1 INFO-VAX	Tue, 07 Oct 2003	Volume 2003 : Issue 555       Contents: Re: %MOUNT-W-INCONSIZE Re: affordable VMS Re: affordable VMS Re: AMD64 sales figures  RE: AMD64 sales figures  RE: AMD64 sales figures 3 Re: Another huge bug: DECwindows MAIL: include file 3 Re: Another huge bug: DECwindows MAIL: include file 3 Re: Another huge bug: DECwindows MAIL: include file 3 Re: Another huge bug: DECwindows MAIL: include file ' Re: Another VMS advertising opporutnity  Re: DCL improvements Re: DCL improvements Re: DS15 systems have arrived @ Re: Fee Based Email (From Re: Process's PreciseMail AntiSpam...)@ Re: Fee Based Email (From Re: Process's PreciseMail AntiSpam...)7 Re: Fixed device names for incoming Telnet connections?  MD5 source code ?  Re: MD5 source code ?  Re: MD5 source code ? $ Re: Missing delta time functionality Re: ODBTP on VMS ?P Problem with CLUSTER_CONFIG? (was: Re: DIFVOLMNT (%X0072832C), then bugcheck wheD Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?D Re: Process's PreciseMail AntiSpam Gateway - any experience so far ? Re: Question about SMTP.CONFIG Re: SHOW DEVICE/FULL DSA Re: Sun takes a hit  Re: Sun takes a hit  Re: Sun takes a hit  Re: Sun takes a hit  Re: Sun takes a hit  Re: Sun takes a hit  Re: Sun takes a hit  Re: Sun takes a hit  Re: Sun takes a hit  Re: Sun takes a hit " Re: Unrecognized printer with DCPS% Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue ) Re: Vax / VMS V5.5-2 External Drive Issue  Re: VMS system on the web   F ----------------------------------------------------------------------  $ Date: Mon, 6 Oct 2003 19:05:16 +0200( From: "H Vlems" <hvlems.nieuw@zonnet.nl> Subject: Re: %MOUNT-W-INCONSIZE 9 Message-ID: <bls7fu$ffno3$1@ID-143435.news.uni-berlin.de>   D "Robert Deininger" <rdeininger@mindspringdot.com> schreef in berichtF news:rdeininger-0510031450070001@user-uinj4rq.dialup.mindspring.com...E > In article <blp55l$eia5c$1@ID-143435.news.uni-berlin.de>, "H Vlems" ! > <hvlems.nieuw@zonnet.nl> wrote:  > - > >"Phillip Helbig---remove CLOTHES to reply" ! <helbig@astro.multiCLOTHESvax.de> 4 > >schreef in bericht news:blov67$2qu$1@online.de...9 > >> Further adventures of a hobbyist cluster, continued.  > >>F > >> Experimenting with a VAX 4000 100A, I have a BA350 hooked up.  ItG > >> appears to have some sort of DSSI-to-SCSI adapter attached, and is D > >> attached to the big multi-function connector on the back of the machine. > >  > >Is it a DWZZA-nn perhaps? > D > DWZZA is differential SCSI to single-ended (narrow) SCSI.  I think! > DSSI-to-SCSI devices are HSDnn.  >   4 Ah, yes indeed, Phillip probably has an HSD05 or so!   > >>4 > >> The system disk on this machine is a DSSI disk. > >>J > >> I stuck in a couple of RZ28-VA disks, powered up the BA350, and did a > >> SYSGEN AUTOCONFIGURE ALL. > >>L > >> SHOW DEVICE now reports a different controller name than for the systemJ > >> disk (i.e. the "node name" in parentheses after the device name).  ItJ > >> also has an allocation class which is the same as another node in the
 > >> cluster. K > >The DSSI/SCSI bus converter behaves similar to an HSC50 in a CI cluster. F > >Where a DSSI disk has its own built-in controller that lets you set ALLCLASSL > >etc., SCSI disks do not have that feature of course. The DWZZA does that.I > >IIRC it also allows you to assign device names to the disks, so if you G > >prefer DUA names instead of DIA then the DWZZA will do that as well.  > F > DWZZA devices are completely passive.  They don't supply ALLCLASS orJ > change device names.  But I don't think he has a DWZZA in this location, > see above.  J Correct, but the HSDnn controllers do allow that feature. When I wrote theJ answer I had the HSD05 in mind and accidentally referred to it as a DWZZA.; Which is as you pointed out an entirely different beast :-)    Hans   ------------------------------  * Date: Mon, 6 Oct 2003 18:22:25 +0000 (UTC)7 From: moroney@world.std.spaamtrap.com (Michael Moroney)  Subject: Re: affordable VMS ( Message-ID: <blsbt1$h3l$1@pcls4.std.com>  % "John Smith" <a@nonymous.com> writes:   
 >Advertising. 
 >Advertising. 
 >Advertising.     G You know, I really have to wonder if that TV commercial that advertises F advertising and uses the Pentium logo/xylophone notes was a subtle digD at DEC/Compaq/HP: "Look, Intel heavily advertised Pentium and now itF rules.  DEC did not advertise Alpha at all and it awaits execution..."   --   -Mike    ------------------------------  # Date: Tue, 07 Oct 2003 02:52:00 GMT # From: "John Smith" <a@nonymous.com>  Subject: Re: affordable VMS K Message-ID: <kTpgb.244540$Lnr1.153610@news01.bloor.is.net.cable.rogers.com>    Michael Moroney wrote:' > "John Smith" <a@nonymous.com> writes:  >  >> Advertising.  >> Advertising.  >> Advertising.  >  > > > You know, I really have to wonder if that TV commercial thatF > advertises advertising and uses the Pentium logo/xylophone notes was@ > a subtle dig at DEC/Compaq/HP: "Look, Intel heavily advertisedF > Pentium and now it rules.  DEC did not advertise Alpha at all and it > awaits execution..."    K It's about buying the correct tool for the job and knowing that you need it  before you begin the job.   & See if you can spot the metaphor......  I Let's say you are a general contractor building a new airport. You need a E lot of concrete and rebar to build the runways to handle fully loaded J 747-400 or A380 freighters. You put an RFP out for tender for the delivery of the concrete and rebar.  J Bidder 1 comes in at the low price - he has a fleet of 1 ton pickup trucksI that he says will do the job just fine. He takes you out for lunch, gives H you tickets to some hot Broadway shows, a 'free' SUV complete with fuzzy5 dice to hang on your rearview mirror, and a keychain.   J Bidder 2 is the mid-price bidder - he has a fleet of  mid-sized trucks. He4 takes you out for several rounds of golf and dinner.  I Bidder 3 comes in at the high price - he has proposed an on-site concrete A plant, delivering the concrete with 30 ton mixers, and a fleet of C semi-trailers to bring the rebar in. His sales pitch is his written H proposal. He is quietly confident that you'll see the correctness of hisI approach and that you'll chose him. He doesn't give you any trinkets, nor  does he wine and dine you.    E You choose Bidder 1 because he's the 'cheapest'. Pretty soon you have G hundreds of broken pickup trucks scattered all over the airport, like a K field of anti-tank obstacles with their harddened concrete loads and broken  suspensions.  K So you fire Bidder 1 and hire Bidder 2 to finish the job. Bidder 2's trucks H don't break down but they just can't haul enough concrete and rebar fastJ enough and pretty soon you realize that the job will never be completed inK time before you, the general contractor, are hit with performance penalties , in your contract with the airport authority.  G Sadly, belately, you realize that you should have hired Bidder 3 in the J first place. You wish that somebody had done a better sales job on you forK Bidder 3 before you made your ill-fated decisions to hire Bidder 1 and then I Bidder 2. But by then the airport authority is thinking of firing you for 5 poor performance, and suing your ass off for damages.    ------------------------------  % Date: Mon, 06 Oct 2003 20:15:01 -0400 * From: "Stanley F. Quayle" <stan@stanq.com>  Subject: Re: AMD64 sales figures/ Message-ID: <3F81CD45.23084.1533A510@localhost>   ) On 6 Oct 2003 at 7:28, Bob Koehler wrote: I >    Charon-VAX looks like VAX hardware to whatever OS is running on it.  J >    Most folks seem to use it to run VMS, but IIRC Ultrix has been booted& >    on it and VAXeln should also run.   Both work perfectly.  4 >    Since special drivers are not used you can haveC >    problems sharing the PC's ethernet port with Windows, they can J >    both talk to the rest of the world, but not to each other.  I believeH >    someone posted a link to a special driver for Windows to get around >    that last one.   F Performance is terrible.  I tell my customers to get a second network 6 card.  My local computer store is selling them for $5.  
 --Stan Quayle  Quayle Consulting Inc.  
 ----------C Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363  Fax: +1 614 868-1671 1 8572 North Spring Ct. NW, Pickerington, OH  43147 = Preferred address:  stan@stanq.com       http://www.stanq.com    ------------------------------   Date: 6 Oct 2003 19:27 CDT' From: carl@gerg.tamu.edu (Carl Perkins)   Subject: RE: AMD64 sales figures, Message-ID: <6OCT200319270614@gerg.tamu.edu>  ' "Tom Linden" <tom@kednos.com> writes... A }Carl, you missed the point.  Take the speed as hypothetical then  }as the question.   5 I disagree. I don't think you know what the point is.   C You still can't guess an answer because any Pentium running at that G speed will have a different core than the current one (the current core F is just about at the end of the line). And it isn't just the core, howF much cache will it have at each of L1, L2, and L3 on chip? Don't know?J Neither do I, or probably anone else here. Now that the current generationF of Pentium 4s have "Hyperthreading" (2 thread SMT), will the next coreJ have the same number of functional units or will they maybe add an integerI unit or two (or maybe even a floating point unit) so as to better be able 1 keep the two threads busy? I don't know - do you?   B Chips with the "Pentium" brand have varied in design considerably.A Even within the "Pentium 4" family, they have varied considerably @ (when they added Hyperthreading, they really should have gone to "Pentium 5", I think).  G Since we don't know the characteristics of a chip that far out, even if L it is just the next one, on what basis should someone formulate their guess?  I It is already pretty hard to say how fast something like that will run on G the P4EE compared to the regular P4 at the same 3.2GHz. The added cache K may make a huge difference, or only a little. Someone has to try it to find F out. (That difference should be pretty much the same as the differenceF between a Xeon and a P4 at the same clock speed and same frontside bus speed. Probably.)   E Why not just ask how fast it is on as system built with the currently C availabe 3.2GHz P4 with 800MHz FSB and dual channel DDR400 memory - B you might also want to specity that it use the Intel 875P chipset,E as I think that is currently the fastest by a slim margin. That would F be the fastest type of P4 system you can currently buy, and as such itE could actually be tested by someone to get a real answer instead of a  guess.  G If you like questions that have no answer, here's one: For running VMS, I what is the equivalent Pentium (runing the emulator) to a 2GHz Alpha EV8?    --- Carl   }>-----Original Message-----0 }>From: Carl Perkins [mailto:carl@gerg.tamu.edu]( }>Sent: Sunday, October 05, 2003 1:04 AM }>To: Info-VAX@Mvb.Saic.Com " }>Subject: RE: AMD64 sales figures }> }>0 }>"Stanley F. Quayle" <stan@stanq.com> writes...+ }>}On 4 Oct 2003 at 8:13, Tom Linden wrote: = }>}> Stan, what is the equivalent Alpha to a 4.3 GHz Pentium?  }>}> (Running VMS, of course)  }>} C }>}I don't have any data at the moment.  Once I know, I'll let you  
 }>}know... }>}  }>}--Stan Quayle }>A }>There is one vital bit of information that gives you the answer @ }>directly: there isn't a 4.3 GHz Pentium. Therefore there is no }>equivalent.  }>E }>The fastest Pentium is the Pentium 4EE (EE = Extreme Edition) which F }>is actually not quite out yet - a few samples were released early toH }>reviewers to spoil the Athlon64 release. The P4EE is clocked at 3.2GHzF }>just like the fastest non EE verion, but the EE is faster because itG }>is really a slightly modified Xeon with "P4" stamped on it instead of H }>"Xeon" - the main difference with the regular P4 being that is has 2MBG }>of L3 cache on chip, the difference between it and the normal Xeon is D }>that they upped the frontside bus to 200MHz (QDR, for an effectiveC }>800MHz data rate - the fastest real Xeon frontside bus is at 533) F }>and they disabled a lot of the internal ECC in the chip (which couldE }>be a bad thing, although probably no worse than a normal P4) to get  }>it to run so fast. }>
 }>--- Carl   ------------------------------  $ Date: Mon, 6 Oct 2003 19:37:21 -0700# From: "Tom Linden" <tom@kednos.com>   Subject: RE: AMD64 sales figures9 Message-ID: <CIEJLCMNHNNDLLOOGNJIEEDCIDAA.tom@kednos.com>    >-----Original Message----- / >From: Carl Perkins [mailto:carl@gerg.tamu.edu] ' >Sent: Monday, October 06, 2003 5:27 PM  >To: Info-VAX@Mvb.Saic.Com! >Subject: RE: AMD64 sales figures  >  > ( >"Tom Linden" <tom@kednos.com> writes...B >}Carl, you missed the point.  Take the speed as hypothetical then >}as the question. > 6 >I disagree. I don't think you know what the point is. > D >You still can't guess an answer because any Pentium running at thatH >speed will have a different core than the current one (the current coreG >is just about at the end of the line). And it isn't just the core, how G >much cache will it have at each of L1, L2, and L3 on chip? Don't know? K >Neither do I, or probably anone else here. Now that the current generation G >of Pentium 4s have "Hyperthreading" (2 thread SMT), will the next core K >have the same number of functional units or will they maybe add an integer J >unit or two (or maybe even a floating point unit) so as to better be able2 >keep the two threads busy? I don't know - do you? > C >Chips with the "Pentium" brand have varied in design considerably. B >Even within the "Pentium 4" family, they have varied considerablyA >(when they added Hyperthreading, they really should have gone to  >"Pentium 5", I think).  > H >Since we don't know the characteristics of a chip that far out, even if@ >it is just the next one, on what basis should someone formulate
