believe most systems that does not do GC the Java/.NET way uses > > reference counting (I know that COM and PHP use that). > > > > With reference counting it is easy to do it deterministic, but > > you can easily get memory leaks. > > I believe the first use of ref counts was in a paper by Freiburghouse on > the Multics PL/I compiler, ca. 1967. Of course, it didn't have many > leaks :-) > > > > Arne > > > > -- > Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Wed, 18 Apr 2007 20:44:02 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: C++ Garbage Collector on VMS? Message-ID: <4626bb27$0$90271$14726298@news.sunsite.dk> Main, Kerry wrote: >> I can not see why GC should be a problem for mission critical apps. > > Well, as I recall, one typically has very little control of when GC > runs, hence a system process that kicks in during peak time for > transactions does not seem like something you would want to risk. There are running a lot of mission critical apps on WAS and WL. It does not take that long time to do a GC. There are different apps, different systems, different java version and different GC options, but as an example: a GC of young generation on my Pentium P4 with a standard SUN Java with no special GC options and a very simple app uses 0.0001 seconds. > It also seems to me that a solution is certainly not a good one when > memory is not properly allocated and de-allocated by design and > depends on system processes to compensate for poor programming > practices. It is very good system design to consider the possibility that programmers make mistakes. Arne ------------------------------ Date: Wed, 18 Apr 2007 20:24:39 -0500 From: David J Dachtera Subject: Re: C++ Garbage Collector on VMS? Message-ID: <4626C4D7.C3FC07CA@spam.comcast.net> Johann 'Myrkraverk' Oskarsson wrote: > > David J. Dachtera writes: > > > Johann 'Myrkraverk' Oskarsson wrote: > >> > >> David J. Dachtera writes: > >> > >> > I have to ask: Why does a programming language (other than an > >> > interpretive environment) need a "garbage collector"??!! > >> > >> Maybe I'm implementing one in C++ and don't want to write my own > >> garbage collector? Boehm GC, afaict, is very popular among such > >> projects. > > > > ...which illustrates the difference in paradigm. > > > > You're looking for something you think you need. > > > > I'm asking why you need something that should never be needed. If a > > compiled/linked program still needs garbage collection, either the > > run-time, the program or both are seriously deficient. > > You assume a GC should never be needed. I don't know what kind of > applications you're thinking of, but as I'm creating a lisp engine, > and the language is designed for a GC (and I believe the first GC was > written for the first lisp engine, but I can be wrong). There are > several run time environments that do GC that work on VMS. Java is > one, Oberon is another. There are probably more that I can't name > righ now? Are those run-time environments *wrong*? Or do you just > want them gone from VMS? > > This is a toy project right now, and doesn't deserve the effort of a > custom built GC. I'd like it to work on VMS, but the lack of aparent > VMS support of the Boehm GC has me stomped. > > http://www.hpl.hp.com/personal/Hans_Boehm/gc/ > > The GC needs some system dependent details, such as register contents, > stack and heap locations, and so forth to function properly. I don't > know enough about VMS to supply and port Boehm right now. If, as another poster suggested, what you're actually describing is a "memory pool defragmenter", VMS does that on its own since about V5.4-3, about the time the VAX and Alpha code streams were merged in an effort to achieve feature parity between the two. VMS's memory management does this without the need for "garbage collectors". That said, if this is an environment like, for example, MUMPS (an "operating environment" running under a host o.s.), then the GC will be running within that environment, and not operating on the o.s.'s memory pools outside of that environment. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Wed, 18 Apr 2007 18:45:18 -0700 From: "Tom Linden" Subject: Re: C++ Garbage Collector on VMS? Message-ID: On Wed, 18 Apr 2007 17:44:20 -0700, Richard Maher wrote: > Hi, > > As an aside, VMS Personae use reference counting so that when you delete > them they don't actually die until everyone (including all outstanding > $qio > i/o) has ceased interest in them. I can't remember looking at that bit of > the source listings but the I can only assume it is the $qio completion > code > (and the completion code of *all* other utilities that could possibly > increment the reference counts) that checks to see if final clean-up of > the > persona structures is required. > > A few extra instructions for *every* i/o? Why not? Isn't CPU fast and > cheap > now? Surely VMS engineering could've come up with a Valkyrie Persona > Grabber > process but instead chose to do the job properly? Could one not also > apply > the same "Procedural languages are past their sell-buy date and cannot > meet > modern business and RAD requirements" arguments of ten years ago to the > almost ancient Javas and C++s of today? Didn't Garbage Collectors come > out > because these languages were just too bloody SLOW to clean-up after > themselves in-line? It wasn't slowness, but the underlying model that was faulty. Isn't it like Rdb's Row-Ca$h where you hide all the > real > work and overhead behind the smoke and mirrors of another process/thread? > > As someone else has already pointed out, if these GCs are so artificially > bloody intelligent and can workout what's live and what's not through > counting, or tracing, or waving a dead chicken over their head, then why > doesn't the User's thread (in the abscence of timely coder instruction) > clean it up there and then? Anyway, I say over-engineering! Maybe we > need an > Object Working-Set for each class? Keep 3 A-blobs around 'cos that's what > this run-unit normally needs but run up 10 B-widget constructors to begin > with. You get what you pay for, and I mean that not in $s but effort expended. > > I know, let's take RMS out of the process and put it in a seperate > server, > let's tell Rdb to go the 'Orrible route and pass the SQL to a seperate > server process, Let's throw VMS out the door all together :-( > > I can understand that all of "those people" who program in Certain > languages, are used to having stack-unwinding save their arses when they > exit a routine (and who simply cannot fathom what I am talking about > when I > refer to *compile-time* initialization of local variables) are clearly > very > grateful that some nanny-state is there to blow their little noses and > tuck > their shirts in! But the rest of us have already got a heap-manager and > it's > call lib$*_vm. Richard, you need to learn PL/I and give up that Cobol crap. It just doesn't get better. Compile-time init is for static, you can't write reentrant code with that concept. (Actually you can if you have fork and spawn, but that is ugly and error prone. > > Cheers Richard Maher > > PS. Also, I personally don't look upon that fact that you don't have > access > to a pointer in these languages as a bonus. But if there are no pointers > then your garbage can't be pointed at I suppose. > > PPS. Why hasn't VMS engineering come up with an EXEC-mode-safe version of > lib$*_vm yet??? "Exciting new functionality in 8.3" my arse! Though I bet > there was > much adolescent giggling around the coffee urn at the naming of f$cunits > :-( > > I put forward a reasonable, easy to retrofit in existing lib$ API, and > low > impact, solution for 2 heaps inner and outer (the 1:mode looked too > hard/costly) and no one was able to debunk it, and yet no one seems > interested in pursuing it. Is there any "real" VMS engineer that would be > willing to look at this suggestion and then pay me the respect to justify > any functional/performance/security reasons why it cannot be implemented? > Shouldn't take too long should it? > > "Tom Linden" wrote in message > news:op.tqz6wvsitte90l@hyrrokkin... >> On Wed, 18 Apr 2007 16:25:50 -0700, Arne Vajhøj wrote: >> >> > Bob Koehler wrote: >> >> In article >> >> > , >> >> "Main, Kerry" writes: >> >>> While the entire Java world uses garbage collection, I have never > found >> >>> anyone (yet) who could explain to me why a properly designed and >> >>> debugged application needs to have a system level process running in >> >>> the >> >>> background cleaning up allocated memory and other app cleanup stuff >> on >> >>> the system. >> >> It doesn't. That's just the Java approach. It's par tof why >> Java >> >> can't be deterministic, even if compiled to native instructions > using >> >> the gnu Java compiler. >> >> But a different approach could have been to automatically free >> >> memory, always at the same time, just as if one had written the >> >> application in C and put the malloc and free calls in. Java seems >> >> able to automatically do the equivalent of malloc/new at the right >> >> time but not willing to do the same for free/delete. A different >> >> lanuage/compiler/Java-implementation could have done it >> >> determinstically. >> >> This would be easier for a language like Java, where String is >> an >> >> instrinsic data type, than for C++ which can't prevent the use >> of >> >> char*, but it also applies to objet creation and deletion when >> >> strings aren't included. >> >> If Java can keep track of objects well enough to guarantee to the >> >> garbage collector that it's OK to free them, then it could free >> them >> >> at the point the guarantee is reached without a separate garbage >> >> collector. That would at least have the benefit of making it >> >> deterministic. >> > >> > I believe most systems that does not do GC the Java/.NET way uses >> > reference counting (I know that COM and PHP use that). >> > >> > With reference counting it is easy to do it deterministic, but >> > you can easily get memory leaks. >> >> I believe the first use of ref counts was in a paper by Freiburghouse on >> the Multics PL/I compiler, ca. 1967. Of course, it didn't have many >> leaks :-) >> > >> > Arne >> >> >> >> -- >> Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ > > -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Wed, 18 Apr 2007 19:06:32 -0700 From: "Tom Linden" Subject: Re: C++ Garbage Collector on VMS? Message-ID: On Wed, 18 Apr 2007 17:18:57 -0700, Arne Vajhøj wrote: > Tom Linden wrote: >> I believe the first use of ref counts was in a paper by Freiburghouse on >> the Multics PL/I compiler, ca. 1967. Of course, it didn't have many >> leaks :-) > > If it only uses reference counting it will be definition > have memory leaks. You have a different paradigm in mind, I will find link and post it. This was for a compiler, not interpreter so different rules. When refernece count drops to 0 memory is freed > > Some reference counting systems supplement reference counting > with a garbage collector to get rid of the special cases. > > Arne > -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: 18 Apr 2007 22:42:36 -0500 From: wb8tyw@qsl.network (John E. Malmberg) Subject: Re: C++ Garbage Collector on VMS? Message-ID: <1HHsEZpaNgq4@eisner.encompasserve.org> In article , Johann 'Myrkraverk' Oskarsson writes: > > You assume a GC should never be needed. I don't know what kind of > applications you're thinking of, but as I'm creating a lisp engine, > and the language is designed for a GC (and I believe the first GC was > written for the first lisp engine, but I can be wrong). There are > several run time environments that do GC that work on VMS. Java is > one, Oberon is another. There are probably more that I can't name > righ now? Are those run-time environments *wrong*? Or do you just > want them gone from VMS? There are many types of garbage collectors. On an interpreted environment where data is handled by descriptors, a garbage collector can totally reorganize storage to consolidate