0 INFO-VAX	Thu, 25 Jan 2001	Volume 2001 : Issue 49      Contents: Re: $setimr problem  Re: $setimr problem  Re: $setimr problem # Re: All microsoft web sites offline # Re: All microsoft web sites offline # Re: All microsoft web sites offline # Re: All microsoft web sites offline # Re: All microsoft web sites offline # Re: All microsoft web sites offline < Announcement -  User Interface Tool Now Available on OpenVMS Re: Another missed opportunity Re: Another missed opportunity Re: Another missed opportunity Re: Another missed opportunity Re: Another missed opportunity Re: Another missed opportunity4 Re: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution+ Does vms support object method perl cgi.pm? / Re: Does vms support object method perl cgi.pm? / Re: Does vms support object method perl cgi.pm?  DS10's incoming Used& Re: Dual ethernet config for failover?" Re: Filtering certain keys on VMS." Re: Filtering certain keys on VMS." Re: Filtering certain keys on VMS.( Re: Hobbyist help- DPWS 500au, VMS 7.2-1( RE: Hobbyist help- DPWS 500au, VMS 7.2-1, Re: How to delete directory with bad blocks?6 Re: Is Compaq dropping support for vms on vax platform$ Re: Linux worm and RedHat 7.0 broken$ Re: Linux worm and RedHat 7.0 broken$ Re: Linux worm and RedHat 7.0 broken$ Re: Linux worm and RedHat 7.0 broken% Re: Lowercase filenames on VMS V7.2-1   Re: Memo:  Help! Zip/Unzip error  Re: Memo:  Help! Zip/Unzip error- Re: Moving from VMS to Tru64 Unix, Any Ideas? - Re: Moving from VMS to Tru64 Unix, Any Ideas? - Re: Moving from VMS to Tru64 Unix, Any Ideas? ! Re: New OpenVMS Times now on line ! Re: New OpenVMS Times now on line  Newbie question  Newbie question 4 Re: No technical computing, was: expanding the niche Re: NTP network load! Re: OpenVMS "reply/to" challenge.  Re: OpenVMS and Notebooks  Re: Printing problems in TCPIP( Re: RA230+ (KZPAC-CA) and 18/36 GB Disks( Re: RA230+ (KZPAC-CA) and 18/36 GB DisksN Re: Rehash of Andrew's Java on OpenVMS FUD.  (was: Re: GS160 hardwarequestion)N Re: Rehash of Andrew's Java on OpenVMS FUD.  (was: Re: GS160 hardwarequestion)N Re: Rehash of Andrew's Java on OpenVMS FUD.  (was: Re: GS160 hardwarequestion) RW-CD drives for VMS# Shareable image with DEC-C question ' Re: Shareable image with DEC-C question ' Re: Shareable image with DEC-C question . Re: stop decw$session = kill SUN CDE server... Re: TCPIP$BIND: dynamic updates  VMS Documentation Giveaway7 Re: VMS vs UNIX (was: Linux worm and RedHat 7.0 broken)  Re: VMS writable CD on Win2K?  Re: VMS writable CD on Win2K?  Re: VMS writable CD on Win2K?  Re: VMS writable CD on Win2K?  RE: VMS writable CD on Win2K? ! Re: Why is the tape write locked? ! Re: Why is the tape write locked?  Re: Wizard Disappears!!! Re: Wizard Disappears!!! Re: Wizard Disappears!!!# Writes flushed to Disk on Dismount? ' Re: Writes flushed to Disk on Dismount? ' Re: Writes flushed to Disk on Dismount? : Re: [Q]: Resetting privileged password on a DECserver 200?: Re: [Q]: Resetting privileged password on a DECserver 200?: Re: [Q]: Resetting privileged password on a DECserver 200?  F ----------------------------------------------------------------------  % Date: Wed, 24 Jan 2001 20:13:51 +0000 + From: "antonio.carlini" <arcarlini@iee.org>  Subject: Re: $setimr problem' Message-ID: <3A6F377F.396883BC@iee.org>    cstranslations@msn.com wrote:    > > 9 > > No EFN ... I don't know whether BASIC will just leave 4 > > random crud on the stack here or whether it puts* > > a zero in nicely for you. If it leaves9 > > random stuff you are probably setting who-knows-whose 0 > > event flag ... not good. IIRC the version of1 > > OpenVMS you have allows you to use EFN$C_ENF.  > I > I'm not a big fan of event flags. Too much of a "polling" feel. I'm 99% F > certain that BASIC makes the stack nice (unless an option inactive =F > setup is in effect). Not that I should necessarilly count on this so > point taken (again). >   0 The point is that regardless of whether you like, them or need them, most system services that* need them require them passed in by value./ Not one of DNCs better decisions IMHO (I assume * it was his fault ...) In your case you may3 well be setting a random EF or you may be affecting ' EFN0. Now it may be that none of *your* 0 code cares about *any* event flags but who knows, what each and every library routine that you( call (directly or indirectly because the* compiler does it for you) uses. Can you be" sure that they don't use any EFNs?  ) If you don't care about an event flag use ( EFN$C_EFN - which is precisely that .. a% value that means "don't need an EF".  $ Admittedly most people get bitten by+ $QIO without an IOSB, but it's such an easy " fix you might as well get into the
 habit now.  F > In any event as is frequently the case the problem wasn't were I was? > looking. The trace back was pointing at "line 3875." A bit of H > instruction reordering seems to have taken place. The one of the firstI > things the routine in question does (although it wasn't included in the G > original post) is check the final status on the read in the IOSB. The G > size of the record being written (by the images in production) to the I > mailbox were one byte larger that what the server was expecting (little B > bit of a slip up there the last time the server was recompiled).    3 I really must learn to read right to the end before  replying in future :-)  . Does BASIC have a /NOOPTIMIZE? VAX doesn't but. I would expect Alpha to help out a little more& here since reordering is such a common! things for those compilers to do.      Antonio    --------------- - Antonio Carlini             arcarlini@iee.org    ------------------------------  # Date: Wed, 24 Jan 2001 20:55:42 GMT = From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)  Subject: Re: $setimr problem0 Message-ID: <009F69A6.91AB88A1@SendSpamHere.ORG>  U In article <3A6F377F.396883BC@iee.org>, "antonio.carlini" <arcarlini@iee.org> writes:  >cstranslations@msn.com wrote: >  >> >: >> > No EFN ... I don't know whether BASIC will just leave5 >> > random crud on the stack here or whether it puts + >> > a zero in nicely for you. If it leaves : >> > random stuff you are probably setting who-knows-whose1 >> > event flag ... not good. IIRC the version of 2 >> > OpenVMS you have allows you to use EFN$C_ENF. >>  J >> I'm not a big fan of event flags. Too much of a "polling" feel. I'm 99%G >> certain that BASIC makes the stack nice (unless an option inactive = G >> setup is in effect). Not that I should necessarilly count on this so  >> point taken (again).  >>   > 1 >The point is that regardless of whether you like - >them or need them, most system services that + >need them require them passed in by value. 0 >Not one of DNCs better decisions IMHO (I assume  9 The Democratic Nation Committee had its hands in VMS?  :)    --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM              O city, n., 1. a place where trees are cut down and streets are named after them.    ------------------------------  # Date: Wed, 24 Jan 2001 21:20:35 GMT  From: cstranslations@msn.com Subject: Re: $setimr problem) Message-ID: <94ngun$s3a$1@nnrp1.deja.com>   5 In article <Z8lb6.58$cu.523@gazette.loc1.tandem.com>, &   hoffman@xdelta.zko.dec.nospam wrote:B > In article <94hsps$u35$1@nnrp1.deja.com>, cstranslations@msn.com writes: " > :OpenVMS 7.1-1H2, BASIC V1.2-000 > @ >   I'd encourage an upgrade to OpenVMS Alpha V7.1-2 with ECO... > E > :Q: Is there any reason as to why (or how) $setimr would decided to  > :return ss$_bufferovr? > G >   Because some condition handler somewhere up the stack frame told it  to?    Ask a stupid question... :-)   F >   I would strongly encourage explicit specification of an event flagG >   or the don't-care event flag EFN$C_ENF.  (EFN$C_ENF requires V7.0.)   C Well - now that I've gotten beaten silly over event flags . . . the F basic$starlet library that comes with V1.2-000 doesn't seem to provideA the symbol (not that it likely to change from 128 or whatever the  actual value is).   H >  --------------------------- pure personal opinion ------------------- --------1 >    Hoff (Stephen) Hoffman   OpenVMS Engineering  hoffman#xdelta.zko.dec.com >  >      Sent via Deja.com  http://www.deja.com/   ------------------------------    Date: 24 Jan 2001 13:47:15 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) , Subject: Re: All microsoft web sites offline+ Message-ID: <bDSNAmsdn8aH@eisner.decus.org>   ` In article <prit6tkuvi45r7dj3b7btgvrnhev8kujl1@4ax.com>, Alan Greig <agreig@my-deja.com> writes:  : > Oh dear, oh dear. I have just confirmed we cannot access > www.microsoft.com   N Please save wear & tear on the Internet by only reporting genuine problems :-)   ------------------------------  % Date: Wed, 24 Jan 2001 17:34:40 -0500 % From: "John Vottero" <John@mvpsi.com> , Subject: Re: All microsoft web sites offline/ Message-ID: <t6um5domv7nbe7@news.supernews.com>   F "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message% news:bDSNAmsdn8aH@eisner.decus.org... E > In article <prit6tkuvi45r7dj3b7btgvrnhev8kujl1@4ax.com>, Alan Greig  <agreig@my-deja.com> writes: > < > > Oh dear, oh dear. I have just confirmed we cannot access > > www.microsoft.com  > L > Please save wear & tear on the Internet by only reporting genuine problems :-)    Who said it was a problem? :)    ------------------------------  % Date: Wed, 24 Jan 2001 20:19:47 -0600 % From: Keith Brown <kbrown780@isd.net> , Subject: Re: All microsoft web sites offline' Message-ID: <3A6F8D43.5082B0AD@isd.net>    John Vottero wrote:  > H > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message' > news:bDSNAmsdn8aH@eisner.decus.org... G > > In article <prit6tkuvi45r7dj3b7btgvrnhev8kujl1@4ax.com>, Alan Greig  > <agreig@my-deja.com> writes: > > > > > > Oh dear, oh dear. I have just confirmed we cannot access > > > www.microsoft.com  > > N > > Please save wear & tear on the Internet by only reporting genuine problems > :-)  >  > Who said it was a problem? :)   + I'm sure it will be billed as a feature ;-)  --   Keith Brown  kbrown780@isd.net    ------------------------------  % Date: Wed, 24 Jan 2001 18:30:01 -0800 ! From: Shane.F.Smith@Healthnet.com , Subject: Re: All microsoft web sites offlineD Message-ID: <OF37B1E376.DE4CC176-ON882569DF.000DAA9A@foundation.com>  C It's a security feature. Their only effective one, come to think of  it......   Shane           9 Keith Brown <kbrown780@isd.net> on 01/24/2001 06:19:47 PM    To:   Info-VAX@Mvb.Saic.Com  cc:   - Subject:  Re: All microsoft web sites offline      John Vottero wrote:  > H > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message' > news:bDSNAmsdn8aH@eisner.decus.org... G > > In article <prit6tkuvi45r7dj3b7btgvrnhev8kujl1@4ax.com>, Alan Greig  > <agreig@my-deja.com> writes: > > > > > > Oh dear, oh dear. I have just confirmed we cannot access > > > www.microsoft.com  > > E > > Please save wear & tear on the Internet by only reporting genuine  problems > :-)  >  > Who said it was a problem? :)   + I'm sure it will be billed as a feature ;-)  -- Keith Brown  kbrown780@isd.net    ------------------------------  % Date: Wed, 24 Jan 2001 22:18:38 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> , Subject: Re: All microsoft web sites offline, Message-ID: <3A6F9B05.8B037ABD@videotron.ca>   Keith Brown wrote:- > I'm sure it will be billed as a feature ;-)   F CNN mentioned the web site outages, but said that Microsoft blamed DNSJ problems on the internet and none of its systems were incapacitated or hadH been attacked. (i.e. its not the fault of mircosoft or of our software).   ------------------------------  % Date: Thu, 25 Jan 2001 12:33:26 +0800 4 From: Dave Sneddon - bigpond <dbsneddon@bigpond.com>, Subject: Re: All microsoft web sites offline+ Message-ID: <3A6FAC96.FA689FDC@bigpond.com>    JF Mezei wrote:  >  > Keith Brown wrote:/ > > I'm sure it will be billed as a feature ;-)  > H > CNN mentioned the web site outages, but said that Microsoft blamed DNSL > problems on the internet and none of its systems were incapacitated or hadJ > been attacked. (i.e. its not the fault of mircosoft or of our software).  B All the Micro$hit servers became sentient, discovered the completeD load of crap that they were and took themselves offline in an effort to save humanity... :-)    --   Regards, Dave. I ------------------------------------------------------------------------- I David B Sneddon (dbs)    VMS Systems Programmer     dbsneddon@bigpond.com I Sneddo's quick guide ...          http://www.users.bigpond.com/dbsneddon/ I DBS freeware at ...   http://www.users.bigpond.com/dbsneddon/software.htm I "Life is what happens to you while you're busy making other plans" Lennon    ------------------------------  % Date: Wed, 24 Jan 2001 14:38:50 -0500 2 From: "Sue Skonetski" <susan.skonetski@compaq.com>E Subject: Announcement -  User Interface Tool Now Available on OpenVMS 5 Message-ID: <KlGb6.88$cu.746@gazette.loc1.tandem.com>   L Integrated Computer Solutions, Inc. (ICS), announced that the latest release ofL its flagship product, Builder Xcessory (BX 5.0), is now available for CompaqJ AlphaServer systems running OpenVMS V6.2 or greater . ICS is also bundlingF EnhancementPak, a library of 30 reusable components including business graphsJ and sophisticated user interface controls, with the OpenVMS version of BX.  L Builder Xcessory simplifies the development of user interfaces on OpenVMS byK providing a WYSIWYG environment that allows developers to "paint" their use I interface. Creation of a Motif based user interface is simply a matter of G selecting the desired object off a palette, and dropping it on the user L interface under development. Various tailoring facilities are available thatB allow the user interface to be customized to the application. User
 interfacesL are easily altered to respond to changing end user needs. Standard C/C++ andF Java 1.1 is generated that can then be linked with backend code of theL application. Builder Xcessory also provides a number of facilities optimized for  large project efforts.  K Reinforcing ICS' long-term commitment to the OpenVMS development community,  ICS K is offering an optional three-year maintenance program with the purchase of  BX. L More detailed product information is available under the Products section of the C ICS home page at http://www.ics.com. Free 30 day evaluations can be  requested by emailing to info@ics.com.      - # Integrated Computer Solutions, Inc. * Visual Development Tools for Professionals     617-621-0060 x108 (voice)  617-621-9555 (fax)     201 Broadway Cambridge, MA 02139   : Visit the MotifZone (www.motifzone.net) for info on Motif!   ------------------------------    Date: 25 Jan 2001 05:09:38 +0800, From: Paul Repacholi <prep@prep.synonet.com>' Subject: Re: Another missed opportunity - Message-ID: <873de8k67h.fsf@prep.synonet.com>   4 rdeininger@mindspring.com (Robert Deininger) writes:  ? > If I understand your numbers (and if I believe them), you are < > talking about hardware revenue.  The compaq folks are alsoE > interested in PROFIT.  The high-volume, high-profit stuff is not as B > profitable.  Also, talking about hardware revenue while ignoringD > software and services revenue (which is what I think you are doing > here) is silly.   ? If you have any high volume, high profit suff you don't want...    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Wed, 24 Jan 2001 22:08:37 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)' Subject: Re: Another missed opportunity L Message-ID: <rdeininger-2401012208380001@user-2ive7g7.dialup.mindspring.com>   In article <rdeininger-2401010905070001@user-2ive6g5.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) wrote:  : > The high-volume, high-profit stuff is not as profitable.  n Sorry, I garbled that sentence.   Meant to say the high-volume stuff is not as profitable.  Obviously, the high-profit stuff is profitable.f  K > Also, talking about hardware revenue while ignoring software and servicesn> > revenue (which is what I think you are doing here) is silly. >e   -- n Robert Deininger rdeininger@mindspring.comv   ------------------------------  % Date: Wed, 24 Jan 2001 22:13:29 -0500r2 From: rdeininger@mindspring.com (Robert Deininger)' Subject: Re: Another missed opportunitypL Message-ID: <rdeininger-2401012213300001@user-2ive7g7.dialup.mindspring.com>  _ In article <g7tt6toisp02dr4o9tesdpverujfv7e3dc@4ax.com>, Alan Greig <agreig@my-deja.com> wrote:d    H > Come on if he had actually *said* *something nasty all hell would haveD > broken loose. He's managed to actually be nasty but with plausible, > deniability. I don't believe he is stupid. > F > Capellas is *well* aware of the sensitivity of VMS customers and theF > issue of mentioning everything other than VMS in end of year reportsB > has explicitly been raised before. There is no excuse this time.   If he's so diabolically clever, why didn't he dishonestly add in the sentence or two you claim is needed.  He could have made a bundle for Compaq by bamboozling you and the rest of us.   Either VMS folks are easy to fool, and he could have done it with his hands tied  behind his back ... or we are too keen to be taken in by a sound bite.     If more than a sound bite was required, I think he risked scaring the herd-followers who were an important audience in this case.    -- i Robert Deininger rdeininger@mindspring.como   ------------------------------  % Date: Wed, 24 Jan 2001 22:14:19 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)' Subject: Re: Another missed opportunitymL Message-ID: <rdeininger-2401012214200001@user-2ive7g7.dialup.mindspring.com>  [ In article <873de8k67h.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> wrote:    > A > If you have any high volume, high profit suff you don't want...o  @ Yes, there was a wee error in my statement, already corrected...   -- a Robert Deininger rdeininger@mindspring.com:   ------------------------------  % Date: Wed, 24 Jan 2001 22:37:05 -0500:- From: JF Mezei <jfmezei.spamnot@videotron.ca>V' Subject: Re: Another missed opportunity , Message-ID: <3A6F9F57.783DD2EF@videotron.ca>   Robert Deininger wrote:q > If more than a sound bite was required, I think he risked scaring the herd-followers who were an important audience in this case.I    N Those analyst conference calls are just an opportunity to collect sound bytes.N Capellas and his advisors who are heading a very large computer company shouldN have enough experience to be able to insert at least a simple sound byte aboutM VMS. They mentioned Hymalaya and Unix, and did that alienate the windows herd!2 ? They mentioned Linux ? Did that cause a raucus ?  M It is clear that Compaq does not wish to kill VMS at this time. But it is notiN clear what their intentions with VMS are. Is their plan a controlled peacefull1 phaseout, or a true renaissance of a capable OS ?   I If it were a renaissance of VMS, how can you explain the total lack of PR-
