0 INFO-VAX	Mon, 05 Feb 2001	Volume 2001 : Issue 71      Contents: Re: Dave Cutler  Re: Dave Cutler  Re: Dave Cutler & Re: Dual ethernet config for failover?3 Re: Firmware upgrade for PWS 600au to use a ZLXp-L1 8 Re: Formal Letter to the Newsgroup from Richard Marcello help fms forms development Re: help fms forms development Re: help fms forms development. I WILL MAKE YOU RICH BEYOND YOUR DREAMS.  8364J Re: Is there a way to execute an Alpha OpenVMS GUI application onaremote s Re: Lisbon Conference ( List of cdroms working with DEC hardware Looking for a good SSH client ! Re: Looking for a good SSH client ! Re: Looking for a good SSH client ! Re: Looking for a good SSH client  Re: MMS/MMK Ghostscript problem  Re: MMS/MMK Ghostscript problem ; Re: NT64-on-Alpha: Microsoft Business Partner = Organ Donor  Re: NTP on VMS 7.1 Re: NTP on VMS 7.1 OpenVMS / UNIX + COE RE: OpenVMS / UNIX + COE OPENVMS 7.3....When? Re: OPENVMS 7.3....When? RE: OPENVMS 7.3....When?- Re: Plenum vs. non-plenum in the data center? - Re: Plenum vs. non-plenum in the data center? - RE: Plenum vs. non-plenum in the data center?  Re: Source listings contents RE: Source listings contents Re: sysuaf question % UCX SMTP duplicate mail sending HELP! + Re: Unrecognized Individual Destination PDU ' Re: Writes flushed to Disk on Dismount? ' RE: Writes flushed to Disk on Dismount? ' RE: Writes flushed to Disk on Dismount? ' RE: Writes flushed to Disk on Dismount?   F ----------------------------------------------------------------------  % Date: Sun, 04 Feb 2001 23:27:55 +0000 ) From: Christof Brass <brass@infopuls.com>  Subject: Re: Dave Cutler, Message-ID: <3A7DE57B.1719FF59@infopuls.com>   JF Mezei wrote:  >  > Christof Brass wrote: ? > > headed Micro$oft systems. Why doesn't Micro$oft support the  > > Alpha processor??  > J > Why should a company spend any money porting to an architecture that hasR > nearly zero marketing and whose markets are being limited to just a few niches ? > O > The Alpha chip may not be dead and there may be a few more iterations left in P > its lifetime, but not only has it lost some of its speed/performance edge, butP > Compaq has no interest in spreading Alpha to the rest of its product line. The/ > chip may not be dead, but its marketplace is.  > O > If Microsoft were to procductize Windows on Alpha, how would Compaq sell it ? O > Since Microsoft is only interested in dealing with the Compaq parts that have N > the permission to grow and advertise heavily, this excludes anything running > on the Alpha.  > L > And of Compaq were to start selling Alphas at the lower end, how would itsE > relationship with Intel go ? And would Compaq make any more money ?  > P > Besides, the current Alpha performance is probably either marginally higher or- > about the same as the 8086. So why bother ?  > L > You know, now that Microsoft is entering the hardware business, I think itP > would be a riot if Microsoft were to start building Alpha based computers, putO > NT on them and sell them with all the marketing might MS has. And it would be O > an even bigger riot if Microsoft were to start selling VMS to all the markets M > that Compaq refuses to sell VMS to. What a slap in the face and kick in the  > butt Compaq would be getting.  > N > But in many ways, that is what Compaq deserves. Someone else should start toP > sell VMS and Alpha in products Compaq refuses to sell them to and prove to theP > world how incomopetent Compaq was when t decided to limit the market for alpha
 > and vms.  @ I don't want to repeat what I replied to Terry's response (after@ you posted that, of course) which should answer the questions of= the first three paragraphs. So I hope you'll be reading that. / Instead I want to add a few different thoughts.   @ Compaq is not depending on Intel. There are several alternatives; and AFAIK Compaq already started selling AMD based systems.   @ Using Alpha has the technical advantage of beeing free of legacy> instructions. Technically there is no doubt that with the same: effort the Alpha can't be beaten by IA-32. With the memory? addressing capabilities of 64 Bit there should also be a market ? place for the Alpha. As you pointed out: marketing is needed to # get your stuff to the market place.   1 Do you know what processor the XBox is featuring?   ; Now that Intel is entering the SW business (they don't tell 8 anybody what OS they are using for their Internet boxes)4 Micro$oft should be very interested in supporting an alternative.   ------------------------------  # Date: Mon, 05 Feb 2001 03:58:48 GMT ; From: Mark Garrett <Mark.Garrett@wedontwantyourspam.com.au>  Subject: Re: Dave CutlerC Message-ID: <B6A46D70.114F8%Mark.Garrett@wedontwantyourspam.com.au>   6 in article 3A7C5BCF.9C67100A@videotron.ca, JF Mezei at7 jfmezei.spamnot@videotron.ca wrote on 04/02/2001 06:28:    > Christof Brass wrote: > >> headed Micro$oft systems. Why doesn't Micro$oft support the >> Alpha processor?? > J > Why should a company spend any money porting to an architecture that hasP > nearly zero marketing and whose markets are being limited to just a few niches > ? K     Its not a matter of porting, I think your just trying to stir otherwise K your not reading. The implication so far is that its been there ie. NT4 and L is still there as they not only started W2K on AXP eg beta's  the posters soH far have said that the Windows 64bit effort is still producing builds on AXP. > O > The Alpha chip may not be dead and there may be a few more iterations left in P > its lifetime, but not only has it lost some of its speed/performance edge, butP > Compaq has no interest in spreading Alpha to the rest of its product line. The/ > chip may not be dead, but its marketplace is. F Don't know why you say that since they have targeted the Tandem sector$ machines to move off MIPS onto ALPHA   > O > If Microsoft were to procductize Windows on Alpha, how would Compaq sell it ? 2                       ^ insert the word AGAIN hereJ And I guess the answer is already part of history. Its really up to COMPAQH rather than MS as CPQ would need to revamp some sort of common referenceH platform with ARC before MS could commit to providing releases. Which isC what they decided to stop doing. Maybe some of the OEM'ers could do K something, hey maybe all you alpha+windows lovers could get together and do K the console hardware support and sell OEM'd motherboards and convince MS it K was worth there time to make the code releases official. But hey, who would J want to buy a little box running an AXP processor with Windows OS when theD intel one is so much cheaper hmmm (lets not forget the 3rd party appK support). CPQ did the right thing when Digital couldn't - get the hell away L from AXP NT all of the other vendors dropped NT long ago. The only reason MSK was ever interested in more than one hardware platform was that at the time I it was beginning UNIX was the Open System leader as it ran on everybody's J hardware and that's where they thought they needed to be to compete. TheseK same people had come from years during the 80's bashing themselves to death J with OS/2 trying to compete against UNIX, they were going to do everythingG they could not to be left out again. Also it was in MS interests to use  Digital L and IBM to pave the way into corporate hearts and legitimise that this PC OSL was OK, even if hard core techos didn't like it. Since all the work has beenK done in this arena there is little value left to MS to build further on the H relationship. Besides I bet there are still plenty of well programmed MSK droid's in CPQ and I don't mean the traditional CPQ which was always wintel I I mean the Digital flavoured ones. Though I have heard hints that they no  longer as popular.           > L > And of Compaq were to start selling Alphas at the lower end, how would itsE > relationship with Intel go ? And would Compaq make any more money ?  > P > Besides, the current Alpha performance is probably either marginally higher or- > about the same as the 8086. So why bother ?  Ok now I know your a DH.              ------------------------------   Date: 4 Feb 2001 23:45:40 -0500 2 From: young_r@eisner.encompasserve.org (Rob Young) Subject: Re: Dave Cutler3 Message-ID: <4+QbLgSGQwNj@eisner.encompasserve.org>   \ In article <3A7C5BCF.9C67100A@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > O > The Alpha chip may not be dead and there may be a few more iterations left in P > its lifetime, but not only has it lost some of its speed/performance edge, butP > Compaq has no interest in spreading Alpha to the rest of its product line. The/ > chip may not be dead, but its marketplace is.  >   < 	Its performance edge hasn't even begun yet.  Everyone talks= 	about leapfrogs in CPU performance, blah , blah , blah.  But > 	EV7 and EV8 and various shrinks and processes associated with< 	each (.18 , .13 , copper, copper-SOI) will stick Alpha back< 	at the top for quite a while.  Interested in learning about: 	cutting edge CPU technologies?  Read up on the three part 	EV8 series in RealWorldTech:   ? http://www.realworldtech.com/page.cfm?ArticleID=RWT121300000000   ? http://www.realworldtech.com/page.cfm?ArticleID=RWT122600000000   ? http://www.realworldtech.com/page.cfm?ArticleID=RWT011601000000   " 	What about present Alpha growth??  @ 	Or better yet, you are overlooking the obvious.  Alpha revenues? 	are going to increase at least 40-60% over the next 3 years or C 	so.  NSK is coming over and natural growth.  Suppose the migration D 	to Alpha scares off 40% of the NSK customer base (highly unlikely),C 	that would still peg NSK hardware revenues in the $600 million per @ 	year range (with no growth 3 years out) adding substantially to@ 	Alpha revenues.  You paint Alpha in decline without a shred of E 	supporting evidence.  Drum something up, anything. . .  Show us the  < 	decline.  Otherwise , what you state in the paragraph above8 	is nothing but pure FUD.  Andrew could do a LOT better.   				Rob    ------------------------------  $ Date: Sun, 4 Feb 2001 15:01:18 -05002 From: "John Gemignani, Jr." <john@ossc.DELETE.net>/ Subject: Re: Dual ethernet config for failover? + Message-ID: <3a7db556$1@newsfeed.vitts.com>   K I did a little looking into this within UCX (and TCPIP works the same way). K While you could have two interfaces on the same subnet, datagrams sent from K the system will always go out over the same interface.  The choice of which I one is an ordering issue, which was defined first or last, I don't recall  which.  K So, you end up receiving datagrams on both adapters but only send responses  out the first one.  E There is a facility called "netrain", which I believe will allow load D balancing and failover.  This has not been implemented in TCPIP yet.   -John   J "Matt Muggeridge" <Matt.Muggeridge@compaq.com.spammffree> wrote in message/ news:F61b6.27$cu.319@gazette.loc1.tandem.com... 5 > > 1. How does DNS resolve a name to 2 IP addresses?  > I > DNS uses round-robin when handing out addresses.  However, this doesn't  giveG > you failover.  In fact, if one of the interfaces fails, then DNS will I > dutifully continue to handout both interfaces in a round-robin fashion, G > which will mean some users will have intermittent trouble connecting.  That > would be a pain. > H > To resolve this, you can add the load-broker and metric server to yourK > network.  The load-broker will poll the metric server, running on each of I > the participating nodes.  If the metric server fails to respond after 3 C > attempts, the load-broker will remove that interface from the DNS 	 database. - > This requires Dynamic DNS to be configured.  > B > This fixes the inbound traffic problem by preventing anyone from