 >their guess?  > J >It is already pretty hard to say how fast something like that will run onH >the P4EE compared to the regular P4 at the same 3.2GHz. The added cacheL >may make a huge difference, or only a little. Someone has to try it to findG >out. (That difference should be pretty much the same as the difference G >between a Xeon and a P4 at the same clock speed and same frontside bus  >speed. Probably.) > F >Why not just ask how fast it is on as system built with the currentlyD >availabe 3.2GHz P4 with 800MHz FSB and dual channel DDR400 memory -C >you might also want to specity that it use the Intel 875P chipset, F >as I think that is currently the fastest by a slim margin. That wouldG >be the fastest type of P4 system you can currently buy, and as such it F >could actually be tested by someone to get a real answer instead of a >guess.   < Fine, I would be happy with at least one point on the curve. > H >If you like questions that have no answer, here's one: For running VMS,J >what is the equivalent Pentium (runing the emulator) to a 2GHz Alpha EV8? > 	 >--- Carl  >  >}>-----Original Message----- 1 >}>From: Carl Perkins [mailto:carl@gerg.tamu.edu] ) >}>Sent: Sunday, October 05, 2003 1:04 AM  >}>To: Info-VAX@Mvb.Saic.Com# >}>Subject: RE: AMD64 sales figures  >}>  >}> 1 >}>"Stanley F. Quayle" <stan@stanq.com> writes... , >}>}On 4 Oct 2003 at 8:13, Tom Linden wrote:> >}>}> Stan, what is the equivalent Alpha to a 4.3 GHz Pentium? >}>}> (Running VMS, of course) >}>}C >}>}I don't have any data at the moment.  Once I know, I'll let you  >}>}know...  >}>} >}>}--Stan Quayle  >}> B >}>There is one vital bit of information that gives you the answerA >}>directly: there isn't a 4.3 GHz Pentium. Therefore there is no  >}>equivalent. >}> F >}>The fastest Pentium is the Pentium 4EE (EE = Extreme Edition) whichG >}>is actually not quite out yet - a few samples were released early to I >}>reviewers to spoil the Athlon64 release. The P4EE is clocked at 3.2GHz G >}>just like the fastest non EE verion, but the EE is faster because it H >}>is really a slightly modified Xeon with "P4" stamped on it instead ofI >}>"Xeon" - the main difference with the regular P4 being that is has 2MB H >}>of L3 cache on chip, the difference between it and the normal Xeon isE >}>that they upped the frontside bus to 200MHz (QDR, for an effective D >}>800MHz data rate - the fastest real Xeon frontside bus is at 533)G >}>and they disabled a lot of the internal ECC in the chip (which could F >}>be a bad thing, although probably no worse than a normal P4) to get >}>it to run so fast.  >}>  >}>--- Carl  >  >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.518 / Virus Database: 316 - Release Date: 9/11/2003  >  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.518 / Virus Database: 316 - Release Date: 9/11/2003   ------------------------------  # Date: Mon, 06 Oct 2003 19:02:40 GMT 9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> < Subject: Re: Another huge bug: DECwindows MAIL: include file/ Message-ID: <k%igb.6415$co.69@news.cpqcorp.net>   L Bug, feature, whatever.  Given that the code in question has worked this way< for over a decade, I will challenge calling it a "huge" bug.  G If you have an issue with the way it currently works, please file a bug H report.  Martin Kirby has already replied in here - which means that theI maintainers are aware of your POV.  If you want them to actually consider K taking an action on the behavior in some known timeframe - you need to take H the next step (that is, the newsgroup isn't an official way to request a functionality change).   Thanks.    ------------------------------  % Date: Mon, 06 Oct 2003 15:57:59 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>< Subject: Re: Another huge bug: DECwindows MAIL: include file) Message-ID: <3F81C946.565217D5@istop.com>    Fred Kleinsorge wrote: > N > Bug, feature, whatever.  Given that the code in question has worked this way> > for over a decade, I will challenge calling it a "huge" bug.  M One may consider this just a "feature" on other operating systems where there M are bugs with much more severe impacts (such as on windows), but since VMS is K of such high quality, of the bugs that remain, this one would be considered  BIG :-) :-) :-) :-) :-) :-)   N Seriously, since this is more of a feature request, as long as the developpersL are aware of it, I trust their expertise and judgement. If they feel it is a7 valid one, they can then implement it at their leasure.   E I am not an important enough customer to demand changes to VMS if the 9 engineers don't themselves feel the change is a good one.    ------------------------------  # Date: Mon, 06 Oct 2003 20:39:47 GMT 9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> < Subject: Re: Another huge bug: DECwindows MAIL: include file1 Message-ID: <nqkgb.6429$0H.2684@news.cpqcorp.net>    > G > I am not an important enough customer to demand changes to VMS if the ; > engineers don't themselves feel the change is a good one.   J We value all of our customers.  Putting this into the bug reporting systemG assures you of some response.  It may not have a high priority - but at K least it will be in the system and someone can make the determination if it  is really a bug or a feature.    ------------------------------  % Date: Tue, 07 Oct 2003 01:05:46 +0100 - From: John Laird <nospam@laird-towers.org.uk> < Subject: Re: Another huge bug: DECwindows MAIL: include file8 Message-ID: <d104ov0oqobeqd5crd0nsaggpnmtlblrti@4ax.com>  H On Mon, 06 Oct 2003 12:04:43 -0400, JF Mezei <jfmezei.spamnot@istop.com> wrote:   >Bob Koehler wrote: F >>    I would say that this is not a bug.  A decision to make access aG >>    potentially incomplete and possibly inconsistent file should take  >>    an extra step.   > K >If the writer of the file has explicitely said that it is ok for others to @ >read the file, then why should a reader decide not to read it ?  K Each "reader" makes up its own mind how flexible it is prepared to be.  Log F files are pretty much a special case in permitting concurrent readers,F because it is very useful to have that ability.  Obvious candidates toK specify share access on reads are then TYPE, SEARCH.  So they work.  As you K say, TPU will work but EDT won't.  CONVERT (with /share) and SORT will too.  But COPY won't.   J If you genuinely think this is a huge bug, then it follows that you expectL every other utility provided in VMS (and almost by implication any availableF file-reading API and any piece of third party software) to be changed.4 That's not going to happen, and nor is it desirable.   --   Mail john rather than nospam...    ------------------------------  % Date: Mon, 06 Oct 2003 19:02:01 +0200  From: "B.Eckstein" <eck@ivu.de> 0 Subject: Re: Another VMS advertising opporutnity- Message-ID: <bls76h$dj8$2@swifty.westend.com>   , John Smith schrub im Jahre 06.10.2003 16:37:  M > Will carly(tm), marcello, et. al. take time from their hunt for leaks (does N > anyone remember the paranoia of the White House 'Plumbers'?) and start doingM > their real job - marketing and promoting the sales of VMS to new customers?   ? It is more likely that Earth will become a flat disk again. ;-)    --  G B.Eckstein, eck@ivu.de         Cheap, Fast, Good - pick any two of them 9 Die FAQ zu de.comp.hardware.netzwerke: http://how.to/dchn G Mozilla-Tips: http://mozilla-anleitung.de/ http://www.holgermetzger.de/    ------------------------------  $ Date: Mon, 6 Oct 2003 19:44:13 +0200( From: "H Vlems" <hvlems.nieuw@zonnet.nl> Subject: Re: DCL improvements 9 Message-ID: <bls9ov$g6fja$1@ID-143435.news.uni-berlin.de>    Guy,  E did you read the post of Brian? He feels that this change would break L existing DCL code. So perhaps the solution is a little more complicated then
 I thought.   Hans1 "Guy Peleg" <guy.peleg@hp.com> schreef in bericht   news:3F814741.DED2BD9D@hp.com...
 > Hi Hans, > < > I'm currently working on WHILE - expected release is V8.2. > I > As for F$GETDVI("AVL") I agree that FALSE should be returned. Could you  please > log @ > a call with your local support center and have it escalated to engineering? I  > will provide you a fix to your; > current version and will make sure to include it in V8.2.  >  > Guy  >  > H Vlems wrote: >  > > Guy, > > G > > now that you convinced us all that DCL is the way to go I'd like to  suggest  > > a few improvements:  > > H > > 1) a WHILE statement that allows building loops a little better than using  > > GOTO's:  > > ' > >     WHILE <boolean expression> [DO] 	 > >     . 	 > >     .  > >     ENDWHILE > > E > >     The keyword DO is optional, it serves no syntactical purpose. G > >     A FOR statement ( la ALGOL60) would be equally fine but offers  little > > more functionality.  > > H > > 2) IIRC you mentioned a CASE statement as well. Though it simplifies codingF > > multiple evaluations of expressions, it may also slow down the DCL5 > > processor. If so, I'd happily do without CASE :-)  > > H > > 3) A CHECK command would be nice: it would check the syntax of a DCL command , > > procedure without actually executing it. > > L > > 4) F$GETDVI("<diskdevice>","AVL") returns an error message when a device isK > > not available, and crash the DCL code. Would it break other software if  it > > would return FALSE instead?  > > G > > 5) Is it possible to implement FUNCTIONs with parameters that allow  return > > values?  > >  > >     FUNCTION FUNC(I,J)	 > >     . 	 > >     .  > >     RETURN  (I+J)  > >  > > Hans Vlems >    ------------------------------   Date: 6 OCT 2003 21:14:28 GMT + From: Dave Greenwood <greenwoodde@ornl.gov>  Subject: Re: DCL improvements 1 Message-ID: <6OCT03.21142891@feda01.fed.ornl.gov>   : In a previous article, Guy Peleg <guy.peleg@hp.com> wrote:
 > Hi Hans, >   < > I'm currently working on WHILE - expected release is V8.2. >   P > As for F$GETDVI("AVL") I agree that FALSE should be returned. Could you please > log O > a call with your local support center and have it escalated to engineering? I   > will provide you a fix to your; > current version and will make sure to include it in V8.2.   C I'll have to agree with the previously noted concern about breaking D existing code with this change.  I'll also note that if you made theC change for AVL I'd expect you to make the same change for *all* the  getdvi item codes.   Dave --------------9 Dave Greenwood                Email: Greenwoodde@ORNL.GOV H Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------   Date: 7 Oct 2003 03:56:30 GMT  From: healyzh@aracnet.com & Subject: Re: DS15 systems have arrived+ Message-ID: <bltdhe15gr@enews4.newsguy.com>   " John Smith <a@nonymous.com> wrote:J > The last keyboard and mouse I purchased a few weeks ago were USB devices0 > which came with a thumb-sized USB-PS2 adapter.  J OK, I'm not sure how to take this, does the adapter convert USB-to-PS2, or PS2-to-USB?   L I've seen a PS2-to-USB converter and have been thinking very seriously aboutM getting one to try on my PowerMac (I want a decent logitech 3-button mouse).  E I'm already using an ADB-to-USB converter so that I can have a decent  keyboard on the system.   J When I buy another PC in the future, I've a feeling I'll need one of theseL for it, since PS/2 ports seem to be soon to be a thing of the past, and I'llI want to continue to use the same keyboard I have for the last 12 years if 6 possible, as well as a decent logitech 3-button mouse.  L I've not seen a USB-to-PS2, which seems to be what you're saying it had, and; quite honestly the idea of such a device seems a bit silly.    			Zane    ------------------------------  * Date: Tue, 7 Oct 2003 01:01:11 +0000 (UTC) From: david20@alpha2.mdx.ac.ukI Subject: Re: Fee Based Email (From Re: Process's PreciseMail AntiSpam...) ) Message-ID: <blt38n$27a$1@news.mdx.ac.uk>   o In article <blrfs6$8l3$1@n.ruf.uni-freiburg.de>, gartmann@non.immunbio.mpg.de.sens (Christoph Gartmann) writes: K >In article <blpfc4$ohi$1@news.mdx.ac.uk>, david20@alpha2.mdx.ac.uk writes: L >>This isn't the case. The TSL/SSL encryption is between central mailhubs itQ >>is NOT end to end encryption of a mail message content like for instance PGP or K >>SMIME. To the end users sending or receiving the mail message their is no Q >>indication that it has been encrypted in it travels - the sender just sends the O >>message. If SPAM is passed to central mailhub A destined for a user served by G >>central mailhub B then it will be passed along the encrypted channel.  > L >Ok, I understand. But as long as SPAM is sent to mailhub A the spammer willO >have to pay and there is no need for an additional fee. This will work as long I >as most e-mails crossing the net are still terminated with a <CR>.<CR> .  > N I don't see this with your scheme. International net company A is not going to7 detect such mail and hence won't charge anybody for it.     K >>How is central mailhub A supposed to know it is SPAM ? Please define SPAM B >>in an unambigouous manner which every recipient would agree on ? > O >Defining SPAM is a real problem. If it would be easy there would be no need to O >force fee-based e-mails. But if the sender or sending host of an e-mail has to L >pay for each message there will be no need any more for mailhub A to detect >SPAM. > 3 >>The way SMTP with TSL works is one of two ways :-  >  >Thanks for the explanation. > K >>In both these cases no intervening system will find any packet containing I >><CR>.<CR> which signals the end of the mail message - it is conceivable N >>that it might pickup encrypted packet(s) not meaning the end of the message 6 >>containing amongst other things the string <CR>.<CR> >  >[...] > M >>As I said you could get around this by looking for connections to port 465, F >>the STARTTLS command and unencrypted mail. But that is rather messy. > P >If we assume that most messages still have the <CR>.<CR> bytes there is time toM >develop a better technical solution for the e-mail fee. It looks as if it is  >still doable. > M You have a little while but support for TSL/SSL is growing rapidly - a number I of freeware MTA's such as EXIM support it as well as commercial MTAs and  K also Microsoft Exchange - though according to my PMDF manual their version  O is "different" hence PMDF has to provide an additional configuration keyword if N you want to support TSL communication to Microsoft Exchange systems as well as standards compliant systems.    
