0 INFO-VAX	Sun, 19 Jan 2003	Volume 2003 : Issue 38      Contents:5 RE: (OT) lots of spam in vmsnet.sdk.openvms.fieldtest ) Re: Altavista Personal Search substitute? % Antique question: 11/730 Console Tape  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX - Re: HP Sets the Stage for Alpha's Last Hurrah - Re: Itanium with alpha goodies on the way ...  Re: MAIL 11 protocol definition & Re: MX 5.3-2 ignoring vmsmail forwards. Re: Network Time Protocol (NTP) for OpenVMS ??. Re: Network Time Protocol (NTP) for OpenVMS ??  F ----------------------------------------------------------------------   Date: 19 Jan 03 12:30:14 +0100) From: p_sture@elias.decus.ch (Paul Sture) > Subject: RE: (OT) lots of spam in vmsnet.sdk.openvms.fieldtest) Message-ID: <2QqEVUgn$3V8@elias.decus.ch>   W In article <01C2BE46.85281560@sulfer.icius.com>, Shane Smith <ssmith@icius.com> writes:    <snip> > C > Incidentally, William, taxidermy may not be involved, but there's C > already an electronic "severed heads" page for usenet abusers who B > tangled with the alt.gothic "special forces". Have a chuckle at:- > http://thingy.apana.org.au/~fun/agsf/heads/  > I > I love the quote on that page: "And lo, the LART did come down from the G > heavens trailing fire and brimstone. Yea verily, the LART did implant J > itself in yon bozo's forehead, its mighty blade cleaving his thick skullJ > in twain and exposing the empty spaces therein for all the world to gaze5 > upon with much mirth and frivolity" - Camille Klein  > ( And I rather like the quote below that:   < "NOBODY expects the SPAMMISH INQUISITION!" - James A. Cooley   --  
 Paul Sture   ------------------------------  % Date: Sun, 19 Jan 2003 11:32:42 +0100 $ From: Michael Unger <unger@decus.de>2 Subject: Re: Altavista Personal Search substitute?+ Message-ID: <00A1A336.F19D4D91.10@decus.de>   4 "PRSTSC::DTL" <Didier.Morandi.nospam@Free.fr> wrote:  D > Doing some drawers cleaning, I found my CD of the Altavista Search > Personal extension '97.  > H > Unfortunately, it does not work properly under XP. The indexing engineE > crashes many times during indexation with an attemp to write into a E > rea-only page (we can recognize Dave Cutler's paw :-) and during an I > attempt to browse the disk with a partial index, ie complains about not D > finding "classes.zip dispatcher", which is actually present in its > directory. > G > Did a product replace this amazing tool to transform a stupid Windows 7 > Search function to a Mac Sherlock-like search engine?   ( Not a direct answer to your question ...  F There is a PC based tool called "superior search" which claims to findE text and files quickly and reliably, even in compressed formats (ZIP, 
 PDF, ...):   http://www.superiorsearch.com   - (page available in English, German and Dutch)   F I had just heard of it but never used it. As far as I know it is about5 30 Euros and there seems to be a trial version of it.    Michael    ------------------------------  % Date: Sun, 19 Jan 2003 10:29:31 -0600 / From: "David Harrison" <djharrid@bellsouth.net> . Subject: Antique question: 11/730 Console Tape6 Message-ID: <0nAW9.44847$Jm2.27574@news.bellsouth.net>  G I'm administrator or a VAX 11/730 that ate its original console tape. I E found another tape but with this tape the system refuses to perform a E default boot to the UDA controller. I tried replacing DEFBOO.CMD with L DUABOO.CMD (see below) and the system hits the disk, but only prints out theK VMS banner before stopping. I then halt the CPU and run DUABOO again and it 
 boots VMS.  I Question: is there something I need to add to the DUABOO file so that the G system will boot from the UDA the first time? I used to have a bunch of 9 11/730 documentation, but it has long since been trashed.    DEFBOO.CMD:      I      D/P/L  F26200  86      D/P/L  F2620C  FFFFFFFF      D/P/L  F26200  6     D/G/L  0  00A80003     D/G/L  1  3      D/G/L  2  3FB86      D/G/L  3  0      D/G/L  4  0      D/G/L  5  0 	     E  SP    DUABOO.CMD:      I      D/G  0  11
     D/G  1  3      D/G  2  3F468 
     D/G  4  0 
