1 INFO-VAX	Sun, 21 Jul 2002	Volume 2002 : Issue 399       Contents:$ RE: HP Lan Console x Itanium servers' Itanic2 - the cHumPaq spin continues... + Re: Itanic2 - the cHumPaq spin continues... + Re: Itanic2 - the cHumPaq spin continues... + Re: Itanic2 - the cHumPaq spin continues... + Re: Itanic2 - the cHumPaq spin continues...  Re: McKinley Cometh... Re: McKinley Cometh... Old vaxes available . Re: outside user can't access host via network. Re: outside user can't access host via network Re: Terminal input from DCL  Re: Terminal input from DCL   Thanks for the Freeware CD issue Re: TKZ60 SCSI ID ( RE: Upgrading from 7.2-1 and latest ECOs( RE: Upgrading from 7.2-1 and latest ECOs! Re: [Hobbyist] User account setup ! Re: [Hobbyist] User account setup ! Re: [Hobbyist] User account setup   F ----------------------------------------------------------------------  % Date: Sun, 21 Jul 2002 11:07:03 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com> - Subject: RE: HP Lan Console x Itanium servers T Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660864@kaoexc01.americas.cpqcorp.net>   Fabio,  E >>>I think with the future of migration of PA-RISC to Itanium, this = @ hardware will be integrated in the Itanium High End Servers. >>>  I Perhaps I am missing something, but are you asking for a remote console =  mgmt solution?  H If so, a number of these exist today for OpenVMS - including web based =B console mgmt solutions. They will typically support any platform =H remotely (system, network, disk controller etc) that has a serial port =I console. You can be sitting at home and using vpn to access an internal = I server, login using secure password and with the application running at = E that server be sitting right at the ">>>" prompt - all sessions are = * typically logged for later review as well.  C ConsoleWorks just recently announced that their web based console = F management product now also supports SSL/SSH console sessions as well.  ( Again, is this what you are looking for?  5 If so, check out these products that support OpenVMS: / http://www.tditx.com/products_consoleworks.html H http://www.tditx.com/news_events_press.html#062502 (ConsoleWorks press = release for SSL / SSH support)2 http://www3.ca.com/Solutions/Product.asp?ID=3D12106 http://www.robomon.com/products/detail_robocentral.htm   Regards   
 Kerry Main Senior Consultant  Hewlett-Packard Canada! Consulting & Integration Services  Voice: 613-592-4660  Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----8 From: Fabio Cardoso [mailto:fabiopenvms@yahoo.com.br]=20 Sent: July 20, 2002 6:05 PM  To: Info-VAX@Mvb.Saic.Com - Subject: Re: HP Lan Console x Itanium servers     	 Robert=20   / The Lan Console is a default hardware in the HP  PA-RISC 1 machines. I think with the future of migration of 4 PA-RISC to Itanium, this hardware will be integrated6 in the Itanium High End Servers. So, as I am imaginig,6 if these machines will run OVMS and HP-UX (I will name6 GSSD - GS Super Dome), I hope this hardware will be=202 ported. It is a good hardware for managing servers3 remotely, or when you have hubdreds of servers in a 5 rack and just one LCD to share with the team. You can # connect to the server by Telnet....   . If it is possible for me to ask this to the HP engineering, tell me how :-)=20     & Nobody here works with HP PA-RISC ????   Reg    FC  =20       0 --- Robert Deininger <rdeininger@mindspring.com> wrote: > In article > 5 <20020715161858.28800.qmail@web20209.mail.yahoo.com>,  > Fabio + > Cardoso <fabiopenvms@yahoo.com.br> wrote:  >=20 > >Just imagining: > > 4 > >As HP Itanium machines will run HP-UX and OpenVMS0 > >in the same box (I hope) , and to mantain the2 > >compatibility  of HP-UX environments, I imagine > that1 > >HP will develop a Lan Console for Itanium (for  > HP-UX 2 > >use). So, it could be used for OpenVMS machines > too. > > , > >Anyone from the PA-RISC x Alpha x Itanium
 > engineering " > >here to answer my question ???? >=205 > Well, you've asked about this a couple of times, so  > it seems nobody here! > knows enough to give an answer.  >=204 > Why don't you request future support for this kind > of widget through 2 > proper channels -- VMS product management and/or > your friendly  > neighborhood VMS ambassador? >=201 > IIRC, Compaq had a similar widget for the Intel  > boxes.  I don't know if it- > ever found it's way into any alpha systems.      =3D=3D=3D=3D=3D L =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D  F=E1bio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazil  fabiopenvms@yahoo.com.brL =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D   2 __________________________________________________ Do You Yahoo!?@ Yahoo! Health - Feel better, live better http://health.yahoo.com   ------------------------------  # Date: Sun, 21 Jul 2002 07:57:31 GMT * From: "Bill Todd" <billtodd@metrocast.net>0 Subject: Itanic2 - the cHumPaq spin continues...C Message-ID: <LPt_8.198334$iB1.10663413@bin4.nnrp.aus1.giganews.com>   J Just took a look at a shiny-new (this month) HP 'technical white paper' (aC title which seems to preclude the defense that "it's just marketing J garbage") called "Inside the Intel Itanium 2 Processor", and found that HPC is continuing Compaq's fine record of lies about the virtues of the I platform.  