1 INFO-VAX	Wed, 21 Dec 2005	Volume 2005 : Issue 709       Contents: A more general suggestion  Re: Another Backup suggestion  Re: Database access from COBOL) Re: Deleting alias files (blocks deleted) ) Re: Deleting alias files (blocks deleted) ) Re: Deleting alias files (blocks deleted) ) Re: Deleting alias files (blocks deleted) ) Re: Deleting alias files (blocks deleted) ) Re: Deleting alias files (blocks deleted) ) Re: Deleting alias files (blocks deleted) ) Re: Deleting alias files (blocks deleted) ) Re: Deleting alias files (blocks deleted)  Re: HP's strategy explained :  Re: HP's strategy explained :  Re: HP's strategy explained :  Re: Jon Andruszkiewicz Re: Jon Andruszkiewicz- Merry Christmas to y'all and a Happy New Year & Re: Moving directories of files to dos Re: Passive node in a cluster ? E Re: PHONE error - Invalid specification of node or person. Try again.  Re: RMS expert needed! Re: RMS expert needed! Re: RMS expert needed! Re: RMS expert needed! Re: RMS expert needed! Re: RMS expert needed! Re: RMS expert needed! Re: shadowing questions  Zip question Re: Zip question  F ----------------------------------------------------------------------    Date: 21 Dec 2005 08:11:09 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) " Subject: A more general suggestion3 Message-ID: <jZiVwari7Enh@eisner.encompasserve.org>   Z In article <00A4E936.A695E514.3@tachysoft.com>, Wayne Sewell <wayne@tachysoft.com> writes:3 >>X-MX-Warning:   Warning -- Invalid "From" header. 9 >>From: "Guy Peleg" <guy.peleg@remove_this_header@hp.com>  >>X-Newsgroups: comp.os.vms ( >>Subject: Re: Another Backup suggestion' >>Date: Tue, 20 Dec 2005 14:55:50 +0200  >  >  >  >>L >>Would it help if I'll modify the message indicating the verification phase9 >>started to include date and time? Something similar to:  >>1 >>Starting verification pass on 20-DEC-2005 07:00  >> >   B    Right now "set prefix" is only valid in batch, so I assume it's)    somehow implemented as a special case.   K    It would make sense to create a standard location (pointer in the PCB?)  I    for a prefix which would be of general use to batch, BACKUP, and other G    utilities, accept FAO strings and perhaps some input (late execution G    of lexical functions?) so we can put in things like time, directory,     custom strings, ...  <    And a user API, perhaps hooked into the message facility.  #    Or maybe it should be a postfix.    ------------------------------  % Date: Wed, 21 Dec 2005 11:06:11 -0500 ( From: Wayne Sewell <wayne@tachysoft.com>& Subject: Re: Another Backup suggestion0 Message-ID: <00A4E9A7.03BB50F6.13@tachysoft.com>  2 >X-MX-Warning:   Warning -- Invalid "From" header.8 >From: "Guy Peleg" <guy.peleg@remove_this_header@hp.com> >X-Newsgroups: comp.os.vms' >Subject: Re: Another Backup suggestion & >Date: Wed, 21 Dec 2005 07:09:35 +0200    ; >"JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message ' >news:43A81E27.592CDC67@teksavvy.com...  >> Guy Peleg wrote: I >> > Would it help if I'll modify the message indicating the verification  >phase< >> > started to include date and time? Something similar to: >> >4 >> > Starting verification pass on 20-DEC-2005 07:00 >>I >> It would be nice if all messages (except individual file messgaes from J >> /LOG were prefixed with something, and then the date and then the date. >> >> eg: >>F >> %%%20-DEC-2005 05:45:11 Please mount volume 2, volume label BACKUP21 >> %%%20-DEC-2005 07:00:15 Volume BACKUP2 Mounted 5 >> %%%20-DEC-2005 07:00:15 Starting verification pass  >> >>J >> This way, in a BACKUP/LOG, you could do a SEARCH file "%%%" and get allI >> the relevant messages. And having the time ahead means that the record C >> is easier to parse with a fixed length header, time and ended by  >> variable length text. >  >Okay, I heard you guys......  > $ >Looks like I have more work now ;-) >  >I will add it to V8.3 (not FT)  >   K I would humbly request that any changes you make to the messaging system be 1 compatible with the backup API and its callbacks.   J For instance, the BCK_EVENT_K_USER_MSG and BCK_EVENT_K_ERROR_MSG callbacksN intercept the generation of messages.  If defined, the API does not output theN message at all, but expects the callback (my code rather than HP's) to do it. J I basically call sys$putmsg the same way that the regular code does, while2 also performing my own tapesys-specifc processing.  L However, if the default message output code uses some mechanism *other* thanH sys$putmsg, I HAVE NO WAY TO DO IT, especially if it uses flags and dataN structures inside of backupshr.  If some fancy new code inside the api outputsN exotic output formats and does not provide an entry point so that I can use itN from my callback, I have lost a capability.  I cannot generate the message the way straight backup does.   O So far, there is nothing that backup.exe can do that I cannot in vmsbu.exe. But E use of routines I can't get to would change that.  I can no longer do % everything that straight backup does.   N Presumably there was a reason for the creation and documentation of the backupJ api.  I embraced it in good faith and am now totally dependent on it.  AnyG functional changes to vms backup should also be available from the api.   M In this particular case, a possible simple solution might be to not *require* L me to actually generate the message myself.  I don't really want to anyway. L The reason for the callback is to do my own processing in *conjunction* withG the message.  Perhaps if I gave you a bad return code or something like K resignal in condition handlers, that would tell you to generate the message ( yourself, which would solve the problem.  N The api documentation talks about someday making use of callback return codes,* but this was apparently never implemented.   Wayne O =============================================================================== N Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html    O =============================================================================== P Jake Blues:"You traded the Caddy for a microphone? ...... Okay, I can buy that."   ------------------------------    Date: 21 Dec 2005 08:04:16 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ' Subject: Re: Database access from COBOL 3 Message-ID: <VoUGAttr0RTC@eisner.encompasserve.org>   W In article <40r57aF1c236sU4@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:  > G > While better than 1978, even 1979 would have proven rather difficult.  >   E    I think IBM was selling PCs for about $4K US at the time.  The IBM C    PC clone was not yet the estabished desktop system.  People were G    still buying a lot of CP-M based systems, Apple was making the Apple C    II, and $4K was in the magnitude of a year's tuition at the same 