     D/G  5  0 	     E  SP    Thanks,    David Harrison djharrid@bellsouth.net   ------------------------------  # Date: Sun, 19 Jan 2003 11:19:27 GMT % From: Tony Lawrence <tony@pcunix.com>  Subject: Re: How to Backup OSX? Message-ID: <3RvW9.615140$GR5.409138@rwcrnsc51.ops.asp.att.net>    Howard S Shubs wrote: B > In article <4wlW9.731686$%m4.3417976@rwcrnsc52.ops.asp.att.net>,) >  Tony Lawrence <tony@pcunix.com> wrote:  >  > I >>I think you are misunderstanding something here. How can a backup know  B >>whether or not a file that it backed up should or should not be G >>restored?  It can't.  There's also the matter of losing one piece of  . >>thev differential, as the article explained. >  > 6 > Okay, I'll explain how it's done, if I can remember. > C > A backup set, be it a full backup or an incremental, stores file  > > attributes as well as data.  By comparing the "date of last J > modification", you can determine if a file on disk is to be replaced by @ > one on tape.  That covers restoring only the latest copy, yes? > J > So, how does it know to -delete- old files?  IIRC, by storing directory E > information as well, a directory being a kind of file.  If a newer  F > version of a directory file doesn't contain file X.Y, then file X.Y @ > shouldn't continue to exist, and it gets deleted.  I might be F > remembering wrong, though.  I'll try cross-posting to a group which 
 > would know.  >   I Sheesh.  I understand the algorithm it could use, what I'm trying to get  I through to you is that it is insufficient.  The whole point of backup is  I recover data.  Sometimes files get deleted that should not have been and  H that fact isn't always discovered immediately.  Sometimes the fact that H something is gone isn't even known: all that is known is that something H that used to work does not. Couple that with some bad timing and a full H restore and this smart differential backup finds the file on its master @ but then deletes it yet again on the differerential.. wonderful.   >  > J >>Differentials are NOT the best way.  They are acceptable if there is no C >>other way for you to do complete backups.  Period.  If you don't  4 >>understand that, you'll someday wish that you did. >  > K > No, that's what I was saying.  We're in violent agreement.  Full backups  8 > followed by a series of incrementals is the way to go.  & No.  FULL BACKUPS EACH AND EVERY TIME.     :-)      >  >  > I >>Trust me: I've seen more disasters than you'd ever want to know about.   >  > I > I haven't, because I've always been able to recover from data loss, or  ; > near as dammit, since my first few weeks in the industry.  >   I I don't mean to offend, but that's incredibly naive.  I've been involved  I with computer systems since 1967.  I have heard the wailing and seen the  F gnashing of teeth.  Bad things happen to backup systems, and when the B shit has really hit the fan and splattered all over the room, the D ability to pull a complete backup from three days back or even last D month can be the difference between a lot of hard work and complete 	 disaster.   I If you can't do full backups because it takes too long or you don't have  H big enough storage, fine: do differentials.  As I have said though, the I usual case is that the static files are often the least of your problems  D in space and time, and large, fast archival systems get cheaper and  cheaper all the time anyway.       --  
 Tony Lawrence , Free SCO, Mac OS X and  Linux Skills Tests: % http://aplawrence.com/skillstest.html    ------------------------------  # Date: Sun, 19 Jan 2003 12:24:43 GMT 0 From: John Santos <john.santos@post.harvard.edu> Subject: Re: How to Backup OSX> Message-ID: <MPG.18946314709c31db9896c2@news.bellatlantic.net>  @ In article <3RvW9.615140$GR5.409138@rwcrnsc51.ops.asp.att.net>,  tony@pcunix.com says...  > Howard S Shubs wrote: D > > In article <4wlW9.731686$%m4.3417976@rwcrnsc52.ops.asp.att.net>,+ > >  Tony Lawrence <tony@pcunix.com> wrote:  > >  > > K > >>I think you are misunderstanding something here. How can a backup know  D > >>whether or not a file that it backed up should or should not be I > >>restored?  It can't.  There's also the matter of losing one piece of  0 > >>thev differential, as the article explained. > >  > > 8 > > Okay, I'll explain how it's done, if I can remember. > > E > > A backup set, be it a full backup or an incremental, stores file  @ > > attributes as well as data.  By comparing the "date of last L > > modification", you can determine if a file on disk is to be replaced by B > > one on tape.  That covers restoring only the latest copy, yes? > > L > > So, how does it know to -delete- old files?  IIRC, by storing directory G > > information as well, a directory being a kind of file.  If a newer  H > > version of a directory file doesn't contain file X.Y, then file X.Y B > > shouldn't continue to exist, and it gets deleted.  I might be H > > remembering wrong, though.  I'll try cross-posting to a group which  > > would know.  > >  > K > Sheesh.  I understand the algorithm it could use, what I'm trying to get  K > through to you is that it is insufficient.  The whole point of backup is  K > recover data.  Sometimes files get deleted that should not have been and  J > that fact isn't always discovered immediately.  Sometimes the fact that J > something is gone isn't even known: all that is known is that something J > that used to work does not. Couple that with some bad timing and a full J > restore and this smart differential backup finds the file on its master B > but then deletes it yet again on the differerential.. wonderful.  > Err, if you know a particular file is missing, you restore it D explicitly.  You don't do a full restore unless you are recovering a
 trashed disk.   C For some reason this thread is cross-posted between comp.os.vms and @ comp.sys.mac.apps.  I don't know anything about Mac backups, but? VMS backup certainly allows you to do selective restores of any A file or group of files from an image, differential or incremental A backup.  If you do a selective restore, it restores the files you ? ask for (including everything if you ask it nicely) and doesn't ? delete anything.  It is explicitly *not* trying to recreate the 8 disk at the time of the most recent backup in this case.   >  > >  > > L > >>Differentials are NOT the best way.  They are acceptable if there is no E > >>other way for you to do complete backups.  Period.  If you don't  6 > >>understand that, you'll someday wish that you did. > >  > > M > > No, that's what I was saying.  We're in violent agreement.  Full backups  : > > followed by a series of incrementals is the way to go. > ( > No.  FULL BACKUPS EACH AND EVERY TIME. >  >  > :-)  >  >  > >  > >  > > K > >>Trust me: I've seen more disasters than you'd ever want to know about.   > >  > > K > > I haven't, because I've always been able to recover from data loss, or  = > > near as dammit, since my first few weeks in the industry.  > >  > K > I don't mean to offend, but that's incredibly naive.  I've been involved  K > with computer systems since 1967.  I have heard the wailing and seen the  H > gnashing of teeth.  Bad things happen to backup systems, and when the D > shit has really hit the fan and splattered all over the room, the F > ability to pull a complete backup from three days back or even last F > month can be the difference between a lot of hard work and complete  > disaster.  > K > If you can't do full backups because it takes too long or you don't have  J > big enough storage, fine: do differentials.  As I have said though, the K > usual case is that the static files are often the least of your problems  F > in space and time, and large, fast archival systems get cheaper and  > cheaper all the time anyway. >  >  >  >    --   John   ------------------------------  # Date: Sun, 19 Jan 2003 14:53:35 GMT * From: James L. Ryan <taliesinsoft@mac.com> Subject: Re: How to Backup OSX@ Message-ID: <0001HW.BA502697000181FDF0284600@news.earthlink.net>  6 On Sun, 19 Jan 2003 8:56:25 -0500, Tony Lawrence wroteB (in message <d8yW9.738625$%m4.3420648@rwcrnsc52.ops.asp.att.net>):  F > The BEST backup is a COMPLETE backup each and every time.  If it is G > impossible to do that because of space or time issues, then AND ONLY  $ > THEN should you use differentials.  H Perhaps it has already been mentioned in this thread of discussion, but L Retrospect (and Retrospect Express) allow one to make three different kinds  of backups:   I 1) an incremental backup which augments the backup with only those items  M which are new or which have changed. Such a backup allows one to selectively  I restore any version of an item that at any time has been included in the  K backup set. The structure of an incremental backup is a proprietary single  K file which requires Retrospect or Retrospect Express when recovering items.   N 2) a duplicate backup which replaces in the backup only those items that have M are new or which have changed. A duplicate backup can only be made to a hard  F disk or to memory. The structure of the duplicate backup is logically N identical to the original and items may be freely moved, without the need for K Retrospect or Retrospect Express, at any time from the backup back into to   the active set.   I 3) an archival backup which is similar to an incremental backup with the  L difference being that designated items are also removed from the active set.  