free space. Such GC implementations can be tightly coupled to the program that uses them, however, I could see a generic implementation. For one that figures out when an allocation of storage is no longer being referenced, in general that is hard to do. For languages implemented to use descriptors to reference all data items, a garbage collector could compare it's list of allocated memory with a list of known descriptors. Where it could help by freeing them, or by flagging a memory leak. The above can be efficiently and portably implemented. > This is a toy project right now, and doesn't deserve the effort of a > custom built GC. I'd like it to work on VMS, but the lack of aparent > VMS support of the Boehm GC has me stomped. > > http://www.hpl.hp.com/personal/Hans_Boehm/gc/ > > The GC needs some system dependent details, such as register contents, > stack and heap locations, and so forth to function properly. I don't > know enough about VMS to supply and port Boehm right now. According to the above link, the Boehm GC appears to work by scanning the stack, registers, and program memory for pointers to memory allocated from the heap, and to pointers in the heap, and then comparing it to the list of allocated pointers of memory that the GC knows it allocated from the heap. If there is memory allocated, but it can not find a reference to it in either the allocated heap, or in active program space, it knows to deallocate it. Some of other references seem to indicate that this is very useful for detecting memory leaks. I do not know how it avoids being fooled by stale pointers in memory that has not been re-initialized. This type of garbage collection looks to be very inefficient for use outside of debugging programs. -John wb8tyw@qsl.network Personal Opinion Only ------------------------------ Date: Thu, 19 Apr 2007 14:33:35 +0800 From: "Richard Maher" Subject: Re: C++ Garbage Collector on VMS? Message-ID: Hi, > Richard, you need to learn PL/I and give up that Cobol crap. It just > doesn't > get better. Compile-time init is for static, you can't write reentrant > code > with that concept. (Actually you can if you have fork and spawn, but that > is > ugly and error prone. Go to sleep JF! If you still feel like baiting me in the morning then re-assemble your goal posts somewhere we all (and GPS!) can see them. (I am now actually a Java/JavaScript/HTML convert! Who cares about interpreters and bollocks garbage collectors on the CPU abundant *client*? Just keep your damned filth away from my nice clean VMS) Reentrant? Of course! AST-reentrant? Yes, depending on alignment and coding. True of all languages no? Recursive? No. But that's why I like $dclast (Which is what I've called my JS setTimeout(x,0) function) *I* know there are things Cobol can't do (nor should it) and for these things God (and John Reagan) gave us MACRO-32! The right tool for the job eh?. If you like PL/I then more power to ya; If you like C then good for you (Just stop pouring *our* VMS license dollars down *your* toilets!) It was like what I was wanting to respond to Arne in a different thread (but I'm gettin' old and find it difficult to remember now) I keep handing him the 1 Wood for the 450m par 5 but he insists on teeing of with his HTTP putter. It's good for the short game but I reakon your gonna get a few people playin' through! (Or is the analogy to the long-game better? Either way, you got me :-( Bastards! :-) Cheers Richard Maher PS. If you insist on me picking up the gauntlet just let me know whether it's the VAX, Alpha or *Itanium* field that I should turn up at. PPS. Is it only me that thinks the operating system should be the OS and the Compiler provide the Executable Image? Every language has it's own VM and memory management. That's what comes from Cromwellian language centric OSs I suppose? PPPS> A couple of months ago I had the pleasure of maintaining a production BASIC program that used recursion to loop through till EOF. Is this common with languages of this ilk? God bless small files I suppose. "Tom Linden" wrote in message news:op.tq0chsb0tte90l@hyrrokkin... > On Wed, 18 Apr 2007 17:44:20 -0700, Richard Maher > wrote: > > > Hi, > > > > As an aside, VMS Personae use reference counting so that when you delete > > them they don't actually die until everyone (including all outstanding > > $qio > > i/o) has ceased interest in them. I can't remember looking at that bit of > > the source listings but the I can only assume it is the $qio completion > > code > > (and the completion code of *all* other utilities that could possibly > > increment the reference counts) that checks to see if final clean-up of > > the > > persona structures is required. > > > > A few extra instructions for *every* i/o? Why not? Isn't CPU fast and > > cheap > > now? Surely VMS engineering could've come up with a Valkyrie Persona > > Grabber > > process but instead chose to do the job properly? Could one not also > > apply > > the same "Procedural languages are past their sell-buy date and cannot > > meet > > modern business and RAD requirements" arguments of ten years ago to the > > almost ancient Javas and C++s of today? Didn't Garbage Collectors come > > out > > because these languages were just too bloody SLOW to clean-up after > > themselves in-line? > > It wasn't slowness, but the underlying model that was faulty. > > Isn't it like Rdb's Row-Ca$h where you hide all the > > real > > work and overhead behind the smoke and mirrors of another process/thread? > > > > As someone else has already pointed out, if these GCs are so artificially > > bloody intelligent and can workout what's live and what's not through > > counting, or tracing, or waving a dead chicken over their head, then why > > doesn't the User's thread (in the abscence of timely coder instruction) > > clean it up there and then? Anyway, I say over-engineering! Maybe we > > need an > > Object Working-Set for each class? Keep 3 A-blobs around 'cos that's what > > this run-unit normally needs but run up 10 B-widget constructors to begin > > with. > > You get what you pay for, and I mean that not in $s but effort expended. > > > > > I know, let's take RMS out of the process and put it in a seperate > > server, > > let's tell Rdb to go the 'Orrible route and pass the SQL to a seperate > > server process, Let's throw VMS out the door all together :-( > > > > I can understand that all of "those people" who program in Certain > > languages, are used to having stack-unwinding save their arses when they > > exit a routine (and who simply cannot fathom what I am talking about > > when I > > refer to *compile-time* initialization of local variables) are clearly > > very > > grateful that some nanny-state is there to blow their little noses and > > tuck > > their shirts in! But the rest of us have already got a heap-manager and > > it's > > call lib$*_vm. > > Richard, you need to learn PL/I and give up that Cobol crap. It just > doesn't > get better. Compile-time init is for static, you can't write reentrant > code > with that concept. (Actually you can if you have fork and spawn, but that > is > ugly and error prone. > > > > Cheers Richard Maher > > > > PS. Also, I personally don't look upon that fact that you don't have > > access > > to a pointer in these languages as a bonus. But if there are no pointers > > then your garbage can't be pointed at I suppose. > > > > PPS. Why hasn't VMS engineering come up with an EXEC-mode-safe version of > > lib$*_vm yet??? "Exciting new functionality in 8.3" my arse! Though I bet > > there was > > much adolescent giggling around the coffee urn at the naming of f$cunits > > :-( > > > > I put forward a reasonable, easy to retrofit in existing lib$ API, and > > low > > impact, solution for 2 heaps inner and outer (the 1:mode looked too > > hard/costly) and no one was able to debunk it, and yet no one seems > > interested in pursuing it. Is there any "real" VMS engineer that would be > > willing to look at this suggestion and then pay me the respect to justify > > any functional/performance/security reasons why it cannot be implemented? > > Shouldn't take too long should it? > > > > "Tom Linden" wrote in message > > news:op.tqz6wvsitte90l@hyrrokkin... > >> On Wed, 18 Apr 2007 16:25:50 -0700, Arne Vajhøj wrote: > >> > >> > Bob Koehler wrote: > >> >> In article > >> >> > > , > >> >> "Main, Kerry" writes: > >> >>> While the entire Java world uses garbage collection, I have never > > found > >> >>> anyone (yet) who could explain to me why a properly designed and > >> >>> debugged application needs to have a system level process running in > >> >>> the > >> >>> background cleaning up allocated memory and other app cleanup stuff > >> on > >> >>> the system. > >> >> It doesn't. That's just the Java approach. It's par tof why > >> Java > >> >> can't be deterministic, even if compiled to native instructions > > using > >> >> the gnu Java compiler. > >> >> But a different approach could have been to automatically free > >> >> memory, always at the same time, just as if one had written the > >> >> application in C and put the malloc and free calls in. Java seems > >> >> able to automatically do the equivalent of malloc/new at the right > >> >> time but not willing to do the same for free/delete. A different > >> >> lanuage/compiler/Java-implementation could have done it > >> >> determinstically. > >> >> This would be easier for a language like Java, where String is > >> an > >> >> instrinsic data type, than for C++ which can't prevent the use > >> of > >> >> char*, but it also applies to objet creation and deletion when > >> >> strings aren't included. > >> >> If Java can keep track of objects well enough to guarantee to the > >> >> garbage collector that it's OK to free them, then it could free > >> them > >> >> at the point the guarantee is reached without a separate garbage > >> >> collector. That would at least have the benefit of making it > >> >> deterministic. > >> > > >> > I believe most systems that does not do GC the Java/.NET way uses > >> > reference counting (I know that COM and PHP use that). > >> > > >> > With reference counting it is easy to do it deterministic, but > >> > you can easily get memory leaks. > >> > >> I believe the first use of ref counts was in a paper by Freiburghouse on > >> the Multics PL/I compiler, ca. 1967. Of course, it didn't have many > >> leaks :-) > >> > > >> > Arne > >> > >> > >> > >> -- > >> Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ > > > > > > > > -- > Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: 19 Apr 2007 09:18:51 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: C++ Garbage Collector on VMS? Message-ID: In article <4626b544$0$90264$14726298@news.sunsite.dk>, =?ISO-8859-15?Q?Arne_Vajh=F8j?= writes: > Tom Linden wrote: >> I believe the first use of ref counts was in a paper by Freiburghouse on >> the Multics PL/I compiler, ca. 1967. Of course, it didn't have many >> leaks :-) > > If it only uses reference counting it will be definition > have memory leaks. > > Some reference counting systems supplement reference counting > with a garbage collector to get rid of the special cases. What technique other than reference counting do these GC use to guarantee the deletes are safe? ------------------------------ Date: Thu, 19 Apr 2007 16:07:48 +0000 From: Johann 'Myrkraverk' Oskarsson Subject: Re: C++ Garbage Collector on VMS? Message-ID: Bob Koehler writes: > In article <4626b544$0$90264$14726298@news.sunsite.dk>, > =?ISO-8859-15?Q?Arne_Vajh=F8j?