1 INFO-VAX	Sat, 27 May 2006	Volume 2006 : Issue 292       Contents: Re: a small problem $ Re: Compaq board member sent to jail$ Re: Compaq board member sent to jail$ Re: Compaq board member sent to jail$ Re: Compaq board member sent to jail	 Re: DN-11 ! Re: Educating potential VMS users ! Re: Educating potential VMS users ! Re: Educating potential VMS users ! Re: Educating potential VMS users ! Re: Educating potential VMS users ! Re: Educating potential VMS users ! Re: Educating potential VMS users ! Re: Educating potential VMS users ! Re: Educating potential VMS users P Re: Educating potential VMS users (was: Re: Unix runs faster, maybe  (was: Re: EP Re: Educating potential VMS users (was: Re: Unix runs faster, maybe (was: Re: EdP Re: Educating potential VMS users (was: Re: Unix runs faster, maybe (was: Re: Ed" Re: Fixing a Corrupt PCSI Database/ Re: GFloat Rounding issue in DEC Basic V1.2-000   Re: Madgoat MX Exchange for VMS.B Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't deleteB Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't deleteB Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't deleteB Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't deleteB Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't delete Re: Question Re: Question0 Re: speeding up LAVC with switch instead of hub?0 Re: speeding up LAVC with switch instead of hub?0 Re: speeding up LAVC with switch instead of hub?D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)D Re: Unix runs faster, maybe (was: Re: Educating potential VMS users)  F ----------------------------------------------------------------------    Date: 26 May 2006 15:32:56 -0700# From: "H Vlems" <hvlems@freenet.de>  Subject: Re: a small problemC Message-ID: <1148682776.922852.234130@i40g2000cwc.googlegroups.com>   @ So you did something like this BACKUP/IMAGE/VERIFY DKA200: DIA0:  followed by a >>> B dia0  right?C If the VS3100 could not establish a connection then I'd suspect the  network first.6  Cable problem, transceiver down, something like that.G Alternatively, your systartup.com procedure has hardcoded references to 2 DKA200 and the 4100 did not boot well as a result.6 But a  network can die even if you didn't touch it....   ------------------------------  % Date: Fri, 26 May 2006 16:50:54 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> - Subject: Re: Compaq board member sent to jail , Message-ID: <44776A2A.BEC450A4@teksavvy.com>   Bill Gunshannon wrote:I > There is virtually no chance that a pair of white-collar criminals like I > this would be considered a threat to flee.  If so, they would have done  > it before the trial.    C Maybe they expected that they could be declared not-guilty and then 8 continue to live comfortably in their existing mansions.  @ These people have probably moved a lot of money to offshore bankD accounts, and they could easily flee to some safe haven that doesn'tG have an extradition treaty with USA.  I would say these people would be F a greater flight risk since they have the means to fly on private jets" and leave the country undetected.   D However, doing so would cause a lot of trouble to his friends in the@ white house. (they are already in way over their heads with intlB affairs). Are prisons private enterprises in the USA ? If so, he'sD probably busy buying the prison he will be staying in and having his cell refurbished :-)   ------------------------------  # Date: Fri, 26 May 2006 23:25:24 GMT ) From: "John Vottero" <JVottero@mvpsi.com> - Subject: Re: Compaq board member sent to jail : Message-ID: <E7Mdg.374$VE1.216@newssvr14.news.prodigy.com>  9 "Andrew" <andrew_harrison@symantec.com> wrote in message  < news:1148659275.899498.168920@38g2000cwa.googlegroups.com... >  > Tom Linden wrote: H >> On Fri, 26 May 2006 05:30:25 -0700, John Vottero <JVottero@mvpsi.com>	 >> wrote:  >>H >> > He's been convicted.  He's going to have to sit in prison while his >> > lawyers >> > appeal. >> >>H >> I don't think so, Bernie Ebbers is still a free man, while his appeal >> is process. > F > He hasn't appealed yet, he is currently on bail awaiting sentencing.F > When the Judge hands down the sentence then he may decide to appeal.  J Before bail would be considered, the convicted felon and his lawyer would H have to convince a judge that there is a high probablity that they will , succeed in their appeal.  That's not likely.  F > And of course the Judge can refuse bail if he thinks that there is a- > high chance that the two accused will flee.    ------------------------------  % Date: Fri, 26 May 2006 20:02:09 -0400 ' From: Dave Froble <davef@tsoft-inc.com> - Subject: Re: Compaq board member sent to jail / Message-ID: <YqGdnWX855SBC-rZ4p2dnA@libcom.com>    JF Mezei wrote:  > Bill Gunshannon wrote:J >> There is virtually no chance that a pair of white-collar criminals likeJ >> this would be considered a threat to flee.  If so, they would have done >> it before the trial.  > E > Maybe they expected that they could be declared not-guilty and then : > continue to live comfortably in their existing mansions.  G If I had some number of millions available to me, and I could get them  F out of the country, where they couldn't be touched, I don't know that G I'd even wait for a verdict.  After a verdict, they might not have the   millions and mansions.  B > These people have probably moved a lot of money to offshore bankF > accounts, and they could easily flee to some safe haven that doesn'tI > have an extradition treaty with USA.  I would say these people would be H > a greater flight risk since they have the means to fly on private jets$ > and leave the country undetected.   G I don't know about undetected.  I'd sure be doing so.  Be nice to keep  
 the jet also.   F > However, doing so would cause a lot of trouble to his friends in the > white house.  G And this matters in which way?  If the choice is prison, dubya and his   cronies are on their own.   5 > (they are already in way over their heads with intl 8 > affairs). Are prisons private enterprises in the USA ?   No!   
 > If so, he's F > probably busy buying the prison he will be staying in and having his > cell refurbished :-)  I The loss of freedom to come and go is a real basic thing.  Even though I  I don't go very often, I know I can do so.  Not being able to would eat at  , you, even if you didn't want to go anyplace.   --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Fri, 26 May 2006 19:06:30 -0500 % From: Dan Foster <usenet@evilphb.org> - Subject: Re: Compaq board member sent to jail 5 Message-ID: <slrne7f606.3uh.usenet@zappy.catbert.org>   X In article <YqGdnWX855SBC-rZ4p2dnA@libcom.com>, Dave Froble <davef@tsoft-inc.com> wrote: > 6 >> (they are already in way over their heads with intl9 >> affairs). Are prisons private enterprises in the USA ?  >  > No!   A I hate to be off-topic, but a number of prisons are indeed run by  private contractors.  # I wonder if they use VMS there. :-)   - (And if so, that'd be the ultimate in irony.)    -Dan   ------------------------------    Date: 26 May 2006 15:24:58 -0700# From: "H Vlems" <hvlems@freenet.de>  Subject: Re: DN-11C Message-ID: <1148682298.414640.196200@i40g2000cwc.googlegroups.com>   < Well that implicitly answered the OP question, didn't it....   ------------------------------   Date: 26 May 2006 18:17:01 GMT From: healyzh@aracnet.com * Subject: Re: Educating potential VMS users, Message-ID: <e57gmt016u0@enews3.newsguy.com>  , Doug Phillips <dphill46@netscape.net> wrote: > healyzh@aracnet.com wrote:0 > > Doug Phillips <dphill46@netscape.net> wrote:K > > > Before you go further, critique the books that are already available. K > > > To save you the trouble of actually looking up their titles, here are : > > > the ones at the link provided by Richard B. Gilbert: > > 4 > > > Duffy, Michael D. Getting Started with OpenVMS& > > > Digital Press, 2002, ~300 pages. > > > ISBN 1-55558-279-6 > > K > > One serious problem is that of all the books you listed, only the one I I > > left above is likely to still be available.  I belive all of the rest I > > are long out of print, and will take some serious looking in order to D > > find a copy, especially an affordable copy.  A good example is II > > recently saw a copy of "Writing Real Programs in DCL, 2nd Ed." listed H > > for $200!  Though at the same time, the ones on using VMS tend to be' > > dirt cheap, *IF* you can find them.  > >  > >       Zane    I > Go to <http://www.amazon.com> or <http://www.barnesandnoble.com> and do G > a search for openvms. Pick one of those. VMS books really aren't that F > hard to find but you do need to look for them. To save everyone evenE > that amount of work, here's the two-page search result from amazon:    > <http://tinyurl.com/jru37>  F > That's about all the help I can offer. Anyone who still thinks thereG > are no VMS books available must be considered hopeless at this point.   J I'm well aware that there are books available, if you go hunting, and thatK some of them are even available *new*.  I was simply pointing out that most K of the ones that you listed are no longer readily available without hunting H up used copies.  The really desirable books that are only available used, practically require selling a kidney to buy.   			Zane    ------------------------------   Date: 26 May 2006 19:58:06 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)* Subject: Re: Educating potential VMS users, Message-ID: <4dp4ueF1bol77U1@individual.net>  C In article <1148672484.632495.319950@g10g2000cwb.googlegroups.com>, 0 	"Doug Phillips" <dphill46@netscape.net> writes: > heal...