 connectingH > to the bad interface, however, there is another problem.  The outboundK > traffic will consult the local routing table to determine which interface  toI > use to send data.  So if the bad interface reamains configured, then it  may K > be selected for outbound traffic, which of course also causes the network  toK > fail.  To address this, you might develop a command procedure which pings K > the default router.  If it fails (after 3 attempts, say), then remove the F > failed interface, (TCPIP SET NOINTERFACE).  This will remove the badI > interface from the routing table, and so you should continue operating. I > (There are other factors too - e.g. are interfaces in the same subnet.)  > J > Then you can make use of the internet cluster alias, which requires yourL > nodes to be configured in a VMS cluster.  If one node in the cluster failsL > (rebooted etc) then one of the other cluster members will become the ownerJ > of the cluster IP address.  (The cluster alias doesn't give you failover9 > between controllers, it is a failover between systems).  > K > Finally, with multiple controllers, you could attach each controller to a H > different subnet, then place a router between the subnets.  Then if anJ > interface fails, the router should be able to dynamically reroute to theK > available interface.  (I haven't tried this configuration, but have heard E > others speak of it, and in theory it should work - you will need an 4 > excellent working knowledge of routing protocols). > 	 > Cheers,  > Matt.  >  >  >    ------------------------------  % Date: Mon, 05 Feb 2001 00:18:36 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)< Subject: Re: Firmware upgrade for PWS 600au to use a ZLXp-L1L Message-ID: <rdeininger-0502010018360001@user-2ivecic.dialup.mindspring.com>  W In article <3A7D77E6.B1D6B32F@infopuls.com>, Christof Brass <brass@infopuls.com> wrote: @ > Thanks a lot!! I started this morning writing a message to you@ > directly because I knew from my searches in Deja and following/ > the threads in cov that you are *the* expert     Yikes!  If Deja gave you the impression I am an expert, it's more broken than I thought.  I'm just throwing out free advice, which is probably worth exactly what it costs...   < There is a Compaq graphics expert who does post sometimes...   > (with an appology = > at the beginning for bothering you by direct mail) but then ? > decided to do further investigation and abandoned writing the ; > email later after finding out that I might be able to use = > several of the current graphics cards within my DS20E. Some @ > months ago a compaq sales person told me that within the Alpha@ > servers there is only one graphics card supported. So I looked@ > around to buy a front-end box to equip with 2 or 3 heads. From? > the latest specs I found at the Compaq web site it seems that ? > the DS20E supports up to four heads with VMS which is awfully 	 > great.     I think I remember reading that some systems have been tested with up to 7 heads.  It probably comes down to the number of PCI slots available to test with.  I expect the driver would work just fine with more heads, but testing was not practical.  Q Note that the new desktop DECwindows stuff is a little rough with multiple heads.   9 > The remaining question is: are these new graphics cards A > SN-PBXGD-AD (102454-B21) P300 and SN-PBXGD-AE (102455-B21) P350 B > of good quality in comparison to the ZLXp-L1? I want to have bigB > colour palettes to avoid having to start several apps with their= > private colour palettes which changes the colour of all the  > other windows.  I don't know these cards.  The newest I have used is the 3D30.  You should be able to get exact specs on the P300 and P350, since they are current.  As far as I know, the PCI ZLXp-L1 is functionally the same as the turbochannel ZLX-L1.  The latter card is described in a 1995 SOC as follows: 24-plane 3D turbochannel color grphics accelerator, 24-bit image, 24-bit Z buffering, 24-bit double buffer, 4-Mpixels total graphics memory.  The order number was PMAGC-DA.  I BTW, all the ZLX-E, -L, and -M cards were listed as having VMS support.      The reason I think the PCI and turbochannel cards were the same is due to a bit of description in the Open 3D docs, where IIRC the two flavors were lumped together.  a I would assume that with 24-bit color, reasonable programs should not have trouble with palettes.    I thought from the first post that you have a ZLXp-L1 in hand, so I suggested some experiments.  If you don't have the card, you probably shouldn't buy it, even at bargain prices, unless you have some assurance that it will work.  Also, mixing different cards in the same machine may be trouble.  On the other hand, these old cards do show up cheap sometimes, and it you can get them to work in a new system, they might meet your needs.  I haven't priced the P300 and P350, but I've seen a lot of complaints from   -- x Robert Deininger rdeininger@mindspring.como   ------------------------------   Date: 5 Feb 2001 03:24:05 GMTa2 From: dilalo@AESOP.RUTGERS.EDU (Gregory J. DiLalo)A Subject: Re: Formal Letter to the Newsgroup from Richard Marcello 1 Message-ID: <95l6cl$inc$1@newsmonger.rutgers.edu>b  Y In article <3A7AA15F.277E709F@bbc.co.uk>, Tim Llewellyn <tim.llewellyn@bbc.co.uk> writes:  >o' >"Brian Schenkenberger, VAXman-" wrote:1 >pi >> In article <95d2gr$hv0$1@newsmonger.rutgers.edu>, dilalo@AESOP.RUTGERS.EDU (Gregory J. DiLalo) writes:. >> >R >> >The sky is falling...the sky is falling...  Come on people!  Why are you doingS >> >this to yourselves?  We know that OpenVMS AlphaClusters are the best damn thing Q >> >there is.  I've been managing VMS systems for nearly 20 years now and I neveriQ >> >thought 20 years ago I would be doing so today.  I wouldn't dream of managingtQ >> >our 80 million dollars of teaching and research funds, our core internet mailtK >> >and network services, and yes, our web servers with anything other than-M >> >an OpenVMS cluster.  I sleep at night.  StorageWorks with FibreChannel is O >> >phenominal technology that moves data like I never imagined anything could.v >> >R >> >Compaq has to compete in multiple markets successfully to remain strong.  TheyO >> >target their marketing dollars where it will have the most impact.  I watch Q >> >my people knock themselves out with Windows 2000 server on the Intel platformcN >> >and just shake my head in amazement.  Likewise, the headaches  our SolarisN >> >admins endure are pitiful.  Me?  I know what you know and my life is good. >> > >iP >Good, I am happy for you, just watch out for those bean counters who are coming >to take your VMS away.... >s >> > --e >i7 >Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project 1 >MedAS at the BBC, Whiteladies Road, Bristol, UK. B >Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uk >eB >I speak for myself only and my views in no way represent those of >MedAS or the BBC. >  >h  O Fortunately, for me, I am a bean counter.  I spent my years in graduate school eI learning applied economics.  When I put a budget together, OpenVMS is our,O lowest cost of ownership platform.  It's an easy sell from the TCO perspective.eM The numbers just don't lie, due primarily to the way AlphaClusters on OpenVMSyL scale.  Anytime Windows or Unix comes in lower, you can bet there are costs O being buried somewhere.  Of course, I had to point out the flaws in methodology : used by a few prominent IT consulting firms along the way.   Greg  P -------------------------------------+------------------------------------------N Gregory J. DiLalo                    | Voice (Work):   (732) 932-1100 Ext. 411F Management Information Services      | Voice (Home):   (732) 257-6969 E Cook College/NJAES                   | FAX:            (732) 932-8887hO Rutgers, The State University of NJ  | Internet:       dilalo@aesop.rutgers.edusP -------------------------------------+------------------------------------------   ------------------------------  # Date: Sun, 04 Feb 2001 23:00:52 GMTa From: allan_agnew@my-deja.comT# Subject: help fms forms developmenta) Message-ID: <95kmv3$iao$1@nnrp1.deja.com>i   Hi, / If this is this wrong forum for my question, mye0 apologies, and would someone mind pointing me to the right one:  - Whether wisely or not, at a point in the pasta( when we wanted to ensure we needed be no) longer dependent on Allin1, we decided to . emulate the Allin1 menu system, enabling us to. use all the existing forms and form libraries./ This worked well, and allowed us to turn Allin1t. off, except that I have found it very handy to- use it myself for continuing form developmentm* only.  Under Allin1 form development, when. updating a form in a form library, you can use1 the "edit named data" option, which opens an EDT-e. like editor, displaying the named data without( line indices, and allowing you to easily0 manipulate the named data, such as inserting new
 data, etc.  % Without Allin1, I need another way of - manipulating the named data, and I can't find-% an easy way within FMS of doing this.M. 1  Using FMS/edit, I can't see how to insert a line../ 2  Using FMS lib/extract, descrip, then editingB- the .flg and using FMS/translate, would work,o- except that the indices to the named data aren/ always presented, and require manually changingo/ when any new lines are inserted, which would bee* very tedious and prone to error. I suppose, changing the indices could be automated, but' I'd rather not have to see them at all.y  , So, is there a way using FMS, that I haven't/ found, or does Allin1 use a stand-alone program 0 which I can run without requiring any other bits