= writes: >> Tom Linden wrote: >>> I believe the first use of ref counts was in a paper by >>> Freiburghouse on the Multics PL/I compiler, ca. 1967. Of course, >>> it didn't have many leaks :-) >> >> If it only uses reference counting it will be definition >> have memory leaks. >> >> Some reference counting systems supplement reference counting >> with a garbage collector to get rid of the special cases. > > What technique other than reference counting do these GC use to > guarantee the deletes are safe? Mark-sweep. AFAIU, it goes through all available pointer space (memory) and marks the regions pointed to. When that is done, it can reclaim the rest. In the simplest case, that requires non-interupted GC. There is also an incremental version, I'm not really sure of the implementation of that. There are, afaict, more methods too. Boehm GC can do, and does afaik by default, incremental mark-sweep and can also run in a separate thread. It is a conservative GC; that means if it finds something it thinks points to a region it doesn't reclaim it. It does not make certian that it's actually looking at a pointer. Johann ------------------------------ Date: Thu, 19 Apr 2007 20:50:21 -0400 From: =?ISO-8859-15?Q?Arne_Vajh=F8j?= Subject: Re: C++ Garbage Collector on VMS? Message-ID: <46280e1f$0$90264$14726298@news.sunsite.dk> Tom Linden wrote: > On Wed, 18 Apr 2007 17:18:57 -0700, Arne Vajhøj wrote: >> Tom Linden wrote: >>> I believe the first use of ref counts was in a paper by Freiburghouse on >>> the Multics PL/I compiler, ca. 1967. Of course, it didn't have many >>> leaks :-) >> >> If it only uses reference counting it will be definition >> have memory leaks. > > You have a different paradigm in mind, I will find link and post it. > This was for a compiler, not interpreter so different rules. When > refernece count drops to 0 memory is freed Yes. But what does it do when object A point to object B and object B points to object A, but nothing else point to A and B ? Arne ------------------------------ Date: Thu, 19 Apr 2007 20:57:47 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: C++ Garbage Collector on VMS? Message-ID: <46280fdd$0$90264$14726298@news.sunsite.dk> Bob Koehler wrote: > In article <4626b544$0$90264$14726298@news.sunsite.dk>, =?ISO-8859-15?Q?Arne_Vajh=F8j?= writes: >> Tom Linden wrote: >>> I believe the first use of ref counts was in a paper by Freiburghouse on >>> the Multics PL/I compiler, ca. 1967. Of course, it didn't have many >>> leaks :-) >> If it only uses reference counting it will be definition >> have memory leaks. >> >> Some reference counting systems supplement reference counting >> with a garbage collector to get rid of the special cases. > > What technique other than reference counting do these GC use to > guarantee the deletes are safe? It checks whether the objects are reachable. It sounds rather inefficient, but algorithms has been developed that makes it work. Most JVM's support multiple GC algorithms. http://h18012.www1.hp.com/java/documentation/1.4.2/ivms/docs/release_notes.html # -XX:+UseParallelGC, use parallel garbage collection, can improve the performance of your application on multiprocessor systems, and is strongly recommended for applications using large heaps. # The option, -XX:+UseConcMarkSweepGC, use the concurrent low pause garbage collector, uses a separate collector thread to do parts of old generation collection concurrently with the application threads. A parallel version of the new generation collector is used as well, to further reduce pause times. # The option, -XX:+AggressiveHeap, instructs the JVM to push memory use to the limit. It sets the overall heap to around 1500 MB, the memory management policy defers collection as long as possible, and some GC activity is done in parallel. # The option, -Xincgc, enables the incremental garbage collector. The incremental garbage collector, which is off by default, will eliminate occasional garbage-collection pauses during program execution. However, it can lead to a roughly 10% decrease in overall performance. # The optimistic garbage collection flag, -Xoptgc, improves garbage collection performance of applications with mostly short-lived objects. A server-side application that creates many short-lived objects for each transaction is likely to benefit greatly with -Xoptgc. However this flag should be used with caution. It is not recommended for applications that build up objects quickly during the run time that are not short-lived. Arne ------------------------------ Date: Thu, 19 Apr 2007 22:31:49 -0400 From: "Main, Kerry" Subject: RE: C++ Garbage Collector on VMS? Message-ID: > -----Original Message----- > From: Arne Vajh=F8j [mailto:arne@vajhoej.dk] > Sent: April 18, 2007 8:44 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: C++ Garbage Collector on VMS? >=20 > Main, Kerry wrote: > >> I can not see why GC should be a problem for mission critical apps. > > > > Well, as I recall, one typically has very little control of when GC > > runs, hence a system process that kicks in during peak time for > > transactions does not seem like something you would want to risk. >=20 > There are running a lot of mission critical apps on WAS and WL. >=20 And there are numerous J2EE app's designed for mission critical = applications that in the end, do not scale anywhere near as high as they = were designed for and in the end they start throwing HW at the final = solution. Some might say it was the nature of OO which makes it more difficult to = design truly scalable, system specific optimized code, some might say it = was all the various tiers and associated network latencies (SAP now = recommends tier consolidation putting App tier on same server as DB = tier) or perhaps it is at least partly do to with "stuff" running in the = background doing things that a well designed app environment would not = need to do. Both .Net and J2EE are OO based strategies and there is a lot of market = momentum to adopt one or both of these, but I am not convinced that OO = strategies are the best ones to adopt for very scalable, very mission = critical systems. Sun and Microsoft and all of the supporting casts on both sides want = Cust's to pick one (.Net or J2EE) and ditch their current 3GL = environments (with all of the 10-20 year investments), but I just do not = believe it is the right way to go for the future. Picking either = approach is an App only driven strategy whereby I believe the future for = mission critical stuff will be much more solution driven i.e. 3GL's or = equiv prod's integrated with OS specific features.=20 > It does not take that long time to do a GC. >=20 > There are different apps, different systems, different java > version and different GC options, but as an example: a GC of > young generation on my Pentium P4 with a standard SUN Java > with no special GC options and a very simple app > uses 0.0001 seconds. >=20 > > It also seems to me that a solution is certainly not a good one when > > memory is not properly allocated and de-allocated by design and > > depends on system processes to compensate for poor programming > > practices. >=20 > It is very good system design to consider the possibility that > programmers make mistakes. >=20 > Arne Kind of like giving everyone a parachute when they board a plane because = "mechanics do make mistakes..." :-) Reminds me of the well known airline exec speech to IBM field types = about the difference in focus on quality between the airline industry = vs. the computer industry. That speech was delivered in the middle = 1960's and the same speech today would not have to be changed much at = all. Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20 OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: 18 Apr 2007 15:56:21 GMT From: Doc Subject: Re: can you handle this? Message-ID: genius@marblecliff.com wrote in news:1176829658.639833.220500@p77g2000hsh.googlegroups.com: > sorry Andrew ... that is an old number ... new one is > unlisted ... I wonder why? You been pissing off other people on the "Internets"? Doc. ------------------------------ Date: 19 Apr 2007 09:28:04 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: can you handle this? Message-ID: In article , Dave Froble writes: > AS he has no real evidence to support his version of a god, > you have no real evidence to be able to claim that there is not a god. There are some of us who believe that second statement is wrong. We feel evidence that there is no deity exists. But all we ask is that others show respect for our religous beleifs that don't include a deity and we will respect thiers even though they do. > What's wrong with saying, there is not enough/any evidence to support > any conclusions, and therefore, "I just don't know"? There is nothing "wrong" with being an agnostic. Some of the worlds most well known leaders, including religious leaders, were agnostic. But I think you'll find no fundamentalists amoungst agnostics or atheists. ------------------------------ Date: Thu, 19 Apr 2007 22:48:35 +0000 (UTC) From: Rick Jones Subject: Re: Dead clock, NTP and 1000 seconds Message-ID: I'm not sure if it is possible in VMS, but on some OSes at least, at bootup they can be told to get time from the timeserver via ntpdate, which IIRC is not subject to the limits. Then the ntp daemon is started and so long as it can keep things in bounds, happiness and joy... rick jones -- Wisdom Teeth are impacted, people are affected by the effects of events. these opinions are mine, all mine; HP might not want them anyway... :) feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH... ------------------------------ Date: Wed, 18 Apr 2007 13:16:52 -0400 From: "David Turner, Island Computers" Subject: Re: ES40 to ES45 Upgrade only $4995 ! Message-ID: <132ckk36c99e2ed@news.supernews.com> FYI - This isn't just internal boards- this is the whole box David "David Turner, Island Computers" wrote in message news:132abiut0tavee9@news.supernews.com... > We currently have some (7) ES45 Model 2 off-lease. > These are 1Ghz EV68C machines. They are upgradeable to 1.2Ghz but that is > still a little pricey ! > > > Barebones system only $4995 > These don't use the same memory as the ES40 but do use pretty much the > same > other parts. > Same power supplies, graphic cards disk cage rackmount kit, etc. > > Memory - 2GB HP Memory $1299 > 4GB HP Memory $2995 > > Uses 133Mhz 200PIN SDRAM JEDEC fyi > > > Call or email if interested > > > -- > David B Turner > Island Computers US Corp > 2700 Gregory St, Suite 180 > Savannah GA 31404 > > T: 877-6364332 x201 > Intl: 001 912 447 6622 > > E: dturner-at-islandco-dot-com > F: 912 201 0402 > W: http://www.islandco.com > > ------------------------------ Date: 19 Apr 2007 04:54:57 -0700 From: genius@marblecliff.com Subject: If you live in California, get out now! Message-ID: <1176983697.132038.110770@y80g2000hsf.googlegroups.com> http://www.10tv.com/?sec=&story=sites/10tv/content/pool/200704/1303690180.html ------------------------------ Date: Thu, 19 Apr 2007 14:14:44 +0200 From: Paul Sture Subject: Re: If you live in California, get out now! Message-ID: In article <1176983697.132038.110770@y80g2000hsf.googlegroups.com>, genius@marblecliff.com wrote: > http://www.10tv.com/?sec=&story=sites/10tv/content/pool/200704/1303690180.html The subject is priceless given that the headline for that article is "Hacker Gets Into Ohio State's Computers" Have a nice day y'all. -- Paul Sture ------------------------------ Date: Thu, 19 Apr 2007 08:51:10 -0700 From: David Mathog Subject: Re: If you live in California, get out now! Message-ID: genius@marblecliff.com wrote: > http://www.10tv.com/?sec=&story=sites/10tv/content/pool/200704/1303690180.html > The link has nothing to do with California. Probably he wants us all out of the state because there is irrefutable evidence that evolution has taken place in California. Regards, David Mathog ------------------------------ Date: 19 Apr 2007 15:54:55 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: If you live in California, get out now! Message-ID: <58phmfF2hhikrU1@mid.individual.net> In article , David Mathog writes: > genius@marblecliff.com wrote: >> http://www.10tv.com/?sec=&story=sites/10tv/content/pool/200704/1303690180.html >> > > The link has nothing to do with California. Probably he wants us > all out of the state because there is irrefutable evidence that > evolution has taken place in California. Knowing how strong his geopgraphy skills are maybe he thinks Ohio State University is in California. bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Thu, 19 Apr 2007 16:47:17 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: If you live in California, get out now! Message-ID: <00A66609.53B05C24@SendSpamHere.ORG> In article , David Mathog writes: > > >genius@marblecliff.com wrote: >> http://www.10tv.com/?sec=&story=sites/10tv/content/pool/200704/1303690180.html >> > >The link has nothing to do with California. Probably he wants us >all out of the state because there is irrefutable evidence that >evolution has taken place in California. Be careful there; Hollywood is a strong argument against evolution. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Thu, 19 Apr 2007 16:49:22 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: If you live in California, get out now! Message-ID: <00A66609.9E3D709B@SendSpamHere.ORG> In article <58phmfF2hhikrU1@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: > > >In article , > David Mathog writes: >> genius@marblecliff.com wrote: >>> http://www.10tv.com/?sec=&story=sites/10tv/content/pool/200704/1303690180.html >>> >> >> The link has nothing to do with California. Probably he wants us >> all out of the state because there is irrefutable evidence that >> evolution has taken place in California. > >Knowing how strong his geopgraphy skills are maybe he thinks Ohio State >University is in California. ...and now, congregation, a reading from the Book of Relocations... -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: 19 Apr 2007 13:49:56 -0500 From: brooks@cuebid.zko.hp.nospam (Rob Brooks) Subject: Re: If you live in California, get out now! Message-ID: <3hk8Z50+jJLz@cuebid.zko.hp.com> VAXman- @SendSpamHere.ORG writes: > ...and now, congregation, a reading from the Book of Relocations... That'd be the linker manual, right? -- Rob Brooks MSL -- Nashua brooks!cuebid.zko.hp.com ------------------------------ Date: Thu, 19 Apr 2007 20:35:01 +0200 From: Paul Sture Subject: Re: If you live in California, get out now! Message-ID: In article <3hk8Z50+jJLz@cuebid.zko.hp.com>, brooks@cuebid.zko.hp.nospam (Rob Brooks) wrote: > VAXman- @SendSpamHere.ORG writes: > > > ...and now, congregation, a reading from the Book of Relocations... > > That'd be the linker manual, right? LOL! -- Paul Sture ------------------------------ Date: 19 Apr 2007 21:05:33 +0200 From: Jean-Marc Bourguet Subject: Re: If you live in California, get out now! Message-ID: <871wig89s2.fsf@news.bourguet.org> VAXman- @SendSpamHere.ORG writes: > Be careful there; Hollywood is a strong argument against evolution. It is just a strong remainder than evolution is not necessarily progress. :-) Yours, -- Jean-Marc ------------------------------ Date: Thu, 19 Apr 2007 22:40:55 -0400 From: Dave Froble Subject: Re: If you live in California, get out now! Message-ID: <0qWdnTVB1oRwurXbnZ2dnUVZ_jKdnZ2d@libcom.com> Rob Brooks wrote: > VAXman- @SendSpamHere.ORG writes: > >> ...and now, congregation, a reading from the Book of Relocations... > > That'd be the linker manual, right? > Having fun, and even back on topic. -- David Froble Tel: 724-529-0450 Dave Froble Enterprises, Inc. E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc. 170 Grimplin Road Vanderbilt, PA 15486 ------------------------------ Date: 19 Apr 2007 16:01:58 -0700 From: genius@marblecliff.com Subject: If you live in California, get out now! (Part 2) Message-ID: <1177023718.376996.258340@q75g2000hsh.googlegroups.com> http://www.worldnetdaily.com/news/article.asp?ARTICLE_ID=55273 ------------------------------ Date: Fri, 20 Apr 2007 00:15:48 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: If you live in California, get out now! (Part 2) Message-ID: <00A66647.FBE52C9D@SendSpamHere.ORG> In article <1177023718.376996.258340@q75g2000hsh.googlegroups.com>, genius@marblecliff.com writes: > > > >http://www.worldnetdaily.com/news/article.asp?ARTICLE_ID=55273 > http://www.sacred-texts.com/neu/mphg/mphg.htm#Scene%2011 -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: 19 Apr 2007 17:53:03 -0700 From: "johnhreinhardt@yahoo.com" Subject: Re: If you live in California, get out now! (Part 2) Message-ID: <1177030383.478677.196160@n76g2000hsh.googlegroups.com> On Apr 19, 8:15 pm, VAXman- @SendSpamHere.ORG wrote: > In article <1177023718.376996.258...@q75g2000hsh.googlegroups.com>, gen...@marblecliff.com writes: > > >http://www.worldnetdaily.com/news/article.asp?ARTICLE_ID=55273 > > http://www.sacred-texts.com/neu/mphg/mphg.htm#Scene%2011 > > -- > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > > "Well my son, life is like a beanstalk, isn't it?" Yours was more fun than Boob's. ------------------------------ Date: 19 Apr 2007 18:46:38 -0700 From: genius@marblecliff.com Subject: Re: If you live in California, get out now! (Part 2) Message-ID: <1177033598.499459.319470@o5g2000hsb.googlegroups.com> On Apr 19, 8:53 pm, "johnhreinha...@yahoo.com" wrote: > On Apr 19, 8:15 pm, VAXman- @SendSpamHere.ORG wrote: > > > In article <1177023718.376996.258...@q75g2000hsh.googlegroups.com>, gen...@marblecliff.com writes: > > > >http://www.worldnetdaily.com/news/article.asp?ARTICLE_ID=55273 > > >http://www.sacred-texts.com/neu/mphg/mphg.htm#Scene%2011 > > > -- > > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > > > "Well my son, life is like a beanstalk, isn't it?" > > Yours was more fun than Boob's. isn't too funny if you living in California with kids ... ------------------------------ Date: Wed, 18 Apr 2007 10:39:14 -0700 From: "Tom Linden" Subject: Intel tips more details on Itanium roadmap Message-ID: http://www.eetimes.com/news/latest/showArticle.jhtml?articleID=199100362 Rick Merritt EE Times (04/17/2007 6:33 PM EDT) SAN JOSE, Calif. — Intel tipped more details about its Itanium processors today, committing to use of a new high-speed interconnect on the high-end server microprocessors starting in 2008. Ultimately, the so-called Common System Interconnect (CSI) will appear in both Itanium and Xeon server CPUs. Tukwilla, Intel's code name for a four-core Itanium processor due in 2008, will be the first to use CSI instead of a traditional Intel front-side bus to link to external components. The new CPU is expected to double the performance of Intel's existing 9000 series Itanium, a two-core chip code named Montecito. The new Itanium CPU is expected to sample this year and will also use a form of simultaneous multi-threading expected to support two threads per core. The technique is similar to the HyperThreading approach Intel discontinued. Jim Fister, a technology strategist in Intel's Digital Enterprise group, sketched out the Itanium roadmap in a keynote address at the Gelato Ice conference here Tuesday (April 17). Intel has been criticized for maintaining Itanium as a separate architecture aimed at the rarified market for mainframe-class systems. Fister defended Itanium which he said now shares about an equal slice of a $25 billion market for high end servers with Sun's Sparc and IBM's Power processors. Fister gave few details about CSI except that it is aimed to leapfrog existing interconnects such as HyperTransport used by archrival Advanced Micro Devices. CSI should have "much higher performance" than Intel current front side bus running at up to 1.3 GHz on its Xeon processors. Intel wants CSI to embrace two different approaches to linking processors in high-end servers. It should enable coherent data transfers between small groups of local processors as well as non-uniform memory links between as many as 128 CPU over a more widely distributed system. The later capability may require a software abstraction layer to reconcile different levels of memory latency within a system. The company is now evaluating in the lab some of its first CPUs made in its 45nm process technology. Based on their performance, Intel expects to disclose more details about CSI as early as this summer. Although both Intel's Itanium and Xeon server CPUs ultimately will use the CSI interconnect, OEMs will not be able to design systems, at least initially, that could accommodate either CPU due to other differences in the architectures. "Long term that [plug and play capability] would be great to have," said Fister. Intel will follow up Tukwilla with Paulson, which may use between 6 and 10 cores in a CPU that has no announced shipping date yet. It is expected to double performance of Tukwilla. Intel is likely to use 65 and 45 nm process technology respectively for the two Itanium generations, although it has not yet announced in what processes the chips will be made. In the short term, Intel plans to release later this year an upgrade of its current Montecito Itanium chip. The Montvale CPU will boost its front side bus to 667 MHz and support demand-based switching, a capability for dynamically lowering processor power based on changing workloads. Separately, Intel launched a range of new products and initiatives at its Intel Developer Forum in Beijing. The company also announced quarterly results that showed revenues down slightly, but profits that have increased.-- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Wed, 18 Apr 2007 17:29:35 -0400 From: JF Mezei Subject: Re: Intel tips more details on Itanium roadmap Message-ID: Tom Linden wrote: > today, committing to use of a new high-speed interconnect on the high-end > server microprocessors starting in 2008. Ultimately, the so-called Common > System Interconnect (CSI) will appear in both Itanium and Xeon server CPUs. This was announced back in 2004. It had been supposed to come out in 2007. But a one year delay is not unheard of. Interestingly, I had heard that the 8086 with CSI would come out in 2008 and IA64 come out either late 2008 or early 2009. Interesting that in his speech, it seems to be a more firm "2008". > Fister defended Itanium which he said now shares about an equal slice of a > $25 billion market for high end servers with Sun's Sparc and IBM's Power > processors. has Intel announced that they managed to make that IA64 thing profitable for intel yet ? Last I had heard, it had been confirmed that it was losing money. And of course, with its 8086 now on accelerated development, it will leapfrog IA64, especially since both will share the same system interconnect technology. > Intel will follow up Tukwilla with Paulson, which may use between 6 and 10 > cores in a CPU that has no announced shipping date yet. It is expected to > double performance of Tukwilla. Now, there's a telling sentence. "No announced shipping date yet". Sounds like EV8 to me. ------------------------------ Date: Thu, 19 Apr 2007 14:35:38 -0400 From: "David Turner, Island Computers" Subject: IT'S THAT TIME AGAIN - FREE DS10L OFFER Message-ID: SEND US YR EMAIL ADDRESS, TEL# AND NAME OF COURSE IF YOU HAVEN'T ALREADY SUBJECT LINE MUST STATE: FREE DS10L -- David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404 T: 877-6364332 x201 Intl: 001 912 447 6622 E: DTURNER-AT-ISLANDCO-DOT-COM islandco.com ------------------------------ Date: 19 Apr 2007 12:21:59 -0700 From: AEF Subject: Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER Message-ID: <1177010519.309545.77500@y80g2000hsf.googlegroups.com> On Apr 19, 2:35 pm, "David Turner, Island Computers" wrote: > SEND US YR EMAIL ADDRESS, TEL# AND > NAME OF COURSE IF YOU HAVEN'T ALREADY Name of course? Say what? > > SUBJECT LINE MUST STATE: FREE DS10L > > -- > David B Turner [...] AEF ------------------------------ Date: 19 Apr 2007 13:32:07 -0700 From: Rich Jordan Subject: Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER Message-ID: <1177014727.577960.202250@b58g2000hsg.googlegroups.com> On Apr 19, 1:35 pm, "David Turner, Island Computers" wrote: > SEND US YR EMAIL ADDRESS, TEL# AND > NAME OF COURSE IF YOU HAVEN'T ALREADY > > SUBJECT LINE MUST STATE: FREE DS10L > > -- > David B Turner > Island Computers US Corp > 2700 Gregory St, Suite 180 > Savannah GA 31404 > > T: 877-6364332 x201 > Intl: 001 912 447 6622 > > E: DTURNER-AT-ISLANDCO-DOT-COM > > islandco.com So those of us who signed up for the original one way back when don't need to do so again, right? Just wanted to