@aracnet.com wrote:/ >> Doug Phillips <dphill46@netscape.net> wrote:  >> > healyzh@aracnet.com wrote: 3 >> > > Doug Phillips <dphill46@netscape.net> wrote: N >> > > > Before you go further, critique the books that are already available.N >> > > > To save you the trouble of actually looking up their titles, here are= >> > > > the ones at the link provided by Richard B. Gilbert:  >> > >7 >> > > > Duffy, Michael D. Getting Started with OpenVMS ) >> > > > Digital Press, 2002, ~300 pages.  >> > > > ISBN 1-55558-279-6  >> > >N >> > > One serious problem is that of all the books you listed, only the one IL >> > > left above is likely to still be available.  I belive all of the restL >> > > are long out of print, and will take some serious looking in order toG >> > > find a copy, especially an affordable copy.  A good example is I L >> > > recently saw a copy of "Writing Real Programs in DCL, 2nd Ed." listedK >> > > for $200!  Though at the same time, the ones on using VMS tend to be * >> > > dirt cheap, *IF* you can find them. >> > > >> > >       Zane  >> >>L >> > Go to <http://www.amazon.com> or <http://www.barnesandnoble.com> and doJ >> > a search for openvms. Pick one of those. VMS books really aren't thatI >> > hard to find but you do need to look for them. To save everyone even H >> > that amount of work, here's the two-page search result from amazon: >> >> > <http://tinyurl.com/jru37>  >>I >> > That's about all the help I can offer. Anyone who still thinks there J >> > are no VMS books available must be considered hopeless at this point. >>M >> I'm well aware that there are books available, if you go hunting, and that N >> some of them are even available *new*.  I was simply pointing out that mostN >> of the ones that you listed are no longer readily available without huntingK >> up used copies.  The really desirable books that are only available used / >> practically require selling a kidney to buy.  >> >  > I > Zane, I didn't mean to attack your post. The list I posted was simply a F > cut & paste from the link posted by someone else.  If my laziness in@ > not checking on the currency of every book caused confusion, I
 > appologize.  > F > My point was that books are available from amazon.com as well as theH > other popular book sites and it doesn't really take a *lot* of looking > to find them.   E Just because they show up in an Amazon search don't assume that means D they actually exist.  A large number of the books returned by AmazonJ searches are from third parties. My last foray into that quagmire resulted6 in no book and 2 months fighting to get my money back.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 26 May 2006 12:41:24 -0700- From: "Doug Phillips" <dphill46@netscape.net> * Subject: Re: Educating potential VMS usersC Message-ID: <1148672484.632495.319950@g10g2000cwb.googlegroups.com>    heal...@aracnet.com wrote:. > Doug Phillips <dphill46@netscape.net> wrote: > > healyzh@aracnet.com wrote:2 > > > Doug Phillips <dphill46@netscape.net> wrote:M > > > > Before you go further, critique the books that are already available. M > > > > To save you the trouble of actually looking up their titles, here are < > > > > the ones at the link provided by Richard B. Gilbert: > > > 6 > > > > Duffy, Michael D. Getting Started with OpenVMS( > > > > Digital Press, 2002, ~300 pages. > > > > ISBN 1-55558-279-6 > > > M > > > One serious problem is that of all the books you listed, only the one I K > > > left above is likely to still be available.  I belive all of the rest K > > > are long out of print, and will take some serious looking in order to F > > > find a copy, especially an affordable copy.  A good example is IK > > > recently saw a copy of "Writing Real Programs in DCL, 2nd Ed." listed J > > > for $200!  Though at the same time, the ones on using VMS tend to be) > > > dirt cheap, *IF* you can find them.  > > >  > > >       Zane >  > K > > Go to <http://www.amazon.com> or <http://www.barnesandnoble.com> and do I > > a search for openvms. Pick one of those. VMS books really aren't that H > > hard to find but you do need to look for them. To save everyone evenG > > that amount of work, here's the two-page search result from amazon:  >  > > <http://tinyurl.com/jru37> > H > > That's about all the help I can offer. Anyone who still thinks thereI > > are no VMS books available must be considered hopeless at this point.  > L > I'm well aware that there are books available, if you go hunting, and thatM > some of them are even available *new*.  I was simply pointing out that most M > of the ones that you listed are no longer readily available without hunting J > up used copies.  The really desirable books that are only available used. > practically require selling a kidney to buy. >     G Zane, I didn't mean to attack your post. The list I posted was simply a D cut & paste from the link posted by someone else.  If my laziness in> not checking on the currency of every book caused confusion, I appologize.   D My point was that books are available from amazon.com as well as theF other popular book sites and it doesn't really take a *lot* of looking
 to find them.   G On topic to the OP, Bill G., his request was for intro-level books, and & there are some of those still current.  D My other point is that publishers track book sales and demand and ifG there is sufficient demand for a book they will print more. If there is G sufficient demand for information not covered by an existing book, they E will work to get a book published. No demand, no sales, no publishing A profit, no book. So, if the VMS community believes that VMS has a B viable future, then there are things we can do to help broaden its@ interest and maybe build a demand for more current publications.  G My personal feeling is that for anyone actually using VMS, its official F documentation is so good and complete at every almost every level thatF only a few highly technical or specialized books might ever be needed.G For handy reference, a "nutshell" *type* of book would be handy to have 4 around. (a real one just isn't going to happen soon)  D People who work with other OS's and love computers of all kinds (theG "real geek") might buy a "nutshell" type of book out of curiosity if it F were sitting in the "Computers" section of the local bookstore. They'd0 probably at least pick it up and look though it.  G Such a generally available newer book (or new revision) would also make B it more obvioius that VMS is still alive, and if the book provided? links to the "free account" sites for hands-on experience while D learning, and had Appendices explaining the hobbiest program and theG emulators (CHARON, SIMH, ...) that would remove the "like to know about - it but can't run it on my computer" argument.   C But, stores won't stock a book like that unless they feel there's a E demand. It's possible for us VMS fans to show that there is a demand. > Isn't it time we did something positive rather than sit around) complaining about HP's lack of marketing?   ? Just some thoughts, not directed to anyone in particular but to G everyone with an interest in VMS. It'd be interesting to hear some more 1 of our resident authors' thoughts on the subject.    ------------------------------    Date: 26 May 2006 13:35:27 -0700- From: "Doug Phillips" <dphill46@netscape.net> * Subject: Re: Educating potential VMS usersC Message-ID: <1148675727.941253.128030@y43g2000cwc.googlegroups.com>    Bill Gunshannon wrote:E > In article <1148672484.632495.319950@g10g2000cwb.googlegroups.com>, 2 > 	"Doug Phillips" <dphill46@netscape.net> writes: > > heal...@aracnet.com wrote:1 > >> Doug Phillips <dphill46@netscape.net> wrote: ! > >> > healyzh@aracnet.com wrote: 5 > >> > > Doug Phillips <dphill46@netscape.net> wrote: P > >> > > > Before you go further, critique the books that are already available.P > >> > > > To save you the trouble of actually looking up their titles, here are? > >> > > > the ones at the link provided by Richard B. Gilbert:  > >> > >9 > >> > > > Duffy, Michael D. Getting Started with OpenVMS + > >> > > > Digital Press, 2002, ~300 pages.  > >> > > > ISBN 1-55558-279-6  > >> > >P > >> > > One serious problem is that of all the books you listed, only the one IN > >> > > left above is likely to still be available.  I belive all of the restN > >> > > are long out of print, and will take some serious looking in order toI > >> > > find a copy, especially an affordable copy.  A good example is I N > >> > > recently saw a copy of "Writing Real Programs in DCL, 2nd Ed." listedM > >> > > for $200!  Though at the same time, the ones on using VMS tend to be , > >> > > dirt cheap, *IF* you can find them. > >> > > > >> > >       Zane  > >> > >>N > >> > Go to <http://www.amazon.com> or <http://www.barnesandnoble.com> and doL > >> > a search for openvms. Pick one of those. VMS books really aren't thatK > >> > hard to find but you do need to look for them. To save everyone even J > >> > that amount of work, here's the two-page search result from amazon: > >>! > >> > <http://tinyurl.com/jru37>  > >>K > >> > That's about all the help I can offer. Anyone who still thinks there L > >> > are no VMS books available must be considered hopeless at this point. > >>O > >> I'm well aware that there are books available, if you go hunting, and that P > >> some of them are even available *new*.  I was simply pointing out that mostP > >> of the ones that you listed are no longer readily available without huntingM > >> up used copies.  The really desirable books that are only available used 1 > >> practically require selling a kidney to buy.  > >> > >  > > K > > Zane, I didn't mean to attack your post. The list I posted was simply a H > > cut & paste from the link posted by someone else.  If my laziness inB > > not checking on the currency of every book caused confusion, I > > appologize.  > > H > > My point was that books are available from amazon.com as well as theJ > > other popular book sites and it doesn't really take a *lot* of looking > > to find them.  > G > Just because they show up in an Amazon search don't assume that means F > they actually exist.  A large number of the books returned by AmazonL > searches are from third parties. My last foray into that quagmire resulted8 > in no book and 2 months fighting to get my money back. >   C So you don't like Amazon? Okay. God forbid you should actually look + anyplace else. You can run down to Powells:   U <http://www.powells.com/biblio?show=TRADE%20PAPER:NEW:1555582796:34.99;show_locs=yes>   G and actually pick this one off the shelf. Oh, they'll order one for you F if I get there first, and so will Barnes & Noble or barnesandnoble.comB or any other frigging bookstore pretty much anywhere in the world.  E Hey, Bill, I'm not trying to fight with you but apparently you're not # really looking for a book, are you?    ------------------------------   Date: 26 May 2006 22:10:34 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)* Subject: Re: Educating potential VMS users, Message-ID: <4dpcmqF1bm5tpU1@individual.net>  C In article <1148675727.941253.128030@y43g2000cwc.googlegroups.com>, 0 	"Doug Phillips" <dphill46@netscape.net> writes: > G > Hey, Bill, I'm not trying to fight with you but apparently you're not % > really looking for a book, are you?   D I'm not.  I was trying to get people to consider other possibilitiesD to bring VMS back into the public eye.  My whole point being that ifA the book isn't sitting on the shelf next to the Linux/BSD/Solaris B books people will continue to believe that VMS is a former OS thatB no longer exists.  How can you expect someone who has never heard F of VMS to go looking for books on that which they have never heard of.  D No more posts on this from me.  I give up.  Let the chips fall whereF they may.  As I have said, I don't make my living with VMS and I don't# expect I ever will at this point.      bill     --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------   Date: 27 May 2006 01:11:09 GMT From: healyzh@aracnet.com * Subject: Re: Educating potential VMS users, Message-ID: <e588vd029vo@enews4.newsguy.com>  ) Bill Gunshannon <bill@cs.uofs.edu> wrote: G > Just because they show up in an Amazon search don't assume that means F > they actually exist.  A large number of the books returned by AmazonL > searches are from third parties. My last foray into that quagmire resulted8 > in no book and 2 months fighting to get my money back.  H This brings up a question of mine.  Was "OpenVMS File System Internals" L by Brian Schenkenberger actually published?  I was actually going to use theJ original as an example of one you just about have to sell a Kidney to get,K when I found out about this.  The HP website says it was published in 2003, J unfortunatly based on Amazon it doesn't appear to be available.  If it was/ published how did I miss knowing about that :^(    		Zane   ------------------------------   Date: 27 May 2006 01:30:13 GMT From: healyzh@aracnet.com * Subject: Re: Educating potential VMS users, Message-ID: <e58a35026q1@enews2.newsguy.com>  , Doug Phillips <dphill46@netscape.net> wrote:I > Zane, I didn't mean to attack your post. The list I posted was simply a F > cut & paste from the link posted by someone else.  If my laziness in@ > not checking on the currency of every book caused confusion, I
 > appologize.   I I was probably a bit touchy thanks to lack of sleep, sorry about that.  I % generally try to be a bit more civil.   F > My point was that books are available from amazon.com as well as theH > other popular book sites and it doesn't really take a *lot* of looking > to find them.   J Part of my problem is I hate using Amazon, I'd much rather pop around to aL locally owned bookstore and buy the books I want.  It does help that for me,K the main locally owned bookstore still around is Powells.  Of course when I G bought Alan's Webserver book, it and the rest of the OpenVMS books were I misfiled in the Unix section and I just happened to get lucky and spot it @ (it thier Technical Bookstore they're in the Mainframe section).  I > On topic to the OP, Bill G., his request was for intro-level books, and ( > there are some of those still current.   Agreed.   F > My other point is that publishers track book sales and demand and ifI > there is sufficient demand for a book they will print more. If there is I > sufficient demand for information not covered by an existing book, they G > will work to get a book published. No demand, no sales, no publishing C > profit, no book. So, if the VMS community believes that VMS has a D > viable future, then there are things we can do to help broaden itsB > interest and maybe build a demand for more current publications.  I Interesting point.  I just happen to have some Amazon Gift Certificates I K need to use...  Is "OpenVMS System Management Guide", by Lawrence Baldwin,  H Steve Hoffman, David Miller worth getting?  Unfortunately I've not had aL chance to thumb through a copy.  The book I really want to get a copy of is,E "The Minimum You Need to Know to be an OpenVMS Developer", but so far + haven't had the spare change to get a copy.   I > My personal feeling is that for anyone actually using VMS, its official H > documentation is so good and complete at every almost every level thatH > only a few highly technical or specialized books might ever be needed.I > For handy reference, a "nutshell" *type* of book would be handy to have 6 > around. (a real one just isn't going to happen soon)  J Exactly what I've been saying.  Though in the case of the "Nutshell" book,L I'd really like to see HP offer a pair for sale, one from a User Standpoint,J the other from an Administrators.  This is my "Base Base Docset" point :^)   			Zane    ------------------------------   Date: 27 May 2006 01:39:53 GMT From: healyzh@aracnet.com * Subject: Re: Educating potential VMS users, Message-ID: <e58al9126q1@enews2.newsguy.com>  , Doug Phillips <dphill46@netscape.net> wrote:E > So you don't like Amazon? Okay. God forbid you should actually look - > anyplace else. You can run down to Powells:   W > <http://www.powells.com/biblio?show=TRADE%20PAPER:NEW:1555582796:34.99;show_locs=yes>   I > and actually pick this one off the shelf. Oh, they'll order one for you H > if I get there first, and so will Barnes & Noble or barnesandnoble.comD > or any other frigging bookstore pretty much anywhere in the world.  L Actually Powells is where Amazon gets a lot of their books, or at least usedJ to.  They also always have a few OpenVMS books in stock, most of mine haveH been purchased either from them, or eBay (for those hard to find books).  L I see your link is to "Getting Started with OpenVMS: A Guide for New Users",H which just proves what I said in my earlier post, they don't know how to4 file VMS books, as this one is in "Database Design".  K Of course it's not just VMS books, a few months ago, I was in there looking H for a couple books on Valve Amplifiers (Tubes), and finally had to ask aH clerk where on earth they were hiding the ones I was after.  He said theI books on the subject are spread all over in places that don't really make 	 sense :^(   J Anyone visiting Portland should consider spending the better part of a dayJ in both the main store (an entire city block of new and used books that isF something like 3 stories high in parts), and the technical bookstore aL couple blocks away.  I try to avoid going more than once or twice a year :^)   		Zane   ------------------------------  # Date: Sat, 27 May 2006 03:33:07 GMT  From: dittman@dittman.net * Subject: Re: Educating potential VMS users) Message-ID: <TLPdg.1837$U_2.954@trnddc05>    healyzh@aracnet.com wrote:+ > Bill Gunshannon <bill@cs.uofs.edu> wrote: I > > Just because they show up in an Amazon search don't assume that means H > > they actually exist.  A large number of the books returned by AmazonN > > searches are from third parties. My last foray into that quagmire resulted: > > in no book and 2 months fighting to get my money back.  J > This brings up a question of mine.  Was "OpenVMS File System Internals" N > by Brian Schenkenberger actually published?  I was actually going to use theL > original as an example of one you just about have to sell a Kidney to get,  I Unfortunately Brian's new revision of _OpenVMS File System Internals_ was  never completed. --   Eric Dittman dittman@dittman.net    ------------------------------  % Date: Fri, 26 May 2006 22:37:17 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> Y Subject: Re: Educating potential VMS users (was: Re: Unix runs faster, maybe  (was: Re: E , Message-ID: <4477BB50.A049A6A6@teksavvy.com>   Bob Harris wrote: B > Of course not.  Lets start a My Editor is Better than Yours War! > Yea!!!!  :-)    C WPSPLUS beats TECO hands down. It is so good that EVE even tried to  emulate it (SET KEYPAD WPS :-)   ------------------------------  # Date: Sat, 27 May 2006 00:41:47 GMT 9 From: Bob Harris <nospam.News.Bob@remove.Smith-Harris.us> Y Subject: Re: Educating potential VMS users (was: Re: Unix runs faster, maybe (was: Re: Ed D Message-ID: <nospam.News.Bob-80E96B.20415626052006@news.verizon.net>  3 In article <Qur1LzKESdrf@eisner.encompasserve.org>, =  koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:   D > In article <Koldg.16$me1.1074@news.uswest.net>, "Michael D. Ober" ( > <obermd.@.alum.mit.edu.nospam> writes: > > N > > That and all the bashing of every other OS in use on the planet.  It's oneF > > thing to compare different OS's, but what frequently passes for OS1 > > comparison here reminds me of a religous war.  > D >    OS is not the only kind of software that tends to inspire "ture >    believers".  B Of course not.  Lets start a My Editor is Better than Yours War!   Yea!!!!  :-)    2                                         Bob Harris   ------------------------------  % Date: Fri, 26 May 2006 21:01:15 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> Y Subject: Re: Educating potential VMS users (was: Re: Unix runs faster, maybe (was: Re: Ed 0 Message-ID: <gqmdnZIWDK5BOerZRVn-qA@comcast.com>   Bob Harris wrote:   5 > In article <Qur1LzKESdrf@eisner.encompasserve.org>, ? >  koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:  >  > D >>In article <Koldg.16$me1.1074@news.uswest.net>, "Michael D. Ober" ( >><obermd.@.alum.mit.edu.nospam> writes: >>M >>>That and all the bashing of every other OS in use on the planet.  It's one E >>>thing to compare different OS's, but what frequently passes for OS 0 >>>comparison here reminds me of a religous war. >>D >>   OS is not the only kind of software that tends to inspire "ture >>   believers". >  > D > Of course not.  Lets start a My Editor is Better than Yours War!   > Yea!!!!  :-) >  > 4 >                                         Bob Harris   EDT forever!!!!  ;-)   ------------------------------  % Date: Fri, 26 May 2006 17:35:15 -0400 + From: Steve Matzura <number6@speakeasy.net> + Subject: Re: Fixing a Corrupt PCSI Database 8 Message-ID: <l0te72h3i6157ve0s9pnqavlh1bbq3mmb6@4ax.com>  , On Sat, 20 May 2006 17:42:58 -0400, JF Mezei% <jfmezei.spamnot@teksavvy.com> wrote:   G >If the database is OK, but warns of a missing product specific file, I F >would fetch the product's original .PCSI file , extract all the files? >from it in a temporary directory and find the product specific G >.pcsi<mumble>  and put them where they belong.  This way, you are sure * >that nothing will mess with your system.   E I believe that's what they're going to wind up doing. I'm not at that D shop any more so cannot say how things are.  It was just told to me,E knowing that I tap into a lot of knowledgeable folks, and that sounds 3 like the right thing to do anyway, given the error.    ------------------------------  # Date: Sat, 27 May 2006 01:05:03 GMT L From: winston@SSRL.SLAC.STANFORD.EDU (Alan Winston - SSRL Central Computing)8 Subject: Re: GFloat Rounding issue in DEC Basic V1.2-0006 Message-ID: <00A56477.603D726F@SSRL.SLAC.STANFORD.EDU>  W In article <4dosu1F1bcom8U1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: * >In article <op.s9524cemlvpiaf@hyrrokkin>,' >	"Tom LINDEN" <tom@kednos.com> writes: L >> On Fri, 26 May 2006 08:40:58 -0700, Bill Gunshannon <bill@cs.uofs.edu>  = >>  	 >> wrote:  >>  H >>> Or to use COBOL which was designed for this task in the first place.$ >>                            POORLY > K >What makes it poorer for financial applications as opposed to PL/I.  COBOL 2 >was specifically designed for these applications,  N As was PL/I, which you may recall was designed to support lessons learned fromD both COBOL and FORTRAN, as well as block-structure stuff from ALGOL.  K It's a pretty decent language, it has the appropriate financial data types, K and it's not hard to learn if you already know a block-structured language.   M (Now, I'm going by a community-college class in PL/I from 25 years ago, but I   quite enjoyed that at the time.)   -- Alan    ------------------------------    Date: 26 May 2006 21:30:34 -02006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)) Subject: Re: Madgoat MX Exchange for VMS. , Message-ID: <4477737a$1@news.langstoeger.at>  c In article <e5700v$49l$1@mailhub227.itcs.purdue.edu>, Chuck Aaron <caaron@ceris.purdue.edu> writes: ; >Can anyone help with how to fix the warning message below? E >26-MAY-2006 09:19:21.33:  MX WARNING: Message queue exceeds allowed  ! >maximum disk space, or disk full    Either free up disk space   > or change the default threshold (IIRC 10% free) to say 1% with  ) $ DEF/SYS/EXE MX_FLQ_DISK_FREE_RESERVED	1   . For more info read MX_DOC:MX_INSTALL_GUIDE.TXT  4 eg. $ DEFINE/SYSTEM/EXEC MX_FLQ_DISK_FREE_ABSOLUTE n   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Fri, 26 May 2006 14:39:47 -0400 ' From: Dave Froble <davef@tsoft-inc.com> K Subject: Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't delete 9 Message-ID: <tbmdnc_IDqsQ1-rZnZ2dnUVZ_tydnZ2d@libcom.com>    rschiemel@gmail.com wrote:! > Compaq Open VMS VAX Version 7.2  > H > It was noticed that the freespace on the system disk was decreasing. A, > search for large files showed a file namedI > SYS$QUEUE_MANAGER.QMAN$JOURNAL;1 which was about 11,000 blocks, existed D > in three separate directories on the system disk, using a total of > about 33,000 blocks. > I > The directories were: [SYS0.SYSCOMMON.SYSEXE], [SYSE.SYSCOMMON.SYSEXE], G > and [VMS$COMMON.SYSEXE]. In an effort to reduce the size of the files # > we issued the following commands:  >> mcr  JBC$COMMAND  > JBC$COMMAND> DIAG 0 7  > H > The result was the original three files disappeared and in their placeE > were three new files each about 1100 blocks in size. We expected to B > find the freespace increased by about 30,000 blocks. Instead the8 > freespace DECREASED by 3,300 blocks (The 3 new files). > * > How do we get those 33,000 blocks free ? >    PURGE    --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Fri, 26 May 2006 15:05:08 -0400 ) From: "Ken Robinson" <kenrbnsn@gmail.com> K Subject: Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't delete G Message-ID: <7dd80f60605261205v27de1b60q9c4e6667948f4f7@mail.gmail.com>   ( ------=_Part_4432_22375965.1148670308543; Content-Type: text/plain; charset=ISO-8859-1; format=flowed + Content-Transfer-Encoding: quoted-printable  Content-Disposition: inline   4 On 5/26/06, Dave Froble <davef@tsoft-inc.com> wrote: >  > rschiemel@gmail.com wrote:# > > Compaq Open VMS VAX Version 7.2  > > J > > It was noticed that the freespace on the system disk was decreasing. A. > > search for large files showed a file namedK > > SYS$QUEUE_MANAGER.QMAN$JOURNAL;1 which was about 11,000 blocks, existed F > > in three separate directories on the system disk, using a total of > > about 33,000 blocks. > > K > > The directories were: [SYS0.SYSCOMMON.SYSEXE], [SYSE.SYSCOMMON.SYSEXE], I > > and [VMS$COMMON.SYSEXE]. In an effort to reduce the size of the files % > > we issued the following commands:  > >> mcr  JBC$COMMAND  > > JBC$COMMAND> DIAG 0 7  > > J > > The result was the original three files disappeared and in their placeG > > were three new files each about 1100 blocks in size. We expected to D > > find the freespace increased by about 30,000 blocks. Instead the: > > freespace DECREASED by 3,300 blocks (The 3 new files). > > , > > How do we get those 33,000 blocks free ? > >  >  > PURGE     C If you do a dir/id you will notice that they are all the same file.    Ken   ( ------=_Part_4432_22375965.1148670308543+ Content-Type: text/html; charset=ISO-8859-1 + Content-Transfer-Encoding: quoted-printable  Content-Disposition: inline   L <br><br><div><span class=3D"gmail_quote">On 5/26/06, <b class=3D"gmail_send=L ername">Dave Froble</b> &lt;<a href=3D"mailto:davef@tsoft-inc.com">davef@ts=L oft-inc.com</a>&gt; wrote:</span><blockquote class=3D"gmail_quote" style=3D=L "border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padd= ing-left: 1ex;">L <a href=3D"mailto:rschiemel@gmail.com">rschiemel@gmail.com</a> wrote:<br>&g=L t; Compaq Open VMS VAX Version 7.2<br>&gt;<br>&gt; It was noticed that the =L freespace on the system disk was decreasing. A<br>&gt; search for large fil= es showed a file namedL <br>&gt; SYS$QUEUE_MANAGER.QMAN$JOURNAL;1 which was about 11,000 blocks, ex=L isted<br>&gt; in three separate directories on the system disk, using a tot=J al of<br>&gt; about 33,000 blocks.<br>&gt;<br>&gt; The directories were: [L SYS0.SYSCOMMON.SYSEXE], [SYSE.SYSCOMMON.SYSEXE],<br>&gt; and [VMS$COMMON.SY=L SEXE]. In an effort to reduce the size of the files<br>&gt; we issued the f=L ollowing commands:<br>&gt;&gt; mcr&nbsp;&nbsp;JBC$COMMAND<br>&gt; JBC$COMMA= ND&gt; DIAG 0 7 L <br>&gt;<br>&gt; The result was the original three files disappeared and in=L  their place<br>&gt; were three new files each about 1100 blocks in size. W=L e expected to<br>&gt; find the freespace increased by about 30,000 blocks. = Instead the L <br>&gt; freespace DECREASED by 3,300 blocks (The 3 new files).<br>&gt;<br>=L &gt; How do we get those 33,000 blocks free ?<br>&gt;<br><br>PURGE</blockqu=L ote><div><br>If you do a dir/id you will notice that they are all the same = file. $ <br><br>Ken <br></div><br></div><br>  * ------=_Part_4432_22375965.1148670308543--   ------------------------------  % Date: Fri, 26 May 2006 15:35:47 -0400  From: norm.raphael@metso.comK Subject: Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't delete Q Message-ID: <OF4439B915.0E0CEE9B-ON8525717A.0069F290-8525717A.006B9DEF@metso.com>   B Dave Froble <davef@tsoft-inc.com> wrote on 05/26/2006 02:39:47 PM:   > rschiemel@gmail.com wrote:# > > Compaq Open VMS VAX Version 7.2  > > J > > It was noticed that the freespace on the system disk was decreasing. A. > > search for large files showed a file namedK > > SYS$QUEUE_MANAGER.QMAN$JOURNAL;1 which was about 11,000 blocks, existed F > > in three separate directories on the system disk, using a total of > > about 33,000 blocks. > > K > > The directories were: [SYS0.SYSCOMMON.SYSEXE], [SYSE.SYSCOMMON.SYSEXE],  > > and [VMS$COMMON.SYSEXE].  . You only have one file in "SYS$COMMON:[SYSEXE]   type $ dire/file/siz=all/nohead4 sys$sysdevice:[*...]SYS$QUEUE_MANAGER.QMAN$JOURNAL;*  > and you will see that all three listings have the same fileid.  K SYS0 is your root,  SYSE is for Standalone backup, vms$common is the actual F location, the other two point to the files in that directory; they are	 "entered" J into the roots.  see help set file/enter.  Caution:  DO NOT MESS WITH THIS