 of Allin1?         Sent via Deja.coms http://www.deja.com/   ------------------------------  % Date: Sun, 04 Feb 2001 21:02:25 -0500s- From: JF Mezei <jfmezei.spamnot@videotron.ca>e' Subject: Re: help fms forms developmenth, Message-ID: <3A7E09B0.56ED9C87@videotron.ca>   allan_agnew@my-deja.com wrote:. > So, is there a way using FMS, that I haven't1 > found, or does Allin1 use a stand-alone programh2 > which I can run without requiring any other bits > of Allin1?   yes.   ------------------------------  % Date: Sun, 04 Feb 2001 21:19:51 -0500b- From: JF Mezei <jfmezei.spamnot@videotron.ca>y' Subject: Re: help fms forms developmente, Message-ID: <3A7E0DC5.F150FF4F@videotron.ca>   allan_agnew@my-deja.com wrote:. > So, is there a way using FMS, that I haven't1 > found, or does Allin1 use a stand-alone programr2 > which I can run without requiring any other bits > of Allin1?  I Well, you didn't expect to get it so easily, did you ? I figured that youoM deserved a bit of anticipation, so decided not to give it to you in the firstsA post and let you suffer a bit of anticipation :-) :-) :-) :-) :-)M  I The following will edit the named data of the form supplied as P1, in theo forms library USER.FLB    	 NAMED.COMp- $FMS/DESC/FULL/output=temp1.flg user/form='p1S  $define/nolog FDND1IN  temp1.flg  $define/nolog FDND1OUT temp1.flx  $define/nolog FDND2OUT TEMP2.FLG $run oa$lib:fdndedt1 $define/user sys$input TTh $edit fdnd1out $run oa$lib:fdndedt2 $fms/translate temp2.flg $fms/lib/replace user temp2.frmh $del temp2.frm;* $del temp1.flg;* $del temp1.flx;* $del temp2.flg;*    0 And shame on you for dropping ALLIN1 :-) :-) :-)   ------------------------------  # Date: Sun, 04 Feb 2001 14:26:18 GMT  From: brucepm@pacbell.neti7 Subject: I WILL MAKE YOU RICH BEYOND YOUR DREAMS.  8364': Message-ID: <eEdf6.13792$GV6.674543@dfiatx1-snr1.gtei.net>  - CONTACT ME FOR DETAILS ON GETTING FILTHY RICHr   BRUCE P. MEYERSa 10517 VALPARAISO STREET  LOS ANGELES, CA 90034o PHONE:(310)202.9005h CELL:(310)213.1148 E-Mail:brucepm@pacbell.net   qsgqxehptrrowsqvcA   ------------------------------  # Date: Sun, 04 Feb 2001 20:37:20 GMTo8 From: Veli =?iso-8859-1?Q?K=F6rkk=F6?= <korkko@decus.fi>S Subject: Re: Is there a way to execute an Alpha OpenVMS GUI application onaremote se( Message-ID: <3A7DB829.71F8DC82@decus.fi>  7 If the original application context does not have to bet8 available to the user whilst (s)he is using the new one, I'd imagine that one can e.q.c  6 - setup number of AUTOLOGIN only account on the target   Alpha OpenVMS system6 - setup ALF database accordingly on the target machine: - have these accounts to start the Alpha app automatically   as part of login procedure - have the application issue$ 	$ define/user sys$input sys$command 	$ set host ALPHAp  3 and it should be somewhat transparent to the users b3 that they are actually now using completely anotherk6 machine since they were not even prompted for username
 and password.    _veli    Norman Woo wrote:  > 1 > On Fri, 02 Feb 2001 13:31:29 +0000, Nigel ArnotR& > <sysmgr@maxwell.ph.kcl.ac.uk> wrote: > 
 > Hi folks > ? > Sorry, I should have meant "CUI" the ole Character-based User 0 > Interface (or is that "Cheap User Interface"?) > E > Basically we have users coming in from PCs (via Telnet sessions) or H > dumb ole VT220 terminals accessing the main application running on theG > DEC 4700 machines (app was built back in the 80's).  From within thatoE > main application, we would need to somehow remote login (rlogin? or B > telnet?) into the Compaq Alpha server and lauching another "CUI"A > application.   This would be displayed on the user's PCs telnetE$ > session or on the VT220 terminals. >  > Hope this is clearer.. >  > Thanks > 
 > >> Hi folksc > >>K > >> We have users running a GUI VAX VMS application on a DEC 4700 (runninggI > >> VAX VMS 5.5-2).   There is also another GUI application that residesd5 > >> on an Alpha DS20E machine running OpenVMS 7.1-2.I > >>K > >> We would like to include in the menu in the application running on the0K > >> DEC 4700 the name of the remote application.  When a user selects thisaJ > >> menu item, we need to somehow automatically log the user to the AlphaH > >> DS20E and start the application.  This should be transparent to theF > >> user (hs/she doesn't even know that the application that they are' > >> accessing are on another machine).  > >>J > >> Can this be done?  Or are there any tools/3rd party packages that can
 > >> do this?  > >> > > @ > >Assuming by GUI, you mean X-windows, it should be quite easy. > >pB > >The menu application first needs to establish where the user is@ > >coming from (ie that which is needed to execute a SET DISPLAYH > >that will send subsequent X-displays back to that user. $SHOW DISPLAYE > >(from an appropriate context!) is one way to get this information.- > >-H > >Next, it has to tell the DS20E to run that application in the contextK > >of the chosen user. The way I'm familiar with is DECNET IV task-to-task.rC > >You define the task in the DS20's task database (a .COM or .EXE)o> > >and fire it up from the menu app just by opening the "file"> > >called DS20"user password"::"task=name" for read and write. > >oF > >This will fire up a process in the user's context running the task-B > >defined .COM, and that should open the file SYS$NET read-write. > >nJ > >There's now a two-way channel between the menu app and the remote task.D > >So the nenu app can write the information necessary to define theF > >display application to that pipe, the task can read it, execute theH > >appropriate SET DISPLAY command to send subsequent X displays back toP > >the appropriate X-display station, and fire up the appropriate X-application. > >eP > >DECnet V is much more complex to set up, but basically similar. AlternativelyJ > >you can use TCP/IP (UCX) and RSH or REXEC. If the DS20 app doesn't haveP > >to run in the context of a variable user, you can define a particular user-IDK > >to run it in the task definition, and miss the out "user password" fieldhN > >(and the associated possible risk of passwords flying around your network).@ > >It's also not needed if the user-id needed on the DS20 is theM > >same as the user-id that the menu-app is running under and provided DECNETr > >proxy logins are enabled. > >tN > >It might be easier (depending on what the user's GUI-display server is/are)O > >to define a shortcut on that platform that goes straight to the DS20, rathereF > >than involving that middle-man menu app. There again, it might not. > >m> > >> Thanks in advance (I'm a VAX newbie so go easy on me ...) > >> > >e > >You're welcome! > >hJ > >Appended an example, all in DCL. If this procedure is defined as a task) > >on a remote machine REMOTE, you can dof > >b > >@tell REMOTE dcl_commandr > >h > >and see the results.e > >  > >       Yours, > >               Nigel Arnot * > >               NRA@MAXWELL.PH.KCL.AC.UK > > G > >               "In the beginning there was nothing, which exploded."i > >fL > >$!-----------------------------------------------------------------------/ > >$ if f$mode().EQS."NETWORK" then goto remotem5 > >$ if P1.EQS."" then inquire/nopunc p1 "_Node:    "g5 > >$ if P2.EQS."" then inquire/nopunc p2 "_Command: "s > >$ p1 = p1 - "::"  > >$ if p1.eqs."" then $exit  > >$ on controly then goto close% > >$ open/write net 'p1'::"task=tell"r8 > >$ write net P2+" ''P3' ''P4' ''P5' ''P6' ''P7' ''P8'" > >$ type net: > >$ close: close netu	 > >$ exits > >$remote:t7 > >$ if "''tell_first_use'".nes."" then goto setup_donesP > >$! there is a bug somewhere that prevents us getting local DEC tables through0 > >$! the user's UAF entry by proxy login. So.... > >$ set command/tables=sys_qec:dcl_tables.exe > >$ tell_first_use == "FALSE" > >$setup_done:. > >$ open/write net sys$net: > >$ read net command & > >$ write sys$output "TELL> ",COMMAND > >$ ass net.out sys$outputt
 > >$ set noonf > >$ 'command' > >$ stat = $statuss > >$ deass sys$outputr > >$ set ono3 > >$ if f$search("net.out").eqs."" then goto no_out  > >$ copy net.out net: > >$ delete net.out;*  > >$ goto out_done > >$no_out:-G > >$ write net "%TELL-I-NO_OUTPUT, command did not write to SYS$OUTPUT"m
 > >$out_done:R0 > >$ if .not.stat then write net f$message(stat) > >$ close net   ------------------------------  % Date: Sun, 04 Feb 2001 23:18:13 +0000m; From: Malcolm MacArthur <malcolmm@rustic-place.demon.co.uk>, Subject: Re: Lisbon Conference8 Message-ID: <3A7DE335.4D39E9EF@rustic-place.demon.co.uk>   Tim Llewellyn wrote: >  > Malcolm MacArthur wrote: >  > >iG > > Mind you though, Cheers used to be on television a lot when I was a  > > kid :=)j > >e > " > and you actually WATCHED it? :-) > H Ah well, we were poor in "them days", and we only had one television ;-)  N Can't believe I made that embarassing mistake about America! (Although I wouldN usually use "the Americas","North America" and "South America" to describe theC continents and "America" to denote the United States of America...)    -M.e   > >n
 > > -Malcolm.  >  > --8 > Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project2 > MedAS at the BBC, Whiteladies Road, Bristol, UK.C > Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukv > C > I speak for myself only and my views in no way represent those ofu > MedAS or the BBC.e   ------------------------------  % Date: Mon, 05 Feb 2001 01:45:20 +0100s1 From: Dennis Grevenstein <cdromlist@pcde.inka.de> 1 Subject: List of cdroms working with DEC hardwareh, Message-ID: <3A7DF7A0.AB99F0C6@pcde.inka.de>   Hi,s  < Some time ago I started a thread about cdroms that work with= DEC hardware. Allthough the results were not so impressiv ando? normally I would not publish it, I got mails from people askingo< me what cdrom they should get for their old DEC workstation.% So I guess it will be usefull anyway.u< If not, it will be just yet another piece of junk online ;-) But I really hope it's usefull.5    G ---------cut here------------------------------------------------------b    J Inofficial and incomplete list of third party cdroms that work with DEC's.J --------------------------------------------------------------------------  A This list is maintained by Dennis Grevenstein. email contact infoh is "cdromlist@pcde.inka.de"c9 You can find an up to date version of this document here:I, http://sites.inka.de/pcde/dec-cdrom-list.txt   I. Introductione  D The question "what cdrom will work with my old DEC?" is heard often.B The problem is that non-x86 workstations and especially DEC's will; normally need special cdroms that support special features.mB Many vendors released their own branded cdroms for their hardware.: But you will have to pay much more money to get a slow DEC> branded cdrom for your workstation than for a fast modern or a$ real cheap old standard cdrom drive.? So here is some sort of an answer for all who are looking for as( third party cdrom for a DEC workstation.A If the list gets longer(I hope so), I may release a html version.e     II. Some general information.   