Just so there's no doubt about the honesty of the company that ' now owns VMS, here are some highlights:   G Page 4:  Note the graph comparing Itanic performance to the PA8x00 over A time.  In 2001 we see Merced appearing at nearly the same integer L performance (about 90%, as close as my eye can tell) as the PA8700.  A quickK examination of spec.org finds two March, 2001 (Merced debuted in May, 2001) K tests of the 8700 yielding SPECint scores of 569/603 and 568/604, while the K best HP (or anyone else) has ever managed for a Merced is a score of 379/-, H i.e., almost exactly 2/3 the integer performance of the 8700 and nowhere nearly 90%.   L We then see Itanic2 (McKinley) appearing just about now at *well over twice*K the integer performance of the 8700.  Of course, *no* official SPEC listing L for McKinley has yet appeared, but even if the 810 SPECint2K figure reportedL by HP is accepted by SPEC it's well under 50% higher than the 750 MHz 8700'sI SPECint base, only about 1/3 higher than the 750 MHz 8700's SPECint peak, L and likely no more than 25% higher than the new (already introduced, but notJ yet SPEC-listed) 875 MHz 8700's SPECint base (or about 15% higher than its
 likely peak).   I And McKinley's integer performance is shown as noticeably higher than the H coming PA8800, in contrast to the estimates I've seen (which if anythingL appear conservative) that PA8800 will debut with a score of about 900 (i.e.,: noticeably higher than, rather than lower than, McKinley).  K In other words, just as Compaq deliberately understated Alpha's performance K relative to Itanic, so HP appears to be doing exactly the same with PA-RISC F (and in both cases leaving aside the question of whether the iterativeL profiling feedback development loop required to achieve Itanic's performanceJ numbers is realistic to expect in real-world applications).  The only partH of the graph that appears to correspond at all closely to reality is theH performance of the Itanic versions with respect to each other:  McKinleyJ does offer over twice the integer performance of Merced, and Madison seemsH likely to offer about 35% - 40% better integer performance then McKinleyE (note that this is considerably less than the more optimistic Madison  boosters have predicted).   K The paper then degenerates into fluff such as "The Itanium Processor Family J has support for almost all major operating systems", which HP then goes onK to enumerate as HP-UX, Win2K (neglecting to mention the 'limited' nature of E this support), and Linux (I won't even bother to list the many other, F unsupported OSs that by any reasonable definition qualify as 'major').I Interestingly (and for once somewhat objectively), it does quote a March, I 2000, Aberdeen Group statement that Itanic "will become the volume leader H for enterprise applications ... within 5 to 7 years of its debut."  EvenH taken at face value that's mid-2006 - mid-2008, and one wonders how muchL Aberdeen might have pushed out these dates after the magnitude of the Merced flop later became known.  K Moving on to page 7, we see another equally mendacious graph (actually, the H graph itself, if interpreted as linear in the vertical axis, may be lessK mendacious than the numbers accompanying its curves).  The assertion to its K left that "The EPIC architecture is a new architecture advancement enabling H processors to break through the limitations of RISC" mirrors the numbersK indicating that while OoO super-scalar RISC achieves instructions-per-clock H values of ">2" EPIC achieves IPC values of ">4" (i.e., close to twice as great).   I However, when we look at the now-aging Alpha EV6 core running at the same E clock speed as Itanic2, we see nothing like a 2:1 integer performance I difference, but only 20% - 30% (depending on whether one looks at base or A peak SPECint scores).  And that aging core is just about to get a G clock-speed boost (whereas both Merced and McKinley have had difficulty L meeting even reduced clock-speed targets, calling into question whether theyJ will offer superior performance even if they do manage to eke out slightlyI greater IPC).  And of course EV7 is about to debut (still using the aging J EV6 core, but with a considerably faster memory subsystem - which McKinleyL has as well) with performance-per-clock that promises to be at least as goodE as McKinley's (and with better absolute performance, as it will clock L faster).  Too bad EV8 won't come along to leave EPIC's performance-per-clockL in the dust, but even EV6/7 are sufficient to make rubbish of EPIC's claimed advantages in that area.  E How about PA-RISC?  The situation there is even more embarrassing for A Itanic:  the 8700 is nipping right at McKinley's heels in SPECint L performance-per-clock (only about 6% behind in SPECint2K base, assuming HP'sI 810 score for Itanic2 is accepted, and dead-even per-clock if you use the  8700's SPECint peak score).   I In fact, McKinley can only barely manage 50% better performance-per-clock H (for SPECint base; only a bit over 30% better for SPECint peak) than theK *in*-order USIII RISC platform.  And as for MIPS, I don't know whether it's J OoO or not, but its SPECint base score just about exactly matches McKinleyL in performance-per-clock and its SPECint peak per-clock score slightly beats; McKinley's SPECint base (HP didn't provide a SPECint peak).   