 David Webb VMS and Unix team leader CCSS Middlesex University    	 >Regards,  >   Christoph Gartmann >  >-- F > Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 > Immunbiologie J > Postfach 1169                 Internet: gartmann@immunbio dot mpg dot de > D-79011  Freiburg, Germany: >               http://www.immunbio.mpg.de/home/menue.html   ------------------------------  * Date: Tue, 7 Oct 2003 02:04:53 +0000 (UTC) From: david20@alpha2.mdx.ac.ukI Subject: Re: Fee Based Email (From Re: Process's PreciseMail AntiSpam...) ) Message-ID: <blt705$455$1@news.mdx.ac.uk>   o In article <blrhnj$8l3$2@n.ruf.uni-freiburg.de>, gartmann@non.immunbio.mpg.de.sens (Christoph Gartmann) writes: K >In article <blpg8h$onk$1@news.mdx.ac.uk>, david20@alpha2.mdx.ac.uk writes: N >>>The spam that I see has mostly around 10 to 20 envelope recipients, most of >>>them directed to our domain.  >>> Q >>You are confusing the To list in the mailbody with the envelope recipient list.  > I >No, I am very aware between the difference between a "To:"-header and an K >envelope-to-header. From time to time we analyze the SPAM that reaches us. H >And we found that the number of envelope headers is very limited. Ok, IK >cannot tell whether the messages came from the original spam host or via a M >relay. But in both cases the fee would be applicable. It would be reduced by K >a factor of 10 or 20 which is not what I like but it would still be a fee.  > N You are still only seeing the envelope recipients which were destined for your
 organisation.   N If I sent a message with 1000 envelope recipients and 999 of them were not forN your company and just one was then all you would see is one envelope recipientI address. What you record for spam reaching any of your systems will be an N unknowably small percentage of the size of the original envelope rcpt to list.      I >>NO. This happens with ALL modern mailhubs. The only difference with an  P >>open-relay is that it allows messages to come in from outside it's domain and L >>be sent out again. Whereas a normal mailhub should reject any unauthorised6 >>external mail not directed to people in it's domain.L >>In the example above mailhub A is not an open-relay the message to the 100= >>users has originated from one of that organisation's users.  > L >Then it is quite ok when mailhub A is charged. Perhaps it is then better toM >switch from the <CR>.<CR> approach to a "RCPT TO:" approach. This would mean O >that we have a slightly higher chance of charging for mails that have not been K >successfully delivered. But I think as it is only a very small fee, normal  >users wouldn't mind.  >   I TSL/SSL encryption will mean that in the longer term you won't be able to K charge on RCPT TO's. You might be able to pick up the initial setting up of 9 the encrypted channel but not any details of the message.   H Note. You may also need to deal with packets destined for port 587 - theO standard message submission port (I'm not sure if there is an encrypted version H of that as well). This is primarily meant for submission of "incomplete"N messages to a local mailhub which then completes the message before forwardingE it on - however it can be used and abused in the same way as port 25. ! For further details see RFC 2476.     N And if you did move to looking for the setting up of encrypted connections youH might also run into problems with the use of the ETRN command which someK systems (which are not permanently connected to the internet ) use to tell  J SMTP servers they have an agreement with to download queued mail to them. N The ETRN command switches the role of sender and receiver after the connectionL to port 25 has been made. Hence with an encrypted connection just seeing whoH made the initial connection is not enough to determine which system is  " the sender and which the receiver.    
 David Webb VMS and Unix team leader CCSS Middlesex University    	 >Regards,  >   Christoph Gartmann >w >-- F > Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 > Immunbiologie J > Postfach 1169                 Internet: gartmann@immunbio dot mpg dot de > D-79011  Freiburg, Germany: >               http://www.immunbio.mpg.de/home/menue.html   ------------------------------  $ Date: Mon, 6 Oct 2003 19:53:15 +0200( From: "H Vlems" <hvlems.nieuw@zonnet.nl>@ Subject: Re: Fixed device names for incoming Telnet connections?9 Message-ID: <blsa9t$fknd5$1@ID-143435.news.uni-berlin.de>F  A "Wilm Boerhout" <w.boerhoutOLD@PAINTplanet.nl> schreef in berichti$ news:bls51s$3cr$1@reader10.wxs.nl... > labadie wrote:C > > "Wilm Boerhout" <w.boerhoutOLD@PAINTplanet.nl> wrote in messageo( > > news:blk26u$75e$1@reader08.wxs.nl... > >uD > >>Way back when, I used to think up solutions for customers on VMSG > >>systems. Nowadays, I seem to be back in the field. For one of thesei& > >>solutions, the following question: > >>E > >>Can I tweak UCX to generate fixed TNAn: device names for incoming5K > >>connections based on, say, the IP address of the originating interface?a3 > >>I used to do this for incoming LAT connections.5 > >>J > >>If Yes to the above, can I have alternae controller names, in order toI > >>generate TNBn: and TNCn: device names? SYSGEN CONNECT TNB0 /WHATEVER, K > >>and of course specify the Whatever and anything else to tell the TelnetrC > >>server that for some IP addresses, TNBn: or TNCn: is to be used  > >> > >>Wilm Boerhoutu > >> > > 	 > > Hello2 > >rF > > I am not sure about what you want to achieve, but, you can do, for example- > >-" > > $ telnet/create alpha1 23 61214 > > this will create on node alpha1 a device tna6121 > >.; > > When you have ended, you should do something similar toD > > $ telnet/delete 6121  > > to delete the device tna6121 > >n > > Look atg > > $ telnet > > help createn > >c > > regardsn > >-
 > > Grard > >u > >iG > I'm trying to extend the life of an 20 yr old application whose userswG > get part of the functionality based on terminal names (TXA3:) that is"H > looked up in a table. Table entries may be changed, but are restricted > to 4 characters. >iF > Users logging in freely via Telnet, would eventually wind up gettingA > TNA10: as login device, and the application method would break.0 >IE > So, creating the TNAn: is possible (provided n <> 0 I presume). How-K > could I force an incoming connection to attach to this particular device?c > J > And, I need more than 9 devices. Is the assumption that I can create and& > attach to TNBn: worth investigating? >rC > I am also investigating other scenarios, involving logical names. J > Assuming that physical device TXB$: is not present in the system, DEFINE > TXA4: as the login device. >  > Thnks, > Wilm >  > w.boerhoutOLD@PAINTplanet.nl* > <remove OLD PAINT from address to reply> >GK In that case, would it help to define a logical name that lets TT: point to ' TXA4: before you start the application?    ------------------------------  # Date: Mon, 06 Oct 2003 17:13:35 GMTt6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) Subject: MD5 source code ?4 Message-ID: <3phgb.81032$qU6.1291982@news.chello.at>  J Does anyone have a source code ready which does a MD5 checksum of an input1 file and puts the hash/checksum in a DCL symbol ?a  M I know I could do it myself but I'm lazy^Wshort of time and ask here first...    TIAo   -- s Peter "EPLAN" LANGSTOEGERU% Network and OpenVMS system specialisti E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  # Date: Mon, 06 Oct 2003 20:51:13 GMT # From: hoff@hp.nospam (Hoff Hoffman)c Subject: Re: MD5 source code ?1 Message-ID: <5Bkgb.6432$YA.1776@news.cpqcorp.net>a  m In article <3phgb.81032$qU6.1291982@news.chello.at>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes: K :Does anyone have a source code ready which does a MD5 checksum of an input)2 :file and puts the hash/checksum in a DCL symbol ? :yN :I know I could do it myself but I'm lazy^Wshort of time and ask here first...  I   The MD5 code is around -- for instance, there are pointers to pre-builtrG   MD5 executables referenced in the OpenVMS Frequently Asked Questions t@   (FAQ), and there is a link from there to the RFC and the code.  K   IIRC, please also realize that MD5 does expose a known bug in the OpenVMSnJ   VAX C compiler's optimizer.  (If this is for use on OpenVMS VAX, compile1   the C code /NOOPT.   Or use one of the images.)-    N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqrN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  # Date: Mon, 06 Oct 2003 20:54:57 GMT / From: Milton <mbhewitt@optonline.SPAMBLOCK.net>o Subject: Re: MD5 source code ?8 Message-ID: <drk3ov814jqfbrs9qtp7b2oqg7l4cvfq0k@4ax.com>  E On Mon, 06 Oct 2003 17:13:35 GMT, peter@langstoeger.at (Peter 'EPLAN'w LANGSTOEGER) wrote:   K >Does anyone have a source code ready which does a MD5 checksum of an input 2 >file and puts the hash/checksum in a DCL symbol ? >eN >I know I could do it myself but I'm lazy^Wshort of time and ask here first...  G The undocumented DCL command CHECKSUM is the usual means,and provides alH rather simple-minded checksum suitable to detect basic file corruptions.  1 CHECKSUM filename $ SHOW SYMBOL CHECKSUM$CHECKSUMs  F For information about an OpenVMS version of the MD5checksum tool, see:> http://www.support.compaq.com/svctools/md5-instructions.html     Cheers,  Milton   ------------------------------  % Date: Mon, 06 Oct 2003 23:39:19 +0200r- From: Marc Van Dyck <marc.vandyck@brutele.be> - Subject: Re: Missing delta time functionalityi9 Message-ID: <mesnews.358b7d3a.c64a4a7d.81.584@brutele.be>d   "H Vlems" a pens trs fort :t> > "Marc Van Dyck" <marc.vandyck@brutele.be> schreef in bericht5 > news:mesnews.1c4f7d3a.2b8d5487.70.584@brutele.be...  >> "H Vlems" avait prtendu :b5 >>> "Guy Peleg" <guy.peleg@hp.com> schreef in bericht $ >>> news:3F7D996D.31EB0B51@hp.com...L >>>> In the technical update day in the Netherlands it was mentioned that it+ >>>> is hard to do delta time calculations. F >>>> In VMS V7.3-2 we added new lexical function F$DELTA_TIME. See the >>>> following example:c >>>> c >>>> BLUSKY> a=f$time()  >>>> BLUSKY> b=f$time()f >>>> BLUSKY> sh sym at$ >>>>   A = " 3-OCT-2003 17:39:58.54" >>>> BLUSKY> sh sym b $ >>>>   B = " 3-OCT-2003 17:40:23.22"* >>>> BLUSKY> write sys$output f$delta(a,b) >>>>    0 00:00:24.68u >>>> tK >>>> Considering F$DELTA and F$CVTIME what other functionality do you think + >>>> is missing for delta time calcualtion?rL >>>> We are currently planning the new features for next VMS version so here) >>>> is your chance to affect the future., >>>> l( >>>> Responde here or just send me mail. >>>> 3
 >>>> Regards,  >>>> Y >>>> Guy Peleg >>>> OpenVMS Engineering >>>>   >>> Guyr >>> L >>> at that particular meeting it was suggested that adding delta times to a" >>> "normal" time would be useful: >>>  >>> $ a=f$time() >>> $ b=f$time() >>> $ dt=f$delta(a,b)o >>> $ ft=f$delta_add(a,dt) >>> O >>> The result stored in ft could then be used to perform tasks at exact times.y >>>  >>> Hans >> n, >> Can't you do that with f$cvtime already ? >> D >> Marc. >3M > Not that I'm aware of. Now I'm also kind of lazy so it may be possible witha' > CVTIME if one really would want that.e  = I don't have my doc at hand but I think you can give F$CVTIMEM@ an argument which is a character string containing a combination? absolute + or - delta. There is just the difficulty of buildinge> the character string argument with the exact syntact, which is rather strict IIRC.e   Marc.      -- a
 Marc Van Dycko   ------------------------------  # Date: Tue, 07 Oct 2003 03:13:37 GMTk6 From: "Craig A. Berry" <craigberry@mac.com.spamfooler> Subject: Re: ODBTP on VMS ?s@ Message-ID: <0f3a6bcdeac34267db40be6962b5d9a8@news.teranews.com>  7 In article <6X3gb.924$Ys.102781@news20.bellglobal.com>,a   "SP" <nospam@nospam.net> wrote:  L >     Has anybody tried using sourceforge.net ODBTP on VMS and would like to= > share their experience ? It sure looks like a nice package.d > G >     It seems to be an TCP ODBC package to access ODBC database on Win- > platforms.  @ No, I am unfamiliar with ODBTP.  However, the slightly misnamed D unixODBC (http://www.unixodbc.org) works fine on VMS, and there are ( several commercial alternatives as well.   ------------------------------  % Date: Mon, 06 Oct 2003 21:13:34 -0400 ( From: Galen Tackett <gtackett@yahoo.com>Y Subject: Problem with CLUSTER_CONFIG? (was: Re: DIFVOLMNT (%X0072832C), then bugcheck whe F Message-ID: <gtackett-C90FD5.21132806102003@news.comcast.giganews.com>  I I now understand how this problem occurred. Pending further information, tB I'd say it appears to be a weakness in the CLUSTER_CONFIG process.  G I reran CLUSTER_CONFIG up until the point were you'd normally boot the uG the new system , which in this case is to be a voting boot server (see l= my original post below for a few more config details). While lC CLUSTER_CONFIG waited for the new node to boot, I examined the new n9 node's parameters in its system-specific ALPHAVMSSYS.PAR.?  F It turns out that the CLUSTER_CONFIG process had set EXPECTED_VOTES=1.  G Now, recall that the new system at first had a bad Ethernet connection cH (and it has no other alternative path for cluster comms). I examined my 9 boot logs from that original attempt to add the new node.r  G Sure enough, when the new noted booted (to run its initial AUTOGEN) it hH saw no other cluster nodes, and it formed a VAXcluster all by itself. I F assume that at this point it corrupted the boot volume's SCB (Storage @ Control Block, the first block in BITMAP.SYS), which led to the @ DIFVOLMNT when we next tried to reboot an existing cluster node.  H SO: Is it reasonable to have CLUSTER_CONFIG set up a new voting node to I initially use EXPECTED_VOTES of 1? Doing so means that there's a risk of PC corrupting the system disk if the new node needs a LAN to boot but   doesn't have a good connection.4  E And since CLUSTER_CONFIG is executing on a cluster with at least one oH vote already present, couldn't CLUSTER_CONFIG.COM set EXPECTED_VOTES to G at least 2? Or perhaps it could look at the other nodes in the cluster t to get their EXPECTED_VOTES?  I Perhaps there's some rationale for having EXPECTED_VOTES of 1, but is it oG worth the possibility of corrupting the system disk in a scenario like rB this? If so, perhaps Hoff or one of the other experts can explain.  C (Sure, you could say that we should have tested the LAN connection rH beforehand, and you're probably right, but wouldn't a safety measure be  desirable here?)  G Fortunately the damage was confined to several log files on the system a0 disk, and we didn't need to restore from backup.  = In article <bdc65a53.0310030347.33bd7031@posting.google.com>,u&  gspamtackett@yahoo.com (Galen) wrote:  H > We've gotten into this situation with our cluster twice recently. (I'mD > not referring to a VOLALRMNT error, which is a different numerical