be sure; my current DS10L is asking for a cluster. ------------------------------ Date: Thu, 19 Apr 2007 16:55:41 -0400 From: "David Turner, Island Computers" Subject: Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER Message-ID: <132flqbqcmtqqaa@news.supernews.com> everyone who has sent their email since the initial draw is on the list dt "Rich Jordan" wrote in message news:1177014727.577960.202250@b58g2000hsg.googlegroups.com... > On Apr 19, 1:35 pm, "David Turner, Island Computers" islandco.com> wrote: >> SEND US YR EMAIL ADDRESS, TEL# AND >> NAME OF COURSE IF YOU HAVEN'T ALREADY >> >> SUBJECT LINE MUST STATE: FREE DS10L >> >> -- >> David B Turner >> Island Computers US Corp >> 2700 Gregory St, Suite 180 >> Savannah GA 31404 >> >> T: 877-6364332 x201 >> Intl: 001 912 447 6622 >> >> E: DTURNER-AT-ISLANDCO-DOT-COM >> >> islandco.com > > > So those of us who signed up for the original one way back when don't > need to do so again, right? Just wanted to be sure; my current DS10L > is asking for a cluster. > ------------------------------ Date: Thu, 19 Apr 2007 17:48:36 -0400 From: norm.raphael@metso.com Subject: Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER Message-ID: AEF wrote on 04/19/2007 03:21:59 PM: > On Apr 19, 2:35 pm, "David Turner, Island Computers" islandco.com> wrote: > > SEND US YR EMAIL ADDRESS, TEL# AND > > NAME OF COURSE IF YOU HAVEN'T ALREADY > > Name of course? Say what? Well, that's why punctuation was invented. "...NAME, OF COURSE, IF...." ;-) > > > > > SUBJECT LINE MUST STATE: FREE DS10L > > > > -- > > David B Turner > [...] > > AEF > ------------------------------ Date: Fri, 20 Apr 2007 00:09:55 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: IT'S THAT TIME AGAIN - FREE DS10L OFFER Message-ID: <00A66647.2942399B@SendSpamHere.ORG> In article , "David Turner, Island Computers" writes: > > >SEND US YR EMAIL ADDRESS, VAXman at "term for the interjection of a word in a compound word" dot com > TEL# 732.363.PINT >AND >NAME OF COURSE IF YOU HAVEN'T ALREADY I've always liked the math courses. How about... Tensor Calculus -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: 19 Apr 2007 03:28:17 -0700 From: palanivel Subject: market resources Message-ID: <1176978497.634623.30400@n59g2000hsh.googlegroups.com> Get the sales you need starting today! Amazing advertising that works! Solo ads, guaranteed visitors, guaranteed signups and more Check it out today http://classifieds.byindia.com/detail.php?siteid=5046&cityid=191&show_still=1&citydetail=1 click on marketing resources General details Sold by sushma kapoor Email Contact seller Location GHANDI CHOWK, ------------------------------ Date: 18 Apr 2007 10:54:50 -0700 From: "brucebrown100@yahoo.com" Subject: Netbackup 5 VMS Client and the VMS backup image switch. Message-ID: <1176918890.669517.152400@n59g2000hsh.googlegroups.com> Hello All Can someone confirm something I wish to do using the NETBACKUP 5 VMS client for netbackup backing up multiple VMS Drives on an EVA 8000 SAN. We are currently using VMS 7.3.2 on a 2 node cluster. "image'. In the context of VMS BACKUP it refers to the mode of backup that results in a complete and functional copy of a disk volume. In the context of NetBackup an "image" refers the collection of files saved during a single NetBackup execution - these files are grouped together in a NetBackup "image" Currently we are using NETBACKUP VMS client to store files but not taking an "image" of the drives. Without using the /image switch I take it I am not storing a true image of the drive but only a file set. With the /image command using netbackup I am getting the following additional information; volume initialization, system disk bootstrap, ACL, etc etc.. Am I correct in this assumption? Thank You in advance. Bruce Brown ------------------------------ Date: 18 Apr 2007 14:28:54 -0700 From: Ian Miller Subject: Re: Netbackup 5 VMS Client and the VMS backup image switch. Message-ID: <1176931734.196244.153910@n76g2000hsh.googlegroups.com> you do need the /IMAGE backup to get a full copy of the logical volume. ------------------------------ Date: Wed, 18 Apr 2007 20:40:43 -0500 From: David J Dachtera Subject: Re: Now I've seen everything Message-ID: <4626C89B.DEE6CE2@spam.comcast.net> Marc Van Dyck wrote: > > David J Dachtera submitted this idea : > > Marc Van Dyck wrote: > >> > >> David J Dachtera formulated on mardi : > >>> Neil Rieck wrote: > >>>> > >>>> Last night on TLC the boys from American Chopper received a phone call > >>>> from HP. It seems that the boys down in Houston want to have an HP- > >>>> Themed chopper built for some such reason. Was it a set-up when Paul > >>>> Sr. smashed a working PC then stepped on it and found that it still > >>>> worked? I don't know but I now can die in peace because I have seen > >>>> everything. > >>> > >>> Well, not quite everything. You haven't seen: > >>> > >>> o A full-page ad in the mainstream media for anything running on or related > >>> to OpenVMS-I64 > >>> > >>> o An I64 SuperDome that can outperform an Alpha EV7z GS1280 > >> > >> Sorry to rain on your parade, Dave, but I think a Superdome with > >> Montecito CPUs does that fingers in the nose. > > > > See the Itanium/Integrity presentations from the most recent HP Tech Forum, > > Fall 2006 in Houston. Let me know if you can't find/access them. > > See the most recent presentations that our friend Guy Peleg has given > in Brussels about one month ago. Any chance you have a URL? Tried searching OpenVMS.org and Encompass/US and came up empty. Even tried searching my e-mail archive of VMS SIG e-mail - no schmae. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Thu, 19 Apr 2007 08:43:08 +0200 From: Marc Van Dyck Subject: Re: Now I've seen everything Message-ID: After serious thinking David J Dachtera wrote : > > Any chance you have a URL? Tried searching OpenVMS.org and Encompass/US and > came up empty. Even tried searching my e-mail archive of VMS SIG e-mail - no > schmae. > Goto http://www.hp-interex.be/wiki/index.php/Main_Page and follow the appropriate links. That page is in english. -- Marc Van Dyck ------------------------------ Date: Thu, 19 Apr 2007 00:10:48 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: OT: OpenVMS Message-ID: <00A6657E.1E66C029@SendSpamHere.ORG> In article <4626a17b$0$90276$14726298@news.sunsite.dk>, =?ISO-8859-1?Q?Arne_Vajh=F8j?= writes: > > >VAXman- @SendSpamHere.ORG wrote: >> In article <462574f1$0$90271$14726298@news.sunsite.dk>, =?ISO-8859-1?Q?Arne_Vajh=F8j?= writes: >>> VAXman- @SendSpamHere.ORG wrote: >>>> BTW, I will have a friend from Denmark (Horsens) here for a week beginning >>>> late next Monday. He's a cricket fan and runs the Danish cricket site. I >>>> going to take him to a baseball game to see if he can follow along. :) >>> I am a lot closer to New Jersey than to Horsens at the moment. >> >> Are you in the states at this time? > >Yep. > >Rhode Island. Say "Hello" to Peter Gryphon if you run into him. :) How long have you been in the states and how long is your stay? -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Wed, 18 Apr 2007 20:21:35 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: OT: OpenVMS Message-ID: <4626b5e2$0$90264$14726298@news.sunsite.dk> VAXman- @SendSpamHere.ORG wrote: > In article <4626a17b$0$90276$14726298@news.sunsite.dk>, =?ISO-8859-1?Q?Arne_Vajh=F8j?= writes: >> Rhode Island. > > Say "Hello" to Peter Gryphon if you run into him. :) RI is a small state but not so small that I know everybody. > How long have you been in the states and how long is your stay? 15 months and unknown. Arne ------------------------------ Date: Wed, 18 Apr 2007 20:19:03 -0500 From: David J Dachtera Subject: Re: OT: OpenVMS Message-ID: <4626C387.E289CE61@spam.comcast.net> AEF wrote: > [snip] > When listing a journal file, you can use the file-selection > qualifiers /BEFORE, /SINCE, and /EXCLUDE to search for specific files. > (In this context, the /BEFORE and /SINCE qualifiers refer to the time > when the save set was created, not the time when the files in the save > set were created.) (Next guess time...) I think it may be possible that the description is misleading. When it says, "In this context, the /BEFORE and /SINCE qualifiers refer to the time when the save set was created", I believe it probably should say "...when the journal file was created...". It may have been assumed (incorrectly) that the journal and saveset creations are the same. ...and maybe not. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Wed, 18 Apr 2007 18:54:32 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: OT: OpenVMS Message-ID: <4626a17b$0$90276$14726298@news.sunsite.dk> VAXman- @SendSpamHere.ORG wrote: > In article <462574f1$0$90271$14726298@news.sunsite.dk>, =?ISO-8859-1?Q?Arne_Vajh=F8j?= writes: >> VAXman- @SendSpamHere.ORG wrote: >>> BTW, I will have a friend from Denmark (Horsens) here for a week beginning >>> late next Monday. He's a cricket fan and runs the Danish cricket site. I >>> going to take him to a baseball game to see if he can follow along. :) >> I am a lot closer to New Jersey than to Horsens at the moment. > > Are you in the states at this time? Yep. Rhode Island. Arne ------------------------------ Date: 19 Apr 2007 11:39:20 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: OT: OpenVMS Message-ID: <58p2n7F2hj0t6U3@mid.individual.net> In article <4626b5e2$0$90264$14726298@news.sunsite.dk>, Arne Vajhøj writes: > VAXman- @SendSpamHere.ORG wrote: >> In article <4626a17b$0$90276$14726298@news.sunsite.dk>, =?ISO-8859-1?Q?Arne_Vajh=F8j?= writes: >>> Rhode Island. >> >> Say "Hello" to Peter Gryphon if you run into him. :) > > RI is a small state but not so small that I know everybody. > >> How long have you been in the states and how long is your stay? > > 15 months and unknown. > If your stay happens to involve a trip out I84 to the west as far as Scranton, PA stop by the University. I'll buy lunch. bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: 19 Apr 2007 05:03:49 -0700 From: AEF Subject: Re: OT: OpenVMS Message-ID: <1176984229.120622.90070@q75g2000hsh.googlegroups.com> On Apr 18, 9:19 pm, David J Dachtera wrote: > AEF wrote: > > [snip] > > When listing a journal file, you can use the file-selection > > qualifiers /BEFORE, /SINCE, and /EXCLUDE to search for specific files. > > (In this context, the /BEFORE and /SINCE qualifiers refer to the time > > when the save set was created, not the time when the files in the save > > set were created.) > > (Next guess time...) > > I think it may be possible that the description is misleading. When it says, "In > this context, the /BEFORE and /SINCE qualifiers refer to the time when the save > set was created", I believe it probably should say "...when the journal file was > created...". Nope. > > It may have been assumed (incorrectly) that the journal and saveset creations > are the same. > > ...and maybe not. Not. It works as advertised for /BEFORE. > > -- > David J Dachtera [...] AEF ------------------------------ Date: Fri, 20 Apr 2007 02:50:42 +0200 From: "Dr. Dweeb" Subject: Re: OT: OpenVMS Message-ID: <46280ea3$0$21928$157c6196@dreader1.cybercity.dk> Arne Vajhøj wrote: > VAXman- @SendSpamHere.ORG wrote: >> In article <4626a17b$0$90276$14726298@news.sunsite.dk>, >> =?ISO-8859-1?Q?Arne_Vajh=F8j?= writes: >>> Rhode Island. >> >> Say "Hello" to Peter Gryphon if you run into him. :) > > RI is a small state but not so small that I know everybody. > >> How long have you been in the states and how long is your stay? > > 15 months and unknown. > > Arne Arne for fanden. Rhode island!?! Hvad i alverden er årsagen til det? Er det VMS relateret ? Dweeb ------------------------------ Date: Thu, 19 Apr 2007 22:01:59 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: OT: OpenVMS Message-ID: <46281ee9$0$90269$14726298@news.sunsite.dk> Dr. Dweeb wrote: > Arne Vajhøj wrote: >> VAXman- @SendSpamHere.ORG wrote: >>> In article <4626a17b$0$90276$14726298@news.sunsite.dk>, >>> =?ISO-8859-1?Q?Arne_Vajh=F8j?