    school.  F    So I went to Rutgers for a few hundred dollars a semester, instead.   ------------------------------  % Date: Wed, 21 Dec 2005 02:29:08 -0500 ( From: Bill Todd <billtodd@metrocast.net>2 Subject: Re: Deleting alias files (blocks deleted)G Message-ID: <7L-dnSe9hIPbmTTenZ2dnUVZ_sSdnZ2d@metrocastcablevision.com>   
 AEF wrote:   ...   G > "Seems"? It's not "seems"; it "does". Why is everyone so reluctant to  > acknowledge that I was right?   D It's a somewhat understandable (though hardly rational) reaction to F someone who is occasionally something of an asshole, even (or perhaps / especially) when he is wrong (since you asked).    ...   F > Actually, on second thought, DELETE also works. It removes the aliasH > and leaves the primary in place. When it says 3 blocks deleted or someE > other positive number the number of blocks that will be reported by # > DIRECTORY will be that much less.   F But since the blocks were not in fact *actually* deleted, it is still J erroneous (unless one accepts a rather contorted definition of 'deleted').   ...   - > I am not familar with "aliases" on the Mac,   I Neither am I, but my guess is (especially in later Mac OS versions) that  D they're the same as Unix 'hard links'.  Hard links are arguably VMS E aliases done right (or at least as right as they can be done):  they  F incorporate a reference count such that the actual file sticks around D until the last remaining hard link (directory entry) to the file is ; removed, at which point the file itself is deleted as well.   F But there's still no good (read:  arbitrarily scalable) way to handle H quotas (I'm not going to go down that moderately complex rat-hole:  you G can take my word for it or work it out yourself), nor to avoid letting  F someone link to a file for which they don't have delete privilege and E then get stuck with it after all other links to it have been removed.      but I don't know what @ > aliases you refer to on Windows. Are you talking about WindowsI > "shortcut icons"? Those are not file aliases. They are files that point E > to the "primary" file. The parallel in VMS would be a DCL file that 
 > contains >  >    $ @blah.com > C > which is not an alias. Please clarify what you mean by "aliases".   I Perhaps the NTFS facility that's essentially identical IIRC to Unix hard  D links ('shortcuts' are essentially equivalent to Unix 'soft links', - which are sometimes called 'symbolic links').    ...   1 > Aliases on VMS are not a user-friendly feature.   H Indeed.  Unix hard links are at least fairly user-friendly, even though . they make system features like quotas awkward.     That's why I thoughtH > they were introduced in an ad hoc manner for cluster system disks. But= > another poster tells me that VMS aliases pre-date clusters.   C Actually, they pre-date VMS itself:  ODS-1 on RSX had Enter/Remove  0 facilities that could be used to create aliases.   ...   1   there is nothing in DIRECTORY output that tells G > you a file is an alias. You have to scan all directories on a disk to E > do that, so I think that's asking a bit much of basic file-handling I > commands like DELETE. Think how long a simple DELETE command would take : > to execute if it had to scan all .DIR files on the disk!  G If DELETE can determine not to delete the actual file because the link  I is an alias, then clearly the information is available (and is available  < to DIRECTORY as well:  it just doesn't bother to report it).  G Hoff already mentioned that there is no 'alias flag', so I'll describe  F my very vague recollection of how this works and let anyone with more ; recent (or more solid) information correct it if necessary.   F ODS-1 worked the way some people here thought ODS-2 worked:  a DELETE G operation through *any* link would delete the actual file contents and  H leave any other links dangling, because in ODS-1 there was no way to be F sure whether a link was the original one or an alias.  In particular, D the RENAME operation did not reliably update the name in the file's F header, so that was not a reliable indicator (and the file header did F not contain the FID of the directory in which the file was originally F created, either - so an alias with the same name in another directory $ would have confused matters anyway).  C ODS-2 both included the FID of the parent directory containing the  A 'primary' link and updated the name in the file header on RENAME  H (presumably only when the file was RENAMEd through the 'primary' link), H and presumably changed the parent-directory FID if the file was RENAMEd G through the primary link to another directory.  So the combination was  H sufficient to ascertain whether a directory entry was an alias or not - C as long as no one screwed around manually with those header fields.   I This does not, however, explain the sequence where an alias was created,  F the original link REMOVEd, and then the file successfully DELETEd via H the alias.  Perhaps DELETE through an alias checks for the existence of G the primary link and if not present figures that it had better get rid  F of the file as well as the alias (just in case it's the *last* alias).   - bill   ------------------------------  % Date: Wed, 21 Dec 2005 02:43:53 -0500 ( From: Bill Todd <billtodd@metrocast.net>2 Subject: Re: Deleting alias files (blocks deleted)G Message-ID: <UISdnRCgFusmmjTenZ2dnUVZ_vidnZ2d@metrocastcablevision.com>    Bill Todd wrote:   ...   I > Hoff already mentioned that there is no 'alias flag', so I'll describe  H > my very vague recollection of how this works and let anyone with more = > recent (or more solid) information correct it if necessary.   H Ah - through the magic of Usenet propagation latency it seems that Hein A had already stepped up to the plate.  