1 INFO-VAX	Thu, 21 Dec 2006	Volume 2006 : Issue 701       Contents: lpr printer and Landscape  Re: lpr printer and Landscape , Re: Open sourced VMS as a business concept ?, Re: Open sourced VMS as a business concept ?, Re: Open sourced VMS as a business concept ?, Re: Open sourced VMS as a business concept ?, Re: Open sourced VMS as a business concept ?, Re: Open sourced VMS as a business concept ?, Re: Open sourced VMS as a business concept ?, Re: Open sourced VMS as a business concept ?$ Re: SAN disk access problem...solved$ Re: SAN disk access problem...solved* Re: Tools to rebuild corrupt indexed files* Re: Tools to rebuild corrupt indexed files Wiki server for VMS?  F ----------------------------------------------------------------------  % Date: Thu, 21 Dec 2006 16:01:47 +0200   From: "Mandi" <mandi@mandat.com>" Subject: lpr printer and Landscape1 Message-ID: <eme3v2$rgb$1@news2.netvision.net.il>    Hi all, L I have a new HP printer which I setup as lpr printer in vms 6.2 environment.I Trying to print "Landscape" doesn't work. Someone said there is a special 6 procedure with form and libraries that should be done.$ Is there any documentation for that?   Thanks in advance    Mandi    ------------------------------  # Date: Thu, 21 Dec 2006 18:51:51 GMT % From: Rob Brown <mylastname@gmcl.com> & Subject: Re: lpr printer and LandscapeE Message-ID: <Pine.LNX.4.61.0612211141280.20642@localhost.localdomain>   ! On Thu, 21 Dec 2006, Mandi wrote:   B > I have a new HP printer which I setup as lpr printer in vms 6.2 F > environment. Trying to print "Landscape" doesn't work. Someone said F > there is a special procedure with form and libraries that should be , > done. Is there any documentation for that?  E I don't know how "lpr printer" fits into all of this, but "forms and  G libraries" is the way to give you landscape and portrait capabilities.  6 On my system here, I have the following forms defined:   $ sho que/form9 Form name                            Number   Description 9 ---------                            ------   ----------- D DEFAULT                                   0   System-defined defaultJ HP_L10 (stock=DEFAULT)                    4   HP LaserJet Landscape 10 CPIJ HP_L12 (stock=DEFAULT)                    5   HP LaserJet Landscape 12 CPIO HP_L16 (stock=DEFAULT)                    6   HP LaserJet Landscape 16.66 CPI 8 2                                                LPII HP_P10 (stock=DEFAULT)                    1   HP LaserJet Portrait 10 CPI I HP_P12 (stock=DEFAULT)                    2   HP LaserJet Portrait 12 CPI P HP_P16 (stock=DEFAULT)                    3   HP LaserJet Portrait 16.66 CPI 8 L1                                                PI    The short story is:   A 1.  create text files containing the escape sequences to put the   printer in the desired mode.  4 2.  add the text files to the device control library  F 3.  define your forms and have them reference the appropriate modules  in the device control library   E See the _OpenVMS System Manager's Manual:  Essentials_.  The chapter  @ of interest is "Setting Up and Maintaining Queues".  Of special G interest are the sections "Using and Creating Forms" and "Using Device   Control Libraries".   - You can find this (and other VMS manuals) at  # <http://www.hp.com/go/openvms/doc>.   	 Have fun.      --    B Rob Brown                        b r o w n a t g m c l d o t c o m6 G. Michaels Consulting Ltd.      (780)438-9343 (voice)4 Edmonton                         (780)437-3367 (FAX)2                                   http://gmcl.com/   ------------------------------  # Date: Thu, 21 Dec 2006 07:20:21 GMT 5 From: rdeininger@mindspringdot.com (Robert Deininger) 5 Subject: Re: Open sourced VMS as a business concept ? Z Message-ID: <rdeininger-2112060220370001@dialup-4.233.149.69.dial1.manchester1.level3.net>  > In article <40d5$45894b69$cef8887a$831@TEKSAVVY.COM>, JF Mezei& <jfmezei.spamnot@vaxination.ca> wrote:   >Ian Miller wrote:& >> ACMS is now supported by a HP team. >  > L >What does "HP team" really mean ? Just people working out of home wherever ; >they may be on this planet, or someone physically at ZKO ?  > I >Back in the days of Digital, was VMS engineering pretty much a physical  L >team in a couple of locations ?  (ZKP, Reading, Valbonnes and what other ?)  G If you include VMS and layered products, Digital had people in multiple G locations in New Hampshire and Massachusetts, as well as contingents in J Colorado, several European countries, and India.  