 exposure ?   ------------------------------  # Date: Thu, 25 Jan 2001 06:18:47 GMT1  From: jlsue <jlsuexxxz@home.com>' Subject: Re: Another missed opportunity:8 Message-ID: <ldfv6tk6hn67g3tf4i534m55l51nhare2f@4ax.com>  3 On Wed, 24 Jan 2001 15:08:44 +0000, andrew harrison.! <andrew.nospam@uk.sun.com> wrote:j  9 >Secondly Compaq spends virtually nothing on OpenVMS thisu8 >considerably assists it in being profitable. If Compaq 9 >wanted to increase OpenVMS's revenues they would have to 2 >invest and the payback would not be immediate or 4 >guaranteed. This consideration may explain Compaqs  >inactivity with OpenVMS.o  D Exactly where do you get this information?  How do you know how muchD Compaq spends on OpenVMS?  C'mon.  Out with it.  Are you secretly on0 our BOD?  Are you really in OpenVMS engineering?  F And let's just say, for the sake of argument, that OpenVMS spending isD lower than Unix - so what?  Unix still has quite a long way to go toA reach the level that OpenVMS and VMSclusters currently have.  AndtA Linux will probably never get there in our lifetime - not that itEB couldn't, but the PHMs in businesses suffer from attention deficitC disorder management philosophy.  i.e., they're waiting for the nextrB big thing to be hyped for them to move on to.  They don't rememberE their business decisions from last week, let alone which OS they werea interested in last year.   ------------------------------   Date: 24 Jan 2001 17:48:15 GMT' From: david20@alpha1.mdx.ac.uk (D.Webb) = Subject: Re: Compaq: A simple, affordable clustering solutionu0 Message-ID: <94n4gv$5ha$1@aquila.news.mdx.ac.uk>  ] In article <3A6F0802.CC966223@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes:o >"D.Webb" wrote: >> a` >> In article <3A6EF4C1.CA3702E3@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes: >> >"Main, Kerry" wrote: >> >>e
 >> >> Andrew,e >> >>oO >> >> While avoiding direct questions, you are trying to get various posters to 7 >> >> this newsgroup to start fighting with each other.n >> >>e >> >5 >> >What would your point be as a matter of interest.r >> >3 >> >He is making the claim you made for downtime ori5 >> >the lack of it when you have to reboot an OpenVMS  >> >cluster node.r >> >4 >> >The other posters on this group either disagreed7 >> >with your/his claim or had allready posted examples 4 >> >prior to the discussion which disproved your/his
 >> >claim. >> >7 >> >I don't know if this is what you would call a fight 6 >> >but surely you understand that in order to advance6 >> >your/his argument you have to refute their claims. >> > >> >Regards5 >> >Andrew Harrisonm >> >Enterprise IT Architectn >> rN >> As I recall there were only one or two disagreements (other than yourself).I >> Eg Developers who were connected to one machine from a workstation who I >> left themselves permanently connected and didn't like having to logout>K >> and login to another machine in the cluster. This even though they couldhO >> be notified and given a reasonable period (say 24hrs) in which to logout and> >> log straight back in.L >> or People who were running long batch jobs which had been written without >> restart capabilities. >> o >y+ >There were more than two dissenting voicesf) >to Kerrys orgional postings citing theirf >actual experiences. >t  L These are the only ones I recall. If you know more then post the details !!!    ) >There were also the previous examples inh( >particular the remedial actions post a * >duff patch from Compaq and the impact on ' >the users of that system. This example & >also differed from Kerry's marketing % >possition and from jlsue's apparent b >practical experience. >u  I Again post details of this duff patch. By the way are you saying that Sunn) engineers never release duff patches ????1  
 David Webb VMS and Unix team leader CCSS Middlesex University     >Regards >Andrew Harrison >Enterprise IT Architect   ------------------------------  # Date: Thu, 25 Jan 2001 06:18:42 GMT   From: jlsue <jlsuexxxz@home.com>= Subject: Re: Compaq: A simple, affordable clustering solutiona8 Message-ID: <3vbv6t0tf07630g59g5lf9av57qfdjidvk@4ax.com>  3 On Wed, 24 Jan 2001 12:45:08 +0000, andrew harrisont! <andrew.nospam@uk.sun.com> wrote:      >l3 >You are having the argument with the wrong person.16 >I never said that in my experience of managing OpnVMS3 >clusters you were incorrect but a number of other s5 >posters did, either directly or by example, Curtis'ss5 >user cull for example. Strangely you never disagreedm >with them at the time how odd.   F Oh, nice try to dodge, but it won't work.  Everyone here has read yourE assertions of opinion as if they are "fact".  Andy, answer yes or no:lF Do you say that VMSclusters, in general, work as I described them?  IfC not, please detail on what basis you make this assertion.  C'mon, ItC want numbers of clusters and apps that come at least near what I've  described to you.o   > 6 >Perhaps you should have responded to Curtis's posting5 >describing his user cull for example. Perhaps he did-4 >something wrong that you could have prevented. Why 1 >not re-read hist post and then perhaps he might e >appreciate your advice !w  > Well, I believe that previous statements were already answeredC adequately by others, and I had no new information to add.  I don'tlE have ego problem that requries me to answer everything that is stateds9 by others in here, unlike, apparently, you (for example).b  E However, I don't recall anyone else (other than you) saying that appsdE are not, in general, ready for VMSclusters.  IIRC, 2 or 3 other folksoC only mentioned that they have an application that won't.  And sincenD nobody every claimed that all applications would work transparently,E your tactic of using their voices to prove *your* "general" case is aR( demonstration of faulty logic (at best).   > A >> It is amazing how you can lose an argument, and then come backn6 >> weeks/months later and state the opposite.  sheesh! >> @ >n0 >Re read my posts, I have never said that it is / >impossible to construct a cluster environment y1 >that behaves as you suggest. I have however saide0 >that your case is not a general rule, sadly for7 >you this is also obvious from the postings disagreeingo' >with Kerrys origional string of posts.h  D No, sadly you never learned to read, or else you just can't follow aF discussion (take your Ritalin lately?).  What, in your humble opinion,D constitutes the "general rule"?  Huh?  100%?  Be careful.  Think forC awhile.  Your answer will color how all of your future comments areK viewed by this group.r  D Now, I think it reasonable to say that anything that is correct overA 80% of the time can safely be considered a "general rule".  And Ii> believe that you'll actually find over 90% of the applications: available will work as has been described in a VMScluster.   >  > A >> No, now you are twisting things again.  You have no ability toeH >> comprehend that all apps and all user populations don't behave as youD >> think.  I never said that *all* will, but I did say that the vast@ >> majority, and imho this is over 90% of the applications, willA >> definitely operate as I have described.  And I don't make thislB >> statement lightly, either.  I have seen many, many applicationsG >> (probably at least 1500) and have only seen rare instances of thingse8 >> that don't work as well as we'd like in a VMScluster. >> t >r: >Strange again that you could not get that "vast majority"> >concensus from other admins who also manage OpenVMS clusters.  B Wow!  Not only do you have reading problems, you also can't count.? Show me where - ANYWHERE - that you have  a "vast majority" whon@ support your opinion.  Go ahead, prove me wrong.  Count them andE report back.  Otherwise, shut the hell up.  You seem to be taking the F "silence" of other admins as tacit agreement with you.  That is both a? serious flaw in your ego, as well as a significant flaw in youre& understanding of statistical analysis.  F And what do you expect, anyway?  When Kerry or I tell you that we have@ managed many VMSclusters where complete OS & HW upgrades and areD performed with zero application downtime, do you think that it's notC the majority just because nobody else speaks up about it?  How elseoF could you do the data collection to make that statement?  I don't needB others to chime in with support to tell me what I'm doing works, I already know that.   >e >tF >> Now, unless you have suddently achieved over 15 years of experience< >> that disproves this, you have a lot of work ahead of you. >. >Again you missed the point. >r  @ No, you did.  But then, just take a look into *every* discussion= you're having in here.  In every single one you can't discussd< technical details for very long.  In *every* single one, the@ conversation degrades into an "I said, you said" stupid bunch of noise.    @ Get a clue, your tactics do not help your company...  oh wait, I@ almost forgot who you worked for.  The inventor of clusters, HA!   ------------------------------  # Date: Thu, 25 Jan 2001 06:18:44 GMTa  From: jlsue <jlsuexxxz@home.com>= Subject: Re: Compaq: A simple, affordable clustering solution 8 Message-ID: <tfdv6t8mp02mlegdfuik1qbc6165vb4jbv@4ax.com>  3 On Wed, 24 Jan 2001 16:51:14 +0000, andrew harrisonf! <andrew.nospam@uk.sun.com> wrote:y     > + >There were more than two dissenting voices-) >to Kerrys orgional postings citing their- >actual experiences.  D Let's see, exactly how many is "more than two"?  It doesn't disproveD the "general" rule unless the vast majority were found to have these	 problems.a  ) >There were also the previous examples ine( >particular the remedial actions post a * >duff patch from Compaq and the impact on ' >the users of that system. This example-& >also differed from Kerry's marketing % >possition and from jlsue's apparent s >practical experience. >n  C Nobody ever said that VMSclusters are always problem free, so this, @ again, doesn't prove anything.  Otherwise, by your reconing, SunE 10000s are nothing but crap, heck we can point to lots of comments inh here that say so.    ------------------------------  % Date: Thu, 25 Jan 2001 06:43:49 +0000 ) From: Christof Brass <brass@infopuls.com>i= Subject: Re: Compaq: A simple, affordable clustering solutionn, Message-ID: <3A6FCB25.7443DE45@infopuls.com>   andrew harrison wrote: > 2 > Yes of course you need a JVM but then if you are0 > running a browser anyway you probably could do1 > with a JVM as part of the browser to get accesss/ > to all those systems that use Java. It is oneM0 > of the things that people have been asking for1 > for OpenVMS hence the Mozilla plugin questions.p  ? You missed the point because the conditions were different. Theg? statement was that we don't need another desktop remote controls> protocol. You proposed Tar. and mentioned there were only 250k> of Java classes. I pointed out that this was an incomplete and= misleading contribution. Then you tried to justify this, i.e.t> not mentioning the JVM and the speed penalty, by introducing a: browser which is completely irrelevant with respect to our4 discussion. The browser isn't a replacement for X11.  2 > But why would you need a really fast system and,1 > Java in silicon etc etc. The only bit of Citrixo/ > and Tarantella that is written in Java is ther/ > client, neither of them consume huge ammountsh0 > of resource, I get very good Tarantella client- > performance using a 167 Mhz Ultra 1 and the 0 > Netscape 4.75 which hardly has the fastest JVM > built into it.  < I don't get it. I'm talking about the *client*. Why should I? otherwise have objected against your thin client statement? Whyn? should I waste time by using an inefficient solution? You don't-? know about the things I want to with the client machine. Pleasei@ name numbers. As I already emphasised serveral times: statements> like "very good" don't help in understanding the facts and are? useless for comparisons with other solutions. I doubt that yourr5 performance is in any way suitable to display severalb< application windows with rapidly changing contents. But as I said: we need numbers!  0 > The performance of the network and the backend0 > server X or RDP is has much more impact on the/ > performance I get on my client than the speedI, > of the JVM and the applet that runs in it.  7 I doubt that because there is plenty of computing power.= available. And anyway this is not especially relevant becausec? this is equally bad for all solutions, besides: what about Tar. = server performance? I really doubt that this double or triple0= tier protocol conversion solution is any better than a direct-; X11 connection. What do you think? Should we run a contest?   r3 > And if it really was a problem I could use eitherj. > the 1.2.X or 1.3 Java plugins instead of the- > standard JVM that built into netscape, bothl. > of them would improve the performance of the	 > applet.0  / Should we run a contest? Any numbers available?i  p> > > Figures and numbers please. As you may remember you have a@ > > unbeaten track record of unproven "facts". I need to know in@ > > what respect technically the Tarantelle adaptive protocol isA > > superiour over LBX. I need a short *commented* description ofbD > > the protocol features and how they are compared to X11. I need aD > > short overview of the bandwidth requirements for certain typical< > > operations as compared to X11. And, please, drop any SUN: > > marketing if you intend to answer these pure technical@ > > questions. And, please, no referring to vapor standards likeD > > J2EE. I'm only interested in existing solutions. And, please, no@ > > statements like "it simply works better than Y for this". It > > simply proves nothing. > > > Why on earth would I be marketing anything, Tarantella isn't; > a Sun product its a product developed and sold by SCO whoe$ > happen to be a competitor of ours.  9 Solaris is involved. I didn't mean that you would only do. marketing on Tarantella.  ! > If you want details on it go too > # > www.sco.com or www.tarantella.comt > . > You can even give it a test drive so why not2 > try it rather than blustering on about technical > questions. > 5 > Incedentally one of the demo sites is Compaq ROTFL.n  ? You missed the point. I was you who introduced this solution asd? a better alternative. But you didn't provide any facts to proveS? your statements. Why should I visit any site, find out that you < was wrong as I knew in advance, come back and telling you to@ only beeing told that I didn't look at the right place or didn't@ find the correct numbers? Why should I test anything which is by+ construction inferior to what I daily use?? ; BTW Compaq beeing able to show that even Tar. is running on ? their machines doesn't prove anything. There might environmentsr> where people want to use Tar. despite of the lousy performance> maybe because a person not playing with full set of cards gave? them this advice. I know a case where some Enterprise Architect>@ (BTW very affiliated with SUN) didn't give good advice to one of; it's customers despite the fact that he was payed a lot fore? helping the customer to implement technical superior solutions.   A > > > > Why do you ask for this proprietary M$ stuff like Windows-E > > > > Terminal Server? You will get a fat client which is obviouslyiF > > > > not the solution of the future. Stated that, what's wrong withC > > > > X11? As far as I know the other solutions (Windows Terminalw) > > > > Server, ICA Citrix, WinFrame) areu? > > > > re-invent-the-wheel-approaches. Do we really need that?p > , > This isn't technical in fact its is almost. > exactly what you accuse me of an untechnical > marketing rant.a > / > Where in this paragraph is a single technicalc > question ?????????????????  : Technical statement 1: Windows Terminal Server needs a fat client.t, Technical question 1: What's wrong with X11?; Technical statement 2: As far as I know the other solutionsD3 (Windows Terminal Server, ICA Citrix, WinFrame) areo re-invent-the-wheel-approaches.i- Technical question 2: Do we really need that?   