K So in marked contrast to the paper's assertions both in its text and in its 8 graph, EPIC enjoys no noticeable advantage whatsoever inL performance-per-clock over *any* OoO RISC competitor (save possibly POWER4 -J but POWER4 clocks enough faster to cancel out any absolute advantage), andH furthermore *will* enjoy no such advantage for at least another 3 years,J since the McKinley core will be with us at least through Montecito in 2004F and very likely through 2005 as well (though it may get some help fromJ peripheral on-chip improvements in that year).  Ironically, *exactly* thisG conclusion is substantiated by the performance-per-MHz graph on page 21 ( (though it doesn't include MIPS or EV7).  E Page 7 then goes on to extoll the virtues of the 'simplicity' of EPIC G relative to OoO RISC.  I think this point has been more than adequately L covered (both here and in comp.arch):  while the original theory of EPIC wasJ that it would be simple, its implementation has turned out to be both moreI complex and less power-efficient than its RISC competition - and the only F avenues forward (after 2005) that have been suggested (increased OoO -I though whether that could still be termed 'EPIC' seems questionable - and L SMT) promise to be even *more* complex than comparable RISC implementations.I And this ignores the massive amount of thread context baggage that Itanic G burdens each context-switch with (which AFAIK is not measured in common 2 benchmarks but can be a real killer in some uses).  F So the paper simply ignores current reality (including the significantL question of whether increased individual processor performance, even if EPIC9 *could* provide it, would be more important than improved L performance-per-Watt and/or increased numbers of processors per die, both ofH which EPIC is conspicuously poor at addressing relative to existing RISCL cores) and instead presents decade-old arguments that have long since proved; invalid.  If this is not a lie, I'm not sure what would be.   J (Just to address the possible contention that improved compiler technologyI will eventually realize EPIC's promises that to date are purely illusory: J the paper itself states, and correctly, that "The Itanium Processor FamilyK requires advanced compilers for optimal performance.  Fortunately this type J of compiler technology has been in development at HP for many years and isI now ready for commercial release."  In other words, don't expect dramatic B improvements on that front:  the main work has already been done.)  G It's not worth deconstructing much of the rest of the paper, since it's J largely "gee, whiz!" talk about details of all the neat features of ItanicK that still don't succeed in making it the superior platform it was promised L to be.  It is, however, worth noting that many of the approaches are equallyI applicable to RISC architectures, and hence do not constitute any kind of H absolute advantage for EPIC (rather, they would if incorporated increaseI RISC's superiority over EPIC):  given funding remotely comparable to that J expended upon EPIC, one could expect any of the good RISC architectures to run away from it.   B But we do glean the new nugget of information on page 29 that "TheG performance of the Itanium 2 processor's IA-32 engine is expected to be K comparable with a 300 MHz Pentium Pro."  Wow - that'll sure cut Hammer IA32 K performance down to size!  Especially since if one runs a code mix heavy on H the IA32 side the IA32 code will consume most of the available processorC time even though only running at about 15% of native performance...   E And the RAS features do not appear significantly different from those L present in Hammer (at least to one who is only superficially conversant withL such areas), so unless others have more detailed information about them thatL particular knock on the low-end-to-mid-range portion of the competition also seems to have evaporated.   E So meet The New HP:  same spinners as the old Compaq.  Hold onto your  wallets.   - bill   ------------------------------  % Date: Sun, 21 Jul 2002 10:05:28 -0400  From: <rob@netcarrier.net>4 Subject: Re: Itanic2 - the cHumPaq spin continues...9 Message-ID: <fgz_8.486$Jn2.419483697@news.netcarrier.net>   	      Bill   &   I suspect you should just let it go.  : Repeating the same stuff over and over will change nothing    !  Let's get back to VMS questions.                             Rob5 "Bill Todd" <billtodd@metrocast.net> wrote in message = news:LPt_8.198334$iB1.10663413@bin4.nnrp.aus1.giganews.com... L > Just took a look at a shiny-new (this month) HP 'technical white paper' (aE > title which seems to preclude the defense that "it's just marketing L > garbage") called "Inside the Intel Itanium 2 Processor", and found that HPE > is continuing Compaq's fine record of lies about the virtues of the K > platform.  Just so there's no doubt about the honesty of the company that ) > now owns VMS, here are some highlights:  >    ------------------------------  # Date: Sun, 21 Jul 2002 15:56:20 GMT * From: "Bill Todd" <billtodd@metrocast.net>4 Subject: Re: Itanic2 - the cHumPaq spin continues...C Message-ID: <DQA_8.254645$Im2.12846118@bin2.nnrp.aus1.giganews.com>   % <rob@netcarrier.net> wrote in message 3 news:fgz_8.486$Jn2.419483697@news.netcarrier.net...  >  >      Bill  > ( >   I suspect you should just let it go. > < > Repeating the same stuff over and over will change nothing  G As long as cHumPaq keeps repeating their drivel, I'll keep refuting it. K Though I won't stoop to their level and lie, they disgust me and deserve to: fail.P  G While the field is complex, the actual comparison process is not rocketeK science.  The EPIC core takes more die area (=> more production cost), morenC power (=> more run-time cost), more programmer effort (in iterative I profiling => more software cost and time), and more complex compilers (=>tH more platform development cost and time) to achieve the same performanceG that competitive RISC processors achieve - and this situation will only F worsen for it over at least the next 3 - 4 years (with no compensatingF up-turn on the horizon even after that point:  they just may then stopI losing ground).  It is not a 'better idea', it is not even an equal idea:y it is a worse idea.u   - bill   ------------------------------  # Date: Sun, 21 Jul 2002 16:30:05 GMT-# From: "John Smith" <a@nonymous.com> 4 Subject: Re: Itanic2 - the cHumPaq spin continues...H Message-ID: <hkB_8.48171$WsS.22529@news01.bloor.is.net.cable.rogers.com>  % <rob@netcarrier.net> wrote in message03 news:fgz_8.486$Jn2.419483697@news.netcarrier.net...> >t >      Bill  >u( >   I suspect you should just let it go. > < > Repeating the same stuff over and over will change nothing    I Perhaps it won't change anything at HP. They have a mammoth investment in I Itanic, and for them to show it in any light that is other than favorablecJ would cause the stock market to drive their stock price down considerably,K as investors would question the sanity of the decisions to kill PA-RISC and H Alpha. This would not be a happy prospect for HP shareholders, or anyone? employed by HP whose compensation is linked to the stock price.a  D Which raises another interesting question....just in the same way asG companies and CEO's are now being raked over the coals for 'cooking theWF books', can a company and CEO be similarly investigated by the SEC forI deliberately posting misleading performance numbers for their technology?hL The numbers they post for Itanic 2 compared to PA-RISC and Alpha may well beC justifiable under certain, but not all, circumstances, but are theyw( sufficiently 'defensible'? I don't know.  L What I suspect is that HP will do everything it can to show that Itanic 2 isH a better performer than current or next to-be-released PA-RISC and AlphaL chips. They have a huge motive to do so, even if it means carefully choosingH the benchmarks they release to fit the message they want to get out. ForC some users, the benchmarks that HP chooses to release may fit their L application/usage perfectly and be an ideal way for them to compare relativeL performance, while for others the published benchmarks may be the wrong onesB entirely and either show nothing of value at all, or show that the' chip/system is not good value for them.t  K That said, all computer companies do/have done this sort of thing. But thattK does not mean that it is correct to do so - it may be similar in that whilenI FASB and GAAP permit some latitude in the categorizing of some accountingiI items, it does not permit the deliberate mis-statements that we have seenwG recently in the books of companies. But then, as we have seen in recenttL discussion here about various benchmarks, some appear to favor one processorJ architecture more than another, or that the way the benchmark is coded can- be 'stretched' to fit a particular processor.   H Probably the only fair benchmarks are to pick 1000 programmers at randomL from a wide variety of customers, give them the benchmark specification, putL them each in a 'clean room' and tell them to code the benchmark and then run9 those 1000 different implementations on all the different I processors/systems. Then take the results and publish them all, calculaterK the mean and std. deviations, and that would give purchasers a more or less I 'real world' approximation of performance based on the skills of a random E sampling of programmers. Not all programmers are experts, not all are H complete losers either. Most are simply average in their skills, even at software/hardware vendors.  I But since HP is not apparently selling OpenVMS to *new* customers (as has L been quoted may times in c.o.v from a public HP document), all this apparentE 'fiddling' of the relative performance numbers, if Bill's analysis isoK correct, may only serve to drive existing VMS customers away faster if they'D were at the point in their decision-making processes where they were: considering whether to stay with, or migrate from OpenVMS.   ------------------------------  # Date: Sun, 21 Jul 2002 16:49:46 GMT 1 From: Ed Wensell III <ewensell3@yahoo.commercial>m4 Subject: Re: Itanic2 - the cHumPaq spin continues...0 Message-ID: <3D3AE62D.5527B8E1@yahoo.commercial>   Bill Todd wrote: > I > As long as cHumPaq keeps repeating their drivel, I'll keep refuting it. M > Though I won't stoop to their level and lie, they disgust me and deserve to2 > fail.2  H But preaching to the comp.os.vms choir will accomplish nothing. EveryoneE that watches COV knows the sermon all too well. And AFAICT, no one in3D the group has the power to print these messages and drop them on the CEO's desk.t   -- l Ed Wensell IIIE NetBSD/Alpha at home - Solaris/SPARC at work - OpenVMS in a past life A E-mail address is valid if you know the appropriate bits to drop.t   ------------------------------  % Date: Sun, 21 Jul 2002 10:54:14 +0200 D From: "Niels J=?ISO-8859-1?B?