L Depending on the user's choice, as many backup sets can be specified as the K user wishes, the backup sets can be a mixture of the above, and the backup   schedule can be automated.      -- James L. Ryan -- TaliesinSoft   ------------------------------  # Date: Sun, 19 Jan 2003 13:56:25 GMT % From: Tony Lawrence <tony@pcunix.com>  Subject: Re: How to Backup OSX@ Message-ID: <d8yW9.738625$%m4.3420648@rwcrnsc52.ops.asp.att.net>   John Santos wrote:B > In article <3RvW9.615140$GR5.409138@rwcrnsc51.ops.asp.att.net>,  > tony@pcunix.com says...  >  >>Howard S Shubs wrote:  >>C >>>In article <4wlW9.731686$%m4.3417976@rwcrnsc52.ops.asp.att.net>, * >>> Tony Lawrence <tony@pcunix.com> wrote: >>>  >>>  >>> K >>>>I think you are misunderstanding something here. How can a backup know  D >>>>whether or not a file that it backed up should or should not be I >>>>restored?  It can't.  There's also the matter of losing one piece of  0 >>>>thev differential, as the article explained. >>>  >>> 7 >>>Okay, I'll explain how it's done, if I can remember.  >>> D >>>A backup set, be it a full backup or an incremental, stores file ? >>>attributes as well as data.  By comparing the "date of last  K >>>modification", you can determine if a file on disk is to be replaced by  A >>>one on tape.  That covers restoring only the latest copy, yes?  >>> K >>>So, how does it know to -delete- old files?  IIRC, by storing directory  F >>>information as well, a directory being a kind of file.  If a newer G >>>version of a directory file doesn't contain file X.Y, then file X.Y  A >>>shouldn't continue to exist, and it gets deleted.  I might be  G >>>remembering wrong, though.  I'll try cross-posting to a group which   >>>would know. >>>  >>K >>Sheesh.  I understand the algorithm it could use, what I'm trying to get  K >>through to you is that it is insufficient.  The whole point of backup is  K >>recover data.  Sometimes files get deleted that should not have been and  J >>that fact isn't always discovered immediately.  Sometimes the fact that J >>something is gone isn't even known: all that is known is that something J >>that used to work does not. Couple that with some bad timing and a full J >>restore and this smart differential backup finds the file on its master B >>but then deletes it yet again on the differerential.. wonderful. >  > @ > Err, if you know a particular file is missing, you restore it F > explicitly.  You don't do a full restore unless you are recovering a > trashed disk.   A I said: " Couple that with some bad timing and a full restore" ..      > E > For some reason this thread is cross-posted between comp.os.vms and B > comp.sys.mac.apps.  I don't know anything about Mac backups, butA > VMS backup certainly allows you to do selective restores of any C > file or group of files from an image, differential or incremental C > backup.  If you do a selective restore, it restores the files you A > ask for (including everything if you ask it nicely) and doesn't A > delete anything.  It is explicitly *not* trying to recreate the : > disk at the time of the most recent backup in this case.   Sigh.  NOT THE POINT.   C Systems crash.  It's funny in a way, but the reliability of modern  I systems contributes to poor backup practice.  Disk failures used to be a  F   fairly common thing for me to deal with, but now I can go a year or C more without seeing even one.  Because of this, people tend to get  I sloppy: I have customers now who have NEVER had a hardware disk failure.  <   So you take someone like that who doesn't have a paranoid D administrator or a PITA consultant  who yells at them when he finds G backups in disarray :-) and you have a recipe for disaster.  Tapes are  D lost, improperly labeled, overwritten, unverified, frozen, whatever.  I Now your "smart differential" does its restore dance, and deletes a file  G that never should have been deleted.  If you are lucky, the problem is  C noticed immediately.  If you are not lucky, the need for that file  G doesn't come up until every tape that had it has been long overwritten.   D Or, you are even less lucky and one of your differentials snaps, is I defective, is lost, mislabeled whatever.  Lucky you.  What now?  Go back  0 to the previous set of course - how old is that?  F The more redundancy, the better.  Period.  I really cannot understand ? why something so simple and obvious has caused such contention.   D The BEST backup is a COMPLETE backup each and every time.  If it is E impossible to do that because of space or time issues, then AND ONLY  " THEN should you use differentials.  