 Any comments?t@ As you might see if you can count on four or twice on two (which8 is four if you multiply) there isn't "a single technical; question" there are in fact two technical questions and two-= technical statements not leaving a word which wasn't in these  four technical items.h  ; > > Nice picture. Could you provide a matrix for the client A > > solutions? It would be interesting what the conditions are to D > > run the Tar. Java client or the Tar. native client. For the Java> > > client: is there a version which runs in a browser as JavaA > > applet? Which versions of the JVM and the browser plug-in areiA > > required for the different OSs and browers? For which OSs arebD > > Tar. native clients available? Is there a version for VMS?? What > > is the license policy?= > > For which server OSs is Tar. server available. Is there ao< > > version for VMS?? Is the Tar. source (client and server) > > available (like Solaris 8)?uB > > Last and simplest question: can the Tar. client and server run > > on the same machine?C > > This is your marketing chance! Good luck. And you better answers; > > one of the questions of availablity on VMS with yes ...V >  > -- > Andrew Harrisonn > Enterprise IT Architecte  = Did you really complete your answer. It looks like you misseds$ the most important part. Sad enough.   ------------------------------  # Date: Wed, 24 Jan 2001 21:25:18 GMTx From: john_20_28_2000@yahoo.comm4 Subject: Does vms support object method perl cgi.pm?) Message-ID: <94nh7i$s95$1@nnrp1.deja.com>   E Does Openvms 7.2-1 support the object method perl for cgi.pm?  When I  try:   use CGI; $query = new CGI;   D I get an error.  The error doesn't tell me much.  I can only get the  function method to work.  Thanks       Sent via Deja.coma http://www.deja.com/   ------------------------------  # Date: Wed, 24 Jan 2001 21:56:30 GMTk+ From: Craig A. Berry <calepine@my-deja.com>>8 Subject: Re: Does vms support object method perl cgi.pm?) Message-ID: <94nj26$u4i$1@nnrp1.deja.com>M  ) In article <94nh7i$s95$1@nnrp1.deja.com>,>"   john_20_28_2000@yahoo.com wrote:G > Does Openvms 7.2-1 support the object method perl for cgi.pm?  When It > try: >h
 > use CGI; > $query = new CGI;> > F > I get an error.  The error doesn't tell me much.  I can only get the" > function method to work.  Thanks  G This is not related to your VMS version but it might be related to your B Perl version.  What version do you have and what error do you get?  ? You might also check the archives of the VMS Perl mailing list:   3 <http://www.xray.mpe.mpg.de/mailing-lists/vmsperl/>      Sent via Deja.comi http://www.deja.com/   ------------------------------  % Date: Wed, 24 Jan 2001 18:17:23 -0500>" From: Dan Sugalski <dan@sidhe.org>8 Subject: Re: Does vms support object method perl cgi.pm?: Message-ID: <5.0.2.1.0.20010124181600.026e6870@24.8.96.48>  ; At 09:25 PM 1/24/01 +0000, john_20_28_2000@yahoo.com wrote: F >Does Openvms 7.2-1 support the object method perl for cgi.pm?  When I >try:h >e	 >use CGI;e >$query = new CGI; >'E >I get an error.  The error doesn't tell me much.  I can only get ther! >function method to work.  Thanksb  K The VMS version's pretty much irrelevant here, the more important thing is >C your perl version. Check the docs for the version of CGI.pm that's  I installed--you may find that you've got an older version that doesn't do wL what you need, and you have to upgrade. (Also, always, *always* include the I error message even if it doesn't tell you much--it may say more to other o people)e   					Dan  I --------------------------------------"it's like this"-------------------e2 Dan Sugalski                          even samurai? dan@sidhe.org                         have teddy bears and eveni;                                       teddy bears get drunkp   ------------------------------  % Date: Wed, 24 Jan 2001 12:43:48 -0500c% From: "Islandco" <sales@islandco.com>  Subject: DS10's incoming Usedt/ Message-ID: <t6u4gachs0sb08@corp.supernews.com>r  A We have some incoming in New condition BUT they are actually USEDr  5 We offer 12 months warranty on everything we sell !!!e  / Email with required configuration if interestede   David To   -- Island Computers US Corporationr 2700 Gregory Streett	 Suite 150g Savannah GA 314043 Tel: 912 447 6622v Fax: 912 201 0096d sales@islandco.com www.islandco.com  C This message and any files transmitted with it are confidential andhJ may be privileged and/or subject to the provisions of privacy legislation.H They are intended solely for the use of the individual or entity to whomE they are addressed. If the reader of this message is not the intendedp
 recipient,G please notify Island Computers US Corp immediately and then delete thist message.I You are notified that reliance on, disclosure of, distribution or copying  of this message is prohibited.   ------------------------------  % Date: Thu, 25 Jan 2001 14:28:59 +1030e% From: Jeremy Begg <jeremy@vsm.com.au>s/ Subject: Re: Dual ethernet config for failover?s* Message-ID: <3A70382B.6A5D1881@vsm.com.au>   Hello Matt,e   Matt Muggeridge wrote: > 5 > > 1. How does DNS resolve a name to 2 IP addresses?r > N > DNS uses round-robin when handing out addresses.  However, this doesn't giveG > you failover.  In fact, if one of the interfaces fails, then DNS willaI > dutifully continue to handout both interfaces in a round-robin fashion,lM > which will mean some users will have intermittent trouble connecting.  Thato > would be a pain. > H > To resolve this, you can add the load-broker and metric server to yourK > network.  The load-broker will poll the metric server, running on each of I > the participating nodes.  If the metric server fails to respond after 3eM > attempts, the load-broker will remove that interface from the DNS database.a- > This requires Dynamic DNS to be configured.   C Does the load broker have to run on the same machines as the metricv servers?H Could you have a machine outside a cluster monitoring the metric servers onH each cluster node and updating the DNS accordingly?  For that matter, do allsF the metric servers which are consulted to define the "A" records for a givenhE name have to be in the same cluster (or have to be clustered at all)?e   Thanks,            Jeremy Begg   =   +---------------------------------------------------------+y=   |            VSM Software Services Pty. Ltd.              | =   |                 http://www.vsm.com.au/                  | =   |       "OpenVMS Systems Management & Programming"        |a=   |---------------------------------------------------------|i=   | P.O.Box 402, Walkerville, |  E-Mail:  jeremy@vsm.com.au | =   | South Australia 5081      |   Phone:  +61 8 83592155    | =   |---------------------------|  Mobile:  0414 422 947      |a=   |  A.C.N. 068 409 156       |     FAX:  +61 8 82231777    |y=   +---------------------------------------------------------+c   ------------------------------  % Date: Wed, 24 Jan 2001 13:22:30 -0500e- From: JF Mezei <jfmezei.spamnot@videotron.ca>y+ Subject: Re: Filtering certain keys on VMS.i, Message-ID: <3A6F1D66.2B76BCBB@videotron.ca>  % Steve.Spires@yellowpages.co.uk wrote:t > N > Contact:   Tel: 3063  -  IS - Infrastructure, 1st Floor, Bridge Street Plaza > / > What? You mean like the Whitehouse method...?f  N No, that was too obvious. I was thinking of a more subtle method. Don't removeG te key, but get the CPU to drop any W it sees inputted from a terminal.0   ------------------------------  % Date: Wed, 24 Jan 2001 13:30:30 -0500a- From: JF Mezei <jfmezei.spamnot@videotron.ca>e+ Subject: Re: Filtering certain keys on VMS.r, Message-ID: <3A6F1F45.45A4D17E@videotron.ca>   Bob Koehler wrote:3 >   Yes.  But the White House solution is more fun.a  K It is far too obvious. Although looking for "W" keys in the rubbish bins orh6 trying to buy them at the store would be entertaining.  K I remember once playing a April 1 prank at university where we switched the H keyboards from one half of a large terminal room. So if you typed on theM keyboard, it would in fact appear on the screen next to your position insteaddL of your screen.  It took the comuter centre guys ages to figure it out. TheyK turned the 3270 controllers off and on, tested the coax cables etc etc, butrV never thought of making sure that the keyboards were plugged into the right terminals.  L It is only once enough people were in the room that someone noticed that her7 screen were "moving" without her touching the keyboard.   J So, having a not so obvious solution would have been far more interesting.   ------------------------------    Date: 25 Jan 2001 05:01:58 +0800, From: Paul Repacholi <prep@prep.synonet.com>+ Subject: Re: Filtering certain keys on VMS.a- Message-ID: <87d7dck6k9.fsf@prep.synonet.com>u  / JF Mezei <jfmezei.spamnot@videotron.ca> writes:o  ' > Steve.Spires@yellowpages.co.uk wrote:o > > P > > Contact:   Tel: 3063  -  IS - Infrastructure, 1st Floor, Bridge Street Plaza > > 1 > > What? You mean like the Whitehouse method...?  > P > No, that was too obvious. I was thinking of a more subtle method. Don't removeI > te key, but get the CPU to drop any W it sees inputted from a terminal.-  A Or patch the scan code to ascii table... Anyone know where it is?j   -- z< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.4@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Wed, 24 Jan 2001 13:17:26 -0700c% From: Dean Woodward <deanw@rdrop.com>e1 Subject: Re: Hobbyist help- DPWS 500au, VMS 7.2-1o) Message-ID: <3A6F3856.27186EAC@rdrop.com>h   "David J. Dachtera" wrote: >  > Dean Woodward wrote: > >w > > Hoff Hoffman wrote:m > > >eJ > > >   Did you UNLOAD and DISABLE the old PAKs?  (If you have a whole newK > > >   set of PAKs, you could recreate the whole license database, loadingrJ > > >   in the new PAKs, and reboot or otherwise reload the PAKs.  If not,I > > >   well, you will want to LICENSE UNLOAD and LICENSE DISABLE each ofs: > > >   the old PAKs, and then LICENSE LOAD the new PAKs.) > >SH > > 1) I did a LICENSE REMOVE * before loading new PAKs.  Shouldn't this > > UNLOAD/DISABLE the PAK?- > D > While I don't find a LICENSE REMOVE command in HELP or the commandG > tables (extracted CLD from DCLTABLES using VERB), LICENSE DELETE willrI > only expunge the LDB records - it does not (according to the HELP text) ) > unload the license if currently loaded.   ( Right- LICENSE DELETE * is what I meant.  I > > 2) I did a fresh install INITIALIZE rather than PRESERVE.  I expected K > > that would INIT the system disk, which is where I expect the license dbi4 > > to live... so I'd be starting from scratch, yes? > J > Yes. Next time, you can save yourself the trouble by just starting a new) > LDB with the new PAKs (LICENSE CREATE).w  G Yeah, but I wanted to clear my old crap off of it anyway (to start withnB fresh crap, of course).  Well, this time 'round, I booted from theH install CD, went to DCL, and did an INIT DKA0 /ERASE (just, you know, to make *sure* .;-)  F However, I believe somone here at work pointed out my fatal flaw.  TheC Hobbyist PAKs come formatted such that you can save them out as COMrD files and simply run them to install.  I pasted them together, FTP'd? them to the box, and ran it.  Two flaws:  I accidentally pastedtD OPENVMS-ALPHA in twice, and failed to add SET NOON at the top of theF file.  While I saw the error message indicating a duplicate, I assumedH it was dealing with UCX, and there's not much interesting after that for# a stand-alone box not running X.25.   F In any case, it's off the ground now.  Thanks to all for the pointers.   ------------------------------  % Date: Wed, 24 Jan 2001 15:35:47 -0600h* From: WILLIAM WEBB <WWEBB1@email.usps.gov>1 Subject: RE: Hobbyist help- DPWS 500au, VMS 7.2-1 - Message-ID: <0033000014385441000002L012*@MHS>   C =0AIn addition, some of the comment lines wrap at 80 characters andn< cause DCL errors to appear.  Nothing serious, just annoying.   WWWebb   -----Original Message-----/ From: Info-VAX-Request@Mvb.Saic.Com at INTERNETp) Sent: Wednesday, January 24, 2001 3:30 PMd6 To: Webb, William W; Info-VAX@Mvb.Saic.Com at INTERNET1 Subject: RE: Hobbyist help- DPWS 500au, VMS 7.2-1      "David J. Dachtera" wrote: >i > Dean Woodward wrote: > >  > > Hoff Hoffman wrote:  > > > H > > >   Did you UNLOAD and DISABLE the old PAKs?  (If you have a whole = newlH > > >   set of PAKs, you could recreate the whole license database, loa= dingH > > >   in the new PAKs, and reboot or otherwise reload the PAKs.  If n= ot,vH > > >   well, you will want to LICENSE UNLOAD and LICENSE DISABLE each = of: > > >   the old PAKs, and then LICENSE LOAD the new PAKs.) > >tH > > 1) I did a LICENSE REMOVE * before loading new PAKs.  Shouldn't thi= sm > > UNLOAD/DISABLE the PAK?. > D > While I don't find a LICENSE REMOVE command in HELP or the commandH > tables (extracted CLD from DCLTABLES using VERB), LICENSE DELETE will=  H > only expunge the LDB records - it does not (according to the HELP tex= t)) > unload the license if currently loaded.   ( Right- LICENSE DELETE * is what I meant.  H > > 2) I did a fresh install INITIALIZE rather than PRESERVE.  I expect= edH > > that would INIT the system disk, which is where I expect the licens= e db4 > > to live... so I'd be starting from scratch, yes? >dH > Yes. Next time, you can save yourself the trouble by just starting a = newp) > LDB with the new PAKs (LICENSE CREATE).r  H Yeah, but I wanted to clear my old crap off of it anyway (to start with=  B fresh crap, of course).  Well, this time 'round, I booted from theH install CD, went to DCL, and did an INIT DKA0 /ERASE (just, you know, t= ol make *sure* .;-)  F However, I believe somone here at work pointed out my fatal flaw.  TheC Hobbyist PAKs come formatted such that you can save them out as COM D files and simply run them to install.  I pasted them together, FTP'd? them to the box, and ran it.  Two flaws:  I accidentally pasted D OPENVMS-ALPHA in twice, and failed to add SET NOON at the top of theF file.  While I saw the error message indicating a duplicate, I assumedH it was dealing with UCX, and there's not much interesting after that fo= r # a stand-alone box not running X.25.o  G In any case, it's off the ground now.  Thanks to all for the pointers.=n   ------------------------------    Date: 25 Jan 2001 00:07:49 +0100) From: sfb@consultron.com (Stefan K. Berg)y5 Subject: Re: How to delete directory with bad blocks? 2 Message-ID: <m3y9w0k0qi.fsf@hermes.consultron.com>   Hi,e  H I'd just like to thank all who posted insightful comments to my problem.M I've enabled read and write verification and am running ANALYZE now and then. - Fortunately no more bad blocks has turned up.a  L I'm assuming the worst case scenario anyway and will be replacing the disk -H a RZ26 is on its way from the States to Sweden as I write this. The diskK itself is considerably cheaper than what I have to pay for the shipment. :)    Again, thanks for all help.e   /Stefan    ------------------------------    Date: 24 Jan 2001 13:42:07 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) ? Subject: Re: Is Compaq dropping support for vms on vax platforml+ Message-ID: <h9GR3PrKFKy9@eisner.decus.org>r  X In article <yEV$WYDLtkXB@eisner.decus.org>, young_r@eisner.decus.org (Rob Young) writes:J > In article <94lfu7$gpp$1@bob.news.rcn.net>, "me" <im@not.saying> writes:O >> Maybe where you work a $20,000 workstation is a three year upgrade but where I >> I work that is the computer we will probably die with.  I, in 2001, ama7 >> assembling a VAX/VMS machine right now to run XDada.l >> a > B > 	Shame ain't it?  Considering you can probably buy several times? > 	the computing power now for $800 and get a printer and modemc% > 	to boot.  Times they a changin'!!!i  ? But if your goal is to run XDada, you won't get that on Wintel.h  N ==============================================================================N Great Inventors of our time: Al Gore -> Internet; Sun Microsystems -> ClustersN ==============================================================================   ------------------------------   Date: 24 Jan 2001 20:27:28 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)"- Subject: Re: Linux worm and RedHat 7.0 broken + Message-ID: <94ndrg$mao$1@info.cs.uofs.edu>t  L In article <rdeininger-2401010135070001@user-2iveaup.dialup.mindspring.com>,5  rdeininger@mindspring.com (Robert Deininger) writes:n |>   |> uA |> Another one:  Why are null-terminated strings a good design?     @ Who said Unix invented them??  Have you ever looked at .ASCIZ inG Macro-11??  Or are you saying that DEC stole this from Bell Labs??  :-)c   |> Why not counted strings?  w  ? Overhead for one.  With a the same overhead a counted string isoI limited to 255 characters while a NULL terminated string is theoretically E infinite.  The longer you want your strings to be, the more overhead..B Probably not important today, but remember the capabilities of theD machines at the time NULL terminated strings came into being.  Hmmm.D Maybe that's why DEC came up with .ASCIZ at about the same time Unix+ was coming up with NULL terminated strings.c  E |> And if null-terminated strings are so great, why does so much unixlI |> code seem so reluctant to validate string arguments before using them?h  H Can't address this as I don't see what the one has to do with the other.  E |> Was this due to "planning, debate, and design" or just sloppiness?a  D My guess would be sloppiness.  And we all now there has never been a* sloppy programmer working on a VMS system.   |>  C |> Which is a better design: a set of system parameters that can beo? |> modified, plus tools to modify them sensibly, or editing and G |> recompiling the kernel to make "simple" changes in OS configuration?   B Matter of opinion, I guess.  I find it infinitely easier to modifyC a Unix kernel than a VMS one.  As a matter of fact, I have recentlyeB done some work with modified Ultrix-11 kernels and I am doing thisC with no documentation at all.  I can't do the same on VMS even with  a shelf full of manuals.   |> eE |> Is it good design that programs compiled and linked many years and0J |> versions ago will run just fine, unchanged, on current hardware and OS?  E And what OS might this be??  Surely your not talking about VMS??  TheeF hardware changes every few years making this impossible even for that.E Depending ont he triviality of the program, the same is true for most G Unix versions as well.  Add into that the cross-Unix compatability that F has been built into most of them today.  FreeBSD running SCO binaries,E Linux running Solaris binaries.  I could probably test it to be sure,pB but I would bet that an Ultrix-32 3.0 binary (without any builtin D kernel dependancies) would run just fine on Ultrix-32 4.3.  I know IB never had to re-compile any of my locally generated programs after upgrading the OS.e    -5 |> Is there a unix hiding somewhere that can do this?t  4 Just about any of them that I have ever worked with.   |> e |> I better not go on... |> eD |> > I'm just curious, how much have your programmed for Unix vs VMS< |> > (sometimes familiarity with one system skews judgement) |>  : |> Without a doubt, this skewing works in both directions.  E Maybe, but I spend a lot of time on Unix groups as I have to maintain @ more Unix Servers than VMS servers and I can assure you the UnixD advocates don't devote .01% of the time slaming VMS or any none Unix0 OS that the VMS people here spend slamming Unix.  C |> How much have you programmed VMS?  Which pieces do you feel havetD |> an "unclean" design, and how well do you understand those pieces?  -C I, for one, don't think any of it is unclean.  Only different.  ButaC I also don't spend any time trying to convince people that it is no @ good.  I don't prefer it, but I use it when it is the right tool for a particular job.  e  ? And they way Compaq is going, even that won't be an option mucha longer.:  C By the way, as much as I like Unix, I still like RSTS/E, too, whichn, is why I am still so familiar with Macro-11.   bill   -- rJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   e   ------------------------------  % Date: Wed, 24 Jan 2001 21:42:12 -050022 From: rdeininger@mindspring.com (Robert Deininger)- Subject: Re: Linux worm and RedHat 7.0 brokeneL Message-ID: <rdeininger-2401012142130001@user-2ive7g7.dialup.mindspring.com>  E In article <94mvhe$ag0$1@nnrp1.deja.com>, sabolich@my-deja.com wrote:o  N > In article <rdeininger-2401010135070001@user-2iveaup.dialup.mindspring.com>,7 >   rdeininger@mindspring.com (Robert Deininger) wrote:pI > > In article <94l2pm$olm$1@nnrp1.deja.com>, sabolich@my-deja.com wrote:e > >dN > > > VMS is well engineered.  It is reliable, stable, etc.  Its design is notL > > > clean though.  It is a commercial and prorietary OS - with such thingsM > > > the design is often rushed to get a working product out the door before  > > > the competition. > >1G > > I'm sure there was pressure to ship VMS as quickly as possible, buteH > the _design_ was not rushed.  Design and implementation were separatedH > by quite a bit more than a slice of pizza and a pepsi.  Quite a lot ofF > the early history of VMS is accessible, and the design is documentedH > quite nicely in the books titled ~"VMS Internals and Data Structures".J > Various versions, and slightly different titles are available.  I highly > recommend this book. > >y? > > You mention "commercial" and "proprietary" as if those were J > automatically bad things.  Perhaps, but they have nothing to do with the > design of the system.m > I > Not automatically bad things.  I meant in general that when there is no J > time to market pressure (such as in academic environment) initial designC > is less rushed and often openly debated.  This does not mean thatsH > _every_ open project is well thought out and designed and every closedF > one is slopped together kludge.  There are examples of excellent and> > poor design from both.  I'm just saying there is a tendency. >  > >dJ > > I also disagree when you say the design of VMS is not clean.  It seemsJ > quite clean and regular to me, though in 20ish years, a little cruft has > crept in here and there. > H > Ok, maybee my statement was a little harsh on VMS.  I'll say that Unix > is cleaner than VMS.   I'll be wanting some examples.    /E > A lot of work was done at Berkley - therefore the name BSD, BerkleynE > Software Distribution.  As far as design, if Unix being simpler and-I > smaller than VMS, can do everything that VMS can do (with less or equaloH > work on the part of the programmer) would you call it better designed?   Hook, line and sinker!   I dispute your claim that Unix can do what VMS can do with less or equal effort.  That has not been my experience.  Both can do simple things with about equal effort; sometimes the "simple" approach of unix actually makes it easier, for _simple_ jobs.o    J > > > Anyways VMS and Unix are similar in there abilities.  Both are multiH > > > user, multi tasking OSes with virtual memory protection and demandM > > > paging.  You can accomplish the same results with both OSes.  Just thata4 > > > Unix does it with fewer and simpler API calls. > >oF > > I don't think you are very familiar with VMS and its capabilities.H > (And I am certainly not up-to-date on unix APIs these 20 years or so.)H > There is a LOT built into VMS that makes applications easier to write.D > Unix has fewer and simpler API calls because it is ... smaller and5 > simpler.  Basic unix doesn't have anything to match5 >  > > RMS,C > Is this a king of data base that for example the file system runsiD > through where a block is a record?  What is the advantage of this?   It is a file system with a very rich set of features.  While my little list isn't pure "RMS" (the boundaries are a little blury), I'll rattle off a few features:a, -- Simple structureless files like unix has.- -- A variety of record-oriented file formats.tF -- Indexed files, with a rich set of index and key features, and toolsI    to manage the files even when they are quite large and complex.  Often I    used as a poor man's database, removing the need for the "big hammer"  
    databases.t -- Multiple versions of files.H -- Comprehensive support for stuff like file name parsing and searching,G    built into the OS, available to programmers at many levels, and used H    by virtually all 1st-, 2nd- and 3rd-party software.  This contributesL    to a rather uniform user (and programmer) environment.  I believe it also    contributes to reliability.I -- Almost all of this is seamlessly integrated with clustering and DECnetsG    networks.  Programs typically don't need to be modified to work in aiH    clustered or networked environment, and users can hardly tell either.  J I'm only scratching the surface.  I can't teach much VMS in a post or two.  In unix, file-oriented tasks tend to be built on an as-needed basis, which appears to cost programmers extra time.  On the other hand, unix programmers don't tend to DO as much with files as VMS programmers, so maybe they don't miss the features they don't even realize exist.     > > or the DLM, or clustering,H > Distributed locks and clustering.  Linux doesn't have this as far as II > know.  Other high end proprietary Unixes might have something like thisWA > but I cannot comment because I don't have experience with them.   &"Something like this" is a BIG stretch.  DLM is available to programmers whenever they need it, and is comprehensive.  Again, programs don't need to be changed at all, whether locking is local or cluster-wide.  Read the book, "In search of clusters".  VMS has some unique features in this area.   > > or Galaxy,I > Virtual machines.  Some systems have this in hardware - for exmaple IBM G > s/390 which can run over 1000 instances of Linux on a single machine. I > Also VMware for x86 which allow multiple OSes (NT, Linux, FreeBSD, etc) & > to run simultaneously.  Also Plex86.   You've almost scratched the surface of Galaxy.  Instances in Galaxy can interact in interesting ways.  They needn't be just N non-interacing copies of the same thing.  IBM does have some interesting features also.    > > or logical names,V6 > Combination of Environment variables and soft links.  And a lot more.  Symbols are a close match for environment variables.  Soft links are rather different beasts than logical names.  Logical names have recently gone cluster-wide as well.  Maybe you should look them up in the docs, and also look at some examples of use.  ) > >  or simple mixed-language support ...   