And I'm sure I'm missing some.    ------------------------------   Date: 21 Dec 2006 13:04:56 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)5 Subject: Re: Open sourced VMS as a business concept ? 0 Message-ID: <4uvf3oF1a6cqlU1@mid.individual.net>  Z In article <rdeininger-2112060220370001@dialup-4.233.149.69.dial1.manchester1.level3.net>,8 	rdeininger@mindspringdot.com (Robert Deininger) writes:@ > In article <40d5$45894b69$cef8887a$831@TEKSAVVY.COM>, JF Mezei( > <jfmezei.spamnot@vaxination.ca> wrote: >  >>Ian Miller wrote: ' >>> ACMS is now supported by a HP team.  >> >>M >>What does "HP team" really mean ? Just people working out of home wherever  < >>they may be on this planet, or someone physically at ZKO ? >>J >>Back in the days of Digital, was VMS engineering pretty much a physical M >>team in a couple of locations ?  (ZKP, Reading, Valbonnes and what other ?)  > I > If you include VMS and layered products, Digital had people in multiple I > locations in New Hampshire and Massachusetts, as well as contingents in L > Colorado, several European countries, and India.  And I'm sure I'm missing > some.   D Well, Digital used to have an office right here in Wilkes-Barre, PA.C But, according to the guy int he office next door, the CE's came in C one morning to find the locks on the door had been changed and they D were all bright enough to read the writting on the wall.  Don't knowC where they went, but I do know none of them still live around here. C (By the way, that being as it was still Digital then, it really was  quite soem time ago.)    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------   Date: 21 Dec 2006 13:15:00 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)5 Subject: Re: Open sourced VMS as a business concept ? 0 Message-ID: <4uvfmkF1a6cqlU2@mid.individual.net>  / In article <emdd38$jsq$03$1@news.t-online.com>, + 	Michael Kraemer <M.Kraemer@gsi.de> writes: 3 > Phillip Helbig---remove CLOTHES to reply schrieb:  >  >>  H >> Simple.  What differentiates VMS from other operating systems is the I >> fact that it is well planned.  Without meaning any harm to Linus, the  H >> fact that a student could write a unix kernel in his spare time says J >> more about the lack of sophistication of the latter than the skills of  >> the former.   > ( > No it says very little about anything.; > The kernel is probably the easiest part of a Unix system,   + Well, it's the easiest part to write badly.   , > in terms of man power. Probably of any OS.6 > (not that I would be able to write one, but a gifted >   CS student should be)   B That can't be true or it would have left Linus out of the picture.  1 > It took about a decade and much more additional 2 > man power until Linux was ready for serious use.  D Linux still isn't "ready for serious use".  And it likely never will. be.  But that won't stop people from doing it.   > > >> I don't see how VMS could continue in an open-source world L >> without a lot of coordination, and I don't see anyone with the knowledge K >> and resources willing to do this for free.  A lot of work is going into  J >> VMS now and this is funded by customers.  Cut off the funding, cut out K >> the development.  Many of the customers would not be willing to pay for   >> an open-source product. >  > Did you ask ?  > I >> The big myth here is that somehow VMS is not everywhere because it is  I >> not free.  Well, Windows is everywhere and it is not free.  There are  H >> free operating systems which occupy a niche, and free ones with many J >> users.  And for hobbyists, VMS IS free.  Are the masses signing up for 3 >> DECUS memberships to get hobbyist licenses?  No.  > . > Well, compared to hobbyist use of other OSs,  A What other OSes?  I am aware of no current OS that has a hobbyist & program that offers anything like VMS.  : > it is still a PITA to deal with PAKs and annual renewal.A > At least the latter is in argument that I heard more than once.   A I have never found that to be the case.  The only problems I have E ever had were the timeing of the EDU license expiration (which really A has nothing to do with a Hobbyist License) and when I am carelees < enough to let mine expire without having asked for new ones.   > E >> And another danger of going the open-source route is the danger of  >> approaching the RMS camp,   > E > I see more danger in software patents, IP religion and DRM rip-off.  > Much more. > 1 >> where not only should all software be free but D >> people who write software for a living are brandmarked as morallyK >> inferior.  Read some RMS speeches; it is difficult to find anything more F >> hateful and inhumane in all of literature.  If he wants to make his >> software free, fine.    > " > RMS, that's Stallmann, I guess ?6 > Come on, you're taking that much too serious. Relax.  E I am amazed to find someone who has read and understood RMS enough to F share my concern.  The real scary part is that too many people alreadyE take him "much too serious".  If more people actually understood what E led to what is called the FSF today (and the true shortcomings of the F whole GNU concept from an RMS perspective) things would likely be much