At least the quality of my  ; recollections and guesses didn't prove unduly embarrassing.    - bill   ------------------------------  % Date: Wed, 21 Dec 2005 07:47:15 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 2 Subject: Re: Deleting alias files (blocks deleted), Message-ID: <43A94ED2.C4B875E6@teksavvy.com>   Hein RMS van den Heuvel wrote:1 > Admittedly the 'X blocks' deleted is confusing.     Many thanks for the explanation.  B > Now if one does a $set file/remove on the directory entry in theG > directory for the backlink, then this remove will blank the backlink. F > Now a delete through any other alias can no longer determine whetherD > there are siblings, and becomes a real delete, again leaving other > alias entries as as orphans.  C so SET FILE/REMOVE on a real file will actually update INDEXF.SYS's  record for that file-ID ?   A Based on your explanation prevous to that paragraph, I would have E thought that it would only play with the directory file and not touch E INDEXF.SYS. Then, when you do a DELETE on the alias, the system would G realise that the backlink in INDEXF.SYS is no longer valid and act as a  real delete.   ------------------------------    Date: 21 Dec 2005 05:48:07 -0800< From: "Hein RMS van den Heuvel" <heinvandenheuvel@gmail.com>2 Subject: Re: Deleting alias files (blocks deleted)C Message-ID: <1135172887.744525.142750@g43g2000cwa.googlegroups.com>   ' Bill... caught in a time warp huh? :-). B AEF... the variant with the same file being entered in a directory; twice was discussed already I taught on (too) quick reading    JF,   D "so SET FILE/REMOVE on a real file will actually update INDEXF.SYS's record for that file-ID ?"& Yes. Takes less than 5 minutes to try:   $ cop nl:  [.a]x.x $ set file/ent=[.b] [.a]x.x  $ dir/file a.dir,b,[.%]x.x  3 $ dir/file a.dir,b,[.%]x.x/nohead/notrai/wid=fil=45 < EISNER$DRA3:[DECUSERVE_USER.HEIN]A.DIR;1       (26735,303,0)< EISNER$DRA3:[DECUSERVE_USER.HEIN]B.DIR;1       (28211,583,0)= EISNER$DRA3:[DECUSERVE_USER.HEIN.A]X.X;1       (38219,1299,0) = EISNER$DRA3:[DECUSERVE_USER.HEIN.B]X.X;1       (38219,1299,0) @ $ pipe dump/head/bloc=coun=0 [.%]x.x | sea sys$pipe "file ident"8     File identification:                  (38219,1299,0)1     Extension file identification:        (0,0,0) 7     Back link file identification:        (26735,303,0) 8     File identification:                  (38219,1299,0)1     Extension file identification:        (0,0,0) 7     Back link file identification:        (26735,303,0)  $ set file/remo [.a]x.x.@ $ pipe dump/head/bloc=coun=0 [.%]x.x | sea sys$pipe "file ident"8     File identification:                  (38219,1299,0)1     Extension file identification:        (0,0,0) 1     Back link file identification:        (0,0,0)   4 "when you do a DELETE on the alias, the system wouldG realise that the backlink in INDEXF.SYS is no longer valid and act as a 
 real delete."    Yes, pretty much. G The back link itself might point to a valid directory, but if that file D id does not live in a directory record in that directory, then it is' believed invalid, and the file deleted. E This is slightly trickier to arrange, and set file/remove updates the 	 backlink.  But witness:  ( and certainly imho not worth the effort,   $ set file/ente=[.a] [.b]x.x3 $ dir/file a.dir,b,[.%]x.x/nohead/notrai/wid=fil=45 < EISNER$DRA3:[DECUSERVE_USER.HEIN]A.DIR;1       (26735,303,0)< EISNER$DRA3:[DECUSERVE_USER.HEIN]B.DIR;1       (28211,583,0)= EISNER$DRA3:[DECUSERVE_USER.HEIN.A]X.X;1       (38219,1299,0) = EISNER$DRA3:[DECUSERVE_USER.HEIN.B]X.X;1       (38219,1299,0) B $! Notice how the new directory is picked up, not the new old [.b]? $ pipe dump/head/bloc=coun=0 [.%]x.x | sea sys$pipe "back link" 7     Back link file identification:        (26735,303,0) 7     Back link file identification:        (26735,303,0)  $ open/read/write a a.dir  $ read a rec/ $ write sys$output %x10000+f$cvsi(10*8,2*8,rec)  38219  $ rec[10*8,2*8]=12345  $ write/update a rec	 $ close a + $ dir/file [.%]x.x/nohead/notrai/wid=fil=45 = EISNER$DRA3:[DECUSERVE_USER.HEIN.A]X.X;1       (12345,1299,0) = EISNER$DRA3:[DECUSERVE_USER.HEIN.B]X.X;1       (38219,1299,0)  $ dele [.B]X.X;1 $ type [.A]x.xF %TYPE-W-OPENIN, error opening EISNER$DRA3:[DECUSERVE_USER.HEIN.A]X.X;1 as input -RMS-E-FNF, file not found  D btw.... RENAME was changed in the VMS 4.0 - 5.0 timeframe from firstF enterring the second alias first and then removing the original to theG current remove first, add next, precisely for the reasons demonstructed A above. By removing first,   a reasonable backlink becomes easy to G manage. Should the operation get interupted mid-stream, then there will ? just be a lost file, which can be found back with no data loss.    Cheers,  Hein.    ------------------------------    Date: 21 Dec 2005 05:44:15 -0800$ From: "AEF" <spamsink2001@yahoo.com>2 Subject: Re: Deleting alias files (blocks deleted)B Message-ID: <1135172655.737305.49470@z14g2000cwz.googlegroups.com>   Bill Todd wrote: > AEF wrote: >  > ...  > I > > "Seems"? It's not "seems"; it "does". Why is everyone so reluctant to ! > > acknowledge that I was right?  > E > It's a somewhat understandable (though hardly rational) reaction to G > someone who is occasionally something of an asshole, even (or perhaps 1 > especially) when he is wrong (since you asked).    Look who's talking.    [...]   D > ODS-2 both included the FID of the parent directory containing theB > 'primary' link and updated the name in the file header on RENAMEI > (presumably only when the file was RENAMEd through the 'primary' link), I > and presumably changed the parent-directory FID if the file was RENAMEd H > through the primary link to another directory.  So the combination wasI > sufficient to ascertain whether a directory entry was an alias or not - E > as long as no one screwed around manually with those header fields.  > J > This does not, however, explain the sequence where an alias was created,G > the original link REMOVEd, and then the file successfully DELETEd via I > the alias.  Perhaps DELETE through an alias checks for the existence of H > the primary link and if not present figures that it had better get ridH > of the file as well as the alias (just in case it's the *last* alias).  E Perhaps you didn't read the other posts that explains this just fine.    >  > - bill   ------------------------------  % Date: Wed, 21 Dec 2005 08:46:11 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 2 Subject: Re: Deleting alias files (blocks deleted), Message-ID: <43A95C9C.EBD2A9EE@teksavvy.com>   Bill Todd wrote:J > Neither am I, but my guess is (especially in later Mac OS versions) thatE > they're the same as Unix 'hard links'.  Hard links are arguably VMS F > aliases done right (or at least as right as they can be done):  theyG > incorporate a reference count such that the actual file sticks around E > until the last remaining hard link (directory entry) to the file is 