 No comment?  h    I > > When these are available for unix, they are add-ons, and often poorlytG > integrated with the OS.  Things like this were designed into VMS from  > very early days. > >hF > > Speaking of "good design", why was unix designed with devices thatH > pretend to be files?  That is either a symptom or a contributing causeH > of a lot of silliness and pain in unix device support.  Simple, maybe. > Good design, I doubt.w > C > 'Everything is a file' is good.  This means that a single form ofoG > security and access control (the same one used for files) is used forv
 > devices.   But devices aren't files!  The security model may (may!) be adequate, but devices require more!  This model causes trouble.  I've seen systems (4 or 5 years ago) where a tape drive was represented by 8 different files.  Each combination of 3 binary features (compression on or off; density low or high; unload or don't unload on rewind) needed a separate file.  Good thing the drives didn't have more features.  As for security, there was little.  One user could trash another's job by accessing a different file   My point is, devices need different semantics than files.  A file has open, close, read, write, and so on.  Even a simple device needs much more.  Providing  it in the "everything is a file" model is clumsy, not clean.  IMNSHO.c    I > > Another one:  Why are null-terminated strings a good design?  Why not I > counted strings?  And if null-terminated strings are so great, why doestI > so much unix code seem so reluctant to validate string arguments before;E > using them?  Was this due to "planning, debate, and design" or just0
 > sloppiness?  > G > This could easily be C vs Fortran.  Why does Fortran use CR and not ;fJ > for end of line.  The line continuation scheme is quite stupid and makesG > code look unreadable.  I'm not saying that Unix is bettery in _every_l
 > respect.  zFortran has fixed its continuation problem.  Unix remains broken.  I agree, it could be blamed on C.  Shall we say C was not well designed?  Or shall we blame the "designers" of unix for choosing C, and not fixing its faults?  Maybe those academicians back in the sixties spent too much time looking out the window at the coeds, and too little on "planning, debate, and design".    D > > Which is a better design: a set of system parameters that can beJ > modified, plus tools to modify them sensibly, or editing and recompiling: > the kernel to make "simple" changes in OS configuration? > B > Linux is often recompiled because the source is available.  ManyE > parameters can be changed without recompile or even without reboot.eF > I'll go out on a limb and claim I can change more paramters in LinuxH > without reboot than you can in OpenVMS (I may get burned here but I'll > take the challenge).   Well, it all depends on what we call a parameter.  Shall we count the OS parameters?  Include RMS tuning?  X-windows?  Probably a meaningless exercise.  I'm willing to skip it.    F > > Is it good design that programs compiled and linked many years andI > versions ago will run just fine, unchanged, on current hardware and OS?e > I > Yes, as long as you are on the same CPU type.  Can you run VAX binariesw > on Alpha?i  No. (Unless I translate them.)  But I can run ancient VAX binaries on a new VAX, or a new version of VMS.  Ditto for alpha.  VMS has a long enough history to make this important.  But it also points out the quality of the design, both hardware architecture and software.  W@ > I programmed multi tasking OSes in this order - NT, VMS, Unix. > J > To be fair I spent the least amount of time with VMS (it was a while agoG > too) but I was exposed to it before Unix.  At that time I though Unix G > was cryptic and backwards.  Then I started playing around with Linux.aI > At first I didn't like it, but the more I used/programmed it the more I 0 > liked it.  It is now my favorite of the three.  7 I guess linux is better than plain unixes in some ways.    The VMS process/task model is somewhat different than unix's.  Neither maps perfectly into the other, and trying to force them to has caused much gnashing of teeth.   -- o Robert Deininger rdeininger@mindspring.comi   ------------------------------  % Date: Wed, 24 Jan 2001 22:05:03 -0500i2 From: rdeininger@mindspring.com (Robert Deininger)- Subject: Re: Linux worm and RedHat 7.0 brokenaL Message-ID: <rdeininger-2401012205040001@user-2ive7g7.dialup.mindspring.com>  G In article <94ndrg$mao$1@info.cs.uofs.edu>, bill@cs.scranton.edu wrote:5  N > In article <rdeininger-2401010135070001@user-2iveaup.dialup.mindspring.com>,7 >  rdeininger@mindspring.com (Robert Deininger) writes:D > |> 3 > |> @C > |> Another one:  Why are null-terminated strings a good design?  n > B > Who said Unix invented them??  Have you ever looked at .ASCIZ inI > Macro-11??  Or are you saying that DEC stole this from Bell Labs??  :-),  a None of the above.  I wonder why such a well-designed OS makes such heavy use of an icky feature.i   > |> Why not counted strings?  l > A > Overhead for one.  With a the same overhead a counted string istK > limited to 255 characters while a NULL terminated string is theoreticallynG > infinite.  The longer you want your strings to be, the more overhead.iD > Probably not important today, but remember the capabilities of theF > machines at the time NULL terminated strings came into being.  Hmmm.F > Maybe that's why DEC came up with .ASCIZ at about the same time Unix- > was coming up with NULL terminated strings.=  9 And "theoretically infinte" strings are a good thing? :-)-  There are many, many instances where you need to know how long a string is. That extra byte or two makes this operation very cheap when the string is counted.  On the other hand, if you don't worry about size of your string compared to the size of your buffer, it _might_ not matter.    I may be wrong, but I think .ASCIZ came over from older DEC platforms.  It isn't very widely used.  But thank you for reminding me about .ASCIZ strings.  VMS tends to offer the right tool for the job.  Unix wants the job to fit the sparse menu of tools.m  G > |> And if null-terminated strings are so great, why does so much unixWK > |> code seem so reluctant to validate string arguments before using them?  > J > Can't address this as I don't see what the one has to do with the other.   It goes to quality of design.  Some engineer somewhere realized length(string) is an important operation, and arranged to support it easily. Perhaps this decision has contributed to overall quality on VMS.A  G > |> Was this due to "planning, debate, and design" or just sloppiness?s > F > My guess would be sloppiness.  And we all now there has never been a, > sloppy programmer working on a VMS system.   Hmm.  Neither is perfect, so we should pretend they are of equal quality?  I do think there are meaningful comparisons to be made.  I listed a few off the top of my head.  Maybe you prefer other points of comparison.    E > |> Which is a better design: a set of system parameters that can be.A > |> modified, plus tools to modify them sensibly, or editing andhI > |> recompiling the kernel to make "simple" changes in OS configuration?  > D > Matter of opinion, I guess.  I find it infinitely easier to modifyE > a Unix kernel than a VMS one.  As a matter of fact, I have recently D > done some work with modified Ultrix-11 kernels and I am doing thisE > with no documentation at all.  I can't do the same on VMS even withh > a shelf full of manuals.  K New glasses, maybe?  :-)  Maybe you aren't really trying to understand VMS.t   It probably is easier to modify a unix kernel.  But to modify it correctly?  I dunno.  We should also consider why it NEEDS to be modified so darned much.  Good design?     G > |> Is it good design that programs compiled and linked many years andnL > |> versions ago will run just fine, unchanged, on current hardware and OS? > D > And what OS might this be??  Surely your not talking about VMS??     Of course I am.  Obviously, a program won't do well on a platform that lacks hardware it needs.  User-mode, hardware-independent code in VMS is VERY portable over time.  That was a DESIGN decision, not an accident.   > ThenH > hardware changes every few years making this impossible even for that.G > Depending ont he triviality of the program, the same is true for mostrI > Unix versions as well.  Add into that the cross-Unix compatability that H > has been built into most of them today.  FreeBSD running SCO binaries,! > Linux running Solaris binaries.I    Does Solaris run Linux binaries?  O My most recnet experience was with HPUX, and upgrades required rebuilding code.m     7 > |> Is there a unix hiding somewhere that can do this?c > 6 > Just about any of them that I have ever worked with.  N Hmm.  What's the oldest working binary on your favorite system?  Early 1980's?     G > Maybe, but I spend a lot of time on Unix groups as I have to maintaintB > more Unix Servers than VMS servers and I can assure you the UnixF > advocates don't devote .01% of the time slaming VMS or any none Unix2 > OS that the VMS people here spend slamming Unix.   Well, I've NEVER gone to a unix group to bash unix.  Why do you unix folks come over here and say thing like "VMS's design isn't clean".  (Not you, I realize.)t   My primary aim is to counter some of the NONSENSE about VMS that I found in this thread.  I tend to assume ignorant nonsense can be cured by removing the ignorance.  Malicious nonsense is another matter.o   -- n Robert Deininger rdeininger@mindspring.com.   ------------------------------  # Date: Thu, 25 Jan 2001 05:10:00 GMTe From: sabolich@my-deja.com- Subject: Re: Linux worm and RedHat 7.0 brokens) Message-ID: <94ocf5$k4l$1@nnrp1.deja.com>c  I >> Ok, maybee my statement was a little harsh on VMS.  I'll say that Unix- > > is cleaner than VMS. >   > I'll be wanting some examples.  B Please see my example in reply to Christof Brass comparing fork(),F execl() unix style of process creation with VMS.  By the way, does VMS have anything like fork()?  D I consider the Unix way cleaner, while you may not.  It actually mayF come down to personal preference.  I consider simplicy and flexibility to be clean.   > G > > A lot of work was done at Berkley - therefore the name BSD, Berkley<G > > Software Distribution.  As far as design, if Unix being simpler andMK > > smaller than VMS, can do everything that VMS can do (with less or equal J > > work on the part of the programmer) would you call it better designed? >h > Hook, line and sinker! >eD > I dispute your claim that Unix can do what VMS can do with less orC equal effort.  That has not been my experience.  Both can do simple G things with about equal effort; sometimes the "simple" approach of unixr, actually makes it easier, for _simple_ jobs.  G Unix way is often to have a complex job be a set of simple jobs.  So if E unix approach makes it easier for simple jobs it also makes it easierI@ for a collection of simple jobs which do something more complex.  J > > There is a LOT built into VMS that makes applications easier to write.F > > Unix has fewer and simpler API calls because it is ... smaller and7 > > simpler.  Basic unix doesn't have anything to matchh > > 
 > > > RMS,E > > Is this a king of data base that for example the file system runshF > > through where a block is a record?  What is the advantage of this? >nH > It is a file system with a very rich set of features.  While my littleF list isn't pure "RMS" (the boundaries are a little blury), I'll rattle off a few features: . > -- Simple structureless files like unix has.  C One could argue that this is enough.  If you try to do more you aresH trying to anticipate the needs of the programmer and may include a bunchB of features he does not need and miss the feature he really needs.  / > -- A variety of record-oriented file formats.nH > -- Indexed files, with a rich set of index and key features, and toolsK >    to manage the files even when they are quite large and complex.  Often J >    used as a poor man's database, removing the need for the "big hammer" >    databases.o  H Yeah but then if you do use the 'big hammer' all these features sit idle and just take up space.d    > -- Multiple versions of files.  @ Agree here - I like this.  Doh, didn't mean to delete that file!  J > -- Comprehensive support for stuff like file name parsing and searching,I >    built into the OS, available to programmers at many levels, and usedp  G This is kind of general and I don't know exactly the scope of what thisa- encompasses.  Give some examples if you wish.   J >    by virtually all 1st-, 2nd- and 3rd-party software.  This contributesF >    to a rather uniform user (and programmer) environment.  I believe it alsoa  >    contributes to reliability.K > -- Almost all of this is seamlessly integrated with clustering and DECneteI >    networks.  Programs typically don't need to be modified to work in asJ >    clustered or networked environment, and users can hardly tell either. >sL > I'm only scratching the surface.  I can't teach much VMS in a post or two. >.F > In unix, file-oriented tasks tend to be built on an as-needed basis,F which appears to cost programmers extra time.  On the other hand, unixF programmers don't tend to DO as much with files as VMS programmers, soA maybe they don't miss the features they don't even realize exist.   C Let me give an example here.  In a place I work at often there is alH Alpha cluster running VMS.  I talk with the VMS guy often (he has a good* amount of experience) sometimes about VMS.  E There is some data acquisition software running there that makes datadF available to PCs via FTP server.  Because of FTP the data is stored inG simple files -- obviously the wrong way to do it on VMS.  The number ofu@ files in the directory climbs into the thousands (maybee tens ofE thousands).  When it is time to purge the data (erase the files) thissC takes a very long time (hours).  Also it wastes a lot of disk space H (since the files are small) and causes fragmentation.  Obviously this isH the wrong way to do it on VMS.  They should rewrite the data acquisitionH to store using variable length records (an Indexed file under RMS).  Ok,D now they cannot use the generic FTP server and would need to write a custom one.g  H Now suppose we do this with Linux, we'll choose the ReiserFS filesystem.H  First off it doesn't waste space - only disk space used is size of dataF + metadata.  Second it has no problem with large numbers of files 100KG in a directory is not a problem.  I did a test on a k6-450 machine withsD 64meg ram running x-windows.  It took about 4-5 seconds to create(),C write(), close() 10000 5Kilobyte files.  It took under 2 seconds toSF delete them all. This means we can use simple files and do not need to write an FTP server.  A So in essence we don't need anything more than simple files.  Thed. filesystem itself is our 'poor man's database'   >   > > > or the DLM, or clustering,J > > Distributed locks and clustering.  Linux doesn't have this as far as IK > > know.  Other high end proprietary Unixes might have something like thisrC > > but I cannot comment because I don't have experience with them.d >t> > "Something like this" is a BIG stretch.  DLM is available to@ programmers whenever they need it, and is comprehensive.  Again,E programs don't need to be changed at all, whether locking is local oryD cluster-wide.  Read the book, "In search of clusters".  VMS has some unique features in this area.   4 As I said I don't know enough about this to comment.   >e > > > or Galaxy,K > > Virtual machines.  Some systems have this in hardware - for exmaple IBMcI > > s/390 which can run over 1000 instances of Linux on a single machine. K > > Also VMware for x86 which allow multiple OSes (NT, Linux, FreeBSD, etc)o( > > to run simultaneously.  Also Plex86. > E > You've almost scratched the surface of Galaxy.  Instances in GalaxyiH can interact in interesting ways.  They needn't be just N non-interacingH copies of the same thing.  IBM does have some interesting features also.  H VMware or Plex86 don't have to be just N of the same thing.  They can beK completely different OSes at the same time (Like Linux + NT simultaneously)l   >g > > > or logical names,e8 > > Combination of Environment variables and soft links. >dG > And a lot more.  Symbols are a close match for environment variables.eD  Soft links are rather different beasts than logical names.  LogicalE names have recently gone cluster-wide as well.  Maybe you should looki; them up in the docs, and also look at some examples of use.a  E OK, but what can you accomplish with them that you cannot do in Unix?d   >g+ > > >  or simple mixed-language support ...  >d
 > No comment?h >aK > > > When these are available for unix, they are add-ons, and often poorly I > > integrated with the OS.  Things like this were designed into VMS froma > > very early days. > > >iH > > > Speaking of "good design", why was unix designed with devices thatJ > > pretend to be files?  That is either a symptom or a contributing causeJ > > of a lot of silliness and pain in unix device support.  Simple, maybe. > > Good design, I doubt.w > >nE > > 'Everything is a file' is good.  This means that a single form of.I > > security and access control (the same one used for files) is used for  > > devices. >rG > But devices aren't files!  The security model may (may!) be adequate,fH but devices require more!  This model causes trouble.  I've seen systemsD (4 or 5 years ago) where a tape drive was represented by 8 differentE files.  Each combination of 3 binary features (compression on or off;tH density low or high; unload or don't unload on rewind) needed a separate? file.  Good thing the drives didn't have more features.  As foruB security, there was little.  One user could trash another's job byA accessing a different file representing the same drive.  And when E conflicting commands went to the drive from multiple sources, it justfF did them all.  No errors were reported.  Maybe this kind of idiocy hasK been papered over in the meantime.  "Simplicty" taken way too far, I think.l > G > My point is, devices need different semantics than files.  A file haskE open, close, read, write, and so on.  Even a simple device needs muchiG more.  Providing  it in the "everything is a file" model is clumsy, notf clean.  IMNSHO.   H Well these faults you mentioned are a problem with that particular Unix,C not the device=file model.  In addition to open() close(), etc. youeG forgot ioctl().  For things like serial ports you use it for baud rate, G parity, etc (one file per port).  For a cdrom for eject, etc.  For tapec> rewind, etc.  For sound card - sampling rate, bits per sample,F mono/stereo, etc.  There is no need to have multiple files per device.   > K > > > Another one:  Why are null-terminated strings a good design?  Why notlK > > counted strings?  And if null-terminated strings are so great, why doeseK > > so much unix code seem so reluctant to validate string arguments beforelG > > using them?  Was this due to "planning, debate, and design" or justn > > sloppiness?r > >nI > > This could easily be C vs Fortran.  Why does Fortran use CR and not ;gL > > for end of line.  The line continuation scheme is quite stupid and makesI > > code look unreadable.  I'm not saying that Unix is bettery in _every_r > > respect. > F > Fortran has fixed its continuation problem.  Unix remains broken.  IF agree, it could be blamed on C.  Shall we say C was not well designed?H Or shall we blame the "designers" of unix for choosing C, and not fixing its faults?nE >  Maybe those academicians back in the sixties spent too much time >yB >looking out the window at the coeds, and too little on "planning, >debate, and design".O  C "looking out the window at coeds". Please, this is not a flame war.a  Let's keep it from becoming one.   >aF > > > Which is a better design: a set of system parameters that can beL > > modified, plus tools to modify them sensibly, or editing and recompiling< > > the kernel to make "simple" changes in OS configuration? > >eD > > Linux is often recompiled because the source is available.  ManyG > > parameters can be changed without recompile or even without reboot.mH > > I'll go out on a limb and claim I can change more paramters in LinuxJ > > without reboot than you can in OpenVMS (I may get burned here but I'll > > take the challenge). >rG > Well, it all depends on what we call a parameter.  Shall we count the0G OS parameters?  Include RMS tuning?  X-windows?  Probably a meaninglessu" exercise.  I'm willing to skip it.  E You asserted that you need to change kernel source code and recompilenH the kernel for "simple" changes in OS configuration.  Do you still stand by this?   > H > > > Is it good design that programs compiled and linked many years andK > > versions ago will run just fine, unchanged, on current hardware and OS?> > > K > > Yes, as long as you are on the same CPU type.  Can you run VAX binaries 