B As most Unix workstations, DEC's will need a cdrom that is capableD of reading 512 bytes/sector, but most PC style cdrom's can only read= 2048 bytes/sector. Often there is jumper to enable this mode.d@ If not, the cdrom must support the "MODE SELECT" command that is5 send by the PROM to enable the 512 bytes/sector mode.tD But often, even if a cdrom supports all of the above and can be usedE with other workstation-platforms e.g. Sun and SGI, the cdrom will not  work with DEC hardware.g    # III. drives that are known to work.-  H I decided to differentiate between VAX, MIPS based DECstations and AlphaG based workstations because of some irregularities. The Alpha section iscE pretty small right now and Alphas should be compatible with cdroms inn; the VAX or DECstation section, but a general rule could be:eC If a drive is not listed as working in your category, don't get it.e     III.1   VAXi  9 Vendor   model #      operating system    hardware testedRM -----------------------------------------------------------------------------a@ Apple    CD300e       VMS                 VAXstation, InfoserverM ----------------------------------------------------------------------------- H HP       C2944D       OpenVMS 6.2         uVAX 3100, VAXstation 4000/90,;                       OpenVMS 7.1         4000/60, 4000/VLCrM -----------------------------------------------------------------------------r6 Pioneer  DR-U124X     OpenVMS 7.2         uVAX 3100-80M ----------------------------------------------------------------------------- 6 Pioneer  DR-U12X      OpenVMS 7.2         uVAX 3100-30M ----------------------------------------------------------------------------->< Pioneer  DR-466       OpenVMS 7.1         VAXstation 3100-40M -----------------------------------------------------------------------------t6 Plextor  12/20        OpenVMS 7.2         uVAX 3100-80M -----------------------------------------------------------------------------mF Plextor  PX-32TSi     OpenVMS 7.1         uVAX3100, VAXstation 4000/60<                                           VAXstation 4000/90M -----------------------------------------------------------------------------o3 Sony     CDU561       Ultrix 4.5          uVAX 3100TM -----------------------------------------------------------------------------e9 Toshiba  XM5701TA     OpenVMS 7.1         VAXstation 3100 M -----------------------------------------------------------------------------n< Toshiba  XM-3401B     OpenVMS 7.1         VAXstation 3100-40M -----------------------------------------------------------------------------e     III.2   DECstations   G DECstations seem to be really tricky, especially if you want to installaC Ultrix. As far as i know Ultrix/MIPS can only be installed using antG original DEC cdrom. Typically your DECstation will boot of a compatiblenH cdrom, but the install programm will crash. If it is not compatible with+ a DECstation the boot will fail completely.oF I can't say much about NetBSD, but my DR-466 does not work with NetBSDD 1.4.1 at all, but works fine with NetBSD 1.5. Other people said thatD some older Toshiba cdroms work with VAXen, but not with DECstations.  9 Vendor   model #      operating system    hardware testedIM ----------------------------------------------------------------------------- = Plextor  PX-12CSi     NetBSD 1.5          DECstation 5000/133eM ------------------------------------------------------------------------------< Plextor  PX-32TSi     NetBSD              DECstation 5000/33M -----------------------------------------------------------------------------M= Pioneer  DR-466       NetBSD 1.5          DECstation 5000/150eM -----------------------------------------------------------------------------r      
 III.3   Alphat  9 Vendor   model #      operating system    hardware testedhM -----------------------------------------------------------------------------t0 Plextor  PX-40TSi     VMS                 MultiaM ----------------------------------------------------------------------------- 1 Apple    CD300e       VMS                 DEC3000lM -----------------------------------------------------------------------------o     IV. Other resourcesn  F The Sun cdrom FAQ may be useful. It provides some general information,E allthough this FAQ does not care, if a cdrom works with DEC hardware.    http://saturn.tlug.org/suncdfaqd    2 V. thanks, suggestions and all the other things...  9 Many thanks to all who contributed to build up this list.iE This list is by no means complete. So if you have ever used a non-DECiE cdrom that is not shown here with a DEC workstation, please send somehF information to "cdromlist@pcde.inka.de". If you have other suggestions) or complains, feel free to mail them too.    ------------------------------  % Date: Sun, 04 Feb 2001 14:18:03 -0500n+ From: Clayton Kroh <ckroh@MAIL.CLARION.EDU> & Subject: Looking for a good SSH client+ Message-ID: <3A7DAAEB.DA5E0EF3@clarion.edu>d  E We are using Multinet 4.3 SSH. Can anyone recommend a good SSH client H running under Windows that does decent VT220 function key emulation or a? Windows client that allows you to define the function key codes 	 yourself?    Thanks   -- Clayton Kroh Manager, Enterprise Services Clarion University of PA ckroh@clarion.edu- (814) 393-1673   ------------------------------  # Date: Sun, 04 Feb 2001 14:16:03 CSTe/ From: Charles Sandmann <sandmann@clio.rice.edu> * Subject: Re: Looking for a good SSH client- Message-ID: <3a7db883.sandmann@clio.rice.edu>   G > We are using Multinet 4.3 SSH. Can anyone recommend a good SSH clientlJ > running under Windows that does decent VT220 function key emulation or aA > Windows client that allows you to define the function key codesa > yourself?e  2 http://www.chiark.greenend.org.uk/~sgtatham/putty/  N Free.  Source available.  Very compatible, small footprint, easy on resources.6 It's better than most commercial software you can buy.   ------------------------------  % Date: Sun, 04 Feb 2001 17:12:26 -0700s% From: Dan O'Reilly <dano@process.com> * Subject: Re: Looking for a good SSH clientB Message-ID: <5.0.2.1.2.20010204171145.00a91a18@pop.clsp.qwest.net>  F There are 2 good ones: SecureCRT (which costs $$$) and TTSSH (which is free).  ) At 12:18 PM 2/4/2001, Clayton Kroh wrote: F >We are using Multinet 4.3 SSH. Can anyone recommend a good SSH clientI >running under Windows that does decent VT220 function key emulation or aE@ >Windows client that allows you to define the function key codes
 >yourself? >m >Thanksd >n >--o
 >Clayton Kroh1 >Manager, Enterprise ServicesM >Clarion University of PA1 >ckroh@clarion.edu >(814) 393-1673d   ------I +-------------------------------+---------------------------------------+rI | Dan O'Reilly                  |                                       | I | Principal Engineer            |  "Those are my principles. If you     |-I | Process Software              |   don't like them I have others."     |eI | http://www.process.com        |                    -- Groucho Marx    |eI +-------------------------------+---------------------------------------+E   ------------------------------  # Date: Mon, 05 Feb 2001 01:51:46 GMTE& From: rmoore@qualcomm.com (Ryan Moore)* Subject: Re: Looking for a good SSH client9 Message-ID: <SGnf6.56863$GV2.13156345@typhoon.san.rr.com>   J In article <3A7DAAEB.DA5E0EF3@clarion.edu>, ckroh@MAIL.CLARION.EDU says...F >We are using Multinet 4.3 SSH. Can anyone recommend a good SSH clientI >running under Windows that does decent VT220 function key emulation or a @ >Windows client that allows you to define the function key codes
 >yourself?  I I use SecureCRT from Van Dyke Technologies (www.vandyke.com).  SecureCRT uM supports vt100 and vt220 emulation complete with VT graphics characters.  It uM will do ANSI color as well.  Supports telnet, rlogin, serial, SSH1, and SSH2 tK protocols.  There is a VT220 keymap you can enable which will give your PC aP keyboard the feel of a VT200 keyborad (editing keys, keypad, etc.).  Or you can I map keys on your own.  It also supports various flavors of scripting (VB g script, PerlScript, etc.).  M It's commercial software, but the price is reasonable.  I use it everyday... r it's well worth the price.   -Ryant   ------------------------------  % Date: Sun, 04 Feb 2001 17:06:35 -0500n3 From: "Edward J. Groth" <groth@pupgg.princeton.edu>d( Subject: Re: MMS/MMK Ghostscript problem3 Message-ID: <3A7D8C1B.661D3381@pupgg.princeton.edu>c   JF Mezei wrote:e > - > Oopps, sorry for type in message header....n >  > Here is the messahe again: > Z >  now that I have decwindows, I decided to give ghostscript a second attempt at building. > M >  Downloaded the 6.0 sources (Merci Patrick Moreau). Unpacked the stuff, ands@ >  edited the build_openvms_decc.com to call MMS instead of MMK. > ! >  Ran it, and got the following:nP >  MMS/Log/Macro=("VAXC=0", "DECWINDOWS1_2=1", "A4_PAPER=0")'macro' [.bin]GS.exe5 >  %MMS-I-DRVOUTFIL, Using output file SYS$OUTPUT:.;.dH >  %MMS-I-DRVNOFMSSUP, DEC/MMS is installed without support for VAX FMS.6 >  %MMS-I-DRVSUBCLI, Using DCL for the subprocess CLI.R >  %MMS-I-DRVQUALIF, Using non-defaulted qualifiers /Log/Macro=("DECWINDOWS1_2=1", >  "A4_PAPER=1"). ? >  %MMS-I-GWKBEGWLK, Starting the build at target [.BIN]GS.EXE.-E >  %MMS-I-GMTIMFND, Time for [.BIN]GS.EXE is 17-NOV-1858 00:00:00.00. S >  %MMS-F-GWKNOPRN, There are no known sources for the current target [.BIN]GS.EXE..> >  %MMS-I-GWKEXESTS, Status of executed command is %X10EE8064.I >  %MMS-F-ABORT, For target MACRO, CLI returned abort status: %X10EE8064.  >  >  This is on VAX  VMS 7.2.- > N >  Any hints on what went wrong or where I should look ? I was hoping to buildA >  this over the weekend (probably will take a fe days to build).-  E Well, this is a really complicated build and I think there are enough C incompatibilities between MMK and MMS, you're not going to have any- success with that route.  7 I just built GS 6.0 on VAX VMS6.2 using the latest MMK..  D My advice is get MMK. It's not very big. However, when I built it byA linking the supplied objects, and then tried to build GS, I got a  message very similar0 to the one you have above (not finding sources).  E So I used MMK to build MMK from sources. That worked and then the new- version also worked on GS.  