 > removed,  F Not in MAC OS Classic. YOu could have aliases pointing to non existentG files. When you try to access it, the OS puts up a dialog and gives you > the chance to find the original file and reset the link to it.   ------------------------------    Date: 21 Dec 2005 08:15:52 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 2 Subject: Re: Deleting alias files (blocks deleted)3 Message-ID: <chmHRMM+h6zv@eisner.encompasserve.org>   \ In article <43A8CDA3.70825379@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: > OK, one more test: > * > Create a 100k block file  DISK:[CAKE]A.A0 > SET FILE DISK:[CAKE]A.A/enter=DISK:[000000]B.B > SHOW DEV DISK  > DELET/LOG [000000]B.B; > SHOW DEV DISK  > J > DELETE says it has deleted 100k blocks. But Show DEV does not change the# > amount of free space on the disk.  >   E    The amount of free space is cached and not guarranteed to be up to C    date unless you exersize a dismount, mount/rebuild cycle, at the B    end of which the actual free space will be more than the listed$    amount by the initial cache size.   ------------------------------    Date: 21 Dec 2005 08:20:22 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 2 Subject: Re: Deleting alias files (blocks deleted)3 Message-ID: <wgWjADGbjOD2@eisner.encompasserve.org>   i In article <1135121818.724852.79800@g47g2000cwa.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes:  > G > I noticed a new oddity: If you have B.B as an alias for A.A, then SET H > FILE/REMOVE A.A;!, you have only an alias left (B.B) which DFU reportsB > as an alias for []A.A. But if you then DELETE B.B, delete reallyH > deletes it (and it cannot be recovered with ANAL/DISK) even though theF > filename in the file header doesn't match the name in the directory. > (Checked on VAX/VMS 6.2)  D    Aliases are NOT links.  If B.B is an alias for A.A and you deleteH    A.A, then the file is gone, including the A.A entry in the directory,G    the A.A entry in the index, the A.A header, and the A.A data blocks.   C    B.B is then a broken alias.  When you delete B.B, only the alias     is removed.  B    This is and always has been the way aliases worked from VMS 0.xD    (where you had to use PIP to manage them), until links were addedF    to ODS-5:  deleting the original is the only way to delete the fileC    and always deletes the file.  Deleting an alias removes only the 4    alias entry and has no affect on the actual file.          ------------------------------    Date: 21 Dec 2005 08:28:51 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 2 Subject: Re: Deleting alias files (blocks deleted)3 Message-ID: <FwXkzSe9XUim@eisner.encompasserve.org>   j In article <1135140046.559809.316900@g14g2000cwa.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes:  D > So you are correct. Your explanation actually covers both cases! I/ > assume by DID you mean the back link file id.   E    The DID is the FID of the directory in which the file was created. E    It can be found (along with the current name of the original file) B    in the header, so the name in the header plus the DID uniquely     identify the original file.  B    An alias in a different directory will not have a directory FID?    mathing the DID.  An alias in the same directory will have a B    different name in at least one field (version number included).  G    Although I'm fairly sure this is all well documented, playing around ?    with aliases, FIDs, and dump/header can be revealing for the     uninitiated.   E    Someday, when I get 8.2 or later on my hobbyist systems, I'll play D    around with links.  Never liked hard links on UNIX, not sure I'llA    want to use them on VMS, they can play havoc with disk quotas.   :    But then I never did like, and rarely use, disk quotas.   ------------------------------  # Date: Wed, 21 Dec 2005 15:00:56 GMT - From: Arthur Entlich <e-printerhelp@mvps.org> & Subject: Re: HP's strategy explained :- Message-ID: <I6eqf.164475$Gd6.88730@pd7tw3no>   > Might be a nice excuse for making you feel better about being G unsupportive to workers and people who are trying to make a living and  B raise their families, but it doesn't convince me, and its a bogus  argument in this case.  C Companies are not necessarily making changes due to "progress" but  I because they think they are evolving the way their stockholders wish, or  I at least, that whatever they do will be perceived as "positive" by those  @ stockholders.  Unfortunately, often stockholders have the ethicsG of sharks, in other words, none.  Many don't even know what it is they  1 "own", only if it is making them a profit or not.   H Its no way to run a company, and in the longer term and bigger picture,  it almost always fails.   F Apparently, there is still a market for those buggy whips... they get B used on the workers, but forcing them to work under lowest common E denominator conditions or they will be replaced with workers who are  > even more desperate and will work for less under even harsher F conditions.  This is the new "global" international workplace we live I with, and it is creating a downward spiral, rather than lifting up those  G who are worse off.  And it is indeed fueled by greedy stockholders who   only care about their returns.   Art    Stephen Grossman wrote:   D > In article <1134600217.498950.61910@z14g2000cwz.googlegroups.com>, > mark_hpq@yahoo.com wrote:  >  >  >>Latest news on the layoffs : >> >>http://hpwfr.blogspot.com/ >  > D > Every time you buy x  instead of y, you are indirectly causing the > downsizing of H > companies which make X. Just think of all the lost jobs in the candle, > horse and buggy,  C > and typewriter companies when anti-social, unfeeling, plutocratic  > consumers startingQ > buying light bulbs, cars and computers! And what about the makers of man-pushed R > plows!! And water buckets balanced on the head!! And spears!! Mud-brick makers!!' > I weep for them all. Solidarity Now!!  >    ------------------------------  # Date: Wed, 21 Dec 2005 15:02:53 GMT - From: Arthur Entlich <e-printerhelp@mvps.org> & Subject: Re: HP's strategy explained :, Message-ID: <x8eqf.156583$ki.33732@pd7tw2no>  > Might be a nice excuse for making you feel better about being G unsupportive to workers and people who are trying to make a living and  B raise their families, but it doesn't convince me, and its a bogus  argument in this case.  C Companies are not necessarily making changes due to "progress" but  I because they think they are evolving the way their stockholders wish, or  I at least, that whatever they do will be perceived as "positive" by those  @ stockholders.  Unfortunately, often stockholders have the ethicsG of sharks, in other words, none.  Many don't even know what it is they  1 "own", only if it is making them a profit or not.   H Its no way to run a company, and in the longer term and bigger picture,  it almost always fails.   F Apparently, there is still a market for those buggy whips... they get B used on the workers, but forcing them to work under lowest common E denominator conditions or they will be replaced with workers who are  > even more desperate and will work for less under even harsher F conditions.  This is the new "global" international workplace we live I with, and it is creating a downward spiral, rather than lifting up those  G who are worse off.  And it is indeed fueled by greedy stockholders who   only care about their returns.   Art    Stephen Grossman wrote:   D > In article <1134600217.498950.61910@z14g2000cwz.googlegroups.com>, > mark_hpq@yahoo.com wrote:  >  >  >>Latest news on the layoffs : >> >>http://hpwfr.blogspot.com/ >  > D > Every time you buy x  instead of y, you are indirectly causing the > downsizing of H > companies which make X. Just think of all the lost jobs in the candle, > horse and buggy,  C > and typewriter companies when anti-social, unfeeling, plutocratic  > consumers startingQ > buying light bulbs, cars and computers! And what about the makers of man-pushed R > plows!! And water buckets balanced on the head!! And spears!! Mud-brick makers!!' > I weep for them all. Solidarity Now!!  >    ------------------------------  # Date: Wed, 21 Dec 2005 15:08:56 GMT - From: Arthur Entlich <e-printerhelp@mvps.org> & Subject: Re: HP's strategy explained :- Message-ID: <ceeqf.164626$Gd6.99117@pd7tw3no>   -  > I think you got your "X" and "Y" mixed up.   G Well, you never know, maybe he was thinking that company X was selling  8 their goods below cost and losing money on each sale ;-)   Art    Charles wrote:  F > On Tue, 20 Dec 2005 16:10:27 GMT, sdgross@att.net (Stephen Grossman) > wrote: >  > D >>In article <1134600217.498950.61910@z14g2000cwz.googlegroups.com>, >>mark_hpq@yahoo.com wrote:  >> >> >>>Latest news on the layoffs :  >>>  >>>http://hpwfr.blogspot.com/  >>D >>Every time you buy x  instead of y, you are indirectly causing the >>downsizing of H >>companies which make X. Just think of all the lost jobs in the candle, >>horse and buggy,  C >>and typewriter companies when anti-social, unfeeling, plutocratic  >>consumers startingQ >>buying light bulbs, cars and computers! And what about the makers of man-pushed R >>plows!! And water buckets balanced on the head!! And spears!! Mud-brick makers!!' >>I weep for them all. Solidarity Now!!  >  >  > , > I think you got your "X" and "Y" mixed up.   ------------------------------    Date: 21 Dec 2005 08:33:06 -0800$ From: "Ed Wilts" <ewilts@ewilts.org> Subject: Re: Jon Andruszkiewicz C Message-ID: <1135182786.726232.198210@g44g2000cwa.googlegroups.com>    Hal Kuff wrote: L > Hi, if you are out there I would like to continue an earlier discussion on' > long distance shadows and journals...   F John's not the only person that has experience with this stuff.  Keith@ Parris can be considered an expert and is active in comp.os.vms.G What's the issue?  My shadows aren't very far apart (1200 feet) but did @ involve Gigaswitches, ATM cards, and all that fun stuff before I< migrated it all to the SAN and now shadow across 1Gbps ISLs.  	    .../Ed    ------------------------------    Date: 21 Dec 2005 08:57:08 -0800 From: john.atoz@gmail.com  Subject: Re: Jon Andruszkiewicz C Message-ID: <1135184228.761301.182950@g47g2000cwa.googlegroups.com>   G That is very true, I am not the only person with experience in HBVS ... G but I was the project leader for several years, until I "retired" on 14  October 2005 :-)  - What earlier discussion are you referring to?   	 Regards,           john   ------------------------------  % Date: Wed, 21 Dec 2005 11:40:55 -0500 C From: "David Turner, Island Computers US Corp" <dbturner@icusc.com> 6 Subject: Merry Christmas to y'all and a Happy New Year8 Message-ID: <usfqf.8546$RZ6.5702@bignews7.bellsouth.net>  " Thanks everyone for a great 2005 !  ) To those that don't believe in Christmas.  Have a nice day !    David        --     David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404  Tel: 912 447 6622 X201 Cell: 912 447 6622 X252  Fax: 912 201 0402  Email: dbturner@icusc.com  Web: http://www.islandco.com% ===================================== < All orders are subject to the following terms and conditions. of sale. These should be read before ordering.% http://www.islandco.com/warranty.html    ------------------------------    Date: 21 Dec 2005 04:58:09 -0800' From: "tadamsmar" <tadamsmar@yahoo.com> / Subject: Re: Moving directories of files to dos B Message-ID: <1135169889.525447.48600@f14g2000cwb.googlegroups.com>   Martin Vorlaender wrote:( > tadamsmar <tadamsmar@yahoo.com> wrote: > > Dave Froble wrote:J > >> If you have network connectivity, what's wrong with FTP.  Some of theE > >> better products will automatically detect text vs binary.  [...]  > > D > > I have used it.  The mget on some dos boxes make you acknowledgeE > > each file which is a hassle and hard to automate.  I do have some D > > automated scripts that just say "Y" over and over again too many