+A==?=rgen Kruse" <nj_kruse@get2net.dk> Subject: Re: McKinley Cometh...m. Message-ID: <UEu_8.25$694.460@news.get2net.dk>  @ I artiklen <3D3831AC.7010902@brussels.sgi.com> , Alexis Cousein  <al@brussels.sgi.com>  skrev:m   > Jan C. Vorbrggen wrote: >%L >>>I actually had a hand in writing the art benchmark and this post hits theO >>>nail on the head.  For this neural net simulation, an array of structures isgF >>>the more natural choice when it comes to conceptualizing the model,. >>>extensibility in a university setting, etc. >>>- >>M >> Well, maybe one of the problems is that people don't conceptualize enough.yL >> It's amazing (and I speak from more than a decade of experience) how muchH >> of this field degenerates into well-known things from linear algebra,M >> statistics and dynamical systems when look more closely. And when you thenrM >> write down the maths properly, the array of structures disappears, and theH  >> stucture of arrays appears... >3 >gO > I'd have to add that in some other fields, arrays of structures are sometimesaK > more cache friendly rather than less -- e.g. if you need all members of anN > structure at once. Esp. if your different arrays are dimensioned with powersO > of two and you've forgotten to add padding in the structure-fo-arrays case...   F Certainly if there are more (accessed) struct members than ways in theL cache. Associativity of L1 in reasonably current designs varies all over the map, from 2-way to 128-way.   J Another issue is automatic hardware prefetch. Each (accessed) member array4 constitutes a stream and only so many are supported.   --8 Mvh./Regards,    Niels Jrgen Kruse,    Vanlse, Denmark   ------------------------------    Date: 21 Jul 2002 12:54:12 +0200- From: Andreas Eder <Andreas.Eder@t-online.de>r Subject: Re: McKinley Cometh...w* Message-ID: <m3sn2djqqz.fsf@elgin.eder.de>  * nmm1@cus.cam.ac.uk (Nick Maclaren) writes:  > > Note that, in none of those cases, does the basic concept of> > RISC have anything to do with the matter.  KISS is the basis@ > of RISC, and I quite agree that it is a good principle.  It is= > the complexity introduced into the ISA in order to simplify B > the implementation of the hardware side that causes the trouble.  + Still another aspect of 'worse is better' !t   'Andreas -- T+ Wherever I lay my .emacs, theres my $HOME.    ------------------------------    Date: 21 Jul 2002 08:49:04 -0700% From: bmeyygoogle@bmegroup.com (Brad)o Subject: Old vaxes available= Message-ID: <4e21eba5.0207210749.32c4b0ff@posting.google.com>h  M I have two older vaxes: one microvax 2000 and the other maybe a little older.eA These are available at very low cost or free to a worthy cause.  h  & Please email me if you are interested.% I would prefer not to throw them out.o   I am in the Philly suburbs.i   Thanks,    Brad   ------------------------------   Date: 21 Jul 02 11:00:49 +0200) From: p_sture@elias.decus.ch (Paul Sture)v7 Subject: Re: outside user can't access host via networkd) Message-ID: <u5+jLczRoJtV@elias.decus.ch>p  Q In article <ahd5t8$f5m$1@husk.cso.niu.edu>, system@niuhep.physics.niu.edu writes:rI > A user has dsl hooked up to a windows 2000 machine and has a windows 98bC > machine hooked to on a local network with local only ip addressesEH > (sorry, I haven't dealt with this for so long I forget teh right name,D > his local network computers think they have ip addresses while the= > gateway machine keeps them separate from teh real internet)l > L > I am running VMS 7.2, TCPIP 5.0A, MX5.2, IUPOP3 2.04 and teh OSU webserver > 2.3? (old I know)e > I > He can not send email via our machine despite my having the appropriateII > "inside address" defined, he has had problems accessing the web server.y+ > He /has/ been able to use the pop server.  > J > But recently he has not been able to acces our machine at all, he can't K > even ping us.  This is from his inner network machines.  He can get othern, > web servers at teh university, but not us. > C > From the machine that is directly connected to the dsl connection  > he can telnet in.  t > G > When I look at the log files I see no sign that he is seen much less   > rejected.u >  > Any thoughts?  > J > He doesn't think he has changed anything, and I haven't chanegd anything > recently.e  H A long shot, but did he ever have his Win98 system networking configuredJ differently, for example with a dial up modem? I had mind numbing problemsE getting Win98 to talk to the outside via an ISDN router. Deleting theaG networking stuff and reinstalling still remembered at least part of theS original setup.   K Reinstalling Win98 from scratch and not mentioning a modem was the only wayi to cure it.i __