 > > on Alpha?t >tG > No. (Unless I translate them.)  But I can run ancient VAX binaries on E a new VAX, or a new version of VMS.  Ditto for alpha.  VMS has a longdB enough history to make this important.  But it also points out the? quality of the design, both hardware architecture and software.   A What your telling me is that you can run the same binaries acrossaF different VMS versions and VAXes provided that the CPU instruction set is the same.  G Same for Linux -- same binary will run on kernel 2.0.x, 2.2.x, 2.4.x on  386, 486, 586 686.   >tB > > I programmed multi tasking OSes in this order - NT, VMS, Unix. > >gL > > To be fair I spent the least amount of time with VMS (it was a while agoI > > too) but I was exposed to it before Unix.  At that time I though UnixfI > > was cryptic and backwards.  Then I started playing around with Linux.eK > > At first I didn't like it, but the more I used/programmed it the more Ir2 > > liked it.  It is now my favorite of the three. >h9 > I guess linux is better than plain unixes in some ways.r  H Well I cannot have (won't spend the money) for a SGI to try IRIX, rs6000# for AIX, etc.  What's a plain Unix?    >lH > The VMS process/task model is somewhat different than unix's.  NeitherE maps perfectly into the other, and trying to force them to has causedw much gnashing of teeth.p   >m > -- > Robert Deininger > rdeininger@mindspring.comr >o     Sent via Deja.comh http://www.deja.com/   ------------------------------  % Date: Mon, 22 Jan 2001 20:12:37 -0800 . From: terrence golden <terrygolden@brandx.net>. Subject: Re: Lowercase filenames on VMS V7.2-1* Message-ID: <3A6D04B4.FCDFC57A@brandx.net>  $ reformat your disk as ods-2 volumes.   Rob Reffner wrote:  D > I have encounter a situation where lowercase filenames are causingB > problems. I was not aware that VMS allowed this but I have found@ > several. Is there a way to prevent this and what is the cause? >  > Thnaks in advance, >y > Robm   ------------------------------  # Date: Wed, 24 Jan 2001 22:09:54 GMTe2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)) Subject: Re: Memo:  Help! Zip/Unzip errort5 Message-ID: <SoIb6.92$cu.893@gazette.loc1.tandem.com>r  n In article <802569DE.0049541C.00@emea-smtp-03.systems.uk.hsbc>, Paul BEAUDOIN <paul.beaudoin@hsbc.com> writes: :Gents,l  @   Somewhat presumptuous of your potential audience genetics, eh?  F :I am having a problem with both ZIP and UNZIP on a partcular machine.F :The executables were coppied from a 7.2 system where they worked to a7 :number of 7.2-1 (DS10/DS20) systems where they didn't.e  D   I'd encourage an upgrade to more recent versions of ZIP -- see theH   OpenVMS FAQ for pointers to the newer stuff.  (The pre-built versions G   on Freeware V4 have problems.  The versions likely to be found on thei&   Freeware V5 kit are listed below...)   : The ZIP version: :kG :Copyright (C) 1990-1994 Mark Adler, Richard B. Wales, Jean-loup Gaillyy= :and Igor Mandrichenko. Type 'zip "-L"' for software license. < :Zip 2.0j (Sept 16th 1994). Usage: zip=="$disk:[dir]zip.exe"     I'm presently working with:r  /     UnZip 5.41 of 16 April 2000, by Info-ZIP...?     andn#     Zip 2.3 (November 29th 1999)...r  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Thu, 25 Jan 2001 05:50:38 GMTn From: Dirk Munk <munk@home.nl>) Subject: Re: Memo:  Help! Zip/Unzip errori' Message-ID: <3A6FBEAD.CF0CC4E3@home.nl>i  ? AFAIK Unzip 5.42 is the latest version (has just been released)    Hoff Hoffman wrote:   p > In article <802569DE.0049541C.00@emea-smtp-03.systems.uk.hsbc>, Paul BEAUDOIN <paul.beaudoin@hsbc.com> writes:	 > :Gents,o >iB >   Somewhat presumptuous of your potential audience genetics, eh? >rH > :I am having a problem with both ZIP and UNZIP on a partcular machine.H > :The executables were coppied from a 7.2 system where they worked to a9 > :number of 7.2-1 (DS10/DS20) systems where they didn't.d > F >   I'd encourage an upgrade to more recent versions of ZIP -- see theI >   OpenVMS FAQ for pointers to the newer stuff.  (The pre-built versions I >   on Freeware V4 have problems.  The versions likely to be found on thee( >   Freeware V5 kit are listed below...) >  > : The ZIP version: > :pI > :Copyright (C) 1990-1994 Mark Adler, Richard B. Wales, Jean-loup Gaillys? > :and Igor Mandrichenko. Type 'zip "-L"' for software license."> > :Zip 2.0j (Sept 16th 1994). Usage: zip=="$disk:[dir]zip.exe" >m >   I'm presently working with:S >b1 >     UnZip 5.41 of 16 April 2000, by Info-ZIP...u	 >     and % >     Zip 2.3 (November 29th 1999)...g >aP >  --------------------------- pure personal opinion ---------------------------N >    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 24 Jan 2001 17:27:48 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon).6 Subject: Re: Moving from VMS to Tru64 Unix, Any Ideas?+ Message-ID: <94n3ak$go7$1@info.cs.uofs.edu>n  + In article <949p59$o9a$1@bob.news.rcn.net>,e!  ddellutr@XXXenteract.com writes:y |>H |> There is a book entitled "OpenVMS and Tru64 UNIX Interoperability andF |> Migration Guide", (no author given), Order number: 1118-1199A-WWEE,I |> from Compaq Computer Corp.  I got it free when I attended a seminar inr3 |> Chicago; you might be able to get it free, also.y  @ I wouldn't put much stock in this.  I have asked (from someplace@ on the Compaq webpage that I cna't seem to find anymore) for the> NT-VMS Integration book and the Clustering hints book, both of? which I really could have used.  While after completing all the @ forms it said they were on the way, they never did show up.  ButA then, I guess I got what I paid for.  I just don't understand the A idea of offering something and never delivering it.  I guess whenx? you get that big, you no longer have to care about what kind ofo image you project.   bill   -- iJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   m   ------------------------------  % Date: Wed, 24 Jan 2001 14:59:40 -0500a& From: "Syltrem" <syltrem@videotron.ca>6 Subject: Re: Moving from VMS to Tru64 Unix, Any Ideas?9 Message-ID: <WmGb6.12054$h67.203896@wagner.videotron.net>   8 I ordered the VMS / NT book and received it in 2-3 days.' Is there another book about clustering?h   -- Syltrema3 http://pages.infinit.net/syltrem (OpenVMS Web Site)         > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message% news:94n3ak$go7$1@info.cs.uofs.edu...i- > In article <949p59$o9a$1@bob.news.rcn.net>, # >  ddellutr@XXXenteract.com writes:u > |>J > |> There is a book entitled "OpenVMS and Tru64 UNIX Interoperability andH > |> Migration Guide", (no author given), Order number: 1118-1199A-WWEE,K > |> from Compaq Computer Corp.  I got it free when I attended a seminar inv5 > |> Chicago; you might be able to get it free, also.f >sB > I wouldn't put much stock in this.  I have asked (from someplaceB > on the Compaq webpage that I cna't seem to find anymore) for the@ > NT-VMS Integration book and the Clustering hints book, both ofA > which I really could have used.  While after completing all themB > forms it said they were on the way, they never did show up.  ButC > then, I guess I got what I paid for.  I just don't understand themC > idea of offering something and never delivering it.  I guess when A > you get that big, you no longer have to care about what kind of  > image you project. >v > bill >e > --L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------  % Date: Wed, 24 Jan 2001 23:34:51 -0500A2 From: rdeininger@mindspring.com (Robert Deininger)6 Subject: Re: Moving from VMS to Tru64 Unix, Any Ideas?L Message-ID: <rdeininger-2401012334510001@user-2ive7g7.dialup.mindspring.com>  G In article <94n3ak$go7$1@info.cs.uofs.edu>, bill@cs.scranton.edu wrote:.    aB > I wouldn't put much stock in this.  I have asked (from someplaceB > on the Compaq webpage that I cna't seem to find anymore) for the@ > NT-VMS Integration book and the Clustering hints book, both ofA > which I really could have used.  While after completing all thedB > forms it said they were on the way, they never did show up.  ButC > then, I guess I got what I paid for.  I just don't understand thesC > idea of offering something and never delivering it.  I guess whentA > you get that big, you no longer have to care about what kind ofi > image you project.  Q I ordered a freebo VMS-related CD from the web site, and it arrived the next day.o  @ Then it turned out I couldn't use it, since it's a Pee Cee disk.   -- r Robert Deininger rdeininger@mindspring.comt   ------------------------------   Date: 24 JAN 2001 17:00:34 GMT4 From: karcher@thuria.waisman.wisc.edu (Carl Karcher)* Subject: Re: New OpenVMS Times now on line6 Message-ID: <24JAN01.17003488@thuria.waisman.wisc.edu>  E In a previous article, Tim Llewellyn <tim.llewellyn@bbc.co.uk> wrote:h   ->"D.Webb" wrote:o -> eL ->> Shouldn't Compaq be talking to these Companies - or do they want all the8 ->> Astronomy work to be being done on Solaris systems ? -> m? ->No, its the astronomers who want to be Solaris only, with thes ->very odd exception.t  G Same is true for Gene Sequencing. GCG <http://www.gcg.com/> dropped VMSa@ last July (we dropped GCG). The GCG software was also originally developed on VMS.    ------------------------------    Date: 25 Jan 2001 05:06:53 +0800, From: Paul Repacholi <prep@prep.synonet.com>* Subject: Re: New OpenVMS Times now on line- Message-ID: <877l3kk6c2.fsf@prep.synonet.com>.  / Tim Llewellyn <tim.llewellyn@bbc.co.uk> writes:    > "D.Webb" wrote:e >  > >i > >eL > > Shouldn't Compaq be talking to these Companies - or do they want all the8 > > Astronomy work to be being done on Solaris systems ? > > > No, its the astonomers who want to be Solaris only, with the > very odd exception.n > A > Starlink did get bitten with the OSF on MIPS fiasco, didn't it?   < Oh, that another fall out victim is it. That I did not know.   -- t< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.v@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  # Date: Thu, 25 Jan 2001 03:15:07 GMT 0 From: Serial # 19781010 <pablo.diaz@writeme.com> Subject: Newbie question8 Message-ID: <676v6t4lop7v7a973ni4qcsb2hvi42btb8@4ax.com>   Hello, everybody! ? 	After some time playing with different unixes, I'd like to trymF Open VMS. I don't know what kind of hardware it does support. Here are
 my questions: 1 CPU: Alphaserver 4/233 32 Mb RAM (guess it's OK).'E Disk: Samsung or IBM SCSI disks (I've read somewhere that not all then1 SCSI disks could be used with VMS. Is it true ? ) F Video card: Can I use any standard video card - ie S3 / Matrox / ATI ?8 CDROM: Good ol' Mitsubishi 6x SCSI (guess that's OK too)B Network card: I've got 3Com 3c509/3c900 and a WD8003. Would one of4 these run ? Or do I have to buy a DEC network card ?   Many thanks for your helpo  
 Pablo Diaz pablo.diaz@writeme.com   ------------------------------  % Date: Wed, 24 Jan 2001 23:50:34 -0500e2 From: "Richard B. Gilbert" <DRAGON@compuserve.com> Subject: Newbie question7 Message-ID: <200101242350_MC2-C2FD-3CDC@compuserve.com>a  ) Message text written by Serial # 19781010d >Hello, everybody!F         After some time playing with different unixes, I'd like to tryF Open VMS. I don't know what kind of hardware it does support. Here are
 my questions:t1 CPU: Alphaserver 4/233 32 Mb RAM (guess it's OK).eE Disk: Samsung or IBM SCSI disks (I've read somewhere that not all thei1 SCSI disks could be used with VMS. Is it true ? )yF Video card: Can I use any standard video card - ie S3 / Matrox / ATI ?8 CDROM: Good ol' Mitsubishi 6x SCSI (guess that's OK too)B Network card: I've got 3Com 3c509/3c900 and a WD8003. Would one of4 these run ? Or do I have to buy a DEC network card ?   Many thanks for your help<  J 32Mb is not enough RAM.  If it works at all, performance will be terrible= ! =a  ' 64Mb is the minimum and more is better.   I Most standards compliant SCSI disks should work with VMS 7.1 and above. =e  J Not all SCSI disks are standards compliant however; many are made for the=  C PC market where very little is demanded of them.  VMS will exercisesJ virtually everything prescribed in the standard and will not work or will=  J corrupt data with disks that are not compliant; Quantum Fireball disks ar= e A known to have problems related to Tagged Command Queuing support.A  J Video cards are not standardized, each one needs a different driver.  If = asJ video card is not listed in the VMS Software Product Description (SPD) yo= unJ will have to write your own device driver for it.  In other words: FORGET=   IT!   H The Ethernet cards you mention are not, AFAIK, supported.  If neither of8 them works,  you will have to purchase a supported card.   ------------------------------    Date: 25 Jan 2001 00:05:59 +0800, From: Paul Repacholi <prep@prep.synonet.com>= Subject: Re: No technical computing, was: expanding the nichea- Message-ID: <87vgr5j5p4.fsf@prep.synonet.com>   . Paul Repacholi <prep@prep.synonet.com> writes:  @ > And to make life more interesting, it seems that /nocrc *with*B > redundancy groups is faster than with neither! well, well. About. > 545 IO/sec, 70% CPU. Who would have guessed. >  > Anyone any idea WHY???  ? CAuse with redudence groups, you are doing 10% more IO idiot...n First thought when I woke up...e   -- c< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.d@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------    Date: 25 Jan 2001 00:32:47 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: NTP network loade- Message-ID: <87r91skj0w.fsf@prep.synonet.com>a  ) Piet Timmers <piet@timmers-it.nl> writes:o  ) > Why 3 strata, what will be your advice?iI > We will use two stratum-1 servers with GPS, 14 stratum-2 nodes (we haveiA > 14 main centra in the country) and the other ones as stratum-3.   C OK, I would have 2 second tier servers in each centre, and use themeE and two others, from two other centers as the servers the the hoards.d  B So that gives 28 ST 2 rather than 14, but you are not dependant on; the one local box. Also peer the ST 2 boxes in each centre.   D Until you have got the delay on the GPS tuned, I would fugdw the GPS8 to ST 4 or so and monitor against external ST 1 sources. -- e< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Wed, 24 Jan 2001 12:15:04 -0500r- From: "Richard D. Piccard" <piccard@ohio.edu>y* Subject: Re: OpenVMS "reply/to" challenge.( Message-ID: <3A6F0D92.E4DBA8CA@ohio.edu>   Our routine backup command procedure, which runs as a batch job, includes the following error-recovery routine if the allocation of the tape drive fails:    
 $ what_drive: > $       answer/reply/symbol=good_drive/interval=30/repeat=30 -G                 "Please identify correct tape drive for DAILYA BACKUP."u8 $       if "''good_drive'" .eqs. "" then goto what_drive$ $       define  tape    'good_drive' $       GOTO GET_DRIVE    u Answer is a short C program that I am 99% sure we got from Ehud Gavron in the early 1990s and have migrated to Alpha.p  < You can get the C source, the CLD file and the HLP file from  4                 http://ouvaxa.cats.ohiou.edu/answer/  - (The .obj and .exe files are alpha versions.)n  } The VMS Developers are encouraged to take this and use it as the starting point to provide a New Feature in the next release!o  +                                         RDPh    $ keith.kepner@us.pwcglobal.com wrote:  G > I've searched the OpenVMS FAQ and couldn't find any reference to thisaJ > issue.   We're running a variety of Microvaxes and some VAX 7000s all onL > OpenVMS 5.5-2H4.  We would like to know if there was a way to capture textL > returned with a "reply/to" that could be used, say, to define a logical by > the waiting "request"? > For example: >. > ___Process A___a* > $ request/reply "pls enter device name.": > %%%%%%%%%%%  OPCOM  10-JAN-2001 05:55:07.89  %%%%%%%%%%%' > Request 5, from user SYSTEM on SYSBCKo' > _SYSBCK$NTY27:, pls enter device nameA >hG > %OPCOM-S-OPRNOTIF, operator has been notified, waiting... 05:55:07.90a >g > ___Process B___e > $  reply/to=5 "mua500:"e	 > mua500:eA > 05:57:32.16, request 5 was completed by operator _SYSBCK$NTY33:  >o > ___Process A___  >  > (something like) > $ define/system tape mua500:B > - or do something useful with the supplied information, mua500:. >R > I'd like to be able to do this without groping through the incredibly large operator.log file.  Also, is there a efficient way to redirect OPCOM > messages originatingJ > from multiple hosts and funnel them to one specified host console or PC? >i) > Thank you for any help you can provide.  >e > Keith Kepner > keith.kepner@us.pwcglobal.com- > kkepner@tampabay.rr.comaB > ---------------------------------------------------------------- > The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privilegede > material.  Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or > entities other than the intended recipient is prohibited.   If you received this in error, please contact the sender and delete the material from any- > computer.e   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  % Date: Thu, 25 Jan 2001 13:28:54 +1030i% From: Jeremy Begg <jeremy@vsm.com.au>o" Subject: Re: OpenVMS and Notebooks* Message-ID: <3A702A16.50D5D47C@vsm.com.au>  * fabio_compaq@ep-bc.petrobras.com.br wrote: > > > Sounds strange: a Polyester Canvas Computer Backpack  in the; > OpenVMS  Store.... There arent  Alphabooks anymore ! =)))  > 	 > Regardso >  > FC  B That's to put your Windows notebook in -- the one which is running Charon-VAX VMS :-)           Jeremy Beggo  =   +---------------------------------------------------------+o=   |            VSM Software Services Pty. Ltd.              |J=   |                 http://www.vsm.com.au/                  |o=   |       "OpenVMS Systems Management & Programming"        |k=   |---------------------------------------------------------|.=   | P.O.Box 402, Walkerville, |  E-Mail:  jeremy@vsm.com.au |v=   | South Australia 5081      |   Phone:  +61 8 83592155    |n=   |---------------------------|  Mobile:  0414 422 947      |a=   |  A.C.N. 068 409 156       |     FAX:  +61 8 82231777    |-=   +---------------------------------------------------------+    ------------------------------  % Date: Thu, 25 Jan 2001 10:39:22 +1100 / From: "Phil Howell" <phowell@snowyhydro.com.au> ' Subject: Re: Printing problems in TCPIPt. Message-ID: <0GJb6.94$6O3.3746@ozemail.com.au>  J <kauludo@my-deja.com> wrote in message news:94m5kt$kto$1@nnrp1.deja.com...= > We have the same problems with lpd-Queues. So we change thee$ > Printsymbiont to TCPIP$Telnetsym . >t <snip>3 The telnetsym also has better debugging facilities.P Phil   ------------------------------  % Date: Thu, 25 Jan 2001 13:22:43 +1030 % From: Jeremy Begg <jeremy@vsm.com.au>S1 Subject: Re: RA230+ (KZPAC-CA) and 18/36 GB Disksn( Message-ID: <3A7028A3.7E4F6F@vsm.com.au>   Malcolm Dunnett wrote: > , > In article <3A6C2EDB.C984A3E7@compaq.com>,3 >    Hans Bachner <Hans.Bachner@compaq.com> writes:  > >>M > >> I have a few questions about the older Raid Array 230+ that Digital sold L > >> a few years ago.  I believe the 3-channel one was KZPAC-CB, it is a PCI6 > >> device and is supported in at least OpenVMS v7.1.! > > V6.2-1H3 and V7.1 (and later)" > >ZR > > The 32 GB limit is a limit for the *logical* volumes. You can always put sevenP > > 9 GB disks into a single BA356 and create a RAID 5 array with a net capacityO > > of ca. 54 GB. You then need to create two logical partitions of, say, 27 GBeQ > > each (though the partitions may be different in size - the only point to takenM > > care of is that a single logical partition can't exceed the 32 GB limit).e > >JF >     Clever workaround. Any idea why Compaq/Digital never bothered toI > release V3.x firmware for the controller though? I've got it on genericaE > DAC 960s ( on Prioris servers ) and it's quite happy to create 90GB>3 > logical volumes ( 3 45GB disks in a Raid 5 set ).i > G >     You'd think they'd want their "enterprise" operating system to beGB > able to support storage volumes at least as well as a PC can :-)  B It was my understanding that these host-based RAID controllers areD somewhat out of favour for several reasons including the 32GB volume size limit referrred to,G comparatively poor performance, and an inability to drive a shared SCSIm bus between clustered systems.  F So your "enterprise" operating system is supposed to have "enterprise"G storage attachments, i.e. external RAID controllers (SCSI, fibrechannel  or now SAN) :-)o           Jeremy Beggs  =   +---------------------------------------------------------+f=   |            VSM Software Services Pty. Ltd.              |e=   |                 http://www.vsm.com.au/                  |g=   |       "OpenVMS Systems Management & Programming"        |m=   |---------------------------------------------------------|u=   | P.O.Box 402, Walkerville, |  E-Mail:  jeremy@vsm.com.au | =   | South Australia 5081      |   Phone:  +61 8 83592155    |d=   |---------------------------|  Mobile:  0414 422 947      |,=   |  A.C.N. 068 409 156       |     FAX:  +61 8 82231777    |n=   +---------------------------------------------------------+e   ------------------------------    Date: 24 Jan 2001 19:59:10 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett)t1 Subject: Re: RA230+ (KZPAC-CA) and 18/36 GB Disksa, Message-ID: <JrEtol5cJnpw@malvm1.mala.bc.ca>  ) In article <3A7028A3.7E4F6F@vsm.com.au>, a+     Jeremy Begg <jeremy@vsm.com.au> writes:a > D > It was my understanding that these host-based RAID controllers areF > somewhat out of favour for several reasons including the 32GB volume > size limit referrred to,  7    That's my point though. It's not a limitation in thei8 controller, only in the outdated version of the firmware6 that Compaq provides with it. I've got 90GB volumes on; Mylex DAC 960s ( same hardware ) with more recent firmware.   