 Sheesh :-)               --  
 Tony Lawrence , Free SCO, Mac OS X and  Linux Skills Tests: % http://aplawrence.com/skillstest.html    ------------------------------  % Date: Sun, 19 Jan 2003 08:58:50 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)6 Subject: Re: HP Sets the Stage for Alpha's Last HurrahL Message-ID: <rdeininger-1901030858500001@user-2ive30k.dialup.mindspring.com>  = In article <9745423a.0301180629.2065e29f@posting.google.com>, . mike.magee@theinquirer.net (Mike Magee) wrote:  E >"Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> wrote in 2 message news:<3e26edfe$1_2@hpb10302.boi.hp.com>...N >> Inquirer reports should always be taken with a large grain of salt.  I haveL >> seen everything from remarkably accurate statements that imply a theft ofH >> internal slides, to statements that can only have been made by peopleN >> without a clue, probably based on some third hand hallway conversation, and' >> also by people with an axe to grind.  > F >I don't mind how much salt you take as long as you read the INQUIRER.5 >But I never steal anything -- and do resent that....   D He didn't say YOU stole the slides, did he?  Do people ever pass youJ information without being authorized to do so?  They would be the thieves.   ------------------------------    Date: 19 Jan 2003 06:34:51 -0800( From: bob@instantwhip.com (Bob Ceculski)6 Subject: Re: Itanium with alpha goodies on the way ...= Message-ID: <d7791aa1.0301190634.72f65155@posting.google.com>   d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<lp-cnTzap9xmRrSjXTWc3A@metrocast.net>... > H > So the sum total of the announcement seems to have been that MontecitoN > slipped a year (and the gap will be filled by a slightly faster Madison withL > a cache-size boost), but the previously unannounced dual-core Montecito inN > 2005 is still on target for that date and its dual-core nature is now public > knowledge. >  > - bill  A and who in the heck cares?  as long as they get the thing running @ well (and they will with the help of alpha team) and it runs vms@ as well as it runs now on alpha with improvements in the future,@ who cares?  amd has had numerous delays ... alpha had delays ...@ as long as the darn thing performs, who cares?  only if it turnsA out to be a dud will someone care, and by the looks of things the C alpha team is slowly using EV7/EV8 alpha designs and slowly putting ? them into itanium and the result is that itanium is starting to B look better and better ... the big event right now is the vms bootB and getting the compilers tuned for itanium ... once that is done, its all downhill ...   ------------------------------  % Date: Sun, 19 Jan 2003 14:33:15 +0100 B From: Michiel Erens <I.dont.want.spam@this.mailaddress.is.invalid>( Subject: Re: MAIL 11 protocol definition7 Message-ID: <3E2AA91B.7552@this.mailaddress.is.invalid>    JF Mezei wrote:  > G > Years ago, I remember having a command procedure one could invoke to  E > send email via DECNET and which could impersonate anyone. The goal  F > was to document/demonstrate the MAIL-11 protocol between nodes (the D > procedure didn't invoke MAIL, it just wrote to a decnet link to a  > remote mail server).   