= writes: >>>> Rhode Island. >>> Say "Hello" to Peter Gryphon if you run into him. :) >> RI is a small state but not so small that I know everybody. >> >>> How long have you been in the states and how long is your stay? >> 15 months and unknown. > > Arne for fanden. Rhode island!?! Hvad i alverden er årsagen til det? Er > det VMS relateret ? Lidt. Men mere Java relateret. Og hvem er bag den anonyme "Dr. Dweeb" ? Arne ------------------------------ Date: 19 Apr 2007 22:35:01 -0700 From: Doug Phillips Subject: Re: OT: OpenVMS Message-ID: <1177047301.370717.251760@n59g2000hsh.googlegroups.com> On Apr 17, 1:53 pm, JF Mezei wrote: > > I have come to the conclusion that the current rash off discussions are > the result of everyone's VMS systems purring nicely without problems and > we can take some time to chat about "other stuff" :-( > > So perhaps it should be seen as a tribute to VMS ... I guess we're kind of like the Maytag repairmen (a TV commercial that was shown in the US; repairmen sitting around with nothing to do because Maytag is so reliable.) Or maybe like firefighters who keep themselves occupied training, taking care of the equipment, cooking, eating, playing games and "chewing the fat" but ready to respond when they're needed. Yeah. At least there *are* some people hanging around here. ------------------------------ Date: 18 Apr 2007 18:01:42 -0700 From: AEF Subject: Re: Problem with BACKUP/LIST/JOU/SINCE/SELECT Message-ID: <1176944502.647610.202630@b58g2000hsg.googlegroups.com> On Apr 18, 8:13 pm, Ken Fairfield wrote: > AEF wrote: > > On Apr 16, 4:31 pm, "AEF" wrote: > >> $ BACK/LIS/JOU=STOR-2007/SINC=1-APR-2007/SELE=BITMAP > >> Listing of BACKUP journal > >> Journal file SYS$SYSDEVICE:[FELDMAN.BACKUP]STOR-2007.BJL;1 on 16- > >> APR-2007 21:29:13.63 > > >> Save set STOR-070131.IMG created on 31-JAN-2007 10:12:32.08 > >> Volume number 1, volume label STOR50 > >> [000000]BITMAP.SYS;1 file header only > > >> Save set STOR-070302.IMG created on 2-MAR-2007 20:37:59.84 > >> Volume number 1, volume label STOR50 > >> [000000]BITMAP.SYS;1 file header only > >> . > >> . > >> . > > >> VMS V6.2 > > >> OK, what am I doing wrong? It works fine for /BEFORE, but seems to > >> completely ignore /SINCE. > > >> Thanks! > > >> AEF > > > It's ON TOPIC. Can anyone help? None of the responses is helpful so > > far, though I do appreciate the efforts. > > Well, looking in the online version of the System Management > Utilities Reference for a detailed description of BACKUP, it > sure looks to me like /SINCE is an qualifier. > In other words, it has no effect on selecting savesets (e.g., > from a backup tape), or files listed within a journal file, nor, > indeed, files restored from a saveset. Read about BACKUP/JOURNAL in the same reference. Here's the relavent snippet: When listing a journal file, you can use the file-selection qualifiers /BEFORE, /SINCE, and /EXCLUDE to search for specific files. (In this context, the /BEFORE and /SINCE qualifiers refer to the time when the save set was created, not the time when the files in the save set were created.) Also, by specifying a file in a multivolume save set, you can search the journal file to find which volume the file is in. You can then mount that volume and restore the file. They should have included /SELECT, perhaps instead of /EXCLUDE. I didn't try /EXCLUDE. > > While one might hope that the CLD for BACKUP would trigger an > error for using /SINCE in this context, one needs to remember > that BACKUP is one of the oldest, and therefore, one of the > most irregular DCL commands on VMS, at least in my experience > (SORT is another). So the fact that this usage isn't caught > and flagged as an error shouldn't be all that surprising... It appears to be a bug. See Doug Phillips' post about this. He found like I did that /BEFORE worked fine but /SINCE had no effect. > > -Ken > -- > Ken & Ann Fairfield > What: Ken dot And dot Ann > Where: Gmail dot Com AEF ------------------------------ Date: 18 Apr 2007 21:59:39 -0500 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: Problem with BACKUP/LIST/JOU/SINCE/SELECT Message-ID: In article <1176944502.647610.202630@b58g2000hsg.googlegroups.com>, AEF writes: > When listing a journal file, you can use the file-selection > qualifiers /BEFORE, /SINCE, and /EXCLUDE to search for specific files. > (In this context, the /BEFORE and /SINCE qualifiers refer to the time > when the save set was created, not the time when the files in the save > set were created.) Also, by specifying a file in a multivolume save > set, you can search the journal file to find which volume the file is > in. You can then mount that volume and restore the file. The format of the journal file was well documented. I do not recall any space in that format for characteristics of individual files, so the date the journal was written is the most one could hope for without accessing the saveset.. ------------------------------ Date: 19 Apr 2007 05:26:21 -0700 From: AEF Subject: Re: Problem with BACKUP/LIST/JOU/SINCE/SELECT Message-ID: <1176985581.432300.229310@y5g2000hsa.googlegroups.com> On Apr 18, 10:59 pm, Kilgal...@SpamCop.net (Larry Kilgallen) wrote: > In article <1176944502.647610.202...@b58g2000hsg.googlegroups.com>, AEF writes: > > > When listing a journal file, you can use the file-selection > > qualifiers /BEFORE, /SINCE, and /EXCLUDE to search for specific files. > > (In this context, the /BEFORE and /SINCE qualifiers refer to the time > > when the save set was created, not the time when the files in the save > > set were created.) Also, by specifying a file in a multivolume save > > set, you can search the journal file to find which volume the file is > > in. You can then mount that volume and restore the file. > > The format of the journal file was well documented. I do not recall > any space in that format for characteristics of individual files, so > the date the journal was written is the most one could hope for > without accessing the saveset.. But the dates of the save sets ARE there. I can see them in the listing. I don't need to lookup the format of the journal file. Remember: There are, in general, multiple save sets in a single journal file, and each save set's date is recorded in the journal file. Example: $ BACK/LIST/JOUR=STOR-2007/SELECT=BITMAP Listing of BACKUP journal Journal file SYS$SYSDEVICE:[FELDMAN.BACKUP]STOR-2007.BJL;1 on 19- APR-2007 14:08:05.39 Save set STOR-070131.IMG created on 31-JAN-2007 10:12:32.08 Volume number 1, volume label STOR50 [000000]BITMAP.SYS;1 file header only Save set STOR-070302.IMG created on 2-MAR-2007 20:37:59.84 Volume number 1, volume label STOR50 [000000]BITMAP.SYS;1 file header only Save set STOR-070306.IMG created on 6-MAR-2007 18:00:35.13 Volume number 1, volume label STOR50 [000000]BITMAP.SYS;1 file header only Save set STOR-070330.IMG created on 30-MAR-2007 15:43:03.29 Volume number 1, volume label STOR50 [000000]BITMAP.SYS;1 file header only Save set STOR-070413.IMG created on 13-APR-2007 16:14:15.78 Volume number 1, volume label STOR50 [000000]BITMAP.SYS;1 file header only End of BACKUP journal $ BACK/LIST/JOUR=STOR-2007/SELECT=BITMAP/BEFORE=4-MAR Listing of BACKUP journal Journal file SYS$SYSDEVICE:[FELDMAN.BACKUP]STOR-2007.BJL;1 on 19- APR-2007 14:11:06.43 Save set STOR-070131.IMG created on 31-JAN-2007 10:12:32.08 Volume number 1, volume label STOR50 [000000]BITMAP.SYS;1 file header only Save set STOR-070302.IMG created on 2-MAR-2007 20:37:59.84 Volume number 1, volume label STOR50 [000000]BITMAP.SYS;1 file header only End of BACKUP journal $. Note that the dates of the save sets are clearly displayed. Also note that /BEFORE works as advertised: it lists the SAVE SETS that were created before the specified date. It appears to be a bug for /SINCE, as Doug Phillips has confirmed, as using /SINCE always lists all the save sets for both him and me. Thanks. AEF ------------------------------ Date: 19 Apr 2007 09:21:49 -0700 From: Doug Phillips Subject: Re: Problem with BACKUP/LIST/JOU/SINCE/SELECT Message-ID: <1176999709.667167.247810@n76g2000hsh.googlegroups.com> On Apr 19, 6:26 am, AEF wrote: > On Apr 18, 10:59 pm, Kilgal...@SpamCop.net (Larry Kilgallen) wrote: > > > In article <1176944502.647610.202...@b58g2000hsg.googlegroups.com>, AEF writes: > > > > When listing a journal file, you can use the file-selection > > > qualifiers /BEFORE, /SINCE, and /EXCLUDE to search for specific files. > > > (In this context, the /BEFORE and /SINCE qualifiers refer to the time > > > when the save set was created, not the time when the files in the save > > > set were created.) Also, by specifying a file in a multivolume save > > > set, you can search the journal file to find which volume the file is > > > in. You can then mount that volume and restore the file. > > > The format of the journal file was well documented. I do not recall > > any space in that format for characteristics of individual files, so > > the date the journal was written is the most one could hope for > > without accessing the saveset.. > > But the dates of the save sets ARE there. I can see them in the > listing. I don't need to lookup the format of the journal file. > Remember: There are, in general, multiple save sets in a single > journal file, and each save set's date is recorded in the journal > file. > > Example: > > $ BACK/LIST/JOUR=STOR-2007/SELECT=BITMAP > Listing of BACKUP journal > Journal file SYS$SYSDEVICE:[FELDMAN.BACKUP]STOR-2007.BJL;1 on 19- > APR-2007 14:08:05.39 > > Save set STOR-070131.IMG created on 31-JAN-2007 10:12:32.08 > Volume number 1, volume label STOR50 > [000000]BITMAP.SYS;1 file header only > > Save set STOR-070302.IMG created on 2-MAR-2007 20:37:59.84 > Volume number 1, volume label STOR50 > [000000]BITMAP.SYS;1 file header only > > Save set STOR-070306.IMG created on 6-MAR-2007 18:00:35.13 > Volume number 1, volume label STOR50 > [000000]BITMAP.SYS;1 file header only > > Save set STOR-070330.IMG created on 30-MAR-2007 15:43:03.29 > Volume number 1, volume label STOR50 > [000000]BITMAP.SYS;1 file header only > > Save set STOR-070413.IMG created on 13-APR-2007 16:14:15.78 > Volume number 1, volume label STOR50 > [000000]BITMAP.SYS;1 file header only > > End of BACKUP journal > > $ BACK/LIST/JOUR=STOR-2007/SELECT=BITMAP/BEFORE=4-MAR > Listing of BACKUP journal > Journal file SYS$SYSDEVICE:[FELDMAN.BACKUP]STOR-2007.BJL;1 on 19- > APR-2007 14:11:06.43 > > Save set STOR-070131.IMG created on 31-JAN-2007 10:12:32.08 > Volume number 1, volume label STOR50 > [000000]BITMAP.SYS;1 file header only > > Save set STOR-070302.IMG created on 2-MAR-2007 20:37:59.84 > Volume number 1, volume label STOR50 > [000000]BITMAP.SYS;1 file header only > > End of BACKUP journal > > $. > > Note that the dates of the save sets are clearly displayed. Also note > that /BEFORE works as advertised: it lists the SAVE SETS that were > created before the specified date. > > It appears to be a bug for /SINCE, as Doug Phillips has confirmed, as > using /SINCE always lists all the save sets for both him and me. > I actually tested this before I posted. The /SINCE qualifier does not work. /BEFORE does. ------------------------------ Date: Thu, 19 Apr 2007 05:51:24 +0800 From: "Richard Maher" Subject: Re: Process Software: SocketException: function not implemented Message-ID: Hi Richard, Ok, I'll pass this on to the Rdb List Server. I would've thought there would've been an official call out with your support people by now? Maybe it's been and gone? fixed itself? To satisfy my curiosity, do you still not support the ucx$c_share socket option for TCPware? Does it still not give you an error at setmode time but rather fall over when the new process tries to take control of the socket? Cheers Richard Maher Here is some more on the original question: - =========================================== Melinda, I am not the owner of the Service Request, but I helped working on it in the background. Fact is that "pretty much any and all testing with the rdb product set on vms uses the HP supplied and supported TCPIP software. We test on nothing else." That's the reason why you have been asked in the SR to try HP TCPIP instead of TCPWare. Jim showed you that we are using the basic call to create a new Server Socket. And that is the line in our code where the error is returned. So why not checking this on your machine with the following simple program? If it fails as well, then our driver can be ruled out. import java.net.