 different.  4 > What I find annoying though, is that the Gnu stuff3 > has turned to be single platform only, basically. 8 > In former times it was relatively easy to compile/port7 > on the platform of choice. Now it's practically Linux  > only.   D Sound like anyone else we know?  And people say proprietary OSes are
 a bad thing!!    > G >> Of course, HP owns VMS and HP makes money from VMS so even thinking  3 >> about them open-sourcing it is idle speculation.  >>   > A > Wasn't it the common opinion in this group that HP doesn't want  > to make money from VMS ?   B Doesn't mean there is a chance to get it open sourced.  Just think? how lucky you are to just have a hobbyist program.  Ask them to 5 allow hobbyist use of Ultrix-32 and see what happens.    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: 21 Dec 2006 08:09:55 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 5 Subject: Re: Open sourced VMS as a business concept ? 3 Message-ID: <+mLNIXdox+xz@eisner.encompasserve.org>   Z In article <emdd38$jsq$03$1@news.t-online.com>, Michael Kraemer <M.Kraemer@gsi.de> writes:3 > Phillip Helbig---remove CLOTHES to reply schrieb:  > ( > No it says very little about anything.; > The kernel is probably the easiest part of a Unix system, , > in terms of man power. Probably of any OS.  D    The kernel in VMS is not so easy.  It's quite complex compared toA    the original time-sharing-only UNIX kernel.  Part of that more G    complex design is why it still has better realtime capabilities than D    the UNIX kernels (like Tru64 and Solaris) that have had realtime G    capabilities added (and are clearly more complex than the original).   A    Linux originally was also a pure timesharing kernel, and quite H    simple.  It's getting realtime pieces but I don't know if it can even    keep up with Solaris yet.  C    How easy a kernel is depends on how simple it's capabilties are. B    Compare the above to Vxworks, which used a simple realtime-onlyC    design and is just now adding the complexity needed for features F    like memory protection that a timesharing or general purpose OS had    starting back in the '60s.   F    A general purpose OS, like VMS, which has both good timesharing and*    good realtime capabilties, is not easy.   ------------------------------    Date: 21 Dec 2006 08:13:15 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 5 Subject: Re: Open sourced VMS as a business concept ? 3 Message-ID: <kNiiFmcvqxEr@eisner.encompasserve.org>   [ In article <4uvfmkF1a6cqlU2@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:     C > What other OSes?  I am aware of no current OS that has a hobbyist ( > program that offers anything like VMS.  C    Thats what we need:  a hobbyist program for Windows!  Sure would     cut into Billy's pockets.  F    Seriously, friends tell me Solaris is pretty easy on their hobbyistI    pocketbooks, too (free download).  Tru64 was for a while at only $99,      but I think that's gone now.    ------------------------------    Date: 21 Dec 2006 07:12:08 -0800- From: "Andrew" <andrew_harrison@symantec.com> 5 Subject: Re: Open sourced VMS as a business concept ? B Message-ID: <1166713928.111788.203920@f1g2000cwa.googlegroups.com>   Bill Gunshannon wrote:1 > In article <emdd38$jsq$03$1@news.t-online.com>, - > 	Michael Kraemer <M.Kraemer@gsi.de> writes: 5 > > Phillip Helbig---remove CLOTHES to reply schrieb:  > >  > >>I > >> Simple.  What differentiates VMS from other operating systems is the J > >> fact that it is well planned.  Without meaning any harm to Linus, theI > >> fact that a student could write a unix kernel in his spare time says K > >> more about the lack of sophistication of the latter than the skills of  > >> the former. > > * > > No it says very little about anything.= > > The kernel is probably the easiest part of a Unix system,  > - > Well, it's the easiest part to write badly.  > . > > in terms of man power. Probably of any OS.8 > > (not that I would be able to write one, but a gifted > >   CS student should be)  > D > That can't be true or it would have left Linus out of the picture. > 3 > > It took about a decade and much more additional 4 > > man power until Linux was ready for serious use. > F > Linux still isn't "ready for serious use".  And it likely never will0 > be.  But that won't stop people from doing it. >  > > ? > >> I don't see how VMS could continue in an open-source world M > >> without a lot of coordination, and I don't see anyone with the knowledge L > >> and resources willing to do this for free.  A lot of work is going intoK > >> VMS now and this is funded by customers.  Cut off the funding, cut out L > >> the development.  Many of the customers would not be willing to pay for > >> an open-source product. > >  > > Did you ask ?  > > J > >> The big myth here is that somehow VMS is not everywhere because it isJ > >> not free.  Well, Windows is everywhere and it is not free.  There areI > >> free operating systems which occupy a niche, and free ones with many K > >> users.  And for hobbyists, VMS IS free.  Are the masses signing up for 5 > >> DECUS memberships to get hobbyist licenses?  No.  > > 0 > > Well, compared to hobbyist use of other OSs, > C > What other OSes?  I am aware of no current OS that has a hobbyist ( > program that offers anything like VMS.  G How about Solaris, its free to download for either SPARC or x86 and you F get the full release including compliers.  You can also get the sourceE code via OpenSolaris.org. The only thnig Sun doesn't have is anything D called a hobbyist program however everything else is actually rather better.    regards  Andrew Harrison  > < > > it is still a PITA to deal with PAKs and annual renewal.C > > At least the latter is in argument that I heard more than once.  > C > I have never found that to be the case.  The only problems I have G > ever had were the timeing of the EDU license expiration (which really C > has nothing to do with a Hobbyist License) and when I am carelees > > enough to let mine expire without having asked for new ones. >  > > G > >> And another danger of going the open-source route is the danger of  > >> approaching the RMS camp, > > G > > I see more danger in software patents, IP religion and DRM rip-off.  > > Much more. > > 3 > >> where not only should all software be free but F > >> people who write software for a living are brandmarked as morallyM > >> inferior.  Read some RMS speeches; it is difficult to find anything more H > >> hateful and inhumane in all of literature.  If he wants to make his > >> software free, fine.  > > $ > > RMS, that's Stallmann, I guess ?8 > > Come on, you're taking that much too serious. Relax. > G > I am amazed to find someone who has read and understood RMS enough to H > share my concern.  The real scary part is that too many people alreadyG > take him "much too serious".  If more people actually understood what G > led to what is called the FSF today (and the true shortcomings of the H > whole GNU concept from an RMS perspective) things would likely be much > different. > 6 > > What I find annoying though, is that the Gnu stuff5 > > has turned to be single platform only, basically. : > > In former times it was relatively easy to compile/port9 > > on the platform of choice. Now it's practically Linux 	 > > only.  > F > Sound like anyone else we know?  And people say proprietary OSes are > a bad thing!!  >  > > H > >> Of course, HP owns VMS and HP makes money from VMS so even thinking5 > >> about them open-sourcing it is idle speculation.  > >> > > C > > Wasn't it the common opinion in this group that HP doesn't want  > > to make money from VMS ? > D > Doesn't mean there is a chance to get it open sourced.  Just thinkA > how lucky you are to just have a hobbyist program.  Ask them to 7 > allow hobbyist use of Ultrix-32 and see what happens.  >  > bill >  > --L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------   Date: 21 Dec 2006 18:29:17 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)5 Subject: Re: Open sourced VMS as a business concept ? 0 Message-ID: <4v023tF19ofcdU1@mid.individual.net>  3 In article <kNiiFmcvqxEr@eisner.encompasserve.org>, > 	koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:] > In article <4uvfmkF1a6cqlU2@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:  >  > D >> What other OSes?  