 STRUCTURE.      0 > > In an effort to reduce the size of the files% > > we issued the following commands:  > >> mcr  JBC$COMMAND  > > JBC$COMMAND> DIAG 0 7  > > J > > The result was the original three files disappeared and in their placeG > > were three new files each about 1100 blocks in size. We expected to D > > find the freespace increased by about 30,000 blocks. Instead the: > > freespace DECREASED by 3,300 blocks (The 3 new files). > > , > > How do we get those 33,000 blocks free ? > >  >  > PURGE  >  > --6 > David Froble                       Tel: 724-529-0450@ > Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com > DFE Ultralights, Inc.  > 170 Grimplin Road  > Vanderbilt, PA  15486    ------------------------------    Date: 26 May 2006 21:39:51 -02006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)K Subject: Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't delete , Message-ID: <447775a7$1@news.langstoeger.at>  _ In article <1148663079.590369.275210@j73g2000cwa.googlegroups.com>, rschiemel@gmail.com writes: H >The directories were: [SYS0.SYSCOMMON.SYSEXE], [SYSE.SYSCOMMON.SYSEXE], >and [VMS$COMMON.SYSEXE]  L Means, they are all the same. (if you add cluster nodes, you get some more).A You should read a little bit about the VMS system disk structure.   G >                          In an effort to reduce the size of the files " >we issued the following commands: >> mcr  JBC$COMMAND  >JBC$COMMAND> DIAG 0 7  K So you know undocumented/unsupported commands but not alias directories....   G >The result was the original three files disappeared and in their place D >were three new files each about 1100 blocks in size. We expected toA >find the freespace increased by about 30,000 blocks. Instead the 7 >freespace DECREASED by 3,300 blocks (The 3 new files).  > ) >How do we get those 33,000 blocks free ?   O You don't. You only get 11k (if you PURGE the old no longer used version ;1)...   H And 11k is not that big for a SYS$QUEUE_MANAGER.QMAN$JOURNAL either. Ok,J my home system has only a 2 Block file, but I've seen 6 figures @work too.+ A good average might be a 2k Block file ;-)    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Fri, 26 May 2006 17:16:03 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> K Subject: Re: Need freespace but SYS$QUEUE_MANAGER.QMAN$JOURNAL won't delete : Message-ID: <hPidnV8HFquO7erZnZ2dnUVZ_tKdnZ2d@comcast.com>   rschiemel@gmail.com wrote:  ! > Compaq Open VMS VAX Version 7.2  > H > It was noticed that the freespace on the system disk was decreasing. A, > search for large files showed a file namedI > SYS$QUEUE_MANAGER.QMAN$JOURNAL;1 which was about 11,000 blocks, existed D > in three separate directories on the system disk, using a total of > about 33,000 blocks. > I > The directories were: [SYS0.SYSCOMMON.SYSEXE], [SYSE.SYSCOMMON.SYSEXE], G > and [VMS$COMMON.SYSEXE]. In an effort to reduce the size of the files # > we issued the following commands:  >  >>mcr  JBC$COMMAND >  > JBC$COMMAND> DIAG 0 7  > H > The result was the original three files disappeared and in their placeE > were three new files each about 1100 blocks in size. We expected to B > find the freespace increased by about 30,000 blocks. Instead the8 > freespace DECREASED by 3,300 blocks (The 3 new files). > * > How do we get those 33,000 blocks free ? >   9 $ ANALYZE /DISK_STRUCTURE /REPAIR /CONFIRM SYS$SYSDEVICE:    ------------------------------  % Date: Fri, 26 May 2006 16:38:29 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: Question , Message-ID: <44776744.39A4F002@teksavvy.com>   > Chuck Aaron wrote: > G > Is there a utilities command file that will search the UAF and report . > a list of id's containing the last time they$ > accessed the system if any at all?   please stop posting in HTML    $ALLIN1/NOINIT oa$ini_initialize H list <for uai$ do merge_line <.username>  <.lastlogin_i>  <.lastlogin_n> exit  9 (the _i is for interactive the _n is for non-interactive)    ------------------------------  % Date: Fri, 26 May 2006 15:48:57 -0700 , From: Ken Fairfield <my.full.name@intel.com> Subject: Re: Question + Message-ID: <e580kq$96t$1@news01.intel.com>   < Not having the original post on my server yet, I'll reply to JF's follow-up...    >> Chuck Aaron wrote:  >>H >> Is there a utilities command file that will search the UAF and report/ >> a list of id's containing the last time they % >> accessed the system if any at all?   ? There are two freeware packages on Hunter Goatley's fileserv at < Process Software, Joe Meadow's UAF and Jim Snyder's SCANUAF.> Try http://vms.process.com/fileserv_search.html and search for@ UAF, or go directly to the ftp server, ftp.process.com, and look under [.VMS-FREEWARE.FILESERV].   D I use the UAF utility to implement "account maintenance" procedures,D i.e., I use UAF to generate a list of accounts that have been "idle"A for a certain period of time and then take appropriate actions on D those.  UAF's interface is not the best (takes a little getting used to), but it does the job.   D I have no experience with SCANUAF.  With any luck, it should performG the same functions as UAF, but perhaps with a more "regular" interface?        Regards, Ken --  6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfield ! D1C Automation VMS System Support " who:   kenneth dot h dot fairfield where: intel dot com   ------------------------------    Date: 26 May 2006 09:03:24 -0500. From: brooks@cuebid.zko.hp.nospam (Rob Brooks)9 Subject: Re: speeding up LAVC with switch instead of hub? , Message-ID: <oxd8fZBGnwno@cuebid.zko.hp.com>  " <johnhreinhardt@yahoo.com> writes:  G > Make sure that the SRM console setting for the NIC have been changed. C >  It's probably a DEC DE500 family card and they are notorious for C > having problems doing an auto-negotiate.  Some switches they will C > auto-negotiate fine with and others they will not.  If you had it H > attached to a 10Mbs hub then it's probably set to "Twisted-Pair" whichI > defaults to 10Mbs/Half Duplex.  When you attach it to the 100Mbs switch I > you want to set it to "FastFD" (>>> set ewa0_mode fastfd) otherwise the F > switch will just match itself to the NIC and you won't see any speed > increase.   O The above info is not quite right.  The DE500-BA and -AA should have no problem I autonegotiating with a correctly-behaving switch (this is coming from the N guy who supports the LAN drivers for VMS [that's not me; I'm just passing thatM statement along]).  If you have a problem with one of these card with respect L to autonegotiation (and you are running a reasonably-current version of VMS H with recent patches), it's likely a switch problem, not a card problem.     L For those who will choose to upgrade to V8.3, you can take advantage of this= updated display (this is not in the field test version) . . .   K There are some subtle differences between the display for a template device ' and a device that is protocol-specific.      SISKO Alpha> sho dev/full ewb0  L Device EWB0:, device type DEGPA, is online, network device, error logging is'     enabled, device is a template only.   O     Error count                    1    Operations completed                  0 O     Owner process                 ""    Owner UIC                      [SYSTEM] O     Owner process ID        00000000    Dev Prot              S:RWPL,O:RWPL,G,W O     Reference count                0    Default buffer size                 512 O     Current preferred CPU Id       1    Fastpath                              1   O Operating characteristics: Link up, Full duplex, Autonegotiation, Jumbo frames.   $     Speed (Mbits/sec)           1000O     Def. MAC addr  00-60-6D-21-1E-DF    Current MAC addr      00-60-6D-21-1E-DF    SISKO Alpha> sho dev ewa7/ful   L Device EWA7:, device type DE500, is online, network device, error logging is     enabled.  O     Error count                    0    Operations completed             118332 O     Owner process           "NETACP"    Owner UIC                      [SYSTEM] O     Owner process ID        40200423    Dev Prot              S:RWPL,O:RWPL,G,W O     Reference count                2    Default buffer size                1498   8 Operating characteristics: Full duplex, Autonegotiation.  $     Speed (Mbits/sec)            100O     Def. MAC addr  00-00-F8-1F-46-6F    Current MAC addr      AA-00-04-00-4A-FD O     Protocol name             DECNET    Protocol type                     60-03    --    H Rob Brooks    VMS Engineering -- Exec Group     brooks!cuebid.zko.hp.com   ------------------------------    Date: 26 May 2006 12:11:33 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> 9 Subject: Re: speeding up LAVC with switch instead of hub? B Message-ID: <1148670693.582467.130090@38g2000cwa.googlegroups.com>   Rob Brooks wrote: $ > <johnhreinhardt@yahoo.com> writes: > I > > Make sure that the SRM console setting for the NIC have been changed. E > >  It's probably a DEC DE500 family card and they are notorious for E > > having problems doing an auto-negotiate.  Some switches they will E > > auto-negotiate fine with and others they will not.  If you had it J > > attached to a 10Mbs hub then it's probably set to "Twisted-Pair" whichK > > defaults to 10Mbs/Half Duplex.  When you attach it to the 100Mbs switch K > > you want to set it to "FastFD" (>>> set ewa0_mode fastfd) otherwise the H > > switch will just match itself to the NIC and you won't see any speed