 						- Ed     --  J /------------------------------------------------------------------------\H | Edward J. Groth, Assoc. Chr. | Phone: 609-258-4361   Fax: 609-258-6853 | H | Physics Dept., Jadwin Hall   | URL: http://pupgg.princeton.edu/~groth/ |-H | Princeton University         | SPAN/HEPNET:  PUPGG::GROTH=44117::GROTH |0H | Princeton, NJ 08544          | Internet:     groth@pupgg.princeton.edu |-J \------------------------------------------------------------------------/   ------------------------------  % Date: Sun, 04 Feb 2001 20:56:44 -0500o- From: JF Mezei <jfmezei.spamnot@videotron.ca> ( Subject: Re: MMS/MMK Ghostscript problem, Message-ID: <3A7E085B.9F90ACCB@videotron.ca>   "Edward J. Groth" wrote:G > Well, this is a really complicated build and I think there are enoughhE > incompatibilities between MMK and MMS, you're not going to have anyu > success with that route.    L Thanks, that is what I had suspected, but wasn't sure. The doc says that the two are compatible:-)   L I learning MMK/MMS considered a valuable asset for someone's CV ? My initialJ reaction was to pick all the .MAK files apart and try to find out what wasI going wrong, but then I realised I might be wasting my time. Learning VMS-- specific stuff is a waste of time these days.x  G > So I used MMK to build MMK from sources. That worked and then the new- > version also worked on GS.  M Thanks, will get to work on that. Hopefully most if it can compile overnight.i   ------------------------------  % Date: Sun, 04 Feb 2001 23:05:55 +0000-) From: Christof Brass <brass@infopuls.com>eD Subject: Re: NT64-on-Alpha: Microsoft Business Partner = Organ Donor, Message-ID: <3A7DE053.3AA5BC8C@infopuls.com>   "Terry C. Shannon" wrote:  > 8 > "Christof Brass" <brass@infopuls.com> wrote in message( > news:3A7C1D15.6C83EFAB@infopuls.com... > > Terry C Shannon wrote: > > > / > > > On Wed, 31 Jan 2001, Dean Woodward wrote:- > > >-L > > > > Which is why I keep hoping against hope that Win64 will be supported > byG > > > > Compaq on Alpha, since that's where all the development happensl	 > anyway.s > > > >  > > >e8 > > > It may not be supported, but Build 2128 exists ;-} > >gB > > *This* sounds really weired. It has been stated that the 64BitC > > clean development is done on the Alpha platform, that IA-64 hassA > > miserable design flaws and that Alpha can easily compete witho? > > current IA-32 systems the only platform left for enterpriseh? > > headed Micro$oft systems. Why doesn't Micro$oft support thee > > Alpha processor??e > K > Easy. Compaq was "asked to contribute" to pay for the port and support by K > its Strategic Partner Micro$oft, even though the Strategic Partner didn'tLJ > give a tinker's damn whether CPQ ever sold a single Alpha system runningM > NT64. Capellas & Co. realized that Microsoft Business Partner = Organ DonornC > and concluded that it wasn't worth spending good money after bad.i > ( >  All is revealed at the following URL. > * > http://www.acersoft.com/SKD/99/V6N25.htm  6 Great analysis in this article (got a new subscriber)!  > Seems that Compaq lost trust into their own capabilities. They< gave up because of the low percentage of AlphaNT systems and= because of the expected close delivery of IA-64 systems. It'sf> like when Micro$oft is announcing it will produce this or that@ even if they didn't start any analysis let alone any development? a lot of companies loose their heads and give up their productsn@ long before Micro$oft is able to deliver the first alpha version: of buggy code. If Compaq has continued they were in a much: better position now because of the delay and the miserable? performance of the IA-64 systems. It is an easy analysis of thet> situation: what Micro$oft addicted people want is the look and? feel, they want to continue to use what they know. The hardware @ is not interesting besides its performance and app availability.@ Migration to IA-64 might introduce the problem that apps are not? available or uncredible slow (may be even not observable faster)> than with FX!32 converted apps on Alphas). If the apps have to@ re-compiled for IA-64 to be of any value than this could be done> for Alpha as well. Buying new HW is necessary in both cases as= long as you don't want to trust these IA-64 emulators. And aso= far as I know AMD is also working on a 64Bit processor design-3 although I don't know anything about compatibility.C  8 Compaq should have known that introducing a new hardware@ platform for any OS will take more than a few years. But I agree@ it's not a good strategy to help Micro$oft for keeping the Alpha> platform if Micro$oft doesn't pay for it. We should not forget< that Micro$oft will get money for each sold OS copy. And the> strategy of Micro$oft for attacking the enterprise business is= not clear to me either. Unless they think that they will gainx> the same power with the IA-32 and later IA-64 systems dropping; Alpha which is the only remaining performance competitor to < Intel seems not a very smart step. Or is the article by Paul@ DeMone about Alpha's EV8 basically wrong? Are there no people at@ Micro$oft who understand processor design? Or do they think they" can change performance by talking?   ------------------------------  $ Date: Mon, 5 Feb 2001 07:36:42 +0100- From: "Ruzsinszky Attila" <aruzsi@mailbox.hu>a Subject: Re: NTP on VMS 7.1k, Message-ID: <95lhko$tmm$1@athena.euroweb.hu>  ; > Try @SYS$MANAGER:UCX$CONFIG; @SYS$STARTUP:TCPIP$CONFIG isu1 > for V5.0 and higher of Digital TCP/IP Services. L OK. It is working now. I found NTP menu point, but I can't setup. Only thing I can.2 do enable NTP. When I try to enable in UCX, I got: UCX> enable service ntpn- %UCX-E-STARTERROR, Error starting NTP serviceo& -UCX-W-NORECORD, Information not found -RMS-E-RNF, record not found  J I'd like to mention I was able to setup SNMP in the same way. SNMP OK, NTP not.   Ruzsia   ------------------------------  $ Date: Mon, 5 Feb 2001 07:47:57 +0100- From: "Ruzsinszky Attila" <aruzsi@mailbox.hu>d Subject: Re: NTP on VMS 7.1o, Message-ID: <95li9v$tne$1@athena.euroweb.hu>  I >   I'd encourage a visit to the documentation website for details of the H >   current version, or a check of any local documentation for the older >   version in use...eJ I found the current version on the WEB and I went to the forest deeper and" deeper. See my UCX$CONFIG problem.  F I don't find any printed documentation, except some words about TCP/IP (telnet, ftp) not NTP, SNMP, DNS etc.  I The DEC documentation maybe on the CD which we got with Alphastation 233.iA The problem: I have never tried to use that and I don't know how.r  K In that cause you can ask: "Why want to setup NTP, if you are so beginner?"eL Because the firm can't solve our problem since last April. (and I think that. solution which was implemented, never worked!)? And where I am the time would be very important. I hope the NetVH community try to help me. :-) The last cause: when the NTP works I learn	 something2 about OpenVMS.   So thanks anyone the help.   Ruzsi    ------------------------------  # Date: Sun, 04 Feb 2001 17:44:38 GMT43 From: "MrSignor" <MrSignor@nospam_bellatlantic.net>s Subject: OpenVMS / UNIX + COEe8 Message-ID: <aygf6.5002$Nx3.1211232@typhoon2.ba-dsg.net>  
 Greetings,H     From what I can tell, IBM's OS/390 supports most if not all standard UnixK     interfaces, and has done so since 1994 - since that time they have beentI     improving it. So that today, OS/390 supports most of the Open Systemsr:     Foundation's Single UNIX Specification ... ie. UNIX98.  J     What I particularly like about the UNIX98, is that it includes POSIX.4D     (supported on OS/390). POSIX.4 is the Asynchronous I/O Interface=     (aio_xxx), which maybe used in conjunction with Pthreads.n  G    I am particularly interested in using POSIX.4 + Pthreads on OpenVMS,p  3    Q) Will COE support provide this functionality ? :    Q) What POSIX standards (if any) is COE a superset of ?>    Q) Exactly where is it described what UNIX system calls are0         required to be supported as part of COE.  C    One last comment, regardless of COE, if OS/390 can support UNIX,r     why can't OpenVMS ?p   Thanks inadvance     -Fred0   ------------------------------  % Date: Sun, 04 Feb 2001 20:13:14 -0600p+ From: "Main, Kerry" <Kerry.Main@compaq.com>t! Subject: RE: OpenVMS / UNIX + COEuN Message-ID: <910612C07BCAD1119AF40000F86AF0D805284D60@kaoexc3.kao.cpqcorp.net>   Fred,m   re: COE and UNIX  C <http://www.openvms.compaq.com/solutions/government/coe/index.html>i   Regards,  
 Kerry Main Senior Consultantt Compaq Canada Inc. Professional Servicesr Voice: 613-592-4660a Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----8 From: MrSignor [mailto:MrSignor@nospam_bellatlantic.net] Sent: February 4, 2001 12:45 PMu To: Info-VAX@Mvb.Saic.Coma Subject: OpenVMS / UNIX + COEU    