 > status.) >  > Configuration is:o > D > A single OpenVMS Alpha V7.3-1 system disk which is very current on
 > patches.? > System disk lives on an HSG80, reached via a SAN core switch.mF > Satellites do not have any shared-storage connections (i.e. no DSSI,# > no FibreChannel, no shared SCSI).>/ > 13 boot servers and 7 satellites, all Alphas. = > Running Storageworks RAID software (not sure how relevant).t > G > In both cases, we had recently run CLUSTER_CONFIG to add a new servers? > node. However, in each case, the new node had no physical LAN F > connection (fiber not hooked up) and took a CLUEXIT bugcheck after a > few minutes. > H > Each time, shortly after the CLUEXIT, we got the node's LAN connectionH > working and re-ran CLUSTER_CONFIG. Just after the new node reached the? > point where it reports there's no pagefile on the system disk % > (%SYSINIT-I-PAGEFILE), it reported:h > < > %SYSINIT-E-Error mounting system device, status = 0072832C >  > We checked these things: > B > * No other clusters with same cluster ID (we only have one other
 > cluster)) > * All systems have VAXCLUSTER set to 2.rF > * The volume label on the system disk has not been changed since the > cluster was last booted. > H > The only solution we've found is to reboot the cluster (not a pleasant > option, of course).@ > H > But we're just as concerned to find out what's causing this. I suspectE > that the CLUEXIT during CLUSTER_CONFIG somehow is involved but haveh; > only a little circumstantial evidence, as described here.t > H > HP software support and the maintainer of the MOUNT code have given usF > a little script to periodically check the volume's SCB and report ifD > its checksum changes. Beyond that, they're out of ideas right now. > G > (FYI, the bad connections occur because our fiber cable plant is very H > badly documented, has a lot of old labels, and some of the fibers haveB > been damaged at one time or another. But this is another issue.) > % > Thanks for any help or suggestions,  >  > Galene   --  
 Galen Tackettn  5 *** To e-mail me just remove the spam from my addressm   ------------------------------  # Date: Tue, 07 Oct 2003 01:02:58 GMTh' From: Don Sykes <anonymous@pacbell.net>CM Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?n+ Message-ID: <3F821157.D2CE295D@pacbell.net>s   <snip>  V > >> >> The receiver is in compete control. If it doesn't like something it sends back1 > >> >> a 5xx response and closes the connection.  > >> >E > >> >The receiver can set aside accepting an email until a time moremD > >> >convenient to them. They're not required to accept mail at theM > >> >convinience of the sender. This allows options that are impossible in aaM > >> >one phase protocol, like retrieving large or low priority emails duringlH > >> >slow times and sending a pre-email notification to the user - e.g.N > >> >       I have a request from smutpeddler@legitdomain.com, do you want toH > >> >accept a 2MB avi file? do you want to charge him? If so, how much?D > >> >that sort of thing. A one phase protocol requires a continuous9 > >> >interaction until the mail is rejected or accepted.t > >> > > >>O > >> You are at least one level away from the sender or receiver at the centralaN > >> mailhub level. Your default policy is you are going to delay all mail not' > >> explicitly requested by the user ?a > >e9 > >Again. That's up to what the domain (ESP) wants to do.s > > T > >> My mail users complain that mail is not instantaneous. They expect it to all beR > >> instantaneous. Adding extra delays is not acceptable. Even if a simple methodR > >> is supplied for them to update what they want to delay on the central mailhubN > >> they won't immediately update it after they get off the phone to that newR > >> contact who is going to mail them - that is if they remember to get his email1 > >> address rather than just telling him theirs.t > >aH > >In your situation, you would probably wouldn't delay based on size orD > >priority. Adding the 2nd phase could take only an extra couple of > >seconds.s > ? > You explicitly stated that the second connection was to allow  >  > " A > The receiver can set aside accepting an email until a time more @ > convenient to them. They're not required to accept mail at theI > convinience of the sender. This allows options that are impossible in aeI > one phase protocol, like retrieving large or low priority emails duringu= > slow times and sending a pre-email notification to the user  > "v >   > Thats not a few seconds delay.  E No it isn't. But the key phrase here is "The receiver CAN". He is notu required to.; This is why I keep saying that implementations will differ."   > N > (As to low priority messages - mailhubs already have facilities to deal withM > that. Mail messages can be sent with "urgent","normal" or "low" priorities.oK > The mailhub's queuing system can be setup to prioritise dealing with sucheO > messages. Suprisingly when this is implemented and starts to slow down normalaB > mail people have a tendency to send all their mail as "urgent").  F Yes, I know we can all mark our emails with "urgent","normal" or "low"G today, but it's pretty meaningless. In my model, the priority would noteH determined by the sender marking as urgent, but by the receiver based on? the phase 1 data he received. Perhaps some implementations will D prioritize based on the number of emails from a given sender, or theF number sent from a given domain. But this is also something that couldF be added to the phase 1 meta data - i.e. priority, which could then be? used to determine cost. In that way those priorities would meansD something - i.e are you willing to pay more for urgent processing...# I'll have to consider this further.s   > J > I would think that my "situation" is pretty much the situation of anyoneL > looking after a central mailhub. I'm not in a high pressure organisation - > just a University. > K > >As for your new customer who just got off the phone with a new customer, K > >he would have told that new customer, "be sure to start the Subject withnI > >'NEW123'", a code which he previously set up for all new customers, soi< > >the receiver ESP will pass it on without charge or delay. > >t > Q > Well if you tell all your friends, colleagues and other mail contacts that thennO > you don't need any new protocol to stop spam - a very simple mail filter will $ > ensure you never receive any spam.  G That would insure I never SAW the spam, but my SMTP service would still C be required to process it all the way to my Netscape filter and then/ bandwidth I pay for will still be impacted too.n   > P > >> Anyone running a central mailhub will immediately turn off any such delays. > >> > >> >>e > >> >> U > >> >> >> At the moment there does exist a small possiblity of spoofing however yoursC > >> >> >> system is exactly as vulnerable as the current standards 	 > >> >> >iQ > >> >> >I don't agree, because in phase 2 the receiver initiates the connection. Q > >> >> >So unless the spoofer can control the DNS I don't see how they will everd% > >> >> >get their message delivered.eP > >> >> >And if they can control the DNS, they can jolly well force all of us to7 > >> >> >a smut site when we try to http to Google.com.c > >> >>oQ > >> >> Yes as I say a small chance of spoofing which is exactly the same for thedE > >> >> current protocol since it can do exactly the same DNS lookup.9U > >> >> Since this is at the central mailhub rather than the desktop sender level thehV > >> >> chances of anyone wanting to spoof this are pretty remote anyway. Any spoofingI > >> >> would be done between the desktop system and the central mailhub.V > >> >K > >> >Give me an example of what you mean. I don't see it. When the desktopaN > >> >user goes to retrieve his mail he logs onto a POP or IMAP server, passesM > >> >his userid & password and downloads the mails that are waiting for him. % > >> >Where would the spoofing occur?e > >> > > >>Q > >> I'm talking about in sending the mail from the desktop to the central server0% > >> in order to avoid being charged.L > ><C > >You mean desktop user 'A' will send an email to the sending ESP,wF > >pretending to be desktop user 'B'? Why don't you require a userid &0 > >password? You must have the same problem now. > >< > >> > I > There is no charging so there is no incentive to go to great lengths toi > avoid being charged. > I > Of course we use userids and passwords to control access to our desktoplK > systems. What has that got to do with someone being able to spoof so thatpD > mail or other connections appears to come from a different system.M > Our logs will show them logged onto machine A. If they then spoof things sooJ > that other machines think they were talking to machine B then it will beQ > extremely difficult if not impossible to prove that a complaint about machine B / > was caused by the user logged into machine A.e  > You mean after they logged onto your mail server as machine A?   > K > We are spending an awful lot of money to improve our security logging but 0 > I have no illusions that it will be foolproof. >  > >> >> T > >> >> As to controlling the DNS and redirecting to smut sites - yes it happens allS > >> >> the time. Though the preference is usually to redirect to their own versiontM > >> >> of a companies site so they can gather credit card and other details.m > >> >>-	 > >> >> >-! > >> >> >> - this can be improvedcR > >> >> >> in the future by use of server certificates and SSL/TSL to provide more- > >> >> >> trustworthy mutual authentication.$
 > >> >> >>V > >> >> >> The problems of identity are to do with tracking within organisations, withP > >> >> >> non-compliant and badly configured mail systems and with open relays.S > >> >> >> These are NOT protocol issues.  Nobody should be running an open-relay -p= > >> >> >> there is even an RFC which states this - RFC 2505.l	 > >> >> >hQ > >> >> >I don't run an open relay and I can't stop the 100's of spams per day or H > >> >> >force anyone to pay if I deliver a spam message to an end user.	 > >> >> >. > >> >> V > >> >> But apart from the fee paying idea which won't work unless you really know whoT > >> >> to charge the fee to your protocol doesn't provide anything extra which will > >> >> control spammers.: > >> >5 > >> >Apart from that, Mrs Lincoln, how was the play?u& > >> >The fee IS the pertenient point. > >> > > G > The fee is only pertinent if you have full end to end accountability.p  G Mail server to mail server accountability, yes, but not necessarily endPF user to end user. Let the ESP take the responsibility of insuring mailC gets to/from his own users. Today anyone can be an ESP and dole oute? email address like candy, yet take no responsibility when thoseh$ addresses are used for spam sending.   >  > >> >>lP > >> >> >I agree that identity problems WITHIN an organisation are not solved byM > >> >> >this, but that' by design. IMO all those problems are implementation M > >> >> >issues, ones, I might add, that are just as likely to cause problemsc > >> >> >using any protocol.	 > >> >> >  > >> >>tW > >> >> What I am saying is your protocol is irrelevent. Communications between centralnT > >> >> mailhubs are already adequately controlled. It is the injection of spam intoQ > >> >> the sending central mailhubs and direct sending of mail bypassing centralO( > >> >> mailhubs which need controlling. > >> >; > >> >How can you bypass the receiving ESP? Example please.b > >>N > >> I'm talking abiout the current setup. Not everybody has central mailhubs,T > >> not everybody has firewall rules blocking direct sending from desktop machines. > > J > >Everybody clearly has a receiving ESP. Bypassing the sending domain ESPH > >wouldn't help you in the new protocol. At best you could pass a bogusI > >phase 1, but when the receiving mail's ESP went to the sending domain,  > >the mail would exist. > >" > O > Not every organisation has a single central mailhub. Some organisations allowLE > mail to be directly addressed to multiple systems in their domain. s  D The key phrase here being "in their domain". I'm not trying to solve= those problems. Granted, I'm sure those problems are real andNF frustrating, but IMO we must first solve the problem of unwanted email from other domains.-   > Each such-J > system may be a multiuser system or a single user's unix workstation etc > L > But even if both organisations have a central mailhub. Then a spammer justN > needs to get the mail onto his own organisation's central mailhub without itN > being traced back to his real login. The message will then be passed betweenL > the two central mailhubs in exactly the same manner whether using your newP > protocol or smtp. In either case you would have no trouble in tracing the mail3 > back to the first organisation's central mailhub.nN > In neither case does SMTP or your new protocol provide any means of tracking > down the real sender.e  G I agree, but as I said I think we must fix the problem with emails from  "out-worlders".n   > L > Now you may think that by implementing fees you can force the owner of theO > sending central mailhub to improve their security to such an extent that theyO- > can track down who really sent the message.I  6 I guarantee that if money is involved they will do so.  