 > > times. > B > FTP has a "prompt" command to switch those acknowledgements off. >   C Thanks for point out "prompt"  that makes my automatic scripts much  cleaner.   > cu, 
 >   Martin > --C >                            | Martin Vorlaender  |  OpenVMS rules! 6 >  UNIX is user friendly.    | work: mv@pdv-systeme.deI >  It's just selective about |   http://www.pdv-systeme.de/users/martinv/ = >  who its friends are.      | home: martin@radiogaga.harz.de    ------------------------------  % Date: Wed, 21 Dec 2005 05:07:17 -0800 ( From: Jeff Cameron <roktsci@comcast.net>( Subject: Re: Passive node in a cluster ?0 Message-ID: <BFCE9385.194DE%roktsci@comcast.net>  J On 12/20/05 6:16 AM, in article 43A8123E.31295310@teksavvy.com, "JF Mezei"% <jfmezei.spamnot@teksavvy.com> wrote:    > Just an idea:  > J > It would be nice to have a node in a cluster which would NOT freeze uponI > loss of quorum. (consider a node that monitors the cluster, disk space, 1 > who has what disks mounted in the cluster etc).  > E > If that node had 0 votes, its own system disk, and would only mount J > cluster disks in a read only mode, could this cause problems in cases of > cluster partitioning ? >  > C > Why am I bringing this up ? I was using one node's serial port to I > monitor the OPA0: of another node. But when quorum was lost, I could no G > longer send command/read text from that OPA0 because 6the workstation E > was also frozen. Having a node which woudln't freeze would be nice.  > D > (yes, I know, make that node outside of the cluster, but there areM > advantages to being inside the cluster in terms of cluster management etc).   L I think you are missing the whole point of having quorum in a cluster. It isI a good mechanism that freezes nodes when many nodes are crashing, so that I they too don't crash. It is certainly easy enough to readjust quorum when 1 you are purposely removing (shutting down) nodes.   K Loosing quorum and freezing is much more desirable than crashing. It allows E you to diagnose the issue and take action to prevent other nodes from  experiencing the same fate.   H Plus with the flexibility of votes and quorum disks, you can achieve any. kind of dependency or non-dependency you wish.   ------------------------------  + Date: Wed, 21 Dec 2005 10:40:39 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)N Subject: Re: PHONE error - Invalid specification of node or person. Try again.$ Message-ID: <dobbf7$v8m$1@online.de>  @ In article <slrndqfrkf.40u.usenet@zappy.catbert.org>, Dan Foster <usenet@evilphb.org> writes:    M > > I disagree.  DECnet, if not always, at least for a long time has been an  6 > > optional product.  PHONE is automatically there.   > J > That's true. I suspect that ultimately it's a 'business case driver' for8 > HP to figure out where it stands in the priority list. > F > PHONE, at one time, was a heavily used feature when the OpenVMS user1 > base was substantially larger than it is today.  > H > It is still useful, but I suspect that it would perhaps be more usefulE > to devote the time elsewhere, unless you really do have quite a few ! > customers screaming for it. :-)  > < > I would classify this as a 'nice to have' feature at best. > H > A workaround does already exist -- one could install DECnet Phase V ifG > that functionality was really required. (As previously noted; even if 9 > it's an optional component, it's still a valid option.)   H Personally, it's not that urgent, even for a hobbyist cluster.  However,A since I want to be able to PHONE between different nodes within a G cluster, installing DECnet Phase IV would be enough---no need for Phase B V or running it over IP.  I don't want to use IP, I want to PHONE I between nodes, and since all have TCPIP already, but as of now no DECnet  H (will change when I find the time), it would be nice if PHONE would run 4 over TCPIP, at least optionally, like, say, MONITOR.   ------------------------------  % Date: Wed, 21 Dec 2005 12:32:44 +0000 0 From: Chris Sharman <chris.sharman@sorry.nospam> Subject: Re: RMS expert needed! 4 Message-ID: <dobi1c$niv$1$830fa17d@news.demon.co.uk>  / Phillip Helbig---remove CLOTHES to reply wrote: C > If anyone wants to experiment, I could send a UUENCODED ZIP file  I > (created with the /VMS qualifier, of course) containing one file which  I > is OK and another which isn't.  Since about 100 files are affected, it  = > would be nice to have a procedure which repairs the damage.   > Either that, or the outputs from a dir/full and a dump/bl=co=1  4 Repair is usually a set file/attr and/or convert/fdl  
 Chris Sharman  (ccagroup co uk)   ------------------------------  % Date: Wed, 21 Dec 2005 05:16:23 -0800 ( From: Jeff Cameron <roktsci@comcast.net> Subject: Re: RMS expert needed! 0 Message-ID: <BFCE95A7.194FF%roktsci@comcast.net>  @ On 12/21/05 4:58 AM, in article dobji2$ln2$1@online.de, "PhillipJ Helbig---remove CLOTHES to reply" <helbig@astro.multiCLOTHESvax.de> wrote:  7 >> Repair is usually a set file/attr and/or convert/fdl  > F > Yes, but I think that is what caused the problem in the first place!3 > Obviously, my command was not completely correct.   L I don't believe that these tools are what caused the problem, but rather the) improper use of the tools is the culprit.   5 Do you have the batch job that did this to the files?    ------------------------------    Date: 21 Dec 2005 05:16:25 -0800< From: "Hein RMS van den Heuvel" <heinvandenheuvel@gmail.com> Subject: Re: RMS expert needed! C Message-ID: <1135170985.681055.290840@o13g2000cwo.googlegroups.com>    There is hope :-).     just use this:    SET FILE /ATTR=RFM=STMLF   F In a variable length file, the first two bytes are a word integer with record length.E IOn your example the first one is 6, the second 0, then next 6 again.   F The the corrupted file we see 'text', which when interpreted as recrodF size is beyond EOF and invalid. However, we see 0Ax, 0Ax following theG first 6 bytes. Those are two new-line record terminators, one for the 6 2 byte first record, one for the 0 byte next record.     Cheers,  Hein.    ------------------------------  % Date: Wed, 21 Dec 2005 08:48:42 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: RMS expert needed! , Message-ID: <43A95D32.C82D12CC@teksavvy.com>  / Phillip Helbig---remove CLOTHES to reply wrote: E > I have several files (.HTML files actually, but that doesn't matter H > here) which I have managed to "corrupt" in the sense that I can't readJ > them anymore.  