 Paul Sture Switzerlandl   ------------------------------  % Date: Sun, 21 Jul 2002 15:57:50 +0100t# From: David Webb <d.webb@mdx.ac.uk>c7 Subject: Re: outside user can't access host via networkf( Message-ID: <3D3ACBEE.4040608@mdx.ac.uk>  $ system@niuhep.physics.niu.edu wrote:  I > A user has dsl hooked up to a windows 2000 machine and has a windows 98IC > machine hooked to on a local network with local only ip addressesrH > (sorry, I haven't dealt with this for so long I forget teh right name,D > his local network computers think they have ip addresses while the= > gateway machine keeps them separate from teh real internet)  >     H  From your description it sounds like he is using private address space > (10.0.0.0/8, 172.16.0.0/12 or 192.168.0.0/16  address ranges).H These private addresses are not supposed to be routable on the internet.E He needs to have a box (router, firewall etc) acting as a NAT device.dF This translates the private internal addresses into external addressesB (often it translates lots of internal addresses into one external D address which is associated with the NAT box - it does this by also . remapping the client port numbers being used).      L > I am running VMS 7.2, TCPIP 5.0A, MX5.2, IUPOP3 2.04 and teh OSU webserver > 2.3? (old I know)  > I > He can not send email via our machine despite my having the appropriatetI > "inside address" defined, he has had problems accessing the web server.k+ > He /has/ been able to use the pop server.o >   D What do you mean by having the appropriate "inside address" defined.D If he is using NAT and you are outside the private network then you E should never see the internal private addresses you should only ever tH receive connections from the internet valid address(es) associated with  the NAT box.  D Can he access you using your IP address rather than your node name -' the problem might be with DNS lookups ?       J > But recently he has not been able to acces our machine at all, he can't K > even ping us.  This is from his inner network machines.  He can get othern, > web servers at teh university, but not us. > C > From the machine that is directly connected to the dsl connectiont > he can telnet in.  i >   < Is that the machine which is acting as the NAT gateway box ?> Get him to look and see whether there is anything in its logs.E It  sounds like that is the point at which the connection is failing.     G > When I look at the log files I see no sign that he is seen much less e > rejected.  >  > Any thoughts?e > J > He doesn't think he has changed anything, and I haven't chanegd anything > recently.o >  > Robert > system@physics.niu.edu >     
 David Webb   VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Sun, 21 Jul 2002 09:19:50 +0200e- From: Didier Morandi <Didier.Morandi@Free.fr>c$ Subject: Re: Terminal input from DCL' Message-ID: <3D3A6093.BA4D39F7@Free.fr>   
 Yes, I do.  P You said: "I am missing featues like: Predefined maximum number of characters inI length, editable default string to start with, handling of function keys,n> character string with the precise characters that are allowed)  ? The Hoffman's routine gives you this, and some coding examples.   M But David's answer has also a good point. You can redefine the whole jeyboardtS with a /terminate, before your input routine call, then redefine it back to normal.i   D.   "Frits A.M. Storms" wrote: > J > If you mean the example at page 349 from this book, it's in every aspect+ > that I wish to realise totally useless...f   ------------------------------  # Date: Sun, 21 Jul 2002 15:12:34 GMTo1 From: "David J. Dachtera" <djesys.nospam@fsi.net>a$ Subject: Re: Terminal input from DCL' Message-ID: <3D3AD3DC.7784BFC8@fsi.net>    Didier Morandi wrote:  >  > Yes, I do. > R > You said: "I am missing featues like: Predefined maximum number of characters inK > length, editable default string to start with, handling of function keys,-@ > character string with the precise characters that are allowed) > A > The Hoffman's routine gives you this, and some coding examples.  > O > But David's answer has also a good point. You can redefine the whole jeyboard1U > with a /terminate, before your input routine call, then redefine it back to normal.e  D Well, not the "whole" keyboard, only those keys that generate escape+ sequences. See HELP DEFINE /KEY Parameters.o  C My point was that without being able to trap individual keystrokes,oG there's no practical way to limit the count of characters entered so as 6 to limit entry into a field to a predetermined length.  @ You can do that in BASIC with INKEY$, but DCL has no equivalent.   -- - David J. Dachterau dba DJE Systems: http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/g   ------------------------------    Date: 21 Jul 2002 08:21:52 -0700/ From: maristella.masiero@libero.it (Maristella)n) Subject: Thanks for the Freeware CD issuen= Message-ID: <8f75104e.0207210721.7749f3b3@posting.google.com>   3 I wish to thank Compaq for the OpenVMS Freeware CD o5 v.5 issue: some monthes ago I discovered MenuFinder, p5 a great menu-driven interface for system management. e2 I now use it daily to find, update and run my DCL  procedures. 7 I wish also to thank the great MenuFinder's programmer b& who make available this tool for free.   Ciao Maristella Masiero MenuFinder programmer's wife ;)  www.itre.com/mf/download.htmls   ------------------------------  # Date: Sun, 21 Jul 2002 15:05:56 GMTc From: dittman@dittman.netS Subject: Re: TKZ60 SCSI ID8 Message-ID: <o5A_8.14166$927.10193@nwrddc01.gnilink.net>  0 David J. Dachtera <djesys.nospam@fsi.net> wrote: : dittman@dittman.net wrote: :> -> :> I've got a TKZ60 tape drive.  