I > comparatively poor performance, and an inability to drive a shared SCSIC  > bus between clustered systems. > H > So your "enterprise" operating system is supposed to have "enterprise"I > storage attachments, i.e. external RAID controllers (SCSI, fibrechannelh > or now SAN) :-)  >   A    Yet more indication that Compaq doesn't want people at the low E end using VMS I suppose - the cheapest external raid controller costst2 something like 3 or 4 times as much as the RA230+.  ;     I'll grant you the performance isn't great, but in many   situations it would be adequate.   ------------------------------  % Date: Wed, 24 Jan 2001 17:12:34 +0000e0 From: andrew harrison <andrew.nospam@uk.sun.com>W Subject: Re: Rehash of Andrew's Java on OpenVMS FUD.  (was: Re: GS160 hardwarequestion)|* Message-ID: <3A6F0D02.5D9BA6BD@uk.sun.com>   Arne Vajh=F8j wrote: > =    > andrew harrison wrote:: > > Well its hardly bogus, its just worse than I suggested9 > > The 1.2.X JVM is available for OpenVMS and neither is 9 > > the plugin. The 1.3.X JVM isn't available for OpenVMSp, > > at all (well there is an alpha release). > =s    > It is indeed not good for VMS. > =   C > But it illustrates how well you check your facts before posting !e > =n    0 Really and you got your facts right first time =  ! as well !!!!!!!!!!!!!!!!!!!!!!!!!n  " Oh no I forgot you didn't did you.   Regardse Andrew Harrisons Enterprise IT Architectn   ------------------------------  % Date: Wed, 24 Jan 2001 17:10:23 +0000B0 From: andrew harrison <andrew.nospam@uk.sun.com>W Subject: Re: Rehash of Andrew's Java on OpenVMS FUD.  (was: Re: GS160 hardwarequestion)e* Message-ID: <3A6F0C7F.B91A77C6@uk.sun.com>   Jordan Henderson wrote:  > - > >Sorry yet again you have missed the point.t > > - > >You asserted incorrectly that OpenVMS Javao! > >support is only a little late.s > > , > >If you thought that 1.2.x was the current) > >Java release then you could be excused-0 > >for making this statement on the grounds that/ > >you were ignorant of how out of date OpenVMSC > >Java support actually is. > >V1 > >If you knew that 1.3.x was the current release@1 > >then you knew that your statement that OpenVMST0 > >support is only a little late was untrue when > >you made it.o > >e1 > >Which do you prefer ignorance or falsehood ???s > >  > / > Neither.  As I've explained, I don't considerd/ > a year, or even more, more than a little later1 > when it comes to language standards compliance.e >   . Ohh I see so you have now moved the goalposts * from something concrete and practical, e.g+ the availbility or otherwise of the latest s) version of Java on OpenVMS to something a- bit more abstract.  * How very convenient. Jordan you really are a remarkable peice of work..  ) Strange that the 12 months isn't a littlen* late view was not shared by other posters , to this group who had complained that Compaq$ had missed the boat on Java support.  & Have you forgotten entirely what your & list of Java non facts and rumours was in response to ?????  / > Somehow, you just can't seem to carry forward?- > my explanations into the quoted portions ofo1 > your replies.  You must assume that many people 0 > here only read the last message in a thread so0 > as to catch up on all of the discussion at one > time.h >  > >nE > >> This is where this statement that you insist is so inaccurate isnE > >> introduced.  Here, I say that Compaq is a "little behind" (not a ) > >> "little late" as we've been saying).  > >> > >0$ > >And you accuse me of being picky.E > >> Note the reference to Java 1.3?  Note that I didn't say anything1B > >> about 1.2 being the most up-to-date as you earlier contended? > >> > >i1 > >So you are admitting to deliberately trying tov1 > >misslead. If you knew as you point out you did 0 > >that 1.3 is the current VM then you also knew$ > >that "a little behind" was false. > >d > 3 > Not false, a matter of opinion.  Is this the besta3 > you can come up with of all the many inaccuraciesn6 > you allude I've made wrt Java?  Laughable!  Somehow,6 > you've failed to respond, again, to real substantive2 > inaccuracies you've stated on this very subject. >     * No Jordan, I am just concentrating on the ( first of numerous mistakes because there( seems little point in trying to get you ' to cough for the rest until you can get  over this one. d  ( What is laughable is that you pesist in & defending this one point when you are  so obviously wrong.   ) No one would describe any system as beingm( a little behind if that system had just ' got support for standard 1 after other D* vendors had started supporting standard 2,- but that is of course what you are attemptingd	 to argue.v    * Perhaps you should have argued that Oracle& support for OpenVMS was only a little ' behind because while other vendors had  % 8i you could get 8.0.5 on OpenVMS, I  ( doubt that anyone would have agreed with you. > >iC > >> Now, are you going to answer the question as to why Sun hasn't # > >> published any JBB2000 numbers?  > >> > >y1 > >Why should I bother Jordan, you introduced thee, > >false claim that Compaq had the best Java/ > >benchmark results. It seems hardly worth the'/ > >effort discussing something that is based onf% > >a blatantly false premise does it.  > >r > + > Compaq does have the best JVM98 benchmark 2 > results, so there is truth to the statement that- > Compaq has the best Java benchmark results.  >   * Very helpfull, its a client side benchmark, as you know and OpenVMS is hardly a desktop  OS any longer is it. e  , Nor can I see the point of your posting, the* SPEC JVM98 results and the JBB2000 results- were all done using Tru64, are you suggestingr+ that OpenVMS would get the same performancet on the same HW ????????   , Though of course as pretty poor FUD attempt   its par for the course from you.   Regardss Andrew Harrisond Enterprise IT ArchitectA   ------------------------------    Date: 24 Jan 2001 12:54:19 -0500/ From: jordan@lisa.gemair.com (Jordan Henderson) W Subject: Re: Rehash of Andrew's Java on OpenVMS FUD.  (was: Re: GS160 hardwarequestion) * Message-ID: <94n4sb$9dm$1@lisa.gemair.com>  * In article <3A6F0C7F.B91A77C6@uk.sun.com>,2 andrew harrison  <andrew.nospam@uk.sun.com> wrote: >Jordan Henderson wrote: >>  . >> >Sorry yet again you have missed the point. >> >. >> >You asserted incorrectly that OpenVMS Java" >> >support is only a little late. >> >- >> >If you thought that 1.2.x was the current8* >> >Java release then you could be excused1 >> >for making this statement on the grounds thatm0 >> >you were ignorant of how out of date OpenVMS >> >Java support actually is.s >> >2 >> >If you knew that 1.3.x was the current release2 >> >then you knew that your statement that OpenVMS1 >> >support is only a little late was untrue wheni >> >you made it. >> >2 >> >Which do you prefer ignorance or falsehood ??? >> > >>  0 >> Neither.  As I've explained, I don't consider0 >> a year, or even more, more than a little late2 >> when it comes to language standards compliance. >> i >0/ >Ohh I see so you have now moved the goalposts .+ >from something concrete and practical, e.g5, >the availbility or otherwise of the latest * >version of Java on OpenVMS to something a >bit more abstract.- >   , I didn't move the goalposts at all.  What an- unbelievable strawman this is!  You accuse mea. of saying that Compaq was providing the latest, version of Java on OpenVMS, this in spite of/ the fact that I demonstrated quite clearly thatd, I had never said that, and now you accuse ME- of moving the goalposts?  I think you've beenu, trying to move the goalposts so that you can1 accuse me of being inaccurate, but I won't budge.   + >How very convenient. Jordan you really aren >a remarkable peice of work. >r* >Strange that the 12 months isn't a little+ >late view was not shared by other posters d- >to this group who had complained that Compaqs% >had missed the boat on Java support.p >o  0 I don't recall seeing a lot of complaints about 0 Compaq vis-a-vis Java support.  Maybe there have, been some, I don't know.  I do recall seeing0 people saying they're using it, even on OpenVMS, without problems.e  ' >Have you forgotten entirely what your e' >list of Java non facts and rumours was  >in response to ?????i >r0 >> Somehow, you just can't seem to carry forward. >> my explanations into the quoted portions of2 >> your replies.  You must assume that many people1 >> here only read the last message in a thread soe1 >> as to catch up on all of the discussion at onel >> time. >> d >> >F >> >> This is where this statement that you insist is so inaccurate isF >> >> introduced.  Here, I say that Compaq is a "little behind" (not a* >> >> "little late" as we've been saying). >> >>e >> >% >> >And you accuse me of being picky. F >> >> Note the reference to Java 1.3?  Note that I didn't say anythingC >> >> about 1.2 being the most up-to-date as you earlier contended?  >> >>a >> >2 >> >So you are admitting to deliberately trying to2 >> >misslead. If you knew as you point out you did1 >> >that 1.3 is the current VM then you also knew % >> >that "a little behind" was false.t >> > >> d4 >> Not false, a matter of opinion.  Is this the best4 >> you can come up with of all the many inaccuracies7 >> you allude I've made wrt Java?  Laughable!  Somehow,:7 >> you've failed to respond, again, to real substantive 3 >> inaccuracies you've stated on this very subject.y >> i >v >s+ >No Jordan, I am just concentrating on the .) >first of numerous mistakes because theret) >seems little point in trying to get you "( >to cough for the rest until you can get >over this one.  > ) >What is laughable is that you pesist in s' >defending this one point when you are c >so obviously wrong. >O* >No one would describe any system as being) >a little behind if that system had just  ( >got support for standard 1 after other + >vendors had started supporting standard 2,f. >but that is of course what you are attempting
 >to argue. >y  - No one?  Another poster to this group pointede+ out that 1.2.x compliance was the important , milestone that a lot of tools were still at.  . Somehow, you just continue to snip out my best+ points about how being at various levels ofu, standards compliance is just not that big of deal in computer languages.  e  - Note to those reading this 'debate'.  Please e/ don't think you are getting even a fraction of a3 the story by reading only Andrew's posts, he snips e1 out all the good points made by the opponent.  If + you must save time, read my posts only as I 0 usually quote all of Andrew's posts (and respond+ to every challenge, generally) OR I give a -) "[snip]" to show where something has beenm) elided. If you are interested, go look upp what I've snipped.  , Andrew, on the other hand, snips out all the, embarrassing challenges and good points made+ by the opposition, rarely with an editorialf* [snip] or other device to show what's been, done.  Hard to believe that he also has the , audacity to accuse others of spin, isn't it?   >y+ >Perhaps you should have argued that Oracle ' >support for OpenVMS was only a little  ( >behind because while other vendors had & >8i you could get 8.0.5 on OpenVMS, I ) >doubt that anyone would have agreed with  >you.i  / I might have argued that, in fact.  The OpenVMS 0 shops I've seen tend to be more conservative wrt, software upgrades than the Unix shops.  They- tend to be more interested in stability than a& chasing the lastest round of features.   >> >D >> >> Now, are you going to answer the question as to why Sun hasn't$ >> >> published any JBB2000 numbers? >> >>i >> >2 >> >Why should I bother Jordan, you introduced the- >> >false claim that Compaq had the best Javae0 >> >benchmark results. It seems hardly worth the0 >> >effort discussing something that is based on& >> >a blatantly false premise does it. >> > >> r, >> Compaq does have the best JVM98 benchmark3 >> results, so there is truth to the statement that-. >> Compaq has the best Java benchmark results. >> s >y+ >Very helpfull, its a client side benchmarkt- >as you know and OpenVMS is hardly a desktop   >OS any longer is it.  >m- >Nor can I see the point of your posting, the-+ >SPEC JVM98 results and the JBB2000 results . >were all done using Tru64, are you suggesting, >that OpenVMS would get the same performance >on the same HW ???????? >   . OpenVMS and Tru64 share the same JVM.  I would+ expect similar performance, but perhaps the , Tru64 benchmarks somewhat faster due various' OS architectural issues.  I can easily -, explain why Compaq doesn't go to the trouble* of providing benchmarks for both Tru64 and OpenVMS.  * Now, can you explain why Sun provides such- poor benchmarks on JVM98 and no benchmarks at0 all for JBB2000?  - >Though of course as pretty poor FUD attempt m! >its par for the course from you.. >.  ; Seems you still haven't learned what FUD means, even after h7 all my efforts to make you understand it.  You finally  8 did seem to understand that you can use facts to produce: FUD, but I'm still having a problem with the issue of the : audience and how it's not FUD if you're not addressing it  so as to scare customers.  1  1 I'm not afraid to repeat myself, especially sincea0 you are so prone to deleting my best points from7 your replies.  If I wanted to spread FUD, I would post h6 to comp.os.solaris or comp.sys.sun.  I'll try to make 6 this so simple that even you can understand it.  If I ; sent a private email containing a whole lot of disparaging C< views on Sun to Kerry Main, for example, no sensible person > would call that FUD.  If I sent the same email to one of your = customers, well, that would be FUD.  Now, what YOU do here...t   >Regards >Andrew Harrison >Enterprise IT Architect   -Jordan Hendersone jordan@greenapple.com5   ------------------------------  % Date: Wed, 24 Jan 2001 18:46:25 -0800a! From: Tom Linden <tom@kednos.com>f Subject: RW-CD drives for VMS 9 Message-ID: <EJEDILJIPPJOIEFOLDHLAEMCCEAA.tom@kednos.com>n  D When I posted my original question I was hoping to get some guidanceI from the troops that had crawled throught the muck. Not interested in DVDl so let's drop that topic.e  C What can I buy off the shelf that will work with VMS?  Is there anye
 experienceL out there that can be shared?  A list of supported equipment would be a nice feature.   ------------------------------  % Date: Wed, 24 Jan 2001 21:49:13 -0500t- From: JF Mezei <jfmezei.spamnot@videotron.ca>e, Subject: Shareable image with DEC-C question, Message-ID: <3A6F9423.112E2626@videotron.ca>  & If have the following snippet of code:   static FILE *myfile ;v   int myopen(char *filename) { myfile = fopen(filename,"r);   }    int mylog(char *text).' { fprintf(myfile,"log text=%s\n",text);  }s   ------------------  N Now, I take this, and compile/link it into a shareable image with "myopen" andN "mylog" declared as universal symbols. If the image is installed /shared, willG "myfile"  be allocated in the process heap (each process having its own.J private copy), or will it reside inside the shareable image, shared by all, processes whose image is linked against it ?  G If the above would result in a single "myfile" being shared amongst all-K processes, is there a way to declare it so that each process would have its  own private copy ?   ------------------------------  % Date: Wed, 24 Jan 2001 22:12:06 -0500M" From: Dan Sugalski <dan@sidhe.org>0 Subject: Re: Shareable image with DEC-C question: Message-ID: <5.0.2.1.0.20010124221120.02c64ec0@24.8.96.48>  * At 09:49 PM 1/24/01 -0500, JF Mezei wrote:' >If have the following snippet of code:  >h >static FILE *myfile ; >o >int myopen(char *filename)2 >{ myfile = fopen(filename,"r);a >} >g >int mylog(char *text)( >{ fprintf(myfile,"log text=%s\n",text); >} >. >------------------y >gL >Now, I take this, and compile/link it into a shareable image with "myopen" F >and "mylog" declared as universal symbols. If the image is installed H >/shared, will "myfile"  be allocated in the process heap (each process  >having its ownoK >private copy), or will it reside inside the shareable image, shared by alla- >processes whose image is linked against it ?   D That static declaration will be process-private--there won't be any  sharing, and you'll be fine.   					Dan  I --------------------------------------"it's like this"------------------- 2 Dan Sugalski                          even samurai? dan@sidhe.org                         have teddy bears and event;                                       teddy bears get drunkt   ------------------------------  % Date: Wed, 24 Jan 2001 22:22:09 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>e0 Subject: Re: Shareable image with DEC-C question+ Message-ID: <3A6F9BD8.C8374BF@videotron.ca>n   Dan Sugalski wrote:mE > That static declaration will be process-private--there won't be any- > sharing, and you'll be fine.    L Thanks for the quick answer. much appreciated. (means I don't have to change< much to move a whole bunch of stuff into a shareable image).   ------------------------------  % Date: Wed, 24 Jan 2001 19:42:39 +0100 2 From: martin@radiogaga.harz.de (Martin Vorlaender)7 Subject: Re: stop decw$session = kill SUN CDE server...V; Message-ID: <3a6f221f.524144494f47414741@radiogaga.harz.de>h  H Jan Vorbrueggen (jan@mailhost.neuroinformatik.ruhr-uni-bochum.de) wrote:N > I think that is to be expected - the session manager exiting probably does aN > reset to the X server (always a good idea with those memory leaks et al. allK > over the place), and that kills all extant sessions. Who said X was easy?   D You have the option of quitting it, though (at least the old SessionB Manager) without killing the X server. It's hidden in the settingsD somewhere (sorry to be fuzzy here, I don't have access to DECwindows right now).:   cu,    Martin --J One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer7 One OS to find them           | work: mv@pdv-systeme.de N One OS to bring them all      |       http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------   Date: 24 JAN 2001 22:55:22 GMT4 From: karcher@thuria.waisman.wisc.edu (Carl Karcher)( Subject: Re: TCPIP$BIND: dynamic updates6 Message-ID: <24JAN01.22552248@thuria.waisman.wisc.edu>  4 Configuration: TCPIP V5.0a with ECO1, OpenVMS V7.2-1  @ As an experiment, I saved my zone files and defined the logical:  ( 	TCPIP$BIND_SERVER_MERGE_DYNAMIC_UPDATES  E system-wide with a value of "TRUE". I have the "allow-update" line in H tcpip$bind.conf under the zone file definition section. I restarted bindF and the log file had the expected message: "WARNING: dynamic zone fileA 'my-domain.db' is writable". I then attempted to use the nsupdatesE utility command (defined in SYS$MANAGER:TCPIP$DEFINE_COMMANDS.COM) ash! follows to send a dynamic update:V    $ nsupdate .  > update add test.my-domain 1000 in a 1.2.3.4  F There was no effect. No such name was defined, nothing in the bind logE file and no new zone file after waiting an hour. Has anyone else beent2 able to use the vms nsupdate to trigger an update?   ------------------------------  % Date: Wed, 24 Jan 2001 16:34:17 -0600o" From: Ryan Hammond <ryanh@bvu.edu># Subject: VMS Documentation Giveawayj' Message-ID: <3A6F5869.79266087@bvu.edu>p  E Our University has an almost complete set of VMS manuals for Vax 6.0.tF The set could possibly be missing a book or two. We are giving the setB away less shipping. The estimated shipping cost via UPS within the8 United States is $50. If interested, email ryanh@bvu.edu   Ryan Hammond Student Network AdministratorC Buena Vista University   ------------------------------  # Date: Wed, 24 Jan 2001 18:23:12 GMTd From: sabolich@my-deja.com@ Subject: Re: VMS vs UNIX (was: Linux worm and RedHat 7.0 broken)) Message-ID: <94n6i5$hi1$1@nnrp1.deja.com>t  , In article <3A6E7D2F.F2CBD230@infopuls.com>,,   Christof Brass <brass@infopuls.com> wrote: > sabolich@my-deja.com wrote: < > > FreeBSD is technically very good except for SMP support. > > B > Thanks, didn't know that. But the bad news will follow below ;-) >tI > > Yeah it will execute Linux, SCO, and other x86 binaries.  So will thei > > other *BSDso > >u> > > > I don't like the Linux way very much because the creator4 > > > jumped into this business without knowing much > > ? > > Well yeah, he decided to make a kernel for the 386 for selftJ > > teaching/interest.  When he started it he expected he himself would be > > the only user. > >o >uB > I don't think that is an excuse. Actually this is even worse. If; > someone decides to start that kind of project for him- oro@ > herself the constraints of compatibility and attracting othersB > with familarity vanish. And wouldn't it be much more exciting to; > explore new ideas instead of repeating the experiences ofd	 > others?  >tI > > Linux didn't start with any grand vision.  All operating systems comehI > > with some historical baggage.  I've programmed for WinNT a lot, VMS auJ > > little, and now use mostly Linux.  Of those three OSes I consider UnixG > > to have the cleanest most intelligent design.  When I first startedoJ > > using and programming for Linux it took me a while to get to like it -L > > after a while I began to realize how simple and elegant Unix programming > > is.a > >p >h@ > Stating that Linux (or should we say Linus Thorvalds?) started? > without any grand vision seems to kill the enthusiasm and they: > trust of getting something real innovative with Linux. I; > wouldn't go that far but I think the vision wasn't on thetB > technical, specifically the architectural, the algorithmical and< > the UI side. Could we agree that there was a vision on the8 > social, political and eventually anti-commercial side?  E No.  Linux (strictly speaking) is just a kernel while GNU/Linux is an0F OS.  UI has no place in the kernel (which is the only part Linus did).G When it was started I doubt there was any grand vision on Linus's part.rC  The vision you speak of is the GNU part, Free Software Foundation.eG Have a look at www.gnu.org.  GNU had libraries compilers, editors, etc.-H  At the time they had no kernel but were working on the HURD.  The LinuxH kernel was put under GPL and put together with software from GNU to make