In the VMS hack FAQ e.g.)  http://vmsbox.cjb.net/VMS/vmsHackFAQ.txt    --   ME Posted by news://news.nb.nu    ------------------------------  # Date: Sun, 19 Jan 2003 11:44:51 GMT . From: peter@langstoeger.at (Peter LANGSTOEGER)/ Subject: Re: MX 5.3-2 ignoring vmsmail forwards 2 Message-ID: <TcwW9.67405$TY.611542@news.chello.at>  i In article <b05t6i$cc0$1$8302bc10@news.demon.co.uk>, "Chris Sharman" <chris.sharman@sorry.nospam> writes: . >Upgraded MX from 5.0 to 5.3 eco 2 last night.0 >It appears it's now ignoring VMS mail forwards:  / Only on your site. Mine is working properly ;-)    >Eg. > " >set forward/user=johnny.cash cash > > >Gets ignored - external mail for johnny.cash is now bouncing. > # >Help ! Anyone know how to fix it ?   B No. I suspect your special installation/setup, which I don't know.   But I've another hint:  : MAIL> SET FORWARD/USER=JOHNNY.CASH SMTP%"CASH@domain.name"  E If you omit the SMTP% (or in your case probably a MX%) then the LOCAL G agent/VMSmail combination inserts a blank line thus corrupting the SMTP G headers. If you read them with VMSmail you might not see the difference J (but you might do), but the POP3 (or IMAP4) client sees it and won't work.  I It might not happen (we tested it many years ago and the common sense was K to always use SMTP%, but I don't remember the details now) if you only have H one forward (user1 => user2) but it surely happens if you have a forward" cascade (user1 => user2 => user3).   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Sun, 19 Jan 2003 11:30:12 +0100 $ From: Michael Unger <unger@decus.de>7 Subject: Re: Network Time Protocol (NTP) for OpenVMS ?? * Message-ID: <00A1A336.9828FD3C.8@decus.de>  $ "Dan Allen" <dallen@nist.gov> wrote:   > [...]  >  > > -----Original Message-----. > > From: Arne Vajhoj [mailto:arne@vajhoej.dk]; > > Subject: Re: Network Time Protocol (NTP) for OpenVMS ??  >  > [...]  > / > > You can either get from a system you trust.  > > 9 > > Or better buy a GPS based appliance to get time from.  > 0 > Why, we're free and reasonably trustworthy ;-)  F But can you guarantee a maximum delay between sending the "time stamp"7 and the arrival of that time stamp at a foreign system?   1 So a GPS based local system may be more accurate.    Michael    ------------------------------  % Date: Sun, 19 Jan 2003 16:51:12 +0100 2 From: martin@radiogaga.harz.de (Martin Vorlaender)7 Subject: Re: Network Time Protocol (NTP) for OpenVMS ?? ; Message-ID: <3e2ac970.524144494f47414741@radiogaga.harz.de>   % Michael Unger (unger@decus.de) wrote: & > "Dan Allen" <dallen@nist.gov> wrote:1 > > > You can either get from a system you trust.  > > > ; > > > Or better buy a GPS based appliance to get time from.  > > 2 > > Why, we're free and reasonably trustworthy ;-) > H > But can you guarantee a maximum delay between sending the "time stamp"9 > and the arrival of that time stamp at a foreign system?   D One nice thing about NTP is that it goes to great lengths to includeG those network delays in its calculations to get an accurate time stamp.    cu,    Martin --  G So long, and thanks        | Martin Vorlaender  |  VMS & WNT programmer 4 for all the books...       | work: mv@pdv-systeme.deG In Memoriam Douglas Adams  |   http://www.pdv-systeme.de/users/martinv/ ;             1952-2001      | home: martin@radiogaga.harz.de    ------------------------------   End of INFO-VAX 2003.038 ************************