I need to change the SCSI ID,< :> but there's not SCSI ID switch on the back.  In fact, the= :> only controls on the tape drive are "UNLOAD" and "FORMAT".n< :> I opened the drive and couldn't find any internal jumpers; :> or switches to change the SCSI ID, so I'm guessing theseh5 :> buttons may be used somehow to change the SCSI ID.r :> e@ :> Can someone tell me whether the SCSI ID is fixed, and if not, :> how to change the SCSI ID?r  J : Well, your description suggests that there's a suffix to the TKZ60 modelD : number (such as TKZ60-VA or equivalent) that would help answer the : question.,   The model number is TKZ60-EA.o -- m Eric Dittman dittman@dittman.net = Check out the DEC Enthusiasts Club at http://www.dittman.net/o   ------------------------------  % Date: Sun, 21 Jul 2002 11:30:09 -0400p' From: "Main, Kerry" <Kerry.Main@hp.com>o1 Subject: RE: Upgrading from 7.2-1 and latest ECOsfT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660865@kaoexc01.americas.cpqcorp.net>   Lyndon,   H Is the issue you are describing different than what was described in the V7.3 cover letter?  G Also as a fyi, for those on V7.2-2, a new update kit is available which-+ ties together a number of previous patches.-   Reference: (one url will wrap)H http://ftp.support.compaq.com/patches/public/vms/axp/v7.2-2/dec-axpvms-v% ms722_update-v0100--4.pcsi-dcx_axpexet  C P.s. VMS V7.3-1 should be out fairly soon now as well (some time in8 August is likely good guess)   Regards>  
 Kerry Main Senior Consultantp Hewlett-Packard Canada! Consulting & Integration Servicesu Voice: 613-592-4660  Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----8 From: Lyndon Bartels [mailto:lbartels@pressenter.com]=20 Sent: July 20, 2002 11:00 PM To: Info-VAX@Mvb.Saic.Coma- Subject: Upgrading from 7.2-1 and latest ECOsl    E Came across a interesting little issue when trying to upgrade to v7.3 F last week. I had installed the very latest ECOs to v7.2-1 for a coupleD reasons. Then, a couple weeks later, I upgraded to v7.3. The upgrade9 went "According to Hoyle" At least it looked like it did.t  E The first time the computer tried to boot, it failed. Turns out, thatrG the latest "sys" eco, and "amacro" eco, improperly write entries in theyF sys$update:vms$remedial_old_files.txt file.  Namely an entry for "apb"G and "apb_debug" (from "sys" eco) and "starlet.mlb" (from "amacro" eco).a  B Any subsequent upgrade would "remove" the ECOs and hence the files mentioned in this file.=20  E This has happened before and the "rename old" eco fixed those issues.v8 This is the same issue. But with new and different ECOs.  G I logged a call, and got the system up and running on v7.3. The guy whoa9 handled the call says it's been elevated to engineering..   D We'll probably see a new "rename old" ECO, or some such pretty soon.     Lyndon     --=20mG My opinions are mine and mine alone. They seldom align with those of myl	 employer.     H The only good thing about putting the cart before the horse is you don't! have to look at the horse's butt.n   ------------------------------  % Date: Sun, 21 Jul 2002 11:43:04 -0400p' From: "Main, Kerry" <Kerry.Main@hp.com>h1 Subject: RE: Upgrading from 7.2-1 and latest ECOsoT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4023D931E@kaoexc01.americas.cpqcorp.net>  0 Oops, minor correction to pointer in last reply-  B The VMS V72-2 update kit release notes are at: (one url will wrap)H http://ftp.support.compaq.com/patches/public/Readmes/vms/dec-axpvms-vms7 22_update-v0100--4.READMEe   Regardsc  
 Kerry Main Senior Consultantu Hewlett-Packard Canada! Consulting & Integration Servicesu Voice: 613-592-4660  Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message----- From: Main, Kerry=20 Sent: July 21, 2002 11:30 AM To: Info-VAX@Mvb.Saic.Com 1 Subject: RE: Upgrading from 7.2-1 and latest ECOs      Lyndon,   H Is the issue you are describing different than what was described in the V7.3 cover letter?  G Also as a fyi, for those on V7.2-2, a new update kit is available whichs+ ties together a number of previous patches.a   Reference: (one url will wrap)H http://ftp.support.compaq.com/patches/public/vms/axp/v7.2-2/dec-axpvms-v% ms722_update-v0100--4.pcsi-dcx_axpexee  C P.s. VMS V7.3-1 should be out fairly soon now as well (some time ine August is likely good guess)   Regardsh  
 Kerry Main Senior Consultante Hewlett-Packard Canada! Consulting & Integration Services  Voice: 613-592-4660s Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----8 From: Lyndon Bartels [mailto:lbartels@pressenter.com]=20 Sent: July 20, 2002 11:00 PM To: Info-VAX@Mvb.Saic.Como- Subject: Upgrading from 7.2-1 and latest ECOst    E Came across a interesting little issue when trying to upgrade to v7.3eF last week. I had installed the very latest ECOs to v7.2-1 for a coupleD reasons. Then, a couple weeks later, I upgraded to v7.3. The upgrade9 went "According to Hoyle" At least it looked like it did.i  E The first time the computer tried to boot, it failed. Turns out, that3G the latest "sys" eco, and "amacro" eco, improperly write entries in thecF sys$update:vms$remedial_old_files.txt file.  