 Greetings,H     From what I can tell, IBM's OS/390 supports most if not all standard UnixK     interfaces, and has done so since 1994 - since that time they have beendI     improving it. So that today, OS/390 supports most of the Open Systemso:     Foundation's Single UNIX Specification ... ie. UNIX98.  J     What I particularly like about the UNIX98, is that it includes POSIX.4D     (supported on OS/390). POSIX.4 is the Asynchronous I/O Interface=     (aio_xxx), which maybe used in conjunction with Pthreads.e  G    I am particularly interested in using POSIX.4 + Pthreads on OpenVMS,a  3    Q) Will COE support provide this functionality ? :    Q) What POSIX standards (if any) is COE a superset of ?>    Q) Exactly where is it described what UNIX system calls are0         required to be supported as part of COE.  C    One last comment, regardless of COE, if OS/390 can support UNIX,      why can't OpenVMS ?u   Thanks inadvance     -Fred    ------------------------------  # Date: Sun, 04 Feb 2001 23:29:01 GMT 0 From: "CHARLES COOPER" <scooper8@houston.rr.com> Subject: OPENVMS 7.3....When? 9 Message-ID: <1Blf6.9877$tS.2498195@typhoon.austin.rr.com>p  I Compaq starting advertising all the great features of OPENVMS 7.3 back in-I September 2000....Does anyone know what the status or release date for it1C is?....perhaps it has been released and I missed my distribution...i thanks,p   Charles Cooper UT School of Public Health   ------------------------------  % Date: Sun, 04 Feb 2001 18:55:30 -0600t7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> ! Subject: Re: OPENVMS 7.3....When? - Message-ID: <3A7DFA02.356E9B66@earthlink.net>e   CHARLES COOPER wrote:i > K > Compaq starting advertising all the great features of OPENVMS 7.3 back innK > September 2000....Does anyone know what the status or release date for itoE > is?....perhaps it has been released and I missed my distribution...i	 > thanks,"  D You could try searching Deja for news, but the least I saw, it stillF slated as simply "Q1, CY2001". I seem to remember some mention that it@ might be worth waiting for V7.3-1, probably summertime (NorthernG Hemisphere) of 2001, depending on what you need and what you're waitingo for.   --   David J. Dachtera: dba DJE Systems. http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/$  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------  % Date: Sun, 04 Feb 2001 19:21:45 -0600 + From: "Main, Kerry" <Kerry.Main@compaq.com>y! Subject: RE: OPENVMS 7.3....When?eN Message-ID: <910612C07BCAD1119AF40000F86AF0D806E03930@kaoexc3.kao.cpqcorp.net>   Charles,   re: VMS V7.3 ..o  F >>> Does anyone know what the status or release date for it is?....<<<  L <http://www.openvms.compaq.com/ebusiness_without_compromise/fact/OPENVMS-V73 .HTML>L "OpenVMS Version 7.3 will be available Q1 2001 directly from Compaq Computer) Corporation and its Business Partners."     J >>> Compaq starting advertising all the great features of OPENVMS 7.3 back in September 2000..<<<  G Just in case anyone thinks this is to early to advertise, check out theb  following ad for Cairo/NT5/W2K -1 http://www8.zdnet.com/eweek/news/0707/11ent5.htmle   :-)o   Regards,  
 Kerry Main Senior Consultante Compaq Canada Inc. Professional Services  Voice: 613-592-46605 Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----5 From: CHARLES COOPER [mailto:scooper8@houston.rr.com]B Sent: February 4, 2001 6:29 PM To: Info-VAX@Mvb.Saic.Comh Subject: OPENVMS 7.3....When?h    I Compaq starting advertising all the great features of OPENVMS 7.3 back inuI September 2000....Does anyone know what the status or release date for itlC is?....perhaps it has been released and I missed my distribution...n thanks,f   Charles Cooper UT School of Public Health   ------------------------------   Date: 4 Feb 2001 13:34:42 -0500-9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) 6 Subject: Re: Plenum vs. non-plenum in the data center?3 Message-ID: <uAap7AtiHKzc@eisner.encompasserve.org>-  O In article <3A7D8ECD.9030907@wi.rr.com>, Scott Vieth <svieth@wi.rr.com> writes:n   > He said that the electricaldI > inspector (or fire inspector) from the city would not allow non-plenum   > cables under the floor inp > the data center.  3 Rule #1 - The local code inspector is always right.   < Rule #2 - If the local code inspector is wrong, see rule #1.   ------------------------------  % Date: Sun, 04 Feb 2001 16:27:54 -0500a- From: JF Mezei <jfmezei.spamnot@videotron.ca>a6 Subject: Re: Plenum vs. non-plenum in the data center?, Message-ID: <3A7DC958.5023FA3F@videotron.ca>   Larry Kilgallen wrote:J > > inspector (or fire inspector) from the city would not allow non-plenum. > > cables under the floor in the data center. > 5 > Rule #1 - The local code inspector is always right.t  G Do you have any fire detection equipment under the floor ? Does the air F conditioning system use the underfloor to distribute the air or is the underfloor just dead space ?    I How about getting Compaq to certify that those cables are safe for plenum0. operations ? Would that please the inspector ?  L Or ask the inspector what it would take for him to approve non plenum cables$ under a raised computer room floor ?  M I assume that this is in a smaller town where raised computer room floors are  not common ?  I This is similar to large data centres with fancy fire suppression systemst; still forced to have water sprinklers over theur equipment.r   ------------------------------  % Date: Sun, 04 Feb 2001 19:21:45 -0600p+ From: "Main, Kerry" <Kerry.Main@compaq.com> 6 Subject: RE: Plenum vs. non-plenum in the data center?N Message-ID: <910612C07BCAD1119AF40000F86AF0D806E03931@kaoexc3.kao.cpqcorp.net>   JF,i  E >>> This is similar to large data centres with fancy fire suppressioneF systems still forced to have water sprinklers over their equipment.<<<  L Actually, this is an insurance thing that I believe has a good reason behind it.I  J What do you do if the fancy fire suppression system discharges its tank of& chemicals and the fire is still going?  J Answer - good old water system kicks in and continues to spray water until the fire really is out.r  C Course, by then the computer room equipments is likely destroyed byuF chemicals and water damage, but at least the building is theoretically
 safe(r) ..   Regards,  
 Kerry Main Senior ConsultantS Compaq Canada Inc. Professional Services  Voice: 613-592-4660k Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----4 From: JF Mezei [mailto:jfmezei.spamnot@videotron.ca] Sent: February 4, 2001 4:28 PM To: Info-VAX@Mvb.Saic.Com 6 Subject: Re: Plenum vs. non-plenum in the data center?     Larry Kilgallen wrote:J > > inspector (or fire inspector) from the city would not allow non-plenum. > > cables under the floor in the data center. > 5 > Rule #1 - The local code inspector is always right.e  G Do you have any fire detection equipment under the floor ? Does the airIF conditioning system use the underfloor to distribute the air or is the underfloor just dead space ?    I How about getting Compaq to certify that those cables are safe for plenum-. operations ? Would that please the inspector ?  L Or ask the inspector what it would take for him to approve non plenum cables$ under a raised computer room floor ?  I I assume that this is in a smaller town where raised computer room floors  are  not common ?  I This is similar to large data centres with fancy fire suppression systems-; still forced to have water sprinklers over theur equipment.    ------------------------------  % Date: Sun, 04 Feb 2001 23:40:20 +0000r) From: Christof Brass <brass@infopuls.com> % Subject: Re: Source listings contentsl, Message-ID: <3A7DE864.F8612AF5@infopuls.com>   "Glenn C. Everhart" wrote: > B > There are also things excluded because they are considered tradeB > secrets (e.g. the GEM code generator), some things because their@ > authors don't want them used as examples (considering them tooD > ugly for public exposure), some things where patent apps are beingB > made, and some where they just don't want to have some info fall? > into competitors' hands (not much of this, but I believe somen@ > bits of cluster code are omitted for this reason). I believe aB > few bits are also excluded because they are considered VERY VERY@ > likely to be redone and they don't want anyone using the calls@ > mentioned, lest their code suddenly and irretrievably break in > a future minor release.  > B > There are probably a few more categories, but I know of examplesA > of all the above. (No, I do not care to be more specific; don'ta% > know as it would be appropriate...)N >  > Glenn Everhart >  > Wayne Sewell wrote:D > > k > > In article <jazs4jZs$EpU@eisner.decus.org>, Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) writes:o^ > > > In article <3A735C46.BB21A735@infopuls.com>, Christof Brass <brass@infopuls.com> writes: > > >  > > > ? > > >> From the FAQ VMS7: "OpenVMS VAX and OpenVMS Alpha source ' > > >> listings CD-ROM sets include thenC > > >> source listings of most of OpenVMS, ...". I also read in covr( > > >> that these sets are not complete. > > >>= > > >> *Why?