I am aware of no current OS that has a hobbyist) >> program that offers anything like VMS.  > E >    Thats what we need:  a hobbyist program for Windows!  Sure would  >    cut into Billy's pockets. > H >    Seriously, friends tell me Solaris is pretty easy on their hobbyistK >    pocketbooks, too (free download).  Tru64 was for a while at only $99,  ! >    but I think that's gone now.   H I don't think that counts as a "Hobbyist Program" as anyone can downloadF it.  even businesses and they can use it for their business with Sun'sG blessing.  Small businesses have been able to do that since long before  this "open source" initiative.   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: 21 Dec 2006 18:33:35 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)5 Subject: Re: Open sourced VMS as a business concept ? 0 Message-ID: <4v02bvF19ofcdU2@mid.individual.net>  B In article <1166713928.111788.203920@f1g2000cwa.googlegroups.com>,0 	"Andrew" <andrew_harrison@symantec.com> writes: >  > Bill Gunshannon wrote: >>D >> What other OSes?  I am aware of no current OS that has a hobbyist) >> program that offers anything like VMS.  > I > How about Solaris, its free to download for either SPARC or x86 and you H > get the full release including compliers.  You can also get the sourceG > code via OpenSolaris.org. The only thnig Sun doesn't have is anything F > called a hobbyist program however everything else is actually rather	 > better.   F My point, exactly.  It isn't a "Hobbyist Program".  Anyone can use it,E even businesses.  I want to know what current, proprietary OSes offer F a Hobbyist Program and how it compares to the VMS Program.  I am awareB of none, although I know of a number of platforms that have active4 Hobbyist communities who would like to see programs.   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: 21 Dec 2006 08:06:58 -0800( From: "rgscomp" <ray@rgscomputing.co.uk>- Subject: Re: SAN disk access problem...solved B Message-ID: <1166717218.469988.96430@n67g2000cwd.googlegroups.com>   rgscomp wrote:    E > I'm struggling with the idea that the console can see dual paths to I > every disk, but VMS can only see the system disk as dual pathed. As its I > the same driver, I can't think how it can be selective over which disks  > to "see" dual-pathed.   E Ok, replying to my own post. It seems that if you present LUN 0 (some F sort of pseudo-device that holds the SAN database) to VMS systems fromE an EMC array, this is precisely what you see. In our case, the system F disk is listed first in the wwidmgr output, followed by LUN 0 and then the rest of the disks.E So we see the system disk multi-pathed, but all disks after the LUN 0 D entry are only single-pathed (along with other assorted connectivity wierdness).   C Our DR site was working and it seems that the SAN guys here did not G folow the instructions to mask off LUN 0, but the DR site did. Also, an B old Digital colleague confirmed he had seen the problem on anotherB customer site recently and again, masking LUN 0 fixed the problem.  2 We are just awaiting a window to apply the change.   Cheers,    Ray.   ------------------------------    Date: 21 Dec 2006 09:00:01 -0800; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> - Subject: Re: SAN disk access problem...solved B Message-ID: <1166720401.142213.245870@79g2000cws.googlegroups.com>   rgscomp wrote: > rgscomp wrote: >  > G > > I'm struggling with the idea that the console can see dual paths to K > > every disk, but VMS can only see the system disk as dual pathed. As its K > > the same driver, I can't think how it can be selective over which disks  > > to "see" dual-pathed.  > G > Ok, replying to my own post. It seems that if you present LUN 0 (some H > sort of pseudo-device that holds the SAN database) to VMS systems fromG > an EMC array, this is precisely what you see. In our case, the system H > disk is listed first in the wwidmgr output, followed by LUN 0 and then > the rest of the disks.G > So we see the system disk multi-pathed, but all disks after the LUN 0 F > entry are only single-pathed (along with other assorted connectivity