*; public class TestServer { public static void main (String[] args) throws Exception { if (args.length != 2) { System.out.println("Usage: java TestServer "); System.exit(1); } System.out.println("Host Address="+args[0]); System.out.println("Port ="+args[1]); ServerSocket ss = new ServerSocket(); ss.bind(new InetSocketAddress(InetAddress.getByName(args[0]), Integer.parseInt(args[1]))); System.out.println("ServerSocket Bound"); ss.accept(); } } $ java "TestServer.java" $ java "TestServer" localhost 1701 Returns for me: Host Address=localhost Port =1701 ServerSocket Bound You need to ctrl-y - stop the program in the case it does not fail. Best regards Wolfgang Medd, Melinda (WSTF-RD)[ESCG] wrote: >We just installed Rdb JDBC version 7.25 and we are having trouble with >installing the JDBC Thin Server. When we tried to test the Thin server >we got this: > >PDIDS2>spawn/nowait/proc=rdbthinsrvtest java -jar rdbthinsrv.jar >%DCL-S-SPAWNED, process RDBTHINSRVTEST spawned >PDIDS2>pipeshow system|search sys$input rdbthinsrvtest >Unable to listen on port 1701 >java.net.SocketException: function not implemented >at java.net.PlainSocketImpl.socketBind(Native Method) >at java.net.PlainSocketImpl.bind(PlainSocketImpl.java:331) >at java.net.ServerSocket.bind(ServerSocket.java:318) >at java.net.ServerSocket.(ServerSocket.java:185) >at java.net.ServerSocket.(ServerSocket.java:141) >at oracle.rdb.jdbc.srv.RdbSrv.getServerSocket(RdbSrv.java:649) >at oracle.rdb.jdbc.srv.RdbThinSrv.(RdbThinSrv.java:127) >at oracle.rdb.jdbc.srv.RdbThinSrv.main(RdbThinSrv.java:251) >%SEARCH-I-NOMATCHES, no strings matched > >Rdb Technical Service couldn't solve the problem expect thinking it was >Process Software (our tcp/ip) the was the problem. We don't think so. > >So I am putting on the list to see if anyway else have the some problem >and maybe a solution! > >Thanks, > >Melinda >Melinda M. Medd >Jacobs Technology Inc. >NASA White Sands Test Facility >Las Cruces, NM 88012 > >"Walking on water and developing software from a specification are easy >if both are frozen." -Edward V. Berard, Life-Cycle Approaches > -- Dr. Wolfgang Kobarg-Sachsse Principal Support Engineer Oracle Support Services ORACLE Deutschland GmbH Hamborner Straße 51 40472 Duesseldorf Germany ================================================ "Richard Whalen" wrote in message news:f054h0$126$1@news.process.com... > I downloaded the Rdb Java kit from Oracle's site and tried it on my > Integrity system running OpenVMS 8.3 with TCPware 5.7-2 and experienced no > problems. I tried it with both Java 4.2 and 5.0. > > > --------------------------- > Richard Whalen > Process Software > > ------------------------------ Date: Thu, 19 Apr 2007 08:46:32 -0400 From: "Richard Whalen" Subject: Re: Process Software: SocketException: function not implemented Message-ID: The comments in the driver say that UCX$_SHARE has been supported as an IO$_SETMODE socket option since late 1994, which is longer than I have been working at Process Software. I am not aware of any outstanding problem reports with UCX$_SHARE. "Richard Maher" wrote in message news:f0646q$nse$1@news-01.bur.connect.com.au... > Hi Richard, > > Ok, I'll pass this on to the Rdb List Server. I would've thought there > would've been an official call out with your support people by now? Maybe > it's been and gone? fixed itself? > > To satisfy my curiosity, do you still not support the ucx$c_share socket > option for TCPware? Does it still not give you an error at setmode time > but > rather fall over when the new process tries to take control of the socket? > > Cheers Richard Maher > > Here is some more on the original question: - > > =========================================== > Melinda, > > I am not the owner of the Service Request, but I helped working on it in > the background. > > Fact is that "pretty much any and all testing with the rdb product set on > vms uses the HP supplied and supported TCPIP software. We > test on nothing else." That's the reason why you have been asked in the > SR to try HP TCPIP instead of TCPWare. > > Jim showed you that we are using the basic call to create a new Server > Socket. And that is the line in our code where the error is returned. So > why not checking this on your machine with the following simple program? > If it fails as well, then our driver can be ruled out. > > import java.net.*; > > public class TestServer { > > public static void main (String[] args) throws Exception { > if (args.length != 2) { > System.out.println("Usage: java TestServer > "); > System.exit(1); > } > > System.out.println("Host Address="+args[0]); > System.out.println("Port ="+args[1]); > ServerSocket ss = new ServerSocket(); > > ss.bind(new > InetSocketAddress(InetAddress.getByName(args[0]), > Integer.parseInt(args[1]))); > > System.out.println("ServerSocket Bound"); > > ss.accept(); > } > } > > $ java "TestServer.java" > $ java "TestServer" localhost 1701 > > Returns for me: > > Host Address=localhost > Port =1701 > ServerSocket Bound > > You need to ctrl-y - stop the program in the case it does not fail. > > Best regards > Wolfgang > > > Medd, Melinda (WSTF-RD)[ESCG] wrote: > >>We just installed Rdb JDBC version 7.25 and we are having trouble with >>installing the JDBC Thin Server. When we tried to test the Thin server >>we got this: >> >>PDIDS2>spawn/nowait/proc=rdbthinsrvtest java -jar rdbthinsrv.jar >>%DCL-S-SPAWNED, process RDBTHINSRVTEST spawned >>PDIDS2>pipeshow system|search sys$input rdbthinsrvtest >>Unable to listen on port 1701 >>java.net.SocketException: function not implemented >>at java.net.PlainSocketImpl.socketBind(Native Method) >>at java.net.PlainSocketImpl.bind(PlainSocketImpl.java:331) >>at java.net.ServerSocket.bind(ServerSocket.java:318) >>at java.net.ServerSocket.(ServerSocket.java:185) >>at java.net.ServerSocket.(ServerSocket.java:141) >>at oracle.rdb.jdbc.srv.RdbSrv.getServerSocket(RdbSrv.java:649) >>at oracle.rdb.jdbc.srv.RdbThinSrv.(RdbThinSrv.java:127) >>at oracle.rdb.jdbc.srv.RdbThinSrv.main(RdbThinSrv.java:251) >>%SEARCH-I-NOMATCHES, no strings matched >> >>Rdb Technical Service couldn't solve the problem expect thinking it was >>Process Software (our tcp/ip) the was the problem. We don't think so. >> >>So I am putting on the list to see if anyway else have the some problem >>and maybe a solution! >> >>Thanks, >> >>Melinda >>Melinda M. Medd >>Jacobs Technology Inc. >>NASA White Sands Test Facility >>Las Cruces, NM 88012 >> >>"Walking on water and developing software from a specification are easy >>if both are frozen." -Edward V. Berard, Life-Cycle Approaches >> > > -- > Dr. Wolfgang Kobarg-Sachsse > Principal Support Engineer > Oracle Support Services > ORACLE Deutschland GmbH > Hamborner Straße 51 > 40472 Duesseldorf > Germany > > ================================================ > > "Richard Whalen" wrote in message > news:f054h0$126$1@news.process.com... >> I downloaded the Rdb Java kit from Oracle's site and tried it on my >> Integrity system running OpenVMS 8.3 with TCPware 5.7-2 and experienced >> no >> problems. I tried it with both Java 4.2 and 5.0. >> >> >> --------------------------- >> Richard Whalen >> Process Software >> >> > > ------------------------------ Date: Fri, 20 Apr 2007 07:51:34 +0800 From: "Richard Maher" Subject: Re: Process Software: SocketException: function not implemented Message-ID: Hi Richard, > The comments in the driver say that UCX$_SHARE has been supported as an > IO$_SETMODE socket option since late 1994, which is longer than I have been > working at Process Software. I am not aware of any outstanding problem > reports with UCX$_SHARE. OK, if you say so, but this from Hunter circa Oct 2000 : - ======================================= Newsgroups: comp.os.vms From: goathun...@goatley.com (Hunter Goatley) Date: 2000/10/17 Subject: Re: time to consolidate the TCP/IP work on VMS On Tue, 17 Oct 2000 14:39:08 GMT, richard_ma...@my-deja.com wrote: >> Conversely, UCX requires OPER or SYSPRV to set the send >> and receive socket buffer sizes, but TCPware and MultiNet do not. > Touche. But surely it is reasonable for an unprivileged user to need >to reduce the connect timeout interval before needing to increase >resource usage? Agreed. I was just pointing another discrepancy that didn't make sense to me. >> >2) TCPWARE accepts, but appears to ignore, socket option >> >ucx$c_share. You get an error when you try to attach >> >another channel. >> Yes, I believe that's true. > My software does not currently make use of this feature but it seems >as though you don't imagine seeing this functionality in TCPWARE in the >forseeable future? FYI I think the error was "Device not ready, not >mounted or unavailable" There's never been a request for the feature, AFAIK, so there are no plans at the moment to implement that functionality (that's the foreseeable future). I'd like to see it added sometime, but there aren't any plans currently to do so (I've never seen an app that uses it). > Please let me take the opportunity to pass on a well done and thanks >to anyone who worked on your _BG: driver! The discrepencies that I >identified are minor (mainly cosmetic) and I'd like to point out that >99% of my, in all modesty fairly complex, code performed consistantly >across UCX and TCPWARE. The fact that the same executables will run >unaltered/linked regardless of TCP/IP vendor/implementation is >remarkable! Thanks again. Thanks! We try hard to catch everything, but some features we don't know about until someone uses them (e.g, the Apache code that used a feature of UCX we didn't know about, but we added it to both MultiNet and TCPware as soon as we learned about the problem). >PS. Any gotchas with muticasting? It is my understanding that TCPWARE >has had this functionality for years although it's new to UCX. Correct, and no gotchas that I know about. >PPS. Is the _BG: driver the interface of choice for your customers? >They're using your _inet: driver where I am at the moment and it looks >nasty. I've always thought the INET interface was much easier to use than the BG interface---you don't have to mess around with different kinds of item lists, etc. Different customers use different interfaces, but the BG interface has been growing in use because of the portability issues (although all 3 stacks support INET, too) and because the DEC C socket routines ultimately use the BG interface, and given all the changes in each DEC C release, we've been encouraging customers writing in C to use the DEC C socket routines instead of the TCPware socket routines. Hunter ------ Hunter Goatley, Process Software, ============================================= Don't suppose you could wander over and ask him could ya? Or does someone have a working example of a Socket being passed to another process through the wizardry of ucx$c_share that they'd like to share with us? (I could try to dig my old one up but I don't have a TCPware test box or site any more) Also, getting back to the original issue, does *anyone* know where or how the text "function not implemented" would bubble up from from TCPware to Java? As in "java.net.SocketException: function not implemented" from a Bind? Cheers Richard Maher "Richard Whalen" wrote in message news:f07o3c$rs4$1@ne ws.process.com... > The comments in the driver say that UCX$_SHARE has been supported as an > IO$_SETMODE socket option since late 1994, which is longer than I have been > working at Process Software. I am not aware of any outstanding problem > reports with UCX$_SHARE. > > > "Richard Maher" wrote in message > news:f0646q$nse$1@news-01.bur.connect.com.au... > > Hi Richard, > > > > Ok, I'll pass this on to the Rdb List Server. I would've thought there > > would've been an official call out with your support people by now? Maybe > > it's been and gone? fixed itself? > > > > To satisfy my curiosity, do you still not support the ucx$c_share socket > > option for TCPware? Does it still not give you an error at setmode time > > but > > rather fall over when the new process tries to take control of the socket? > > > > Cheers Richard Maher > > > > Here is some more on the original question: - > > > > =========================================== > > Melinda, > > > > I am not the owner of the Service Request, but I helped working on it in > > the background. > > > > Fact is that "pretty much any and all testing with the rdb product set on > > vms uses the HP supplied and supported TCPIP software. We > > test on nothing else." That's the reason why you have been asked in the > > SR to try HP TCPIP instead of TCPWare. > > > > Jim showed you that we are using the basic call to create a new Server > > Socket. And that is the line in our code where the error is returned. So > > why not checking this on your machine with the following simple program? > > If it fails as well, then our driver can be ruled out. > > > > import java.net.