1 INFO-VAX	Sun, 27 May 2001	Volume 2001 : Issue 292       Contents: Re: Affordable VMS Workstations  Re: Affordable VMS Workstations  Re: DPWS cpu freq ; Re: error: DECW$DEVICE-I_NODEVICE, no graphics device found  Re: IRCD for VMS Re: IRCD for VMS Rack mounted stuff.  Re: Rack mounted stuff. 6 Unable to Telnet & Rlogin from Solaris/HPUX to OpenVMS. Re: Very odd:  Init/erase is controller-based?. Re: Very odd:  Init/erase is controller-based?. Re: Very odd:  Init/erase is controller-based?- Re: Very odd: Init/erase is controller-based? 6 VMS Home Automation (was: Affordable VMS Workstations)  F ----------------------------------------------------------------------  % Date: Sun, 27 May 2001 11:03:44 -0400 ( From: Bill Gunshannon <bill@cs.uofs.edu>( Subject: Re: Affordable VMS WorkstationsK Message-ID: <Pine.LNX.4.10.10105271100200.8921-100000@triangle.cs.uofs.edu>   F Now that's ambitious.  And here is was trying to decide if I could getF the components for a homebrew computer controlled solar heating system for my swimming pool.   F Of course, if I do it, I'll use a PDP-11 rather than an Alpha.  EasierK to interface to, easier to program and designed for this kind of work.  :-) C And I have enough 11/02's to keep it going for the rest of my life.    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: Sun, 27 May 2001 11:14:47 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)( Subject: Re: Affordable VMS WorkstationsL Message-ID: <rdeininger-2705011114470001@user-2ive7o2.dialup.mindspring.com>  ; In article <3B106BDD.F1AA3072@fsi.net>, "David J. Dachtera"  <djesys.nospam@fsi.net> wrote:    K > > I'm not so sure.  Decent quality sensors for this sort of thing are not K > > too cheap.  Signal conditioners, A/D modules, and all the little extras J > > can be expensive too.  I don't think most folks would want their house9 > > monitored and controlled by junk-quality components.   > F > You may be over-complicating these. Simple devices are less prone toA > failure. Look at the temperature sensors on mobos, for example.   J We recently re-instrumented a large magnet at work.  This system has a fewJ dozen sensors: temperatures, voltages, and water flowmeters.  Many sensorsG put out very small signals; they have to be boosted if you run them any J distance in a noisy environment.  A house probably doesn't have as much EMI noise as our lab environment, but it's not like being inside a nice metal I computer case.  We had to use various amplifiers and signal conditioners, H good quality shielded cable, and fairly expensive 16-channel A/D modulesH in the computer.  The amplifiers need a chassis to supply power, etc.  IJ think all the "extras", not counting the sensors and the computer, came to around $200/channel.  E A home system wouldn't need quite the accuracy or reliability that we G demanded, so you could cut corners .Still, you wouldn't like it if your B house got 10 degrees colder every time your neighbor turned on hisJ dishwasher.  EM noise is evil, and it seems to seek out anything connectedH to a computer.   Cut too many corners, and your house would be broken as often as a Windoze PC.  I If I had to pull a number out of the sky, I'd guess you could get down to I $50/channel for household sensors and control circuits. That would add up  fairly fast.   > > It would be REAL; > > annoying when it broke, and maybe unsafe in some cases.  > H > The system would be self-diagnosing, issuing appropriate warnings when > components fail.  E Ok, then you have to double or triple the price for the components...   J And now you're starting to talk about real software, not the mickey-clickyH snotware that most of the industry is churning out.  It has to be almostD completely autonomous, yet allow Joe and Jane homemaker to adjust itH easily.  It should be reliable and self-diagnosing.  It should be safe. I This is not $49.95 software.  If you go to the trouble of making reliable H software, you find that peecees themselves are too quirky, the OS is tooE unstable, and the commercial application software that supports weird E electronics is usually a joke.  A reliable computer (alpha) running a F reliable OS (VMS) would be a dream in this environment.   But it's too, expensive to use in a household environment.  I But if you decide to do it yourself as a hobby, then money and time don't 