 > > increase.  > Q > The above info is not quite right.  The DE500-BA and -AA should have no problem K > autonegotiating with a correctly-behaving switch (this is coming from the P > guy who supports the LAN drivers for VMS [that's not me; I'm just passing thatO > statement along]).  If you have a problem with one of these card with respect M > to autonegotiation (and you are running a reasonably-current version of VMS I > with recent patches), it's likely a switch problem, not a card problem.  >   G I stand corrected. :-)  My troubles with these  were some time ago back E in the V6.2/V7.0 timeframe.  Probably a combination of early switches  and new NIC drivers.   > N > For those who will choose to upgrade to V8.3, you can take advantage of this? > updated display (this is not in the field test version) . . .  > M > There are some subtle differences between the display for a template device ) > and a device that is protocol-specific.  >  >   > SISKO Alpha> sho dev/full ewb0 > N > Device EWB0:, device type DEGPA, is online, network device, error logging is) >     enabled, device is a template only.  > Q >     Error count                    1    Operations completed                  0 Q >     Owner process                 ""    Owner UIC                      [SYSTEM] Q >     Owner process ID        00000000    Dev Prot              S:RWPL,O:RWPL,G,W Q >     Reference count                0    Default buffer size                 512 Q >     Current preferred CPU Id       1    Fastpath                              1  > Q > Operating characteristics: Link up, Full duplex, Autonegotiation, Jumbo frames.  > & >     Speed (Mbits/sec)           1000Q >     Def. MAC addr  00-60-6D-21-1E-DF    Current MAC addr      00-60-6D-21-1E-DF  >  > SISKO Alpha> sho dev ewa7/ful  > N > Device EWA7:, device type DE500, is online, network device, error logging is >     enabled. > Q >     Error count                    0    Operations completed             118332 Q >     Owner process           "NETACP"    Owner UIC                      [SYSTEM] Q >     Owner process ID        40200423    Dev Prot              S:RWPL,O:RWPL,G,W Q >     Reference count                2    Default buffer size                1498  > : > Operating characteristics: Full duplex, Autonegotiation. > & >     Speed (Mbits/sec)            100Q >     Def. MAC addr  00-00-F8-1F-46-6F    Current MAC addr      AA-00-04-00-4A-FD Q >     Protocol name             DECNET    Protocol type                     60-03  >   D That's great.  Shows pretty much all the of the typical things you'd want to know in one display.   ------------------------------  % Date: Fri, 26 May 2006 17:13:59 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> 9 Subject: Re: speeding up LAVC with switch instead of hub? : Message-ID: <hPidnVwHFqsH8urZnZ2dnUVZ_tKdnZ2d@comcast.com>   Rob Brooks wrote:   $ > <johnhreinhardt@yahoo.com> writes: >  > G >>Make sure that the SRM console setting for the NIC have been changed. C >> It's probably a DEC DE500 family card and they are notorious for C >>having problems doing an auto-negotiate.  Some switches they will C >>auto-negotiate fine with and others they will not.  If you had it H >>attached to a 10Mbs hub then it's probably set to "Twisted-Pair" whichI >>defaults to 10Mbs/Half Duplex.  When you attach it to the 100Mbs switch I >>you want to set it to "FastFD" (>>> set ewa0_mode fastfd) otherwise the F >>switch will just match itself to the NIC and you won't see any speed >>increase.  >  > Q > The above info is not quite right.  The DE500-BA and -AA should have no problem K > autonegotiating with a correctly-behaving switch (this is coming from the P > guy who supports the LAN drivers for VMS [that's not me; I'm just passing thatO > statement along]).  If you have a problem with one of these card with respect N > to autonegotiation (and you are running a reasonably-current version of VMS J > with recent patches), it's likely a switch problem, not a card problem.  >   G I'm no longer able to verify this but I seem to recall having to force  I the Fast FD setting with DE500-AA NICs and and a relatively modern Cisco  G Switch (Catalyst 6500????).  I no longer work there so I can't go look   and can't swear to the model.   E I thought that the autonegotiation stuff was hardware/firmware level  F rather than VMS driver level.  Not so?   The systems in question were   running VMS V6.2-1H3 and V7.2-1.   ------------------------------   Date: 26 May 2006 18:07:38 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) , Message-ID: <4doufaF1bcom8U3@individual.net>  = In article <VZ-dncv1BLIK3OrZRVn-iA@metrocastcablevision.com>, + 	Bill Todd <billtodd@metrocast.net> writes:  > Bob Koehler wrote: >>  L >>    With UNIX you get no choice other than to buy a megabucks DBMS if you  >>    need reliable I/O. > F > It's this kind of incompetent and arrogant ignorance that gives VMS + > bigots a bad name (and justifiably so).     G I've been saying this for quite some time now.  People here either know G nothing about Unix or their last experience was with Ultrix-32 V3.0 and E they base their comments on this as if while VMS continued to develop  Unix just stood still.  E >                                         The Unix fsync() operation  F > performs essentially the same function that the RMS FLUSH operation I > does:  it forces all pending writes for the associated file to disk if   > they are not already there.   F And mounting the file system with the "sync" operation will cause thisG to happen automatically on all writes.  There is a performance hit, but ( even that is not as bad as it is on VMS.  I VMS has advantages.  If people want to see VMS survive they need to start G concentrating on those advantages and stop posting drivel that everyone G in the industry knows is hogwash.  It just makes them and VMS look bad.    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Fri, 26 May 2006 13:57:42 -0400 ( From: Bill Todd <billtodd@metrocast.net>M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) = Message-ID: <VZ-dncv1BLIK3OrZRVn-iA@metrocastcablevision.com>    Bob Koehler wrote:Q >  "Richard B. Gilbert" <rgilbert88@comcast.net> wrote on 05/25/2006 10:39:32 AM:  > 4 >> Unix runs faster than VMS on comparable hardware. > I >    Intentionally misleading FUD.  A subtract instruciotn, for instance, ; >    depends only on the hardware and not at all on the OS.  > K >    What you get with VMS is the option to determine whether you want more @ >    reliable I/O or faster I/O since you control the buffering.  * As, of course, you can do on Unix as well.  	    If you 8 >    don't tell VMS otherwise, it will do realiable I/O.  H Not with sequential files (i.e., the vast majority of all files), which G by default are buffered (just as they are on Unix - save that VMS does  ' the job nowhere nearly as efficiently).       You get what you 
 >    need.  I No, regardless of what your specific needs may be you get what VMS gives  I you by default, just as you do with Unix.  However, VMS by default gives  I you mediocre sequential file performance with no guarantee that what you  G write will wind up on disk immediately, while Unix gives you excellent  G sequential file performance with no guarantee that what you write will   wind up on disk immediately.   > K >    With UNIX you get no choice other than to buy a megabucks DBMS if you   >    need reliable I/O.   D It's this kind of incompetent and arrogant ignorance that gives VMS D bigots a bad name (and justifiably so).  The Unix fsync() operation D performs essentially the same function that the RMS FLUSH operation G does:  it forces all pending writes for the associated file to disk if   they are not already there.    - bill   ------------------------------  % Date: Fri, 26 May 2006 17:42:17 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) 0 Message-ID: <6uSdncyfpf-n6-rZRVn-jA@comcast.com>   Bill Todd wrote:   > Bob Koehler wrote: > F >>  "Richard B. Gilbert" <rgilbert88@comcast.net> wrote on 05/25/2006  >> 10:39:32 AM:  >>5 >>> Unix runs faster than VMS on comparable hardware.  >> >>J >>    Intentionally misleading FUD.  A subtract instruciotn, for instance,< >>    depends only on the hardware and not at all on the OS. >>L >>    What you get with VMS is the option to determine whether you want moreA >>    reliable I/O or faster I/O since you control the buffering.  >  > , > As, of course, you can do on Unix as well. > 
 >   If you > 9 >>    don't tell VMS otherwise, it will do realiable I/O.  >  > J > Not with sequential files (i.e., the vast majority of all files), which I > by default are buffered (just as they are on Unix - save that VMS does  ) > the job nowhere nearly as efficiently).  >  >   You get what you >  >>    need.  >  > K > No, regardless of what your specific needs may be you get what VMS gives  K > you by default, just as you do with Unix.  However, VMS by default gives  K > you mediocre sequential file performance with no guarantee that what you  I > write will wind up on disk immediately, while Unix gives you excellent  I > sequential file performance with no guarantee that what you write will   > wind up on disk immediately. >   I I won't argue the performance issue but when $QIO completes with success  D status in the IOSB then VMS has written the data to the disk.  When D write() returns success in Unix, it just means that it has promised G write your data to disk someday if the power doesn't fail before then.  G If you lose power before Unix feels like actually writing your data to  F disk that's your tough luck.  fsck is usually able to repair the file % structure but your data may be toast.    ------------------------------  % Date: Fri, 26 May 2006 18:02:12 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) 0 Message-ID: <pKOdnQ2lpct45-rZRVn-tw@comcast.com>   Bob Koehler wrote:  Q >  "Richard B. Gilbert" <rgilbert88@comcast.net> wrote on 05/25/2006 10:39:32 AM:  >  > 3 >>Unix runs faster than VMS on comparable hardware.  >  > I >    Intentionally misleading FUD.  A subtract instruciotn, for instance, ; >    depends only on the hardware and not at all on the OS.  > K >    What you get with VMS is the option to determine whether you want more H >    reliable I/O or faster I/O since you control the buffering.  If youK >    don't tell VMS otherwise, it will do realiable I/O.  You get what you  