E > I doubt it could be made to work since the charging would be acrosscO > international boundaries. However if you wanted to do it I'd say extend ESMTP 8 > to include charging rather than invent a new protocol. >   H Noted, but I don't think the international boundaries issue is really anG issue. I buy things from all over the world using my VISA card and thatp works out quite well.    <snip>     --     Have VMS, Will Traveln Wire paladin, San Francisco    (paladinATalphaseDOTcom)   ------------------------------  * Date: Tue, 7 Oct 2003 03:03:36 +0000 (UTC) From: david20@alpha2.mdx.ac.ukM Subject: Re: Process's PreciseMail AntiSpam Gateway - any experience so far ?r) Message-ID: <bltae8$4ht$1@news.mdx.ac.uk>   U In article <3F821157.D2CE295D@pacbell.net>, Don Sykes <anonymous@pacbell.net> writes:l >a >t ><snip>e >> sL >> >As for your new customer who just got off the phone with a new customer,L >> >he would have told that new customer, "be sure to start the Subject withJ >> >'NEW123'", a code which he previously set up for all new customers, so= >> >the receiver ESP will pass it on without charge or delay.- >> > >> uR >> Well if you tell all your friends, colleagues and other mail contacts that thenP >> you don't need any new protocol to stop spam - a very simple mail filter will% >> ensure you never receive any spam.f >8H >That would insure I never SAW the spam, but my SMTP service would stillD >be required to process it all the way to my Netscape filter and the0 >bandwidth I pay for will still be impacted too. >eK Whatever mechanism you use to tell your ESP about that subject prefix couldp7 also be used with current SMTP based central mailhubs.  N True the message data would still need to be accepted and only rejected at theM end according to the RFCs but there is nothing to stop the receiving mailhub  L directing it to null as it comes in once it has determined the subject line.J Most mailhubs can't do this but some such as EXIM can if provided with the right filter program.n         >> gQ >> >> Anyone running a central mailhub will immediately turn off any such delays.o >> >>d >> >> >> >> >> >>V >> >> >> >> At the moment there does exist a small possiblity of spoofing however yourD >> >> >> >> system is exactly as vulnerable as the current standards
 >> >> >> >R >> >> >> >I don't agree, because in phase 2 the receiver initiates the connection.R >> >> >> >So unless the spoofer can control the DNS I don't see how they will ever& >> >> >> >get their message delivered.Q >> >> >> >And if they can control the DNS, they can jolly well force all of us to 8 >> >> >> >a smut site when we try to http to Google.com. >> >> >>R >> >> >> Yes as I say a small chance of spoofing which is exactly the same for theF >> >> >> current protocol since it can do exactly the same DNS lookup.V >> >> >> Since this is at the central mailhub rather than the desktop sender level theW >> >> >> chances of anyone wanting to spoof this are pretty remote anyway. Any spoofingsJ >> >> >> would be done between the desktop system and the central mailhub. >> >> >eL >> >> >Give me an example of what you mean. I don't see it. When the desktopO >> >> >user goes to retrieve his mail he logs onto a POP or IMAP server, passes N >> >> >his userid & password and downloads the mails that are waiting for him.& >> >> >Where would the spoofing occur? >> >> >v >> >>CR >> >> I'm talking about in sending the mail from the desktop to the central server& >> >> in order to avoid being charged. >> >D >> >You mean desktop user 'A' will send an email to the sending ESP,G >> >pretending to be desktop user 'B'? Why don't you require a userid &a1 >> >password? You must have the same problem now.r >> > >> >>D >> uJ >> There is no charging so there is no incentive to go to great lengths to >> avoid being charged.  >> aJ >> Of course we use userids and passwords to control access to our desktopL >> systems. What has that got to do with someone being able to spoof so thatE >> mail or other connections appears to come from a different system. N >> Our logs will show them logged onto machine A. If they then spoof things soK >> that other machines think they were talking to machine B then it will beoR >> extremely difficult if not impossible to prove that a complaint about machine B0 >> was caused by the user logged into machine A. > ? >You mean after they logged onto your mail server as machine A?. >   9 What do you mean by logon to a mail server as a machine ?AM In the scenario I described above the user is logged onto a desktop machine A8F They spoof the IP address so packets appear to come from machine B andL obviously also spoof the mail from address. To the mailhub the mail message $ appears to have come from machine B.       >> mL >> We are spending an awful lot of money to improve our security logging but1 >> I have no illusions that it will be foolproof.e >> b >> >> >>U >> >> >> As to controlling the DNS and redirecting to smut sites - yes it happens all T >> >> >> the time. Though the preference is usually to redirect to their own versionN >> >> >> of a companies site so they can gather credit card and other details. >> >> >>
 >> >> >> >" >> >> >> >> - this can be improvedS >> >> >> >> in the future by use of server certificates and SSL/TSL to provide more . >> >> >> >> trustworthy mutual authentication. >> >> >> >> W >> >> >> >> The problems of identity are to do with tracking within organisations, withtQ >> >> >> >> non-compliant and badly configured mail systems and with open relays.eT >> >> >> >> These are NOT protocol issues.  Nobody should be running an open-relay -> >> >> >> >> there is even an RFC which states this - RFC 2505.
 >> >> >> >R >> >> >> >I don't run an open relay and I can't stop the 100's of spams per day orI >> >> >> >force anyone to pay if I deliver a spam message to an end user.n
 >> >> >> > >> >> >>W >> >> >> But apart from the fee paying idea which won't work unless you really know whodU >> >> >> to charge the fee to your protocol doesn't provide anything extra which willd >> >> >> control spammers. >> >> >p6 >> >> >Apart from that, Mrs Lincoln, how was the play?' >> >> >The fee IS the pertenient point.  >> >> >a >> gH >> The fee is only pertinent if you have full end to end accountability. > H >Mail server to mail server accountability, yes, but not necessarily endG >user to end user. Let the ESP take the responsibility of insuring mail D >gets to/from his own users. Today anyone can be an ESP and dole out@ >email address like candy, yet take no responsibility when those% >addresses are used for spam sending.S >S >> r >> >> >>Q >> >> >> >I agree that identity problems WITHIN an organisation are not solved bytN >> >> >> >this, but that' by design. IMO all those problems are implementationN >> >> >> >issues, ones, I might add, that are just as likely to cause problems >> >> >> >using any protocol.d
 >> >> >> > >> >> >>X >> >> >> What I am saying is your protocol is irrelevent. Communications between centralU >> >> >> mailhubs are already adequately controlled. It is the injection of spam intoeR >> >> >> the sending central mailhubs and direct sending of mail bypassing central) >> >> >> mailhubs which need controlling.  >> >> > < >> >> >How can you bypass the receiving ESP? Example please. >> >>eO >> >> I'm talking abiout the current setup. Not everybody has central mailhubs,oU >> >> not everybody has firewall rules blocking direct sending from desktop machines.s >> >K >> >Everybody clearly has a receiving ESP. Bypassing the sending domain ESPaI >> >wouldn't help you in the new protocol. At best you could pass a bogus5J >> >phase 1, but when the receiving mail's ESP went to the sending domain, >> >the mail would exist.e >> > >> tP >> Not every organisation has a single central mailhub. Some organisations allowF >> mail to be directly addressed to multiple systems in their domain.  >tE >The key phrase here being "in their domain". I'm not trying to solve6> >those problems. Granted, I'm sure those problems are real andG >frustrating, but IMO we must first solve the problem of unwanted email, >from other domains. >  This is in response to your )  "Everybody clearly has a receiving ESP."-H I was just pointing out that everybody does not clearly have an existingL central mailhub and not everybody enforces all incoming mail to pass through/ such a central mailhub even if they do have it.r       >> Each suchK >> system may be a multiuser system or a single user's unix workstation etc@ >> rM >> But even if both organisations have a central mailhub. Then a spammer justnO >> needs to get the mail onto his own organisation's central mailhub without itkO >> being traced back to his real login. The message will then be passed betweenIM >> the two central mailhubs in exactly the same manner whether using your newqQ >> protocol or smtp. In either case you would have no trouble in tracing the mailn4 >> back to the first organisation's central mailhub.O >> In neither case does SMTP or your new protocol provide any means of tracking  >> down the real sender. >,H >I agree, but as I said I think we must fix the problem with emails from >"out-worlders". >S  + Sorry I don't understand your meaning here.sL In the paragraph above the receiver has received a spam message from another
 organisation.r       >> :M >> Now you may think that by implementing fees you can force the owner of therP >> sending central mailhub to improve their security to such an extent that they. >> can track down who really sent the message. >s7 >I guarantee that if money is involved they will do so.h >yF >> I doubt it could be made to work since the charging would be acrossP >> international boundaries. However if you wanted to do it I'd say extend ESMTP9 >> to include charging rather than invent a new protocol.  >>   >iI >Noted, but I don't think the international boundaries issue is really anoH >issue. I buy things from all over the world using my VISA card and that >works out quite well. >aL With VISA you have one party acting as an intermediary between yourself and I the stores in different countries. Visa has negotiated with each of thoseeO stores so that they allow purchasing using the Visa card. You can use Visa in abK lot of places - but you can't use it everywhere. With Visa the contract to tN purchase is made directly between buyer and seller. You agree to the price the vendor is asking.            e  L With your system you do not have a single entity like Visa instead you have ? bilateral agreements between pairs of intermediaries your ESPs.dK Thus you have 4 different parties involved in any charging. The sender, his.L ESP, the receiver and his ESP. There is also no direct contract between the  senders and receivers.I Also negotiations on costs have taken place between the intermediate ESPsi- rather than between the sender and receiver.  L This assumes a message to a single recipient. The number of parties involvedN explodes if the message is sent to multiple recipients who use different ESPs.  N Throw in differing national laws and I think the lawyers will be the only ones going home happy.d    
 David Webb VMS and Unix team leader CCSS Middlesex University       ><snip>  >n >I >--  >  >Have VMS, Will Travel >Wire paladin, San Francisco >. >(paladinATalphaseDOTcom)E   ------------------------------  # Date: Tue, 07 Oct 2003 01:49:16 GMT@+ From: lewis@e.thundermaker.net (Spud Demon) ' Subject: Re: Question about SMTP.CONFIG 1 Message-ID: <wYogb.412$gG3.47552@news.uswest.net>c  { JF Mezei <jfmezei.spamnot@istop.com> writes in article <3F80AE23.749DFDCC@istop.com> dated Sun, 05 Oct 2003 19:50:17 -0400:sK >Correct. I also have the tcpip$smtp_local_aliases.txt populated with locallG >host names. This allows those hosts to relay. (but not sure if this iso >included in Relay-zones:)  	 Thanks!     # >> Any other anti-spam suggestions?h  O >Read the TCPIP Services management manual. It has full description of the SMTPi# >configuration rules for anti-spam.   A I did read it before I asked.  I just wasn't sure if I understoodDL everything.  This seems to be working reasonably well as is, I don't think IL need to use RBLs, which seems to be the only anti-spam feature which doesn't3 come on by default once you create the config file.n  @ Now if I only knew why my Usenet propagation was so piss-poor...  . --Spud Demon		spud_demon -at- thundermaker.net   ------------------------------  $ Date: Mon, 6 Oct 2003 19:49:27 +0200( From: "H Vlems" <hvlems.nieuw@zonnet.nl>! Subject: Re: SHOW DEVICE/FULL DSAw9 Message-ID: <blsa2p$fq12j$1@ID-143435.news.uni-berlin.de>   F "Michael Moroney" <moroney@world.std.spaamtrap.com> schreef in bericht" news:blq6pt$c2d$1@pcls4.std.com..., > "H Vlems" <hvlems.nieuw@zonnet.nl> writes: > H > >> No, the requirement for identical drives was dropped long ago.  TheG > >> current requirement is that device MAXBLOCKS must be the same, butb% > >> even this will change real soon.t >0 > >On AXP/VMS 7.3: >a > >$ ini/sys dka100 user8w' > >$ mou/sys dsa2:/shad=$8$dka100 user8v, > >%MOUNT-I-MOUNTED, USER8 mounted on _DSA2:G > >%MOUNT-I-SHDWMEMFAIL, _$8$DKA100: (ERBIUM) failed as a member of theI shadow > >set6 > >-SYSTEM-F-INCSHAMEM, incompatible shadow set memberF > >%MOUNT-I-ISAMBR, _$8$DKA400: (ERBIUM) is a member of the shadow set > >$8 > >$ pipe sho dev ds/full | sear sys$pipe "total blocks"< > >    Total blocks            17774160    Sectors per track > >32U< > >    Total blocks             4110480    Sectors per track > >86l< > >    Total blocks             2055035    Sectors per track > >108 > >$L > >The devices listed in the PIPE command are DSA0, DSA1 and DSA2. $8$DKA100 is > >a 9 GB disk.a >V >DG > You don't show what the block count of $8$DKA100 so I can't tell whatlE > shadowing is complaining about, but I was part of the shadow driverVJ > rewrite project in VMS Engineering several years ago and the requirementL > for identical drives was already gone at that time.  (I regularly shadowedI > RF73 and RZ73 disks.  Entirely different interfaces (DSSI and SCSI) butiF > same HDA under it all and thus same MAXBLOCKS.  Others shadowed real > drives and memory "disks") >nI > Remember, currently MAXBLOCKS must be _exactly_ the same, thus disks of7H > the "same" size (but not exactly the same) such as RZ28M's and RZ28C's > might not work together. > -- t > -Mike>  G That is right, somehow I was under the impression that you were writing3K about dissimilar device support for volume shadowing. Probably because AndyBI Goldstein mentioned it in his presentions during the VMS technical updateCK seminar. The example I posted had no value, the disks used were dissimilar.i   Hans   ------------------------------  # Date: Mon, 06 Oct 2003 16:43:11 GMTa9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>a Subject: Re: Sun takes a hit1 Message-ID: <zYggb.6382$Da.1900@news.cpqcorp.net>>  7 "Paul Sture" <nospam@sture.homeip.net> wrote in messageg* news:3F7FF321.64F09CCE@sture.homeip.net... > Keith Parris wrote:g > > : > > young_r@encompasserve.org (Rob Young) wrote in message/ news:<POztxxIPltF8@eisner.encompasserve.org>...>: > > >       It isn't starting to set, it has been setting. > >e? > > TheStreet.com says It's been a very, very bad week for Sun.m@ > > http://www.thestreet.com/_yahoo/tech/kcswanson/10117341.html > >?4 > > Motley Fool says it's been Sun's Nightmare Week. > > L http://www.fool.com/news/mft/2003/mft03100317.htm?source=eptyholnk303100&log  visit=y&npu=y&bounce=y&bounce2=y > > C > > "On Monday, the computer maker issued an earnings warning whilenI > > tacking on a $1.05 billion charge to last quarter. Yesterday, MerrilluD > > Lynch analyst Steve Milunovich sent a hostile open letter to theD > > company's board stating that 'Sun faces a crisis.' And now this:C > > Friday morning, Hewlett Packard announced that it would pay SunM@ > > customers $25,000 to switch to HP computers running Linux OS > > freeware." > B > But this is something I don't get. Why is HP pushing Linux as anH > alternative to Solaris rather than HP-UX, or even, heaven forbid, VMS? >oA > Why not give those away and go for the software maintenance andD > consulting revenue?k >o