 > wierdness).  > E > Our DR site was working and it seems that the SAN guys here did not I > folow the instructions to mask off LUN 0, but the DR site did. Also, an D > old Digital colleague confirmed he had seen the problem on anotherD > customer site recently and again, masking LUN 0 fixed the problem. > 4 > We are just awaiting a window to apply the change. > 	 > Cheers,  >  > Ray.  D Oops.  Yeah, presenting LUN 0 to a VMS system is not good.  IIRC VMSG uses LUN 0 as a control path to the storage controller.  Actual storage B units need to start out with LUN 1 at a minimum. HSG80's have someC other rules, but obviously, using EMC units you don't have to worry  about those.   ------------------------------    Date: 21 Dec 2006 01:33:34 -06004 From: cornelius@encompasserve.org (George Cornelius)3 Subject: Re: Tools to rebuild corrupt indexed files 3 Message-ID: <SHYrEP8wtKXy@eisner.encompasserve.org>   c In article <sZydneAq64aMthfYnZ2dnUVZ_vyunZ2d@libcom.com>, Dave Froble <davef@tsoft-inc.com> writes: J > Actually, RMS isn't too hard to work with.  It's actually a simple file E > structure, and some simple programming can process the buckets and  , > attempt to determine what's in the bucket. > J > No guarantees, but I've recovered a few RMS files.  As mentioned above, F > it's rather trivial when it's secondary key structures that are bad.  G Is the converse true?  Can you read sequentially by an alternate key if % the primary key structure is corrupt?   M Hopefully RMS does not do something like keeping only one copy of the primary K key and omitting it from the data portion.  Not that I would blame them for I doing that as an implementation strategy, since the key(s) can constitute N a large part of some data sets.  Also: there's a technique I've seen elsewhereL (DSM) of only storing the portion of the key that differs from the precedingK (and possibly B-tree parent?) key value - very helpful for large data sets, I but another complication when trying to recover from database corruption.    --? George Cornelius              cornelius@SPAMNOTeisner.decus.org 7                               cornelius@SPAMNOTmayo.edu    ------------------------------    Date: 21 Dec 2006 07:20:17 -0800< From: "Hein RMS van den Heuvel" <heinvandenheuvel@gmail.com>3 Subject: Re: Tools to rebuild corrupt indexed files B Message-ID: <1166714417.839888.235020@80g2000cwy.googlegroups.com>   George Cornelius wrote: e > In article <sZydneAq64aMthfYnZ2dnUVZ_vyunZ2d@libcom.com>, Dave Froble <davef@tsoft-inc.com> writes: K > > Actually, RMS isn't too hard to work with.  It's actually a simple file F > > structure, and some simple programming can process the buckets and. > > attempt to determine what's in the bucket.  E Right. And in the 'bonus' directory for the rms_tools freeware you'll  find some of that.   > > K > > No guarantees, but I've recovered a few RMS files.  As mentioned above, H > > it's rather trivial when it's secondary key structures that are bad.  D It is also trivial when the primary index structure is somewhat bad.  E A simple convert (or $type) reads: prologue, 'left' side of the index G from top to bottom, data buckets following next bucket pointer. note... G rms records the frist data bucket vbn, so I does not really need to use & the index to find that... but it does.     > I > Is the converse true?  Can you read sequentially by an alternate key if ' > the primary key structure is corrupt?   $ Much like waht Bill already replied.  C The Secondary index record contain 'RFAs', not pprimary key values. E An RFA is a VBN number plust record number. No index or key structure  needed.      > O > Hopefully RMS does not do something like keeping only one copy of the primary , > key and omitting it from the data portion.  F The primary key is extracted from the data record and put there as the first bytes.F It is not repeated. With key compression on, any matching portion from> the prior key is represented as a byte count and not repeated.  / > Also: there's a technique I've seen elsewhere N > (DSM) of only storing the portion of the key that differs from the precedingM > (and possibly B-tree parent?) key value - very helpful for large data sets,   - Yes, rms key and index compression does this.   K > but another complication when trying to recover from database corruption.    SMOP   Cheers,  Hein.    ------------------------------    Date: 21 Dec 2006 08:08:59 -0800( From: "Rich Jordan" <jordan@ccs4vms.com> Subject: Wiki server for VMS? B Message-ID: <1166717339.077740.156290@f1g2000cwa.googlegroups.com>  C Has anyone gotten an open source Wiki running under VMS/Apache?  Is E this something like other PHP (or Perl?) packages that can almost run . without any modifications?  Database required?   Pointers appreciated.  Thanks!     Rich Jordan  CCS    ------------------------------   End of INFO-VAX 2006.701 ************************