That is, EDT or EVE doesn't show anything, neither does aI > web browser when accessing them through a web server, TYPE doesn't show  > anything.  >  > I can, however, DUMP them.   Can you post :   DIR/FULL for one file    and  DUMP/BLOCK=(COUNT=1)  1 You can also try DUMP/RECORD to see what it says.    ------------------------------    Date: 21 Dec 2005 05:52:45 -0800< From: "Hein RMS van den Heuvel" <heinvandenheuvel@gmail.com> Subject: Re: RMS expert needed! B Message-ID: <1135173165.045256.23720@g49g2000cwa.googlegroups.com>   Right Chris.G Minor tweak... You can use 'the shortest fdl file in the world" for the  convert:   $CONVERT/FDL=NL: old new  : For more FDL fun, see slide set attached to last reply to:K http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=965966   / Of course you 'had to be there' for max effect.    :-)    ------------------------------  + Date: Wed, 21 Dec 2005 14:12:44 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: RMS expert needed! $ Message-ID: <dobnss$ve9$3@online.de>  D In article <dobnj9$ve9$2@online.de>, helbig@astro.multiCLOTHESvax.de3 (Phillip Helbig---remove CLOTHES to reply) writes:     > > There is hope :-). > >  > > just use this: > >  > >  SET FILE /ATTR=RFM=STMLF  > > J > > In a variable length file, the first two bytes are a word integer with > > record length.I > > IOn your example the first one is 6, the second 0, then next 6 again.  > > J > > The the corrupted file we see 'text', which when interpreted as recrodJ > > size is beyond EOF and invalid. However, we see 0Ax, 0Ax following theK > > first 6 bytes. Those are two new-line record terminators, one for the 6 6 > > byte first record, one for the 0 byte next record.  C Is there a quick and easy way to see if the file is "really" STMLF?    ------------------------------    Date: 21 Dec 2005 08:11:28 -0800< From: "Hein RMS van den Heuvel" <heinvandenheuvel@gmail.com> Subject: Re: RMS expert needed! C Message-ID: <1135181488.687180.256780@z14g2000cwz.googlegroups.com>   0 >> First, why do some files work and some don't,  D Because the contents does not match the description. Variable length@ record files have a specific, unique internal structure chainingC records through word aligned record length words. Not all bytes are ; data. You will often find binary zero bytes in those files.   @ Stream-lf files are just a bunch of bytes which may or might notG represent individual records. all bytes are data. Some are special data E (record terminator). Most applications dealing with those file to not " take kindly to 'binary zero bytes'  3 >> I'll BACKUP everything first then give it a try.   , Sure, but it's a waste of time in this case.2 You are not touching the data, only the meta-data.8 And there is an antidote for every set file/atrr action.$ (in this case SET FILE/ATTR=RFM=VAR)    F >> Is there a quick and easy way to see if the file is "really" STMLF?  F There was an attempt made in OpenVMS 7.1 to add 'Heuristic' support inF RMS to recognize files which are not what they claim to be, notably toG solve some Pathworks issues. The main documentation was pulled, the new G features still mentioned it, and the sysgen param is there: $MCR SYSGEN  SHOW/RMS  --> RMS_HEURISTIC.# I don't think it will help you any.   D For 'Normal' text files it is enough to check the second byte in the file. G If it is binary 0, then the file is likely to be a variable length text 3 file, with the first record shorter than 256 bytes. F A better test is for the value of the first word to be larger then the largest expected record.G A more extensive test would be to follow that offset, round up, and see A if at that location there is an other 'reasonable' record length. B If the first bytes are plain Ascii, then you could hunt for the LFF and/or CR  characters in the first  block or so to see whether either,1 or the combo CR-LF is a likely record terminator.     F IMHO you are better of practising 'safe computing' and egt a handle on= where the files are coming from, how they need to be treated.     more than you wanted to know no? :-).   Hein.    ------------------------------    Date: 21 Dec 2005 07:04:01 -0800 From: john.atoz@gmail.com   Subject: Re: shadowing questionsC Message-ID: <1135177441.889082.175260@f14g2000cwb.googlegroups.com>   D >> The Shadowing developer is also looking into the possibility thatK >> Shadowing may be able to, at the point where a member has to be removed, E >> convert a mini-merge bitmap to a mini-copy bitmap for that removed K >> member, and thus track all the changes subsequent to its loss, and allow ? >> a mini-copy operation to reintegrate it later. This would be H >> particularly handy to allow mini-copies in disaster-tolerant clustersI >> after a failure which results in downtime of either one site or of the  >> inter-site link.    >Yes!   B Unfortunately, in your case, if the VAX (which I assume is runningF V7.3, or a variant that supports mini copy) has the DSA device (a.k.a.D as a Virtual Unit ... VU) mounted, you will not be allowed to enableD Host Based Mini Merge (HBMM) on that VU.  The new V8.3 feature, thatB automatically "converts" a HBMM bitmap to a mini copy bitmap, willG require that HBMM is enabled on the VU to function correctly.  This was 8 a decision that was made very early in the HBMM project.  $ John Andruszkiewicz (a.k.a. AtoZ :-)   ------------------------------  # Date: Wed, 21 Dec 2005 15:09:21 GMT ; From: "Jeffrey H. Coffield" <jeffrey@digitalsynergyinc.com>  Subject: Zip question = Message-ID: <Beeqf.34879$7h7.4202@newssvr21.news.prodigy.com>   E Is there a version of zip on either VMS or Linux that can zip a file  	 over 2GB?   D Alternately, is there a way to get VMS Backup to write multiple save0 sets so that each is below a given maximum size?  E I am backing up several systems by making save sets, zipping them and D then burning them to a dvd on Linux system. (I have not been able toB get a dvd to work on VMS and not all the systems are new enough to have an ide bus.)   C A related question, can someone tell me a scsi dvd drive that would  work with VMS?  