 >    need. > K >    With UNIX you get no choice other than to buy a megabucks DBMS if you   >    need reliable I/O.  >   G Things like process creation are a hell of a lot slower on VMS than on  I Unix.  VMS also does some sanity checking that Unix doesn't bother with;  H it does mandatory locking which is not built into Unix, etc.  VMS has a G fairly complex system of rights and privileges which we like but which   takes time.   I Sure, a subtract is a subtract regardless of the O/S but there is a hell  : of a lot more to most applications than math instructions.  D Ever see VMS grind to a halt when some asshole puts thirty thousand H files in a directory?   Ever spend three or four days watching a DELETE G *.*;* clean up those thirty thousand files?  I don't recommend putting  I thirty thousand files in a directory under ANY O/S but it does not cause  H the same problems in Unix.  This was years ago under V6.2 and I believe G that the situation is not quite as bad as it used to be but it's still  	 not good.    ------------------------------  % Date: Fri, 26 May 2006 18:15:07 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) : Message-ID: <UbWdnVzg55lx4OrZnZ2dnUVZ_uudnZ2d@comcast.com>   Bob Koehler wrote:  r > In article <hvCdnejkK5FEtuvZnZ2dnUVZ_vednZ2d@comcast.com>, "Richard B. Gilbert" <rgilbert88@comcast.net> writes: > B >>Strangely enough, Unix, or at least Solaris, tends to be pretty  >>reliable.  >  > J >    We've just gone through a series of security hits.  Every one of themA >    on up to date, patched, firewall protected Solaris machines.  > E >    Personnaly I was a bit surprised by this because I was under the ; >    impression Solaris was doing much better in this area.  > . >    Without security there is no reliability. >   E Would you care to provide a few details?  Did these "hits" originate  H externally or inside your firewall?  Were the vulnerabilities exploited E in the O/S itself or in a Web server?  Were vulnerabilities actually  9 exploited or did you simply detect unsuccessful attempts?   G The last time I had a PC directly connected to the internet, ZoneAlarm  G detected about fifteen or twenty attempts per hour to connect to my IP  " address.  It's probably worse now.   ------------------------------    Date: 26 May 2006 16:09:10 -0700 From: bob@instantwhip.com M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) B Message-ID: <1148684950.606935.88490@j33g2000cwa.googlegroups.com>  4 except with unix you need to run 88000 boxes and you9 need to become a member of the patch of the week club ...    CERTS don't lie!   ------------------------------    Date: 26 May 2006 16:06:55 -0700 From: bob@instantwhip.com M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) B Message-ID: <1148684815.085633.98440@i40g2000cwc.googlegroups.com>  ( OpenVMS is a firewall!  And on vms IP or) web vulnerabilities just throw back those  pesky    "ACCESS VIOLATION"  - errors ... you don't need extra firewalls and ' 88000 boxes to run a vms web server ... ( and you don't need to become a member of the patch of the day club!   ------------------------------  # Date: Sat, 27 May 2006 00:40:01 GMT 9 From: Bob Harris <nospam.News.Bob@remove.Smith-Harris.us> M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) D Message-ID: <nospam.News.Bob-F397A0.20400626052006@news.verizon.net>  0 In article <6uSdncyfpf-n6-rZRVn-jA@comcast.com>,5  "Richard B. Gilbert" <rgilbert88@comcast.net> wrote:    > Bill Todd wrote: >  > > Bob Koehler wrote: > > H > >>  "Richard B. Gilbert" <rgilbert88@comcast.net> wrote on 05/25/2006  > >> 10:39:32 AM:  > >>7 > >>> Unix runs faster than VMS on comparable hardware.  > >> > >>L > >>    Intentionally misleading FUD.  A subtract instruciotn, for instance,> > >>    depends only on the hardware and not at all on the OS. > >>N > >>    What you get with VMS is the option to determine whether you want moreC > >>    reliable I/O or faster I/O since you control the buffering.  > >  > > . > > As, of course, you can do on Unix as well. > >  > >   If you > > ; > >>    don't tell VMS otherwise, it will do realiable I/O.  > >  > > L > > Not with sequential files (i.e., the vast majority of all files), which K > > by default are buffered (just as they are on Unix - save that VMS does  + > > the job nowhere nearly as efficiently).  > >  > >   You get what you > > 
 > >>    need.  > >  > > M > > No, regardless of what your specific needs may be you get what VMS gives  M > > you by default, just as you do with Unix.  However, VMS by default gives  M > > you mediocre sequential file performance with no guarantee that what you  K > > write will wind up on disk immediately, while Unix gives you excellent  K > > sequential file performance with no guarantee that what you write will    > > wind up on disk immediately. > >  > K > I won't argue the performance issue but when $QIO completes with success  F > status in the IOSB then VMS has written the data to the disk.  When F > write() returns success in Unix, it just means that it has promised I > write your data to disk someday if the power doesn't fail before then.  I > If you lose power before Unix feels like actually writing your data to  H > disk that's your tough luck.  fsck is usually able to repair the file ' > structure but your data may be toast.   C And the same is true for a UNIX file that has been opened with the  C O_SYNC option ( fd = open("file", O_RDWR|O_SYNC); ) or if the file  A system has been mounted with the sync option (mount -o sync /dev  C /mnt).  No promises.  Data is on the disk, or the write returns an   error.  ? And when it comes to doing OpenVMS I/O, there are not a lot of  C people that do $QIO.  Most likely few even use $RMS.  Most use the  @ language I/O package which today is mostly C and the C Run-time A library.  And that uses buffered RMS files which are also just a   promise, unless you do a flush.    Don't pick on UNIX I/O.   = I worked the PATHWORKS for OpenVMS (Macintosh) DECshare file  : server, and wrote $QIOs that not even a mother could love.  C I've also worked on the Tru64 UNIX AdvFS file system, and until HP  @ cancelled the project, I worked on the port of AdvFS to HP-UX.    ? I know of what I speak when it comes to file system I/O.  Both   UNIX and OpenVMS.   B Like someone else said, don't pick on UNIX weaknesses, because it = is not as weak as you may think.  Better to focus on OpenVMS  > Strengths.  Clustering, Security, Reliability, Scaling, etc...  2                                         Bob Harris   ------------------------------  % Date: Fri, 26 May 2006 20:42:43 -0500 / From: pechter@pechter.dyndns.org (Bill Pechter) M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) 0 Message-ID: <W6WdnUeXPdkOM-rZRVn-pw@comcast.com>  0 In article <6uSdncyfpf-n6-rZRVn-jA@comcast.com>,2 Richard B. Gilbert <rgilbert88@comcast.net> wrote: >Bill Todd wrote:  >  >> Bob Koehler wrote:  >>  G >>>  "Richard B. Gilbert" <rgilbert88@comcast.net> wrote on 05/25/2006   >>> 10:39:32 AM: >>> 6 >>>> Unix runs faster than VMS on comparable hardware. >>>  >>> K >>>    Intentionally misleading FUD.  A subtract instruciotn, for instance, = >>>    depends only on the hardware and not at all on the OS.  >>> M >>>    What you get with VMS is the option to determine whether you want more B >>>    reliable I/O or faster I/O since you control the buffering. >>   >>  - >> As, of course, you can do on Unix as well.  >>   >>   If you  >>  : >>>    don't tell VMS otherwise, it will do realiable I/O. >>   >>  K >> Not with sequential files (i.e., the vast majority of all files), which  J >> by default are buffered (just as they are on Unix - save that VMS does * >> the job nowhere nearly as efficiently). >>   >>   You get what you  >>   >>>    need. >>   >>  L >> No, regardless of what your specific needs may be you get what VMS gives L >> you by default, just as you do with Unix.  However, VMS by default gives L >> you mediocre sequential file performance with no guarantee that what you J >> write will wind up on disk immediately, while Unix gives you excellent J >> sequential file performance with no guarantee that what you write will  >> wind up on disk immediately.  >>   > J >I won't argue the performance issue but when $QIO completes with success E >status in the IOSB then VMS has written the data to the disk.  When  E >write() returns success in Unix, it just means that it has promised  H >write your data to disk someday if the power doesn't fail before then. H >If you lose power before Unix feels like actually writing your data to G >disk that's your tough luck.  fsck is usually able to repair the file  & >structure but your data may be toast. >     G I don't believe this is true... at least not in some situations I know. G We run mirror disks and the mirroring ( I thought ) guarantees that the E write() operation must complete successfully to both drives before it  returns an OK status...   E (Read returns when the first available mirror segment is successfully  read...)  - At least that's what I thought... and taught.   8 Now linux with Async mounts on ext2 -- all bets are off.   Bill   --   --  H   d|i|g|i|t|a|l had it THEN.  Don't you wish you could still buy it now!#   pechter-at-ureachtechnologies.com    ------------------------------  % Date: Fri, 26 May 2006 20:44:56 -0500 / From: pechter@pechter.dyndns.org (Bill Pechter) M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) 0 Message-ID: <W6WdnUaXPdmFMurZRVn-pw@comcast.com>  0 In article <pKOdnQ2lpct45-rZRVn-tw@comcast.com>,2 Richard B. Gilbert <rgilbert88@comcast.net> wrote: >Bob Koehler wrote:  > E >>  "Richard B. Gilbert" <rgilbert88@comcast.net> wrote on 05/25/2006 