* Is anybody out there who possesses the sources?s > > > G > > > They exclude material covering pending patents, material covering 0 > > > unannounced features and the LMF listings. > > F > > Ah, yes.  LMF, the license *management* tool that is not a license > > *enforcement* tool.  > >m > > --S > > =============================================================================== Q > > Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxt< > > http://www.tachysoft.xxx/www/tachyon.html and wayne.htmlO > > change .xxx to .com in addresses above, assuming you are not a spambot  :-)rS > > ===============================================================================cF > > Cute Girl, to Curly: "Oh, what a beautiful head of bone you have!"  > Fine with me. Reverse engineering the cluster code as probably? been done by SUN and IBM. This can be done without any sources. @ This also holds for the patented parts (and of course all others@ where it doesn't matter). I think that including the robustness,= cluster technique and other VMS specific things into Solaris,f9 Linux, OS/400 ... isn't close to impossible because thesew@ features have to be designed in. I know that clusters came later@ but I'm not sure if they hadn't something like this in mind when? they designed VMS first or if they rewrote major parts and withf> help of their clean design it was possible. There are features? (e.g. multi-threading) which can't be implemented afterwards ift0 you have forgotten them. Better start over then.  5 The other reasons call for publishing the sources :-)    ------------------------------  % Date: Sun, 04 Feb 2001 14:51:48 -0600 + From: "Main, Kerry" <Kerry.Main@compaq.com>T% Subject: RE: Source listings contentssN Message-ID: <910612C07BCAD1119AF40000F86AF0D805284D5A@kaoexc3.kao.cpqcorp.net>  
 Christof -  ? >> Thanks. Does anybode know the price or where it is listed?<<e  4 <http://www.openvms.compaq.com/wizard/wiz_1725.html>  5 QB-MT1AB-E8      OpenVMS Alpha Source Listings CD-ROMe= QT-MT1AB-Q8      OpenVMS Alpha Source Listings CD-ROM Updatese3 QB-001AB-E8      OpenVMS VAX Source Listings CD-ROMr; QT-001AB-Q8      OpenVMS VAX Source Listings CD-ROM Updates    Regards,  
 Kerry Main Senior Consultanth Compaq Canada Inc. Professional ServicesC Voice: 613-592-4660e Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----0 From: Christof Brass [mailto:brass@infopuls.com] Sent: February 4, 2001 10:28 AMp To: Info-VAX@Mvb.Saic.Como% Subject: Re: Source listings contents      Wayne Sewell wrote:o > - > In article <jazs4jZs$EpU@eisner.decus.org>,r; Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) writes:m? > > In article <3A735C46.BB21A735@infopuls.com>, Christof Brassw <brass@infopuls.com> writes: > >m > >p= > >> From the FAQ VMS7: "OpenVMS VAX and OpenVMS Alpha sourcet% > >> listings CD-ROM sets include theOA > >> source listings of most of OpenVMS, ...". I also read in cov & > >> that these sets are not complete. > >>; > >> *Why?* Is anybody out there who possesses the sources?m > >bE > > They exclude material covering pending patents, material covering7. > > unannounced features and the LMF listings. > D > Ah, yes.  LMF, the license *management* tool that is not a license > *enforcement* tool.  >  > -- >UL ============================================================================ ===e: > Wayne Sewell, Tachyon Software Consulting  (281)812-0738 wayne@tachysoft.xxxt: > http://www.tachysoft.xxx/www/tachyon.html and wayne.htmlH > change .xxx to .com in addresses above, assuming you are not a spambot :-)  >XL ============================================================================ ===tD > Cute Girl, to Curly: "Oh, what a beautiful head of bone you have!"  : Thanks. Does anybode know the price or where it is listed?   ------------------------------  $ Date: Mon, 5 Feb 2001 09:07:29 +13006 From: "antony wardle" <antony.wardle@nospam.met.co.nz> Subject: Re: sysuaf question1 Message-ID: <Oxif6.1503$sS4.40249@ozemail.com.au>o  + that might be what I have to do in the end.l  ( There is a decus programme called sysuaf1 and that has a bit in it where it can extract theB, encrypted password, then insert it back into a new sysuaf  0 I have the vax version, but not the axp version.   cheers   antony      0 "Veli Krkk" <korkko@decus.fi> wrote in message" news:3A7D5CBD.3E729AE0@decus.fi...< > You could always extract the relevant records using simple > DCL, like: >e1 > $ if f$trnlnm("SYSUAF",,,"LNM$SYSTEM").eqs."" -a4 >           then define sysuaf sys$system:sysuaf.dat( > $ open/read/write/share=write s sysuaf! > $ read/key="USERNAME " s usrrech > $ close sw > $ open/write tmp usr.txt > $ write/symbol tmp linee
 > $ close tmpv > : > Now move usr.txt to the other system and insert the user > record > like >e1 > $ if f$trnlnm("SYSUAF",,,"LNM$SYSTEM").eqs."" -o4 >           then define sysuaf sys$system:sysuaf.dat( > $ open/read/write/share=write s sysuaf > $ open/read tmp usr.txto > $ read tmp line_
 > $ close tmpt > $ write/symbol s lineI > $ close ss >-> > Or make a copy of the SYSUAF, remove non-needed records from > that copy,8 > then move the cleaned copy to the other system and use > CONVERT/MERGEf6 > to merge contents to the systems current SysUAF.dat. >e > _veli  > antony wardle wrote: > >,0 > > anyone got an axp version of the sysuaf exe? > >i: > > I am particularly interested in the bit that transfers > >i+ > > the encrypted password to another node.l > >PC > > Don't want to copy the sysuaf file, and I don't want to try andt > >e3 > > figure out /recreate everyones password either.  > >a
 > > cheers > >a
 > > antony >r   ------------------------------  $ Date: Sun, 4 Feb 2001 22:16:06 -0500- From: "Jerry Alan Braga" <jabraga@golden.net>s. Subject: UCX SMTP duplicate mail sending HELP!5 Message-ID: <N%of6.250$xc3.18960527@radon.golden.net>e  H Is there any way with UCX 4.2 or other TCP$IP versions to have SMTP mailD filter out duplicates before sending just as standard VMS mail does.  G I would like to leverage our VMS system to become a true smtp/pop emaile. server for our company but here is my problem.  3 1) VMS char mode mail can't handle mime attachments/F     - figured that one out by using MAILBOX software suggested on this. newgroups and using TOP_HEADERS in smtp config  E 2) SMTP does not filter out duplicate messages that are not containedi exactly in to list9     - ie: using server side distribution groups stored inl sys$specific:[ucx_smtp]2D         {copy of vms mail dis groups usng RFC names user@domain.com}K     - problem is that if DIS1.DIS constains references to DIS2.DIS and user]2 'A' happens to be in both then with SMTP mail userI         'A' will get the email twice if sending email to dis1@domain.com. D This is costly as users have to filter the correct vs incorrect mail  I 3) having to maintain a client distribution list for outlook express typem. clients is nothing but a pain to keep in sync.  K If there was some way to intercept smtp mail before it was sent and have it.L go through a user collection process and filter out duplicates?  This is howJ VMS mail does it now and I would have thought VMS would have put somethingH into UCX TCP$IP software to emulate that through a logical or something.  D I have heard of the product on this group called DELIVER but have no information on it at all.    Thanks in advanceQ   ------------------------------  % Date: Sun, 04 Feb 2001 19:41:57 -0800 ! From: Koloth <koloth@tmisnet.com>l4 Subject: Re: Unrecognized Individual Destination PDU+ Message-ID: <3A7E2105.B8A2D7A3@tmisnet.com>   K These messages occurs during boot up when DECnet has started but TCP/IP has P not.  Remember DECnet needs to be started first.  The Data Link Header shows theP destination address AA-00-04-00-02-84  or %x8402 SCSSYSTEMID or 33,794 or DECnet 33.2  J The sender is AA-00-04-00-3C-84 or %x 843C SCSSYSTEMID or 33,852 or DECnet- 33.60.  The Ethernet Protocol was 08-00 whichk
 is TCP/IP.  J You will get these messages between the time that DECnet starts up and theO TCP/IP stack starts up.  Your startup is hanging during startup.  It looks like 1 it is hanging before the TCP/IP stack is started.-   Regards,   Cass Witkowski SAIC   "Gnter Gratzer" wrote:m   > Hi!JF > After rebooting our Alpha server we got the following message during