 count. ;-)  G (One of the ways VMS became so widespread in labs (and probably also in F factories, process control, etc.)  was by supporting a huge variety of@ hardware via Q-bus, and providing an incredibly stable, reliableG platform.  I still see a surprising amount of lab equipment attached to I decade-old microvaxes, still working just fine.  That doesn't happen with  PCs.)    --   Robert Deininger rdeininger@mindspring.com    ------------------------------  # Date: Sun, 27 May 2001 08:34:44 GMT  From: Dirk Munk <munk@home.nl> Subject: Re: DPWS cpu freq& Message-ID: <3B10BC23.3D912FF@home.nl>  & --------------BD999F351857867CE76278F1* Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit   ' These are the official switch settings:   D                 switch settings         system freq.    mult. factor system          8 7 6 5 4 3 2 1   9 433 a/au        d u d u u d d u         61.9 MHz        7 9 500 a/au        d u u u u d d u         62,5 MHz        8 9 600 a/au        d d u d d u d u         66,7            9    d = down = on , u = up      
 mhr wrote:  J > I'm trying to find info on the cpu switch settings for a DPWS setup. AllM > info gives 433 settings; what is the setting for 500,533 and 600 MHz chips?  >     Thanks >                     mhr   & --------------BD999F351857867CE76278F1) Content-Type: text/html; charset=us-ascii  Content-Transfer-Encoding: 7bit   > <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html>9 <tt>These are the official switch settings:</tt><tt></tt> a <p><tt>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; F switch settings&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; system) freq.&nbsp;&nbsp;&nbsp; mult. factor</tt> F <br><tt>system&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 8 7 6 5 4 3 2 1</tt><tt></tt> G <p><tt>433 a/au&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; d u d u u d d d u&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 61.9 MHz&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 7</tt>F <br><tt>500 a/au&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; d u u u u df d u&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 62,5 MHz&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 8</tt>F <br><tt>600 a/au&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; d d u d d uz d u&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 66,7&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 9</tt><tt></tt> " <p><tt>d = down = on , u = up</tt> <br><tt></tt>&nbsp;  <br><tt></tt>&nbsp;<tt></tt> <p><tt>mhr wrote:</tt>L <blockquote TYPE=CITE><tt>I'm trying to find info on the cpu switch settings for a DPWS setup. All</tt>H <br><tt>info gives 433 settings; what is the setting for 500,533 and 600 MHz chips?</tt> & <br><tt>&nbsp;&nbsp;&nbsp; Thanks</tt>z <br><tt>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; mhr</tt></blockquote>  <tt></tt></html>  ( --------------BD999F351857867CE76278F1--   ------------------------------  # Date: Sun, 27 May 2001 08:09:25 GMT 8 From: Veli =?iso-8859-1?Q?K=F6rkk=F6?= <korkko@decus.fi>D Subject: Re: error: DECW$DEVICE-I_NODEVICE, no graphics device found( Message-ID: <3B0FE956.19909A1C@decus.fi>   IIRC it would be OPEN3D V3.6.    _veli    Fred Kleinsorge wrote: > H > The 3D30 (aka TGA2 8-plane) is generally supported in the base system.J > Open3D replaces the driver/server DDX with ones that contain 3D support.N > However, you are on an OLD, OLD version of VMS (V6.2) so I don't remember ifM > 3D30 support was in the base OS or not (initial support for some cards came N > out first in Open3D).  You will need an OLD version of Open3D, I'm not quite- > sure how old, but it needs to support V6.2.  > M > With any luck, you still have the distribution CD laying around, and it has A > the same version of Open3D on it that was originally installed.  > = > martin wrote in message <3B098289.B07D9AC5@netscape.com>... G > >Does the OPEN3D kit include the graphics drivers? My video card is a  > >powerstorm 3d30.  > > 	 > >cheers 	 > >martin  > >  > >Chris Scheers wrote:  > >  > >> > >> >7 > >> > DECW$DEVICE-I_NODEVICE, no graphics device found  > >> >, > >> > Any suggestions would be appreciated. > >>J > >> Depending on your graphics card, you may need to load the OPEN3D kit. > >>L > >> -----------------------------------------------------------------------) > >> Chris Scheers, Applied Synergy, Inc.  > >>G > >> Voice: 817-237-3360            Internet: chris@applied-synergy.com  > >>   Fax: 817-237-3074 > >    ------------------------------  % Date: Sun, 27 May 2001 17:33:59 +0100 - From: "Edward Brocklesby" <ejb@leguin.org.uk>  Subject: Re: IRCD for VMS > Message-ID: <A1aQ6.9599$Hd6.167558@news11-gui.server.ntli.net>  J "Brian Schenkenberger, VAXman-" <system@SendSpamHere.ORG> wrote in message* news:009FC97F.A0F527AA@SendSpamHere.ORG...H > In article <1VPP6.5204$Hd6.101469@news11-gui.server.ntli.net>, "Edward' Brocklesby" <ejb@leguin.org.uk> writes: K > Gee, that's nice.  Perhaps it would even nicer if you explained what IRCD K > is.  Those that have not heard of this package might take an interest and  > give it a try.  " Sorry, I didn't think about that..K www.irchelp.org and similar sites explain exactly what it is.. other people  have already given the general idea.   L I'd be very interested in testing performance of ircd on VMS. Current serverJ platforms on the large IRC networks are Intel P2/P3.. it would probably be good= for VMS if someone could show a VMS server handling it well..        -larne-    ------------------------------  # Date: Sun, 27 May 2001 17:28:08 GMT = From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)  Subject: Re: IRCD for VMS 0 Message-ID: <009FCA39.5527206C@SendSpamHere.ORG>  W In article <9ep71o$1fcq$1@node21.cwnet.roc.gblx.net>, dsf@frontiernet.net (Dan Foster)  , crawled out from under some rock and writes:7 >In article <1010526163442.37731A-100000@Ives.egh.com>, = >>> VAXman- OpenVMS APE certification number: AAA-0001 wrote:  >>> N >>> Gee, that's nice.  Perhaps it would even nicer if you explained what IRCD M >>> is.  Those that have not heard of this package might take an interest and  >>> give it a try. >>> -- > I >My guess is that the initial announcement was more targeted toward those K >who had an existing interest and knowledge. Not an inherently unreasonable J >proposition. Sarcasm not necessary - if you were truly curious, could ask( >while checking the sarcasm at the door. > G >I'd answer your question, but looks like Mr. Santos has already nicely 3 >answered it, so I'll add just a little more to it.   I Perhaps reading comprehension was not an emphasized skill set during your J primary education but I had stated that "Those that have not heard of thisJ package..."  I recognized the significance of IRC in IRCD but it _is_ onlyJ a name.  For all we know, it may have been something completely different.  I I've asked that messages stating "I have ported Claptrap V1.0 to OpenVMS; J get your copy here!" contain a brief explaination of the package's intent, function or audience.     J BTW, this IRCD barfs in its build complaining CC-E-NOTEXPECTING errors.  IJ wonder why?  Before you answer this, the question was rhetorical.  Hmm.  II compiled on OpenVMS 7.2 Alpha which "IRCD is known to work on".  I wonder H if it could be that no compiler version was specified?  Gee!  No versionH was/is specified in the post or in the README.VMS so I guess any version is sufficient?  K I applaud the efforts to port utilities to VMS, but many too many require a I build on VMS and fail to cover the details necessary to build them.  What K happens then?  Bandwidth consumed with silly questions about how to get the I package to compile/build successfully.  Of course, C is portable so there ! shouldn't be any problems, right?    --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM              O city, n., 1. a place where trees are cut down and streets are named after them.    ------------------------------  % Date: Sun, 27 May 2001 13:02:16 +0100 ; From: "Leigh G. Bowden" <LGBowden@bowdenfamily.fsnet.co.uk>  Subject: Rack mounted stuff.. Message-ID: <9eqqd4$78d$1@news8.svr.pol.co.uk>  I We want to put a lot of VAX and Alpha equipment into racks - say six to a I rack. On the top shelf we want to put a VT as console but use a switch to - select which machine is active on the screen.   E Does anybody make a system that can switch do this or is it easier to " knobble a PC multi-printer switch?   ------------------------------  % Date: Sun, 27 May 2001 11:21:38 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)  Subject: Re: Rack mounted stuff.L Message-ID: <rdeininger-2705011121380001@user-2ive7o2.dialup.mindspring.com>  @ In article <9eqqd4$78d$1@news8.svr.pol.co.uk>, "Leigh G. Bowden"* <LGBowden@bowdenfamily.fsnet.co.uk> wrote:  K > We want to put a lot of VAX and Alpha equipment into racks - say six to a K > rack. On the top shelf we want to put a VT as console but use a switch to / > select which machine is active on the screen.  > G > Does anybody make a system that can switch do this or is it easier to $ > knobble a PC multi-printer switch?  J Digital made such a switch at one time.  I think I've seen a 4-port one on* ebay.  You might find something like this.  H Another way is to use a terminal server.  Attach each computer's consoleJ port to a port on the server, and attach the VT to another port.  ProperlyI configured, the terminal can connect to any of the other ports.  Terminal B servers also connect to ethernet, so you could access the consoles remotely if you needed to.  G Used terminal servers are fairly cheap, and I think a few companies are F still making new ones.  For this application, you'd want a server that@ could be could be configued to work without a remote boot from aJ computer.  You don't want the situation where the terminal server is inert2 until you get a computer going to boot the server.   --   Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Sun, 27 May 2001 20:18:57 +0530 , From: Shiv <Sivakumar.Thangavelu@oracle.com>? Subject: Unable to Telnet & Rlogin from Solaris/HPUX to OpenVMS * Message-ID: <3B1113D9.471692C3@oracle.com>   Gurus,  L I'm unable to connect to our VMS servers running DIGITAL TCP/IP Services forN OpenVMS Alpha Version V5.0A - ECO 2 / OpenVMS V7.2-1H1  using Telnet or Rlogin! over WAN from Solaris/HPUX boxes.   L But Telnet works fine if we connect from WindowsNT/OpenVMS boxes on the same network.  H Also Telnet & Rlogin works fine with servers on the same cluster runningI DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A - ECO 1 / OpenVMS % V7.2-1H1. Could ECO 2 be the culprit?   0 Any help/insights will be greatly appreciated...   TIA, Shiv   ---*---                        ~\\\~                       (`O-O')+     _________.oOo_____( ^ )____oOo.________ ,     \                                      \-      \          Sivakumar Thangavelu        \ .       \    (Platform Technologies Group)     \.       |--------------------------------------|.       |      Oracle Corporation, (IDC)       |.       |       Oracle Technology Park         |.       |       #3, Bannerghatta Road          |.       |      Bangalore-560029, India         |.       |--------------------------------------|.      /  Phone: +91 (80) 552 8335 x 1056      /-     /Email: Sivakumar.Thangavelu@oracle.com / ,    /____________.oooO___Oooo.______________/                 (   )   (   )                   \ (     ) /                   \_)   (_/    ------------------------------  # Date: Sun, 27 May 2001 13:52:37 GMT9$ From: Ed Wilts <ewilts@mediaone.net>7 Subject: Re: Very odd:  Init/erase is controller-based?S< Message-ID: <FE7Q6.37403$V6.1890243@typhoon.mn.mediaone.net>   Scott Vieth wrote:  C > I expected that if I did an init/erase on a dozen disks at a time J > (in 12 separate batch jobs), that it would bury the AlphaServer io-wise. > E > Boy, was I wrong.  Instead of doing tons of i/o to erase the disks,?( > the system hardly does any i/o at all. > B > It seems that the init/erase command sends a command to the HSJsJ > that says "do an erase operation on the whole disk and report back to me > when you are finished."- > B > I could have erased all of our disks at one time and not had any > impact on the host.u  L That depends on what else is happening on your hosts and drives.  Obviously I you're saturating your controllers so any other I/O on those controllers w would be impacted.  A > The more drives that were being erased, the lower the per-driveaJ > data throughput.  The total amount of data being moved by the controllerE > wouldn't go any higher than 8MB/sec.  I don't know where that limitr > is coming from....  K That's the write limit of the HSJs and is clearly documented.  The HSJ-50s  > are a bit faster than the 40s, but that's still about right.    F > Anyway, thought I would share this little tidbit with the group.  It > seemsoA > (at least with HSJs), that INIT/ERASE is handled totally by thee > controllers.  H It's been this way for years and years...  Even the HSCs did init/erase * locally (I know, I'm dating myself again).   -- i Ed Wilts, Mounds View, MN, USA mailto:ewilts@mediaone.net   ------------------------------  # Date: Sun, 27 May 2001 17:52:34 GMTo$ From: Scott Vieth <svieth@wi.rr.com>7 Subject: Re: Very odd:  Init/erase is controller-based?y) Message-ID: <3B113F59.4B923DE4@wi.rr.com>o   Ed Wilts wrote:i  C > > The more drives that were being erased, the lower the per-driveiL > > data throughput.  The total amount of data being moved by the controllerG > > wouldn't go any higher than 8MB/sec.  I don't know where that limitl > > is coming from.... >tL > That's the write limit of the HSJs and is clearly documented.  The HSJ-50s> > are a bit faster than the 40s, but that's still about right.  H I did a little more testing and found that I could move 8 MB/sec throughC one controller of an HSJ50 pair while the other controller was also.E doing 8MB/sec.  So it pays to try to "load balance" your disks acrossy% the two controllers in an HSJxx pair.n  
 -Scott :^)   ------------------------------  # Date: Sun, 27 May 2001 17:56:33 GMT $ From: Scott Vieth <svieth@wi.rr.com>7 Subject: Re: Very odd:  Init/erase is controller-based?s) Message-ID: <3B114048.967ADEF6@wi.rr.com>    >,C > > The more drives that were being erased, the lower the per-drivemL > > data throughput.  The total amount of data being moved by the controllerG > > wouldn't go any higher than 8MB/sec.  I don't know where that limit  > > is coming from.... >uL > That's the write limit of the HSJs and is clearly documented.  The HSJ-50s> > are a bit faster than the 40s, but that's still about right.  J Q1. Does anyone what the write limit is for the HSG80 controllers?  If so,! please share it with the group...t  < I can't do an INIT/ERASE on a bunch of disks on our ESA12000? as all 48 disk drives are now in production.  I'll have to wait > until I add the third BA370 shelf and then I'll INIT/ERASE all 24 of those disks at one time.   -Scott2 "and what did *you* do for Memorial Weekend?"  ;^)   ------------------------------  # Date: Sun, 27 May 2001 16:42:10 GMT 7 From: moroney@world.std.spaamtrap.com (Michael Moroney)w6 Subject: Re: Very odd: Init/erase is controller-based?& Message-ID: <GE06EA.Co9@world.std.com>  & Scott Vieth <svieth@wi.rr.com> writes:   >All:   I >I was using init/erase this week to scrub any useful data off the dozensh  E >and dozens of RZ disks that we're going to be retiring this weekend.d  B >I expected that if I did an init/erase on a dozen disks at a timeI >(in 12 separate batch jobs), that it would bury the AlphaServer io-wise.o  D >Boy, was I wrong.  Instead of doing tons of i/o to erase the disks,' >the system hardly does any i/o at all.b  K Yes.  INIT/ERASE will issue an IO$_DSE command rather than an IO$_WRITExBLKsE to the disk. (DSE=DATA SECURITY ERASE)  The blockcount is most likelyVD whatever the max block count of the disk is.  There is no data to beD transferred.  This gets converted by DUDRIVER to the appropiate MSCP9 command to the HSJ.  This is the extent of the IO on VMS.e  I The HSJ is responsible for converting the erase command to the appropiateeH writes of zeros to the SCSI drives, unless there is a SCSI erase command (I don't think there is).i  H You may find that INIT/ERASE is *much* less load intensive for HSJ disksK than fibrechannel disks, if there is no SCSI/fibrechannel erase equivalent.a; You may find out if you issue INIT/ERASE of your new disks.i   -Miken   ------------------------------    Date: 27 May 2001 08:43:17 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)R? Subject: VMS Home Automation (was: Affordable VMS Workstations)-3 Message-ID: <ytUipHwMhxtb@eisner.encompasserve.org>p  [ In article <3B100CB3.C8E45BF6@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:n  / > I can see developing *TRUE* home automation. t > D > No, I'm not just talking about turning lights on and off by remoteI > control, or making coffee before you get up (a timer will suffice here) F > or cranking up the heat/cooling before you arise or return home fromJ > work (an automatic thermostat will suffice here, within limits), either. > J > I'm talking about sounding an alarm or automatically taking action under > certain circumstances:  A There is a starting point (VMS-based) provided by John Covert at:   > 	http://eisner.decus.org/DECUServe/DECnotes/VMS/3064/SDIR.HTML   ------------------------------   End of INFO-VAX 2001.292 ************************