 >10:39:32 AM:  >>   >>  4 >>>Unix runs faster than VMS on comparable hardware. >>   >>  J >>    Intentionally misleading FUD.  A subtract instruciotn, for instance,< >>    depends only on the hardware and not at all on the OS. >>  L >>    What you get with VMS is the option to determine whether you want moreI >>    reliable I/O or faster I/O since you control the buffering.  If you L >>    don't tell VMS otherwise, it will do realiable I/O.  You get what you  >>    need.  >>  L >>    With UNIX you get no choice other than to buy a megabucks DBMS if you  >>    need reliable I/O. >>   > H >Things like process creation are a hell of a lot slower on VMS than on J >Unix.  VMS also does some sanity checking that Unix doesn't bother with; I >it does mandatory locking which is not built into Unix, etc.  VMS has a  H >fairly complex system of rights and privileges which we like but which  >takes time. > J >Sure, a subtract is a subtract regardless of the O/S but there is a hell ; >of a lot more to most applications than math instructions.  > E >Ever see VMS grind to a halt when some asshole puts thirty thousand  I >files in a directory?   Ever spend three or four days watching a DELETE  H >*.*;* clean up those thirty thousand files?  I don't recommend putting J >thirty thousand files in a directory under ANY O/S but it does not cause I >the same problems in Unix.  This was years ago under V6.2 and I believe  H >that the situation is not quite as bad as it used to be but it's still 
 >not good. >     5 Well... try delete *.* on Unix and try to rm -rf *...   F (you better hack in xargs and stuff....the command line metacharactersD will break  the shell on thirty thousand files... or thirty thousand directories...)   E I'm a Unix sysadmin and I know VMS is not any worse than Unix on this  one.     Bill --   --  H   d|i|g|i|t|a|l had it THEN.  Don't you wish you could still buy it now!#   pechter-at-ureachtechnologies.com    ------------------------------   Date: 27 May 2006 02:02:14 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) , Message-ID: <4dpq96F1936phU1@individual.net>  0 In article <W6WdnUaXPdmFMurZRVn-pw@comcast.com>,2 	pechter@pechter.dyndns.org (Bill Pechter) writes:2 > In article <pKOdnQ2lpct45-rZRVn-tw@comcast.com>,4 > Richard B. Gilbert <rgilbert88@comcast.net> wrote: >>Bob Koehler wrote: >>F >>>  "Richard B. Gilbert" <rgilbert88@comcast.net> wrote on 05/25/2006 >>10:39:32 AM: >>>  >>> 5 >>>>Unix runs faster than VMS on comparable hardware.  >>>  >>> K >>>    Intentionally misleading FUD.  A subtract instruciotn, for instance, = >>>    depends only on the hardware and not at all on the OS.  >>> M >>>    What you get with VMS is the option to determine whether you want more J >>>    reliable I/O or faster I/O since you control the buffering.  If youM >>>    don't tell VMS otherwise, it will do realiable I/O.  You get what you   >>>    need. >>> M >>>    With UNIX you get no choice other than to buy a megabucks DBMS if you   >>>    need reliable I/O.  >>>  >>I >>Things like process creation are a hell of a lot slower on VMS than on  K >>Unix.  VMS also does some sanity checking that Unix doesn't bother with;  J >>it does mandatory locking which is not built into Unix, etc.  VMS has a I >>fairly complex system of rights and privileges which we like but which  
 >>takes time.  >>K >>Sure, a subtract is a subtract regardless of the O/S but there is a hell  < >>of a lot more to most applications than math instructions. >>F >>Ever see VMS grind to a halt when some asshole puts thirty thousand J >>files in a directory?   Ever spend three or four days watching a DELETE I >>*.*;* clean up those thirty thousand files?  I don't recommend putting  K >>thirty thousand files in a directory under ANY O/S but it does not cause  J >>the same problems in Unix.  This was years ago under V6.2 and I believe I >>that the situation is not quite as bad as it used to be but it's still   >>not good.  >> >  > 7 > Well... try delete *.* on Unix and try to rm -rf *...  > H > (you better hack in xargs and stuff....the command line metacharactersF > will break  the shell on thirty thousand files... or thirty thousand > directories...)  > G > I'm a Unix sysadmin and I know VMS is not any worse than Unix on this  > one.  G That's a shell shortcoming and has little. if anything, to do with Unix 2 or the file system.  And, it's easy to get around.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Fri, 26 May 2006 22:59:10 -0400 ( From: Bill Todd <billtodd@metrocast.net>M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) G Message-ID: <d_adnSdccu8dXerZnZ2dnUVZ_tKdnZ2d@metrocastcablevision.com>    Bill Gunshannon wrote:? > In article <VZ-dncv1BLIK3OrZRVn-iA@metrocastcablevision.com>, - > 	Bill Todd <billtodd@metrocast.net> writes:    ...   F >>                                         The Unix fsync() operation G >> performs essentially the same function that the RMS FLUSH operation  J >> does:  it forces all pending writes for the associated file to disk if  >> they are not already there. > H > And mounting the file system with the "sync" operation will cause thisI > to happen automatically on all writes.  There is a performance hit, but * > even that is not as bad as it is on VMS.  H Hmmm - I'd kind of expect forcing write-through operation on all writes G to be about equal in performance on any OS, and would certainly expect  I that forcing it on Unix would result in poorer performance (at least for  A writes - reads would retain their advantages) than the *default*  # (non-write-through) VMS mechanisms.   H But forcing all writes for a file system to execute synchronously is an C awfully brute-force approach to most needs, which require far less  B frequent or pervasive persistency checkpoints.  FLUSH/fsync-style C mechanisms satisfy most needs with the minimum amount of increased  G overhead (ISTR that RMS may also offer a per-$PUT option to force only  F that particular write and its associated metadata changes to disk, as H Windows - gak! - does, which is kind of nice in avoiding the need for a I separate FLUSH after each such $PUT as would be the case on Unix but not  D all that critical given the low cost of system calls in modern OSs).  G IIRC some Unixes also offer the option to specify write-through access  H for an individual file when it is opened (as distinct from 'direct I/O' 8 facilities which bypass the Unix system cache entirely).   - bill   ------------------------------  % Date: Fri, 26 May 2006 22:43:42 -0400 ( From: Bill Todd <billtodd@metrocast.net>M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) G Message-ID: <1cmdnVLkx6V9IerZnZ2dneKdnZydnZ2d@metrocastcablevision.com>    Richard B. Gilbert wrote:    ...   F >> No, regardless of what your specific needs may be you get what VMS D >> gives you by default, just as you do with Unix.  However, VMS by B >> default gives you mediocre sequential file performance with no I >> guarantee that what you write will wind up on disk immediately, while  J >> Unix gives you excellent sequential file performance with no guarantee 8 >> that what you write will wind up on disk immediately. >> > K > I won't argue the performance issue but when $QIO completes with success  F > status in the IOSB then VMS has written the data to the disk.  When F > write() returns success in Unix, it just means that it has promised I > write your data to disk someday if the power doesn't fail before then.  I > If you lose power before Unix feels like actually writing your data to  H > disk that's your tough luck.  fsck is usually able to repair the file ' > structure but your data may be toast.   G Don't confuse low-level (and rare, as was just observed elsewhere) QIO  I access with the normal VMS access mechanism via RMS:  the latter is what  I I was discussing above, which should have been obvious (since QIO has no  I understanding of file type, sequential or otherwise, nor of intermediate   buffering).   > Many (I think most, but I haven't done an actual census) Unix H implementations will also give you direct, unbuffered access to disk if G you really want it.  Just as is the case on VMS, few applications take   advantage of that facility.    = bill   ------------------------------    Date: 27 May 2006 06:21:29 +01002 From: "Dave Weatherall" <djw-nothere@nospam.nohow>M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) ? Message-ID: <DTiotGxQ0bj6-pn2-u4qNcEfztSPR@dave2_os2.home.ours>   F On Thu, 25 May 2006 16:25:16 UTC, sms@antinode.org (Steven M. Schweda) wrote:   > From: bob@instantwhip.com  > - > > [...]  ... get your facts straight before  > > posting lies ... >  >    What good would that do?  > I >    Now, if you were to change "before" to "instead of", it might make a , > little more sense.  (That'd be a novelty.)  > Take your point Steven. OTOH it is valid thing to say someone D promoting statistics or delivering propaganda :-)  No reflection on B Richard there - just another (inapproprate) interpretation of the  sentence snippet.    --   Cheers - Dave W.   ------------------------------    Date: 27 May 2006 06:21:31 +01002 From: "Dave Weatherall" <djw-nothere@nospam.nohow>M Subject: Re: Unix runs faster, maybe (was: Re: Educating potential VMS users) ? Message-ID: <DTiotGxQ0bj6-pn2-ZnLoVYzqEGOr@dave2_os2.home.ours>   7 On Thu, 25 May 2006 22:12:40 UTC, "Richard B. Gilbert"   <rgilbert88@comcast.net> wrote:    B > Strangely enough, Unix, or at least Solaris, tends to be pretty K > reliable.  It is not as user friendly as VMS and is laden with traps for  K > the unwary but it will generally run as long as you remember to pay your  I > electric bill.  Unskilled tinkering as root can and will bring it down  H > but, as with VMS, there is no defense against privilege.  The Solaris I > systems at my last job ran like the VMS systems and came down only for  I > hardware maintenance or for power failures lasting longer than the run   > time of our UPS.  YMMV!   F It seems to help if you reboot 'em twice a week - well ours anyway :-)    --   Cheers - Dave W.   ------------------------------   End of INFO-VAX 2006.292 ************************