 > startup: >-: > %%%%%%%%%%%  OPCOM   3-FEB-2001 14:27:07.32  %%%%%%%%%%%$ > Message from user SYSTEM on DPMS01I > Event: Unrecognized Individual Destination PDU from: Node LOCAL:.DPMS01a > CSMA-CD Station CSMACD-0, / >         at: 2001-02-03-14:27:07.322+01:00Iinfn: >         Data Link Header='AA0004000284AA0004003C840800'H9 >         eventUid   A1359B79-F9E0-11D4-A64C-44504D533031t9 >         entityUid  D185CB35-F9DB-11D4-8115-AA0004000284o9 >         streamUid  D47D2478-F9DB-11D4-8160-AA0004000284a > L > The message was comming continously. Can anybody explain me the meaning of > this message?  >.I > The server was hanging during startup. I don't know exactly where but InM > think it was after starting the network and starting the Queues. The server/N > is running with another alpha server in a cluster with a common system disk./ > The second server was runnning during reboot. J > We could only solve this problem with shuting down the second server andM > reboot the server with the pending startup. After the second reboot all was- > ok again.-L > There was just a strange thing with the batch queues on the second server.M > After reboot the daily backup job was in state executing with a submit time1A > 14.01.2001! At this day we had the same problem after a reboot.e0 > Is there a known problem in the queue manager? >s > We are running VMS V7.2-1. >< > Thanks >w > Gnter   ------------------------------  % Date: Sun, 04 Feb 2001 23:30:17 +0000V) From: Christof Brass <brass@infopuls.com>l0 Subject: Re: Writes flushed to Disk on Dismount?, Message-ID: <3A7DE609.28E050E8@infopuls.com>  ! "Koska, John C. (LNG-MBC)" wrote:M >  > > -----Original Message-----7 > > From: Paul Repacholi [mailto:prep@prep.synonet.com]e- > > Sent: Saturday, February 03, 2001 8:09 AMK > > To: Info-VAX@mvb.saic.com/4 > > Subject: Re: Writes flushed to Disk on Dismount? > >e@ > > "Koska, John C. (LNG-MBC)" <John.C.Koska@bender.com> writes: > >3A > > > I have been trying to follow this thread, since it is great  > > interest to me.-# > > > Please excuse as I jump in...o > > >o= > > > If an Operations or electrician guy drops power to yourb > > HGS80 and diskA > > > arrays and you have cache batteries, are you not ok as longa > > as the cache@ > > > batteries have power?  Which is quite a few hours (over 20 > > hours as I recalli? > > > from my site's requirement of having the systems down fori > > over 24 hours), to > > > keep the cache intact. > >lI > > So, how much are those cute lil suckers worth now. You will take REALr > > good care of them I hope.n > < > They may not be worth as much as when we first got'em, but< > that tends to be the nature of computer hardware overtime. > ? > I believe we spent a little over $100K US dollars for a dozenhE > of them and another $13K US dollars for the 256 meg cache upgrades.  > = > They are all on a service contract for 2 hour response overs= > 7 by 24 by 365.  So we get fairly good care over time, even       ^^^^^^^^^^^^^^  ??????????????  B > if they are not the same HSG80s we started out with; not that we@ > have had a lot of failures outside of our little power fiasco. >  > :) jck   ------------------------------  % Date: Sun, 04 Feb 2001 15:44:19 -0500h: From: "Koska, John C. (LNG-MBC)" <John.C.Koska@bender.com>0 Subject: RE: Writes flushed to Disk on Dismount?K Message-ID: <3D35AD137AAAD411A6BA0008C7B1B12D6DBDA6@MBCALBEXC03.BENDER.COM>o   > -----Original Message-----5 > From: Paul Repacholi [mailto:prep@prep.synonet.com]o+ > Sent: Saturday, February 03, 2001 9:36 PMt > To: Info-VAX@mvb.saic.comd2 > Subject: Re: Writes flushed to Disk on Dismount? >  > > > "Koska, John C. (LNG-MBC)" <John.C.Koska@bender.com> writes: > = > > > So, how much are those cute lil suckers worth now. You W > will take REAL > > > good care of them I hope.e > >  a> > > They may not be worth as much as when we first got'em, but> > > that tends to be the nature of computer hardware overtime. > > B > > I believe we spent a little over $100K US dollars for a dozen G > > of them and another $13K US dollars for the 256 meg cache upgrades.c > H > No, no, not the HSG, the batteries? $10? $20? ( other than from a well > know vendor that is. ) > > > So that's how many dollars for a failure per dollar or NiCd?  G A lot.  But I believe there are dual redundant batteries, which can be 1E swapped out hot, such that one could keep the cache "live" in theory  B for a very long time.  At least that is what my testing showed.  IA could drop one cache battery at a time, and replace with another e charged one with no problems.i  r? > I just can NOt believe, after all the 'we go to all this.... d > quallification'cE > stuff ( yes guys who do it, I know you do ) thats why it is SO muche > better and cost more.r > H > Then they dump a load of crap like this. BTW, and one for a sweepstake# > on a new version of HSG firmware?h  I Well, I guess the odds are in my flavor so far (and Compaq's by design)  eJ with this, then with other vendors (EMC, etc.), since I have lost data on  with the others.   > --  > > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda.-B >                                              West Australia 60760 > Raw, Cooked or Well-done, it's all half baked. >    :) jck   ------------------------------   Date: 4 Feb 2001 23:51:47 -0500a2 From: young_r@eisner.encompasserve.org (Rob Young)0 Subject: RE: Writes flushed to Disk on Dismount?3 Message-ID: <T$Z+OK3AYhHe@eisner.encompasserve.org>e  X In article <1010203162846.3117E-100000@Ives.egh.com>, John Santos <JOHN@egh.com> writes:! > On 2 Feb 2001, Rob Young wrote:o >  >> g >> 	Scenario.... >> o@ >> 	You wish to make things fast.  You hear through grapevine or; >> 	CETS sessions that setting CACHE_FLUSH_TIMER to 65000+  C >> 	"increases performance."  So blindingly, you make that setting.rI >> 	In your environment, you quiesce your database product/app , dismount B >> 	a contributing shadowset, back up that shadowset and return it% >> 	to be resynched.  The problem is:f >>  3 >> 			1)  On dismount, the HSG does not flush cache 1 >> 				writes to disk.  Conversely, the HSJ does.  >> d; >> 			2)  Because writes aren't flushed, there are a number-/ >> 				that aren't on that shadowset member you: >> 				just dismounted.t > L > Shouldn't matter, if there are no failures!  When BACKUP reads the removedG > member, it will get data that hasn't hit the disk yet from CACHE, nothL > from the disk.  Remember *HOST*-based volume shadowing...  The HSG doesn'tJ > know this disk is part of a shadow set.  It should be keeping its cachesI > on a per-drive basis, and satisfying reads for cached, unwritten blocksm$ > from the cache, not from the disk. > 6 >> 			3)  Your backups aren't worth plug nickel... but/ >> 				you are still okay if your service leveli0 >> 				agreement contains language to the effect& >> 				"backups are write-only"  (HA!) > I > Your backups are fine, assuming the cache (or HSG) doesn't fail between < > the time you dismounted and the time you write the backup. > B > What worries me is if you dismount the disk, then pull it out toB > transport it without waiting for the cache to flush, or power it > down for some reason.l > G > (I've seen dead batteries in HSx's and UPS that interrupted.  All youd# > need is both at the same time...)r >   > 	Yes.... and if you read further in the thread, I came to that5 	conclusion after Glenn Everhart straightened me out.r   				Robn   ------------------------------  # Date: Mon, 05 Feb 2001 06:34:29 GMTr% From: Uwe Zessin <zessin@my-deja.com> 0 Subject: RE: Writes flushed to Disk on Dismount?) Message-ID: <95lhhl$40q$1@nnrp1.deja.com>D  5 In article <1010203162846.3117E-100000@Ives.egh.com>, #   John Santos <JOHN@egh.com> wrote: ! > On 2 Feb 2001, Rob Young wrote:t > >  > > 	Scenario....  > >yA > > 	You wish to make things fast.  You hear through grapevine oru; > > 	CETS sessions that setting CACHE_FLUSH_TIMER to 65000+ D > > 	"increases performance."  So blindingly, you make that setting.A > > 	In your environment, you quiesce your database product/app ,eE > >     dismount a contributing shadowset, back up that shadowset and=0 > > 	return it to be resynched.  The problem is: > >l2 > >		1)  On dismount, the HSG does not flush cache0 > >			writes to disk.  Conversely, the HSJ does. > >n: > >		2)  Because writes aren't flushed, there are a number. > >			that aren't on that shadowset member you > >			just dismounted. >=D > Shouldn't matter, if there are no failures!  When BACKUP reads theD > removed member, it will get data that hasn't hit the disk yet fromF > CACHE, not from the disk.  Remember *HOST*-based volume shadowing...D > The HSG doesn't know this disk is part of a shadow set.  It shouldF > be keeping its caches on a per-drive basis, and satisfying reads for= > cached, unwritten blocks from the cache, not from the disk.  > 5 > >		3)  Your backups aren't worth plug nickel... but . > >			you are still okay if your service level/ > >			agreement contains language to the effectt% > >			"backups are write-only"  (HA!)e >nA > Your backups are fine, assuming the cache (or HSG) doesn't faileD > between the time you dismounted and the time you write the backup. > B > What worries me is if you dismount the disk, then pull it out toB > transport it without waiting for the cache to flush, or power it > down for some reason.b   Then don't do it that way!. Do a 'SET unit NORUN', was has been said here,B delete the unit, delete the storagesets + delete the DISK entries. Now remove the physical disks.  @ And the manual says you should never pull a disk while the power is down.   [...]>   --
 Uwe Zessin3 (If you want to send mail, please use user "zessin" / who lives at "decus.decus.de", not my-deja.com)r     Sent via Deja.comM http://www.deja.com/   ------------------------------   End of INFO-VAX 2001.071 ************************