 > ???????? > 4 > It all sounds like dot com boom economics to me... > I > Come on Keith, we know you aren't daft. Please explain the economics oft > this one.   I I don't know the specifics of this, but it would make sense to me to push/H Linux or HP-UX as an alternative to Solaris in general - as they are allJ more-or-less UNIX.  Trying to push VMS onto a Solaris user is likely to be. as popular as pushing Solaris onto a VMS user.  J Given that.  I would guess that the license for VMS is less than the $25k.E So even if VMS were not in the offer, take it and buy the VMS license0
 ala-carte.   ------------------------------  % Date: Mon, 06 Oct 2003 09:57:40 -0700t/ From: Greg Cagle <news@removethisgregcagle.com>a Subject: Re: Sun takes a hit/ Message-ID: <vo37oap4po9l87@corp.supernews.com>n   Fred Kleinsorge wrote:  9 > "Paul Sture" <nospam@sture.homeip.net> wrote in message , > news:3F7FF321.64F09CCE@sture.homeip.net...  B >>But this is something I don't get. Why is HP pushing Linux as anH >>alternative to Solaris rather than HP-UX, or even, heaven forbid, VMS? >>A >>Why not give those away and go for the software maintenance andh >>consulting revenue?. >>
 >>???????? >>4 >>It all sounds like dot com boom economics to me... >>I >>Come on Keith, we know you aren't daft. Please explain the economics ofo >>this one.t >  > K > I don't know the specifics of this, but it would make sense to me to push(J > Linux or HP-UX as an alternative to Solaris in general - as they are allL > more-or-less UNIX.  Trying to push VMS onto a Solaris user is likely to be0 > as popular as pushing Solaris onto a VMS user.  B It's actually very simple. You flush your $$$ Sun servers down the> toilet and replace them with (perceived) cheap Linux machines.B HP-UX machines are not as cheap, and the migration from Solaris to? Linux is less arduous than that from Solaris to HP-UX. Besides,i? there's always storage and other stuff to think about; many Suns? installations are set up with direct attached disks that can bet4 replaced with SAN or NAS boxes as part of the shift.  C Also, the migration path depends on the market segment. Some placesrE Linux doesn't make sense. You can't today replace a UE10K with a rackrF of Proliants running Linux in any sensible way, for example. And LinuxH doesn't yet scale onto big MP machines. The sweet spot are the thousandsE of old 4-8 way Solaris machines that can be outperformed by a two-wayx Proliant running a "free" OS.u   - Greg --  
 Greg Cagle gregc at gregcagle dot com   ------------------------------   Date: 6 Oct 2003 12:31:15 -0500u+ From: young_r@encompasserve.org (Rob Young)a Subject: Re: Sun takes a hit3 Message-ID: <Aw0U2Ta5plWb@eisner.encompasserve.org>    In article <bls4jv$4h2$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > Rob Young wrote:a >> In article <vnppv12drd8vef@corp.supernews.com>, "Dave Gudewicz" <k9jdk@NOSPAMarrl.net> writes:d >> eN >>>Permit me to rephrase.  I certainly don't know, but it appears that the sun >>>is starting to set. >> v >> u >>  B >> 	It isn't starting to set, it has been setting.  What cracks meA >> 	up is the recent downgrades.  As if there was a big surprise.n >> eH >> 	I've been saying for a few years - the only place they are going is D >> 	down.  It is a Dell/Windows and Linux effect.  Sparc/Solaris is B >> 	proprietary and moribound.  The VAX/VMS of the 21st century.  F >> 	They'll breath life into it by ditching SPARC and the overhead of K >> 	maintaining that infrastructure.  It's staring Scotty in the face, but  % >> 	it is hard to abandon your baby.   >> t >> s >  > 2 > Now the division of the company that you need to3 > stay operating because it produces the OS and kite4 > that your moon appears to revolve around is losing1 > ~ 1 billion a year off its bottom line, and youI1 > are excited about Sun taking a 1 billion dolalr  > accounting charge. >   = 	But HP has plenty of other divisions to prop it up and makes E 	money every quarter.  The excitiment isn't about a 1 time accountingm: 	charge.  It is how everyone else is piling on as if Sun's< 	troubles are something new or a surprise.  Sun hasn't grown? 	revenues in 9 quarters, the most recent $1 billion charge justtD 	reflects structural defects.  Because if it, a flurry of headlines:  < http://www.thestreet.com/_yahoo/tech/kcswanson/10117341.html   Sun's Very, Very Bad Week   3 http://biz.yahoo.com/rb/031002/tech_sunmicro_4.htmln  $ Analyst: Sun Micro Facing a 'Crisis'   http://www.marketwatch.com/news/yhoo/story.asp?source=blq/yhoo&siteid=yhoo&dist=yhoo&guid=%7B977A973A%2DDAC6%2D4BF7%2DAA1C%2D317C2DFC29F0%7D  '  Merrill rips Sun Micro, urges changes '  o> 	It isn't just me getting excited.  But really, why the suddenA 	surprise by the analyst community?  They see 9 straight quarters @ 	of declining revenue and an accounting charge (based on futures 	by the way) is a surpirse?a  2 > To cap it all we now have rumours that Intel are1 > being forced to adopt x86-64 by MS for Prescotto. > which will if true be the end of Itanium and. > the end of HP's Enterprise Systems division.    = 	Nah.  I don't see the tie-in.  You would be naive to suggestC; 	the Alpha team won't make something extremely fast.  Intelx< 	will sell it at a price that makes sense.  IA64 isn't dead.    2 > I would suggest that you spend a whole load less3 > time sicking up boring rehashes of your origional>4 > predictions about Sun and more time worrying about3 > the death of Itanium which Sun isn't involved in.n  > 	The death of SPARC is more topical.  Hard to find articles on 	the death of Itanium.   > 3 > It would appear that whatever has happened in the / > last 5 years it hasn't altered you ability tor! > not see the wood for the trees.i >   A 	No, your ability to feebly spin things.  Sun is in deep trouble,u
 	admit it.   				Rob    ------------------------------  % Date: Mon, 06 Oct 2003 21:34:14 +0200c* From: Paul Sture <nospam@sture.homeip.net> Subject: Re: Sun takes a hit0 Message-ID: <3F81DFD6.4079A2E8@sture.homeip.net>   Dave Weatherall wrote: > G > On Sun, 5 Oct 2003 08:32:01 UTC, Paul Sture <nospam@sture.homeip.net>e > wrote: >  > > Keith Parris wrote:e > > >al > > > young_r@encompasserve.org (Rob Young) wrote in message news:<POztxxIPltF8@eisner.encompasserve.org>...< > > > >       It isn't starting to set, it has been setting. > > >aA > > > TheStreet.com says It's been a very, very bad week for Sun. B > > > http://www.thestreet.com/_yahoo/tech/kcswanson/10117341.html > > >e6 > > > Motley Fool says it's been Sun's Nightmare Week.r > > > http://www.fool.com/news/mft/2003/mft03100317.htm?source=eptyholnk303100&logvisit=y&npu=y&bounce=y&bounce2=y > > >fE > > > "On Monday, the computer maker issued an earnings warning whileiK > > > tacking on a $1.05 billion charge to last quarter. Yesterday, MerrilljF > > > Lynch analyst Steve Milunovich sent a hostile open letter to theF > > > company's board stating that 'Sun faces a crisis.' And now this:E > > > Friday morning, Hewlett Packard announced that it would pay SuneB > > > customers $25,000 to switch to HP computers running Linux OS > > > freeware." > >gD > > But this is something I don't get. Why is HP pushing Linux as anJ > > alternative to Solaris rather than HP-UX, or even, heaven forbid, VMS? > >rC > > Why not give those away and go for the software maintenance andw > > consulting revenue?  > >w > > ???????? > >a6 > > It all sounds like dot com boom economics to me... > > K > > Come on Keith, we know you aren't daft. Please explain the economics ofA
 > > this one.  > 	 > HI Paulh> >                have you read this month's C'T yet? Stiller'sE > Prozessorgefluester (Processor News) talks about Sun actively goingdF > after Dell and HP's server sales and making an attack on Microsoft'sF > Desktop pitch. The latter with a Suse Linux variant. The question is > who is pre-empting whom? >   B Thanks for the pointer. Also fin the German press, a short articleB reports that since MS took over RAV, the linux antivirus firm, andH announced the discontinuation development on Linux, the staff don't wantG to learn Windows, so are taking their expertise and hopfully former RAVo customers to Kaspersky.d  \ > http://www.vnunet.de/testticker/news/detail.asp?ArticleID=9688&Topic=Newsticker&Ref=pc-pro   (text in German)   -- i
 Paul Sture   ------------------------------  % Date: Mon, 06 Oct 2003 15:46:25 -0400n* From: JF Mezei <jfmezei.spamnot@istop.com> Subject: Re: Sun takes a hit) Message-ID: <3F81C690.97711A13@istop.com>e   Rob Young wrote:C >         charge.  It is how everyone else is piling on as if Sun'soE >         troubles are something new or a surprise.  Sun hasn't grownhH >         revenues in 9 quarters, the most recent $1 billion charge justM >         reflects structural defects.  Because if it, a flurry of headlines:s  N Nortel, another company which bet and lost on the .com has begun to come back.N But that is because it retained its original business (telephony), not because/ of its acquired .com assets (bay networks etc).3  M For me, the health of "Sun" isn't so much measured by the corporation itself, N but perhaps by the market for Sun products. The .com collapse generated a glutN of servers. Until that glut is gone, I personally don't expect Sun to start to show significant growth again.  M Also, if you look at Sun as appliances (web servers), then once the applianceiK is installed, as long as the traffic on your web site doesn't grow anymore,pM you don't need to upgrade. The one brigt spot is SMTP software. They may needsK to upgrade their SMTO servers to handle the extra volume generated by thosedR microsoft viruses. (takes much mroe cpu to scan through the contents of messages).   ------------------------------   Date: 6 Oct 2003 15:38:26 -0500>+ From: young_r@encompasserve.org (Rob Young)u Subject: Re: Sun takes a hit3 Message-ID: <dVLQ7QEwZ8Qa@eisner.encompasserve.org>e  V In article <3F81C690.97711A13@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes: > Rob Young wrote:D >>         charge.  It is how everyone else is piling on as if Sun'sF >>         troubles are something new or a surprise.  Sun hasn't grownI >>         revenues in 9 quarters, the most recent $1 billion charge justsN >>         reflects structural defects.  Because if it, a flurry of headlines: > P > Nortel, another company which bet and lost on the .com has begun to come back.P > But that is because it retained its original business (telephony), not because1 > of its acquired .com assets (bay networks etc).p    = 	Right.  They are $4 and change today, climbing back to $100 n 	a share level of January 2001.e  O > For me, the health of "Sun" isn't so much measured by the corporation itself, P > but perhaps by the market for Sun products. The .com collapse generated a glutP > of servers. Until that glut is gone, I personally don't expect Sun to start to  > show significant growth again.    ? 	That is why it is very important to look at fundamentals, i.e.n< 	how are they doing total revs and profits, how much kit are> 	they shipping?  That kind of thing is hard to hide from.  The= 	fact that Dell has been increasing share (numbers in several = 	directions), IBM doing quite well (server numbers in several = 	directions) makes Sun's arguments all that more hollow.  Suno: 	is being eaten in several directions.  High-end IBM makes> 	MUCH more sense (HP too) if for no other reason very powerfulB 	CPUs which makes those Oracle per-cpu licensing go a lot further.$ 	Dell is eating them on the low-end.     				Rob    ------------------------------  # Date: Tue, 07 Oct 2003 02:20:49 GMTa# From: "John Smith" <a@nonymous.com>o Subject: Re: Sun takes a hitK Message-ID: <5qpgb.244170$Lnr1.224370@news01.bloor.is.net.cable.rogers.com>i   Rob Young wrote: > In articleE > <cqfgb.123778$3r1.6890@news02.bloor.is.net.cable.rogers.com>, "John ! > Smith" <a@nonymous.com> writes:r >> Rob Young wrote:n6 >>> In article <3F80B824.F46CCFE1@istop.com>, JF Mezei' >>> <jfmezei.spamnot@istop.com> writes:e >>>> Rob Young wrote:iH >>>>>         you're going to switch Unixes , doesn't make much sense toB >>>>>         move to a Unix that isn't Linux and isn't running on >>>>> commodity hardware.t >>>>F >>>> If you're being forced to migrate your VMS infrastructure to that >>>> IA64 thing,C >>>> it doesn't make sense to migrate to an expensive non-commodityoB >>>> hardware/software especially when the IA64 future is clouded. >>>> >>>vB >>> Not as clouded as SPARC's future.  A dying competitor to IA64. >> >>G >> Rob, remind me how many IA-64 systems shipped last year and what thesH >> real forecast is for shipments this year. What are Sun's comparables? >> >r> > SPARC growth rate is negative, has been for a while.  We see= > a flurry of reports, including several about Sun in crisis,h; > etc.  and recent downgrades, Wall Street analysts writingn; > open letters to Sun urging something be done.  As this is # > all a surprise.  Yet end of July:e >e > L http://groups.google.com/groups?selm=bybJ6pVnhmln%40eisner.encompasserve.org &oe=UTF-8&output=gplainn >wB > Sun is shrinking itself out of existence.  Sun did $18.2 billion< > in sales year ending June 2001.  Sun did ~$12.5 billion in> > sales year ending June 2002.  An amazing growth rate of -32%