Namely an entry for "apb"G and "apb_debug" (from "sys" eco) and "starlet.mlb" (from "amacro" eco).a  B Any subsequent upgrade would "remove" the ECOs and hence the files mentioned in this file.=20  E This has happened before and the "rename old" eco fixed those issues. 8 This is the same issue. But with new and different ECOs.  G I logged a call, and got the system up and running on v7.3. The guy whok9 handled the call says it's been elevated to engineering..k  D We'll probably see a new "rename old" ECO, or some such pretty soon.     Lyndon     --=201G My opinions are mine and mine alone. They seldom align with those of myg	 employer.     H The only good thing about putting the cart before the horse is you don't! have to look at the horse's butt.b   ------------------------------  % Date: Sun, 21 Jul 2002 09:13:30 +0200 - From: Didier Morandi <Didier.Morandi@Free.fr>a* Subject: Re: [Hobbyist] User account setup' Message-ID: <3D3A5F16.9E02CD73@Free.fr>s  ; Hmmm... can you ever log in locally with any other account? 1 If no, please reinstall VMS, booting from the CD.    D.   Alder wrote: >  > Didier Morandi wrote:t% > > $ set audit/ala/ena=file=fail=all  > > $ reply/enable > >o# > > on another console, gives what?m >  > Took a while, eh?o > J > Here's the entrails from a login attempt to a "bad" account named BORIS. > : > %%%%%%%%%%%  OPCOM  20-JUL-2002 12:58:54.35  %%%%%%%%%%%% > Message from user INTERnet on HOBBYn: > TELNET Login Request from Remote Host: KIRALY Port: 1428 >  > $ : > %%%%%%%%%%%  OPCOM  20-JUL-2002 12:59:08.80  %%%%%%%%%%%) > Message from user AUDIT$SERVER on HOBBYeJ > Security alarm (SECURITY) and security audit (SECURITY) on HOBBY, system > id: 1.< > Auditable event:          Remote interactive login failure3 > Event time:               20-JUL-2002 12:59:08.80 $ > PID:                      0000214E! > Process name:             BORISe! > Username:                 BORIS ) > Process owner:            [USERS,BORIS] G > Terminal name:            _TNA162:, Host: 192.168.0.1      Port: 1428tL > Image name:               HOBBY$DKA0:[SYS0.SYSCOMMON.][SYSEXE]LOGINOUT.EXEF > Status:                   %LOGIN-F-CLIFAIL, error activating command > interpreter !AS   ------------------------------   Date: 21 Jul 02 10:26:58 +0200) From: p_sture@elias.decus.ch (Paul Sture)0* Subject: Re: [Hobbyist] User account setup) Message-ID: <JNQt7BTu$Kbd@elias.decus.ch>   W In article <3D39B070.8000203@spammotel.com>, Alder <PGDEHMKOKIMD@spammotel.com> writes:) > Carl Perkins wrote:, >> oF >> P.S. Are you the same person I just sent an e-mail to about gettingE >> SAMBA working on your system? If so, and if it is the same system, : >> then this problem could also be why that isn't working. > H > How can you tell? :-)  I'll be giving that post a close read a little ( > later and trying out your suggestions. > J > I can't get over the amount of assistance this group can provide.  It's H > humbling and quite a departure from the typical advice on the Windows = > newsgroups to visit someone's webpage or reinstall Windows.d >  >  > 		  THANK YOU EVERYONE!e >   J Likewise it is a pleasure to see someone who is appreciative of the advice0 given. Please keep us informed of your progress.  r __
 Paul Sture Switzerlando  J PS May I quote your comments? It's always good to have a bit of ammunition* about the quality of the VMS community :-)   ------------------------------  # Date: Sun, 21 Jul 2002 15:20:21 GMT-1 From: "David J. Dachtera" <djesys.nospam@fsi.net>l* Subject: Re: [Hobbyist] User account setup& Message-ID: <3D3AD5B0.30795F8@fsi.net>   Alder wrote: >  > Didier Morandi wrote: % > > $ set audit/ala/ena=file=fail=all. > > $ reply/enable > >c# > > on another console, gives what?h >  > Took a while, eh?0 > J > Here's the entrails from a login attempt to a "bad" account named BORIS. > : > %%%%%%%%%%%  OPCOM  20-JUL-2002 12:58:54.35  %%%%%%%%%%%% > Message from user INTERnet on HOBBYo: > TELNET Login Request from Remote Host: KIRALY Port: 1428 >  > $a: > %%%%%%%%%%%  OPCOM  20-JUL-2002 12:59:08.80  %%%%%%%%%%%) > Message from user AUDIT$SERVER on HOBBYIJ > Security alarm (SECURITY) and security audit (SECURITY) on HOBBY, system > id: 1i< > Auditable event:          Remote interactive login failure3 > Event time:               20-JUL-2002 12:59:08.80 $ > PID:                      0000214E! > Process name:             BORISy! > Username:                 BORISm) > Process owner:            [USERS,BORIS]TG > Terminal name:            _TNA162:, Host: 192.168.0.1      Port: 1428 L > Image name:               HOBBY$DKA0:[SYS0.SYSCOMMON.][SYSEXE]LOGINOUT.EXEF > Status:                   %LOGIN-F-CLIFAIL, error activating command
 > interpreterI >   !ASi  B *SIGH* Someone else suggested setting an Audit ACE (Access Control> Entry) in the ACL (Access Control List) on both of DCL.EXE andF DCLTABLES.EXE. That's kind of heavy stuff for a newbie, but maybe that+ resulting data would yield a further clue. f  E There's not enough info. in HELP to determine the exact the syntax of9G the appropriate ACE, but if I can review the on-line doc.'s and come upuA with it, I'll post it later. Perhaps someone who does such thingsl8 "everyday" (Larry K.?) can post it from memory. Dunno...   -- m David J. Dachteraa dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/0   ------------------------------   End of INFO-VAX 2002.399 ************************