1 INFO-VAX	Sat, 01 Apr 2006	Volume 2006 : Issue 181       Contents: Re: DVD+RW drives on Alpha 8.x Re: EFN$C_ENF Chalk or Cheese? Re: OPA0 on DS10L problem  Re: OPA0 on DS10L problem  Re: OPA0 on DS10L problem . Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMSG Re: The Minimum You Need to Know to Be an OpenVMS Application Developer   F ----------------------------------------------------------------------  $ Date: Sat, 1 Apr 2006 16:13:41 +0200? From: "Eberhard Heuser-Hofmann" <vaxinf@chemie.uni-konstanz.de> ' Subject: Re: DVD+RW drives on Alpha 8.x 4 Message-ID: <010d01c65596$7ba45360$0301a8c0@athlon1>   Hi,   ; Because DVD-R(W) needs different commands. The same problem < arises if you want to burn DVD-RAM. No to forget Dual/double layer DVDs.   = If you want to use all media (DVD+-R(W), DVD-RAM, DVD+-R DL),  you should buy DVDwrite.   Eberhard  D PS: The first blue ray drive are avaliable, someone is interested to burn 25 GB/50 GB on one disk??   ----- Original Message -----  6 From: "Alan Frisbie" <Usenet02_REMOVE@Flying-Disk.com> To: <Info-VAX@Mvb.Saic.Com> & Sent: Saturday, April 01, 2006 6:04 AM' Subject: Re: DVD+RW drives on Alpha 8.x      > Hoff Hoffman wrote:  > F >>   There is a PCSI installation kit containing CD and DVD recording L >> software updates and diagnostics past what was shipped in and documented A >> in OpenVMS F8.3, and yes, the kit includes full documentation.  > E > Just out of curiosity, why does it only claim to support DVD+RW and  > not DVD-R(W)?  > 	 > Thanks,  > Alan     ------------------------------  $ Date: Sat, 1 Apr 2006 15:33:49 +08003 From: "Richard Maher" <maher_rj@hotspamnotmail.com> ' Subject: Re: EFN$C_ENF Chalk or Cheese? 1 Message-ID: <e0labp$ib6$1@news-02.connect.com.au>   	 Hi Steve,   E >   When a real event flag is raised, you'll get local processing for D > the IOSB, to see if it's gone non-zero.  If the IOSB remains zero,* > then the event flag is assumed spurious.  D So would it be fare to say that "When a real event flag is raised" =1 "whenever *any* of the 128 event flags are set."?    or  I "a $synch with efn$c_enf is equivalent to a $wflor of all 128 event flags  followed by the iosb check"   E >   Again, I cite that event flags are not my prefered solution when D > coding.  (It took some very severe application pain to convince meF > of this, but I did eventually learn.  I also receieved the requisiteG > volume of pain around locally-written and application-specific memory  > management, too.)   I I know exactly what you mean. I keep telling people that "threads" is not H only the most lamentable and deprecated programming paradigm of the lastH millenium, but also an officially recognized symptom of self-harm. MutexG this, agoraphobic-mutex that, and debugging is a complete nightmare. If L you're porting poorly written crap from Unix to VMS then it's great that VMSE does it's best to prop-up such shoddy practises, but if you ever even D contemplate the pssibility of developing a threads-based system fromL scratch, then it's time to reach-out. (Or get the hammer and nail out again,4 'cos it will be much less painfull in the long-run.)  K [Unfortunately, about *2-3% of all VMS CPU usage these days is spent on the I identity crisis of "Is process threaded?". Every System Service, nook and I cranny constantly asking itself this same bullshit question before it can I actually do anything. I think the loyal VMS client base out there needs a * performance boost from vmsNT (No Threads!)  L The good news is that John Reagan's lovely .Branch_Likely comes into play asD does "Snowflake's chance in hell" to mitigate the performance loss.]  E >   BTW, using your own sys$synch with EFN$C_ENF is somewhat unusual. D > (sys$qiow with an IOSB and EFI$C_ENF handles all this for you, for > instance.)  K I was about to say, "It's useful when you want to $synch at an outer mode." L but in the one example I have at hand, I can't for the life of me understandG why the same thing can't be achieved with a completion AST. Oh well, no J matter. I'd say it's useful anywhere $synch is useful and you haven't beenJ able to call lib$get_ef. Perhaps eliminating the false negatives discussedK in the docs? Or how 'bout parallelism? Your doing more than one I/O or more I than one lock but you don't need the first one to complete before kicking K off the second one and so on. Or you know your $enq roughly takes about one J I/O and a Lotto number generation. So you fit them in between the $enq and! the $synch? Wow is that ice thin!    Regards Richard Maher   G * I asked a guy in the street if he thought 2 to 3 percent sounds about  right.  ! <hoff@hp.nospam> wrote in message , news:4RDWf.5559$3R1.4868@news.cpqcorp.net...C > In article <e02gtr$r4a$1@news-02.connect.com.au>, "Richard Maher" % <maher_rj@hotspamnotmail.com> writes:  >  > |> How does EFN$C_ENF work?  > B >   It works, it helps performance, and it's really rather simple. > E >   I prefer to use it, and I use it in most every piece of code that D > I write now.  I particularly prefer to avoid all the SCHED fun andC > (particularly when dealing with common event flags, though events C > are also reported for local event flags) the rescheduling events.  > E >   When a real event flag is raised, you'll get local processing for D > the IOSB, to see if it's gone non-zero.  If the IOSB remains zero,* > then the event flag is assumed spurious. > E >   Again, I cite that event flags are not my prefered solution when D > coding.  (It took some very severe application pain to convince meF > of this, but I did eventually learn.  I also receieved the requisiteG > volume of pain around locally-written and application-specific memory  > management, too.)  > E >   BTW, using your own sys$synch with EFN$C_ENF is somewhat unusual. D > (sys$qiow with an IOSB and EFI$C_ENF handles all this for you, for > instance.) > E >   If details of the implementation are of interest, do see the IDSM  > and the source listings. >  >  >     ------------------------------  $ Date: Sat, 1 Apr 2006 10:46:56 +0200& From: "H Vlems" <nospam@what.ever.com>" Subject: Re: OPA0 on DS10L problem; Message-ID: <c98a1$442e3e43$513b818b$4992@news.versatel.nl>   0 "Tom Linden" <tom@kednos.com> schreef in bericht# news:op.s67bn5mvzgicya@hyrrokkin... F > Using a DB9 null modem cable I can connect from a PC using HypertermH > to either TTA0 or OPA0 on an XP1000, but only TTA0 on a DS10L.  I haveG > confirmed that the settings for OPA0 are identical for the two.  If I : > loop the cable on the DS10L from TTA0 to OPA0 I can both > * > SET HOST/DTE OPA0  and SET HOST/DTE TTA0 >  > Any ideas?   What kind of cable do you use?F The reason I ask is that you might be using an MMJ cable and DB9 - MMJ converters. J There are several models of the latter, the H8571-B and the -J. Use a pair of the same  model and you're fine.F Another reason is that the terminal may send noise and LOGINOUT.EXE is already K running, the $ SHO USER    command displays the terminal device and <login>  as the username.I Kill that process and you can (usually) log on, unless the terminal noise  immediately starts another copy of LOGINOUT.EXE   Hans   ------------------------------  $ Date: Sat, 1 Apr 2006 07:36:02 -0500* From: "FredK" <fred.nospam@nospam.dec.com>" Subject: Re: OPA0 on DS10L problem, Message-ID: <442e73b3$1@usenet01.boi.hp.com>  K "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote in message 0 news:442E0891.C58ABDE9@NeOaSrPtAhMlNiOnWk.net... > Tom Linden wrote:  > > 9 > > On Wed, 29 Mar 2006 17:51:20 -0800, David J. Dachtera 1 > > <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote:  > >  > > > Tom Linden wrote:  > > >>K > > >> Using a DB9 null modem cable I can connect from a PC using Hyperterm H > > >> to either TTA0 or OPA0 on an XP1000, but only TTA0 on a DS10L.  I haveL > > >> confirmed that the settings for OPA0 are identical for the two.  If I? > > >> loop the cable on the DS10L from TTA0 to OPA0 I can both  > > >>/ > > >> SET HOST/DTE OPA0  and SET HOST/DTE TTA0  > > >> > > >> Any ideas?  > > > $ > > > I'm not clear on the question. > > > K > > > Can you get the serial console on the DS10L? Which COM port is it on?  Is# > > > there more than one COM port?  > > > K > > > If DECwindows loads at startup, try disabling that and see if the COM  > > > port becomes OPA0. > > > < > > > Just a guess... Don't have a DS10L to experiment with. > > >  > > L > > The problem is that I can't connect to the OPA0 port (COM2) on the DS10LD > > but can to the TTA0 port (COM1) on same machine from a W2K using
 Hyperterm. > > F > > This works with an XP1000 with identical settings, as displayed byK > > SET TERM/FULL OPA0.  This has nothing to do with DECWindows.  Typically H > > when you connect in this manner you will get a console login prompt. > E > I think I'm being confused by the use of the word "connect" in this 
 > context. > I > It seems like you're saying that a DB9 <-> DB9 null modem cable between 2 > two Alphas works but not from the Alpha to a PC. >  > First guess time: pinouts? > I > Another possible guess would be that perhaps the PC terminal program is C > expecting the modem control signals DTR, CD, CTS and/or RTS to be J > asserted. They likely aren't. VT terminals had a setting for "data leadsJ > only". See if the terminal program has a similar setting. It may just beG > as simple as selecting XON/XOF flow control rather than hardware flow 