 Jeff Coffield    ------------------------------  + Date: Wed, 21 Dec 2005 09:36:36 -0600 (CST) * From: sms@antinode.org (Steven M. Schweda) Subject: Re: Zip question 2 Message-ID: <05122109363609_2024C3E0@antinode.org>  ; From: "Jeffrey H. Coffield" <jeffrey@digitalsynergyinc.com>   G > Is there a version of zip on either VMS or Linux that can zip a file   > over 2GB?   F    Not an officially released version, but the latest beta source kitsH for Zip 3 and UnZip 6 should do the job.  (And complaints are welcome if you have problems.)  Try:   2       ftp://ftp.info-zip.org/pub/infozip/OLD/beta/  =       ftp://ftp.info-zip.org/pub/infozip/OLD/beta/unz600c.ann =       ftp://ftp.info-zip.org/pub/infozip/OLD/beta/unz600c.zip   <       ftp://ftp.info-zip.org/pub/infozip/OLD/beta/zip30e.zip  D If you lack a C compiler on VMS, I can supply Alpha binaries easily,E and, after a few days' delay (probably), VAX.  (IA64, too, I suppose,  using an HP TestDrive system.)  F > Alternately, is there a way to get VMS Backup to write multiple save2 > sets so that each is below a given maximum size?  $    In general, no, so far as I know.  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------   End of INFO-VAX 2005.709 ************************                                                                                                                                                                                                                                                                                                                                                                                                                      \0)NM.)1]/p\)-۸.kb$+rL:29
ueEuՈ(*Ad@H1?;Z6[+՜${)`ˬ̧&ABzT\ЪlVSd4{R^jD1Q֬}ry
qI 
]8]kewzlSzYӲRm*Ê3tFMWb&2l\25*j7̜sL(&5}|ѱ0-T/5}5Ugk]V[t"n/8یQQȷIeO5բrv\X0i3(0Zo-P$L㏲/$d(@Y^Ѵt5!{ysM[aSgF%oܦ&tT	Lz܏M^RE*%rcR(BC]
/E{mc4@]xwwkcS(gr,>0ąIFO#=N*IwM]o̍/!3Ӓ`h$'=85f4 Lys%\h#gk!سbJwÓӣ/UMq*'P1S8f)21Fqv}0Zu!ӵzx<LP'wpzd۶n??Ƽ]ʮ7 rk.N{\7s
It2.t,~9(6jmcݮx.SG|vK`SU(!FG,`Bt#Vep4QФQ!x
]WRN,4jL=޶۫u9џ8TNnY,*cqO<`VDQK
&i\d/ԗ]pSs/uQLU`Hܸ^ΚaĲVAayz<~^?O _3Ua_J49]9Wm7><xjf
H5/;_ܾ	JW	W2X2!cmV[ NlN*txekĭ?}=pQP[+ҏL(68?>mH,요BikR
y[䵝k; li-]۱kE~tNVx׏;kE) qjk::],dmDyXW96LL-iy`EᬟYx6@rHՁ]LYtL.[rϏ^2k y#cEjGӒk}9Qt~s@?7;elw)5?av0<LUܘ=?,6{}pݡl8W<ҋA;.r\n˭su.rܥ1K#o_Pn_ĺN+?-܂u_Ys/<
#sd'ٻ^܉aF?c0._oD1o8M޲\l8/L.mk sSRxZ\贘RUޮI)`&9sMLa(n_-olUYM?k_`ZƘ;%AV憅AT5(].E^cJqQgGTew~$YJtSHJqR/܉7	P78dscPY>ߔ`S&`RmhaVZ/GqB9 mMS67ޚ/ګMS	ArdF{09&Ѥ3'}Œ[?`[
垮 U}wU=xI`$(/0?A5Ǵ?L!o;ɖÈ)\m.2:a'"/$tef.>JN DAlE6K^|6tYad7+0R=Sb|މ|X-c#*QcOY灹OrrN"M2Au<c?[#>O] 0Yu8
9sjY:P*3
3&5r֋B<y}8alF;d$pڰscU<e<tSJ##ĂĤ6|d!`u`0r`8xK߬#uN㑟̀6UD&%-|c;]qk!M\̡OfG_ŴGtFo]R&Wk\Y9>4j0If!Ԉ9^NOpA6U"kc4Sp.b3u|3ÕtѵSsh:7΍sl4̳[g4ay|
h]7*.dk ?dD`z y<fVk0se2Fd_W#73}D:W2rR.1Eլ҉;jSd