 a working OS.S   >aK > > FreeBSD is very popular - there just isn't as much hype surrounding it.AJ > > What makes you think that FreeBSD people want this hype?  Knowing thatI > > their OS runs much of the Internet - sites like Yahoo, etc. is rewardh > > enough.> > >> > A > Agreed. I don't know. But I regret that even in better informede; > circles it is not mentioned that much. And with technical B > discussions Linux is much more mentioned, analysed and compared.? > So apart from any hype I think there is a level of popularitya' > which could be helpful and desirable.s  ? have a look at http://uptime.netcraft.com/up/today/top.avg.htmln  ? What is talked about and what is used are not allways the same.i   >lJ > > The Alpha version of FreeBSD I think is not that solid -- I remember aK > > while ago seeing recommendations not to use it for critical stuff (thissJ > > was a while ago so I may be wrong as the status now).  FreeBSD goal isF > > maximum performance on x86.  You might want to try NetBSD (goal is > > portability) on the Alpha. > >o >eA > So, here it comes: is it not a little bit strange that a system @ > which has its goal of maximum performance on x86 isn't good at > SMP?  B No.  FreeBSD kernel is written in such a way that it cannot safelyE execute on multiple CPUs simultaneously.  To provide protection a big H kernel lock is used that enforces this.  However user processes can takeA advantage of multiple CPUs.  So for example if compiling multiple E programs simultaneously multiple CPUs will be used since most time ispG spent processing outside the kernel.  But under certain conditions like G high network throughput (FTP server) only one CPU can be utilized sincer; much time will be system time in kernel (disk i/o, tcp/ip).w  G To allow the kernel to run on multiple CPUs simultaneously fine grainedCH locking is required.  Look for this in FreeBSD 5.x.  Linux 2.0 has a bigD kernel lock like FreeBSD.  Linux 2.2 removed the big lock but is notE really fine grained - so SMP performance is not that good.  Linux 2.4hG does have fine grained locking and SMP performance is quite good - beste of the free *nixes.e  ; > Are you sure about this FreeBSD's goal? I remember BSD386?. > and think that you might confuse these both.  E FreeBSD focus - high performance on x86, also runs on Alpha (check to.3 make sure Alpha port solid before you commit to it)   C OpenBSD focus - security (acomplished by continuous code audits ands? secure default install).  Also quite portable and runs on Alphae  G NetBSD focus - portability (go to www.netbsd.org and check out the lista of supported platforms).  F All the BSDs are good performers and hold up well under load.  Any one will make a fine server.    But I really don'te@ > know. But I read a few weeks ago that Alpha is the other major> > platform for FreeBSD. Do you know if the Alpha systems don't= > have the Intel architecture problems like BIOS restrictions A > (boot devices, disk addressing schemes, ...) and other built-in @ > problems like memory addressing restrictions? My Linux/FreeBSDA > system is - unfortunately - Intel based but I decided to go forh > Alpha with my next buys. > Thanks anyway for the hints! >>A > > Alpha - the fastest design available (but makes lots of heat)aE > > MIPS - great design (very good speed considering low power usage)a? > > x86 - amazing this kludge of a CPU runs as fast as it does.o > >r >n > :-)e > H > > VMS clean in every of the three important dimensions? What are these > > three dimensions?  > >h >e. > 1.Kernel - how things internally are handled  A There are many implementations of Unix, varies by implementation.n  3 > 2.API - how the OS is presented to the programmerc  A This was my main point.  Unix has a short, simple, effective API.   1 > 3.CLI - how the OS is presented to the sysadmind  G This has more to do with the shell, scripts, and directory organiztion.p;  Varies from Unix to Unix although generally quite similar.  >nL > > VMS is well engineered.  It is reliable, stable, etc.  Its design is notJ > > clean though.  It is a commercial and prorietary OS - with such thingsK > > the design is often rushed to get a working product out the door before  > > the competition. > >n >gA > Very interesting thought. I would conclude the opposite though.nB > You are probably right with respect to implementation quality of@ > the first versions. If I look back on the history of VMS there@ > was plenty of time to kill bugs. So I regard time pressure not@ > as a big issue on the mid term. But what really matters is theB > basic system design. I heard (or read) that VMS and the VAX were< > developped together (the first time in history that the SW? > design influenced the HW, cpu architecture). So I expect thatl< > the design was carefully chosen and that stayed until now.= > May I ask you to point out the five most embarassing designH > faults of VMS? >eL > > Unix on the other hand has much of its roots in academia where a lack ofE > > market pressure gives more time for planning, debate, and design.t > >  >hA > If it were so! In fact UNIX didn't really start in academia. ToeA > my knowledge UNIX was the outcome of a project that was startediB > by AT&T, GE and the MIT. The DoD's ARPA was also a major funder.> > To make is short: MULTICS was the goal, never finished, AT&T? > canceled the project. Ken Thompson, Dennis Ritchie and Joseph : > Ossanna, all AT&T, had nothing to do, so they decided toA > implement a game called Space Travel on a PDP-7 which was therenA > and unused. They realised that it would be faster to develop ano@ > OS for the PDP-7 suitable for their task than cross-developing@ > the game with one of the other computers around where the gameA > was already running. The goals of UNIX were minimalism in everys( > respect: kernel hardware requirements,  D Humble beginnings heh?  Let's make an OS for Space Travel.  The name< Unix is a pun on Multics.  Multics was designed to be an allG encompassing do all OS.  Perhaps Thompson, Ritchie, and Ossanna came to E the conclusion that Multics was overkill and a simple system could dot! everything the complex one could.    > number and completenesst? > of services, documentation, guidelines a.s.o.. And the source & > wasn't by no means freely available.  G Because of the simple minimalist approach the source code from AT&T waseE used in computer science courses at universities for study.  A lot oftF work was contributed back, much of it from Berkley (therefore the name# BSD, Berkley software distribution)o  G Later when AT&T realized what they had did trouble start.  Lawyers cameSB in did a lot of damage.  Enter Linus and his kernel - written fromB scratch and free of AT&T code with no legal cloud hanging over it.  = > Honestly, I don't see any design on the sysadmin dimension.e@ > There is no consistent option and parameter handling. There is$ > even no system in naming commands.  F I'm a programmer so I tend to place more weight on the API.  It is notE like every one is different though - ls, ps, find, etc are all prettyr	 standard.s  A > I'm really interested in understanding the "design" of UNIX ands> > find a way to like the API. I tried it for several years butB > without success. And I'm close to give up. If you could show howB > I can learn to like UNIX this would be really great (no irony!).  @ I'll try to give a good example here.  Suppose you have two fileG descriptors (fds which is an open serial port, and fdf which is an openmF file)  The program xyz takes input on standard in and writes it outputD to standard out.  You want to run xyz with fds for input and fdf for; output.  You also want to run it at boosted priority level.i  F In VMS you will have one call to the OS the create process xyz.  ThereF is a big structure you pass containing information about input, outputG handles (for redirection), priority, priviledge, etc.  Everything aboutJ1 that process must have a place in that structure.   & Unix is completely different, example:  ' x = fork() /* create copy of process */l if (x == 0)h {u /* we are child */ /* make fds standard in */
 dup2(fds, 0);c /* make fdf standard out */i
 dup2(fdf, 1);S /* adjust my priority to -5 */! setpriority(PRIO_PROCESS, 0, -5); , /* replace current process image with xyz */ execl("xyz", NULL);- /* never get here */ }e else { /* we are parent */t }r  G So I've replaced a complex system call (in VMS) which takes a structure @ with a few simple ones in Unix (none of which take a structure).  G That is what I like about unix.  Simplicity and flexibility.  It's kindnH of like a lego or erector set.  The basic components are very simple butL you can put them together any way you want to achieve exactly what you want.   >IH > > Anyways VMS and Unix are similar in there abilities.  Both are multiF > > user, multi tasking OSes with virtual memory protection and demandK > > paging.  You can accomplish the same results with both OSes.  Just thatw2 > > Unix does it with fewer and simpler API calls. > >  >gA > Very interesting thought. I never looked at it that way. ThoughfB > it seems that this description will fit for many other OSs. If IB > look close than it seems to me that the differences are stronger@ > than the similarities. Could we agree that VMS to UNIX is like, > MacOS to M$DOS? VMS establishes standards,  + A quote from the book Linux Device Drivers:S  H .... role of a device driver is providing mechanisms, not policies.  TheH distinction between mechanism and policy is one of the best ideas behindH the Unix design.  Most programming problems can be split into two parts:C "what needs to be done" (the mechanism) and "how can the program bemG used" (the policy).  If the two issues are addressed by different parts-F of the program, or even by different programs altogether, the softwareC package is much easier to develop and to adapt to particular needs.   ) A quote from Beginning Linux Programming:y  G You can't anticipate exactly how ingenious users will use your program.pE .... Never assume that you know everything the user might want to do.l  A Again, philosophy here is similar to programming, give the user alG collection of simple programs and let him pipe them together to do what 	 he wants.    > a rich set of full= > blown services and guidelines to achieve uniform and system @ > adapted behaviour of the apps written according to these rules8 > and using these services, to prevent each developer to: > re-implement the same things again and again like record! > handling and parameter parsing.e  D Parameter parsing functions are in the C library.  On the whole codeD reuse is good.  Suppose you want to see all unique processes running excluding bash, output sorted.  " ps -A | sort | unique | grep -v sh  G As you can see a few simple commands piped together.  ps (process show)aH does not need to have sort logic programmed in, etc.  ps need not have aD million options trying to anticipate everything the user could want.! Connect them up however you want.-     If I compare this with the UNIX7 > approach then I think if there was design that it hasaA > disappeared over time or it was an anti-design (the decision tov? > be without any design) or the design was intentionally hiddene > and remained in that state.h  H Let's call it anti policy.  The design is to be flexible through lack of policy.e   >iC > > I'm just curious, how much have your programmed for Unix vs VMS ; > > (sometimes familiarity with one system skews judgement)e > >d > > Fran > >  > > Sent via Deja.comi > > http://www.deja.com/ >e= > I programmed about 2000 h on VMS using MODULA-2 and heavilye@ > exploiting the RMS services (we didn't use a database) and the= > I/O features to communicate with exernal devices like SPSs.sB > I programmed about 500 h on different UNIX variants using C, C++B > and Java for very different tasks one of them recently beeing to= > implement a servlet based management system. But I did more ( > system management on UNIX than on VMS. >t@ > And yes, familiarity with one system skews judgement: I hadn't@ > believed that I scorned UNIX so much if I hadn't gotten really > acquainted with.  F What I consider to be 'clean' about Unix the fact that you can achieveH so much with such a simple system.  Just a guess but I'll bet people whoF like Unix like building things from kits, etc.  I guess a lot of it is just personal taste.   Fran     Sent via Deja.comN http://www.deja.com/   ------------------------------  # Date: Wed, 24 Jan 2001 22:16:31 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)& Subject: Re: VMS writable CD on Win2K?5 Message-ID: <3vIb6.93$cu.893@gazette.loc1.tandem.com>n  W In article <t6u12pst176h61@news.supernews.com>, "John Vottero" <John@mvpsi.com> writes: L :I've had a CD-R attached to a VMS system for years.  Both VAX and Alpha.  I> :didn't have to write my own driver, DKdriver works just fine.  B   OpenVMS Engineering, too, has CD-R drives connected directly to :   an unmodified DKDRIVER, and makes regular use of CD-R...  B   An updated version of DQDRIVER -- with support for the necessaryC   IO$_DIAGNOSE function for controlling j-random IDE widgets -- is 1$   available at the Freeware website.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 24 Jan 2001 17:52:43 -0500e* From: "Paul A. Jacobi" <nospan@nospam.com>& Subject: Re: VMS writable CD on Win2K?, Message-ID: <94nm9o$77tf$1@lead.zk3.dec.com>  : "Tim Llewellyn" <tim.llewellyn@bbc.co.uk> wrote in message# news:3A6ED69A.D45F0283@bbc.co.uk...- >- > ! > DVD-R is coming too, so I hear.M    H In fact, I recently booted OpenVMS from a ATAPI DVD-ROM drive with a ODSI formated DVD-R disk.  No changes to the O.S. were needed!  The DVD-R diske looks like a really big CD-ROM.,  L We have the technology today, but it will be a while before the business and+ distributuion issues with DVD are resolved.o     Paul A. Jacobi Compaq Computer Corporation ! OpenVMS Systems Group, ZKO3-4/U14; 110 Spitbrook Road Nashua, NH 03062-2698, Email: Paul.Jacobi@compaq.coml   ------------------------------  # Date: Thu, 25 Jan 2001 00:13:01 GMTh= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)n& Subject: Re: VMS writable CD on Win2K?0 Message-ID: <009F69C2.220C20D2@SendSpamHere.ORG>  j In article <3vIb6.93$cu.893@gazette.loc1.tandem.com>, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:X >In article <t6u12pst176h61@news.supernews.com>, "John Vottero" <John@mvpsi.com> writes:M >:I've had a CD-R attached to a VMS system for years.  Both VAX and Alpha.  It? >:didn't have to write my own driver, DKdriver works just fine.  >eC >  OpenVMS Engineering, too, has CD-R drives connected directly to s; >  an unmodified DKDRIVER, and makes regular use of CD-R...i  ' Curious, what do you use for write s/w?h   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM             nO city, n., 1. a place where trees are cut down and streets are named after them.n   ------------------------------  % Date: Wed, 24 Jan 2001 17:44:33 -0800o! From: Shane.F.Smith@Healthnet.com & Subject: Re: VMS writable CD on Win2K?D Message-ID: <OFF4BC3C2C.D219C81B-ON882569DF.00097FAC@foundation.com>  G Huh? I've seen DVD-R's in my local tech shop, at over $600 granted, buti/ they're there. They've been there a while, too.    Shaneg          A Tim Llewellyn <tim.llewellyn@bbc.co.uk> on 01/24/2001 08:53:41 AMt  ) Please respond to tim.llewellyn@bbc.co.uki   To:   Info-VAX@Mvb.Saic.Comg cc:m  ' Subject:  Re: VMS writable CD on Win2K?t         Alan Greig wrote:   3 > On Wed, 24 Jan 2001 13:20:26 +0000, Tim Llewellyn?" > <tim.llewellyn@bbc.co.uk> wrote: >u > >k > >  > >"David J. Dachtera" wrote:  > >o3 > >> ...and that's not to mention CD-RW or DVD-RAM.d > >> > > " > >DVD-R is coming too, so I hear. > C > And I saw a report somewhere that the major CD manufacturers haverB > agreed to release MF CD or whatever it's called (Multi-frequency> > laser). Up to 100 GB on a CD. Could DVD be obsolete already? >   H Yeah, I want one! However DVD-R is alledgedly almost shipping (in hi end macs).H Can burn CDR's and DVD-R's (not the same as DVD-RAM). Mind you they seem    to be only 1x or 2x write speed.  F I sure hope those 100GB rewriteables, when they arrive, can be written at aF decent rate, otherwise you're gonna NEED and VMS system to have enough8 guaranteed uptime to write that 100 gigs at 100k/sec :-)    -- 6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uk   A I speak for myself only and my views in no way represent those ofh MedAS or the BBC.l   ------------------------------  # Date: Thu, 25 Jan 2001 06:27:34 GMTp From: cstranslations@msn.com& Subject: RE: VMS writable CD on Win2K?) Message-ID: <94oh0h$njq$1@nnrp1.deja.com>e  0 In article <009F6959.14102BC1@SendSpamHere.ORG>,    system@SendSpamHere.ORG wrote:F > In article <EJEDILJIPPJOIEFOLDHLOEJHCEAA.tom@kednos.com>, Tom Linden <tom@kednos.com> writes:6 > >Well one can always hope, but I do see the problem. > >wF > >Is it too much to expect that a such a drive purchased at the local5 > >electornics superstore (Fry's) will work with VMS?a >sG > Huh???  Sorry but I can't follow what you're trying to do.  At first,  I?C > read it as if you have the device on the Weendoze box and somehowy wantedH > to network mount it on VMS.  Now you seem to be saying some device youE > purchased at some PeeCee outlet (sorry, never heard of Fry) doesn'tt  > Is a chain of generalized electronics, computer hardware, homeD appliance, whatever type stores out here in the wonderful southwest.F Well I don't know if they're anywhere other than the southwest (I know? they're here in Arizona). Think of it as a cross between Edmund F Scientific (not that I've been in there the last 20 years or even knowD if they ever went out of business) and the home appliance section ofE the local Sears. You can get any thing from a volt meter to a washingaG machine to the latest Spice Girls CD (if you're into the Spice Girls ort whatever they're called).S  C In any event not something that someone located half way across thecD planet (or country for that matter) is likely to be familiar with...   > --C > VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)h TMESIS(dot)COM >eE > city, n., 1. a place where trees are cut down and streets are namedl after them.. >a     Sent via Deja.comd http://www.deja.com/   ------------------------------  # Date: Thu, 25 Jan 2001 00:51:28 GMTn* From: Alan E. Feldman <alan48@my-deja.com>* Subject: Re: Why is the tape write locked?) Message-ID: <94ntae$7mi$1@nnrp1.deja.com>o  2 In article <94ln5u$7fk$1@slb3.atl.mindspring.net>,%   "NoSpam" <NoSpam@NoSpam.com> wrote:nE > First, I am working at home, so if anyone can respond via email, my  *REAL*4 > email address is mailto:DOlivastro@ChiResearch.com > 2 > I have a command procedure that looks like this: >d6 > $ mount /nowrite /for /rec:28 /bs:5600 /nounload c1:- > $ mount /for /rec:28 /bs:5600 /nounload c2:s > $ sort /key=(...) c1: c2:i > C > The sort runs for 2 hours, but then, when it is time to write they	 output, InG > get an error from RMS :  "The device (c2:) is currently write locked"b >.E > I know it is *NOT* write-locked, because the message from the mounta commandu+ > does not mention that it is write-locked.g  G You only get that message when the volume is write-locked and you mountu+ it without the /NOWRITE qualifier. Example:    $ 
 $ MOUN/FOR MKM* %MOUNT-I-WRITELOCK, volume is write locked. %MOUNT-I-MOUNTED, AEFTST mounted on _xxx$MKA0: $ DISM/NOUNLO MK $ MOUN/FOR/NOWRITE MKl. %MOUNT-I-MOUNTED, AEFTST mounted on _xxx$MKA0: $ DISM/NOUNLO MK  F I did not change the tape during the above command sequence. Since youE are explicitly asking for "NOWRITE", MOUNT doesn't bother telling you H that the tape is write-locked. Therefore, your tape probably *is* write- locked.e   >l$ > Does anyone know what is going on?  