 > control. >   L I think that he probably has a good cable instead of a cheap cable.  A cheapD cable would have 3 or 4 wires and loop the modem signals back at the9 connector.  A good cable would run all the modem signals.   K The "driver" for OPA0 is in the firmware (that is, we indirectly talk to it L through the firmware) so there could be a difference in how it is respondingK to the signals.  Since VMS-to-VMS loopback works, I would suspect something - on the PC emulator.  Which one are you using?    ------------------------------  % Date: Sat, 01 Apr 2006 05:53:26 -0800 # From: "Tom Linden" <tom@kednos.com> " Subject: Re: OPA0 on DS10L problem) Message-ID: <op.s7b0vcg8zgicya@hyrrokkin>   7 On Fri, 31 Mar 2006 20:58:57 -0800, David J. Dachtera   - <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote:    > Tom Linden wrote:  >>8 >> On Wed, 29 Mar 2006 17:51:20 -0800, David J. Dachtera0 >> <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote: >> >> > Tom Linden wrote: >> >> J >> >> Using a DB9 null modem cable I can connect from a PC using HypertermI >> >> to either TTA0 or OPA0 on an XP1000, but only TTA0 on a DS10L.  I    >> have K >> >> confirmed that the settings for OPA0 are identical for the two.  If I > >> >> loop the cable on the DS10L from TTA0 to OPA0 I can both >> >> . >> >> SET HOST/DTE OPA0  and SET HOST/DTE TTA0 >> >>  >> >> Any ideas? >> ># >> > I'm not clear on the question.  >> >L >> > Can you get the serial console on the DS10L? Which COM port is it on?   >> Is " >> > there more than one COM port? >> >J >> > If DECwindows loads at startup, try disabling that and see if the COM >> > port becomes OPA0.  >> >; >> > Just a guess... Don't have a DS10L to experiment with.  >> > >>K >> The problem is that I can't connect to the OPA0 port (COM2) on the DS10L E >> but can to the TTA0 port (COM1) on same machine from a W2K using   
 >> Hyperterm.  >>E >> This works with an XP1000 with identical settings, as displayed by J >> SET TERM/FULL OPA0.  This has nothing to do with DECWindows.  TypicallyG >> when you connect in this manner you will get a console login prompt.  > E > I think I'm being confused by the use of the word "connect" in this 
 > context. > I > It seems like you're saying that a DB9 <-> DB9 null modem cable between 2 > two Alphas works but not from the Alpha to a PC. >  > First guess time: pinouts? > I > Another possible guess would be that perhaps the PC terminal program is C > expecting the modem control signals DTR, CD, CTS and/or RTS to be J > asserted. They likely aren't. VT terminals had a setting for "data leadsJ > only". See if the terminal program has a similar setting. It may just beG > as simple as selecting XON/XOF flow control rather than hardware flow 
 > control. > 5 > Too bad you don't have a real VT terminal to try...  > L Perhaps I wasn't clear, as you don't understand the problem, Please see my   exchange with John Santos.    ------------------------------   Date: 1 Apr 2006 18:08:17 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)7 Subject: Re: Opinion: I was just trying to sell OpenVMS + Message-ID: <497tshFn5dabU1@individual.net>   0 In article <fvidnXhtwdkrJLPZRVn-rw@bresnan.com>,% 	GreyCloud <mist@cumulus.com> writes:  > F > On bigger setups, like the Seattle hospitals, somebody turned their K > network of PCs into spambots.  Don't know if that can happen to VMS, but  7 >   I highly doubt that it can happen from the outside.   A We'll see.  As more and more people start putting PHP crap on VMS @ based web servers it is only a matter of time.  You can make theA OS as secure as you want, but if the users start leaving the barn 4 door open don't be surprised to find the horse gone.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Sat, 01 Apr 2006 10:48:38 -0700 " From: GreyCloud <mist@cumulus.com>7 Subject: Re: Opinion: I was just trying to sell OpenVMS 0 Message-ID: <CvOdnRBaAMBkIbPZRVn-hg@bresnan.com>   Michael Kraemer wrote:   > GreyCloud schrieb: >  >>K >> Yet no one could crack VMS at the Las Vegas DEFCON trials.  How is that?  >  > 4 > That's nothing, because it doesn't prove anything,= > except that at that particular event people didn't succeed. . > And as others (e.g. BillG) have pointed out,1 > the most probable reason is, that those hackers - > simply are not interested in an obscure OS. 5 > Anyway, following scientific logic, you can't prove 4 > that something doesn't exist, e.g. a hack for VMS.7 > You can't even prove that Santa Clause doesn't exist. L > (Although Santa Clause's existence is much less likely than a hacked VMS).. > If you want to do your favorite OS a favour,8 > don't come up with that old story over and over again.G > Better find something positive, more visible to a potential end user.  >   > Tell that to those that use VMS in the stock market exchanges.= Tell that to US military that still has contracts for JStars.      --   Where are we going?   And why am I in this handbasket?   ------------------------------   Date: 1 Apr 2006 18:15:47 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)7 Subject: Re: Opinion: I was just trying to sell OpenVMS + Message-ID: <497uajFn5dabU2@individual.net>   0 In article <CvOdnRBaAMBkIbPZRVn-hg@bresnan.com>,% 	GreyCloud <mist@cumulus.com> writes:  > Michael Kraemer wrote: >  >> GreyCloud schrieb:  >>   >>> L >>> Yet no one could crack VMS at the Las Vegas DEFCON trials.  How is that? >>   >>  5 >> That's nothing, because it doesn't prove anything, > >> except that at that particular event people didn't succeed./ >> And as others (e.g. BillG) have pointed out, 2 >> the most probable reason is, that those hackers. >> simply are not interested in an obscure OS.6 >> Anyway, following scientific logic, you can't prove5 >> that something doesn't exist, e.g. a hack for VMS. 8 >> You can't even prove that Santa Clause doesn't exist.M >> (Although Santa Clause's existence is much less likely than a hacked VMS). / >> If you want to do your favorite OS a favour, 9 >> don't come up with that old story over and over again. H >> Better find something positive, more visible to a potential end user. >>   > @ > Tell that to those that use VMS in the stock market exchanges.  ; The stock market exchange machines are not on the INTERNET.   ? > Tell that to US military that still has contracts for JStars.    JStars is not on the INTERNET.   What exactly was your point?  F And, just as a point of reference, JStars is ancient (or should I justF use the "legacy" word!)  None of the current replacement projects thatI are being worked on (that I have become aware of) utilize VMS in any way.    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------   Date: 1 Apr 2006 20:09:02 +0100 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)7 Subject: Re: Opinion: I was just trying to sell OpenVMS , Message-ID: <442eddde$1@news.langstoeger.at>  U In article <CvOdnRBaAMBkIbPZRVn-hg@bresnan.com>, GreyCloud <mist@cumulus.com> writes: ? >Tell that to those that use VMS in the stock market exchanges.   B I know of a stock exchange and a bank where VMS is on its way out.B Not because they like, more because they need to (applications)...   It is still time to market VMS.    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Sat, 01 Apr 2006 12:09:36 -0600 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>P Subject: Re: The Minimum You Need to Know to Be an OpenVMS Application Developer6 Message-ID: <442EC1E0.108F2EBD@NeOaSrPtAhMlNiOnWk.net>   "Dr. Dweeb" wrote: >  > Well Arne, >  > You clearly love MySQL.    No doubt, is has its uses.  B > However, it is a toy database for toy applications - and explain; > exactly how MySQL is *not* "any other ISAM file library".   H Is there an RTL that you can link against, ala RMS on RSTS/E? Perhaps anH .OLB? Is it integrated into any 3GLs, ala RMS on RSTSE/E, VMS, etc. suchF that a language built-in can be used rather than a call to an external routine?  G > Until recently (and maybe still), one did not have to go far into the G > manual to realise that the developers were totally clueless as to the D > requirements of a database per se, and of a relational database in
 > particular.  > I > Faulty understanding of the issues involved led to fundamentally flawed B > product.  the fact that InnoBase even exists, is, well, proof byE > existence, that the original and current MySQL is seroiusly flawed.  > H > And I am sorry, the notion that dat integrity is the responsibility ofH > the application, as the developers of MySQL so fondly recounted in theG > manuals says exactly everything you need to know about their level of H > understanding of the theory, or for that matter, practice of databases4 > in general and relational databases in particular.  D What protection does RMS provide against assigning, for example, theH content of string field into a quadword integer? ...or the content of anB unsigned longword to the memory space occupied by a floating point field?  % > If you want a glorified file system   F Is there "a glorified file system" underlying or provided with Oracle?? Seems to me it depends only on the extreme low-level primitives E underlying ODS, but side-steps enough of it that it can't even handle  bound volume-sets.  A AFAIK, Oracle provides no file system services outside of itself.   ! > with a toy SQL on top if it for E > non-critical applications - fine, use MySQL.  I can conceive of few E > applications with even the most limited data integrity and security 2 > requirements that would be well served by MySQL.  H If sloppy programming practice is the rule of the day, even Oracle won't help.    --   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/   ------------------------------   End of INFO-VAX 2006.181 ************************