*; > > > > public class TestServer { > > > > public static void main (String[] args) throws Exception { > > if (args.length != 2) { > > System.out.println("Usage: java TestServer > > "); > > System.exit(1); > > } > > > > System.out.println("Host Address="+args[0]); > > System.out.println("Port ="+args[1]); > > ServerSocket ss = new ServerSocket(); > > > > ss.bind(new > > InetSocketAddress(InetAddress.getByName(args[0]), > > Integer.parseInt(args[1]))); > > > > System.out.println("ServerSocket Bound"); > > > > ss.accept(); > > } > > } > > > > $ java "TestServer.java" > > $ java "TestServer" localhost 1701 > > > > Returns for me: > > > > Host Address=localhost > > Port =1701 > > ServerSocket Bound > > > > You need to ctrl-y - stop the program in the case it does not fail. > > > > Best regards > > Wolfgang > > > > > > Medd, Melinda (WSTF-RD)[ESCG] wrote: > > > >>We just installed Rdb JDBC version 7.25 and we are having trouble with > >>installing the JDBC Thin Server. When we tried to test the Thin server > >>we got this: > >> > >>PDIDS2>spawn/nowait/proc=rdbthinsrvtest java -jar rdbthinsrv.jar > >>%DCL-S-SPAWNED, process RDBTHINSRVTEST spawned > >>PDIDS2>pipeshow system|search sys$input rdbthinsrvtest > >>Unable to listen on port 1701 > >>java.net.SocketException: function not implemented > >>at java.net.PlainSocketImpl.socketBind(Native Method) > >>at java.net.PlainSocketImpl.bind(PlainSocketImpl.java:331) > >>at java.net.ServerSocket.bind(ServerSocket.java:318) > >>at java.net.ServerSocket.(ServerSocket.java:185) > >>at java.net.ServerSocket.(ServerSocket.java:141) > >>at oracle.rdb.jdbc.srv.RdbSrv.getServerSocket(RdbSrv.java:649) > >>at oracle.rdb.jdbc.srv.RdbThinSrv.(RdbThinSrv.java:127) > >>at oracle.rdb.jdbc.srv.RdbThinSrv.main(RdbThinSrv.java:251) > >>%SEARCH-I-NOMATCHES, no strings matched > >> > >>Rdb Technical Service couldn't solve the problem expect thinking it was > >>Process Software (our tcp/ip) the was the problem. We don't think so. > >> > >>So I am putting on the list to see if anyway else have the some problem > >>and maybe a solution! > >> > >>Thanks, > >> > >>Melinda > >>Melinda M. Medd > >>Jacobs Technology Inc. > >>NASA White Sands Test Facility > >>Las Cruces, NM 88012 > >> > >>"Walking on water and developing software from a specification are easy > >>if both are frozen." -Edward V. Berard, Life-Cycle Approaches > >> > > > > -- > > Dr. Wolfgang Kobarg-Sachsse > > Principal Support Engineer > > Oracle Support Services > > ORACLE Deutschland GmbH > > Hamborner Straße 51 > > 40472 Duesseldorf > > Germany > > > > ================================================ > > > > "Richard Whalen" wrote in message > > news:f054h0$126$1@news.process.com... > >> I downloaded the Rdb Java kit from Oracle's site and tried it on my > >> Integrity system running OpenVMS 8.3 with TCPware 5.7-2 and experienced > >> no > >> problems. I tried it with both Java 4.2 and 5.0. > >> > >> > >> --------------------------- > >> Richard Whalen > >> Process Software > >> > >> > > > > > > ------------------------------ Date: 19 Apr 2007 19:25:34 -0700 From: george.constantinides@gmail.com Subject: Running NFS client on Open VMS Message-ID: <1177035934.732893.60550@y80g2000hsf.googlegroups.com> I want to run NFS client on a VMS box to access (read only) files on a Linux box. What do I need on the VMS side to make this happen. I am running OpenVMS V7.2-2, TCPIP V5.2 Thanks in advance ------------------------------ Date: Thu, 19 Apr 2007 22:36:59 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Running NFS client on Open VMS Message-ID: <4628271d$0$90267$14726298@news.sunsite.dk> george.constantinides@gmail.com wrote: > I want to run NFS client on a VMS box to access (read only) files on > a Linux box. > What do I need on the VMS side to make this happen. > I am running OpenVMS V7.2-2, TCPIP V5.2 Start by enabling NFS Client in TCPIP. :-) It is all in the docs - on the web see http://ftp.openvms.compaq.com/doc/83final/6526/6526pro_contents_006.html#toc_chapter_23 Arne ------------------------------ Date: Fri, 20 Apr 2007 05:21:30 +0200 From: "P. Sture" Subject: Re: Running NFS client on Open VMS Message-ID: In article <4628271d$0$90267$14726298@news.sunsite.dk>, Arne Vajhøj wrote: > george.constantinides@gmail.com wrote: > > I want to run NFS client on a VMS box to access (read only) files on > > a Linux box. > > What do I need on the VMS side to make this happen. > > I am running OpenVMS V7.2-2, TCPIP V5.2 > > Start by enabling NFS Client in TCPIP. > > :-) > > It is all in the docs - on the web see > http://ftp.openvms.compaq.com/doc/83final/6526/6526pro_contents_006.html#toc_c > hapter_23 > > Arne Here's a sample TCPIP command to mount a remote NFS share /SYSTEM $ tcpip mount dnfs2:[data] /host="unixbox" - /path="/usr/data" /system -- Paul Sture ------------------------------ Date: 19 Apr 2007 20:58:03 -0700 From: george.constantinides@gmail.com Subject: Re: Running NFS client on Open VMS Message-ID: <1177041483.602193.171060@b75g2000hsg.googlegroups.com> On Apr 20, 12:36 pm, Arne Vajh=F8j wrote: > george.constantini...@gmail.com wrote: > > I want to run NFS client on a VMS box to access (read only) files on > > a Linux box. > > What do I need on the VMS side to make this happen. > > I am running OpenVMS V7.2-2, TCPIP V5.2 > > Start by enabling NFS Client in TCPIP. > > :-) > > It is all in the docs - on the web seehttp://ftp.openvms.compaq.com/doc/8= 3final/6526/6526pro_contents_006.h... > > Arne Thanks, I will go through the on line docs. ------------------------------ Date: 18 Apr 2007 14:50:20 -0700 From: Ian Miller Subject: Re: SSM ? Message-ID: <1176933020.717695.211670@n76g2000hsh.googlegroups.com> I see the SWROL http://h71000.www7.hp.com/openvms/os/swroll/VAROL.HTML only mentions SSM 1.8 on VMS V8.2 but the SPD for SSM 1.8 http://h18000.www1.hp.com/info/SP6264/SP6264PF.PDF has the following list so I think you are OK "SSMgr V1.8 supports all operations on ODS-5 disks on systems running: OpenVMS Alpha: V7.3-2 (SPD 25.01.XX) V8.2 (SPD 82.35.XX) V8.3 (SPD 82.35.XX) OpenVMS I64: V8.2 (SPD 82.35.XX) V8.2-1 (SPD 82.35.XX) V8.3 (SPD 82.35.XX) " ------------------------------ Date: 19 Apr 2007 08:06:41 +0200 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Re: SSM ? Message-ID: <46272311$1@news.langstoeger.at> In article <1176933020.717695.211670@n76g2000hsh.googlegroups.com>, Ian Miller writes: >I see the SWROL >http://h71000.www7.hp.com/openvms/os/swroll/VAROL.HTML > >only mentions SSM 1.8 on VMS V8.2 but the SPD for SSM 1.8 > >http://h18000.www1.hp.com/info/SP6264/SP6264PF.PDF > >has the following list so I think you are OK > >"SSMgr V1.8 supports all operations on ODS-5 disks on >systems running: > >OpenVMS Alpha: V7.3-2 (SPD 25.01.XX) >V8.2 (SPD 82.35.XX) >V8.3 (SPD 82.35.XX) >OpenVMS I64: V8.2 (SPD 82.35.XX) >V8.2-1 (SPD 82.35.XX) >V8.3 (SPD 82.35.XX) So, only the VAROL is wrong. Somebody needs to fix it... Thanks a lot. -- Peter "EPLAN" LANGSTOEGER Network and OpenVMS system specialist E-mail peter@langstoeger.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ Date: 18 Apr 2007 22:39:37 +0200 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: [OpenVMS V8.3] MMOV ? Message-ID: <46269e29@news.langstoeger.at> Does anybody know the status of MMOV with OpenVMS Alpha V8.3? Will it run? Will it be supported (later)? TIA -- Peter "EPLAN" LANGSTOEGER Network and OpenVMS system specialist E-mail peter@langstoeger.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ Date: Wed, 18 Apr 2007 15:49:38 -0500 From: Dan Foster Subject: Re: [OpenVMS V8.3] MMOV ? Message-ID: In article <46269e29@news.langstoeger.at>, Peter 'EPLAN' LANGSTOeGER wrote: > Does anybody know the status of MMOV with OpenVMS Alpha V8.3? > > Will it run? Will it be supported (later)? 2.2 and 2.2-1 does run on OpenVMS/Alpha V8.3. I thought I remembered MMOV came as a part of either the OpenVMS dist CD set or the SPD. http://h71000.www7.hp.com/openvms/products/mmov/index.html in case you need it. The SPD mentions it works on OpenVMS/Alpha V7.1-2 or V7.2-1 and higher. http://h18000.www1.hp.com/info/SP6424/SP6424PF.PDF -Dan ------------------------------ Date: Wed, 18 Apr 2007 23:33:28 +0200 From: Paul Sture Subject: Re: [OpenVMS V8.3] MMOV ? Message-ID: In article <46269e29@news.langstoeger.at>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) wrote: > Does anybody know the status of MMOV with OpenVMS Alpha V8.3? > > Will it run? Will it be supported (later)? > > TIA I have MMOV 2.2-1 working on Alpha V8.3 (on my PWS 600 au). -- Paul Sture ------------------------------ Date: 19 Apr 2007 08:09:20 +0200 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Re: [OpenVMS V8.3] MMOV ? Message-ID: <462723b0$1@news.langstoeger.at> In article , Paul Sture writes: >In article <46269e29@news.langstoeger.at>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) wrote: >> Does anybody know the status of MMOV with OpenVMS Alpha V8.3? >> >> Will it run? Will it be supported (later)? >> >> TIA > >I have MMOV 2.2-1 working on Alpha V8.3 (on my PWS 600 au). So now we have 2 for "run" but 0 for "is supported"/"will be supported ..." I can live with it... Thanks a lot folks -- Peter "EPLAN" LANGSTOEGER Network and OpenVMS system specialist E-mail peter@langstoeger.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ Date: 19 Apr 2007 09:37:42 +0100 From: pmoreau@ath.cena.fr (Patrick MOREAU, DTI Athis ex CENA, Tel: 01.69.57.68.40) Subject: Re: [OpenVMS V8.3] MMOV ? Message-ID: In article <46269e29@news.langstoeger.at>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) writes: > Does anybody know the status of MMOV with OpenVMS Alpha V8.3? > > Will it run? Will it be supported (later)? > > TIA Hello, MMOV (the last version) runs fine under VMS 8.3 on DS15A. I don't know if it is supported (the DS15 sound card is not officially supported under VMS, only Tru64) but works. Patrick -- =============================================================================== patrick.moreau@aviation-civile.gouv.fr DSNA/DTI/EOS (ex SDER/CENA) ______ ___ _ Pôle XPE / / / / /| /| Athis-Mons France / /___/ / / | / | __ __ __ __ BP 205 / / / / |/ | | | |__| |__ |__| | | 94542 ORLY AEROGARE CEDEX / / :: / / | |__| | \ |__ | | |__| http://www.ath.cena.fr/~pmoreau/ http://membres.lycos.fr/pmoreau/ =============================================================================== ------------------------------ Date: 18 Apr 2007 22:54:58 +0200 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: [OpenVMS V8.3] SSM ? Message-ID: <4626a1c2$1@news.langstoeger.at> Does anybody know the status of Save Set Manager with OpenVMS Alpha V8.3? Will it run? Will it be supported (later)? TIA -- Peter "EPLAN" LANGSTOEGER Network and OpenVMS system specialist E-mail peter@langstoeger.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ End of INFO-VAX 2007.215 ************************