 > in revs. > @ > Sun is doing very poorly as has been rehashed here in the last@ > 2 weeks.  Declining revenues for 9 straight quarters (and yes,E > we now know what view to take when comparing quarters).  Sun hasn'ts/ > grown revenues at all in the last 9 quarters.t >f > ---n >aD > Sun is dying.  Itanium is getting started.  Compare the two a yearD > from now, and two years from now.  You can be assured that ItaniumG > will be growing far faster than Sun and Sun will *still* be shrinkinga< > (if it exists at all, i.e. isn't taken over).  Speaking of> > taking over ... read a recent analysis.  When Sun was flyingC > high 4-5 years ago, they had a market cap of $200 billion.  Todayl> > it is around $10 billion and they have $5 billion of cash on> > hand.  A hostile takeover would be a very timely event about5 > now, that sure is a nice pot of gold Sun has there.s    ) So what's stopping HP from digesting Sun?n  G Offer 15% premium to the current stock price, which if your numbers arenK correct would seem generous to a fault for a company that is imploding like> a red giant.  K HP can always tell the FTC that Linux is growing and that the concentration D of Tru64/HP-UX/Solaris is more than offset by Linux growth and IBM'sG (remember them...largest computer company on the planet with far deeper ) pockets and more profitable than HP) AIX.P  K Then HP can migrate all Solaris customers to Tru64 (aka. HP-UX) or Linux ontI IA-64 (if, as, and when), and make money on the services and maintenance.    ------------------------------  # Date: Tue, 07 Oct 2003 02:24:06 GMTv# From: "John Smith" <a@nonymous.com>  Subject: Re: Sun takes a hitK Message-ID: <atpgb.244205$Lnr1.221630@news01.bloor.is.net.cable.rogers.com>    Rob Young wrote:   >n> > But HP has plenty of other divisions to prop it up and makes > money every quarter.  G Some day somebody at HP may make a Jack Welch type of decision - if theo@ division isn't in the top 2 in its field it gets sold or closed.   ------------------------------   Date: 6 Oct 2003 21:56:52 -0500 + From: young_r@encompasserve.org (Rob Young)a Subject: Re: Sun takes a hit3 Message-ID: <hmTtySZHnJH2@eisner.encompasserve.org>   q In article <5qpgb.244170$Lnr1.224370@news01.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes:    > + > So what's stopping HP from digesting Sun?  >   = 	Confusion it will cause.  Boat anchor situation.  On anothert 	round of takeover rumors:  / http://money.cnn.com/2003/05/02/technology/sun/n   Sun jumps on takeover rumors y   M Company declines to comment; stock most active on Nasdaq; analysts skeptical., May 2, 2003: 4:54 PM EDT      H "I can't think of anyone who would want to buy Sun," said Michael Cohen,M director of research for PacificAmerican Securities, which does no investment K banking. "Nobody is going to spend more than $10 billion for a company thatoH might go away on its own." Sun's current stock market value is about $12# billion, including Friday's gains. d  L If anything, Cohen said, Sun would be wise to try and acquire a company thatN could help boost its competitive position. Despite its troubles, Sun does haveJ a strong balance sheet, with $5.5 billion in cash and just $1.5 billion in long-term debt.   ; 	What might happen is Sun buys something instead of getting < 	bought.  Otherwise, they risk someone snatching them up for@ 	the cash and carving it up and selling/spinning off the pieces.  I > Offer 15% premium to the current stock price, which if your numbers are7M > correct would seem generous to a fault for a company that is imploding likeo > a red giant. > M > HP can always tell the FTC that Linux is growing and that the concentrationaF > of Tru64/HP-UX/Solaris is more than offset by Linux growth and IBM'sI > (remember them...largest computer company on the planet with far deeperm+ > pockets and more profitable than HP) AIX.o > M > Then HP can migrate all Solaris customers to Tru64 (aka. HP-UX) or Linux on K > IA-64 (if, as, and when), and make money on the services and maintenance.>  > 	Yep.  But... the confusion and energy required to pull it offH 	would be much too hard on HP/IBM/Dell but who knows a crazy industry.  C 	Someone other than computer industry (RJR Nabisco - i.e. Reynolds  C 	tobacco buying Nabisco years ago) OR a Fujitsu (read about Fujitsu > 	being a possible candidate recently) is a more likely suitor.  A 	But I think it is more likely that HP and IBM are smelling bloodi@ 	in the water and will greatly ratchet up their SunAway programs7 	and HP's recent program is probably one of several news) 	ones we will see in the next few months.r   				Robs   ------------------------------  % Date: Mon, 06 Oct 2003 22:52:04 -0400e* From: JF Mezei <jfmezei.spamnot@istop.com> Subject: Re: Sun takes a hit) Message-ID: <3F822A38.603FBC76@istop.com>    John Smith wrote:sI > Some day somebody at HP may make a Jack Welch type of decision - if theuB > division isn't in the top 2 in its field it gets sold or closed.  J Good wall street PR. But the real solution is to hire competent people and9 give them enough empowerement to make that division work.>  G IN VMS terms, it means giving VMS management the right to advertise andr: replacing those who don't believe VMS should be advertise.   ------------------------------  # Date: Mon, 06 Oct 2003 17:20:40 GMTe* From: Paul Anderson <paul.anderson@hp.com>+ Subject: Re: Unrecognized printer with DCPSo5 Message-ID: <061020031320409531%paul.anderson@hp.com>   ? In article <rl23ov080o1q0ma04414th49bfc5ijrepc@4ax.com>, Jeisonl <acanom@iberdrola.es> wrote:  E > I've some problems with DCPS ( dec print supervisor ) and a toshiba>D > printer model e-studio45. DCPS does not recognize this printer andD > when i try to print there is an error such a DCPS-F-CONTERMINATED.  H This printer will not work with any currently released version of DCPS. G We tried to add support for the Toshiba eSTUDIO series of printers but,pF because the printer and/or network card did not support bi-directional# communication, it was not possible..  F You are likely to get the printer to work with DCPS V2.3, currently inE field test and scheduled for release soon, by using a DCPS LPD queue.P   Paul   -- ,  Paul Anderson   OpenVMS Engineeringa   Hewlett-Packard Company    ------------------------------  " Date: Mon, 6 Oct 2003 17:15:33 GMT+ From: "BigFish" <johnsonronald@hotmail.com>e. Subject: Vax / VMS V5.5-2 External Drive Issue' Message-ID: <HMCILw.nC@news.boeing.com>>  F I just formatted two pc hdd for use on the VAX using the >>> t/ut scsiK command. Both drives formatted successfully, however, once I brought up theaL operating system (v5.5-2) I was not able to initialize either drives I get aL init-f-drverr fatal drive error message. One drive is 4.2 GB's and the other@ is 4.55 GB well underneath VAX v.5-5.2 8GB disk size limitation.  
 Any thoughts?r     -- Ronald Johnson JohnsonRonald@hotmail.coms   ------------------------------  % Date: Mon, 06 Oct 2003 13:21:52 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com>i2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue' Message-ID: <3F81CEE0.5090900@MMaz.com>k   BigFish wrote:  G >I just formatted two pc hdd for use on the VAX using the >>> t/ut scsi L >command. Both drives formatted successfully, however, once I brought up theM >operating system (v5.5-2) I was not able to initialize either drives I get a.M >init-f-drverr fatal drive error message. One drive is 4.2 GB's and the otheraA >is 4.55 GB well underneath VAX v.5-5.2 8GB disk size limitation.  >y >  f >eG You didn't mentioned which VAX you're using, but older ones have a sizesH limit on the system disk.  Also, the older VMS will most likely need the5 DKDRIVER patch to deal with other SCIS page issues...g   Barry    -- m  > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028   ------------------------------  % Date: Mon, 06 Oct 2003 19:46:43 +0100r) From: Antonio Carlini <arcarlini@iee.org>s2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue? Message-ID: <mMigb.8184$QH3.7458@newsfep4-winn.server.ntli.net>>   BigFish wrote:  1 > I just formatted two pc hdd for use on the VAX r  9 Why? I've swapped disks between PC and VAX and I've never>: needed to format a SCSI disk. Try putting them back on the# PC to see what you've done to them.>  = What VAX did you use to format them? The early MicroVAX 3100so; and all the VAXstation 3100s cannot "see" past 1.073GB intoo7 the disk, so you may have partially formatted the disk.s  M > command. Both drives formatted successfully, however, once I brought up theoN > operating system (v5.5-2) I was not able to initialize either drives I get aN > init-f-drverr fatal drive error message. One drive is 4.2 GB's and the otherB > is 4.55 GB well underneath VAX v.5-5.2 8GB disk size limitation.  G You weren't running OpenVMS when you formatted them - the console is a d whole different beast.   Antonio    -- r   --   --------------->- Antonio Carlini             arcarlini@iee.orgp   ------------------------------  % Date: Mon, 06 Oct 2003 15:51:34 -0400i* From: JF Mezei <jfmezei.spamnot@istop.com>2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue) Message-ID: <3F81C7C5.4A2E994D@istop.com>    Antonio Carlini wrote:; > Why? I've swapped disks between PC and VAX and I've never < > needed to format a SCSI disk. Try putting them back on the% > PC to see what you've done to them.s  H a real format is a good idea since it will work its magic to replace badM sectors etc. Would you trust a PC to correctly handle bad sector replacementsv etc ?   ? > What VAX did you use to format them? The early MicroVAX 3100sl= > and all the VAXstation 3100s cannot "see" past 1.073GB intoa9 > the disk, so you may have partially formatted the disk.   % This applies only to the system disk.s  L Also, the actual format is essentially a single scsi command. The drive thenL formats itself (I think it may then issue warnings during the format, but am8 not sure). So the drive should know how o format itself.  P > > init-f-drverr fatal drive error message. One drive is 4.2 GB's and the otherD > > is 4.55 GB well underneath VAX v.5-5.2 8GB disk size limitation.  N Have you taken a look at the error log under VMS ?  Are you sure parity is set properly on the drives ?   ------------------------------  $ Date: Mon, 6 Oct 2003 22:56:12 +0100( From: "John Travell" <john@jomatech.com>2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue9 Message-ID: <blsoe0$g7fkl$1@ID-120847.news.uni-berlin.de>   7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messagem# news:3F81C7C5.4A2E994D@istop.com...> > Antonio Carlini wrote:= > > Why? I've swapped disks between PC and VAX and I've never > > > needed to format a SCSI disk. Try putting them back on the' > > PC to see what you've done to them.n >eJ > a real format is a good idea since it will work its magic to replace badB > sectors etc. Would you trust a PC to correctly handle bad sector replacements > etc ?  >iA > > What VAX did you use to format them? The early MicroVAX 3100sb? > > and all the VAXstation 3100s cannot "see" past 1.073GB intoe; > > the disk, so you may have partially formatted the disk.n >o' > This applies only to the system disk.    This is not true.c  D At the console level on 'affected' machines it applies to ALL disks.L The problem is that the relevant console firmware was created at a time whenI space in ROM chips was very restricted, so one of the tricks used to save G space was to use only 21 bits of a longword to store an lbn to use whenoL accessing a disk, e.g. during a system boot. The other 11 bits were used forL something else. This means that the console can only access lbn's within the% first 1Gb of the disk (1.073decimal).iC You CAN use a larger disk as a system disk *IF* you use manual filetI placement to ensure that all relevant files are wholly in lbn's below theoE 21bit lbn boundary, BUT, do a backup and restore and all of your file.G placement goes out the window, and so does your ability to boot off the  disk."H Hence disks larger than a true 1Gb(binary) are not supported as a system disk on these machines.i  J The reason we only view it as a system disk issue is that the console doesK not access other disks, VMS does. And while older versions of VMS had their>G own restrictions, they did not have the same limitations as the console 	 firmware.n     -- John Travell" Independent VMS crashdump analyst. john- at - jomatech - dot - comi +44-(0)23-92552229 http://www.jomatech.com/           --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.522 / Virus Database: 320 - Release Date: 29/09/2003    ------------------------------  % Date: Mon, 06 Oct 2003 15:04:39 -0700e+ From: "Barry Treahy, Jr." <Treahy@MMaz.com>t2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue' Message-ID: <3F81E6F7.5090907@MMaz.com>y   John Travell wrote:a  8 >"JF Mezei" <jfmezei.spamnot@istop.com> wrote in message$ >news:3F81C7C5.4A2E994D@istop.com... >  t >s >>Antonio Carlini wrote: >>     >>< >>>Why? I've swapped disks between PC and VAX and I've never= >>>needed to format a SCSI disk. Try putting them back on thei& >>>PC to see what you've done to them.	 >>>        >>>eJ >>a real format is a good idea since it will work its magic to replace badB >>sectors etc. Would you trust a PC to correctly handle bad sector >>     >>