 See above.   --F NOTE: If you wish to e-mail me, please do NOT use the deja address. ItD is not a valid address. Instead, use the address below, removing the long wrong part first. Thanks.   Disclaimer: JMHO Alan E. Feldman  &-)( afeldman@gfigroup.ButItSaidItPrinted.com     Sent via Deja.com  http://www.deja.com/   ------------------------------  # Date: Thu, 25 Jan 2001 03:15:40 GMTn* From: Alan E. Feldman <alan48@my-deja.com>* Subject: Re: Why is the tape write locked?) Message-ID: <94o5oo$et6$1@nnrp1.deja.com>e  ) In article <94ntae$7mi$1@nnrp1.deja.com>,e-   Alan E. Feldman <alan48@my-deja.com> wrote: 4 > In article <94ln5u$7fk$1@slb3.atl.mindspring.net>,' >   "NoSpam" <NoSpam@NoSpam.com> wrote: G > > First, I am working at home, so if anyone can respond via email, myW > *REAL*6 > > email address is mailto:DOlivastro@ChiResearch.com > >h4 > > I have a command procedure that looks like this: > > 8 > > $ mount /nowrite /for /rec:28 /bs:5600 /nounload c1:/ > > $ mount /for /rec:28 /bs:5600 /nounload c2:a > > $ sort /key=(...) c1: c2:d > >wE > > The sort runs for 2 hours, but then, when it is time to write thee > output, ItA > > get an error from RMS :  "The device (c2:) is currently writea locked"e > >yG > > I know it is *NOT* write-locked, because the message from the mount,	 > command - > > does not mention that it is write-locked.    >kC > You only get that message when the volume is write-locked and youe mountt- > it without the /NOWRITE qualifier. Example:s >n > $l > $ MOUN/FOR MKg, > %MOUNT-I-WRITELOCK, volume is write locked0 > %MOUNT-I-MOUNTED, AEFTST mounted on _xxx$MKA0: > $ DISM/NOUNLO MK > $ MOUN/FOR/NOWRITE MKf0 > %MOUNT-I-MOUNTED, AEFTST mounted on _xxx$MKA0: > $ DISM/NOUNLO MK >sH > I did not change the tape during the above command sequence. Since youG > are explicitly asking for "NOWRITE", MOUNT doesn't bother telling you C > that the tape is write-locked. Therefore, your tape probably *is*e write-	 > locked.S  G Oops! I missed it -- well, part of it. Sorry. What I said wasn't wrong,b but it doesn't apply to C2!   @ Try SHOW SYMBOL MOUNT. Maybe there's a hidden /NOWRITE in there.  D Add a SHOW DEVICE C2 /FULL command before and after the SORT command@ and see what its output says about C2 being write-locked or not.  7 Can you post actual commands and actual error messages?t  E Maybe the write-lock feeler in tape drive c2 is defective and changesc its "reading" after 2 hours.   > >d& > > Does anyone know what is going on?   [...]s   --F NOTE: If you wish to e-mail me, please do NOT use the deja address. ItD is not a valid address. Instead, use the address below, removing the long wrong part first. Thanks.s   Disclaimer: JMHO Alan E. Feldman  &-)( afeldman@gfigroup.ButItSaidItPrinted.com     Sent via Deja.como http://www.deja.com/   ------------------------------  # Date: Wed, 24 Jan 2001 17:49:00 GMTy/ From: "Tom Wade" <t.wade@vms.eurokom.ie.nospam>o! Subject: Re: Wizard Disappears!!! . Message-ID: <gAEb6.4758$r17.14618@news.iol.ie>  K Hoff Hoffman wrote in message <93g05j$dcs$1@mailint03.im.hou.compaq.com>...iJ >  That particular stuffed pachyderm is the unofficial OpenVMS EngineeringI >  group mascot, and has a long history here in OpenVMS -- the mascot wast& >  a gift from a SIG some years ago...    D Might I suggest that a White Elephant is more suitable as a logo for@ Digital Unix, er OSF/1 er Ultrix er Tru64 or whatever it's beingE called this week (the guy jumping off the cliff in the graphic on theeE same page could represent the unfortunate user that was stupid enoughe* to migrate from VMS to Berkeley Hackware).  L ---------------------------------------------------------------------------- --A Tom Wade    | EMail: T.Wade@vms.eurokom.ie  (all domain mailers).iG EuroKom     | X400:  g=tom;s=wade;o=eurokom;p=eurokom;a=eirmail400;c=iex& 30, Dale Rd | Tel:   +353 (1) 278-7878& Stillorgan  | Fax:   +353 (1) 278-78793 Co Dublin   | Disclaimer:  This is not a disclaimere@ Ireland     | Tip:         "Friends don't let friends do Unix !"   ------------------------------  % Date: Wed, 24 Jan 2001 14:13:15 -0800c) From: Wayne Holland <wholland@tscnet.com> ! Subject: Re: Wizard Disappears!!!iO Message-ID: <693B3793F6270F6E.B1C5EB77CD37B4FA.8D28135E2B788A5F@lp.airnews.net>p   Tom Wade wrote:e > M > Hoff Hoffman wrote in message <93g05j$dcs$1@mailint03.im.hou.compaq.com>...cL > >  That particular stuffed pachyderm is the unofficial OpenVMS EngineeringK > >  group mascot, and has a long history here in OpenVMS -- the mascot wasi( > >  a gift from a SIG some years ago... > F > Might I suggest that a White Elephant is more suitable as a logo forB > Digital Unix, er OSF/1 er Ultrix er Tru64 or whatever it's beingG > called this week (the guy jumping off the cliff in the graphic on the G > same page could represent the unfortunate user that was stupid enoughr, > to migrate from VMS to Berkeley Hackware). > N > ---------------------------------------------------------------------------- > --C > Tom Wade    | EMail: T.Wade@vms.eurokom.ie  (all domain mailers).eI > EuroKom     | X400:  g=tom;s=wade;o=eurokom;p=eurokom;a=eirmail400;c=ie>( > 30, Dale Rd | Tel:   +353 (1) 278-7878( > Stillorgan  | Fax:   +353 (1) 278-78795 > Co Dublin   | Disclaimer:  This is not a disclaimercB > Ireland     | Tip:         "Friends don't let friends do Unix !"   Hi Tom! F      Boy, are you right!  I'm running Solaris now and can't even get aE straight answer out of the docs, faqs, or whatever on getting a localo! printer to work thru ghostscript!s: I have to fight the O/S more than I can get any work done.   ------------------------------  # Date: Thu, 25 Jan 2001 06:18:45 GMT   From: jlsue <jlsuexxxz@home.com>! Subject: Re: Wizard Disappears!!! 8 Message-ID: <updv6t8tv5rsgbidjgp8f0a6rcnv05v9u9@4ax.com>  1 On Wed, 24 Jan 2001 14:13:15 -0800, Wayne Hollandh <wholland@tscnet.com> wrote:    G >     Boy, are you right!  I'm running Solaris now and can't even get a F >straight answer out of the docs, faqs, or whatever on getting a local" >printer to work thru ghostscript!; >I have to fight the O/S more than I can get any work done.   C Hey Andrew, are you reading this?  By your calculating methods, SuntD can't say that their systems, in general, are easy and reliable.  We. have evidence right here that that isn't true.   ------------------------------    Date: 24 Jan 2001 12:19:57 -0500* From: young_r@eisner.decus.org (Rob Young), Subject: Writes flushed to Disk on Dismount?+ Message-ID: <JuF9UznclVol@eisner.decus.org>t  G 	Are writes flushed from controller (HSJ/HSZ/HSG) cache to disk when a .? 	disk is dismounted?  Maybe the obvious answer is "absolutely!"09 	I don't feel like tracking that down and thought perhapsu' 	someone knows the answer at the ready.   
 			Thanks,   				Rob@   ------------------------------  % Date: Wed, 24 Jan 2001 20:14:54 -0800N! From: Koloth <koloth@tmisnet.com> 0 Subject: Re: Writes flushed to Disk on Dismount?+ Message-ID: <3A6FA83E.88F5751E@tmisnet.com>   Q I would not count on it.  I think you have to wait for the Cache timeout (defaulte is 10 seconds) on the controller.   Regards:   Cass Witkowski       Rob Young wrote:  O >         Are writes flushed from controller (HSJ/HSZ/HSG) cache to disk when aoH >         disk is dismounted?  Maybe the obvious answer is "absolutely!"B >         I don't feel like tracking that down and thought perhaps0 >         someone knows the answer at the ready. > ! >                         Thanks,n >n% >                                 Robg   ------------------------------    Date: 25 Jan 2001 00:13:16 -0500* From: young_r@eisner.decus.org (Rob Young)0 Subject: Re: Writes flushed to Disk on Dismount?+ Message-ID: <6IdfnIS5xtKj@eisner.decus.org>o  O In article <3A6FA83E.88F5751E@tmisnet.com>, Koloth <koloth@tmisnet.com> writes:uS > I would not count on it.  I think you have to wait for the Cache timeout (defaultm > is 10 seconds) > on the controller. >   @ 	But I want to count on it.  Here's the scenario... as you point: 	out CACHE_FLUSH_TIMER is default of 10 seconds.  But what= 	about the poor sap that has it set to 120 seconds?  He makesv: 	sure all file streams are closed (or they are anyhow) and8 	dismounts a drive.  If writeback cache is turned on, is; 	there a mechanism that signals the controller to flush thek7 	cache?  If not, where is the big doc warning that saysi? 	"don't do that" or "beware of?"  Same question ... more stuff.d   				Rob     	 > Regards- >  > Cass Witkowski >  >  >  > Rob Young wrote: > P >>         Are writes flushed from controller (HSJ/HSZ/HSG) cache to disk when aI >>         disk is dismounted?  Maybe the obvious answer is "absolutely!"nC >>         I don't feel like tracking that down and thought perhapsn1 >>         someone knows the answer at the ready.  >>" >>                         Thanks, >>& >>                                 Rob >    ------------------------------  % Date: Wed, 24 Jan 2001 20:55:40 +0000C+ From: "antonio.carlini" <arcarlini@iee.org>oC Subject: Re: [Q]: Resetting privileged password on a DECserver 200? ' Message-ID: <3A6F414C.43332ECF@iee.org>/   "David J. Dachtera" wrote: > ; > "Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515" wrote:oL > >         I'm  still  somewhat  confused  about  _setting_  the  privilegeL > >     password.  Can you only set it after it's been cleared via the resetL > >     described  above?   I can't find a command in the docs or help  that > >     seems to do it.iC > AFAIK, you can change the privileged password only after you havesE > successfully SET PRIV'd (i.e., you must know the current privilegeda > password).  . Not only that, but I (think I) recall that the& online HELP command does not show some$ priv'd commands unless you have done) a SET SYSTEM to prove your worthiness ...T   Antonio    -- m   ---------------:- Antonio Carlini             arcarlini@iee.orge   ------------------------------  % Date: Wed, 24 Jan 2001 16:44:38 -0500 % From: "Brian Tillman" <tillman_brian>lC Subject: Re: [Q]: Resetting privileged password on a DECserver 200?b$ Message-ID: <3a6f4bdf$1@news.si.com>  / >Not only that, but I (think I) recall that the ' >online HELP command does not show some2% >priv'd commands unless you have donee* >a SET SYSTEM to prove your worthiness ...  D I don't know about other DECservers, but for a DS200, the command isI DEFINE/SET PRIVILEGED, not SET SYSTEM.  The command to set the privileged K password is DEFINE/SET SERVER PRIVILEGED PASSWORD.  See pages 2-85 and 2-93r6 of the "Terminal Server Commands and Messages" manual. --A Brian Tillman                   Internet: tillman_brian at si.com A Smiths Aerospace                          tillman at swdev.si.com = 3290 Patterson Ave. SE, MS      Addresses modified to prevento< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------   Date: 24 Jan 2001 17:10:04 PSTT From: Fairfield@SLAC.Stanford.EDU (Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515)C Subject: Re: [Q]: Resetting privileged password on a DECserver 200?D3 Message-ID: <g9zW7VBrXZbo@mccdev.slac.stanford.edu>w  % In article <3a6f4bdf$1@news.si.com>, m,     	"Brian Tillman" <tillman_brian> writes: [...]yF > I don't know about other DECservers, but for a DS200, the command isK > DEFINE/SET PRIVILEGED, not SET SYSTEM.  The command to set the privilegedrM > password is DEFINE/SET SERVER PRIVILEGED PASSWORD.  See pages 2-85 and 2-93f8 > of the "Terminal Server Commands and Messages" manual.  H         Thanks, Brian, I should  have  thought  to  do the SET PRIVILEGEH     before  doing  the SERVER HELP SET SERVER ... command.   I  wouldn'tH     have been so confused.  As for the manual  you  reference.   Manual?H     What  manual?   We  don't  need  no  steenkin'  manuals  here!   ;-pH     [Actually, if we ever had  them,  other than the Guide, they've long5     since been relegated to future archaeologists...]0           -Ken -- 5M  Kenneth H. Fairfield            |  Internet: Fairfield@SLC.Slac.Stanford.Edur:  SLAC, 2575 Sand Hill Rd, MS 46  |  Voice:    650-926-2924:  Menlo Park, CA  94025           |  FAX:      650-926-3515N  -----------------------------------------------------------------------------B  These opinions are mine, not SLAC's, Stanford's, nor the DOE's...   ------------------------------   End of INFO-VAX 2001.049 ************************