 >replacementsa >  y >  >>etc ?h >> >>     >>@ >>>What VAX did you use to format them? The early MicroVAX 3100s> >>>and all the VAXstation 3100s cannot "see" past 1.073GB into: >>>the disk, so you may have partially formatted the disk.	 >>>      l >>>e' >>This applies only to the system disk.  >>     >> >d >This is not true. >sE >At the console level on 'affected' machines it applies to ALL disks.d >  l >sF Hair splitting... A disk is only a 'system' disk if you can boot from E it.  That the limit applies to all disks is true but also immaterial oG because once VMS is up and running, the limit on 5.5-2 is increased to CB just a little more than 8GB per mounted volume.  Therefore, JF is G essentially correct that it only 'applies' to system disks because the m/ limit is only applicable to bootable devices...,   Barry0   -- 9  > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                        9   ------------------------------  # Date: Mon, 06 Oct 2003 20:45:03 GMTo# From: hoff@hp.nospam (Hoff Hoffman)t2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue1 Message-ID: <jvkgb.6431$YA.3309@news.cpqcorp.net>u  U In article <HMCILw.nC@news.boeing.com>, "BigFish" <johnsonronald@hotmail.com> writes:CG :I just formatted two pc hdd for use on the VAX using the >>> t/ut scsinL :command. Both drives formatted successfully, however, once I brought up theM :operating system (v5.5-2) I was not able to initialize either drives I get a M :init-f-drverr fatal drive error message. One drive is 4.2 GB's and the otherfA :is 4.55 GB well underneath VAX v.5-5.2 8GB disk size limitation.  :e :Any thoughts?  D   I would encourage you to read the frequently asked questions (FAQ)F   document as a start, as there is a reasonable chance this problem isG   covered there -- some of the more common causes of problems with SCSInG   devices on older releases are covered.  There is a section in the FAQ3F   on third-party widgets which should help you gain some understandingG   of this area -- you have set out to become far more of a SCSI expert,sI   and you are not the first and won't be the last on this path. :-)  (PutcJ   another way, please also remember to check the archives and the FAQ for B   various of the previous discussions on this and related topics.)   G   BTW, there is no such thing as a "PC HDD" (and anything named a "SCSIdI   standard" is seemingly just as hard to find :-), and there can and does G   exist substantial variations among SCSI command requirements and SCSIhH   implementations -- SCSI is a collection of optionally implemented (andH   occasionally optionally misimplemented) features, and rather less of aG   traditional "standard".  SCSI is big and hairy, and older versions ofoI   the SCSI documents had, um, even greater room for interpretation.  (ATA F   and ATAPI (sometimes called IDE) is big and hairy and different, and:   still has seemingly endless room for interpretation. :-)  G   Your OpenVMS VAX release is designed for and expects a contemporarily I   ancient SCSI interface and particularly SCSI devices -- upgrades to thesI   SCSI command interface that brought OpenVMS and its SCSI device drivershH   into line with then-current SCSI expectations were made circa 1995 andJ   1998, to allow OpenVMS to better adapt to arbitrary SCSI devices.  BeingI   your release dates from 1992, it lacks various of these updates and you K   will have to more carefully configure your disks to meet the expectationssI   of that and equivalently old or older releases.  (Or you could considerBD   an OpenVMS VAX upgrade to V6.2, to V7.2 or to current, obviously.)  F   SYS$ETC:SCSI_INFO appeared in V6.2, and can assist here -- this toolD   allows you to review the internal settings of arbitrary SCSI disk 9   devices -- for instance.  Also see RZTOOLS and related.p  I   If you want this to auto-configure and operate with a minimum of deviceiG   adjustments, you'll want to look around for a SCSI disk device -- onesF   of the older RZ-series widgets -- that was formally supported by theH   OpenVMS VAX V5.5-2 release.  (These were configured as expected by theE   then-current OpenVMS VAX releases, and this can largely avoid your aC   current need to test and to troubleshoot the SCSI configuration.)r  J   My guess is that this is something weird with this PC HDD, or with this J   configuration, or this is a mismatch with the ARRE and AWRE expectationsH   present within ancient OpenVMS SCSI drivers and referenced in the FAQ.  J   And congrads on your decision to learn about SCSI -- but realize this isI   an effort that is seemingly highly effective at creating gray hair  :-)e  J   Good luck with this, and remember to bring the rubber chicken needed to H   bless the resulting SCSI chain.  :-)   Sorry, that's a long story. :-)  N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqnN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.como   ------------------------------  " Date: Mon, 6 Oct 2003 21:49:44 GMT+ From: "BigFish" <johnsonronald@hotmail.com> 2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue( Message-ID: <HMCvAu.G4B@news.boeing.com>   Update:     I am using a VaxStation 4000-90.   -Ron   -- Ronald Johnson JohnsonRon@email.uophx.edu      6 "BigFish" <johnsonronald@hotmail.com> wrote in message! news:HMCILw.nC@news.boeing.com...cH > I just formatted two pc hdd for use on the VAX using the >>> t/ut scsiI > command. Both drives formatted successfully, however, once I brought upt theaL > operating system (v5.5-2) I was not able to initialize either drives I get a H > init-f-drverr fatal drive error message. One drive is 4.2 GB's and the otheriB > is 4.55 GB well underneath VAX v.5-5.2 8GB disk size limitation. >o > Any thoughts?r >  >e > -- > Ronald Johnson > JohnsonRonald@hotmail.come >  >i >  >t >I   ------------------------------  * Date: Mon, 6 Oct 2003 17:02:01 -0500 (CDT) From: sms@antinode.org2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue) Message-ID: <03100617020139@antinode.org>   ( From: "John Travell" <john@jomatech.com>9 > "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messagee% > news:3F81C7C5.4A2E994D@istop.com...s > > Antonio Carlini wrote:	 > > [...]PC > > > What VAX did you use to format them? The early MicroVAX 3100s2A > > > and all the VAXstation 3100s cannot "see" past 1.073GB intoe= > > > the disk, so you may have partially formatted the disk.e > >n) > > This applies only to the system disk.> >  > This is not true.n > [...]t  E    Let's not waste too nuch (more) time on this particular argument.  C While the original questioner did not specify his VAX model, he dideH mention his format command, ">>> t/ut scsi".  If any of the systems withG the 1GB limit could deal with this command, it's news to me.  Certainly H on a VAXstation 3100 model 30/40/38/48, it's ">>> TEST 75", not the more7 modern "TEST /UTILITY" (or whatever "ut" is short for).S  C    And, as someone said, this sort of formatting involves sending anG single "format yourself" command to the drive, and the drive presumablyrD knows its own size, whether or not the VAX firmware can cope with it$ (and in this case, it probably can).  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode.org     Saint Paul  MN  55105-2547k   ------------------------------  $ Date: Mon, 6 Oct 2003 23:21:21 +0100( From: "John Travell" <john@jomatech.com>2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue9 Message-ID: <blspt3$fuk7k$1@ID-120847.news.uni-berlin.de>r  6 "Barry Treahy, Jr." <Treahy@MMaz.com> wrote in message! news:3F81E6F7.5090907@MMaz.com...n > John Travell wrote:t >t: > >"JF Mezei" <jfmezei.spamnot@istop.com> wrote in message& > >news:3F81C7C5.4A2E994D@istop.com... > >e > >u > >>Antonio Carlini wrote: > >> > >>> > >>>Why? I've swapped disks between PC and VAX and I've never? > >>>needed to format a SCSI disk. Try putting them back on theg( > >>>PC to see what you've done to them. > >>>0 > >>>WL > >>a real format is a good idea since it will work its magic to replace badD > >>sectors etc. Would you trust a PC to correctly handle bad sector > >> > >> > >replacementsr > >  > > 	 > >>etc ?p > >> > >> > >>B > >>>What VAX did you use to format them? The early MicroVAX 3100s@ > >>>and all the VAXstation 3100s cannot "see" past 1.073GB into< > >>>the disk, so you may have partially formatted the disk. > >>>t > >>>n) > >>This applies only to the system disk.  > >> > >> > >  > >This is not true. > >hG > >At the console level on 'affected' machines it applies to ALL disks.e > >  > > G > Hair splitting... A disk is only a 'system' disk if you can boot fromaF > it.  That the limit applies to all disks is true but also immaterialH > because once VMS is up and running, the limit on 5.5-2 is increased toC > just a little more than 8GB per mounted volume.  Therefore, JF isnH > essentially correct that it only 'applies' to system disks because the1 > limit is only applicable to bootable devices...   L it also applies to anything other than booting that the console firmware canD do to a disk, possibly including formatting if the console has to do0 anything beyond merely starting the process off.H I cannot remember every detail of the console capability of such ancientI machines (neither can I remember everything my Alpha firmware can do, and' that is much newer!).t     -- John Travell" Independent VMS crashdump analyst. john- at - jomatech - dot - come +44-(0)23-92552229 http://www.jomatech.com/       ---r& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.522 / Virus Database: 320 - Release Date: 29/09/20037   ------------------------------  # Date: Mon, 06 Oct 2003 22:29:21 GMTt# From: hoff@hp.nospam (Hoff Hoffman)M2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue1 Message-ID: <51mgb.6443$%H.1272@news.cpqcorp.net>s  V In article <HMCvAu.G4B@news.boeing.com>, "BigFish" <johnsonronald@hotmail.com> writes:  ! :I am using a VaxStation 4000-90.m  J   While we are dragging out the details, and what SCSI disks are involved?  H   This system does not have the SCSI console addressing limits discussed   elsewhere in the thread.  I   This is most likely the incompatibility between the expectations of thelI   OpenVMS VAX V5.5-2 release and what the SCSI peripheral is providing torK   the host, as described in my earlier response -- and specifically, in the I   OpenVMS FAQ.  (This is the ancient SCSI drivers present in this equallyp/   ancient OpenVMS VAX release, in other words.)e  I   Boot a recent OpenVMS VAX distribution off CD-ROM as a test, and go forlG   a look as a test -- this will tell you if OpenVMS VAX V7.3 can accesssG   the disks, or if this is a problem with the disks or elsewhere withinoD   your SCSI chain.  With length, termination, signal integrity, etc.     --  I   The VAXstation 3100 discussed elsewhere in the thread applies to system G   disks, and particularly to system disk bootstraps and particularly tonG   the system disk writes that can occur during a system crash; both use,I   the same (and limited) console drivers.   As you might guess, there are K   also details on this particular subject within the OpenVMS FAQ, including1I   a list of specific VAXstation 3100 and MicroVAX 3100 series models thatsF   are affected by this particular console-level SCSI addressing limit.    N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqeN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Tue, 07 Oct 2003 01:15:40 +0100 - From: John Laird <nospam@laird-towers.org.uk>n2 Subject: Re: Vax / VMS V5.5-2 External Drive Issue8 Message-ID: <q014ovg3clv1iap81tqc7cbkvkbun50r1r@4ax.com>  L On Mon, 6 Oct 2003 22:56:12 +0100, "John Travell" <john@jomatech.com> wrote:  8 >"JF Mezei" <jfmezei.spamnot@istop.com> wrote in message$ >news:3F81C7C5.4A2E994D@istop.com... >> Antonio Carlini wrote:  >>B >> > What VAX did you use to format them? The early MicroVAX 3100s@ >> > and all the VAXstation 3100s cannot "see" past 1.073GB into< >> > the disk, so you may have partially formatted the disk. >>( >> This applies only to the system disk. >u >This is not true. >,E >At the console level on 'affected' machines it applies to ALL disks. M >The problem is that the relevant console firmware was created at a time whennJ >space in ROM chips was very restricted, so one of the tricks used to saveH >space was to use only 21 bits of a longword to store an lbn to use whenM >accessing a disk, e.g. during a system boot. The other 11 bits were used for M >something else. This means that the console can only access lbn's within thep& >first 1Gb of the disk (1.073decimal).  L I thought it was related to the use of an early format of scsi command whichK could not specify large block numbers.  I'd guess the prom fix swapped to aeL longer command format.  (That's a little more believable than having console@ rom programmers saving bits down to that extreme, to be honest.)   -- e Mail john rather than nospam...e   ------------------------------  # Date: Mon, 06 Oct 2003 20:19:04 GMTM4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>" Subject: Re: VMS system on the web0 Message-ID: <3F81CD04.8B8BF12C@blueyonder.co.uk>   Paul Sture wrote:x >  > Doc.Cypher wrote:e > >o > > all we're missing is CMS > H > I shall see if I can dig a kit up for you. I assume youe want both VAX3 > and Alpha, so that might take a little more time.e > 5 > Not sure I can go back to a pre 7.3 version though.3  N I can if you can't. In fact I have condist media going back to V5 and V6 days.  5 I'm sure the CMS kit it won't take too long over DSL.o     -- r tim.llewellyn@blueyonder.co.uk   ------------------------------   End of INFO-VAX 2003.555 ************************